Multiparameter flow cytometry in the diagnosis of hematologic malignancies [1 ed.] 9781107503830, 1107503833

Citation preview

Cambridge University Press 978-1-107-50383-0 — Multiparameter Flow Cytometry in the Diagnosis of Hematologic Malignancies Edited by Anna Porwit , Marie Christine Béné Frontmatter More Information

Multiparameter Flow Cytometry in the Diagnosis of Hematologic Malignancies

© in this web service Cambridge University Press

www.cambridge.org

Cambridge University Press 978-1-107-50383-0 — Multiparameter Flow Cytometry in the Diagnosis of Hematologic Malignancies Edited by Anna Porwit , Marie Christine Béné Frontmatter More Information

Multiparameter Flow Cytometry in the Diagnosis of Hematologic Malignancies Edited by

Anna Porwit Lund University, Sweden

Marie-Christine Béné University of Nantes, France

© in this web service Cambridge University Press

www.cambridge.org

Cambridge University Press 978-1-107-50383-0 — Multiparameter Flow Cytometry in the Diagnosis of Hematologic Malignancies Edited by Anna Porwit , Marie Christine Béné Frontmatter More Information

University Printing House, Cambridge CB2 8BS, United Kingdom One Liberty Plaza, 20th Floor, New York, NY 10006, USA 477 Williamstown Road, Port Melbourne, VIC 3207, Australia 314–321, 3rd Floor, Plot 3, Splendor Forum, Jasola District Centre, New Delhi – 110025, India 79 Anson Road, #06-04/06, Singapore 079906 Cambridge University Press is part of the University of Cambridge. It furthers the University’s mission by disseminating knowledge in the pursuit of education, learning and research at the highest international levels of excellence. www.cambridge.org Information on this title: www.cambridge.org/9781107503830 DOI: 10.1017/9781316218549 © Cambridge University Press 2018 his publication is in copyright. Subject to statutory exception and to the provisions of relevant collective licensing agreements, no reproduction of any part may take place without the written permission of Cambridge University Press. First published 2018 Printed in the United Kingdom by Clays, St Ives plc A catalogue record for this publication is available from the British Library. Library of Congress Cataloging-in-Publication Data Names: Porwit, Anna, editor. | Bene, Marie-Christine, editor. Title: Multiparameter low cytometry in the diagnosis of hematologic malignancies / edited by Anna Porwit, Marie Christine Bene. Description: Cambridge, United Kingdom ; New York, NY : Cambridge University Press, 2018. | Includes bibliographical references and index. Identiiers: LCCN 2017042275 | ISBN 9781107503830 (paperback) Subjects: | MESH: Hematologic Neoplasms—diagnosis | Flow Cytometry—methods Classiication: LCC RC280.H47 | NLM WH 525 | DDC 616.99/418—dc23 LC record available at https://lccn.loc.gov/2017042275 ISBN 978-1-107-50383-0 Paperback Cambridge University Press has no responsibility for the persistence or accuracy of URLs for external or third-party internet websites referred to in this publication and does not guarantee that any content on such websites is, or will remain, accurate or appropriate. Every efort has been made in preparing this book to provide accurate and up-to-date information that is in accord with accepted standards and practice at the time of publication. Although case histories are drawn from actual cases, every efort has been made to disguise the identities of the individuals involved. Nevertheless, the authors, editors and publishers can make no warranties that the information contained herein is totally free from error, not least because clinical standards are constantly changing through research and regulation. he authors, editors and publishers therefore disclaim all liability for direct or consequential damages resulting from the use of material contained in this book. Readers are strongly advised to pay careful attention to information provided by the manufacturer of any drugs or equipment that they plan to use.

© in this web service Cambridge University Press

www.cambridge.org

Cambridge University Press 978-1-107-50383-0 — Multiparameter Flow Cytometry in the Diagnosis of Hematologic Malignancies Edited by Anna Porwit , Marie Christine Béné Frontmatter More Information

Contents List of Contributors vii Preface ix List of Abbreviations xi

1

Flow Cytometry in Clinical Haematopathology: Basic Principles and Data Analysis of Multiparameter Data Sets 1 Francis Lacombe and Marie-Christine Béné

2

Antigens 14 Marie-Christine Béné and Anna Porwit

3

Flow Cytometry of Normal Blood, Bone Marrow and Lymphatic Tissue 36 Anna Porwit and Marie-Christine Béné

4

5

6

Reactive Conditions and Other Diseases Where Flow Cytometric Findings May Mimic Haematological Malignancies 61 Wolfgang Kern, Marie-Christine Béné and Anna Porwit Examples of Immunophenotypic Features in Various Categories of Acute Leukaemia 75 Marie-Christine Béné and Anna Porwit Acute Lymphoid Leukaemias (All) and Minimal Residual Disease in All 89 Giuseppe Basso, Barbara Buldini and Andrea Zangrando

7

Immunophenotyping of Mature B-Cell Lymphomas 105 Olof Axler and Anna Porwit

8

Plasma Cell Myeloma and Related Disorders 128 Ruth M. de Tute, Andrew C. Rawstron and Roger G. Owen

9

Mature T-Cell Neoplasms and Natural Killer-Cell Malignancies 140 Anne Tierens

10 Flow Cytometric Diagnosis of Hodgkin’s Lymphoma in Lymph Nodes 161 Lori Soma, Brent L. Wood and Jonathan R. Fromm 11 Minimal Residual Disease in Acute Myeloid Leukaemia 171 Gerrit J. Schuurhuis, Angèle Kelder, Gert J. Ossenkoppele, Jacqueline Cloos and Wendelien Zeijlemaker 12 Ambiguous Lineage and Mixed Phenotype Acute Leukaemia 191 Anna Porwit and Marie-Christine Béné 13 Flow Cytometry in Myelodysplastic Syndromes 199 heresia M. Westers and Arjan A. van de Loosdrecht 14 Future Applications of Flow Cytometry and Related Techniques 215 Marie-Christine Béné and Francis Lacombe

Index 231

v

© in this web service Cambridge University Press

www.cambridge.org

Cambridge University Press 978-1-107-50383-0 — Multiparameter Flow Cytometry in the Diagnosis of Hematologic Malignancies Edited by Anna Porwit , Marie Christine Béné Frontmatter More Information

Contributors

Olof Axler Lund University, Sweden Giuseppe Basso University of Padova, Italy Barbara Buldini University of Padova, Italy Jacqueline Cloos VU University Medical Center, Amsterdam, he Netherlands Ruth M. de Tute Leeds University, UK Jonathan R. Fromm University of Washington, Seattle, Washington, USA Angèle Kelder VU University Medical Center, Amsterdam, he Netherlands Wolfgang Kern MLL Munich Leukemia Laboratory, Munich, Germany

Gert J. Ossenkoppele VU University Medical Center, Amsterdam, he Netherlands Roger G. Owen Leeds University, UK Andrew C. Rawstron Leeds University, UK Gerrit J. Schuurhuis VU University Medical Center, Amsterdam, he Netherlands Lori Soma University of Washington, Seattle, Washington, USA Anne Tierens University of Toronto, Canada Theresia M. Westers VU University Medical Center, Amsterdam, he Netherlands Brent L. Wood University of Washington, Seattle, Washington, USA

Francis Lacombe Bordeaux University, France

Andrea Zangrando University of Padova, Italy

Arjan A. van de Loosdrecht VU University Medical Center, Amsterdam, he Netherlands

Wendelien Zeijlemaker VU University Medical Center, Amsterdam, he Netherlands

vii

© in this web service Cambridge University Press

www.cambridge.org

Cambridge University Press 978-1-107-50383-0 — Multiparameter Flow Cytometry in the Diagnosis of Hematologic Malignancies Edited by Anna Porwit , Marie Christine Béné Frontmatter More Information

Preface

Together with cytological examination, low cytometry is oten the irst exploration step in patients with clinical symptoms suggesting haematological malignancy or with fortuitously discovered anomalies in a whole blood cell count. Depending on the healthcare organisation, low cytometry results will stand alone and be discussed later during a diagnostic conference, or be integrated in a comprehensive set of investigations including bone marrow biopsy morphology, cytogenetics and sophisticated molecular studies. Over the years, knowledge and skills have developed so that in many cases the subtleties of the sets of markers, as well as their expression of absence, have become familiar to clinicians expecting a diagnosis. Yet, the thousands of references in the literature, dealing with this speciic part of laboratory haematology, provide a good idea of the puzzlement that may overwhelm any novice in the ield. hinking about the outlines of this book, we placed ourselves in the position of a young laboratory haematologist or haematopathologist and wondered which questions would need an answer likely to be found in a single document. We then asked Expert Friends to work with us with this aim in mind. Moreover, we wanted to focus on the new 8- and 10-colour methodologies. We decided to start with basic characteristics of the structure and functions of low cytometers, trying to provide a clear explanation of what sometimes seems to be very complex. We also depicted the analysis tools available in current sotware to make the most of acquired data (Chapter 1).

We then collected pertinent information about the structure, function and expression of a large number of the antigens investigated in low cytometry, all mentioned somewhere in this book, together with a brief history of the way they were characterised or discovered (Chapter 2). Before tackling pathological issues, we thought that readers would appreciate some information about what to expect when low cytometry is applied to normal samples of blood, bone marrow or lymphatic tissue (Chapter 3). We also listed a series of non-malignant conditions where the hypothesis of malignancy is plausible and must be ruled out (Chapter 4). In Chapter 5, we present a collection of typical low cytometry graphs characteristic for various categories of acute leukaemia. From Chapters 6 to 13, the authors considered speciic sets of diseases and their idiosyncratic low cytometry features. Finally, Chapter 14 provides a glimpse at what lays ahead, in the already foreseeable developments of the versatile and powerful technology of cell analysis. We built this book, not only as a manual that may be read through while starting to work with low cytometry diagnostics, but also as a reference document to consult when interested in any aspect of low cytometry diagnostics of haematological malignancies. We hope that, together with our co-authors, we have reached that goal. Anna Porwit and Marie-Christine Béné

ix

© in this web service Cambridge University Press

www.cambridge.org

Cambridge University Press 978-1-107-50383-0 — Multiparameter Flow Cytometry in the Diagnosis of Hematologic Malignancies Edited by Anna Porwit , Marie Christine Béné Frontmatter More Information

Abbreviations

7-AAD AIHA AITL ALL AML APC APL ATLL BCP ALL BCR BCR-ABL

BDCA BF BM BPDCN BR CALLA CAR T-cell CBF CCR CD ChIP CLL CLPD CML CMML CRLF2 Cy DAPI DC DLBCL DNA

7-aminoactinomycin D Auto immune haemolytic anaemia Angio-immunoblastic T-cell lymphoma Acute lymphoblastic leukaemia Acute myeloid (or myeloblastic) leukaemia Allophycocyanin Acute promyelocytic leukemia Adult T-cell leukemia/lymphoma B-cell progenitor acute lymphoblastic leukaemia B-cell receptor Breakpoint cluster region-abelson [t(9;22) also called Philadelphia chromosome] Blood-derived dendritic cell antigens Body luid Bone marrow Blastic plasmacytoid dendritic cell neoplasm Blast region Common acute lymphoblastic leukemia antigen Chimeric antigen receptor T-cell Core binding factor Chemokine receptor Cluster of diferentiation Chromatin immunoprecipitation Chronic lymphocytic leukemia Chronic lymphoproliferative disorders Chronic myeloid leukaemia Chronic myelomonocytic leukaemia Cytokine receptor-like factor Cyanin 4′,6-diamidino-2-phenylindole Dendritic cell Difuse large B-cell lymphoma Deoxyribonucleic acid

© in this web service Cambridge University Press

ECD

EDTA ETP-ALL FCM FITC FNA FL FLAER FLT3-ITD FSC HCL GFP GvL HIV HL HLA-DR HSL HSCT HTLV-1 ICOS Ig IL JAK KIR KrO LAIP LCA LGL LSC Lin LPD MAPK

Energy coupled dye (phycoerythrine-Texas red conjugate) Ethylene diamine tetraacetic acid Early T-cell precursor acute lymphoblastic leukaemia Flow cytometry Fluorescein isothiocyanate Fine-needle aspirate Follicular lymphoma Fluorescein-labelled proaerolysin FMS-like tyrosine kinase-3 inversion tandem duplication mutation Forward scatter Hairy cell leukaemia Green luorescent protein Grat versus leukaemia Human immunodeiciency virus Hodgkin lymphoma Human leukocyte antigen – antigen D related Hepatosplenic lymphoma Haematopoeitic stem cell transplantation Human T-cell lymphotropic virus-1 Inducible costimulatory Immunoglobulin Interleukin Janus kinase Killer immunoglobulin-like receptors Krome orange Leukemia associated immunophenotype Leucocyte common antigen Large granular lymphocyte Leukemic stem cell Lineage Lymphoproliferative disorder Mitogen activated protein kinase

xi

www.cambridge.org

Cambridge University Press 978-1-107-50383-0 — Multiparameter Flow Cytometry in the Diagnosis of Hematologic Malignancies Edited by Anna Porwit , Marie Christine Béné Frontmatter More Information

Abbreviations

Monoclonal B-cell lymphocytosis Mantle cell lymphoma Myelodysplastic syndrome Mycosis fungoïdes Mean luorescence intensity Major histocompatibility complex MLL/KMT2A Mixed lineage leukaemia/lysine methyl transferase 2A MRD Minimal residual disease m-TOR mammalian transporter of rapamycin MZL Marginal zone lymphoma MPAL Mixed phenotype acute leukemia MRD Minimal residual disease NF Nuclear factor-kappa B NGS Next generation sequencing NK Natural killer NHL Non-hodgkin lymphoma NPM Nucleophosmin PB Peripheral blood Pbl Paciic blue PC Plasma cell PCA Principal component analysis PCM Plasma cell myeloma MBL MCL MDS MF MFI MHC

PCR PD-1 Percp PE RPI PI3K PML-RARA

PMT RNA RBC SC SLL SS SSC STAT TCR TdT TK Tregs WBC WHO

Polymerase chain reaction Programmed death-1 Peridinin chlorophyll-A protein Phycoerythrin Propidium iodide Phosphoinositide 3 kinase Promyelocytic leukemia/ retinoic acid receptor A [t(15;17) translocation] Photomultiplier Ribonucleic acid Red blood cell Sézary cell Small lymphocytic lymphoma Sézary syndrome Side scatter Signal transducer and activator of transcription T-cell receptor Terminal deoxynucleotidyl transferase Tyrosine kinase Regulatory T-cells White blood cell World Health Organization

xii

© in this web service Cambridge University Press

www.cambridge.org

Chapter

1

Flow Cytometry in Clinical Haematopathology: Basic Principles and Data Analysis of Multiparameter Data Sets Francis Lacombe and Marie-Christine Béné

Introduction Flow cytometry (FCM), as indicated by its name, is a semi-automated method, which combines two basic approaches, cytometry and flow, respectively. Cytometry is by essence the measurement of cell characteristics. More broadly, it can be applied to various types of particles. In fact, the very first application of automated cytometry was invented by Wallace and Joseph Coulter with the first objective of counting paint particles [1]. Before that era, cell counts were performed manually under a microscope with specific calibrated slides called haemocytometers usually bearing the names of their inventors (Malassez, Thoma, Neubauer, Nageotte and others) [2]. Such devices also allow us to recognise some cell types based on their size and granularity, detectable without any staining, by simple phase contrast. Other properties of cells can be examined with an optical bright field microscope after preparing smears or cytospins where the cells are spread in a thin mono­ layer fixed on a slide. Such preparations are then stained, most frequently with May Grünwald Giemsa (MGG) or Wright stains [3]. These panoptic stains contain eosin and methylene blue, plus azure for MGG, and thus make acidic components appear blue and basic components appear orange-red to violet. Smears or cell suspensions can also be labelled with antibodies conjugated to fluorochromes and examined under UV light in specially equipped microscopes. The flow component of FCM is a liquid sheath that allows to convert a cell suspension in a narrow linear flux individualizing cells/particles. The flow part of the instrument performs what is called hydrofocusing, i.e. single cell alignment. Hydrodynamic focusing is achieved by injecting the cell suspension in the core of sheath fluid at a slightly higher pressure and at a point when the channel becomes smaller. The acceleration of the fluids through this narrow channel and the ­different speed of the sheath and the cell suspension result in cell alignment. Of note, there is no mixing of 02

the cells/particles suspension with the sheath liquid. The latter can therefore very well be pure water, its main characteristic being to be devoid of any particle. The flux of cells/particles is guided in a specific device called a flow cell, through which a source of light will illuminate each cell as it passes in front of it. The major advantages of FCM, compared to the methods briefly mentioned above, are that larger numbers of cells will be counted and that many parameters including immunological characterisation of cells will be examined. Moreover, all results will be electronically stored, remaining available for analysis at any time after data acquisition.

Cell Counting in FCM Haemocytometers allow us to count cells in a welldefined chamber of 0.1 mm3 using unmanipulated suspensions (i.e. cerebrospinal fluid) or diluted samples where red blood cells have been lysed. Data are then converted by calculation in the usual measurement units of events per mm3 or per litre. For stained cells, typically, between 100 and 500 cells are counted manually when performing cell differentials. Both these methods are prone to errors linked to the small number of events actually taken into account and thus lack precision. This has been well established by Rümke, who designed a table displaying the decreasing level of incertitude associated with larger numbers of events counted [4]. Flow cytometry, that examines several thousands of events in a few minutes, provides a high level of sensitivity and exactitude. The relative numbers (proportions) of each cell subsets acquired will therefore be accurate. However, for exact absolute counting, flow cytometers require the use of standardised bead suspensions with a known number of beads per microlitre mixed volume/volume with the cells/particles preparation. When this known number of beads has been recorded by the instrument, it can therefore be concluded that 1 μL of sample has been examined. This can 19:40:10

1

Chapter 1: Flow Cytometry in Clinical Haematopathology

then be extrapolated to the other cells recorded during the same time. Another possibility is to use calibrated volumetric systems built in the flow cell. Because of these large numbers, it also allows to identify minute populations, likely to be missed with smaller counts.

Photodiode No cell: no signal

Laser

FCM and Light

2

(a)

Besides counting particles, FCM allows to appreciate their physical, chemical or biological properties. The major physical properties of particles/cells exploited by FCM are their ability to diffract, reflect and refract the coherent monochromatic light of a laser beam. Different types of lasers are available. Flow cytometers were initially equipped with gas lasers using argon, krypton, helium-neon or helium-cadmium, some requiring a cooling system. Solid-state lasers are crystals [ruby, yttrium aluminium garnet (YAG)] or ions such as titanium or chromium. More recent instruments use laser diodes based on semi-conductors and similar to light-emitting diodes (LEDs) [5]. When the narrow and focused coherent light of the laser encounters a cell/particle (so-called event), its diffraction intensity is proportional to the size of the ‘event’. Flow cytometers are equipped with photodiodes collecting the light diffracted by the cell in the path of the laser beam, of forward scatter (FSC). The instruments are also equipped with a device (mask) blocking the laser light from the FSC photodiode when no particle crosses it. The beam is widened proportionally to the size of the cell as one enters the laser’s path and the FSC photodiode can collect and transform it in an electronic signal proportional to the size of the cell/ particle (Figure 1.1). Concomitantly, a second detector (photodiode or photomultiplier tube (PMT)) collects the light reflected by the surface of the cell/particle as well as by any surface inside it (i.e. organelles, vesicles, granules, etc.) at a defined angle, lateral to the path of the laser beam. The intensity of this side scatter (SSC) signal will thus be proportional to the granularity of the cell. Typically, in a blood sample, the small erythrocytes with no nuclei will provide very small signals while those generated by the larger granulocytes will be more intense. The pattern of scatter signals will differ slightly between instruments, depending on the angle of the SSC detector and the number of display channels (see below in signal acquisition). The voltage applied to the detector will also modify the intensity of light collected. It must be adapted to the 02

Photodiode Cell: light diffraction → signal

Laser (b)

Figure 1.1  Forward scatter measurement of cell size. (a) The laser light is blocked by the mask when no cell passes through the beam. (b) Each cell, by diffracting light, allows for a signal proportional to its size to be collected on the detector around the mask.

type of cells investigated, i.e. it will have to be higher to see small particles such as platelets and lower to see larger cells such as granulocytes. Chemical parameters can also be measured by flow cytometers, typically based on the properties of fluorochromes. The latter are chemical substances able to absorb light at a defined wavelength and re-emit it at a higher and defined wavelength. This is based on the fact that, in these molecules, absorption of a photon will result in a modification of electrons, moving from a ground state to an excited state. When electrons return to their ground state, they restore the energy by going through transition stages resulting in the emission of a quantum of light with a higher energy and thus higher wavelength than the emission light [6]. Basically, in FCM, lasers provide excitation light and fluorochromes emission light. To collect emitted light from each fluorochrome, flow cytometers are equipped with dichroic mirrors and bandpass filters before each signal is registered by a dedicated PMT. Dichroic mirrors reflect light at a specific wavelength while letting all other light pass through. They are positioned at an angle from the emission source so that reflected beams make a 90° angle to the mirror and get directed towards the relevant PMT. Just before PMTs, filters of a specific wavelength will narrow the beam of light collected, ideally at the level of peak fluorescence. 19:40:10

Chapter 1: Flow Cytometry in Clinical Haematopathology

Table 1.1  Selection of commonly used fluorochromes

Fluorochrome

Excitation wavelength

Laser

Emission wavelength

Pacific Blue

405

Violet

455

Brilliant Violet421™

405

Violet

421

BD HorizonV450™

405

Violet

455

Alexa Fluor405™

405

Violet

421

BD HorizonV500™

405

Violet

500

Krome Orange

405

Violet

528

Pacific Orange™

405

Violet

551

DAPI

405

Violet

461

R-Phycoerythrin (PE)

488

Blue

575

PE-Texas Red/ECD

488

Blue

620

PE-Cy5/PC5

488

Blue

670

PE-Cy5.5/PC5.5

488

Blue

692

PerCP-Cy5.5™

488

Blue

695

PE-Cy7/PC7

488

Blue

770

GFP

488

Blue

510

Fluorescein isothiocyanate (FITC)

488

Blue

520

Alexa Fluor 488™

488

Blue

520

PI

540

Blue

637

7-AAD

550

Blue

647

APC

643

Red

660

APC-Cy7

650

Red

667

Alexa Fluor 700™

650

Red

720

APC-Alexa Fluor 700™

643

Red

720

APC-Alexa Fluor 750™

650

Red

774

Cy = Cyanin, APC = Allophycocyanin, PerCP = Peridinin chlorophyll.

All of the above is valid for instruments with one laser. However, the development of multiparameter FCM has led to build flow cytometers with multiple lasers, thus broadening the possibilities of staining by using fluorochromes excitable at different wavelengths. Although the basic FSC and SSC parameters are usually measured on the 488 nm blue laser, separate pathways are then used to channel the emitted lights generated by the different lasers [7].

Fluorochromes Fluorochromes (Table 1.1), also called fluorophores, are aromatic polycyclic carbohydrates, which can be found naturally in some algae (i.e. phycoerythrin or rhodamine) or organic synthetised compounds 02

(i.e. fluorescein isothiocyanate) [8]. Some proteins are also fluorescent, such as the green fluorescent protein (GFP), which will stain cells transfected with its gene  [9]. Some fluorochromes are used individually, such as propidium iodide (PI), which will fluoresce in red once intercalated in the hydrophobic environment of ­deoxyribonucleic acid (DNA) and is widely used together with annexinV to study cell death and ­apoptosis [10], or thiazole orange, which will stain both DNA and RNA and is widely used for the analysis of reticulocytes and platelets [11, 12]. Nowadays, fluorochromes are mostly used conjugated to monoclonal antibodies, allowing visualisation of the structures specifically recognised by the latter. Most of such conjugates (monoclonal ­antibody/ fluorochrome combination) use single fluorochromes. 19:40:10

3

Chapter 1: Flow Cytometry in Clinical Haematopathology

0V

(b)

103

100

101

102

900,000 channels

000

(a)

102

90,000 channels

010 001

101

900 channels

2.5 V

100

15,000 channels

100 011

1,500 channels

101

5V

20 bits = 1,048,576 channels

14 bits = 16,384 channels

9,000 channels

110

150 channels

7.5 V

111

15 channels

10 V 8.75 V

103

(c)

Figure 1.2  Principle of digital voltage partitioning and resulting sensitivity of fluorescence expression. (a) Example of a 3 bit partitioning of the signal between 0 and 10 V. (b) Number of channels obtained by using 14 or 20 bits digitalisation.

4

In some combinations, the properties of two different compounds are used in ‘tandem’ fluorochromes. This allows detection of emitted light from a fluorochrome, which cannot be excited by the available laser. The principle of such tandems is to use the light emitted by the first molecule to excite the second fluorochrome tightly bound to it and then measure the second emitted fluorescence. The principle used is also called fluorescence resonance energy transfer (FRET) [13]. A general very important property of fluorochromes is their sensitivity to light. For this reason, fluorochrome conjugates are provided in light-proof dark vials. Great care must be taken while manipulating these reagents to avoid exposure to direct light at all stages of the experiment. Some fluorochromes are also sensitive to pH, yet stable in the neutral conditions associated with most biological applications of FCM [14]. Finally, tandems can be unstable and get degraded. In this case, the fluorescence collected with them will thus erroneously be that of the first fluorochrome [15]. Among the numerous fluorochromes available, the major properties to examine are the wavelengths of their excitation and emission lights (Table 1.1), assuring that they are adapted to the parameters of the flow cytometer to be used. It is also important to know their level of brightness to choose a conjugate best adapted to the level of expression of the molecule to be stained.

by analogic/digital converters (ADC), which generate binary signals of 0 or 1. Typically, no or a low signal is 0 and a higher signal is 1. The sensitivity of the signals collected can be improved by increasing the number of bits, which are the combinations of 0 and 1 that are used to partition the range of voltage of the signal. For instance, a signal between 0 and 10 V can be d ­ ichotomised as 0 between 0 and 5 V and 1 between 5 and 10 V in a 1 bit digitalisation. Three bits digitalisation (i.e. all permutations between 000 and 111) will divide this voltage range in eight levels (23). Typically, flow cytometers use either 14 or 20 bits combinations, which yield 16,384 or 1,048,576 channels. The second combination p ­ rovides a better discrimination for weak signals. Indeed, since the graphic representation of these signals generally uses logarithmic scales, on a four decades scale the first configuration will provide ~16 channels in the first ­decade and the second one ~900 channels (Figure 1.2). The difference will be minimal in the high voltages but will be visible in the low values. These technical considerations may impact the settings of the instruments. When lower numbers of bits are used, the dynamic range of the logarithmic scale can be limited and the choice of PMT values can impact the resolution of the signals at low intensities.

Signal Acquisition

A number of software packages have been developed to analyse the signals provided by flow cytometers, some being incorporated in the instrument and used during acquisition and others (or portable versions of the same) used to perform analyses at distance [17]. The simplest way to analyse FCM signals is a monoparametric histogram where the signal intensity is displayed in the abscissa and the number of events in the ordinate. Because most applications of FCM are

The signals sent to PMTs are very low and need to be amplified, hence the name of these collectors which increase the signal provided by collected photons. Amplification is performed by increasing the tension applied from 100 to 1,000 V [16]. Linear amplification can also be obtained by applying a gain value, usually for FSC and SSC. Each signal is transformed electronically 02

Signals Displays

19:40:10

Chapter 1: Flow Cytometry in Clinical Haematopathology

100

101

102

103

Fluorescence intensity Figure 1.3  Monoparametric representation of the eight Rainbow® beads of different fluorescence intensities. Note that the signal gets narrower in the high fluorescence channels. Abscissa: fluorescence intensity. Ordinate: particle count.

to biological systems, this usually results in Gaussian peaks, the breadth of which depends on the variability of expression of the marker investigated. For fluorescent beads used to check the proper alignment of the laser, this peak should be very narrow, especially for bright fluorescence, and reflects the sensitivity of the instruments at low fluorescence (see above in Signal Acquisition; Figure 1.3). In a properly aligned instrument, the coefficient of variation of each peak (standard error divided by the mean) should be as low as possible,

Neutrophils [Time] FS INT / SS INT

1000

800

800

600

600

SS INT

SS INT

1000

400

0

0

200

400 600 FS INT

Lymphocytes (a)

[A] CD64 ECD / SS INT

800

102

400

0

1000

[B] CD45 / CD19

103

101 100

200

200

CD45bright/CD19+ B-Lymphocytes

CD45dim/CD19+ Blasts

Activated neutrophils

CD19

Debris, Fat

independently of the apparent width of the peak which will logically appear larger at low fluorescence values. Biparametric histograms also called scattergrams or dot plots are used to display events based on two of their properties. They will thus appear as dots (each event resulting in a dot at the intersection between its X and Y values) forming one or more clusters depending on the populations present in the sample. The most basic biparametric histogram to check for proper selection of FSC and SSC settings (Figure 1.4a) uses these two parameters, most frequently on linear scales. As mentioned above, this is also used by other instruments such as cell counters and does not depend on fluorescence. Displaying each fluorescence against SSC (Figure 1.4b) allows for a good appreciation of autofluorescence versus specific signals in a given population. With such a display for the analysis of examined samples (such as peripheral blood (PB), bone marrow (BM) or lymph node cell suspension), unstained cells provide an excellent internal control. Biparametric histograms are also used to examine the relationship between two different fluorochromes (Figure 1.4c). The results can become extremely pertinent to small subsets if a succession of selection of relevant subsets is performed by drawing gates. Such gates can be coloured/painted and the use of a homogeneous colour code on a single platform greatly facilitates data interpretation.

100

Monocytes

101 102 CD64 ECD

103

100

Monocytes (b)

(c)

Other leukocytes

101 CD45

102

103

CD45bright/CD19– Lymphocytes

Figure 1.4  Flow cytometry displays: (a) biparametric plot of FSC and SSC, i.e. light diffraction without considering any fluorescence; ­leukocyte subsets are easily distinguishable and debris can be gated out from gate A; (b) biparametric plot of FL1 against SSC allowing for an easy distinction of positive subsets, here CD64+ monocytes (blue) and a subset of activated neutrophils (lighter blue), by comparison to unstained other subsets blood and (c) relationship between two fluorochromes, here CD45 and CD19 in a sample containing B-blasts, T and NK lymphocytes, B-lymphocytes and neutrophils.

02

19:40:10

5

Chapter 1: Flow Cytometry in Clinical Haematopathology

[Cells] CD45 KO / SS INT 1000

600

[Lymphos] CD3 APC / CD19 APC750 103 D-+ D++

400

102

200 0

(a)

100

101

102

103

CD19 APC750

SS INT

800

101 5 0

CD45 KO

D--

D+-

[CD3] CD8 / CD4

103 D-+

D++

2

–5

100

(b)

101

102

103

CD3 APC

10

101 D--

D+-

1 0 [Cells] CD45 KO / SS INT

[Cells] CD45 KO / SS INT 1000

800

800

600

600

SS INT

SS INT

1000

400 200 0

(d)

–1

(c)

0

2

101 CD8

102

103

400 200

100

101 102 CD45 KO

0

103

(e)

100

101 102 CD45 KO

103

Figure 1.5  (a) Backgating of coloured leukocyte subsets on a CD45/SSC ‘cartography’ of BM; (b) gating on lymphocytes allows to distinguish CD19+ and CD3+ lymphocytes; (c) further gating on the CD3+ population allows to display T-cell subsets CD4 and CD8; (d) density plot representation of the same plot as in (a) and (e) contour representation of the same plot as in (a).

6

In the example provided in Figure 1.5, the specific staining of leukocytes by a monoclonal antibody directed to CD45 is displayed (see also Chapter 3). Gating strategies have been used in this example to colour granulocytes in red, monocytes in green and lymphocytes in magenta. These subsets have then been backgated on the CD45/SSC scattergram. Selecting only the lymphocyte gate, and a biparametric histogram based on CD3 and CD19 expression, the major lymphocyte subsets are displayed. Gating on CD3+ lymphocytes and displaying CD4 and CD8 allow for analysis and visualisation of the main T-cell subsets. The gating hierarchy must be kept in mind when performing complex analyses. Most software allows for a visualisation of this hierarchy and labels histograms based on the parameters used. Although software will basically provide letters to identify gates, it is advised 02

to design and save protocols where subsets are more precisely named. Counts and percentages (called gate statistics), based on the chosen reference population, can be obtained by various means. For monoparametric histograms, integration cursors can be placed encompassing the Gaussian peak of interest. For bi- or multiparameter histograms, gates of different shapes (polygons, squares, rectangles, circles, freehand. . .) will provide their X and Y coordinates as well as the number and percentage of events they contain. For  wellseparated subsets, quadrants can be used, dividing each histogram into four regions the size and shape of which can be adapted to best delineate the various subsets visualised (Figure 1.5b and c). It is important to consider populations and subpopulations as Gaussian clusters and not to divide 19:40:10

Chapter 1: Flow Cytometry in Clinical Haematopathology

them either by too short integration cursors of quadrants bisecting clusters. A frequent mistake is to rely too strictly on controls such as irrelevant isotype ­antibodies stained with the same fluorochrome and consider the brightest signal provided by the Gaussian of such a negative control as the beginning of significant fluorescence. Although this may be the case for brightly stained subsets, often the peak of the population of interest presents with a shift overlapping the ‘negative’ peak [18]. In the examples of biparametric histograms shown so far, individual dots are only seen on the periphery of coloured clusters. A better idea of the density of events in such clusters can be provided by monoparametric histograms but also by variations of biparametric representations such as density or contour plots (Figure 1.5d and e). This can prove very useful to properly delineate cell subsets. The number of cells displayed can also be chosen or the resolution of the plots modified. Some software packages also provide various approaches of multiparameter representations in spaces with more than one dimension. This can be modelled by mathematical calculation of principal component analysis or drawn by moving the length and angle of the various vectors to best individualise populations of interest (see also Chapter 14). Finally, for samples assayed with the same antibody combinations, the merge function of some software allows for a direct comparison of different conditions (i.e. diagnosis and relapse) or for the concomitant analysis of larger numbers of events. For example, merging six different samples of normal BM will provide a good idea of what normal is, by smoothing out individual variations (Figure 1.6). An important parameter for such displays is the time of acquisition, which must be systematically recorded. This allows, during merge analysis, to always individualize each sample.

Compensation Emission spectra of fluorochromes appear as semiGaussian curves with a maximal peak that dictates the choice of the bandpass filters placed before the PMTs (Figure 1.7a). Yet, there is frequent overlap of these emission spectra and it may be necessary to eliminate the contaminating signal [19]. This is achieved mathematically by subtracting a percentage of the overlapping signal(s) in each fluorescence channel (Figure 1.7b). 02

Basically, compensations will be calculated by performing single staining of cells or beads with each of the antibodies intended to be used in a panel. Each preparation will then be examined in each possible emission fluorescence channel, and the compensation percentage will be adjusted to remove overlapping signals. It is important to have in the preparations a negative control which will provide a reference baseline signal. Currently, beads coated with antibodies to mouse immunoglobulins allow compensation of settings for any antibody/fluorochrome combination  [18]. If different conjugates are obtained from the same manufacturer, using the same fluorochromes, a compensation matrix adequate for different panels can be used. The beads stained with the antibodies are mixed with uncoated beads, which will provide the baseline signal. Each biparametric histogram combining the fluorescence tested versus each other channel will be ­examined to provide for the absent fluorochromes the same signal as that of the negative beads. It is also possible to perform or check for proper compensation by using the ‘fluorescence minus one’ or FMO method. This technique consists of testing a relevant sample in a series of tubes where, for each, one of the markers intended to be used in the panel is missing. This allows us to visualise/check the overspill or spreading of other emitted fluorescences in the fluorescence of interest where it is known that no antibody has been added [20]. When checking for proper compensation, or building a compensation matrix with cells, it is wise to use a biexponential or logical representation of low signals since some staining will lead to an unavoidable spreading or trumpet effect (Figure 1.7c). This means that light from labelled cells will spill over in a different channel with a broader yet symmetrical display, which is also called signal distortion or spreading [21–23]. Compensations could initially be performed ‘manually’ with specific tools provided in the software of the instrument or in external software. Nowadays, with the development of multiparameter FCM, the interferences are too complex and most software provides a ‘wizard tool’ that automatically performs compensations using the acquisition file of each single labelling.

FCM Settings For any given experiment, the parameters of the instruments must be defined beforehand. The PMTs must be adjusted in such a way that the signals recorded are well defined yet do not saturate in the brightest decades. 19:40:10

7

Chapter 1: Flow Cytometry in Clinical Haematopathology [B] CD45-KO / SS INT

800

800

800

600

600

600

600

WBC

SS INT

800

SS INT

1000

400

400

400

400

200

200

200

200

0

200

400

600

800

0

1000

100

101

102

0

103

100

CD45-KO

FS INT [B] CD13-ECD / SS INT

101

102

0

103

[B] CD34-PC7 / SS INT

[B] CD33-PC5.5 / SS INT

800

800

800

600

600

600

600

0

100

101

102

0

103

CD13-ECD [B] CD7-A700 / SS INT

400

200

100

101

102

0

103

CD33-PC5.5 [B] CD11b-APC-A750 / SS INT

200

100

101

102

0

103

1000

800

800

800

800

600

600

600

600

FS INT

1000

SS INT

19:40:10

1000

400

400

400

400

200

200

200

200

100

101

CD7-A700

102

103

0

100

101

102

CD11b-APC-A750

100

103

0

100

101

CD16-PB

102

101

102

103

CD117-APC [B] TIME / FS INT

1000

0

103

400

CD34-PC7 [B] CD16-PB / SS INT

SS INT

02

200

SS INT

800

SS INT

1000

200

102

[B] CD117-APC / SS INT

1000

400

101

CD14-PE

1000

400

100

CD65-FITC

1000

SS INT

SS INT

[B] CD14-PE / SS INT

1000

0

SS INT

[B] CD65-FITC / SS INT

1000

SS INT

SS INT

[Ungated] FS INT / SS INT 1000

103

0 0

200

400

600

800

1000

TIME

Figure 1.6  Example of the merge of six PB samples. The bottom right histogram shows the merged samples according to a ‘time’ abscissa that allows to discriminate each sample as a single column. All cells in this histogram (i.e. merged samples) have been used then to examine the florescence of each of the markers in this common tube using SSC as ordinate. The top left histogram allows for eliminating debris on a FSC/SSC display. The second top histogram is the ‘cartography’ of CD45/SSC display with backgating of neutrophils in red, monocytes in green and lymphocytes in magenta. The third histogram shows the strong positivity of neutrophils for CD65 and intermediate labelling of monocytes. The fourth histogram shows the strong positivity of monocytes for CD14. On the second row, histograms show CD13-positive monocytes and neutrophils, strongly CD33-positive monocytes with lower staining of neutrophils and, in the last two histograms, absence of immature myeloid cells expressing CD34 or CD117, respectively. The three bottom left histograms show CD7-positive lymphocytes, CD11b-positive monocytes and neutrophils and CD16bright neutrophils and finally CD16intermediate NK-lymphocytes.

8

Chapter 1: Flow Cytometry in Clinical Haematopathology

Laser : 488 nm FITC

PE

ECD

PC5

FL1

FL2

FL3

FL4

500

600

700

(a) 103

FITC

102 101 500

600

700

800

FITC 525 nm 100%

(b)

FITC 575 nm 12%

0

FITC 620 nm 12%

–2

FITC 675 nm 0.1%

50%

2

50% 100

(c)

101

102

103

Figure 1.7  (a) The spectral overlap of four commonly used fluorochromes, ordered by emission wavelength (abscissa and spectrum below); (b) the strategy for compensation calculations, indicating the percentage of overspill of FITC in higher wavelengths and (c) ­distortion or spreading or ‘trumpet effect’: compensations are perfect with 50% of the spread on both sides of the negative signal for the ­fluorescence on the ordinate.

A good way of defining this is to use unstained lysed blood and adjust all PMTs so that over 85% of the cells can be seen in the first decade, i.e. above the first channel. Such adjustments can then be used to identify the fluorescence channels where a specific batch of beads fluorescing in all channels will appear. This strategy, adopted in the Harmonemia initiative, allows for excellent reproducibility between instruments, new channels being recorded at each change of beads batch [24]. Lasers’ alignment with the flow cell must be checked regularly, although current instruments are usually extremely stable. This is performed by recording the fluorescence emitted by specific beads, with target limits of coefficients of variation of the peaks obtained [25]. Such controls belong to the daily assessment of laser alignment and result in the generation of traceable data 02

in the form, most frequently, of Levy-Jennings graphs. The mandatory pre-analytical precautions to ensure that proper data are generated have been extensively described by an expert working group [25]. The sheath fluid is usually an isotonic solution, but since there is no contact between this liquid channel and the cell suspension it guides, water can also be used. The mandatory condition is that this fluid is completely devoid of any particle that could generate a light signal. Depending on the number of cells that have to be analysed and the precision of the detection, the pressure of the sample fluid can be adjusted between low for the more precise measurements and intermediate or high for less sensitive signals. The tubing of the instrument must be kept extremely clean, and regular cleansing cycles with chlorine and several tubes of distilled water are recommended 19:40:10

9

Chapter 1: Flow Cytometry in Clinical Haematopathology

on top of the automatic cleanses performed by most machines between samples.

Panel Design

10

Multiparameter FCM presents the great advantage of being able to test numerous characteristics of a suspension at the same time. Besides FSC and SSC, 8 and 10 colours flow cytometers now very frequently provide up to 10 or 12 parameters. This has proven to be extremely valuable in immunophenotyping for the diagnosis of haematological malignancies especially when only small samples such as fine-needle aspirates (FNA) or cerebrospinal fluid are available. Two types of considerations can be taken into account when designing a panel [23]. One is to use a variety of markers allowing characterisation of a maximum of different subsets, such as markers for granulocytes, monocytes, lymphocytes and progenitors. The other option is to investigate the co-expression of several markers on the same cells. This is applied if the goal is to define the maturation stage or activation status of a given subset. In that case, it is important to choose wisely the fluorochromes associated with the various monoclonal antibodies used. In all cases, bright fluorochromes must be preferred for antigens expressed at low levels, and dimmer fluorochromes for densely expressed antigens. When co-expression is expected, the overlap of fluorochromes must be taken into account to avoid the necessity of too much (or impossible) compensation. Several rules apply for this case. The first channel of the blue laser (488 nm excitation), generally used for FITC conjugates, is never impacted by other fluorochromes. The same is true for the first channel of the violet laser (405 nm excitation) used for Pacific Blue® or Brilliant Violet 421®. These two channels can be used for weakly expressed antigens, since no compensation needs to be applied. The second channel of the violet laser conversely never overlaps other channels. It can thus be used for bright markers and/or ‘parent’ markers that will contain subsets. This is for instance the case for the pan-leukocyte CD45. The degree of spillover will then vary between fluorochrome combinations which should also guide the choice of combinations. ‘Parent’ and ‘children’ relationships between markers characterise the fact that all ‘children’ are also stained by the ‘parent’ marker. Spillover can then be accepted from a ‘child’ to the ‘parent’, since the primary staining will prevail. Conversely, ‘parents’ should have no or low 02

spray on ‘children’ to avoid erroneous appreciation or fuzzy images [26]. It is also important to know that coexpressed markers can lead to dot plots with an angled shape suggesting poor compensation although it is just due to the dual staining. Compensations are usually easily done for lights emitted after excitation with blue and violet lasers. They are more complicated for excitation with the red laser and mutually exclusive markers can be preferred for these channels, where spillover will not interfere.

Sample Handling Flow cytometry by definition requires cells in suspension, implying that PB or BM are collected on anticoagulated tubes [27]. Depending on local habits and on other tests liable to be performed on the same sample, ethylene diamine tetraacetic acid (EDTA) or heparin can be used. It is also recommended to perform FCM analyses as rapidly as possible after collection. For fragile samples such as cerebrospinal fluid, it may be interesting to add a specific preservative solution to collect more accurate information and avoid cell loss in the inhospitable low-protein content of this liquid [28]. Because staining and acquisition are relatively rapid, this allows rapid answers to the clinicians, nearly at the same time as morphologic analyses. When transport is needed, the time to processing for samples other than cerebrospinal fluid should not exceed 72 h [27]. Of note, the characterisation of leukaemic cells in a heavily infiltrated sample usually will not be modified by some delay possibly leading to some apoptosis. Conversely, the search for minute subsets in the context of minimal residual disease can be impacted by extended delays. Also, some tumour cells such as high-grade B-cell lymphoma or plasma cells in myeloma may be more prone to apoptosis. Another pre-analytical aspect of possible impact is the haemodilution of BM samples [29]. It is important to remember that this will not impact the characterisation of malignant cells which can be isolated by gating strategies. However, the great quality of FCM to be able to provide exact counts is lost in haemodiluted samples. It therefore cannot be used to make a diagnosis of leukaemia or myelodysplasia where blast percentages impact the result. However, this is well put to use to enumerate mobilised CD34+ progenitors in PB before stem-cell transplantation [30]. The choice of the panel of antibodies to test will depend on clinical information provided when sending the sample. Screening or disease-specific panels 19:40:10

Chapter 1: Flow Cytometry in Clinical Haematopathology

(described in detail in other chapters) are used. Those applied on a daily basis can be prepared in advance (i.e. weekly on Monday) but have to be carefully stored in the dark and validated for use on several consecutive days. It is also important that each antibody is titrated to use the most adequate concentration for saturation yet limiting non-specific labelling (see below, next paragraph). This is also true when antibody cocktails are prepared extemporaneously. Great care must be taken in these preparations in terms of exposure to light, meticulous pipetting (changing pipetting cones between each antibody to avoid contamination) and strict control of the sequential addition of antibodies, not to forget or duplicate any given marker. Interesting alternatives have appeared which provide customised or generic ready-to-use antibody combinations [31, 32]. The latter can be premixed cocktails allowing for a single pipetting step, or individual FCM tubes containing lyophilised or dried antibodies. Thorough vortexing of the sample and the antibody mixture must be performed before incubation in the dark for 10–15 min at room temperature (+20°C). Incubation at +4°C avoids the undesirable patching and capping of the antibodies that may lead to erroneous labelling negativity, but requires a 15–20 min incubation. For most applications, whole blood or BM is the sample of choice. They can be used directly by mixing an aliquot of 50 to 100 μL with the antibody mixture. Various lysis reagents can be applied after the incubation time. Home-made ammonium chloride (NH4Cl) has long been used but there are many different versions of this lysis that may not lead to highly reproducible data. Very stable lysis solutions are available commercially, some of them being highly selective in interfering only with the physiology of red cells (Table 1.2). After lysis, the sample can be directly processed for acquisition. Such lysis-no wash methods allow us to examine the entire cell population of the sample. To minimise non-specific staining, it is recommended to use smaller amounts of antibodies, determined after titrating the reagents as mentioned above. Otherwise, a washing step may follow the lysis at the risk of losing some cells [25]. Only when investigating for surface immunoglobulins, a pre-incubation wash is mandatory to eliminate plasma antibodies [25]. Washing steps are also necessary when permeabilisation is required for the investigation of intracytoplasmic or nuclear antigens. Finally, if a search for a small population is required, bulk lysis of a larger sample can be performed prior to incubation, followed 02

Table 1.2  Commercially available major lysing reagents

Manufacturer

Name

Comments

Abcam

Flow Cytometry Lysing Solution

With fixative. Wash or no wash

BD Biosciences

BD FACS Lysing Solution

Proprietary hypotonic buffer, wash or no wash

BD Biosciences

BD Pharm Lyse™ Lysing Buffer

Ammonium chloride

Beckman Coulter

OptiLyse®

No wash

Beckman Coulter

Versalyse™

Wash or no wash, red blood cells-specific

Miltenyi Biotec

Red Blood Cell Lysis Solution

No wash

Thermo Fischer

Cal-Lyse™

With fixative

Thermo Fischer

High-Yield Lyse

Fixative-free, no wash

by centrifugation to increase the cell concentration. It must however be remembered that each washing step induces random cell losses and can change the proportions of cell subsets. Some lysing preparations may contain fixatives that will stabilise the staining by modifying the plasti­ city of plasmatic membranes. This is interesting when performing large batches of staining in that it allows a delayed acquisition. It is however mandatory, even after fixation, to prevent exposure of the samples to light since the fluorochromes retain their sensitivity.

Post-Analytical Procedures As mentioned above, FCM experiments generate acquisition files which can be analysed or re-analysed at any time. It is good practice to establish analysis matrices (protocols), which will easily provide all the information needed in any specific condition in a reproducible fashion. As also mentioned, the use of systematic colour codes makes interpretation easier. For immunophenotypic patterns, especially in leukaemia, the pathological clone must be identified at best on scatter properties and CD45 expression (see also Chapter 5). This will usually be enough for acute leukaemia or myelodysplasia where the population involved is that of progenitors. The latter can be defined using a Boolean equation excluding from the whole sample the mature granulocytes, monocytes and lymphocytes. This can be helped by a combination 19:40:10

11

Chapter 1: Flow Cytometry in Clinical Haematopathology

of CD11a, CD16 and CD14 that will stain the first two subsets. Lymphocytes are characterised by their low SSC and bright CD45 expression, and are usually easy to gate, especially when using a density plot display [33]. For lymphoproliferative disorders of B-lineage, a CD19 gate can be useful, better established on the whole ­sample not to miss proliferations with a high SSC such as hairy cells or diffuse large B-cell lymphomas. A lymphocyte gate will be preferable for T-lineage proliferations since not all express surface CD3. Once the malignant population is isolated and enumerated, percentages should not be used afterwards to describe the immunophenotype, except for clear partial expressions. The report should mention the type of sample ­analysed, the specificities tested and those that yielded a relevant positive or negative signal contributing to the interpretation, together with a conclusion on the likely diagnosis. Some recommendations can be found in the literature regarding suggested report formats [34, 35]. In any case, an integrated report, at least combining morphology and FCM and further incorporating cytogenetics and molecular findings when available, is recommended.

Conclusion

12

3.

6. 7. 8.

9. 10.

11. 12.

14.

15. 16.

17.

18.

References 2.

5.

13.

Many of the principles and technologies described in this chapter are embedded in the instruments and monitored without necessarily knowing what happens technically. However, a good knowledge of them can prove useful for troubleshooting when unexpected data are obtained. The most sensitive part is probably the stability of fluorochromes and great care must be taken when manipulating labelled reagents to preserve their quality. At the end of the day, what will in fact matter for haematological malignancies is the result obtained and provided to the clinicians in charge of the patient’s management. Flow cytometry results will also be important to guide further analyses as stated above. Production of an integrated conclusive report following the WHO classification is certainly an aim to target.

1.

4.

E. Simson. Wallace Coulter’s life and his impact on the world. Int J Lab Hematol; 35 (2013):230–6. L.M. Sandhaus. Is the hemocytometer obsolete for body fluid cell counting? Am J Clin Pathol; 145 (2016):294–5. B. Houwen. Blood film preparation and staining procedures. Clin Lab Med; 22 (2002):1–14.

19. 20.

02

C.L. Rümke, P.D. Bezemer and D.J. Kuik. Normal values and least significant differences for differential leukocyte counts. J Chronic Dis; 28 (1975):661–8. W.G. Telford. Lasers in flow cytometry. Methods Cell Biol; 102 (2011):375–409. U. Noomnarm and R.M. Clegg. Fluorescence lifetimes: fundamentals and interpretations. Photosynth Res; 101 (2009):181–94. S.P. Perfetto, D. Ambrozak, R. Nguyen, et al. Quality assurance for polychromatic flow cytometry. Nat Protoc; 1 (2006):1522–30. P.K. Chattopadhyay, C.M. Hogerkorp and M. Roederer. A chromatic explosion: the development and future of multiparameter flow cytometry. Immunology; 125 (2008):441–9. P.J. Cranfill, B.R. Sell, M.A. Baird, et al. Quantitative assessment of fluorescent proteins. Nat Methods; 13 (2016):557–62. N. Atale, S. Gupta, U.C. Yadav and V. Rani. Celldeath assessment by fluorescent and nonfluorescent cytosolic and nuclear staining techniques. J Microsc; 255 (2014):7–19. J. Nygren, N. Svanvik and M. Kubista. The interactions between the fluorescent dye thiazole orange and DNA. Biopolymers; 46 (1998):39–51. S. Rapi, A. Ermini, L. Bartolini, et al. Reticulocytes and reticulated platelets: simultaneous measurement in whole blood by flow cytometry. Clin Chem Lab Med; 36 (1998):211–14. J. Szöllosi, S. Damjanovich and L. Mátyus. Application of fluorescence resonance energy transfer in the clinical laboratory: routine and research. Cytometry; 34 (1998):159–79. J.K. Sugden. Photochemistry of dyes and fluorochromes used in biology and medicine: some physicochemical background and current applications. Biotech Histochem; 79 (2004):71–90. R. Hulspas, D. Dombkowski, F. Preffer, et al. Flow cytometry and the stability of phycoerythrin–tandem dye conjugates. Cytometry A; 75 (2009):966–72. M.P. Bristow, D.H. Bundy and A.G. Wright. Signal linearity, gain stability, and gating in photomultipliers: application to differential absorption lidars. Appl Opt; 34 (1995):4437–52. K.R. Chi; Let the data flow. The Scientist (2010). www.the-scientist.com/?articles.view/articleNo/29066/ title/Let-the-Data-Flow/ (last accessed 21 January 2017). H.T. Maecker and J. Trotter. Flow cytometry controls, instrument setup, and the determination of positivity. Cytometry A; 69 (2006):1037–42. J.W. Tung, D.R. Parks, W.A. Moore, et al. New approaches to fluorescence compensation and visualization of FACS data. Clin Immunol; 110 (2004):277–83. J.W. Tung, K. Heydari, R. Tirouvanziam, et al. Modern flow cytometry: a practical approach. Clin Lab Med; 27 (2007):453–68. 19:40:10

Chapter 1: Flow Cytometry in Clinical Haematopathology

21. M. Roederer. Spectral compensation for flow cytometry: visualization artifacts, limitations, and caveats. Cytometry; 45 (2001):194–205. 22. D. Novo, G. Grégori and B. Rajwa. Generalized unmixing model for multispectral flow cytometry utilizing nonsquare compensation matrices. Cytometry A; 83 (2013):508–20. 23. R. Nguyen, S. Perfetto, Y.D. Mahnke, et al. Quantifying spillover spreading for comparing instrument performance and aiding in multicolor panel design. Cytometry A; 83 (2013):306–15. 24. F. Lacombe, E. Bernal, D. Bloxham, et al. Harmonemia: a universal strategy for flow cytometry immunophenotyping – A European LeukemiaNet WP10 study. Leukemia; 30 (2016):1769–72. 25. S. Tanqri, H. Vall, D. Kaplan, et al. ICSH/ICCS Working Group. Validation of cell-based fluorescence assays: practice guidelines from the ICSH and ICCS – part III – analytical issues. Cytometry B Clin Cytom; 84 (2013):291–308. 26. E. Lugli, M. Roederer and A. Cossarizza. Data analysis in flow cytometry: the future just started. Cytometry A; 77 (2010):705–13. 27. B.H. Davis, A. Dasgupta, S. Kussick, et al. ICSH/ICCS Working Group. Validation of cell-based fluorescence assays: practice guidelines from the ICSH and ICCS – part II – preanalytical issues. Cytometry B Clin Cytom; 84 (2013):286–90. 28. A.H. de Jongste, J. Kraan, P.D. van den Broek, et al. Use of TransFix™ cerebrospinal fluid storage tubes prevents cellular loss and enhances flow cytometric detection of malignant hematological cells after 18 hours of storage. Cytometry B Clin Cytom; 86 (2014):272–9.

29. M.R. Loken, S.C. Chu, W. Fritschle, et al. Normalization of bone marrow aspirates for hemodilution in flow cytometric analyses. Cytometry B Clin Cytom; 76 (2009):27–36. 30. C. Schmidt-Lucke, S. Fichtlscherer, A. Aicher, et al. Quantification of circulating endothelial progenitor cells using the modified ISHAGE protocol. PLoS One; 5 (2010):e13790. 31. B.D. Hedley, M. Keeney, J. Popma and I. Chin-Yee. Novel lymphocyte screening tube using dried monoclonal antibody reagents. Cytometry B Clin Cytom; 88 (2015):361–70. 32. R.C. Chan, J.S. Kotner, C.M. Chuang and A. Gaur. Stabilization of pre-optimized multicolor antibody cocktails for flow cytometry applications. Cytometry B Clin Cytom; (22 March 2016). 33. C. Arnoulet, M.C. Béné, F. Durrieu, et al. Four- and five-color flow cytometry analysis of leukocyte differentiation pathways in normal bone marrow: a reference document based on a systematic approach by the GTLLF and GEIL. Cytometry B Clin Cytom; 78 (2010):4–10. 34. L. Del Vecchio, B. Brando, F. Lanza, et al. Italian Society for Cytometry. Recommended reporting format for flow cytometry diagnosis of acute leukemia. Haematologica; 89 (2004):594–8. 35. O. Hrušák, G. Basso, R. Ratei, et al. AIEOP-BFM Flow Network. Flow diagnostics essential code: a simple and brief format for the summary of leukemia phenotyping. Cytometry B Clin Cytom; 86 (2014): 288–91.

13

02

19:40:10

Chapter

2

Antigens Marie-Christine Béné and Anna Porwit

Introduction

14

For a long time, the immunophenotype of leukaemic blast cells has remained unknown. The first indication that lymphoblastic cells could be of the B lineage came when ultraviolet (UV) light microscopic immunofluorescence identified surface immunoglobulins (sIgs) on the very small subset of mature B-cell acute lymphoblastic leukaemia (ALL) as it did on normal B-cells [1]. At about the same time, mixing lymphocytes with sheep red blood cells induced the formation of ‘rosettes’ of erythrocytes, attached on the membrane of T-cells [2], and T-lineage ALL behaved similarly. The first breakthrough came when polyclonal antibodies were produced towards the nuclear ­antigen terminal deoxynucleotidyl transferase (TdT) [3] and towards a protein that was called CALLA for Common Acute Lymphoblastic Leukaemia Antigen [4]. TdT could be detected in blast nuclei on blood or bone marrow (BM) smears by immunofluorescence after fixation/permeabilisation by methanol. CALLA expression could also be investigated on the surface of cells in suspension, in indirect immunofluorescence and by UV light microscopy examination of a drop of suspension between a glass slide and coverslip. Then the revolution of monoclonal antibodies changed everything. The technology was initially developed by George Köhler to obtain homogeneous antibodies directed to sheep red blood cells [5] in Cesar Milstein’s laboratory, where research was extensively conducted in order to understand whether the ­specificity of immunoglobulins was acquired by gene mutations. A few years later, Susumu Tonegawa [6] discovered that the rearrangement of genes from several repertoires was in fact responsible for the generation of this extensive diversity. Meanwhile, at the beginning of the 1980’s, increasing numbers of laboratories began to implement the technology of monoclonal antibodies generation. The fusion of a B-lymphocyte, obtained from the spleen of an immunised animal (usually a mouse), with a cell from a non-secreting 03

myeloma cell line allows to produce an immortalised cell resulting in a proliferating clone of antibodysecreting cells. Such clones synthetise large amounts of identical immunoglobulins hence called ‘monoclonal’ antibodies. By immunising mice with human leucocytes, many unknown structures were discovered on these white blood cells. To organise this explosion of novelties, a series of international workshops on Human Leucocyte Differentiation Antigens or HLDA was initiated [7]. The rules were that a new molecule could be named when it had been identified by at least two different clones produced in two different laboratories and the molecule’s molecular mass was characterised, usually by western blot. Later, identification of the encoding gene was also required. After gathering information from many laboratories, statistical methods were used to group antibodies with the same specificities, with a clustering program. Such groups were called ‘clusters of differentiation’ or CDs. Molecules so far unknown were thus given a number and called CD‘n’. Many of them, for example, CD4, still have only that name. With 10 HLDA workshops and 371 CDs characterised, the exercise has slowed down [8, 9]. Yet it is estimated that around 7,000 proteins can be encoded by the human genome [10]. Thus, many of them have not yet been characterised as CDs and some remain unknown. The CD saga had a strong impact on the definition of haematological malignancies, bringing to laboratories an impressive array of tools allowing for the characterisation of each patient’s malignant cells. Although the dream of discovering differentiation antigens with highly selective lineage specificity was not fulfilled, many of the CDs display preferential expression on given cell types. Short characteristics of various antigens commonly used in immunophenotyping haematological malignancies are presented below. Differential expression of these antigens during cell maturation is described in more detail in Chapter 3. 19:42:15

Chapter 2: Antigens

CD43 CD45

CD10

COOH

E

NH2

CD38 COOH

E

P

NH2

P

Figure 2.1  Ubiquitous antigens. CD10 is a broadly expressed ectopeptidase present on normal mature neutrophils and on lymphocytes of the B-lineage, early during their maturation (haematogones) and later in secondary lymphoid organs’ germinal centres. It is also an important marker for the diagnosis of acute lymphoblastic leukaemia. CD38 is also an enzyme (ribosyltransferase) important to discriminate early progenitors, highly expressed on early precursors during B-cell maturation and later a marker of plasma cells. It is also a marker of natural killer-cells, activated T-cells and monocytes. CD43 also known as sialophorin or leucosialin is an ubiquitous antigen investigated to differentiate lymphoproliferative syndromes. CD45 is a key pan-leucocyte tyrosine phosphatase with different levels of expression on lymphocytes, monocytes, neutrophils and ­progenitors. It has become a mandatory antigen to delineate these cell subsets in peripheral blood, bone marrow or other fluids. Of note, CD45 is rapidly lost during erythroid maturation.

The antigen nowadays investigated in all immunophenotyping panels is CD451 (Figure 2.1). CD45 [11] is also known as the leucocyte common antigen. It is the prototype of receptors of the protein tyrosine phosphatases (PTP) subfamily, involved in the regulation of tyrosine kinases activity, hence its other acronym of PTPRC5 (PTP receptor type C). The large intracytoplasmic domain of the molecule has a phosphatase enzymatic function removing an inhibitory phosphate residue from activating tyrosine kinases. CD45 is expressed by all nucleated haematopoietic cells except erythropoietic precursors and some plasma cells. It has become an indispensable marker to identify leucocytes and their subsets since its level of expression varies according to cell types. A classical representation of leucocytes in flow cytometry uses a bi-parametric histogram combining side scatter CD45 is encoded by a long gene comprising 34 exons. Three of them are responsible for the generation of eight different isoforms, respectively CD45RA, CD45RB, CD45RC, CD45RAB, CD45RAC, CD45RBC, CD45R(ABC) and CD45RO; the latter results from different splicing retaining only some of these three exons.

1 

03

and CD45 expression (see Chapter 3). The multiple roles of this molecule, physiologically, involve the regulation of lymphocyte activation, both in the T and B lineages. CD45 isoforms, resulting from alternative splicing, can be identified by specific antibodies targeting unique epitopes of these molecules. CD45 R0 and CD45RA are the most frequently explored in immunology and haematology. CD45RA defines naïve and activated CD4 and CD8 T-cells. CD45RO appears after a T-lymphocyte has encountered an antigen and is one of the markers of effector-memory cells. Both can be expressed on some T-lineage lymphomas (see Chapter 9).

Immature Cell Antigens

CD34: Although the precise immunophenotype of the pluripotent haematopoietic stem cell (HSC) is still a matter of debate, it became obvious in the early 1990s that CD34 was a good marker to identify HSC (Figure 2.2). This led to the now widespread use of counting CD34+ cells in BM, mobilised peripheral blood or cord blood intended to perform autologous or allogeneic stem cell transplantation [12]. CD34 is a heavily glycosylated 19:42:15

15

Chapter 2: Antigens

CD34

CD117

CD164

CD38 COOH

CD133

E

NH2

K

Figure 2.2  Immature cell differentiation antigens. CD34, a heavily glycosylated sialomucin, is the hallmark of stem cells, notably used to appreciate their number in haematopoietic stem cell grafts. CD38, already presented in Figure 2.1, is also expressed early on stem cells, yet the most immature compartment seems to be the CD34+CD38− compartment of bone marrow cells. CD117, receptor of the stem cell factor, is a tyrosine kinase expressed on early progenitors. CD133 is an integral transmembrane protein with five membrane spans present on stem cells of many lineages and associated to cancer metastasis. CD164 is a sialomucin close to CD34 and also expressed on progenitors.

16

molecule with a protein core, containing three different epitopes dubbed I, II and III [13]. The latter were determined by assessing the sensitivity of CD34 to enzymatic digestion. Class I epitopes are sensitive to cleavage by neuraminidase, chymopapain and a glycoprotease from Pasteurella haemolytica, while class II epitopes are resistant to neuraminidase and class III resist cleavage by all three enzymes. Specific monoclonal antibodies can selectively recognise each of these isoforms but the most widely used are directed to class III. CD164 [14–16] CD 164 is a soluble and transmembrane sialomucin with a structure close to that of CD34, also known as endolyn or MUC24 (for MUltiglycosylated Core protein 24). It is expressed on stem and progenitor cells, together with CD34 and may act as ligand to selectins through its first mucin domain. Such interactions are involved in cell adhesion and proliferation. At later stages of differentiation, CD164 has also been reported on the surface of Sézary syndrome malignant T-cells. It also has been shown to be expressed by basophils upon allergen recognition. CD133 [17] has been characterised as a pentaspan transmembrane protein expressed on HSCs and mouse neuroepithelial cells. It stands as a good candidate for stem cell marker of many lineages. CD133 is also 03

expressed on a large type of cancer cells, and CD133+ cells isolated from tumours are capable of developing in long-term spheroids or growing in xenografts models. CD133 is thus also suspected to be involved in cancer metastasis. In malignant haematological disorders, the role and value of CD133 as a biomarker or potential tumour target is still ill-defined. HLA-DR. The major histocompatibility complex (MHC) class II molecule HLA-DR is also a hallmark of immature haematopoietic cells, although its expression remains on antigen-presenting cells, B-cells and activated T-lymphocytes [18]. The major role of HLA-DR is indeed to present exogenous peptides to T-lineage lymphocytes that will recognise an antigen’s epitope in the MHC molecule groove. No antigen presentation is supposed to be performed by immature haematopoietic cells. CD38. This molecule is a ribosyltransferase, expressed on numerous cell types and also considered to be absent from the most immature HSC [19]. It appears at an early differentiation stage and high CD38 expression is a key feature of normal B-cell precursors (see later). CD38 will remain on monocytes, granulocytic precursors, some natural killer (NK)-cells and on B-cells with high expression occurring again at their 19:42:15

Chapter 2: Antigens

CD11a

MAJOR MYELOID DIFFERENTIATION ANTIGENS

CD11b CD11c

CD18 Mg Mg Mg

CD13 COOH

CD15

3-fucosyl-N-acetyl-lactosamine NH2

β2 integrins CD16

CD33

CD16a

CD65

ceramide dodecasaccharide

GPI anchor

Figure 2.3  Major myeloid differentiation antigens. β-2 integrins are essential adhesion molecules of, among others, cells of the myeloid lineage, allowing their mobility. They act as receptors of extracellular matrix molecules through their N-terminal portion, involved in interactions relying on such ions as calcium or manganese. Their intracytoplasmic portion is in close association with glycoproteins of the cytoskeleton. Illustrated here are the three β-2 integrins, each containing an α chain (CD11a for leucocyte function antigen 1, CD11b for Mac-1 and CD11c with no other name) linked to the same β-2 chain. CD11b is frequently used to characterise monocytes and neutrophils. CD11c is used to differentiate CD123− dendritic cells (DC)1 from DC2. CD13 is an ectopeptidase sharing many similitudes with CD10, but restricted to the myeloid lineage, with a variable expression during maturation. CD15 and CD65 are two carbohydrates that attach to glycoproteins essentially on granulocytes and more weakly on monocytes. CD16 is a receptor for the Fc fragment of IgG, present in large amounts on neutrophils, with a glycerophosphatidyl inositol anchor. IgG ­recognition serves only to favour phagocytosis and does not seem to provide any signalling to neutrophils. CD33 is expressed early on myeloid progenitors, remains at high levels on monocytes and at lower levels on granulocytes.

terminal differentiation stage of plasma cells. CD38 is seen on activated T-cells. CD38 expression is also frequent on leukaemic blast cells of any lineage.

Myeloid Antigens Early Myeloid Antigens

early erythroid progenitors [21]. This tyrosine kinase, encoded by the KIT gene, has five extracellular domains of the Ig superfamily (IGSF)2. The tyrosine kinase activity is exerted by the intracytoplasmic portion of

Such domains are structures of about 110 amino acids organised in antiparallel β sheets linked by loops. Initially discovered in Igs, they are in fact some sort of a basic unit with adhesive properties involved in cell-cell communication. These units are used by a large number of molecules thereby enclosed in the abovementioned ‘immunoglobulin superfamily’.

2 

Three CDs appear early upon commitment of a haematopoietic progenitor towards the myeloid lineage (Figures 2.3 and 2.4). CD117 (Figure 2.2) is likely to be the earliest myeloid-associated antigen [20], also expressed on 03

19:42:15

17

Chapter 2: Antigens

CD91 CD163 CD304 CD14

L L L L L L L L L L

CD64

CD32

CD303

CD36

ITIM

Figure 2.4  Major differentiation antigens of monocytes and dendritic cells. CD14 is the hallmark of monocytes and myeloid-derived dendritic cells. This glycerophosphatidyl inositol-anchored molecule acts together with Toll-like receptor 4 in the recognition of bacterial lipopolysaccharide. CD32 and CD64 are also receptors for the Fc fragment of IgG involved in the phagocytosis of immune complexes and IgG-coated bacteria or cells. CD36 is a scavenger glycoprotein with multiple functions in the management of lipids. It is also the receptor of thrombospondin. CD91 is an intriguing large molecule of the superfamily of low-density lipoprotein receptors. It also acts, together with CD163, to limit the toxicity of free haemoglobin by binding haemoglobin-haemopexin complexes. CD163 belongs to the scavenger-receptor cysteine-rich superfamily. It is also involved in limiting the toxicity of free haemoglobin by binding haemoglobin-haptoglobin complexes. CD303 and CD304 are blood-derived dendritic cell antigens BDCA-2 and BDCA-4, specifically expressed by plasmacytoid dendritic cells. CD303 is a lectin that binds carbohydrates. CD304 has three types of domains, including two with homology to coagulation factors, ­interacting with semaphorins.

18

the molecule, while the extracellular domains bind the stem cell factor also called the steel factor. Activation through this cytokine activates cell signalling and favours cell proliferation and differentiation. CD117 is also known as c-kit, the cellular homologue of the viral oncogene v-kit. Besides early myeloid progenitors, CD117 is highly expressed on mature mast cells and its spontaneous activation through mutations is seen in mastocytosis [22]. CD33. Also a molecule of the IGSF, CD33 is a homodimer of two units carrying each two extracellular domains. CD33 is believed to play a role in cell– cell adhesion, especially between myeloid progenitors and stromal cells, through its capacity to bind sugars. CD33 indeed also belongs to the sialoadhesin SIGLEC family, proteins with lectin-like properties, that is, ability to bind carbohydrates [23]. The intracellular portion of CD33 carries immunoreceptor tyrosine-based inhibitory motifs (ITIMs) downregulating tyrosine phosphorylations. 03

CD13. This type-II transmembrane homodimeric protein (with an intracytoplasmic NH2 extremity) ­displays, among other properties, an ectopeptidase activity [24]. This means that this molecule is able to cut off two amino acids at the end of many proteins. This activity is mostly exerted by CD13 expressed on the epithelial cells of renal tubules or on enterocytes. On myeloid cells, this role is more obscure but is believed to help these cells to deal with proliferation/ differentiation factors, activating or inactivating them. CD33 and CD13 remain on cells engaged in monocytic differentiation and are co-expressed by mature monocytes. CD33 expression is characteristically higher on monocytes than on myeloid cells. During granulocytic maturation, CD33 expression decreases while that of CD13 increases. For the latter, there is also a transient decrease before full expression of the molecule occurs on mature granulocytes. These two markers are useful in identifying immature granulocytes which will be CD33hi/CD13lo. 19:42:15

Chapter 2: Antigens

Myelomonocytic Antigens Integrins3 appear progressively on differentiating cells both in granulocytic and monocytic lineage. β-2 integrins are heterodimers composed of the CD18 β chain and one of three α chains CD11a, CD11b and CD11c [25]. CD11b and CD11c are the most frequently investigated chains in haematological malignancies. In monocytic differentiation, they become expressed at the stage of promonocyte and are strongly expressed in mature monocytes. CD11b becomes expressed in granulocytic differentiation at the stage of myelocyte and increases during maturation to neutrophils [26]. The association CD11a/CD18 is also known as leucocyte function antigen 1 (LFA1) and CD11b/CD18 as Mac-1. CD14. By its function as the lipopolysaccharide (LPS) receptor, CD14 is involved in the recognition of bacteria by mature monocytes [27]. This molecule acts together with Toll-like receptor 4 (TLR4) to activate phagocytosis. CD14 is a very good marker of mature monocytes. It is bound to the cell surface by a glycerophosphatidyl inositol (GPI) anchor.4 Of note, ­myeloid-derived dendritic cells (DCs) can be produced from CD14+ peripheral monocytes. CD36. This small transmembrane molecule with both COOH– and NH2 terminuses in the cytoplasm and a short heavily glycosylated extracellular loop is expressed in the haematopoietic system on monocytes, platelets, immature erythroid cells and some plasmacytoid DCs [28]. It belongs to the family of scavenger molecules and acts as receptor for thrombospondin, long chain fatty acids, amyloid β and oxidised lipoproteins. It is broadly involved in the management of food lipids and its deregulation is considered in the pathogenesis of atherosclerosis as well as in thrombosis. It also recognises bacterial cell wall components, thus The large superfamily of integrins is composed of heterodimers comprising an α chain and a β chain. A given β chain can bind to different α chains. These molecules all share two important functions. Their extracellular portion is able to bind molecules of the extracellular matrix. Their intracytoplasmic portion is linked to molecules of the cytoskeleton. Integrins can thus be seen as the displacement machinery of cells, allowing them to crawl on the extracellular matrix to reach inflammatory/infected areas. 4  GPI anchors allow several proteins to be expressed on cells’ surfaces without being fully transmembrane. The GPI anchor is loosely bound to membrane phospholipids. Alteration in the enzyme manufacturing such bonds is responsible for paroxysmal nocturnal haemoglobinuria. 3 

03

taking part in innate immunity and in inflammation. It has anti-angiogenic properties. CD36 is involved in the clearance by the retinal pigment epithelium of shed photoreceptor outer segments. Finally, it is used as an entry way by Plasmodium falciparum. Receptors of the constant portion of IgG or Fc-gamma receptors are also expressed by monocytes, in order to facilitate the phagocytosis of opsonised particles, covered by antibodies specific to their surface antigens [29]. FcγRI or CD64 and FcγRII or CD32 are constitutively expressed on monocytes. Conversely, FcγRIII or CD16 is only expressed on activated inflammatory monocytes. Investigation of the expression of CD16 and CD14 delineates three monocyte subsets (see Chapter 3) [30]. These three molecules belong to the IGSF. Being professional antigen-presenting cells, monocytes express HLA-DR. Monocytes, at various stages of maturation, and also acute myeloid leukaemia (AML) of myelomonocytic differentiation may express low levels of the lymphoid antigens CD4 and CD19 (see below).

Other Granulocytic Antigens

CD10, which will be discussed later in the B-cell ­section, is also a marker of mature neutrophils [31]. Granulocytes use β-2 (CD18) integrins [26] to move on the extracellular matrix towards inflammatory areas where they are needed to resolve infections. These integrins are markers of mature neutrophil granulocytes. CD16 (FcγRIII), a member of the IGSF with two extracellular domains, appears progressively on granulocytic precursors and is strongly expressed on mature granulocytes. Of note, CD16 uses a GPI anchor on the granulocytic lineage. Eosinophils typically lack CD16 expression [31]. CD15. This carbohydrate residue (3-fucosyl-Nacetyl-lactosamine) binds to proteins expressed by neutrophil granulocytes and CD15s is also known as the blood group Sialyl LewisX [32]. Another carbohydrate is a hallmark of neutrophils, the ceramide dodecasaccharide CD65 [33]. When expressed on blast cells, these two markers are indicative of the commitment of leukaemic cells towards the granulocytic lineage. CD24. This small protein of only 12 amino acids and a GPI anchor, with a large carbohydrate attachment, is brightly expressed on mature neutrophil granulocytes [34]. It belongs to the same family as CD52, target of the Campath therapeutic antibody developed at Cambridge University and mostly expressed on ­lymphocytes and monocytes [35]. 19:42:15

19

Chapter 2: Antigens

Basophils are characterised by the expression of CD22 and CD123 (see later) together with CD203 or CRTH2 [36]. Eosinophils typically express higher levels of CD45 (see below) than neutrophils and lack CD16 as mentioned previously. By flow cytometry, eosinophils also show a characteristic autofluorescence in Pacific Blue, which may be helpful in immunophenotyping.

Dendritic Cells Antigens

20

One subset of DCs can derive from monocytes (DC1) and a second subset (DC2), also called plasmacytoid DC, is believed to differentiate from a lymphoid precursor [37]. CD123 [38] is the α chain of the receptor of ­interleukin-3 (IL-3). This cytokine is crucial in haematopoiesis and is particularly important to differentiate DC2 [39]. CD123 forms a heterodimer with the common β-c subunit of interleukin receptors CD131, shared by receptors for IL-5 and the granulocyte-­macrophage colony-stimulating factor in the β-common receptors family. These three molecules are all involved in haematopoiesis, immune responses and inflammation. IL-3 is involved in the development of multiple cell types including basophils, neutrophils, eosinophils, macrophages, erythroid cells, megakaryocytes and DC, as well as endothelial cells outside of the haematopoietic system. CD123 is expressed early on haematopoietic progenitors. It is also characteristic of hairy cell leukaemias and blastic plasmacytoid DC neoplasms only expressing CD4 and CD56 among classically explored antigens, in the absence of lineage-specific molecules [40]. CD123 is present at various levels on a number of cases of acute lymphoblastic or myeloblastic leukaemias, and is thus considered a good candidate for targeted therapy by MAbs, immunotoxins or CAR T-cells. Some antigens characterising DC belong to the group of blood-derived DC antigens (BDCAs) [41]. BDCA-2 (CD303 also known as C-type lectin domain family 4 member C or CLEC4C) and BDCA-4 (CD304 or neuropilin-1) are expressed by plasmacytoid DC together with high levels of CD123. CD303 is a transmembrane glycoprotein with a single lectin domain. Neuropilin-1 is a transmembrane molecule with only four intracytoplasmic amino acids, acting as a ­co-receptor for semaphorins. It has a curious extracellular structure with two terminal complement-binding homology domains, two coagulation factor V and VII homology domains and a meprin domain, which is involved in the binding of semaphorins. It is mostly 03

involved in the growth of neurons and in angiogenesis by being able to bind dimers of vascular endothelial cell growth factor (VEGF) via its coagulation factor homology domains. BDCA-3 (CD141), also known as thrombomodulin, is restricted to a small population of myeloid CD11c+/CD123− DCs called MDC2 [41]. These antigens are strongly modulated upon activation of DCs, which leads to the loss of CD303 and expression of CD141. CD141 is also expressed on endothelial cells and is involved in the regulation of coagulation. It is a transmembrane 74 kDa protein with an extracellular portion composed of six tandem EGF (epidermal growth factor) repeats and a serine-/threonine-rich spacer. Because of its expression on DCs, CD1c, one of the first antigens identified with MAbs and an isotype of the CD1 cluster with an MHC class I-like structure, is also called BDCA-1 [41].

Erythroid Antigens Blood groups are the hallmark of red blood cells but appear rather late in erythroid maturation (Figure 2.5). Early erythroblasts freshly differentiated from progenitors still carry CD45 and CD117 and will progressively lose these markers as they acquire CD36, CD71, CD105 and CD235 expressions. CD35, also known as CR1 (complement receptor 1), is a receptor from the gene family of regulators of complement activity, with an extended extracellular domain composed of 30 short consensus repeats or SCR. On red blood cells, CD35 participates in the elimination of immune complexes by the spleen. It is also expressed on B-cells, follicular DCs, neutrophils, macrophages, DCs and renal podocytes. It is also used by Plasmodium falciparum as a receptor [42]. As previously mentioned, CD36 is the thrombospondin receptor present on all erythroblasts with the brightest intensity on basophilic erythroblasts [43]. CD71, the transferrin receptor 1, is expressed by all cells requiring to use iron, including enterocytes. It is characteristic of normal erythroid cells but can also be expressed by proliferating cells, including leukaemic cells of various lineages [44, 45]. It is a homodimer type II protein, with an intracytoplasmic NH2 ­terminus. CD71 is used by arenaviruses to enter human cells. CD105, also known as endoglin, is a homodimer glycoprotein, member of the TGFβ-receptor family. It is expressed transiently during erythroid maturation at the proerythroblast stage [46] but is mostly a component of the vascular endothelium. It is responsible for the TGFβ-dependent growth of endothelial cells. 19:42:15

Chapter 2: Antigens

CD42

CD41

CD61

Ca Ca Ca Ca

CD35

CD36 CD71 CD105

COOH

NH2

Figure 2.5  Major differentiation antigens of platelets and red blood cells. CD41 is an integrin α chain, also known as gpIIb. It forms a heterodimer with CD61, an integrin β-3 chain also known as gpIIIa. This integrin is involved in clot formation through binding fibrinogen. CD42 or gpIb is one of the four sub-clusters of CD42. It binds Von Willebrand factor and is also involved in clot formation in primary haemostasis. CD35 is a long molecule acting as complement receptor, involved in the transport of immune complexes by ageing red blood cells reaching the spleen. CD71 is the transferrin receptor that is also present on enterocytes. It is a homodimer of still not completely resolved structure. CD105, also known as endoglin, is a homodimer of the transforming growth factor (TGF)-β receptor family and is also involved in angiogenesis. CD36 appears on this figure between the two types of receptors, as it is expressed both on platelets and erythroblasts. It is the same ­molecule as shown being a monocytic marker, with numerous functions.

CD235, also known as glycophorin A, is a highly glycosylated sialoglycoprotein present at all stages of erythroblasts differentiation [47]. It interacts with the band 3 protein of erythrocytes’ cytoskeleton [48]. CD241 or Rhesus-associated glycoprotein (RHAG) is a molecule with 12 transmembrane domains, making it an integral part of the erythrocytes membrane [49]. Four blood groups are associated with CD241 as epitopes of this protein: Duclos, DSLK, OIa and RHAG4. It is part of a macromolecular membrane complex, associated with the cytoskeleton through ankyrin, crucial to maintain the integrity of erythrocytes. It is also a member of the family of ammonium transporters regulating the entrance of NH3 and CO2 in erythrocytes. Two mutations of this gene have been reported, associated with overhydrated stomatocytosis related to monovalent cations leak [50]. 03

Thrombocytic Antigens

Platelets use a specific β-3 integrin, gpIIb/IIIa, which was discovered and named before the CD era (Figure 2.5). It is also called CD41/CD61 since both chains were later assigned CD numbers. This integrin binds fibrinogen and participates in platelets aggregation and clot formation [51]. Platelets also carry surface CD42b or gpIba, which binds Von Willebrand factor in the early stages of primary haemostasis together with gpV and gpIX [52]. These three molecules are expressed on the surface and in the cytoplasm of megakaryoblasts. CD42 exists as four sub-cluster molecules (CD42a, b, c and d), all characterised by the presence of extracellular leucine-rich glycoprotein (LRG) domains. CD42b is the most targeted in flow cytometry. It has a heavily glycosylated part prolonged, on the extracellular part of the 19:42:15

21

Chapter 2: Antigens

molecule, by a series of seven LRG domains. Anomalies in the structure of these proteins are responsible for such diseases as Glanzmann’s thrombasthenia (CD41/61) or Bernard-Soulier syndrome (CD42b) [53]. CD36, already mentioned as a monocytic antigen, by its role as thrombospondin receptor, is expressed as early as on megakaryoblasts and remains present on platelets [54]. It is also known as gpIV in the terminology of platelets’ glycoproteins.

Lymphoid Antigens B-Lineage Antigens

22

The hallmark of B-cells is their antigen receptor (B-cell receptor (BCR)) which is a surface immunoglobulin (sIg) (Figure 2.6). The latter is a symmetrical double heterodimer composed of two identical heavy chains and two identical light chains (κ or λ). Heavy chains have three or four constant domains defining their isotype, while κ or λ light chains have only one constant domain. The heavy and light chains have an NH2 terminal variable domain called variable heavy (VH) or variable light (VL). As alluded to in the introduction and elucidated by Susumu Tonegawa [6], the protein structure of these domains results from the rearrangement of gene fragments. Chromosome 14 carries a repertoire of V, D and J segments, one of each will be used by each mature heavy chain. On chromosomes 2 and 22, separate repertoires contain V and J segments, one of each being used to define a κ or λ VL. The B-cells genome will therefore undergo c­ omplex rearrangements in order for each successful progenitor to express a mature BCR. Autoreactive cells resulting from this random generation of diversity will be eliminated before they leave the BM to reach lymphoid ­tissues [55]. The presence of sIgs, most frequently investigated by the expression of membrane light chains, is a late event in B-cell maturation and only pertains to the mature forms of B-cell malignancies. Yet, a number of differentiation antigens are expressed sequentially during B-cell maturation. CD19. Commitment to the B-lineage is characterised by the surface expression of CD19. This molecule, belonging to the IGSF, has one extracellular domain and a long intracytoplasmic tail with immunoreceptor tyrosine-based activation motifs (ITAMs) involved in cell activation upon phosphorylation [56]. CD19 appears very early and will remain on the cell surface throughout its life. It is a pan-B antigen. However, as previously 03

mentioned, lineage infidelity exists for this antigen, and it may be also present at low levels on monocytic or monocytoid cells as well as in some myeloid leukaemias. CD79. The most specific-lineage commitment indicator of B-cells is the intracytoplasmic expression of CD79. CD79 is a CD with two members called CD79a and CD79b or Ig-α and Ig-β [57]. They both have an extracellular IGSF domain and a long intracytoplasmic tail with ITAMs. Both CD79a and CD79b are mandatory to allow the newly formed Igs to reach the cell surface. Some CD79a MAbs also bind an epitope expressed in T-cells lymphoblastic leukaemias and AML with RUNX1-RUNX1T1 [58–60]. CD20 is a linear molecule with four transmembrane domains, reminiscent of tetraspanins, although it does not belong to this molecular family. The function of CD20 is still unknown although there are indications that it could be involved in ion transport, especially chloride ions [61]. This molecule has been the first immensely successful target of biotherapies with anti-CD20 MAbs for the treatment of B-lineage haematological malignancies [62]. CD21 is a long molecule with cysteine-rich repeats which acts as a complement receptor [63]. On mature B-cells, it catches immune complexes and brings them in the vicinity of the BCR for possible epitope recognition and B-cell activation during humoral immune responses. CD21 is also used by the Epstein-Barr virus (EBV) to enter B-cells and immortalise them in EBVassociated malignancies [64]. CD21 is expressed at very low levels on some T-cells, explaining the occurrence of EBV-related T-cell lymphomas. CD22 is a member of the IGSF, a heterodimer of two very similar chains with five or seven extracellular domains and a long cytoplasmic tail carrying both ITAMs and antagonistic ITIMs . Both molecules appear to result from differential splicing of the same transcript. CD22 is a member of the sialoadhesins family, depending on sialic acid to exert its adhesive functions [21]. CD22 is initially expressed in the cytoplasm of B-cell progenitors, then reaches the cell surface. It acts as a regulator of B-cell activation [65]. It is also expressed on basophils. CD23 is a type II transmembrane glycoprotein with a lectin-like extracellular C terminus. It is a receptor for IgE. Its extracellular portion can be cleaved and provide proliferation signals to B-cells. It is also involved in cell-cell interactions via its adhesive properties [66]. CD24, mentioned earlier as a strong marker of mature neutrophil granulocytes, is also brightly expressed on B-cells [34]. 19:42:15

Chapter 2: Antigens

CD21

CD22

ou

CD19 CD20

ITIM

ITAM ITIM

ou

ITAM ITIM ITIM

CD138 COOH

CD23

CD200 CD24

CD79a

NH2

CD79b

ITAM ITAM ITAM ITAM

Figure 2.6  Major B-lineage antigens. CD19 is a pan-B antigen expressed very early in B-lineage commitment. CD19 will remain present throughout the life of B-cells, even normal plasma cells. It is a member of the immunoglobulin (lg) superfamily with an activation intracytoplasmic motif. CD20 is an integral transmembrane molecule with four spans, expressed later during B-cell maturation and remaining on B-cells up to the plasma cell stage. It is the target of successful monoclonal antibody-based immunotherapy. CD21 is a long molecule with cysteine-rich repeats, acting as a complement receptor 2 (CR2) and therefore capturing immune complexes. It is also used by Epstein-Barr virus as entry to B-cells. CD22, a molecule of the lg superfamily, exists in two forms with inhibitory intracytoplasmic motifs, involved in the regulation of B-cell ­activation. It is also expressed on basophils. CD23 is a lectin type II transmembrane protein, a hallmark of chronic lymphocytic leukaemia diagnosis. Its cleaved soluble form acts as a B-cell growth factor. CD24 is a very small heavily glycosylated protein with a glycerophosphatidyl inositol anchor, expressed on both mature B-cells and neutrophils. CD79a and CD79b form two heterodimers on each side of the sIg acting as antigen receptor on B-cells. They carry several activation motifs in their long intracytoplasmic portion. CD138, also known as semaphorin, is specific of the latest stage of B-cell maturation, that is, plasma cells and is also found on epithelial cells. CD200 is a member of the immunoglobulin superfamily, useful for differential diagnosis of lymphoproliferative disorders.

03

19:42:15

23

Chapter 2: Antigens

24

CD81 [67] is an integral transmembrane molecule of the tetraspanin family. CD81 has thus four transmembrane domains, both NH2 and COOH extremities in the cytoplasm and two extracellular loops. CD81 belongs to the activation complex of B-cells by being very close to CD19, CD21 and CD225 and participates in the regulation of B-cell responses upon binding of their antigen receptor (BCR) to a specific antigen. Tetraspanins are ancestral molecules likely to have been involved in the appearance of multicellularity, most likely because of their close association with integrins. CD81 has recently been shown to be able to change conformation in order to harbour or not a molecule of cholesterol in an intramembrane pocket [67]. It is also used by the hepatitis C virus to enter cells. CD81 is downregulated in lymphoproliferative disorders such as chronic lymphocytic leukaemias (CLLs). CD9, also a member of the tetraspanin family that interacts with CD19, is brightly expressed on B-cell precursors, downregulated on CD10-negative B-cells and upregulated again on plasma cells. Other members of the tetraspanin family, CD53 and CD37, show low expression on B-cell precursors and are upregulated on CD10− B-cells [68]. CD10 [69]. This molecule has broad similarities with CD13 and is expressed transiently on maturing B-cells. CD10 was initially identified as the common acute lymphoblastic leukaemia antigen or CALLA, already alluded to in this chapter [4]. This ectopeptidase is supposedly also involved, by its ability to chop off dipeptides from proteins, in the activation/­ inactivation of differentiation/proliferation cytokines. As for CD13, it is strongly expressed on enterocytes and renal tubules epithelial cells. CD138 [70] or syndecan-1 is a heavily glycosylated membrane proteoglycan from the family of semaphorins. These molecules are rich in heparin sulphate, close to heparin, which mediates adhesion to a large number of cell types and can also be shed in a soluble form. There are four syndecan genes, with high homology to that of drosophilae, suggesting conservation and duplication during evolution. The intracytoplasmic part of syndecans interacts with the cell’s cytoskeleton. In haematology, CD138 is a key marker of plasma cells. However, CD138 is also broadly involved in organogenesis and in the adhesion properties of epithelial cells. When a maturing B-cell has achieved a proper rearrangement of the genome encoding its future VH domain, it sort of ‘tests’ whether the resulting protein will be correct. At this pre-B stage, RNA derived from the 03

rearranged DNA will encode a μ chain (heavy chain of an IgM) and the resulting protein will be detectable in the cytoplasm of the cell [71]. This is a characteristic of pre-B ALL [72]. Physiologically, expression of minute amounts of this heavy chain with surrogate light chains will help the BM to discard early any cell involved in self-reactivity [73]. This pre-BCR will also trigger the rearrangement steps necessary to obtain a proper light chain, from the four possibilities offered by repertoires on two chromosomes 2 and two chromosomes 22. At that stage, CD10 becomes unnecessary and is no longer expressed. It will return later in the life of B-cells involved in the germinal centres of secondary lymphoid tissues, upon successful encounter of their antigen. These cells will proliferate, undergo an ­additional selection step and, if successful, further differentiate in Ig-secreting plasma cells or memory B-cells. The role of CD10 at this later stage is also probably to monitor growth/differentiation proteins. When the B-cell is able to produce both heavy and light chains, it assembles a BCR which is carried to the cell surface. Because of the order of the genes coding for constant domains and differential splicing, the naïve cell resulting from all these maturation steps will co-express surface IgM and IgD of the same specificity (same VH and VL). These are called μδ naïve B-cells which will reach secondary lymphoid organs and reside in the mantle zone of reactive germinal centres [74].

T-Lineage Antigens

CD2 is a pan-T antigen of the IGSF [75] with one canonical domain and one pseudo-IGSF (Figure 2.7). It is also known as LFA2 and was demonstrated to be the sheep red blood cell receptor responsible for the ‘rosette-­forming’ phenomenon mentioned in the introduction of this ­chapter, by binding its ligand CD58 on these erythrocytes. Its main functions are adhesion and signalling. CD3. As for B-cells, the hallmark of T-cells is the TCR or T-cell receptor [76]. This is a single heterodimer, composed either of an α chain and a β chain, or of a γ chain and a δ chain. Each of those chains belongs to the IGSF and has a constant domain and a variable domain. Variable domains are composed from VDJ and VJ segments positioned as repertoires on chromosomes 7 and 14. To be expressed on the cell surface and be able to activate T-cells after recognition of an antigen, the TCR requires the CD3 complex [77]. The latter is composed of several subunits. CD3 gamma, delta and epsilon chains have a single extracellular IGSF domain and a long intracytoplasmic tail. These are associated to two other chains, lacking an 19:42:15

Chapter 2: Antigens

TCR proper γδ or αβ

CD1 α3

CD2

CD3

γ

δ

ε

β2 microglobulin

ITAM ITAM ITAM

ε

zz or zη

ITAM ITAM ITAM ITAM ITAM ITAM ITAM

CD4

CD5 CD7

CD8

Figure 2.7  Major T-lineage antigens. CD1 is a molecule with several isoforms sharing structure homology with class I molecules of the major histocompatibility complex (MHC). CD1a is the isoform most involved in T-cell maturation, highly expressed on cortical thymocytes. CD2 is a molecule of the immunoglobulin superfamily with adhesion properties. Its ligand is CD58, expressed notably on sheep erythrocytes and responsible for the ability of T-cell to form rosettes with sheep red blood cells. CD3 is a multi-molecular complex allowing expression of the T-cell receptor (TCR) on the surface of mature T-cells having successfully ­rearranged this antigen-recognition heterodimer (γδ or αβ). It is equipped with 10 activation motifs (immunoreceptor tyrosine-based ­activation motifs) involved in T-cell activation upon antigen recognition. CD4 and CD8 define the major subsets of mature T-cells. They respectively check that a peptide is presented to the TCR in the groove of an MHC class I or MHC class II molecule, respectively. CD5 and CD7, respectively a scavenger (of the same family as CD163) and an adhesion molecule of the immunoglobulin superfamily, are early pan-T antigens that will remain on these cells through their whole life.

extracellular domain, yet with even longer cytoplasmic tails, which can be a ζ-ζ homodimer or a ζ-η heterodimer. In total, CD3 molecules associated to the TCR carry 10 ITAMs. They function as dimeric signalling molecules respectively δε, γε and ζζ or ζη, which become phosphorylated by the Src-family kinase Lck as the earliest signal upon antigen recognition [78]. The classically used detection of cytoplasmic CD3 to identify cells of the T-lineage at their early stage relies on the identification of the ε chain. Of note, this will also identify some NK-cells [79]. CD4 and CD8 are TCR co-receptors [80]. They will respectively check that a peptide presented to the TCR in the groove of a molecule of the MHC is associated to either class II MHC, presenting exogenous peptides, or class I MHC, presenting endogenous peptides. To this avail, CD4 and CD8 have the same size, in order 03

to span the two IGSF domains of the TCR and the two IGSF-like domains of MHC molecules. CD4 has four domains, three of the IGSF and one ‘pseudo-IGSF’. CD8 is a dimer, with a long stem topped by an IGSF domain. CD8 comes as an α-α homodimer or an α-β ­heterodimer. CD4dim expression is also seen on ­monocytes [81]. CD5, a pan-T antigen, is a type I transmembrane glycoprotein belonging to the scavenger-­ receptor cysteine-rich (SRCR) superfamily with three extracellular cysteine domains [82]. Its intracytoplasmic tail comprises a pseudo-ITAM. It is expressed on T-cells since the stage of early precursors as well as on a subset of B-cells (B1a) [83]. This molecule appears to exert a negative effect on TCR and BCR signalling [82], thereby participating in the growth advantage of some tumours where it remains overexpressed on lymphocytes. 19:42:15

25

Chapter 2: Antigens

CD7 is a pan-T antigen which appears to be the first to be expressed on the surface of early T-cells, at a stage where it is possible for these cells to reverse to the myeloid lineage [84]. Thus, CD7 expression is not T-cell specific. Structurally, it carries an extracellular domain composed of four tandem repeats of a consensus amino acid sequence ending with an NH2 IGSF domain. CD1a belongs to a family of five MHC class I antigens, capable of presenting lipid antigens to T-cells and thus mediating adaptive immunity to the vast range of microbial lipid antigens [85]. CD1a is characteristic of maturing T-cells located in the thymus cortex where it provides a checkpoint after which differentiation into NK or DCs will no longer be possible [86]. This molecule is also found on Langerhans cells, notably in the epidermis [87].

NK-Cells’ Antigens

26

NK-cells have long been thought to be devoid of surface antigens and were for a time called ‘nude lymphocytes’ (Figure 2.8). In fact, they display such a large array of receptors that their characterisation is limited to specialised laboratories involved in this peculiar subset of cells [88]. Two antigens are the most commonly used to define NK-cells: CD16 and CD56 [89]. CD16 on NK-cells is the same extracellular molecule as that described earlier on granulocytes, but it is fully transmembrane and has a cytoplasmic tail with ITAMs instead of being attached by a GPI anchor. CD56, initially identified on neural cells, is also called neural cell adhesion molecule. It has five extracellular distal IGSF domains and two proximal fibronectin-like type III domains. CD56 is capable of homodimerisation either between two cells or on the same cell. It provides important signals for neurite growth and neuromuscular interactions. Its role on NK-cells is probably of adhesion, but the level of expression of CD56 allows differentiation between cytotoxic NK-cells (low intensity) and those producing cytokines (bright intensity). CD56 is also expressed on activated CD3+ T-cells. Other surface antigens of NK-cells are the carbohydrate CD57 [90] and the killer Ig-like receptors or KIRs ([91], see below). Curiously, CD57, or glucuronic acid 3-sulphate is, like CD56, an important marker in neural cells differentiation, expressed on neural crest cells and best characterised initially in the neuroscience community [92]. CD94 is a C-lectin transmembrane glycoprotein which forms homodimers or heterodimers with NKG2A on the surface of NK-cells [93]. It displays a 03

selective affinity for MHC class-I HLA-E antigens. The CD94–NKG2A complex has a strong inhibitory activity impairing NK cytotoxicity upon ligand binding. CD94 homodimers may act differently yet still providing killing inhibition, a property possibly exploited by viruses to escape NK antiviral activity. CD158 is a CD with a number of isoforms (CD158a to CD158k), belonging to the KIR family of Ig-like receptors [94]. They possess ITIMs on their cytoplasmic part, which makes them inhibitory receptors, halting NK activation. They mostly recognise epitopes of the HLA-C MHC class-I antigens. CD160 is an IGSF molecule with a single extracellular Ig domain. It is present either as a GPI-anchored or transmembrane glycoprotein on cytolytic NK-cells, gamma-delta T-cells and a subset of CD8bright α-β T-cells containing granzyme and perforin [95]. It is also expressed on intestinal intraepithelial CD8+ T-lymphocytes. On NK-cells, CD160 is preferentially expressed on the cytotoxic CD56dim/CD16+ subset. Extra cysteine residues in the Ig domain allow for the formation of multimers linked by disulphide bonds. CD160 displays activating properties, favouring the cytotoxic activity of the cells expressing it as well as triggering the production of proinflammatory cytokines such as interferon γ, IL-6, IL-8 and tumour necrosis factor (TNF)-α. It should be noted that CD160 expressing B-cells derived from CLL use a similar CD160-mediated pathway to enhance their survival and cellular activation leading to IL-6 production [96]. CD161 [97] is a lectin essentially expressed on NK-cells and cytotoxic T-cells as well as on follicular DCs. Its ligand is the very similar (100 amino acids difference) lectin called Lectin-like transcript 1 (LLT1). Both depend on calcium for their functions and thus belong to the family of C-lectins. LLT1 is expressed on osteoblasts, germinal centre B-cells and plasmablasts. Although some conflicting results have been published, CD161 is globally recognised as a molecule inhibiting NK cytotoxicity and cytokine secretion, with the possible exception of interferon-γ production.

Other Antigens of Relevance in the Immunophenotyping of Haematological Malignancies

CD25 is the α chain of the receptor for IL-2 [45]. It is a marker of T-cell activation, required for the proliferation induced during immune responses by the 19:42:15

Chapter 2: Antigens

CD56

CD16

CD57 CD94

CD16b

CD160

COOH

γ/ζ

CD161

glucuronic acid 3-sulfate

NH2 ITAM ITAM ITAM

ITAM ITAM ITAM

KIR family i.e. CD158 isoforms

KIR2DS KIR2DL

KIR3DS

KIR3DL

Figure 2.8  NK-cells’ antigens. CD16 is the same molecule as on granulocytes, but with a full transmembrane portion. It acts in association with γ or ζ chains with activation motifs. CD56 or neural cell adhesion molecule (N-CAM), an adhesion molecule, is also present on cells of the nervous system and on plasmacytoid dendritic cells. CD57 is a carbohydrate associated with NK and CD8+ cytotoxic cells. CD94 is a lectin homodimer acting together with NKG2A to inhibit the cytotoxic activity of NK-cells upon recognition of MHC class I antigens on the target cell. CD160 is an immunoglobulin superfamily (IGSF) molecule expressed as shown as a transmembrane molecule or glycerophosphatidyl inositol-anchored. It is associated with the cytotoxic function of NK-cells. CD161 is a calcium-dependent lectin associated with the inhibition of NK-cells cytotoxicity and cytokine production. KIR families: Killer immunoglobulin-like receptors (KIR) have two or three extracellular domains of the IGSF and either short (S) or long (L) intracytoplasmic tails. Depending on the adaptor to which they interact, and on the motifs present on their intracytoplasmic portion, they can display activating or inhibitory functions on NK-cells cytotoxicity.

autocrine and paracrine secretion of IL-2. CD25 is constitutively expressed at a high level on regulatory T-cells (T-regs), which are also characterised by downregulation of the IL-7 receptor (CD127) and nuclear expression of the transcription factor FoxP3 (Forkhead box P3 or scurfin). The IL-2 receptor comprises three 03

chains, α (CD25), β (CD122) and the γ chain (CD132) common to several IL receptors. CD25 is the most commonly explored. It is expressed also on activated monocytes and aberrantly expressed in some lymphoproliferative disorders of the B-lineage such as hairy cell leukaemia (Figure 2.9). 19:42:15

27

Chapter 2: Antigens

TNF RECEPTOR FAMILY IGSF COSTIMULATORY MOLECULES

CHEMOKINE TETRASPANINS RECEPTORS

Figure 2.9  Differentiation antigen families. This cartoon exemplifies: – the costimulatory molecules belonging to the immunoglobulin superfamily, involved in T-cell activation/regulation, which carry one or two extracellular domains; – the structure of molecules of the tumour necrosis factor-receptor family, also often involved in cell regulation; – the characteristic structure of chemokine receptors with their numerous membrane spans that earned them, early on, the nickname of ‘serpentines’; and – the structure of tetraspanins, with, as indicated by their name, four membrane spans and a specific signature peptide in the ­intracytoplasmic loop. They interact notably with integrins and lipid rafts.

28

CD26 [44, 98] is another marker of T-cell activation and a co-stimulatory molecule preferentially expressed on CD4 T-cells. It is also known as dipeptidyl peptidase IV which designates its intrinsic enzymatic activity. As other peptidases (i.e. CD10 or CD13), it is expressed on a variety of epithelial cells, notably in the gut, kidney and liver. It can also be expressed by lymphocytes, haematopoietic progenitors and fibroblasts. It selectively cleaves two amino acids at the NH2 terminus of peptides possessing a proline or alanine as penultimate residue. This versatile serine protease can also act, besides its enzymatic activity, as a receptor, especially for the adenosine deaminase (ADA). It binds collagen and fibronectin and also appears to be involved in apoptosis. By removing N-terminal amino acids on several cytokines, it interferes with a number of networks involved in cell activation and regulation [99]. CD26’s lack of expression is a characteristic of the T-lymphocytes of Sézary syndrome and mycosis ­fungoides (See Chapter 9). CD27 [45] is a co-stimulatory molecule. It is a trimer, member of the TNF-receptor superfamily. Preferentially expressed on T-lymphocytes, it can also be present on the surface of B- and NK-cells. CD27, together with its ligand CD70, is involved in cell survival and apoptosis and plays a role in the regulation of B-cells and Ig synthesis. 03

CD28 [45] is another co-stimulatory molecule of T-lymphocytes, also known as B7, constitutively expressed on naïve and activated cells. It is a member of the IGSF, with one extracellular domain, and its ligands are CD80 and CD86, also IGSF members, with two extracellular domains. CD30 [45], also a member of the TNF-receptor superfamily is a trimer expressed on activated T-cells, with co-stimulatory properties. CD30 was discovered as marker of Reed-Sternberg cells in Hodgkin’s lymphoma (see Chapter 10), but it is present on several T-cell lymphomas and some B-cell lymphomas. CD30 can be therapeutically targeted by the immunotoxin brentuximab vedotin. CD40 [100] is another member of the TNF-receptor superfamily and its ligand (CD40L or CD154) belongs to the TNF superfamily. The latter is expressed by CD4+ T-cells and CD40 is broadly expressed by many cell types liable to activate T-lymphocytes. It was initially discovered on B-cells and was shown to be involved in the hyper IgM syndrome following a pathogenic mutation. CD40 is also a key component of antigen-presenting cells such as DCs and monocytes, but it can also be expressed by endothelial cells, fibroblasts or epithelial cells. CD40CD40L interactions are critical in inflammatory 19:42:15

Chapter 2: Antigens

reactions and drive the production of proinflammatory cytokines such as IL-12. Both CD40 and CD40-L exist as soluble forms with cytokine-like functions. CD43 [101] is also known as sialophorin or leucosialin. It is a heavily glycosylated transmembrane glycoprotein composed in 50% of carbohydrates (sialic acid, galactose and N-acetylgalactosamine, essentially). Intercellular adhesion molecule-1 (ICAM-1) (CD54), galectin and CD45 have been shown to interact with CD43 as well as with ezrin and moesin in the cell’s cytoplasm. CD43 has many functions such as adhesion and anti-adhesion, locomotion, cellular activation, differentiation, proliferation and apoptosis. It is mostly expressed on T-cells but is also present on granulocytes, monocytes and some B-cell subsets. CD43 can be upregulated in some B-lineage lymphoproliferative disorders. CD49 is an integrin α-chain with numerous isoforms (CD49a to CD49f) which form heterodimers with the β-1 integrin CD29 [102, 103]. CD49 isoforms are also called very late antigens. The major role of these integrins is to favour neutrophil movements and thus migration. Some isoforms are also expressed on lymphocytes (see also Chapter 7). CD54 [45], also known as ICAM-1, is a member of the IGSF. CD54 is expressed by activated endothelial cells and some stromal cells. It is involved in interactions with leucocyte integrins of the CD11/CD18 family (LFA1, Mac-1 and CD11cD18) favouring cell adhesion to blood vessels and diapedesis. CD58 [104] is the ligand of CD2 and the molecule responsible for the phenomenon of ‘rosette formation’ between human T-lymphocytes and sheep red blood cells, since it is expressed on the latter. CD58 is also known as LFA3. It is a molecule of the IGSF, very similar to CD2, suggesting some gene duplication at some time during evolution. CD58 exists as a transmembrane form and as a GPI-anchored protein. It allows interactions between T-cells and antigen-presenting cells in the immunological synapse, as well as stable binding between CD8+ T-cells or NK-cells and their targets during cytotoxicity. CD60 [105] is an acidic glycosphingolipid (or ganglioside) which can be differently acetylated on its sialic residues. CD60a is the non-acetylated form of the ganglioside, while O-acetylation of carbons 9 or 7 of the sialic acid results in, respectively, 9-O-acetylated GD3 (CD60b) and 7-O-acetylated GD3 (CD60c). These three molecules are involved in the finely tuned pathways controlling B- and T-lymphocyte’s proliferation and apoptosis. 03

CD66 [106] is a member of the family of carcinoembryonic antigens, also members of the IGSF. There are five isoforms of CD66, respectively CD66a to CD66e. These antigens are expressed on neutrophils, epithelial cells and some leucocytes with isotype selectivity and engage in homophilic or heterophilic intercellular adhesion activities. CD66 isoforms can also bind carbohydrate motifs on several types of bacteria and thus appear to be involved in the protection of mucosa and phagocytosis/destruction of microorganisms by neutrophils. CD86 [45], as mentioned earlier, is one of the ligands of CD28. It can also bind the negatively regulating molecule CTLA4, thus driving T-cell responses towards activation or anergy, respectively. CD86 is a molecule of the IGSF with two extracellular domains. CD91 [107, 108] is a member of the superfamily of low-density lipoprotein (LDL) receptors and was first named LDL receptor-related protein (LRP) for this property. It recognises, in fact, a multitude of ligands, besides LDL, including heat shock proteins, extracellular matrix, cytokines, viruses and more (up to 40 described). It is expressed by macrophages but also numerous other cell types such as hepatocytes, fibroblasts or neurons. It is a large glycoprotein with an extracellular portion of 31 complement-type repeats, 23 EGF repeats and 8 so-called YWTD propeller domains. As other scavenger receptors, it has the ability to induce the production of proinflammatory cytokines by macrophages and antigen-presenting cells. Moreover, it acts together with CD163 to limit the toxicity of free heme by binding and internalising the haemopexin–heme complex and is thus involved in the recycling of iron. CD95 [45, 109], also known as Fas, Fas receptor or apoptosis antigen 1 is a death receptor appearing on the surface of activated T-cells at the end of an immune response. It is involved in the apoptosis of effector cells through binding Fas ligand present on numerous cell types. The Fas pathway of apoptosis depends on the formation of the death complex DISC (death inducing signalling complex) and is an alternative to mitochondrial apoptosis. It is a member of the TNFreceptor family of trimers involved in programmed cell death. Other molecules of this family with close relationship and similarity of activity are TRAIL5-R1 and TRAIL-R2 which directly interact with the Fas-related death domain to activate the caspase pathway leading to a possible mitochondrial-independent apoptosis. 5 

TRAIL = TNF-related apoptosis-inducing ligand. 19:42:15

29

Chapter 2: Antigens

30

CD135 [110], also known as FMS-like tyrosine kinase 3 (FLT3), is expressed on myeloid cells. It is a member of the IGSF with five extracellular domains and has two kinase intracytoplasmic domains. An active kinase is formed upon homodimerisation after binding to the FLT3 ligand. CD135 is important in the development of lymphocytes. Its level of expression is low in acute myeloblastic leukemias [111]. CD148 [112], together with CD45, belongs to the family of receptor-like protein tyrosine phosphatases (RPTPs). It acts in a similar way as CD45 in the regulation of cell activation by interacting with src kinases. At variance with CD45, CD148, however, seems to be determinant for the function of B1 cells, closer to innate immunity than classical B2 B-lymphocytes. CD152 [113], also known as CTLA4, is a molecule of the IGSF with a single extracellular domain. It forms homodimers on the cell’s surface and also exists as soluble monomers. As CD28, CD152 binds CD80 and CD86, but at variance with CD28, this provides a negative regulatory signal. CD152 is constitutively expressed by T-regs and is upregulated on the surface of activated T-cells. This leads to the extinction of immune responses, in a manner resembling that exerted by the PD-1/PD-L1/2 system (programmed death and programmed death ligand). CD163 [114] is, like CD5, a heavily glycosylated glycoprotein, a member of the SRCR superfamily with nine extracellular cysteine-rich domains. CD163 is expressed at low levels on the surface of monocytes and at much higher levels after their transformation in macrophages. It also exists in soluble form (extracellular part) in most body fluids. CD163 is expressed at its highest levels during the resolution phase of inflammation and tissue repair. It is upregulated by glucocorticoids and IL-10. Conversely, proinflammatory signals dampen CD163 expression. CD163 has been shown to be a specific receptor of the haptoglobin–­haemoglobin (Hp-Hb) complex but binds neither of these molecules outside of such complexes. Upon endocytosis of Hp-Hb, CD163-expressing macrophages produce high levels of anti-inflammatory cytokines such as IL-10, amplifying the loop of anti-inflammation. They also, thus, enter in the cycle of iron recycling. Other receptors of CD163 have been described such as TWEAK (TNF-like weak inducer of apoptosis), also involved in the regulation of innate and immune reactions. Soluble CD163 also participates in the regulation of immune responses by impairing T-cell proliferation. 03

CD180 [115, 116], initially known as RP105 (radioprotective 105 kDa), is a TLR homologue involved in the recognition of bacterial LPS by TLR-4 and therefore implicated in innate immunity on not only dendritic cells but also B-cells. It is a type I glycoprotein with many leucine-rich repeats in its extracellular solenoidal portion, very similar to TLR-4. It functions together with MD-2, a member of the lipid recognition family. CD194 and CD197 [117–119] are chemokine receptors, respectively, specific for CCR4 and CCR7. As chemokine receptors, they have a typical ‘serpentine’ structure of at least seven membrane spanning portions which makes them integral membrane proteins. Both are expressed on many cell types including epithelial cells as well as B- and T-lymphocytes. Abnormal expression of these chemokine receptors has been reported in several diseases, from asthma and autoimmune diseases to various types of cancer, including lymphomas. By driving cell chemotaxis, they could be involved in metastasis. CD200 [120] or OX2 is a broadly expressed transmembrane molecule interacting with an inhibitory receptor (CD200R) present on myeloid cells. The interaction between these two molecules is probably involved in the downregulation of inflammation. The expression of CD200 on tumour cells could also be a way to protect them from anti-tumour cytotoxicity mediated by myeloid cells. Both CD200 and CD200R are members of the IGSF with two heavily glycosylated extracellular domains. CD200 can be applied to differentiate between chronic lymphocytic leukaemia (CLL) and mantle cell lymphoma within CD19+CD5+ lymphoproliferative disorders (see Chapter 7). CD229 [121] or Ly9 is a member of the IGSF with four extracellular domains. It belongs also to the CD150 (signalling lymphocyte-activating molecule (SLAM)) superfamily of receptors involved in the regulation of T and NK-cells functions. CD229 engages in homophilic adhesion processes since it is expressed both on T-cells and antigen-presenting cells. In this context, it is a component of the immunological synapse between T- and B-cells. CD278 [122], also known as ‘inducible costimulatory’ or ICOS, is an IGSF member with one extracellular domain. It shares much similarity with CD28 and also acts as a CD4+ T-cell co-stimulation receptor, yet preferentially on activated T-cells, while CD28 is present also on naïve and resting T-cells. Both molecules 19:42:15

Chapter 2: Antigens

have an intracytoplasmic motif6 able to bind the kinase Phosphoinositide 3 (PI3-kinase) YMNM in CD28 or YMFM in ICOS. The one amino acid difference in these motifs is sufficient to activate different pathways upon engagement of CD28 or CD278. CD279 [123] is the immune checkpoint glycoprotein PD-1, mostly expressed on the surface of activated T-cells, initially identified as an apoptosis-associated molecule. PD-1 is an important regulator of cellular immune responses, inducing apoptosis or anergy by ‘exhaustion’ on lymphocytes having expanded in the course of an immune reaction. Binding to its ligands, PDL-1 or PDL-2 triggers this arrest of T-cell activation. Although important physiologically, this mechanism is diverted by many tumours in their attempt to escape anti-tumour immune responses. These molecules are good therapeutic targets to restore functional immunity, especially in the context of malignancies. PD-1 is a member of the IGSF with a single extracellular domain and carries some homology with CD28, CTLA4 (CD152) and ICOS (CD278). CD300e [124], or IREM-2, belongs to the superfamily of seven CD300 molecules in humans. These molecules are also members of the IGSF with one extracellular domain. CD300a and CD300f have long intracellular domains with ITIMs and CD300e shares with the other members of the cluster a short intracytoplasmic tail and interaction with adaptor proteins. CD300e is expressed by monocytes and myeloid DCs and functions with the adapter molecule DAP12. It participates in the activation of DCs and hence T-cellrecognising antigens on the latter. CD317 [125, 126], also known as tetherin or bone marrow stromal cell antigen 2, is a curious type II glycoprotein (intracytoplasmic N terminus) with a GPI anchor at its C terminus and a coil-coiled extracellular structure. Expressed as a dimer, it is part of lipid rafts and binds virions-containing vesicles through its GPI anchor. CD317 thus prevents the budding of viral vesicles and viral dissemination. It is also involved in survival in poor nutriment conditions by interacting with the mitochondrial pathway to inhibit apoptosis. CD317 is present on cells of the B-lineage in haematopoietic tissues. CD317 expression is enhanced by interferon α. CD319 [127] is another member of the CD2 family, also known as CS1 (CD2 subset 1) or CD2-like

receptor-activating cytotoxic cells (CRACC). CD319 is expressed on activated B-lymphocytes, plasma cells, NK-cells, CD8+ T-lymphocytes and mature DCs. It is a member of the IGSF with two extracellular domains. By homophilic interactions, it activates the cytotoxic properties of NK-cells. It is also involved in B-cell growth upon antigen recognition through the L1 isoform expressed on B-lymphocytes and by increasing the amount of autocrine cytokines. CD319 is the target of elotuzumab, an anti-SLAM antibody proposed for the treatment of multiple myeloma. CD371 [128], also known as CLEC12A or CLL-1, is a type II protein of the C-lectin family. As other molecules of this family, it has the properties of a pattern-recognising receptor (PRR) capable of binding pathogen-associated molecular patterns (PAMPs) on pathogens or death-associated molecular patterns (DAMPs) on host cells having suffered non-­homeostatic death. Upon cell death, the soluble uric acid which is released is transformed in monosodium urate crystals by combining with soluble sodium ions. CD371 is capable of binding these crystals and thus reducing inflammation through the ITIMs on its intracytoplasmic tail that prevent activation of inflammatory cells such as granulocytes, monocytes, macrophages or DCs. In AML and myelodysplastic syndromes (MDS), CD371 has also been shown to discriminate between normal and leukaemic stem cells [129].

The motifs are described by the letters corresponding to the amino acids composing them. Y=Tyrosine, M=Methionine, N=Asparagine, F=Phenylalanine.

5.

6 

03

References 1.

2.

3.

4.

E.R. Unanue, H.M. Grey, E. Rabellino, P. Campbell and J. Schmidtke Immunoglobulins on the surface of lymphocytes. II. The bone marrow as the main source of lymphocytes with detectable surface-bound immunoglobulin. J Exp Med; 133 (1971):1188–98. M. Jondal, G. Holm and H. Wigzell. Surface markers on human T and B lymphocytes. I. A large population of lymphocytes forming nonimmune rosettes with sheep red blood cells. J Exp Med; 136 (1972): 207–215. A.V. Hoffbrand, K. Ganeshaguru, G. Janossy, et al. Terminal deoxynucleotidyl-transferase levels and membrane phenotypes in diagnosis of acute leukaemia. Lancet; 2 (1977):520–3. G. Janossy and M.F. Greaves. Diagnostic use of an antiserum made against acute lymphoid leukemia associated antigen. Bibl Haematol; 45 (1978): 156–160. G. Köhler and C. Milstein. Continuous cultures of fused cells secreting antibody of predefined specificity. Nature; 256 (1975):495–7. 19:42:15

31

Chapter 2: Antigens

6.

7.

8.

9.

10. 11.

12.

13.

14. 15. 16. 17. 18.

19. 20.

32

21.

H. Sakano, R. Maki, Y. Kurosawa, W. Roeder and S. Tonegawa. Two types of somatic recombination are necessary for the generation of complete immunoglobulin heavy-chain genes. Nature; 286 (1980):676–83. A. Bernard and L. Boumsell. The clusters of differentiation (CD) defined by the First International Workshop on Human Leucocyte Differentiation Antigens. Hum Immunol; 11 (1984):1–10. H. Zola and B. Swart. The human leucocyte differentiation antigens (HLDA) workshops: the evolving role of antibodies in research, diagnosis and therapy. Cell Res; 15 (2005):691–4. G. Clark, H. Stockinger, R. Balderas, et al. Nomenclature of CD molecules from the tenth human leucocyte differentiation antigen workshop. Clin Transl Immunology; 5 (2016):e57. S. Hashimoto, S. Nagai, J. Sese, et al. Gene expression profile in human leukocytes. Blood; 101 (2003): 3509–13. H.J. Nam, F. Poy, H. Saito and C.A. Frederick. Structural basis for the function and regulation of the receptor protein tyrosine phosphatase CD45. J Exp Med; 201 (2005):441–52. D.R. Sutherland, L. Anderson, M. Keeney, R. Nayar and I. Chin-Yee. The ISHAGE guidelines for CD34+ cell determination by flow cytometry. International Society of Hematotherapy and Graft Engineering. J Hematother; 5 (1996):213–26. M. Maynadié, L. Gerland, S. Aho, et al. GEIL. Clinical value of the quantitative expression of the three epitopes of CD34 in 300 cases of acute myeloid leukemia. Haematologica; 87 (2002):795–803. T. Muramatsu and H. Muramatsu. Carbohydrate antigens expressed on stem cells and early embryonic cells. Glycoconj J; 21 (2004):41–5. E. Guenova, D. Ignatova, Y.T. Chang, et al. Expression of CD164 on Malignant T cells in Sézary Syndrome. Acta Derm Venereol; 96 (2016):464–7. A. Wolanczyk-Medrala, W. Barg and W. Medrala. CD164 as a basophil activation marker. Curr Pharm Des; 17 (2011):3786–96. Z. Li. CD133: a stem cell biomarker and beyond. Exp Hematol Oncol; 2 (2013):17. S. Huang and L.W. Terstappen. Lymphoid and myeloid differentiation of single human CD34+, HLA-DR+, CD38− hematopoietic stem cells. Blood; 83 (1994):1515–26. E. Ferrero and F. Malavasi. The metamorphosis of a molecule: from soluble enzyme to the leukocyte receptor CD38. J Leukoc Biol; 65 (1999):151–61. M. Miettinen and J. Lasota. KIT (CD117): a review on expression in normal and neoplastic tissues, and mutations and their clinicopathologic correlation. Appl Immunohistochem Mol Morphol; 13 (2005):205–20. M.O. De Jong, G. Wagemaker and A.W. Wognum. Separation of myeloid and erythroid progenitors 03

22. 23.

24. 25.

26.

27.

28. 29. 30. 31. 32. 33.

34.

35. 36. 37.

based on expression of CD34 and c-kit. Blood; 86 (1995):4076–85. D. Ribatti. The development of human mast cells. An historical reappraisal. Exp Cell Res; 342 (2016):210–15. S. Kelm, R. Schauer and R.R. Crocker. The Sialoadhesins – a family of sialic acid-dependent cellular recognition molecules within the immunoglobulin superfamily. Glycoconj J; 13 (1996):913–26. P. Mina-Osorio. The moonlighting enzyme CD13: old and new functions to target. Trends Mol Med; 14 (2008):361–71. C.G. Gahmberg, M. Tolvanen and P. Kotovuori. Leukocyte adhesion – structure and function of human leukocyte beta2-integrins and their cellular ligands. Eur J Biochem; 245 (1997):215–32. G.S. Kansas, M.J. Muirhead and M.O. Dailey. Expression of the CD11/CD18, leukocyte adhesion molecule 1, and CD44 adhesion molecules during normal myeloid and erythroid differentiation in humans. Blood; 76 (1990):2483–92. F. Stelter, M. Pfister, M. Bernheiden, et al. The myeloid differentiation antigen CD14 is N- and O-glycosylated. Contribution of N-linked glycosylation to different soluble CD14 isoforms. Eur J Biochem; 236 (1996):457–64. Y.M. Park. CD36, a scavenger receptor implicated in atherosclerosis. Exp Mol Med; 46 (2014):e99. A. Tamm and R.E. Schmidt. IgG binding sites on human Fc gamma receptors. Int Rev Immunol; 16 (1997):57–85. K.L. Wong, W.H. Yeap, J.J. Tai, et al. The three human monocyte subsets: implications for health and disease. Immunol Res; 53 (2012):41–57. M.T. Elghetany. Surface antigen changes during normal neutrophilic development: a critical review. Blood Cells Mol Dis; 28 (2002):260–74. P. Tetteroo and A.D. Geurts van Kessel. Expression of CD15 (FAL) on myeloid cells and chromosomal localization of the gene. Histochem J; 24 (1992):777–82. M. Noguchi, N. Sato, H. Sugimori, K. Mori and K. Oshimi. A minor E-selectin ligand, CD65, is critical for extravascular infiltration of acute myeloid leukemia cells. Leuk Res; 25 (2001):847–53. C. Ayre, N.K. Pallegar, N.A. Fairbridge, et al. Analysis of the structure, evolution, and expression of CD24, an important regulator of cell fate. Gene; 590 (2016):324–37. A. Domagała and M. Kurpisz. CD52 antigen--a review. Med Sci Monit; 7 (2001):325–31. P. Valent. Immunophenotypic characterization of human basophils and mast cells. Chem Immunol; 61 (1995):34–48. M.L. Kapsenberg, C.M. Hilkens, E.A. Wierenga and P. Kalinski. The paradigm of type 1 and type 2 antigen-presenting cells. Implications for atopic allergy. Clin Exp Allergy; 29 (1999):33–6. 19:42:15

Chapter 2: Antigens

38. U. Testa, E. Pelosi and A. Frankel. CD 123 is a membrane biomarker and a therapeutic target in hematologic malignancies. Biomark Res; 2 (2014):4. 39. V.I. Patel and J.P. Metcalf. Identification and characterization of human dendritic cell subsets in the steady state: a review of our current knowledge. J Investig Med; 64 (2016):833–47. 40. K. Laribi, N. Denizon, A. Besançon, et al. Blastic plasmacytoid dendritic cell neoplasm: from origin of the cell to targeted therapies. Biol Blood Marrow Transplant; 22 (2016):1357–67. 41. A. Dzionek, A. Fuchs, P. Schmidt, et al. BDCA-2, BDCA-3, and BDCA-4: three markers for distinct subsets of dendritic cells in human peripheral blood. J Immunol; 165 (2000):6037–46. 42. M. Krych-Goldberg and J.P. Atkinson. Structure function of complement receptor type I. Immunol Rev; 180 (2001):112–22. 43. M. Fajtova, A. Kovarikova, P. Svec, E. Kankuri and J. Sedlak. Immunophenotypic profile of nucleated erythroid progenitors during maturation in regenerating bone marrow. Leuk Lymphoma; 54 (2013):2523–30. 44. P. Aisen. Transferrin receptor 1. Int J Biochem Cell Biol; 36 (2004):2137–43. 45. M. Shipkova and E. Wieland. Surface markers of lymphocyte activation and markers of cell proliferation. Clin Chim Acta; 413 (2012): 1338–49. 46. F. Nassiri, M.D. Cusimano, B.W. Scheithauer, et al. Endoglin (CD105): a review of its role in angiogenesis and tumor diagnosis, progression and therapy. Anticancer Res; 31 (2011):2283–90. 47. J.P. Cartron and C. Rahuel. Human erythrocyte glycophorins: protein and gene structure analyses. Transfus Med Rev; 6 (1992):63–92. 48. K. Giger, I. Habib, K. Ritchie and P.S. Low. Diffusion of glycophorin A in human erythrocytes. Biochim Biophys Acta; 1858 (2016):2839–45. 49. G. Matassi, B. Chérif-Zahar, V. Raynal, P. Rouger and J.P. Cartron. Organization of the human RH50A gene (RHAG) and evolution of base composition of the RH gene family. Genomics; 47 (1998):286–93. 50. C. Badens and H. Guizouarn. Advances in understanding the pathogenesis of the red cell volume disorders. Brit J Haematol; 174 (2016):674–85. 51. J.S. Bennett. Regulation of integrins in platelets. Biopolymers; 104 (2015):323–33. 52. E.E. Gardiner and R.K. Andrews. Structure and function of platelet receptors initiating blood clotting. Adv Exp Med Biol; 844 (2014):263–75. 53. K.J. Clemetson and J.M. Clemetson. Molecular abnormalities in Glanzmann’s thrombasthenia, Bernard-Soulier syndrome, and platelet-type von Willebrand’s disease. Curr Opin Hematol; 1 (1994):388–93. 03

54. J.R. Nofer and M. van Eck. HDL scavenger receptor class B type I and platelet function. Curr Opin Lipidol; 22 (2011):277–82. 55. K.P. Lam and K. Rajewsky Rapid elimination of mature autoreactive B cells demonstrated by Cre-induced change in B cell antigen receptor specificity in vivo. Proc Natl Acad Sci U S A; 27 (1998):13171–75. 56. K. Wang, G. Wei and D. Liu. CD19: a biomarker for B cell development, lymphoma diagnosis and therapy. Exp Hematol Oncol; 1 (2012):36. 57. E.M. Fuentes-Pananá, G. Bannish, F.G. Karnell, J.F. Treml and J.G. Monroe. Analysis of the individual contributions of Igalpha (CD79a)- and Igbeta (CD79b)-mediated tonic signaling for bone marrow B cell development and peripheral B cell maturation. J Immunol; 177 (2006):7913–22. 58. M. Hashimoto, Y. Yamashita and N. Mori. Immunohistochemical detection of CD79a expression in precursor T cell lymphoblastic lymphoma/ leukaemias. J Pathol; 197 (2002):341–7. 59. R. Lai, J. Juco, S.F. Lee, S. Nahirniak and W.S. Etches. Flow cytometric detection of CD79a expression in T-cell acute lymphoblastic leukemias. Am J Clin Pathol; 113 (2000): 823–30. 60. R.C. Johnson, L. Ma, A.M. Cherry, D.A. Arber and T.I. George. B-cell transcription factor expression and immunoglobulin gene rearrangement frequency in acute myeloid leukemia with t(8;21)(q22;q22). Am J Clin Pathol; 140 (2013):355–62. 61. J.K. Riley and M.X. Sliwkowski. CD20: a gene in search of a function. Semin Oncol; 27 (2000):17–24. 62. P. Boross and J.H.W. Leusen. Mechanisms of action of CD20 antibodies. Amer J Cancer Res; 2 (2012); 676–90. 63. J.P. Hannan. The structure-function relationships of complement receptor type 2 (CR2; CD21). Curr Protein Pept Sci; 17 (2016):463–87. 64. E. Levy, J. Ambrus, L. Kahl, et al. T lymphocyte expression of complement receptor 2 (CR2/ CD21): a role in adhesive cell-cell interactions and dysregulation in a patient with systemic lupus erythematosus (SLE). Clin Exp Immunol; 90 (1992):235–44. 65. S. Sato, J.M. Tuscano, M. Inaoki and T.F. Tedder. CD22 negatively and positively regulates signal transduction through the B lymphocyte antigen receptor. Semin Immunol; 10 (1998):287–97. 66. J.Y. Bonnefoy, S. Lecoanet-Henchoz, J.F. Gauchat, et al. Structure and functions of CD23. Int Rev Immunol; 16 (1997):113–28. 67. B. Zimmerman, B. Kelly, B.J. McMillan, et al. Crystal Structure of a full-length human tetraspanin reveals a cholesterol-binding pocket. Cell; 167 (2016):1041–51. 68. S. Barrena, J. Almeida, M. Yunta, et al. Aberrant expression of tetraspanin molecules in B-cell chronic lymphoproliferative disorders and its correlation 19:42:15

33

Chapter 2: Antigens

69.

70. 71.

72. 73. 74.

75.

76. 77.

78. 79.

80. 81.

82.

34

83.

with normal B-cell maturation. Leukemia; 19 (2005):1376–83. V. Maguer-Satta, R. Besançon and E. BachelardCascales. Concise review: neutral endopeptidase (CD10): a multifaceted environment actor in stem cells, physiological mechanisms, and cancer. Stem Cells; 29 (2011):389–96. M. Bernfield, M. Götte, P.W. Park, et al. Functions of cell surface heparan sulfate proteoglycans. Annu Rev Biochem; 68 (1999):729–77. P. Guglielmi and J.L. Preud’homme. Immunoglobulin expression in human lymphoblastoid cell lines with early B cell features. Scand J Immunol; 13 (1981):303–11. J.C. Brouet, J.L. Preud’homme, C. Penit, et al. Acute lymphoblastic leukemia with pre-B-cell characteristics. Blood; 54 (1979):269–73. R. Übelhart, M. Werner and H. Jumaa. Assembly and function of the precursor B-cell receptor. Curr Top Microbiol Immunol; 393 (2016):3–25. E.L. Mather, K.J. Nelson, J. Haimovich and R.P. Perry. Mode of regulation of immunoglobulin mu- and delta-chain expression varies during B-lymphocyte maturation. Cell; 36 (1984):329–38. P. Sayre and E.L. Reinherz. Structure and function of the erythrocyte receptor CD2 on human T lymphocytes: a review. Scand J Rheumatol; 76 (1988):131–44. K.W. Wucherpfennig. The first structures of T cell receptors bound to peptide-MHC. J Immunol; 185 (2010):6391–3. M.E. Call, K.W. Wucherpfennig and J.J. Chou. The structural basis for intramembrane assembly of an activating immunoreceptor complex. Nat Immunol; 11 (2010):1023–9. A. Weiss and D.R. Littman. Signal transduction by lymphocyte antigen receptors. Cell; 76 (1994): 263–74. L.L. Lanier, C. Chang, H. Spits and J.H. Phillips. Expression of cytoplasmic CD3 epsilon proteins in activated human adult natural killer (NK) cells and CD3 gamma, delta, epsilon complexes in fetal NK cells. Implications for the relationship of NK and T lymphocytes. J Immunol; 149 (1992):1876–80. D.J. Leahy. A structural view of CD4 and CD8. FASEB J; 9 (1995):17–25. W. Gorczyca, Z.Y. Sun, W. Cronin, et al. Immunophenotypic pattern of myeloid populations by flow cytometry analysis. Methods Cell Biol; 103 (2011):221–66. M. Tabbekh, M. Mokrani-Hammani, G. Bismuth and F. Mami-Chouaib. T-cell modulatory properties of CD5 and its role in antitumor immune responses. Oncoimmunology; 2 (2013):e22841. R. Berland and H.H. Wortis. Origins and functions of B-1 cells with notes on the role of CD5. Ann Rev Immunol; 20 (2002):253–300. 03

84. R.E. Ware, R.M. Scearce, M.A. Dietz, et al. Characterization of the surface topography and putative tertiary structure of the human CD7 molecule. J Immunol; 143 (1989):3632–40. 85. D.C. Barral and M.B. Brenner. CD1 antigen presentation: how it works. Nat Rev Immunol; 7 (2007):929–41. 86. H. Spits, B. Blom, A.C. Jaleco, et al. Early stages in the development of human T, natural killer and thymic dendritic cells. Immunol Rev; 165 (1998): 75–86. 87. J.T. Elder, N.J. Reynolds, K.D. Cooper, et al. CD1 gene expression in human skin. J Dermatol Sci; 6 (1993):206–13. 88. H.J. Pegram, D.M. Andrews, M.J. Smyth, P.K. Darcy and M.H. Kershaw. Activating and inhibitory receptors of natural killer cells. Immunol Cell Bio; 89 (2011):216–24. 89. A. Poli, T. Michel, M. Thérésine, et al. CD56bright natural killer (NK) cells: an important NK cell subset. Immunology; 126 (2009):458–65. 90. H. Kared, S. Martelli, T.P. Ng, et al. CD57 in human natural killer cells and T-lymphocytes. Cancer Immunol Immunother; 65 (2016):441–52. 91. V. Varbanova, E. Naumova and A. Mihaylova. Killercell immunoglobulin-like receptor genes and ligands and their role in hematologic malignancies. Cancer Immunol Immunother; 65 (2016):427–40. 92. I. Morita, S. Kakuda, Y. Takeuchi, T. Kawasaki and S. Oka. HNK-1 (human natural killer-1) glyco-epitope is essential for normal spine morphogenesis in developing hippocampal neurons. Neuroscience; 164 (2009):1685–94. 93. S.A. Cassidy, K.S. Cheent and S.I. Khakoo. Effects of peptide on NK cell-mediated MHC I recognition. Front Immunol; 5 (2014):133. 94. A.K. Purdy and K.S. Campbell. Natural killer cells and cancer: regulation by the killer cell Ig-like receptors (KIR). Cancer Biol Ther; 8 (2009):2211–20. 95. P. Le Bouteiller, J. Tabiasco, B. Polgar, et al. CD160: a unique activating NK cell receptor. Immunol Lett; 138 (2011):93–6. 96. F.T. Liu, J. Giustiniani, T. Farren, et al. CD160 signaling mediates PI3K-dependent survival and growth signals in chronic lymphocytic leukemia. Blood; 115 (2010):3079–88. 97. A. Llibre, P. Klenerman and C.B. Willberg. Multifunctional lectin-like transcript-1: a new player in human immune regulation. Immunol Lett; 177 (2016):62–9. 98. C. Morimoto and S.F. Schlossman. The structure and function of CD26 in the T-cell immune response. Immunol Rev; 161 (1998):55–70. 99. M. Metzemaekers, J. Van Damme, A. Mortier and P. Proost. Regulation of chemokine activity – a focus on the role of dipeptidyl peptidase IV/CD26. Front Immunol; 7 (2016):483. 19:42:15

Chapter 2: Antigens

100. C. van Kooten and J. Banchereau. CD40-CD40 ligand. J Leukoc Biol; 67 (2000):2–17. 101. G. Pedreza-Alva and Y. Rosenstein. CD43 – one molecule, many tales to recount. Signal Trans; 7 (2007):372–85. 102. K. Futosi, S. Fodor and A. Mócsai. Neutrophil cell surface receptors and their intracellular signal transduction pathways. Int Immunopharmacol; 17 (2013):638–50. 103. V.C. Ridger, B.E. Wagner, W.A. Wallace and P.G. Hellewell. Differential effects of CD18, CD29, and CD49 integrin subunit inhibition on neutrophil migration in pulmonary inflammation. J Immunol; 166 (2001):3484–90. 104. J.H. Wang, A. Smolyar, K. Tan, et al. Structure of a heterophilic adhesion complex between the human CD2 and CD58 (LFA-3) counter receptors. Cell; 97 (1999):791–803. 105. D. Wipfler, G.V. Srinivasan, H. Sadick, et al. Differentially regulated expression of 9-O-acetyl GD3 (CD60b) and 7-O-acetyl-GD3 (CD60c) during differentiation and maturation of human T and B lymphocytes. Glycobiology; 21 (2011):1161–72. 106. M. Kuroki, H. Abe, T. Imakiirei, et al. Identification and comparison of residues critical for celladhesion activities of two neutrophil CD66 antigens, CEACAM6 and CEACAM8. J Leukoc Biol; 70 (2001):543–50. 107. M.J. Nielsen, H.J. Møller and S.K. Moestrup. Hemoglobin and heme scavenger receptors. Antioxid Redox Signal; 12 (2010):261–73. 108. A. Bolhassani and S. Rafati. Mini-chaperones: potential immuno-stimulators in vaccine design. Hum Vaccine Immunother; 9 (2013):153–61. 109. D. Siegmund, I. Lang and H. Wajant. Cell deathindependent activities of the death receptors CD95, TRAILR1, and TRAILR2. FEBS J; 8 (2017):1131–59. 110. K. Verstraete, G. Vandriessche, M. Januar, et al. Structural insights into the extracellular assembly of the hematopoietic Flt3 signaling complex. Blood; 118 (2011):60–8. 111. A.A. Abd El-Ghaffar, R.A. El-Gamal, N.N. Mostafa and N.M. Abou Shady. FLT3 (CD135) and interleukin-2 receptor alpha-chain (CD25) expression in acute myeloid leukemia: improving the correspondence to FLT3 – internal tandem duplication mutation. Int J Lab Hematol; 38 (2016):e69–72. 112. K.M. Skrzypczynska, J.W. Zhu and A. Weiss. Positive regulation of lyn kinase by CD148 is required for B cell receptor signaling in B1 but not B2 B Cells. Immunity; 45 (2016):1232–44. 113. W.A. Teft, M.G. Kirchhof and J. Madrenas. A molecular perspective of CTLA-4 function. Annu Rev Immunol; 24 (2006):65–97. 114. K. Kowal, R. Silver, E. Sławińska, et al. CD163 and its role in inflammation. Folia Histochem Cytobiol; 49 (2011):365–74. 03

115. S. Divanovic, A. Trompette, L.K. Petiniot, et al. Regulation of TLR4 signaling and the host interface with pathogens and danger: the role of RP105. J Leukoc Biol; 82 (2007):265–71. 116. M.L. Ortiz-Suarez and P.J. Bond. The structural basis for lipid and endotoxin binding in RP105-MD-1, and consequences for regulation of host lipopolysaccharide sensitivity. Structure; 24 (2016):200–11. 117. R. Solari and J.E. Pease. Targeting chemokine receptors in disease – a case study of CCR4. Eur J Pharmacol; 763 (2015):169–77. 118. M.A. Mishan, N. Ahmadiankia and A.N. Bahrami. CXCR4 and CCR7: two eligible targets in targeted cancer therapy. Cell Biol Int; 40 (2016):955–67. 119. S. Florian, K. Sonneck, M. Czerny, et al. Detection of novel leukocyte differentiation antigens on basophils and mast cells by HLDA8 antibodies. Allergy; 61 (2006):1054–62. 120. D. Hatherley, S.M. Lea, S. Johnson and A.N. Barclay. Structures of CD200/CD200 receptor family and implications for topology, regulation, and evolution. Structure; 21 (2013):820–32. 121. X. Romero, N. Zapater, M. Calvo, et al. CD229 (Ly9) lymphocyte cell surface receptor interacts homophilically through its N-terminal domain and relocalizes to the immunological synapse. J Immunol; 174 (2005):7033–42. 122. Y.Y. Acosta, M.P. Zafra, G. Ojeda, et al. Biased binding of class IA phosphatidyl inositol 3-kinase subunits to inducible costimulator (CD278). Cell Mol Life Sci; 68 (2011):3065–79. 123. K. Bardhan, T. Anagnostou and V.A. Boussiotis. The PD1:PD-L1/2 pathway from discovery to clinical implementation. Front Immunol; 7 (2016):550. 124. F. Borrego. The CD300 molecules: an emerging family of regulators of the immune system. Blood; 121 (2013):1951–60. 125. D.Y. Evans, R. Serra-Moreno, R.K. Singh and J.C. Guatelli. BST-2/tetherin: a new component of the innate immune response to enveloped viruses. Trends Microbiol; 18 (2010):388–96. 126. X. Li, G. Zhang, Q. Chen, et al. CD317 Promotes the survival of cancer cells through apoptosis-inducing factor. J Exp Clin Cancer Res; 35 (2016):117. 127. J.K. Lee, S.O. Mathew, S.V. Vaidya, P.R. Kumaresan and P.A. Mathew. CS1 (CRACC, CD319) induces proliferation and autocrine cytokine expression on human B lymphocytes. J Immunol; 179 (2007): 4672–8. 128. S. Yamasaki. Clec12a: quieting the dead. Immunity; 40 (2014):309–11. 129. M. Toft-Petersen, L. Nederby, E. Kjeldsen, et al. Unravelling the relevance of CLEC12A as a cancer stem cell marker in myelodysplastic syndrome. Br J Haematol; 175 (2016):393–401.

19:42:15

35

Chapter

3

Flow Cytometry of Normal Blood, Bone Marrow and Lymphatic Tissue Anna Porwit and Marie-Christine Béné

Introduction

36

Knowledge of the levels and expression patterns of various differentiation antigens by flow cytometry (FCM) on normal haematopoietic cells present in various tissues provides a frame of reference for the recognition of abnormal differentiation patterns in disease states. Expression of the leucocyte common antigen (CD45) displayed together with the right-angle light scatter (side scatter (SSC)) is usually applied as a first step to delineate the major populations present in the examined sample (peripheral blood (PB), bone ­marrow (BM), body fluid (BF) or tissue biopsy) [1, 2] (Figure 3.1): • Erythropoietic precursors are CD45− and have low SSC (normally not seen in normal PB or lymphatic tissue); • Early haematopoietic precursors of various lineages, including CD34+ stem cells, are characterized by low/intermediate CD45 expression and low SSC (also called blast region (BR), CD45dim or ‘bermudes’ as described in detail in [1]); • Granulopoietic precursors and granulocytes have an intermediate CD45 expression and a variable SSC (intermediate to high) spanning a large part of the SSC axis; • Mature lymphocytes are characterised by low SSC and strong CD45 expression; • Monocytes have a higher SSC than lymphocytes and equally strong CD45 expression. The localisation of these subpopulations on the CD45/SSC plot can be refined by multi-colour backgating of various lineage-associated antigens together with CD45 and visualisation of cell clusters positive for given antigen combinations on the CD45/SSC plot (see Figure 3.1 and Chapter 1, Figure 1.5) using back-gating and colour coding [1]. The combination of CD14/CD16/CD11b and CD45 is an example of a panel that makes it possible to precisely localise mature cells on a CD45/SSC plot (Figure 1.5). A colour 04

coding proposed in the literature defines monocytes in green, ­granulocytes in red, lymphocytes in magenta and blasts in cyan [1]. A CD14/SSC dot plot is used to define the monocyte region as CD14+SSCintermediate and for back-gating on the CD45/SSC dot plot. Similarly, back-gating of the CD16+CD11b+/SSChigh population is applied for visualising the localization of mature neutrophil granulocytes. Lymphocytes are defined as the SSClow/CD45bright cell cluster. The BR on the CD45/ SSC dot plot is the result of the Boolean gate ‘CD45dim/ SSClow AND NOT monocytes AND NOT lymphocytes AND NOT granulocytes’ (Figure 3.1).

Antigen Expression Patterns in Normal Blood Blood immunophenotyping is one of the most common requests addressed to FCM laboratories. Complete blood counts and white blood cell (WBC) differentials are usually obtained using haematology analysers but FCM can be performed along with ­morphological evaluation as a screening to exclude haematological malignancy and in search for signs of infection, autoimmune diseases and/or various immunodeficiency states. Proposals to replace the haematology analyser/manual differential with an FCM differential have been published [3–9] (Table 3.1). The most widely used CytoDiff® reagent in the HematoFlow® system combines six antibodies in five-colour FCM and allows for the recognition of 13 WBC subsets (see also Chapter  14, Figure 14.6) [7, 10]. FCM differentials can be successfully applied to low WBC PB samples [8]. In the five-colour assay published by van de Geijn et al. [3], 10 antibodies are combined to provide a differential of 13 populations. The eight-colour assay of Cherian et al. [5] combines 10 antibodies with the DNA-binding dye Hoechst 34580 and TruCount beads to generate absolute values for eight populations: lymphocytes, granulocytes, monocytes, eosinophils, basophils, immature granulocytes, blasts and nucleated red 19:43:00

1000

1000

800

800

600

600 SSC

SSC

Chapter 3: Flow Cytometry of Normal Blood, Bone Marrow and Lymphatic Tissue

400 CD7+ T-cells : 16.68%

1000

200 0

100

101 CD7

102

200

800

103

CD14+ Monocytes : 1.74%

400

0

100

101

102

103

CD14

SSC

600 400 103

200 0

1000 800

100

101

102

CD45

101

103

600 SSC

CD45

102

CD45- : 3.53%

100

400

CD16+ granulopoietic cells: 16.15% CD16+ neutrophils: 38.77%

200 CD34+ cells : 0.74%

100

101

102

CD16+ NK cells

0

103

100

101

102

103

CD16

CD34

Figure 3.1  Main subpopulations of haematopoietic cells in normal bone marrow (BM) visualised on the CD45/SSC plot. Cell subsets are gated on the expression of characteristic markers: the pan-T antigen CD7 (violet dots, top left histogram) allows to locate the lymphocyte region where T-cells are predominant. CD34+ progenitors (dark blue dots, bottom left histogram) are located in the CD45dim region (blast region or ‘bermudes’ [1]). CD14 (green dots, top right histogram) allows to visualise mature monocytes. Granulopoietic cells are gated on their SSCintermediate/high characteristic and CD16 expression (brown and red dots, bottom right histogram). CD45− cells (orange dots), are erythroblasts. The larger middle histogram shows a back-gating of all these subsets on the CD45/SSC plot (‘cartography’) of normal BM.

blood cells (Figure 3.2). The 10-colour assay combining 13 antibodies proposed by Melzer et al. [11] defines neutrophils, eosinophils, classical and non-classical monocytes, B-lymphocytes, CD4+ T-lymphocytes, CD4 regulatory T-cells (Tregs), CD8+ T-lymphocytes, double-positive T-lymphocytes and natural killer (NK)cells (Table 3.2).

Lymphocytes Reference Values for Lymphocyte Subsets in Blood Examples of FCM panels that have been proposed to evaluate lymphocyte subsets in PB are given in Table 3.3. Some laboratories use separate tubes for the 04

evaluation of B-cells and T- and NK-cells, respectively [12, 17]. Others utilise panels with combinations of antibodies against B-, T- and NK-cells in a single tube [13, 18, 19]. To combine several antibodies in one tube, antibodies to antigens that are not seen on the same cell populations in normal conditions may be combined in the same fluorochrome (see examples in Chapter  7, Table. 7.1). Basic screening panels usually aim to evaluate the frequency of B-cells (CD19+, CD20+) and to confirm the normal distribution of κ and λ surface immunoglobulin (sIg) light chain expression (­normal κ:λ ratio between 1 and 3, range 0.8–4). In addition, the vast majority of PB normal B-lymphocytes lack CD5, CD10 and/or CD23 expression. These panels also examine the frequency of CD3+ T-cells, which constitute the majority of PB lymphocytes, and 19:43:00

37

Chapter 3: Flow Cytometry of Normal Blood, Bone Marrow and Lymphatic Tissue

Table 3.1  Examples of flow cytometry panels for blood differential

Flow cytometer/ Ref #

Reagents

Leukoflow

CD4 CD14 FITC

CD16 CD34 CD56 PE

CD19 ECD

CD45 CD138 PE-PC5

CD3 CD71 PE-PC7

FC500 [3]

CytoDiff

CD36 FITC

CD2 CD294 PE

CD19 ECD

CD16 PE-Cy5

CD45 PE-Cy7

FC500 [4, 7–10]

Björnsson et al.

CD36 FITC

CD203 CD138 PE

CD45 ECD

CD16 CD56 PE-Cy5

DRAQ5

FC500 [6]

Cherian et al.

CD16 CD19 FITC

CD123 PE

CD33 CD64 PE-Cy5

CD34 CD117 APC

CD45 APC-Cy7

HLA-DR PE-CY7

CD38 A594

Hoechst 34580

Melzer et al.

CD8 CD14 CD19 FITC

CD69 PE

CD25 ECD

HLA-DR APC

CD38 PE-CY5.5

CD16 CD56 PE-CY7

CD127 APC-Ax700

CD45 PBl

04

Method

19:43:00

38

LSRII [5]

CD4 APC H7

CD3 KrO

Navios [11]

Chapter 3: Flow Cytometry of Normal Blood, Bone Marrow and Lymphatic Tissue

SSC-H

HLA-DR PE-Cy7

SSC-H

CD16+19 FITC

CD33+64 PE-CY5

CD123 PE

CD45 APC-CY7

CD33+64 PE-CY5

(A)

CD16+19 FITC

HLA-DR PE-Cy7

(E) CD45 APC-Cy7

CD123 PE

CD123 PE

(B)

CD16+19 FITC

HLA-DR PE-Cy7

SSC-H

SSC-H

CD34+117 APC

CD16+19 FITC

CD45 APC-Cy7

CD33+64 PE-Cy5

(C)

CD45 APC-Cy7

HLA-DR PE-Cy7

CD45 APC-Cy7

CD33+64 PE-Cy5

F3C-H

CD33+64 PE-Cy5

CD16+19 FITC

CD33+64 PE-Cy5

(D)

SSC-H

CD45 APC-Cy7

HO-B PB

Figure 3.2  Example of eight-colour flow cytometry blood differential (from Cherian et al.(5) with permission). (A) A gate was drawn around TruCount beads on the CD123 versus side scatter (SSC) histogram as shown. Then monocytes (pink) were isolated from the viable nucleated cells using CD33/CD64, HLA-DR and CD16+CD19. (B) Monocytes were then excluded, and of the remaining cells, basophils (purple) and plasmacytoid dendritic cells (light blue) were identified and then excluded using expression of CD123 (bright on both populations) and HLA-DR, which is positive on plasmacytoid dendritic cells and negative on basophils. (C) Of the remaining cells, CD45 and SSC were used to generate a rough lymphoid cell gate that was purified by removing CD33+CD64+ events. The lymphocytes were then separated into B-cells (CD19+, HLA-DR+ shown in green), natural killer -cells (CD16+, HLA-DR− shown in dark blue) and T-cells (lymphoid cells lacking CD19 and CD16, shown in red), which were combined to give the ­lymphocyte enumeration. (D) The lymphocytes were then excluded leaving mature neutrophils, immature granulocytes, eosinophils, blasts and nucleated red blood cells. Mature neutrophils (green) were identified on the basis of CD16 expression. Eosinophils (orange) were ­separated from immature granulocytes and blasts on the basis of CD33+CD64 and CD45 expression in conjunction with SSC. Eosinophils were then excluded and the immature granulocytes were isolated from the remaining cells on the basis of SSC versus expression of CD33+CD64 with forward scatter low debris being excluded from the immature granulocyte cell gate (this latter plot is not shown). (E) Blasts were identified using CD45 versus SSC characteristics in conjunction with expression of CD34+CD117 (blue). Finally, nucleated red cells (aqua) were isolated from the remaining nucleated cells on the basis of expression of DNA-binding dye versus forward scatter.

the CD4+:CD8+ ratio (­normally 1–3, range 0.8–6). NK-cell-related antigens such as CD16, CD56 and/or CD57 may also be included. An example of a normal blood analysis in such a screening panel is shown in Figure 3.3. This kind of panel allows a quick evaluation 04

of samples with a normal distribution of lymphocyte subsets. It is also useful to detect clonal/abnormal B-cell populations or an abnormal distribution of T- and/or NK-cells. Depending on the type of screening panel, preliminary results and clinical requests, 19:43:00

39

Chapter 3: Flow Cytometry of Normal Blood, Bone Marrow and Lymphatic Tissue

Table 3.2  Reference values for blood leucocyte subsets absolute numbers (109/L) according to [11]

Distributions of Leukocyte Subset Counts in Males and Females Males

Females

Cell types

Median (P02.5-P97.5)

Median (P02.5-P97.5)

P values

WBC

5.6500 (3.4270-9.0290)

5.4700 (3.5900-8.8765)

0.564

Neutrophils

3.3524 (1.6204-6.1134)

3.2242 (1.7868-5.8517)

0.296

Eosinophils

0.1069 (0.0301-0.3738)

0.1016 (0.0301-0.4506)

0.420

Monocytes

0.2630 (0.1087-0.5525)

0.2320 (0.1070-0.4392)

0.000

L

nonclassical m.

0.0244 (0.0056-0.0665)

0.0167 (0.0059-0.0510)

0.000

L

classical m.

0.2262 (0.0845-0.4908)

0.2084 (0.0945-0.3870)

0.005

Lymphocytes L

T-lymphocytes

1.5232 (0.6918-2.8322)

1.6006 (0.7372-2.8969)

0.035

0.9460 (0.3585-2.0142)

1.0810 (0.4040-2.0754)

0.001

 L Tc cells

0.1541 (0.0328-0.5404)

0.1627 (0.0432-04375)

0.806

 L DPT

0.0058 (0.0013-0.0360)

0.0074 (0.0019-0.0268)

0.058

L

  Th cells

0.6773 (0.2436-1.4411)

0.8072 (0.2975-1.6976)

0.000

  L regulatory T cells

0.0549 (0.0210-0.1298)

0.0604 (0.0229-0.1336)

0.008

 L NKT cells L

B-lymphocytes

 L CD19+ B cells L

NK cells

0.0868 (0.0207-0.3045)

0.0921 (0.0212-0.3324)

0.891

0.1833 (0.0694-0.4573)

0.2084 (0.0679-0.5077)

0.000

0.1563 (0.0516-0.4311)

0.1857 (0.0555-0.4718)

0.000

0.2885 (0.1114-0.6908)

0.2520 (0.0817-0.5960)

0.000

Median (2.5% percentile to 97.5% percentile) for leukocyte subset counts in males and females. Differences between males and females are tested using the Mann-Whitney U test. Outliers were not removed for the analysis. Cell counts in 109/I. Abbreviations: Tc: T cytotoxic, DPT: Double positive T-cells, Th T helper, WBC, white blood cell, m. monocytes, P02.5, 2.5% percentile, P97.5, 97.5% percentile.

40

further immunophenotyping panels can be applied, as described in detail in the respective chapters dealing with B-cell lymphomas (Chapter 7), T-cell lymphomas (Chapter 9) and acute leukaemias (Chapters 6 and 11). Each laboratory should establish its own reference values for lymphocyte subsets. Examples of normal reference values from the literature are given in Tables 3.2 and 3.4 [11, 20, 21]. In a large study [11] from Germany (Table 3.2) dealing with healthy subjects aged 40 years, significantly higher cell counts for monocytes (P75% positive

>95% negative

75

CD19

>70% positive

>95% negative

96

CD56

5% of the total PC population define a good prognostic subgroup of patients, with prolonged progression-free survival (PFS) and overall survival (OS) [20]. Furthermore, examination of peripheral blood samples can detect circulating neoplastic PCs in approximately 80% of patients with PCM at presentation and the level of circulating neoplastic PCs in newly diagnosed patients is a predictor of PFS and OS [21, 22]. 19:51:58

133

Chapter 8: Plasma Cell Myeloma and Related Disorders

Quantification of circulating neoplastic PCs may also be useful in predicting risk of transformation of MGUS and smouldering myeloma cases [23–25].

Phenotypic Characteristics with Prognostic Significance

134

The phenotype of neoplastic PC populations can be heterogeneous, with differences seen both between patients and within the tumour population in a single patient. Several groups have investigated the relationship between immunophenotype and a range of genetic abnormalities commonly seen in myeloma. The level of expression of a number of antigens has been reported to correlate with key genetic lesions and to have prognostic significance. Expression of CD117 has been shown to be associated with improved PFS and OS in several independent studies [26, 27]. Studies in large cohorts of patients have shown that cases of non-hyperdiploid PCM have a higher frequency of CD28 and CD20 expression and a relatively high incidence of CD117 and CD56 negativity. Deletions of 13q and chromosomal rearrangements involving immunoglobulin heavy chains (IGHs) genes, particularly t(4;14), demonstrate a strong association to CD117 negativity. CD28 expression has been shown to correlate with deletion of 17p and t(14;16) and is associated with shorter PFS and OS [28]. CD20+CD56− cases are associated with t(11;14) and extramedullary disease/PC leukaemia, while cases of other IGH translocations are almost all CD20− [29, 30]. CD81 expression has been shown to be associated with poor prognosis in patients, with positivity related to shorter PFS and OS [16]. The same study also analysed a cohort of patients with smouldering myeloma and demonstrated a shorter time to progression in those cases with CD81 expression. CD45 is frequently downregulated in neoplastic PCs and lack of CD45 expression has been shown to impact on OS in a study of patients with PCM receiving high-dose therapy. Patients without CD45 expression had a median survival of 42 months versus median not reached in the CD45+ cohort [31]. All these data support the potential application of FCM immunophenotyping to assist in the identification of high-risk patients. Newer studies to assess the significance of immunophenotypic characteristics in the era of novel therapies will be of interest. 09

Staging of Plasmacytoma Isolated plasmacytoma of bone is currently treated with locally targeted radiotherapy but there is significant heterogeneity in this patient group in terms of progression and outcome. Almost two-thirds of patients develop progressive disease in the form of further plasmacytomas or systemic PCM. Outcome for the patients who do not progress is good, with most staying disease-free for at least 10 years. High-sensitivity FCM can be applied to analyse the BM of patients with plasmacytoma and it demonstrates that a high proportion of patients have low levels of neoplastic PCs present in the BM [32, 33]. BM involvement is strongly predictive of outcome with progression in >70% of cases with neoplastic PCs present, compared to only 8%–12% in those individuals without BM disease. Evaluation of BM aspirates by FCM is critical to determine prognosis in these patients. It also offers the potential for improved management of patients with plasmacytoma, with clinical trials currently in development to use FCM analysis to identify a cohort of patients who may benefit from early systemic therapy.

Identification and Quantitative Assessment of Therapeutic Targets High-sensitivity FCM can be applied to both myeloma cell lines and patient samples to interrogate many antigens simultaneously in the search for novel therapeutic targets. Therapeutic monoclonal antibodies already in use, including daratumumab (anti-CD38) and elotuzumab (anti-CD319), have shown promising results in myeloma clinical trials. Quantitative surface expression data assessed by FCM show that there is a variability of expression of both CD319 and CD38 when comparing diagnostic categories such as PCM, MGUS and cases with normal PCs (see Figure 8.4). There is also considerable heterogeneity between patients which could affect efficacy [34]. As the choice of monoclonal antibody therapies increases, FCM may be an important tool in assessing the best fit for individual patients based on quantitation of antigen expression.

MRD Analysis Response to therapy in PCM has historically relied upon assessment of paraprotein or serum free light chain concentration, in conjunction with clinical parameters such as renal function and degree of 19:51:58

Chapter 8: Plasma Cell Myeloma and Related Disorders

25000

160000

Median fluorescence intensity

Median fluorescence intensity

140000 20000

15000

10000

5000

120000 100000 80000 60000 40000 20000

0

Normal

Myeloma

MGUS

0

WM

Normal

Myeloma

MGUS

Diagnosis

Diagnosis

CD319 expression

CD38 expression

WM

Figure 8.4  Flow cytometry allows quantitative analysis of antigens which may relate to treatment efficacy. Assessment of 32 patients with myeloma/monoclonal gammopathy of undetermined significance /Waldenström’s macroglobulinaemia and 17 normal bone marrow samples showed significant heterogeneity in expression of CD38 and CD319, both of which are target antigens for therapeutic antibodies. About 66% of myeloma cases had CD319 expression below normal levels and myeloma plasma cells (PCs) showed lower levels of CD38 expression than normal PCs in all cases. This heterogeneity in surface expression seen could potentially affect efficacy of antibody treatment.

anaemia. Although all these factors are an important part of patient response, they do not offer a direct assessment of disease burden. There can be a delay in the reduction of particular paraprotein isotypes with longer half-lives, and BM recovery times can vary between patients. Treatment regimens have also become considerably more complex, and in order to assess individual components of therapeutic schedules with multiple agents, a direct assessment of the neoplastic population is required.

Applicability of FCM to Myeloma MRD FCM assays such as those previously described can readily discriminate between neoplastic and normal PCs and are therefore applicable in the assessment of post-treatment BM samples. An example of an MRD analysis can be seen in Figure 8.5, and Table 8.2 gives examples of recently published 8- and 10-colour antibody combinations [35–37]. It has become clear over recent years that this type of assessment is highly informative and has powerful prognostic significance. Numerous studies have demonstrated that assessment of MRD by FCM represents a more sensitive measure of response than conventional morphological criteria [38–43]. It has been shown to be an independent predictor of PFS and OS in many different patient groups and clinical settings. Immunophenotypic complete response (i.e., undetectable MRD at the 10−4 level in 09

the BM) has been shown to be one of the most relevant prognostic factors for patients undergoing autologous stem-cell transplantation, as well as in non-transplant eligible patients treated with novel agents [40–43]. Accurate quantification of residual disease is critical as it has recently been demonstrated that the level of MRD is a more powerful predictor of PFS and OS than a categorical approach with MRD-negativity based on a threshold [44]. Reproducible quantification is an absolute requirement for comparison of results across different centres in clinical trials and different treatment strategies. There has been a huge increase in novel treatments for myeloma over the past decade and current therapeutic approaches result in an OS of more than five years for most newly diagnosed patients. This means that randomised Phase III trials now take several years to show benefit when measured by the conventional end points, such as OS. FCM represents a potential surrogate for evaluation of response which would allow accelerated assessment of novel therapies.

The Importance of International Consensus The need for alternative methods for assessment of response in myeloma and the demonstration of the importance of accurate quantification of MRD rather than a categorical positive/negative result has increased the need for a robust approach to myeloma MRD 19:51:58

135

Chapter 8: Plasma Cell Myeloma and Related Disorders

0

–2,460 M –807

103 0

103

104

M –304

105

CD138 APC R660/20-A

50

0 102

103

104

105

50

CD45 APC-H7 R780/60-A

CD45 CD19 PerCP-Cy5-5 B670LP-A

CD19

CD19 PerCP-Cy5-5 B670LP-A

100

150

–2,460 0

CD138

105 104 103 0

–1,224 M –304

200

105 104 103 0

–1,224 0 102

103

104

105

CD56 BV421 V450/50-A

M –304

CD56

200 250 (x 1,000)

100 150 FSC-A

FSC CD19 PerCP-Cy5-5 B670LP-A

CD38

103

104

(x 1,000)

104

250

105

SSC-A

CD38 PE-Cy7 B780/60-A

CD38 PE-Cy7 B780/60-A

105

0 102

103

104

105 104 103 0

–1,224 M –304

105

CD27 BV510 V510/50-A

CD27

0 102

103

104

105

CD81 FITC B530/30-A

CD81

Normal

Neoplastic

Figure 8.5  Representative flow plots showing minimal residual disease (MRD) in a plasma cell myeloma (PCM) patient. MRD analysis was conducted for a patient with PCM three months post autologous transplant. Total events collected were in excess of 1 ­million. Plasma cells represent 0.6% of leucocytes, of which 1.4% (0.008% of leucocytes) have a neoplastic phenotype, with strong CD56 expression and weaker expression of CD27 and CD81. Table 8.2  Recently published 8- and 10-colour antibody combinations for assessment of minimal residual disease in multiple myeloma PETHEMA/ GEM2010 MAS65 MRD STUDY

FITC

PE

PerCP-Cy5.5

PE-Cy7

APC

APC-H7

PacB

OC-515

CD38

CD56

CD27

CD19

CD117

CD81

CD45

CD138

UK MYELOMA XI TRIAL

FITC

PE

PerCP-Cy5.5

PE-Cy7

APC

APC-H7

BV421

BV510

CD81

CD117

CD19

CD38

CD138

CD45

CD56

CD27

EUROFLOW

FITC

PE

PerCP-Cy5.5

PC7

APC

AF700

APC-C750

BV421

8-colour

CD38

CD56

CD45

CD19

CD117

CD81

CD138

CD27

CD38

CD56

CD45

CD19

cIgκ

cIgλ

CD138

CD27

CD38

CD56

CD45

CD19

cIgλ

CD81

CD138

CD27

10-colour

136

which is applicable across different patient cohorts, therapeutic strategies and laboratories. To be clinically relevant, this assay needs to be backwards compatible with established assays and needs to be applicable in a significant number of central laboratories. 09

cIgκ

BV510

BV605

CD117

The International Clinical Cytometry Society and European Society for Clinical Cell Analysis recognised the need for a consensus approach and this led to the publication of a set of manuscripts detailing recommended approaches to data acquisition, analysis and 19:51:58

-807

105 CD81 FITC B530/30-A 0 103 104

CD81

NK-cells

-807

Q2

B-cells

0 103 104 105 CD56 BV421 V450/50-A

-807

B-progenitors

101 102 103 104 105 CD45 APC-H7 R780/60-A

CD45

0

103

104

CD56

-807 -1.1639

103

0

104

105

105 104 103

CD117 PE 8585/42-A

Mast cells

0

Myeloid progenitors

CD117

CD117

SSC

-807

105

200 250 (x1,000)

104

150 SSC-A

103

100

CD117 PE 8585/42-A

50

0

-807

CD45

CD45 APC-H7 R780/60-A

105

Q1

-807

Mononuclear cells

CD19 PerCP-Cy5-5 B670LP-A 0 103 104 105

CD19

Chapter 8: Plasma Cell Myeloma and Related Disorders

50

100

150 SSC-A

CD27 V500 V510/50-A

CD27

200

250 (x1,000)

SSC

Figure 8.6  Internal controls which have utility in sample quality assessment. The antigens contained within the plasma cell (PC) minimal residual disease assay can be used to assess the presence of bone marrow (BM) elements, such as myeloid and B-lymphoid progenitors and mast cells. Gated mononuclear cells following exclusion of PCs are shown in the first plot. The subsequent plots demonstrate which populations can be assessed in order to determine the quality of the BM specimen.

quality control [11, 12, 45]. Increased sensitivity provided by the use of a greater number of colours simultaneously, more sophisticated gating strategies and acquisition of 1–5 million events per assay mean this generation of FCM assays can reach a limit of detection of 0.001% of leucocytes.

Pros and Cons of MRD Analysis by FCM Flow cytometric assessment of MRD is widely applicable, with virtually all patients assessable using the same combination of antigens developed around disease-specific immunophenotypes. It also benefits from the ability to assess the composition of the whole sample, thereby allowing a check of the quality of the sample cellularity (see Figure 8.6). This is critical to 09

eliminate false-negative results which could be generated due to haemodiluted aspirate samples. In contrast, quantitative real-time polymerase chain reaction (qPCR) approaches, including allelespecific oligonucleotide polymerase chain reaction (ASO-PCR), require additional checks for sample quality and have, until recently, required the development of a specific assay for each patient. The development of patient-specific primers is both expensive and time-consuming. Newer high-throughput sequencing methods with potentially higher sensitivity than FCM assays are now becoming more accessible but are still costly and represent a not fully validated approach which will require extensive prospective application before it is readily applicable in the routine setting. 19:51:58

137

Chapter 8: Plasma Cell Myeloma and Related Disorders

Flow-MRD assays have become more sensitive in recent years (10−5) and are directly quantitative with the same lower limits of detection and quantification in every case. At this time, FCM still represents the most applicable approach to MRD testing in myeloma and other PC disorders.

References

S.H. Swerdlow, E. Campo, N.L. Harris, et al. WHO classification of tumours of haematopoietic and lymphoid tissues. (4th ed. Lyon, Internat. Agency for Research on Cancer; 2008). 2. S.V. Rajkumar, Updated diagnostic criteria and staging system for multiple myeloma. Am Soc Clin Oncol Educ Book, 36 (2016), e418–23. 3. S.V. Rajkumar, O. Landgren and M-V. Mateos. Smoldering multiple myeloma. Blood, 125 (2015), 3069–75. 4. M. Drayson, L.X. Tang, R. Drew, G.P. Mead, et al. Serum free light-chain measurements for identifying and monitoring patients with nonsecretory multiple myeloma. Blood, 97 (2001), 2900–2. 5. V. Simeon, K. Todoerti, F. Rocca, et al. Molecular classification and pharmacogenetics of primary plasma cell leukemia: an initial approach toward precision medicine. Int J Mol Sci, 16 (2015), 17514–34. 6. R. Soutar, H. Lucraft, G. Jackson, et al. Guidelines on the diagnosis and management of solitary plasmacytoma of bone and solitary extramedullary plasmacytoma. Br J Haematol, 124 (2004), 717–26. 7. M.A. Gertz, Immunoglobulin light chain amyloidosis: 2011 update on diagnosis, risk-stratification, and management. Am J Hematol, 86 (2011), 180–6. 8. A. Dispenzieri, POEMS syndrome: 2014 update on diagnosis, risk-stratification, and management. Am J Hematol, 89 (2014), 214–23. 9. A.C. Rawstron, R.M. de Tute, J. Haughton and R.G. Owen, Measuring disease levels in myeloma using flow cytometry in combination with other laboratory techniques: lessons from the past 20 years at the Leeds Haematological Malignancy Diagnostic Service: MRD DETECTION BY FLOW CYTOMETRY AND OTHER LABORATORY TECHNIQUES. Cytometry B Clin Cytom, 90 (2016), 54–60. 10. A.C. Rawstron, A. Orfao, M. Beksac, et al. Report of the European Myeloma Network on multiparametric flow cytometry in multiple myeloma and related disorders. Haematologica, 93 (2008), 431–8. 1 1. M. Stetler-Stevenson, B. Paiva, L. Stoolman, et al. Consensus guidelines for myeloma minimal residual disease sample staining and data acquisition. Cytometry B Clin Cytom, 90 (2016), 26–30. 12. M. Arroz, N. Came, P. Lin, et al. Consensus guidelines on plasma cell myeloma minimal residual disease analysis and reporting. Cytometry B Clin Cytom, 90 (2016), 31–9. 1.

138

09

13. B. Paiva, J. Almeida, M. Pérez-Andrés, et al. Utility of flow cytometry immunophenotyping in multiple myeloma and other clonal plasma cell-related disorders. Cytometry B Clin Cytom, 78 (2010), 239–52. 14. J.J.M. van Dongen, L.L. Lhermitte, S. Böttcher, et al. EuroFlow antibody panels for standardized n-dimensional flow cytometric immunophenotyping of normal, reactive and malignant leukocytes. Leukemia, 26 (2012), 1908–75. 15. D. Liu, P. Lin, Y. Hu, et al. Immunophenotypic heterogeneity of normal plasma cells: comparison with minimal residual plasma cell myeloma. J Clin Pathol, 65 (2012), 823–9. 16. B. Paiva, N-C. Gutiérrez, X. Chen, et al. Clinical significance of CD81 expression by clonal plasma cells in high-risk smoldering and symptomatic multiple myeloma patients. Leukemia, 26 (2012), 1862–9. 17. P.R. Tembhare, C.M. Yuan, D. Venzon, et al. Flow cytometric differentiation of abnormal and normal plasma cells in the bone marrow in patients with multiple myeloma and its precursor diseases. Leuk Res, 38 (2014), 371–6. 18. K.R.M. Raja, L. Kovarova and R. Hajek. Review of phenotypic markers used in flow cytometric analysis of MGUS and MM, and applicability of flow cytometry in other plasma cell disorders. Br J Haematol, 149 (2010), 334–51. 19. N. Robillard, G. Jego, C. Pellat-Deceunynck, et al. CD28, a marker associated with tumoral expansion in multiple myeloma. Clin Cancer Res, 4 (1998), 1521–6. 20. B. Paiva, M-B. Vidriales, G. Mateo, et al. The persistence of immunophenotypically normal residual bone marrow plasma cells at diagnosis identifies a good prognostic subgroup of symptomatic multiple myeloma patients. Blood, 114 (2009), 4369–72. 21. B. Paiva, M. Pérez-Andrés, M-B. Vídriales, et al. Competition between clonal plasma cells and normal cells for potentially overlapping bone marrow niches is associated with a progressively altered cellular distribution in MGUS vs myeloma. Leukemia, 25 (2011), 697–706. 22. B. Paiva, T. Paino, J-M. Sayagues, et al. Detailed characterization of multiple myeloma circulating tumor cells shows unique phenotypic, cytogenetic, functional, and circadian distribution profile. Blood, 122 (2013), 3591–8. 23. S. Kumar, S.V. Rajkumar, R.A. Kyle, et al. Prognostic value of circulating plasma cells in monoclonal gammopathy of undetermined significance. J Clin Oncol, 23 (2005), 5668–74. 24. W.I. Gonsalves, S.V. Rajkumar, A. Dispenzieri, et al. Quantification of circulating clonal plasma cells via multiparametric flow cytometry identifies patients with smoldering multiple myeloma at high risk of progression. Leukemia. (26 July 2016), doi: 10.1038/ leu.2016.205. 19:51:58

Chapter 8: Plasma Cell Myeloma and Related Disorders

25. G. Bianchi, R.A. Kyle, D.R. Larson, et al. High levels of peripheral blood circulating plasma cells as a specific risk factor for progression of smoldering multiple myeloma. Leukemia, 27 (2013), 680–5. 26. R. Bataille, C. Pellat-Deceunynck, N. Robillard, et al. CD117 (c-kit) is aberrantly expressed in a subset of MGUS and multiple myeloma with unexpectedly good prognosis. Leuk Res, 32 (2008), 379–82. 27. M. Schmidt-Hieber, M. Perez-Andres, B. Paiva, et al. CD117 expression in gammopathies is associated with an altered maturation of the myeloid and lymphoid hematopoietic cell compartments and favorable disease features. Haematologica, 96 (2011), 328–32. 28. G. Mateo, M.A. Montalbán, M-B. Vidriales, et al. Prognostic value of immunophenotyping in multiple myeloma: a study by the PETHEMA/GEM cooperative study groups on patients uniformly treated with high-dose therapy. J Clin Oncol, 26 (2008), 2737–44. 29. G. An, Y. Xu, L. Shi, et al. t(11;14) multiple myeloma: a subtype associated with distinct immunological features, immunophenotypic characteristics but divergent outcome. Leuk Res, 37 (2013), 1251–7. 30. G. Mateo, M. Castellanos, A. Rasillo, et al. Genetic abnormalities and patterns of antigenic expression in multiple myeloma. Clin Cancer Res, 11 (2005), 3661–7. 31. P. Moreau, N. Robillard, H. Avet-Loiseau, et al. Patients with CD45 negative multiple myeloma receiving high-dose therapy have a shorter survival than those with CD45 positive multiple myeloma. Haematologica, 89 (2004), 547–51. 32. B. Paiva, M. Chandia, M-B. Vidriales, et al. Multiparameter flow cytometry for staging of solitary bone plasmacytoma: new criteria for risk of progression to myeloma. Blood, 124 (2014), 1300–3. 33. Q.A. Hill, A.C. Rawstron, R.M. de Tute and R.G. Owen. Outcome prediction in plasmacytoma of bone: a risk model utilizing bone marrow flow cytometry and light-chain analysis. Blood, 124 (2014), 1296–9. 34. J. Shingles, R.M. de Tute, A.C. Rawstron and R.G. Owen. CD319 and CD38 expression patterns in Waldenstrom’s macroglobulinaemia (WM), myeloma and MGUS: implications for antibody therapy. Br J Haematol, 173S1 (2016), 79, Abstract 190. 35. J. Flores-Montero, R. de Tute, B. Paiva, et al. Immunophenotype of normal vs. myeloma plasma cells: toward antibody panel specifications for MRD detection in multiple myeloma. Cytometry B Clin Cytom, 90 (2016), 61–72.

36. E. Domingo, C. Moreno, A. Sánchez-Ibarrola, et al. Enhanced sensitivity of flow cytometry for routine assessment of minimal residual disease. Haematologica, 95 (2010), 691–2. 37. B. Paiva, M-T. Cedena, N. Puig, et al. Minimal residual disease monitoring and immune profiling in multiple myeloma in elderly patients. Blood, 127 (2016), 3165–74. 38. J.F. San Miguel, J. Almeida, G. Mateo, et al. Immunophenotypic evaluation of the plasma cell compartment in multiple myeloma: a tool for comparing the efficacy of different treatment strategies and predicting outcome. Blood, 99 (2002):1853–6. 39. A.C. Rawstron, F.E. Davies, R. DasGupta, et al. Flow cytometric disease monitoring in multiple myeloma: the relationship between normal and neoplastic plasma cells predicts outcome after transplantation. Blood, 100 (2002), 3095–100. 40. A.C. Rawstron, J.A. Child, R.M. de Tute, et al. Minimal residual disease assessed by multiparameter flow cytometry in multiple myeloma: impact on outcome in the Medical Research Council Myeloma IX Study. J Clin Oncol, 31 (2013), 2540–7. 41. B. Paiva, M-B. Vidriales, J. Cerveró, et al. Multiparameter flow cytometric remission is the most relevant prognostic factor for multiple myeloma patients who undergo autologous stem cell transplantation. Blood, 112 (2008), 4017–23. 42. B. Paiva, J. Martinez-Lopez, M-B. Vidriales, et al. Comparison of immunofixation, serum free light chain, and immunophenotyping for response evaluation and prognostication in multiple myeloma. J Clin Oncol, 29 (2011), 1627–33. 43. B. Paiva, N.C. Gutiérrez, L. Rosiñol, et al. Highrisk cytogenetics and persistent minimal residual disease by multiparameter flow cytometry predict unsustained complete response after autologous stem cell transplantation in multiple myeloma. Blood, 119 (2012), 687–91. 44. A.C. Rawstron, W.M. Gregory, R.M. de Tute, et al. Minimal residual disease in myeloma by flow cytometry: independent prediction of survival benefit per log reduction. Blood, 125 (2015), 1932–5. 45. T.A. Oldaker, P.K. Wallace and D. Barnett. Flow cytometry quality requirements for monitoring of minimal disease in plasma cell myeloma. Cytometry B Clin Cytom, 90 (2016), 40–6.

139

09

19:51:58

Chapter

9

Mature T-Cell Neoplasms and Natural Killer-Cell Malignancies Anne Tierens

Introduction Mature T-cell and Natural Killer (NK)-cell neoplasms are classified in distinct disease entities according to the World Health Organization (WHO) criteria (Box 9.1) [1]. T-cell and NK-cell neoplasms are relatively infrequent diseases and account for 10%–15% of all nonHodgkin’s lymphomas (NHLs). They usually affect middle-aged and older adults and more frequently men Box 9.1  Mature T- and Natural Killer (NK)-cell neoplasms (Revised WHO classification of lymphoid neoplasm; according to Swerdlow SH et al. [1]) T-cell prolymphocytic leukaemia* T-cell large granular lymphocytic leukaemia Chronic lymphoproliferative disease of NK-cells Aggressive NK-cell leukaemia Systemic Epstein- Barr virus + T-cell lymphoma of childhood Hydroa vaccineforme-like lymphoproliferative disorder Adult T-cell leukaemia/ lymphoma (ATLL) Extranodal NK/T- cell lymphoma, nasal type Enteropathy-associated T-cell lymphoma Monomorphic epitheliotropic intestinal T-cell lymphoma Indolent T-cell lymphoproliferative disorder of the gastrointestinal tract Hepatosplenic T-cell lymphoma Subcutaneous panniculitis-like T-cell lymphoma Mycosis fungoides Sézary syndrome Primary cutaneous CD30 + T-cell lymphoproliferative disorders Lymphomatoid papulosis Primary cutaneous anaplastic large-cell lymphoma (ALCL) Primary cutaneous γ/δ T-cell lymphoma Primary cutaneous CD8 + aggressive epidermotropic cytotoxic T-cell lymphoma Primary cutaneous acral CD8 + T-cell lymphoma Primary cutaneous CD4 + small/medium T-cell lymphoproliferative disorder Peripheral T-cell lymphoma, not otherwise specified (NOS) Angioimmunoblastic T-cell lymphoma (AITL) ALCL, ALK+ ALCL, ALK− Breast implant-associated ALCL

140

*The disease entities in bold are discussed in this chapter.

10

than women. The majority of mature T-cell chronic lymphoproliferative disorders (T-CLPDs) primarily involve lymph nodes and extranodal tissues but some T-cell disease entities present with a predominant leukaemic picture. NK-cell neoplasms are classified into three disease entities including extranodal NK/T-cell lymphoma, nasal type (NKTCL), aggressive NK-cell leukaemia (ANKL) and chronic lymphoproliferative disorder of NK-cells (CLPD-NK). The diagnosis and classification of mature T-cell neoplasms, especially those with leukaemic involvement, may be challenging because of overlapping clinical and pathological features [2]. Therefore, the integration of clinical and genetic features with the morphologic and immunophenotypic findings is needed for classification. The identification of disease-specific markers by molecular analysis or multiparameter flow cytometry (FCM) assists in the diagnostic work-up. In this chapter, mature T- and NK-cell disease entities with primary or secondary leukaemic involvement are discussed with respect to their i­mmunophenotypic profiles and cell of origin, as well as genetic and clinical features.

Antibody Panels for T-Cell and NK-Cell Lymphoproliferative Diseases Various antibody combinations of pan T-cell and NK-cell antigens (CD2, CD3, CD5, CD7, CD16, CD56 and CD57) can be used to identify the malignant cells of mature T-cell and NK-cell diseases. Examples of antibody combinations are given in Table 9.1. Normal patterns of antigen expression are presented in Chapter 3. Aberrant expression of two and often more T-cell and NK-cell-associated antigens is very common in T-cell and NK-cell lymphoproliferative diseases. However, since many of the aberrancies are not specific for a disease entity, additional markers that are differentially expressed in specific disease entities have been evaluated to improve diagnostic accuracy. In addition, the 19:53:02

Chapter 9: Mature T-Cell Neoplasms and Natural Killer-Cell Malignancies Table 9.1  Antibody panels used for T- and Natural Killer (NK)-cell phenotyping

T-cell panels

FITC

PE

T-/NK-cell subsets

CD16+CD57

Sézary syndrome

CD7

T-cell maturation

PE-TR ECD

PercPCy5.5 PeCy7 PeCy5.5

APC

APC700/ APCH7 V450 APC APCCy7 PB AF700 APC AF750 BV421

CD7

CD4

CD2

CD56

CD3

CD5

CD8

Aggarwal N et al. [3]

CD26

sCD3

CD2

CD28

CD8

CD4

CD45

van Dongen JJM et al. [4]

CD27

CD197

sCD3

CD45RO

CD45RA

CD8

CD4

CD45

T-PLL

CD5

CD25

sCD3

HLA-DR

cy TCL1

CD8

CD4

CD45

Cytotoxic T-cells/ALCL/AILT

CD57

CD30

surface CD3

CD279

CD11c

CD8

CD4

CD45

Cytotoxic T-cells/T-LGL

cy perforin

cy granzyme B

surface CD3

CD16

CD94

CD8

CD4

CD45

AILT

CD57

CD278

CD5

CD279

CD10

CD3

CD4

CD8

Baseggio L et al. [5]

T-/NK-cell subsets

CD2

CD5

CD34

CD56

CD3

CD7

CD45

CD8

CD4

Chen X et al. [6]

CD5

CD45

CD4

CD4

CD5

CD30

V500 PO KO/A594

CD7

CD34

CD8

CD3

CD56

CD57

CD11c

CD8

CD3

CD2

CD56

CD45

Porwit A. [7]

T-NK-cell subsets

CD4

CD26

CD8

CD3

CD7

CD14

CD45

Loghavi S et al. [8]

CD57

CD94

CD8

CD3

CD56

CD16

CD45

10

CD2 T-/NK-cell subsets

CD7

TCRαβ

TCRγδ

CD5

CD3

CD2

CD4

CD45

CD4

CD30

CD8

CD3

CD10

CD25

CD45

CD7

CD26

sCD3

CD56

CD5

CD19

CD2

CD45

CD57

CD25

sCD3

CD56

CD11c

CD19

CD16

CD45

cy perforin

cy granzyme B

sCD3

CD56

CD94

CD19

HLA-DR

CD45

CD2

CD7

CD56

CD3

CD16

CD5

CD56

CD3

CD57

CD11c

CD56

CD3

CD38

CD11b

CD56

CD3

CD94

CD161

CD56

CD3

CD11a

HLA-DR

CD56

CD3

CD158a

CD158e

CD56

CD3

CD45RA

CD45RO

CD56

CD3

CD27

CD26

CD56

CD3

NK-cell panels 19:53:02

van Dongen JJM et al. [4]

Lima A et al. [9]

A594: AlexaFluor 594; APC A700: APC-Alexa Fluor 700; APC Cy7: APC-cyanine 7; APC H7: APC-Hilite7; APC: allophycocyanin; BV421: Horizon Brilliant Violet 421;cy: cytoplasmic; FITC: fluorescein isothiocyanate; Horizon V450 and V450; KO: Krome Orange; PB: Pacific Blue; PE: Phycoerythrin; PECy5.5: PE-Cyanine 5.5; PECy7: PE-cyanine 7; PercPCy5.5: peridinin-chlorophyll cyanine 5.5; PO: Pacific Orange; s: surface.

141

Chapter 9: Mature T-Cell Neoplasms and Natural Killer-Cell Malignancies

assessment of maturation stage-associated and activation markers may be helpful for diagnosis. Although analysis of disease- and maturation-associated markers may contribute to diagnosis, international recommendations are not yet published regarding the use of these markers. A new approach to data analysis was proposed by the EuroFlow consortium to improve the immunophenotypic diagnosis and classification of T-CLPDs and NK-CLPDs [4, 10–13]. Data analysis was performed using the automated pattern-guided principal component analysis (PCA) of merged and calculated data files of T-CLPD or NK-CLPD. By comparing patient files with reference data files of normal cell populations and files of the respective disease entities, it was possible to correctly classify CD4+ disease entities; however, the CD8+ and CD8−/CD4− lymphoproliferative disorders showed overlapping immunophenotypic profiles [4, 10–13]. Tables 9.2, 9.3, and 9.4 provide the immunophenotypic characteristics of T-CLPDs and NK-CLPDs.

Assessment of T-Cell Receptor Clonality by FCM analysis Polymerase chain reaction of T-cell receptor (TCR)γ and TCRβ rearrangements followed by capillary electrophoresis is the reference method to detect clonal T-cells. Alternatively, the restricted expression of specific TCRVβ proteins can be analysed by using antibodies against TCRVβ families covering about 70% of the TCRVβ repertoire [30–32]. The presence of a clonal T-cell subset is suspected when more than 50% of the

analysed population is positive for a single TCRVβ or its TCRVβ expression exceeds 10-fold its normal ­maximum [33]. It is a useful tool to determine the clonality of T-cell populations with abnormal immunophenotype at diagnosis and for disease monitoring [34]. For example, the assessment of TCRVβ2 levels within the CD4+ T-cells in peripheral blood (PB) is a sensitive and specific screening assay for the identification of clonal T-cells in patients with Sézary syndrome [35].

T-Cell Lymphoproliferative Disorders T-Cell Prolymphocytic Leukaemia T-cell prolymphocytic leukaemia (T-PLL) is a rare T-cell lymphoproliferative neoplasm identified by the rearrangement involving the T-cell leukaemia 1 (TCL1) gene at 14q32 or much less frequently the mature T-cell proliferation 1 (MTCP-1) gene at Xq28 and the TCRα/δ locus at 14q11.2 [36–38]. Inv(14)(q11.2;q32) and t(14;14)(q11.2;q32) are demonstrated in 80% and 10% of cases, respectively [39]. Additional abnormalities of chromosomes 8 and 12p, as well as deletions of 11q23 are also frequently seen [39]. The 11q23 is the locus of the ataxia telangiectasia (AT)-mutated (ATM) gene mutated in patients with AT who may develop T-PLL at an earlier age than in sporadic cases [40, 41]. Deletions of chromosomes 6 and 17p are detected in 33% and 26% of cases, respectively [42]. Genome sequencing has shed more light in the pathogenesis of T-PLL. Such analyses have revealed a high frequency of gain-of-function mutations in Janus kinases (JAK) 1/3

Table 9.2  Immunophenotypic patterns of Natural Killer (NK)-cells in acute and chronic activation states and chronic ­lymphoproliferative disorders of NK-cells

142

CD56dim+/ CD16+ NK-cells

Acute activation

Chronic activation

Clonal NK-cells

CD56/CD16

CD56 dim+; CD16 +

CD56 dim+; CD16 +

CD56−/dim+; CD16 +

CD2/CD7

CD2bimodal −/+; CD7bright+

CD11a/CD11b/CD11c

CD11a ; CD11b CD11cheterogeneous low+

CD94

CD94heterogeneous+

bright+

CD38/HLA-DR

CD38 ; HLA-DR

CD45RA/CD45RO CD57

+

CD2bimodal −/+; CD7heterogeneous+ ;

low+

CD11a ; CD11b CD11cbimodal−/+ bright+

heterogeneous low/−

CD94heterogeneous+

CD2bright+; CD7heterogeneous dim+/partial − ;

CD11abright+; CD11bheterogeneous low+/−; CD11cbright+ CD94bright homogeneous+

−/heterogeneous low+

CD38 ; HLA-DRheterogeneous low

CD38 low+; HLA-DRheterogeneous+

CD45RA +; CD45ROheterogeneous+

CD45RA +; CD45ROheterogeneous low+

CD45RA +; CD45ROheterogeneous+

CD57 −/ small subset+

CD57heterogeneous+

CD57bimodal −/+

heterogeneous+

Data from Lima M et al. [14, 15]

10

19:53:02

Chapter 9: Mature T-Cell Neoplasms and Natural Killer-Cell Malignancies Table 9.3  Immunophenotypic characteristics of T-cell lymphoproliferative disorders

T-prolymphocytic leukaemia

Immunophenotypic profiles

Frequently observed Disease entity-specific markers changes of T-cell-associated and marker combination antigens

References

CD2+CD3+CD5+CD7+CD45RO+TCRαβ+

CD3 increased

Chen and Cherian [6]

+

CD4 (75%)

CD5 increased

CD4+CD8+ (35%)

CD2 decreased

cyTCL1

CD8+CD4− (15%–17%) Sézary syndrome

CD2+CD3+CD5+CD7heterogeneous+TCRαβ+ +

CD2 lowCD3lowCD26−CD28+ +

Kelemen et al. [16]

−

CD4

CD4 decreased

CD164 CD26 (early and advanced SS)

Sokolowska-Wysocka et al. [17]

CD8+ (rare)

CD2 loss/CD2 decreased

FCRL3+ CD26− (advanced SS)

Novelli et al. [18]

+

Adult T-cell leukaemia/ lymphoma

CD3 decreased

+

CCR7 CD27

increased scatter

CD26−CD28+

CD3 and CD5 loss (rare)

CD2+CD3+CD5+CD7−/lowTCRαβ+

CD3 decreased

10

CD4+ (majority)

CD5 decreased

CD8+ (rare) and CD4+CD8+ (rare)

CD7 partial positivity

+

CD158k (advanced SS)

Moins-Teisserenc et al. [19]

CD3lowCD25+CD26−HLA-DR+FOXP3+

Shao et al. [20] Tian et al. [21]

CD27+ CD25+CD26−HLA-DR+ 19:53:02

Angioimmunoblastic T-cell lymphoma

T-Large granular lymphocytic leukaemia

CD2+CD3+CD5+CD7+CD4+

sCD3 loss

cyCD3+ or sCD3 lowCD5+ CD4+CD10+

Singh et al. [22]

CD10 subset

sCD3 decreased

cyCD3+ or sCD3 lowCD5+ CD4+ ICOS+

Marafioti et al. [23]

+

cyCD3 / sCD3 ICOS+PD-1+

+

+

+

CD5 CD4 CD10

Baseggio et al. [5]

ICOS subset

CD5 increased

PD-1 subset

CD7 decreased

CD2+CD3+CD5low+CD7+

CD5 loss

Cy Granzyme B

van Dongen et al. [4]

TCRab+ CD8+ or CD4−CD8− (rare)

CD7 decreased or loss

Cy Perforin

Lundell et al. [24]

low

(cont.)

143

Chapter 9: Mature T-Cell Neoplasms and Natural Killer-Cell Malignancies Table 9.3  (cont.)

Immunophenotypic profiles

TCRgd+ CD8low+ +

+

+

CD57 (often partial) CD16 or CD56 *

Frequently observed Disease entity-specific markers changes of T-cell-associated and marker combination antigens

References

CD2 decreased

Morice et al. [25]

CD3 increased

CD45RA+CD45RO−CD27− CD28− CD94+ monotypic CD158a/b/e (subset of cases) Variants TCRαβ+CD4+CD8 low +

−

preferential usage of TCRVβ 13.1

−

−

TCRγδ CD4 CD8 (rare) Hepatosplenic T-cell lymphoma

CD2+CD3+CD5−CD7+ or CD7− CD4−CD8 −/ variable+TCRγδ+CD56+

−

# CD5 CD57 (TCRγδ variant) CD5 loss

CD3+CD5−CD56+CD57−TCRγδ+

CD57−CD16+ or CD16− 10

cy perforin− granzyme B− CD38− CD94 homogeneous+ monotypic CD158a/b/e (subset of cases)

19:53:02

*CD16 and CD56 are seldom co-expressed; # Immunophenotype is similar to the immunophenotype of hepatosplenic T-cell lymphoma.

144

Lima et al. [26] Chen et al. [27]

Chapter 9: Mature T-Cell Neoplasms and Natural Killer-Cell Malignancies

Table 9.4  Immunophenotypic characteristics of natural killer (NK)-cell lymphoproliferative disorders

Chronic lymphoproliferative disorder of NK-cells

Immunophenotypic profiles

Frequently observed changes of T/NK-cellassociated antigens

CD2+CD3−CD7+CD16+ CD56− or low (2/3 of cases)

CD7 decreased

Chan et al. [28]

Skewed expression of CD158a/CD158b/CD158e

CD38 decreased

Lima et al. [9]

CD94 homogeneous+

CD11c increased

Morice et al. [29]

+

−

+

Disease entityspecific markers/ marker combination

References

−

CD2 CD3 CD7 CD16 CD56bright+ (1/3 of cases] KIRs− CD94bright+ Extranodal NK/T-cell lymphoma, nasal type

CD2+CD3−CD16 low (infrequent) CD7 lowCD8variable low

Aggressive NK-cell leukaemia

CD56bright+ CD94bright+

EBV+

Lima et al. [9]

CD45RA+CD45ROheterogeneous+ CD38 heterogenous+ HLA-DRheterogeneous+CD28+

and in the signal transducer and activator of transcription protein 5 (STAT5b). Besides, inactivating mutations have been disclosed in the epigenetic modulator enhancer of zeste homolog 2 (EZH2), a member of the polycomb group family of transcriptional repressors, as well as in DNA repair genes such as checkpoint kinase 2, ATM and TP53 [43, 44]. T-PLL usually affects older patients with a median age of 65 years. Presenting clinical features are lymphocytosis and hepatosplenomegaly. Half of the patients also have generalized lymphadenopathy whereas 15% to 25% show skin lesions, including predominantly maculopapular rash and oedema, but seldom erythroderma [39, 45, 46]. There are two morphologic variants. The prolymphocytic variant is characterised by smallto-medium-sized lymphocytes with round to irregular and sometimes even cerebriform nuclei with a distinct nucleolus. The small-cell variant, seen in around 25% of the cases, is predominantly composed of small lymphocytes with round nuclear contours without conspicuous nucleolus. Irrespective of the variant, the presence of cytoplasmic blebs is often a morphologic feature. T-PLL has a naïve/central memory phenotype with 75% of the cases derived from CD4+ TCRαβ+ CD45RO+ T-cells of which half do also express CD8 [6, 47]. 10

Between 15% and 17% of cases are CD8+/CD4− [6, 47]. Double-negative CD4−CD8− T-PLL is extremely rare [6]. T-PLL usually consists of an immunophenotypically homogenous population. However, cases with distinct abnormal subsets have been described [6]. Although most T-PLL populations show normal expression of the pan T-cell markers CD2, CD3, CD5 and CD7, abnormal dim or bright expression is not infrequent. In a study of 20 cases, abnormal expression of CD7, CD5, CD2 and CD3 was observed in 31%, 27.6%, 20.7% and 13.8% of the cases, ­respectively [6]. Higher expression levels than normal are predominantly identified for CD3 and CD5, whereas lower expression levels are seen for CD2. The expression of CD7 can be either increased or decreased. Some T-PLL may also be positive for CD25 at low levels. The expression of cytoplasmic (cy) TCL1 in 70%–80% of cases is a distinct immunophenotypic feature in the differential diagnosis with other mature CD4+ T-cell neoplasms [4, 48]. Its overexpression is associated with a more aggressive course often presenting with high cell counts and hyperproliferation [48]. The analysis of TCL1 and of the identified Vβ gene by FCM can be employed for the detection of residual T-PLL [4]. In CD4+CD8+ or CD4−CD8− cases without expression of cyTCL1, T-cell lymphoblastic lymphoma/leukaemia 19:53:02

145

Chapter 9: Mature T-Cell Neoplasms and Natural Killer-Cell Malignancies

needs to be excluded by analysing the immature markers CD34, CD1a and terminal deoxynucleotidyl transferase (TdT) which are negative in T-PLL. Figure 9.1 shows characteristic immunophenotypic features of a CD4+ T-prolymphocytic leukaemia.

Adult T-Cell Leukaemia/Lymphoma

1E6

1E6

8E5

1E5

1E5

6E5 400000 200000

CD7 APC-A700:FL7-A

1E6

CD3 PC5.5:FL4-A

SS INT LIN:SS-A

Adult T-cell leukaemia/lymphoma (ATLL) is a unique CD4+ T-cell disease entity caused by infection by the human T-cell lymphotropic virus-1 (HTLV-1), which is endemic in the southwestern region of Japan, the Caribbean islands and parts of Central Africa and South America [49, 50]. In other parts of the world, ATLL is extremely rare and only observed in first- and second-generation immigrants from these endemic regions. HTLV-1 is a single-stranded RNA virus that is

integrated in the host’s genome. It encodes multiple regulatory genes of which the p40 tax protein is postulated to play a role in early ATLL pathogenesis by activation of the JAK–STAT and NF-κB signalling pathways, but it is not sufficient for transformation [51–53]. Notably, an integrated molecular analysis of ATLL cases showed mutations in the genes coding for proteins interacting with Tax in HTLV-1 infected T-cells and involved in T-cell migration [54]. The most frequently mutated genes in the former group include PLCG1, PRKC8, CARD11 and STAT3 which belong to the TCR-NF-κB signalling pathway. Mutations in the chemokine receptors CCR4 and CCR7 have been demonstrated in 40% of the cases and are associated with increased Phosphoinositide 3 kinase (PI3K) signalling. In addition, loss-of-function and gain-of-function mutations in RHOA could also be involved in the pathogenesis of

1E4 1E3 1E2 0

1E3 1E4 1E5 CD45 K0:FL10-A

TCR ab PE:FL2-A

(a)

–1E2 0 1E2

1E6

1E3 1E2 0 –1E2

0 1E2

(b)

1E3 1E4 1E5 CD5 PB:FL9-A

1E6

1E6

1E5

1E5 CD8 ECD:FL3-A

0

1E4

1E4 1E3 1E2

1E6

(c)

1E3 1E4 1E5 1E6 –1E2 0 1E2 CD2 PC7:FL5-A

1E4 1E3

0

0 –1E2 0 1E2

(d)

146

1E3

1E4

1E5

CD25 PB:FL9-A

–1E2 0 1E2

1E6

(e)

1E3 1E4 1E5 1E6

CD4 APC-A750:FL8-A

Figure 9.1  The following antibody combinations were used: CD57FITC-CD11cPE-CD8ECD-CD3PCy5.5-CD2PeCy7-CD56APC-CD7APCAF700CD4APCAF750-CD5PB-CD45KO and CD30FITC-TCRαβPE-CD8ECD-TCRγδPCy5.5-CD3PeCy7-CD1aAPC-CD7APCAF700-CD4APCAF750-CD25PBCD45KO. Data analysis was performed using Infinicyt™ (version 1.7). The T-cell prolymphocytic leukaemia cells were gated as follows: CD3+ events were selected in a CD3 versus SSC dot plot, followed by back-gating on the clustered low FSC and SSC events and CD45bright+ events (Figure 9.1a). Subsequently, total CD4+ events among T-cells were gated, followed by selecting the CD3, CD5 (Figure 9.1b) and CD7bright (Figure 9.1c) events or TCRαβbright+ and CD3bright+ (Figure 9.1d) events among total CD4+ T-cells (Figure 9.1e). The black, green and red dots correspond to all leucocytes, reactive T-cells and T-prolymphocytes, respectively. The expression of CD3 and CD7 is increased in the neoplastic population and there is a low expression of CD25.

10

19:53:02

Chapter 9: Mature T-Cell Neoplasms and Natural Killer-Cell Malignancies

ATLL [55]. Although recurrent numerical and structural chromosomal abnormalities have been demonstrated, none are specific for ATLL [56]. In 50% of the cases, loss of heterozygosity of chromosome 6q15–21 suggests the inactivation of a tumour-suppressor gene in ATLL [57]. ATLL is a disease of adults aged between 20 and 80 years old with an average in the mid-60s. About 65%–75% of the patients present with systemic disease and leukaemia, while the remaining patients manifest with lymphoma [58]. There are four subtypes depending on the clinical presentation: acute, chronic, lymphomatous and smouldering. The acute and lymphomatous ATLL are the most aggressive forms with survival rates between 6 and 10 months. The acute form is characterised by generalised lymphadenopathy and hepatosplenomegaly, hypercalcaemia with or without osteolytic lesions, leukaemic involvement and high levels of lactate dehydrogenase (LDH). In patients presenting with lymphoma, no or minimal blood involvement is seen despite extensive disease. Skin lesions and eosinophilia are common in these variants. In chronic ATLL, skin rash and a variable lymphocytosis are the presenting features. In most patients, the lymphocytosis may remain stable for a long period of time. The smouldering variant manifests as exfoliative skin rashes and pulmonary lesions. Lymphocyte counts are usually normal and circulating abnormal lymphocytes are few. The morphology of ATLL cells is pleomorphic with respect to size and nuclear chromatin features. Lymphoma cells may be small and medium-to-large cells exhibiting irregular nuclear contours of which some have a flower-like appearance and coarse to dispersed nuclear chromatin. Anaplastic cell variants have been described. ATLL is postulated to arise from regulatory T-cells hence the characteristic immunophenotype of CD3+ CD4+CD25+FoxP3+CCR4+ T-cells [59, 60]. The high incidence of opportunistic infections caused by T-cell immunodeficiency in the patients correlates with this immunophenotype. Tumour cells are consistently positive for CD2, CD3, CD4, CD5, CD27 and brightly positive for CD25, while they are often negative for CD7 and CD26. Although most ATLL cases are negative for CD7, partial or dim expression is observed. There are usually normal expression levels of CD2, whereas the expression of CD3 is decreased in the majority of cases [21, 61]. Also, dim expression of CD5 was observed in 22% of cases in one study [20]. While the majority is CD4+, occasional CD8+ and double CD4+CD8+ 10

cases have been demonstrated [49]. The most distinctive immunophenotypic features of the tumour cells are dim expression of CD3 and loss of CD26 [20, 21]. Identification of the latter immunophenotypic changes is useful in the distinction with normal CD3+CD4+CD25+ T-cells as shown in a study investigating residual lymphoma cells in PB of patients with ATLL [20]. Indeed, lymphoma cells as low as 0.29% of total leucocytes could be detected using this approach. Characteristic immunophenotypic features of ATLL are shown in Figure 9.2.

Mycosis Fungoides and SS Mycosis fungoides (MF) and Sézary syndrome (SS) are the most common cutaneous T-cell lymphomas characterised by skin infiltration of small-to-mediumsized T-lymphocytes. MF and SS account for approximately 50% and 5% of cutaneous T-cell lymphomas, respectively [62, 63]. While MF remains confined to the skin, SS is a systemic disease defined by erythroderma, lymphadenopathy and circulating Sézary cells (SC) characterised by their cerebriform nuclei. The International Society for Cutaneous Lymphomas and the European Organization of Research and Treatment of Cancer Classification determined four clinical stages depending on the extent of skin infiltration and blood involvement [64]. Defining features for the classification include the extent of the patches or plaques, skin involvement with erythrodermic lesions, the degree of lymphadenopathy and a cut-off of circulating SC at 1 × 109/L. Alternative criteria used instead of absolute counts of SC are the increase of abnormal CD4+ T-cells with a CD4:CD8 ratio of more than10 and the presence of >40% CD7− T-cells or >30% CD26− T-cells [64]. Both MF and SS occur in older patients over the age of 60 but MF may be observed in young adults and children. While MF is an indolent disease with slow progression, SS is aggressive with a 5-year s­urvival between 10 and 20%. Complex karyotypes are frequently present and recurrent structural chromosomal abnormalities have been demonstrated [65–67]. Gains at 8q24 and losses at 17p, 19p and chromosome 10 are often found in SS. Genomic profiling identified somatic mutations in genes involved in epigenetic regulation, T-cell survival and differentiation, including NF-κB, MAPK (mitogen activated protein kinase) and JAK-STAT signalling pathways [68–71]. SC are derived from central-memory CD4+CD3+ T-cells expressing CD45RO, CD197 (CCR7), CD27 19:53:02

147

1E6

1E6

8E5

1E5

1E5

6E5 400000 200000

–1E2 0 1E2

1E3 1E2 0

1E3 1E4 1E5 1E6

CD45 K0:FL10-A

(a)

1E4 1E3 1E2 0 –1E2

–1E2 0 1E2

1E3

1E4 1E5 1E6

CD5 PB:FL9-A

(b)

–1E2 0 1E2

1E6

1E5

1E5

1E5

1E3 1E2 0 –1E2 –1E2 0 1E2

1E3 1E4 1E5 1E6

CD25 PB:FL9-A

CD26 PE:FL2-A

1E6

1E4

1E4 1E3 1E2

1E4 1E3 1E2

0

0

–1E2

–1E2 –1E2 0 1E2

(e)

–1E2 0 1E2

1E3 1E4 1E5 1E6

CD4 APC-A750:FL8-A

(f)

1E3 1E4 1E5 1E6

CD2 PC7:FL5-A

(c)

1E6

CD8 ECD:FL3-A

TCR ab PE:FL2-A

1E4

–1E2

0

(d)

CD7 APC-A700:FL7-A

1E6

CD3 PC5.5:FL4-A

SS INT LIN:SS-A

Chapter 9: Mature T-Cell Neoplasms and Natural Killer-Cell Malignancies

1E3

1E4

1E5 1E6

CD4 FITC:FL1-A

Figure 9.2  The following antibody combinations were used: CD57FITC-CD11cPE-CD8ECD-CD3PCy5.5-CD2PECy7-CD56APC-CD7APCAF700-CD4APC-AF750-CD5PB-CD45KrO, CD30FITC-TCRαβPE-CD8ECD-TCRγδPCy5.5-CD3PECy7-CD1aAPC-CD7APC-AF700-CD4APC-AF750CD25PB-CD45KrO and CD4FITC-CD26PE-CD3PECY7-CD45KrO. Data analysis was performed using Infinicyt™ (version 1.7). The adult T-cell leukaemia/lymphoma (ATLL) cells were gated as follows: CD3+ events were selected in a CD3 versus SSC dot plot, followed by back-gating on the clustered low FSC and SSC events and CD45bright+ events (Figure 9.2a). Subsequently, total CD4+ events among total T-cells were gated followed by selecting CD3dim, CD5bright (Figure 9.2b) and CD7dim events (Figure 9.2c) or CD25+ and TCRαβdim+ events (Figure 9.2d) among total CD4+ T-cells (Figure 9.2e). The black, green and red dots correspond to all leucocytes, reactive T-cells and ATLL cells, respectively. The ­expression of CD3 and CD7 is decreased, while the expression of CD5 is increased. CD25 is weakly positive and CD26 is negative (Figure 9.2f).

148

and CD28 as well as the skin-homing common lymphocyte antigen (CLA, CD162) and chemokine receptors CCR4 (CD194) and CCR10 [72, 73]. The most frequently reported immunophenotypic aberrancies of SC are the loss of CD26 and variable CD7 expression [16–18, 74–80]. Weak and infrequently bright expression of CD3, CD2 and CD4 was reported in 76%, 48% and 44% of the cases, respectively, in a retrospective study of 107 patients [18]. While loss of CD2 has been reported in approximately one-fifth of the cases, loss of CD3 or CD5 is rare [18]. An increased scatter signal is another immunophenotypic feature of cutaneous T-cell lymphoma [81]. Since all of the above-­mentioned immunophenotypic changes are not constant findings and may be observed in reactive dermatoses, additional surface molecules have been investigated for the detection of SC. The killer immunoglobulin-like 10

receptor CD158k/KIR3DL2 is expressed at elevated levels on SC. CD158k+ (KIR3PDL1)/CD4+ T-cells were demonstrated in 40% to 97% of patients [19, 82–85]. The differences in percentages of CD158k+ SC reported in the different studies may be explained by the use of different antibody clones and staining methods as well as the stage of the disease. Indeed, the percentage of CD158k+CD4+ T-cells strongly correlates with the percentage of atypical lymphocytes in the blood film and the presence of clonal T-cells indicating a more advanced stage of the disease [85]. Similarly, syndecan-4 (SD-4) and ganglioside GD3 (CD60) are expressed at elevated levels in advanced SS [86, 87]. A microarray analysis of global gene expression on samples from MF/SS, reactive dermatosis and healthy donors demonstrated a significantly higher mRNA expression of CD164, a sialomucin adhesion receptor, 19:53:02

Chapter 9: Mature T-Cell Neoplasms and Natural Killer-Cell Malignancies

and of the Fc-receptor-like 3 (FCRL3). The mRNAs of T-plastin and SD-4 were also increased confirming previous findings [88]. Of interest, the surface expression of CD164 and FCRL3 analysed by FCM correlate with the loss of CD26 expression in SS, but not in MF. CD164 is significantly increased in all patients, including those with low tumour burden. In contrast, the expression of FCRL3 is mainly present in patients with high tumour burden. Moreover, CD164+ CD4+ T-cells display cerebriform morphology and are larger in size in keeping with SC. Taken together, CD164 might be a specific marker for diagnosis and disease monitoring of patients with SS. Although the intracellular protein T-plastin is exclusively expressed in malignant CD4+ T-cells of SS, no specific antibodies are available for diagnostic use.

Angioimmunoblastic T-Cell Lymphoma Angioimmunoblastic T-cell lymphoma (AITL) is a distinct entity of peripheral T-cell lymphoma with generalised lymphadenopathy, hepatosplenomegaly, polyclonal hypergammaglobulinaemia and circulating immune complexes [89, 90]. Pleural effusions, ascites and arthritis are also common manifestations. Minimal BM involvement and PB dissemination are frequent in patients with AITL. AITL is characterised by a polymorphic infiltrate of small-to-medium-sized lymphocytes with pale cytoplasm admixed with B-immunoblasts, plasma cells, eosinophils and histiocytes, a proliferation of high endothelial venules and expansion of the follicular dendritic cell network. The pattern of lymph node involvement varies between preservation of the architecture with the presence of hyperplastic follicles and variable effacement of the architecture with depleted or complete absence of B-follicles. Epstein-Barr virus (EBV)-positive B-cells are nearly always present [91]. Gene expression analysis of AITL has shown a derivation from follicular T-helper (Tfh) cells being CD4+CD45R0+ and characterised by the expression of a number of Tfh-specific markers, of which CD10, BCL6 and CXCL13 are the most useful for the diagnosis of AITL in lymph node biopsies [92–94]. AITL is further characterised by isocitrate dehydrogenase (IDH) 2 mutations identified in 20%–45% of cases [95, 96]. The IDH2 R172 mutation defines a unique subgroup of AITL and has been associated with DNA and histone hypermethylation of genes involved in TCR signalling and differentiation. 10

Although the morphologic features in lymph nodes are well characterised, the diagnosis may be challenging because of overlapping features with reactive conditions. Multiparameter FCM analysis of lymph node cell suspension and PB can assist in the diagnosis. Indeed, the expression of CD10 and frequent loss of surface CD3 expression are distinct immunophenotypic features of the neoplastic T-cells [97–101]. Other immunophenotypic aberrancies of T-cell-associated antigens have also been observed, but they are not specific for AITL [102, 103]. The most frequent changes after the loss of surface CD3 are complete or partial loss of CD7 and overexpression of CD5 and CD45RO [22]. Weak expression of CD4 and bright expression of CD2 have been infrequently observed. Although CD10 is positive in the majority of cases, CD10 expression is often partial and may be completely absent in approximately 20% of cases. Therefore, the expression of additional Tfh markers such as programmed death-1 (PD-1) and inducible costimulatory (ICOS) has been evaluated in both lymph node cell suspensions and PB samples [5, 23, 104]. In a study of 15 cases, ICOS was positive in 87% of the lymph node cell suspensions using a threshold of 5% of CD4+ T-cells, whereas ICOS+CD4+ T-cells were not identified in the control group [5]. In contrast, PD-1 (CD279) was positive in 60% of reactive hyperplasia and in 93% of AITL using a threshold of 20% of CD4+ T-cells. Although the expression of PD-1 was higher in AITL, it could not be used to discriminate AITL from reactive hyperplasia. Co-expression of CD10, ICOS and PD-1 was demonstrated in 67% of AITL but not in control cases. Also, in the 13 analysed PB samples, the expression of ICOS was specific for AITL whereas the expression of PD-1 was not. Of interest, the ICOS+ subset in PB is smaller than the CD10+ subset of the lymph nodes and PB, indicating clonal heterogeneity. It remains to be demonstrated whether ICOS is a useful marker in CD10− cases.

Hepatosplenic T-Cell Lymphoma Hepatosplenic T-cell lymphoma (HSTL) is a rare extranodal T-cell lymphoma of the liver and spleen affecting young and predominantly male adults at a median age of 35 years [105–108]. Up to 20% of HSTL arise after prolonged immunosuppressive therapy for solid organ transplantation or chronic antigen stimulation of unknown origin. HSTL may develop secondary to treatment with infliximab and azathioprine in teenagers and young adults treated for Crohn’s disease and 19:53:02

149

Chapter 9: Mature T-Cell Neoplasms and Natural Killer-Cell Malignancies

ulcerative colitis. Patients typically present with hepatosplenomegaly, cytopenias and often B-symptoms yet no or minimal lymphadenopathy. Histology of the spleen and liver shows sinusoidal infiltration by medium-sized lymphoid cells with variably condensed nuclear chromatin without conspicuous nuclei. BM involvement is a constant finding, and lymphoma cells, although rare, may be identified in the PB. A marked thrombocytopenia is often present, but anaemia and leucopenia are also common. Upon disease progression, transformation to large atypical and blast-like cells may be seen. Isochromosome 7q is the characteristic cytogenetic abnormality along with frequent loss of chromosome Y and trisomy 8 [109, 110]. The lymphoma cells have the immunophenotype of inactive cytotoxic T-cells, predominantly expressing TCRγ/δ. They are positive for CD2, CD3, CD38 and CD56 and are usually negative for CD5, CD8 and CD57 [111]. The expression of CD16 and CD7 may be positive or negative. In keeping with its inactive cytotoxic phenotype, HSTL is negative for the cytotoxic effector proteins perforin and granzyme B. It is however positive for T-cell-restricted intracellular antigen-1 (TIA-1) and granzyme M, a marker of lymphocytes of innate immunity [112, 113]. Of interest, aberrant expression of multiple killer immunoglobulin-like (KIR) receptors as well as the expression of CD94 has also been demonstrated [113].

T-Cell Large Granular Lymphocytic Leukaemia

150

T-cell large granular lymphocytic (T-LGL) leukaemia is an indolent lymphoproliferative disorder of cytotoxic T-lymphocytes characterised by a clonal large granular lymphocytosis of at least 0.5 × 109/L, but typically between 2 and 20 × 109/L for a period of six months, cytopenias and frequent autoimmune diseases [114, 115]. Mild-to-moderate splenomegaly is observed in 20%–60% of the patients. Hepatomegaly and lymphadenopathy are rare. T-LGL leukaemia is found in older patients with a median age of 60 years and is usually diagnosed months to years after the first clinical symptoms. The predominant clinical presentation is mild-to-severe neutropenia in nearly 80% of the patients and high frequencies of bacterial infections. Chronic anaemia largely due to autoimmune haemolytic anaemia (AIHA) and idiopathic thrombocytopenia are seen in 50% and 20% of the patients, respectively. T-LGL leukaemia seems to be the most 10

frequent cause of pure red cell aplasia (PRCA). It is also reported in aplastic anaemia and myelodysplastic syndrome [116]. Rheumatoid arthritis and Felty’s syndrome are the most common autoimmune disorders reported in 25%–40% of T-LGL. A diagnosis of T-LGL leukaemia requires a detailed clinical history that should be integrated with laboratory findings. Indeed, healthy older adults may h ­ arbour clonal expansions of large granular lymphocytes with immunophenotypes indistinguishable from the indolent T-LGL leukaemia [117]. In addition, a transient increase of clonal large granular lymphocytes is observed secondary to viral infections [118]. Persistent large granular lymphocytosis with evidence of clonality in some cases has been reported following s­ plenectomy, solid organ transplantation, allogeneic BM transplantation and autologous stem cell transplantation in multiple myeloma [119]. The latter manifestations of large granular lymphocytosis are usually not associated with cytopenias or autoimmune disorders. The aetiology and pathogenesis of T-LGL leukaemia is due to antigen-driven activation of effector memory T-cells along with increased survival. Major dysfunctions in large granular lymphocytic leukaemia include the disturbed Fas-Fas ligand signalling and activation of the signalling PI3K-Akt and JAK-STAT pathways [116]. Constitutive activation of the STAT3 protein is a recurrent feature in T-LGL leukaemia. Of interest, somatic activating mutations in the SH2 dimerization and activation domain of the STAT3 gene have been demonstrated in 30%–50% of patients with T-LGL leukaemia [120, 121]. Many patients harbour multiple clones each with different STAT3 mutations, of which some have been associated with PRCA and aplastic anaemia [122]. STAT5b mutations have been demonstrated in 2% of the patients and seem to predict a more aggressive course [123]. Karyotypic abnormalities are rare in T-LGL leukaemia. PB findings are pathognomonic but may not be appreciated if the lymphocyte count is low. Large granular lymphocytes are medium-sized cells with round-to-oval nuclei with condensed chromatin and abundant pale cytoplasm with azurophilic granules [115]. Occasionally, a dispersed chromatin pattern with the presence of a nucleolus is documented. In T-LGL leukaemia, large granular lymphocytes are the predominant lymphocyte population, whereas in normal donors they do not exceed 15% of total lymphocytes. In the BM, involvement by T-LGL leukaemia may be minimal. By using immunohistochemical staining 19:53:02

Chapter 9: Mature T-Cell Neoplasms and Natural Killer-Cell Malignancies

for CD3 and cytotoxic molecules such as TIA-1 and granzyme B, interstitial infiltration of single or small clusters of T-cells is often demonstrated and sinusoidal infiltration may be seen as linear arrays of T-cells. In addition, T-LGL may be associated with a left shift in the granulopoiesis. T-LGL cells are terminal effector memory T-cells with the following immunophenotype: CD3+, CD8+, CD57+, CD45RA+, CD45RO−, CD4−, CD27− and CD28−. While most T-LGL are positive for TCRαβ, 10% of the cases express TCRγδ. The majority of T-LGL has aberrant expression of two or more pan T-cell-associated markers. The most characteristic features are decreased expression or loss of CD5 and CD7, the latter findings being observed in 70% of cases [24, 25, 124]. Weak expression of CD2 and bright expression of CD3 is seen in 40% and 30% of the patients, respectively [24]. Although CD57 is a universal immunophenotypic marker of T-LGL leukaemia, it is often only partially positive and occasionally completely negative. In contrast, a uniform positive expression of CD16 or CD56 is reported in approximately 78% of cases with a higher relative number of cases expressing CD16 compared to CD56 [24]. Of interest, the coexpression of CD16 and CD56 seems rare. The analysis of NK receptors (NKRs) such as the lectin-like receptor CD94 and class I MHC receptor molecules including the killer immunoglobulin-like receptors CD158a, CD158b, and CD158e may be helpful in establishing the diagnosis since only small subsets of T-cells and CD57+ T-cells express these markers [125–128]. A homogenous expression of CD94 was found in 42% and 96% of cases in two different studies [24, 123]. The monotypic or less frequently combined expression of two different KIR receptors is a common finding and is observed in more than half of the patients [24, 25]. In addition, the demonstration of cytoplasmic granzyme B and perforin confirms the active cytotoxic phenotype [4]. Last but not least, a search for clonality is recommended to confirm the clonal nature of the large granular lymphocytosis [33, 129]. A typical example of a TCRαβ+ T-LGL leukaemia is shown in Figure 9.3. TCRγδ+ T-LGL leukaemias are predominantly CD8+, displaying a similar immunophenotypic profile as the TCRαβ+ T-LGL [129, 130]. However, rare cases of CD4−CD8− cases are described. In a study of seven cases it was shown that CD4− CD8− TCRγδ leukaemias have similar but also strikingly different features with their CD8+TCRαβ or CD8+TCRγδ counterparts [27]. The former almost never presents with an increase of 10

LGL in the PB despite prominent involvement of the BM and spleen and is associated with a higher incidence of AIHA and PRCA. In addition, they share immunophenotypic features of HSTL being negative for CD57 and CD5. However, γδ + T-LGL leukaemia can be differentiated from HSTL based on morphology and their pattern of infiltration in the BM and the spleen. Intrasinusoidal infiltration of the spleen and BM is characteristic for HSTL while it is much less prominent in γδ + T-LGL leukaemia, and, if present, only as single arrays of T lymphocytes. HSTL comprise medium-sized cells with a blast-like appearance in contrast to the mature morphology of the LGL of TCRγδ+ T-LGL leukaemia. Immunophenotypic features of reactive TCR γδ+ T-cells and a TCRγδ+ T-LGL leukaemia and are shown in Figures 9.4 and 9.5, respectively. Another variant of LGL-leukaemia is the CD4+/ CD8dim+/− TCRαβ T-LGL leukaemia [26]. This variant is clinically different from the CD8+ T-LGL leukaemia. It is mostly associated with other malignancies, including concomitant B-cell lymphoproliferative disorders and non-haematologic malignancies. The triad of cytopenia, autoimmune diseases and splenomegaly of CD8+ T-LGL leukaemia is lacking. Of interest, and in contrast with CD8+ T-LGL, a restricted TCRVβ repertoire with a preferential usage of the TCRVβ 13.1 p ­ rotein has been demonstrated in CD4+ T-LGL leukaemia [131]. This finding, taken together with a high degree of homology of the complementary determining region 3 of the rearranged Vβ genes, suggests an antigen-driven origin. It was indeed demonstrated that monoclonal CD4+ CD8−/dim+ TCR αβ T-LGL leukaemia cells recognise CMV antigens, indicating a role of chronic CMV infections in the ontogenesis of the leukaemia [132].

NK-Cell Lymphoproliferative Disorders Chronic Lymphoproliferative Disorders of NK-Cells The CLPD-NK is a provisional entity in the WHO classification of haematopoietic tumours. It is a rare, indolent disorder with similar clinical disease characteristics and common pathogenesis as the TCRαβ or TCRγδ CD8+ T-LGL leukaemias [14, 29, 133, 134]. In keeping with normal NK-cells, they can be divided into CD56bright and CD56−/dim+ subsets with immunophenotypic features similar to their respective normal ­counterparts. The latter subset seems to be more frequent and accounted 19:53:02

151

1E6

8E5

1E5

6E5 400000

200000

1E6

CD7 APC -A700:FL7-A

1E6

CD3 PC5.5:FL4-A

SS INT LIN:SS-A

Chapter 9: Mature T-Cell Neoplasms and Natural Killer-Cell Malignancies

1E4 1E3 1E2 0

1E4

1E3 0 –1E3

–1E2

0 1E3

1E4

1E5

1E6

CD8 ECD:FL3-A

CD45 K0:FL10-A

–1E2 0 1E2

1E3

1E6

1E6

1E5

1E5

1E4 1E3 1E2 0 –1E2

1E5 1E6

–1E2 0 1E2

1E3

1E4

1E5 1E6

CD2 PC7:FL5-A

(c)

1E4 1E3 0 –1E3

–1E3 0

(d)

1E4

CD5 PB:FL9-A

(b)

CD11c PE:FL2-A

0 1E2

(a)

1E5

1E3

1E4

1E5

1E6

CD4 APC-A750:FL8-A

–1E2

(e)

1E2

1E3

1E4

1E5 1E6

CD57 FITC:FL1-A

Figure 9.3  The following antibody combination was used: CD57FITC-CD11cPE-CD8ECD-CD3PCy5.5-CD2PECy7-CD56APC-CD7APCAF700CD4APCAF750-CD5PB-CD45KO. Data analysis was performed using Infinicyt™ (version 1.7). T-cells were gated as follows: CD3+ events were selected in a CD3 versus SSC dot plot, followed by back-gating on the clustered low FSC and SSC events and CD45bright+ events (Figure 9.3a). Subsequently, total CD8+ events among total T-cells were gated followed by selecting CD5dim (Figure 9.3b) and CD2dim (Figure 9.3c) events. The black, green and red dots correspond to all leucocytes, reactive T-cells and large granular lymphocytes, respectively. The CD8+ large granular lymphocytic cells (Figure 9.3d) are double-positive for CD57 and CD11c (Figure 9.3e). The expression of CD5 and CD7 is decreased (Figure 9.3b and c).

152

for approximately two-thirds of 36 cases in one study [29]. Presenting clinical features are persistent lymphocytosis of large granular lymphocytes, neutropenia and anaemia, splenomegaly and autoimmune diseases. In contrast to T-LGL leukaemia, more patients are asymptomatic at diagnosis and severe neutropenia is less common. The association with rheumatoid arthritis is less prevalent, while the frequency of associated solid neoplasms is higher in CLPD-NK [29]. Of interest, the CD56−/dim+ NK-cell leukaemia is more frequently associated with cytopenia and clinical symptoms than the CD56bright NK-cell leukaemia [29, 135]. CLPD-NK are CD2, CD7 and CD16+ NK-cells showing variable expressions of CD56, CD57 and CD8 and a skewed expression of the KIR receptors CD158a, CD158b and CD158e [136, 137]. Clonal and polyclonal CD56low NK-cells have similar immunophenotypic 10

profiles as late-stage activated NK-cells and are characterised by the decreased expression of CD7, CD38 and CD11b and increased expression of CD2, CD57, CD94, HLA-DR and CD11c compared with NK-cells from healthy adults [15, 29, 135, 138]. Barcena et al. used unsupervised hierarchical clustering analysis of markers assessed by FCM analysis on CD56low NK-cells from patients with monoclonal and polyclonal NK-cell expansions and from normal adults to build a predictive model [139]. It was demonstrated that overexpression CD94, HLA-DR and CD45RO as well as downregulation of CD158a, CD161, CD11b and CD38 are the most relevant parameters to distinguish clonal from polyclonal/normal NK-cells. Moreover, the bright homogenous expression of CD94 and HLA-DR is restricted to clonal CLPD-NK. Furthermore, the value of CD94 and HLA-DR expression in the separation 19:53:02

1E6

1E6

8E5

1E5

1E5

6E5 400000 200000

1E3 1E2 0

1E4 1E3

0

–1E2 –1E2 0 1E2

1E3

1E4 1E5 1E6

–1E2 0 1E2

(b)

CD45 K0:FL10-A

CD8 ECD:FL3-A

(a)

1E4

1E4 1E5

0

1E6

(c)

CD5 PB:FL9-A

1E6

1E6

1E5

1E5

1E4 1E3 0

–1E3

1E3

1E4

1E5 1E6

CD2 PC7:FL5-A

1E4 1E3 0

–1E3 –1E3

(d)

1E3

CD56 APC:FL6-A

0

CD7 APC-A700:FL7-A

1E6

CD3 PC5.5:FL4-A

SS INT LIN:SS-A

Chapter 9: Mature T-Cell Neoplasms and Natural Killer-Cell Malignancies

0

1E3

1E4

1E5 1E6

CD4 APC-A750:FL8-A

–1E2 1E2

(e)

1E3

1E4

1E5 1E6

CD57 FITC:FL1-A

Figure 9.4  The following antibody combination was used: CD57FITC-CD11cPE-CD8ECD-CD3PCy5.5-CD2PECy7-CD56APC-CD7APCAF700CD4APCAF750-CD5PB-CD45KO and CD30FITC-TCRαβPE-CD8ECD-TCRγδPCy5.5-CD3PeCy7-CD1aAPC-CD7APCAF700-CD4APCAF750-CD25PBCD45KO. Data analysis was performed using Infinicyt™ (version 1.7). The reactive TCRγδ+ T-cells were gated as follows: CD3+ events were selected in a CD3 versus SSC dot plot, followed by back-gating on the clustered low FSC and SSC events and CD45bright events (Figure 9.4a). Subsequently, CD3bright/CD5dim (Figure 9.4b) and CD7dim events (Figure 9.4c), among all T-cells, were selected and confirmed to represent TCRγδ+ T-cells (red dots). The TCRγδ+ cells show heterogeneous expression of CD8 and CD57 (Figure 9.4d and e, respectively). The black, green and red dots in the bivariate dot plots correspond to all leucocytes, TCRαβ+ T-cells and TCR γδ+ T-cells, respectively.

of clonal versus polyclonal/normal cases was confirmed in principal component analysis. Of interest, CD56low NK-cells of most monoclonal CLPD-NK are negative for CD158a, CD158b and CD158e. In contrast, more than 50% of the cases with polyclonal expansions of CD56low NK-cells show a predominant expression of one KIR. It is interesting to note that CD56low NK-cells of clonal CLPD-NK acquire immunophenotypic features akin to CD56bright NK-cells.

Aggressive NK-Cell Leukaemia and Extranodal NKTCL Extranodal NKTCL (ENKTCL) and aggressive NK-cell leukaemia (ANKL) are relatively frequent in Asia, Central and South America, accounting for 10% of NHLs but are very uncommon in Europe and 10

North America where they represent B-cells) and variable numbers of histiocytes, eosinophils and plasma cells [2, 3]. The HRS cells are often rosetted (rimmed) by T-cells. Although traditionally CHL has been immunophenotypically characterised by immunohistochemical (IHC) ­ studies, flow cytometry (FCM) has recently been shown capable of ­characterising Hodgkin’s cells with high sensitivity and specificity [4–7]. FCM has the unique ability to immunophenotypically characterise small, distinct populations within a heterogeneous mixture, with the more common clinical application being minimal residual disease (MRD) testing. Given that Hodgkin’s cells (lacunar cells, mummified cells and other mononuclear variants) and Reed-Sternberg cells (­binucleate/ multinucleate forms), collectively referred to as HRS cells, are essentially a ‘rare event’ application, evaluation for CHL by FCM is a well-suited derivative of MRD technology. Since techniques are already available to aid in the morphologic diagnosis of CHL, some may consider FCM irrelevant for diagnosing this disease. However, evaluation for HRS cells by FCM is well-suited to working with small amounts of tissue and offers a faster, more objective and more economical option than immunohistochemistry, while providing additional information to help further refine diagnostic challenges.

11

Historically, FCM evaluation of CHL was considered problematic due to an inability to identify the HRS cells, presumably secondary to fragility of the large HRS cells that were lost during specimen processing [8] in combination with their low number in the specimen. While these factors likely play a role, a perhaps more important feature is the recent recognition that HRS cells in these preparations exist as cellular aggregates. Specifically, T-cells, bound to the HRS cells, result in a composite immunophenotype between the T- and HRS cells, resulting in apparent T-cell antigen expression by the T-cell-HRS-cell aggregates (rosetting). T-cell rosetting may cause exclusion of some of the HRS cells in a typical FCM gating strategy (given they will be omitted when gating on non-coincidence events), in addition to confounding the immunophenotype as the HRS cells may appear to express T-cell antigens and thus may be excluded. T-cell rosetting can be visualised on flowsorted HRS cells when antibodies to disrupt T-cell– HRS-cell rosettes (‘blocking antibodies’) are not used (Figure 10.1). With the advent of newer approaches, the ability to detect low numbers of HRS cells by FCM has proven to be a highly sensitive and specific adjunct to the morphologic impression [4–7].

The Immunophenotype of HRS Cells The ‘classic’ immunophenotype of the HRS cells in tissue sections includes expression of CD15, CD30 and weak PAX5 with lack of CD20 and CD45. Evaluation of the B-cell repertoire and other antigens has demonstrated an absence of other B-cellassociated markers (including CD19, CD22, CD79a, Bob-1, Oct-2, Pu.1) and expression of other markers such as BCL6 and human germinal centre-associated lymphoma, ­ leucocyte-specific protein-1, multiple myeloma oncogene-1, CD40 and CD95 (members of the tumour necrosis ­factor receptor (TNFR) superfamily) [2, 9–12]. 19:54:13

161

Chapter 10: Flow Cytometric Diagnosis of Hodgkin’s Lymphoma in Lymph Nodes

Figure 10.1  Cytomorphologic evaluation of sorted Hodgkin and Reed-Sternberg (HRS) cells, some with T-cell rosettes. HRS cells from classical Hodgkin’s lymphoma cases were purified by flow cytometry (FCM) cell sorting and cytocentrifuge preparations of the resultant cells were made. The first and fourth rows show cells FCM-sorted in the absence of T-cell blocking antibodies and the second and third rows show cells sorted with T-cell blocking antibodies. Rows 1 and 2 are from the same case, demonstrating the effect of blocking antibodies (row 2). ­Modified from [3] (© 2006 American Society for Clinical Pathology).

162

Identification of HRS cells by FCM requires a selection of antigens that the cells express, as well as of antigens expected to be negative, to allow for a refined gating strategy that will ultimately isolate the HRS population hidden within the inflammatory milieu. For example, in addition to the expression of CD15 and CD30, other antigens are commonly present on HRS cells at abnormally high levels including CD40, CD95 and CD71, while CD64 is consistently negative (although HRS cells display autofluorescence in the FL1 channel mimicking weak CD64 expression). CD20 can also be used to exclude normal mature B-cells by bright or uniform expression. CD45 shows more variability in expression by FCM than IHC; however, uniform and bright CD45 is not expected on HRS cells. Evaluation of T-cell markers may seem problematic secondary to the presence of T-cell rosettes giving 11

apparent expression by HRS cells. However, the evaluation of T-cell antigens can be a useful adjunct to the collective immunophenotypic impression, showing both ‘positive’ and negative populations representing the rosetted and non-rosetted HRS cells, respectively. Given that CD40, CD71 and CD95 are not antigens typically evaluated in lymphoid proliferations, we will discuss these three antigens in more detail. CD40 is a member of the TNFR superfamily and is found on the surface of a variety of haematopoietic and epithelial cells. In lymph nodes, CD40 expression is expected on B-cells (involved in germinal centre formation and memory B-cell development), dendritic cells and follicular dendritic cells. CD40 interacts with CD40 ligand, which is also expressed on a variety of cells, but is felt to be the most important for T-cells, as the interaction is critical for a T-cell immune response [13]. CD40 has 19:54:13

Chapter 10: Flow Cytometric Diagnosis of Hodgkin’s Lymphoma in Lymph Nodes

been reported to show strong expression by IHC in all cases of CHL [9] and is therefore felt to be a good marker to exploit by FCM for HRS cells. CD71 (transferrin receptor) is expressed by erythroid lineage and proliferating cells [12]. However, CD71 expression has also been reported in HRS cells (moderate-to-marked intensity in 80%–85% of cases studied) and is expected to be present at low levels in the associated background cells [14]. CD95 (Fas), like CD40, is a member of the TNFR superfamily, and is a cell surface protein which binds CD95 ligand and induces apoptosis [15]. It is expressed at higher levels on activated B and T-cells, as well as various epithelial and connective tissue cells, and the majority of HRS cells [10].

Development of FCM Assay for CHL At the University of Washington, a Hodgkin’s cell line (L1236 cells) was used for the initial development of the FCM assay for CHL, with subsequent validation using residual material from patient samples as positive and negative controls (the morphologic and IHC findings were considered reference standards) [4]. Antigens reportedly expressed by IHC were evaluated by FCM, with HLA-DR, CD15, CD40, CD54, CD58, CD71, CD86 and CD95 showing the strongest expression by the L1236 cells (CD30 showed low-level cytoplasmic expression). When L1236 cells were mixed with normal lymphocytes, T-cell rosetting and apparent T-cell marker ‘expression’ could be identified on subsets of the L1236 cells (while isolated L1236 showed no T-cell marker expression). Given that CD40, CD71 and CD95 are not expected to be as brightly expressed by normal lymphocytes, eosinophils or plasma cells, as by L1236 cells or HRS cells, these markers were used in conjunction with CD20, CD30 and CD45 to isolate the small HRS population with a sequential gating strategy. Based on the L1236 data, cases of CHL, NHL and reactive hyperplasia were evaluated by FCM. Two tubes were used with the following antibody combinations: tube 1 (CD3, CD15, CD19, CD40, CD45, CD64, CD71, CD86, DAPI (4’-6 diamidino-2-phenylindole), HLA-DR) and tube 2 (CD3, CD19, CD20, CD30, CD40, CD45, CD64, CD71, CD95, DAPI), where DAPI (viability dye) is used to exclude non-viable events. Of the 27 CHL cases (diagnosis confirmed by morphologic and IHC findings), 24 cases were considered either diagnostic (n = 18) or suspicious (n = 6) for HRS cells. Diagnostic cases demonstrated a discrete population with the expected immunophenotype: 11

non-rosetted/‘naked’ HRS cells = CD3−, CD15+, CD19−, CD20−, CD30+, CD40+, CD45+/−, CD64−, CD71+, CD95+, HLA-DR+, with a subset of rosetted HRS cells having apparent expression of CD3. None of the reactive lymph nodes (n = 23) or non-CHL lymphoid neoplasms (n = 29) demonstrated a diagnostic or suspicious HRS cell population. Of note, all of the cases were evaluated by FCM prior to the knowledge of the morphologic and IHC impression. Three cases of CHL were considered indeterminate (n = 1) or negative (n = 2) by FCM. Those specimens were felt to be false-negatives secondary to delayed processing and/or degeneration of the HRS cells. Of note, it was determined that the use of blocking antibodies to reduce HRS aggregate formation was not required for HRS cell identification using the reagent combinations employed. Subsequent prospective validation of the Hodgkin’s FCM assay in 420 viable tissue samples was performed, of which 53 samples were morphologically confirmed as CHL [6]. The validation was performed using a single tube nine-colour assay including antibodies to CD5, CD15, CD20, CD30, CD40, CD45, CD64, CD71 and CD95, which demonstrated a sensitivity and specificity of 88.7% and 100%, respectively [6]. In addition, a sixcolour single-tube FCM assay was recently validated that utilised CD3, CD20, CD30, CD40, CD64 and CD95. This panel demonstrated a sensitivity of 85.4% and specificity of 99.7% for the detection of HRS cells [7]. Of note, we have only validated the Hodgkin’s tube in tissue samples, and the assay has not been properly validated in bone marrow aspirate or disaggregated marrow biopsy samples (or other specimen types).

Reagent Selection and Gating Strategy in the FCM Laboratory For a more in-depth discussion of the sample preparation and procedures, we refer the readers elsewhere [16, 17]. As noted earlier, in clinical practice, T-cell rosetting of HRS cells usually does not prevent the immunophenotyping of CHL, and in fact this phenomenon can provide additional supportive evidence for HRS cells. Therefore, we do not routinely use T-cell blocking strategies in our clinical assay, although one would consider using them for purifying HRS cells by FCM cell sorting. The cytometers we use for routine clinical practice (modified LSRII, BD Biosciences, San Jose, CA, USA) are able to simultaneous interrogate up to 19:54:13

163

(C)

CD20 PE-Cy7 CD71 APC-A700

CD15 APC

CD30 PE

CD95 PB

CD5 APC-Cy7

CD20 PE-Cy7

CD71 APC-A700

CD15 APC

CD40 PE-Cy55

CD95 PB

CD5 PE-TR

CD20 PE-Cy7

CD95 PB

CD30 PE

CD64 FITC SSC-H

CD15 APC

CD40 PE-Cy55

CD64 FITC SSC-H

CD45 PE-TR

(B)

CD30 PE

CD40 PE-Cy55

SSC-H

CD45 PE-TR

(A)

CD64 FITC

CD45 APC-H7

Chapter 10: Flow Cytometric Diagnosis of Hodgkin’s Lymphoma in Lymph Nodes

CD71 APC-A700

CD5 APC-Cy7

Figure 10.2  Three representative examples of Hodgkin and Reed–Sternberg (HRS) cells (red population, emphasized) by flow cytometry. In cases 2A and 2B, the population is quite small. These three cases highlight that all HRS cells are positive for CD30, CD40, CD95 and CD71 and are negative for CD64. However, HRS cells show more variation with CD5 (T-cell rosetting), CD15, CD20 and CD45 expression.

164

10 antigens. Our routine CHL panel is a single-tube, nine-colour assay, which includes CD5-PE-TR, CD15APC, CD20-PE-Cy7, CD30-PE, CD40-PE-Cy5.5, CD45-APC-H7, CD64-FITC, CD71-APC-A700 and CD95-PB. HRS cells are typically positive for CD15, CD30, CD40 (bright), CD71 (variable, most moderate to bright), and CD95 (moderate to bright), with CD45 often showing low/variable expression (Figure 10.2). Although CD5 expression is negative on HRS cells, we generally see binding of CD5 by HRS cells in a bimodal pattern secondary to T-cell rosetting (population with and without CD5 ‘expression’) or in a linear expression pattern when evaluating CD5 versus CD45. CD20 should be low to negative and CD64 negative, remembering that HRS cells invariably show increased autofluorescence in the FITC channel (which will appear as low-level CD64 expression). In a typical evaluation, 200,000–500,000 events are collected, if possible. The sequential gating strategy is explained in detail in Figure 10.3. 11

Given that CHL generally has a mixed inflammatory background, the histiocytes present within the sample can be a confounder given that they are larger cells, falling in the same region as HRS cells. The CD30 and CD64 combination is particularly effective in separating the HRS population (CD30+, CD64−) from the histiocytes (CD30−, CD64+), remembering that HRS cells will show increased autofluorescence. This autofluorescence also helps separate the HRS cells from reactive CD30+ immunoblasts which do not show autofluorescence at the level of an HRS cell [16]. As mentioned previously, if a discrete population with increased scatter characteristics, FITC autofluorescence and expression of CD40, CD95 and CD30 without CD64 or CD20 is identified by the gating schema, this population meets diagnostic criteria for HRS cells (Table 10.1). Expression of CD15 certainly makes one more confident in making the assessment of HRS cells. However, given that approximately 20% of cases lack expression by immunohistochemistry, it is not 19:54:13

Chapter 10: Flow Cytometric Diagnosis of Hodgkin’s Lymphoma in Lymph Nodes

Singlets

FSC-H

SSC-H

CD45 APC-H7

Real

FSC-A

CD30 PE

SSC-H

CD40 PE-Cy55

CD64 FITC

CD40 PE-Cy55

FSC-A

CD20 PE-Cy7

CD95 PB

Figure 10.3  Gating strategy for Hodgkin and Reed-Sternberg (HRS) cells. The singlet gate is expanded to capture the larger and rosetted HRS cells (highlighted in red). The viability gate is expanded to capture larger cells (cells with increased forward and side scatter). Notice some events are hard against the far right with extremely high forward scatter. The first inclusive gate selects for cells with increased side scatter compared to the associated lymphocytes. Next CD30+ cells are selected that appear to express dim CD64 but in fact show autofluorescence in the FITC channel. The third step isolates cells that are negative or show low-level CD20 expression and bright expression of CD40. The final gate requires that the HRS cells express CD40 at least as bright as reactive B-lymphocytes and moderate-to-bright CD95.

Table 10.1  Hodgkin and Reed-Sternberg cell diagnostic criteria (all five criteria must be met) 1

Increased light scatter properties (forward scatter and side scatter) compared to normal lymphocytes

2

Form a discrete cluster of events in multidimensional space (≥10 events)

3

Expression of CD30 (moderate to bright), CD40 (bright: the same or greater intensity as reactive B-cells) and CD95 (moderate to bright)

4

Lack expression of intermediate or bright CD20 (variable low expression may be seen)

5

Demonstrate increased autofluorescence in the FITC channel mimicking low CD64 expression

a diagnostic requirement [2]. It should be noted that CD20 is usually not expressed (or expressed weakly), although rare cases show expression approaching that of normal B-cells [6]. When the criteria are met, the findings are consistent with the presence of HRS cells, however, the flow findings must be correlated with the morphologic evaluation (haematoxylin and eosin stained slides), to confirm the diagnosis of CHL. 11

Not only is it important to evaluate for an HRS cell population, there are other findings in the background B-cell or T-cell populations that will further support a diagnosis of CHL, including the exclusion of an abnormal B- or T-cell population. Given that T-cell rosetting is essentially only seen in CHL, nodular lymphocyte predominant Hodgkin’s lymphoma (NLPHL) and some cases of T-cell histiocyte-rich large B-cell lymphoma (TCHRLBCL), this phenomenon can be exploited to further confirm the presence of HRS cells. The identification of T-cell markers and CD45 on the HRS population in a bimodal fashion suggests T-cell rosetting in some, but not all of the HRS cells. The composite HRS cell/T-cell rosette immunophenotype would include the rosetted HRS population (CD5+, CD45high), and the ‘naked’ HRS population (CD5−, CD45low)[6]. CD5 is never expressed on the non-rosetted (dim CD45) HRS population. Comparing CD45 with CD5 is also a useful adjunct, confirming T-cell–HRS interactions. In many of our samples, we see a linear pattern (occasionally going off-scale), when evaluating CD45 versus CD5 [16]. When a diagonal pattern is seen, one usually considers non-specific reactivity, but it is also seen in correlated expression of antigens, as expected 19:54:13

165

CD5 PE-TR

CD45 APC-H7

CD45 PB

CD45 APC-H7

CD45 APC-H7

Chapter 10: Flow Cytometric Diagnosis of Hodgkin’s Lymphoma in Lymph Nodes

CD5 PE-TR

CD5 APC-Cy7

CD5 PE-TR

Figure 10.4  Evaluation of T-cell antigens on Hodgkin and Reed-Sternberg (HRS) cells. Patterns of CD5 and CD45 on HRS cells (red dots), from left to right: (i) few HRS cells with T-cell rosettes, (ii) bimodal pattern of HRS cells, with and without T-cell rosettes, (iii) CD5− HRS cells without rosettes and (iv) linear pattern of HRS cells with and without rosettes.

when T-cells expressing CD5 and CD45 bind HRS cells, contributing to the HRS cell immunophenotype. This pattern represents another manifestation of the T-cell–HRS cell interaction (please refer to Figure 10.4 for dot plots demonstrating the different T-cell–HRS cells interactions).

Potential Immunophenotypic and Morphologic Confounders

166

Potential diagnostic challenges to consider when evaluating for HRS cells by FCM, include anaplastic large cell lymphoma (ALCL), CD30+ diffuse large B-cell lymphoma (DLBCL), CD30+ peripheral T-cell lymphoma (PTCL) and CD30+ reactive immunoblasts. Although ALCL is CD30+, it lacks CD15 and shows more variability with CD40 (Figure 10.5A) [18, 19]. Only about a third of ALCL cases will express CD40, and the expression (when present) is more variable and less intense when compared to CHL. Experience suggests that CD40 expression on ALCL by FCM (if present) is weak and always less than that seen with small, reactive B-cells. As ALCL may show expression of CD71, CD95 and autofluorescence similar to HRS cells, it is extremely important to use the criteria in Table 10.1, that the population must demonstrate CD40 at or greater than the level of reactive B-cells to consider a population HRS cells [7]. Similarly, CD30+ PTCLs evaluated by FCM lack expression of CD40 [6]. Unlike CHL, CD30+ DLBCL is expected to demonstrate preservation of the B-cell repertoire, showing uniform CD19 and/or CD20 expression, and often is surface light-chain restricted. Additionally DLBCL will show more variable expression of CD30, generally not at the level of HRS cells (Figure 10.5B). DLBCLs typically have increased side scatter but not to the level seen with HRS cells. Reactive B-cell and T-cell 11

immunoblasts also show more variable and lower CD30 expression compared to HRS cells, in addition to demonstrating expression of B- or T-cell antigens (Figure 10.5C) [20]. When compared to HRS cells, CD30+ immunoblasts show lower expression of CD40 and CD95, lower light scatter properties and lower autofluorescence in the FITC channel [6]. Two NHLs that pose a significant problem when using this FCM assay are PTCL with HRS-like cells [21, 22] and chronic lymphocytic leukaemia/small lymphocytic lymphoma with HRS-like cells [23, 24], (Figure 10.6). These findings emphasise that a diagnosis of CHL should not be made by FCM in the presence of an abnormal T- or B-cell population and additionally must be correlated with the morphologic examination. NLPHL, TCHRLBCL and primary mediastinal large B-cell lymphoma (PMLBCL) can lead to a morphologic challenge, and at times IHC evaluation can be problematic. The neoplastic cells of NLPHL are expected to display CD20 and lack CD30, features that would exclude them from consideration as HRS cells by FCM. NLPHL cells cannot be consistently identified by the flow assays described in this manuscript. Likewise, if the neoplastic cells of TCHRLBCL are present in sufficient numbers for characterisation, they are expected to express relatively bright and uniform CD20 and CD45 and usually do not express significant CD30. Additionally, CHL should have associated polyclonal B-cells, whereas TCHRLBCL should not have a significant associated polyclonal B-cell population [2]. PMLBCL often shows expression of CD30 and may be surface immunoglobulin negative; however, the population should express uniform CD20 and CD45 with lack of CD15. Additionally, autofluorescence is lower and the expression of CD30 is usually less than what is typically seen in HRS cells – features that distinguish this neoplasm from CHL. 19:54:13

(C)

CD20 PE-Cy7

CD15 APC

CD40 PE-Cy55

CD71 APC-A700

CD30 PE

CD95 PB

CD5 PE-TR

CD20 PE-Cy7

CD15 APC

CD40 PE-Cy55

CD95 PB

CD5 PE-TR

CD20 PE-Cy7

CD71 APC-A700

CD15 APC

CD64 FITC CD64 FITC SSC-H

CD95 PB

CD30 PE

SSC-H

CD45 APC-H7

(B)

CD30 PE

CD40 PE-Cy55

SSC-H

CD45 APC-H7

(A)

CD64 FITC

CD45 APC-H7

Chapter 10: Flow Cytometric Diagnosis of Hodgkin’s Lymphoma in Lymph Nodes

CD71 APC-A700

CD5 PE-TR

+

Figure 10.5  Flow cytometric findings with non-Hodgkin and Reed-Sternberg (HRS) CD30 populations. (A) Anaplastic large cell lymphoma case is shown, where tumour cells are visualised as red dots. Although CD5, CD30, CD71 and CD95 can be observed similar to that of HRS cells, there is no bright CD40 expression. (B) CD30+ diffuse large B-cell lymphoma, tumour cells are shown visualised as red dots. Although CD30, CD40, CD71 and CD95 are at a level that can be seen with HRS cells, the CD20 is bright and side scatter is lower than would be expected for the HRS population. Additionally, there is no variability in CD5 expression. (C) HRS cells are highlighted in red, B-cells in green, T-cells in yellow, monocytes and granulocytes in blue. Notice the T-cell ­immunoblasts (arrow) showing weaker CD30 expression than HRS cells in continuity with the other small reactive lymphocytes without significant ­autofluorescence in the FITC channel. Note some CD5+ HRS cells rosetting in the CD5/CD20 plot.

Diagnostic Utility of the Inflammatory Background in CHL Although the characterisation of hematopoietic neoplasms by immunohistochemistry or FCM usually focuses on the immunophenotype of the neoplastic cells, the background inflammatory milieu is additionally important not only in the biology of the disease, but in diagnosis and classification. Increased CD4 to CD8 ratios have been reported in some but not all studies of reactive T-cells in CHL [3, 25, 26]. Additionally, one manuscript proposed using increased T-regulatory cells, identified by FCM, to support a diagnosis of CHL [27]. We, and others, have identified overexpression 11

of CD7 (as well as other T-cell antigens) and CD45 by CD4+ T-cells in CHL as findings that can suggest or support the diagnosis of CHL [28, 29].

FCM for NLPHL As was noted earlier, the FCM combinations described do not allow for the neoplastic cells of NLPHL to be consistently identified. While outside the scope of this review, unpublished work in our laboratory has identified unique antibody combinations that do afford the detection of these cells. In addition, a feature that can be seen by FCM in NLPHL is related to its associated inflammatory background. Increased numbers 19:54:13

167

(B)

SSC-H

CD19 PE-Cy7

CD20 PE-Cy7

CD15 APC

CD40 PE-Cy55

CD71 APC-A700

Kappa FITC

CD5 PE-TR

CD38 A594

CD95 PB

Lambda PE

CD5 PE-Cy5

CD30 PE

CD10 APC

SSC-H

CD45 APC-Cy7

(A)

CD64 FITC

CD45 APC-H7

Chapter 10: Flow Cytometric Diagnosis of Hodgkin’s Lymphoma in Lymph Nodes

CD20 V450

CD10 APC

Figure 10.6  Presence of Hodgkin and Reed-Sternberg (HRS) cells in chronic lymphocytic leukaemia (CLL)/small lymphocytic lymphoma. Upper panel shows the HRS cell in red (all remaining cells in blue) and the lower panel shows the κ-restricted CD19+ CD20+ CD5+ CLL ­population in blue (λ expressing B-cells = red, T-cells = green).

of dual CD4:CD8+ T-cells have been reported in approximately 60% of NLPHL, accounting for 10%– 38% of the T-cells [30, 31].

Conclusions and Future Directions

168

In routine clinical practice, a diagnosis of CHL should not be based on FCM alone, as the findings must be correlated with histologic evaluation. In addition, we never evaluate for HRS cells in isolation, always in concert with the FCM evaluation for abnormal B-cell and T-cell populations to exclude concurrent NHL. A T-cell tube can be used to evaluate for increased CD7 and CD45 on a subset of the CD4+ T-cells, a finding that may support the diagnosis of CHL. This analysis would prompt the evaluation for HRS cells when a diagnosis of CHL is not suspected. If a distinct population with the expected phenotype is detected that meets all of the criteria in Table 10.1, the presence of an abnormal CD30+ haematopoietic cell population can be reported discussing the meaning of the findings in a comment. However, as mentioned, the cluster must confidently meet the five criteria listed in Table 10.1 to feel the sample is diagnostic for HRS cells and lack an abnormal B- or T-cell population to suggest the immunophenotypic diagnosis of CHL. A diagnosis of CHL must be morphologically confirmed. In cases that are morphologically CHL, where the flow studies demonstrate no abnormal B- or T-cell population with the presence of 11

a diagnostic HRS cell population, experience suggests the diagnosis of CHL can be made without additional IHC studies. An area of potential future interest is the evaluation of markers for prognosis in CHL and NLPHL by FCM. In tissue sections, a variety of markers are of prognostic significance in CHL. Findings reported to be associated with an adverse prognosis include increased tissue macrophages associated with the HRS cells [32], CD20 expression on the HRS cells [33, 34], increased activated cytotoxic T-cells [35] and expression of Bcl-2 by the HRS cells [36]. Reactive B-cells suggest a better prognosis when increased in number [32, 37]. Any of the previously mentioned markers (and others) could be evaluated in patients with CHL to determine if these FCM findings can also predict outcome. Evaluation is underway for novel markers in CHL that will add diagnostic confidence in difficult cases. There are a variety of reports regarding expression of antigens on HRS cells not routinely used in clinical practice, such as chemokine receptors (CCR7 and CXCR4) [38], adhesion molecules (CD54 and CD58) [4], T-cell signalling (PD-L1 and PD-L2) [39] and cyclin-dependent kinases (cyclin E1) [40], some of which have shown potential for diagnostic use (unpublished data from our laboratory). We have recently shown that expression of CD123 on HRS cells can be demonstrated by FCM [41] and continue to evaluate other markers. FCM is a fast, reliable and cost-effective method for 19:54:13

Chapter 10: Flow Cytometric Diagnosis of Hodgkin’s Lymphoma in Lymph Nodes

diagnosing many cases of typical CHL and is a useful adjunct in cases that may pose a diagnostic dilemma. Expanded use of this FCM assay will add confidence to the use of FCM as a principal modality for establishing a diagnosis of CHL. The ability to identify HRS cells for clinical practice certainly has many benefits; however, combining this assay with FCM cell sorting allows for countless possibilities at the research bench [4, 16].

References 1.

R. Kuppers, A. Engert, and M.L. Hansmann. Hodgkin lymphoma. J Clin Invest, 122 (2012), 3439–47. 2. S. Swerdlow, ed. World Health Organization Classification of Tumours of Haematopoietic and Lymphoid Tissues. (2008, IARC press, Lyon, France). 3. S.D. Hudnall, E. Betancourt, E. Barnhart, et al., Comparative flow immunophenotypic features of the inflammatory infiltrates of Hodgkin lymphoma and lymphoid hyperplasia. Cytometry B Clin Cytom, 74 (2008), 1–8. 4. J.R. Fromm, S.J. Kussick, and B.L. Wood. Identification and purification of classical Hodgkin cells from lymph nodes by flow cytometry and flow cytometric cell sorting. Am J Clin Pathol, 126 (2006), 764–80. 5. J.R. Fromm, D.E. Sabath, and B.L. Wood, Diagnostic usefulness of flow cytometry for immunophenotyping classical Hodgkin lymphoma. Am J Clin Pathol, 136 (2011), 157–8. 6. J.R. Fromm, A. Thomas, and B.L. Wood. Flow cytometry can diagnose classical Hodgkin lymphoma in lymph nodes with high sensitivity and specificity. Am J Clin Pathol, 131 (2009), 322–32. 7. J.R. Fromm and B.L. Wood. A six-color flow cytometry assay for immunophenotyping classical Hodgkin lymphoma in lymph nodes. Am J Clin Pathol, 141 (2014), 388–96. 8. J. Irsch, S. Nitsch, M.L. Hansmann, et al. Isolation of viable Hodgkin and Reed-Sternberg cells from Hodgkin disease tissues. Proc Natl Acad Sci U S A, 95 (1998), 10117–22. 9. A. Carbone and A. Gloghini. The role of current and new discriminating markers in the immunodiagnosis of Hodgkin’s disease and other phenotypically related lymphomas. Pathologica, 88 (1996), 169–74. 10. L.H. Kim, G.I. Eow, S.C. Peh, et al. The role of CD30, CD40 and CD95 in the regulation of proliferation and apoptosis in classical Hodgkin’s lymphoma. Pathology, 35 (2003), 428–35. 1 1. N. Masir, T. Marafioti, M. Jones, et al., Loss of CD19 expression in B-cell neoplasms. Histopathology, 48 (2006), 239–46. 12. S. Tedoldi, A. Mottok, J. Ying, et al., Selective loss of B-cell phenotype in lymphocyte predominant Hodgkin lymphoma. J Pathol, 213 (2007), 429–40. 11

13. J. Banchereau, F. Bazan, D. Blanchard, et al., The CD40 antigen and its ligand. Annu Rev Immunol, 12 (1994), 881–922. 14. M.S. Dorreen, J.A. Habeshaw, A.G. Stansfeld, et al. Characteristics of Sternberg-Reed, and related cells in Hodgkin’s disease, an immunohistological study. Br J Cancer, 49 (1984), 465–76. 15. M.E. Peter, R.C. Budd, J. Desbarats, et al., The CD95 receptor, apoptosis revisited. Cell, 129 (2007), 447–50. 16. J.R. Fromm and B.L. Wood, Strategies for immunophenotyping and purifying classical Hodgkin lymphoma cells from lymph nodes by flow cytometry and flow cytometric cell sorting. Methods, 57 (2012), 368–75. 17. D. Wu, B.L. Wood and J.R. Fromm, Flow cytometry for non-Hodgkin and classical Hodgkin lymphoma, in Lymphoma, Methods and Protocols, Methods in Molecular Biology, R. Kuppers, Editor. (2013, Berlin: Springer Science), 1–21. 18. A. Carbone, A. Gloghini, V. Gattei, et al. Expression of functional CD40 antigen on Reed-Sternberg cells and Hodgkin’s disease cell lines. Blood, 85 (1995), 780–9. 19. M.V. Kesler, G.S. Paranjape, S.L. Asplund, et al., Anaplastic large cell lymphoma, a flow cytometric analysis of 29 cases. Am J Clin Pathol, 128 (2007), 314–22. 20. G.H. Segal, C.R. Kjeldsberg, G.P. Smith, et al. CD30 antigen expression in florid immunoblastic proliferations. A clinicopathologic study of 14 cases. Am J Clin Pathol, 102 (1994), 292–8. 21. T.S. Barry, E.S. Jaffe, L. Sorbara, et al. Peripheral T-cell lymphomas expressing CD30 and CD15. Am J Surg Pathol, 27 (2003), 1513–22. 22. L. Quintanilla-Martinez, F. Fend, L.R. Moguel, et al. Peripheral T-cell lymphoma with Reed-Sternberg-like cells of B-cell phenotype and genotype associated with Epstein-Barr virus infection. Am J Surg Pathol, 23 (1999), 1233–40. 23. H. Kanzler, R. Küppers, S. Helmes, et al. Hodgkin and Reed-Sternberg-like cells in B-cell chronic lymphocytic leukemia represent the outgrowth of single germinal-center B-cell-derived clones, potential precursors of Hodgkin and Reed-Sternberg cells in Hodgkin’s disease. Blood, 95 (2000), 1023–31. 24. H. Momose, E.S. Jaffe, S.S. Shin, et al., Chronic lymphocytic leukemia/small lymphocytic lymphoma with Reed-Sternberg-like cells and possible transformation to Hodgkin’s disease. Mediation by Epstein-Barr virus. Am J Surg Pathol, 16 (1992), 859–67. 25. O. Hernandez, T. Oweity and S. Ibrahim, Is an increase in CD4/CD8 T-cell ratio in lymph node fine needle aspiration helpful for diagnosing Hodgkin lymphoma? A study of 85 lymph node FNAs with increased CD4/CD8 ratio. Cytojournal, 2 (2005), 14. 26. J.M. Martin and R.A. Warnke. A quantitative comparison of T-cell subsets in Hodgkin’s 19:54:13

169

Chapter 10: Flow Cytometric Diagnosis of Hodgkin’s Lymphoma in Lymph Nodes

27.

28.

29.

30.

31.

32. 33.

34.

disease and reactive hyperplasia. Frozen section immunohistochemistry. Cancer, 53 (1984), 2450–5. D.S. Bosler, V.K. Douglas-Nikitin, V.N. Harris et al. Detection of T-regulatory cells has a potential role in the diagnosis of classical Hodgkin lymphoma. Cytometry B Clin Cytom, 74 (2008), 227–35. J.R. Fromm, A. Thomas and B.L. Wood. Increased expression of T cell antigens on T cells in classical Hodgkin lymphoma. Cytometry B Clin Cytom, 78 (2010), 387–8. A.C. Seegmiller, N.J. Karandikar, S.H. Kroft et al. Overexpression of CD7 in classical Hodgkin lymphoma-infiltrating T lymphocytes. Cytometry B Clin Cytom, 76 (2009), 169–74. A. Rahemtullah, N.L. Harris, M.E. Dorn et al. Beyond the lymphocyte predominant cell, CD4+CD8+ T-cells in nodular lymphocyte predominant Hodgkin lymphoma. Leuk Lymphoma, 49 (2008), 1870–8. A. Rahemtullah, K.K. Reichard, F.I. Preffer et al. A double-positive CD4+CD8+ T-cell population is commonly found in nodular lymphocyte predominant Hodgkin lymphoma. Am J Clin Pathol, 126 (2006), 805–14. C. Steidl, T. Lee, S.P. Shah et al. Tumor-associated macrophages and survival in classic Hodgkin’s lymphoma. N Engl J Med, 362 (2010), 875–85. C.S. Portlock, G.B. Donnelly, J. Qin, et al., Adverse prognostic significance of CD20 positive ReedSternberg cells in classical Hodgkin’s disease. Br J Haematol, 125 (2004), 701–8. A. Tzankov, J. Krugmann, F. Fend, et al. Prognostic significance of CD20 expression in classical Hodgkin lymphoma, a clinicopathological study of 119 cases. Clin Cancer Res, 9 (2003), 1381–6.

35. J.J. Oudejans, N.M. Jiwa, J.A. Kummer, et al. Activated cytotoxic T cells as prognostic marker in Hodgkin’s disease. Blood, 89 (1997), 1376–82. 36. S.J. Sup, C.A. Alemañy, B. Pohlman, et al., Expression of bcl-2 in classical Hodgkin’s lymphoma, an independent predictor of poor outcome. J Clin Oncol, 23 (2005), 3773–9. 37. B. Chetaille, F. Bertucci, P. Finetti, et al., Molecular profiling of classical Hodgkin lymphoma tissues uncovers variations in the tumor microenvironment and correlations with EBV infection and outcome. Blood, 113 (2009), 2765–75. 38. U.E. Hopken, H.D. Foss, D. Meyer, et al. Up-regulation of the chemokine receptor CCR7 in classical but not in lymphocyte-predominant Hodgkin disease correlates with distinct dissemination of neoplastic cells in lymphoid organs. Blood, 99 (2002), 1109–16. 39. P. Armand, M.A. Shipp, V. Ribrag, et al., Programmed death-1 lockade with pembrolizumab in patients with classical Hodgkin lymphoma after brentuximab vedotin failure. J Clin Oncol, (2016). pii: JCO673467. 40. A. Tzankov, A. Zimpfer, A. Lugli, et al. Highthroughput tissue microarray analysis of G1-cyclin alterations in classical Hodgkin’s lymphoma indicates overexpression of cyclin E1. J Pathol, 199 (2003), 201–7. 41. J.R. Fromm, Flow cytometric analysis of CD123 is useful for immunophenotyping classical Hodgkin lymphoma. Cytometry B Clin Cytom, 80 (2011), 91–9.

170

11

19:54:13

Chapter

11

Minimal Residual Disease in Acute Myeloid Leukaemia Gerrit J. Schuurhuis, Angèle Kelder, Gert J. Ossenkoppele, Jacqueline Cloos and Wendelien Zeijlemaker

Introduction Acute myeloid leukaemia (AML) is characterized by a fast accumulation of blast cells in the bone marrow (BM) and usually also in the peripheral blood (PB). General clinical symptoms are fatigue, infections and bleeding disorders. Hidden under these rather uniform clinical symptoms, there is a huge heterogeneity on the cellular level, both genotypically and phenotypically. Cytogenetic aberrancies have been recognized for a long time and, together with clinical features such as blast count and age, are at the basis of current risk group stratification protocols. Best known are the corebinding protein aberrancies t(8;21)/RUNX1-RUNX1T1, inv(16)/ CBFB-MYH11, complex karyotypes and specific deletions. Over the past two decades there has been an increase in knowledge of molecular aberrancies, which now have found their way to the risk group stratification protocols as well. Mutations in FLT3 and NPM1 are examples, but more recent mutations in IDH, TET and DNMT3A may become part of it too. Flow cytometry (FCM) studies have included numerous studies on the prognostic impact of multidrug resistance-related gene and protein expression together with drug efflux pump activity as well as apoptosis resistance studies. These studies showed very strong independent prognostic impact[1–4], but efforts to implement these factors in risk group stratification were in vain probably due to the need for ample practical experience, which is presumably too labour-intensive for daily practice in a routine laboratory. A similar fate awaited the FCM immunophenotyping, which history goes back to the 1980s. Immunophenotyping is important for the diagnosis of AML, especially to discriminate from acute lymphoblastic leukaemia (ALL), but, despite numerous efforts,

aberrancies did not become part of risk group protocols. In part, the field itself is to blame: for example, the simplest question as to the prognostic impact of one parameter, that is CD34 expression, could not be solved in two decades. Drastic changes in clinical protocols, which might have led to changes in prognostic impact, could not be blamed, since AML treatment did not rigorously change during that time. Although the prognostic significance of expression of individual markers in AML is under debate, the aberrant expression of various markers has led to clinical application: FCM minimal residual disease (MRD) is nowadays used not only for prognostic purposes (risk group definition) but also for clinical decision-making. The basis for immunophenotypical MRD was laid in the 1980s when several pioneers defined combinations of markers to specifically detect leukaemic cells [5, 6]. The term ‘minimal residual disease’ was coined to define conditions under which remaining tumour cells were present at lower percentages compared to the morphologic criterion of 5% blast cells, which had been established as a marker for complete remission (CR). The latter threshold was based on the observation that normal BM may contain up to 5% of normal precursor blast cells [7]. It was subsequently realised that the specificity of immunophenotypical aberrancies used for MRD detection, that is, ‘background’ presence of these aberrancies in normal BM or BM regenerating after therapy, would be the most important potential hurdle for successful application. How slowly new insights may find their way into clinical application is perhaps best illustrated by the fact that the 5% criterion to define a CR is still in use, although times are a-changing [8]. One of the drives to pursue the MRD issue was probably the limited sensitivity of morphology. Also,

We thank Willemijn Scholten, Yvonne Brockhoff, Sander Snel and Dennis Veldhuizen for their valuable practical contributions. 12

19:55:26

171

Chapter 11: Minimal Residual Disease in Acute Myeloid Leukaemia

the sensitive detection of residual cells, far below the level of 5% blasts, could offer a post-diagnosis sumup of effects of all resistance mechanisms [8]. We now know, and we will explain in this chapter, that unfortunately this view is too simplistic. It was not until 1997 that the first MRD study with a considerable number of patients was published [9], followed by others [10–13], ultimately resulting in prospectively assessed MRD [14, 15]. Cytogenetic and molecular aberrations opened the way to molecular MRD assessment, but despite their relatively fast entrance into diagnostic practice, such as t(15;17)/PML-RARα in acute promyelocytic leukaemia, there are still controversies as to the applications of other aberrations used for MRD detection. Molecular MRD in core-binding factor AMLs, especially t(8;21)/RUNX1-RUNX1T1 was informative, although not highly sensitive (three-log reduction cut-off time points which is more or less similar to FCM) [16]. In another study, when evaluated together with inv(16)/CBFB-MYH11, it was considered non-­ informative as compared to FCM [17]. NPM1 mutations have recently been incorporated in clinical decision-making [18, 19]. Other targets for molecular MRD, for example, WT1 gene overexpression [20], with years of often-contradictory results, still have not found their way to clinical decision-making, especially due to specificity problems [21]. In addition, FLT3-ITDs have been described as a target for MRD. Although FLT3-ITD MRD has prognostic impact [22], sensitivity was lower as compared to NPM1 [19], and most importantly, the possibility of disappearance, emergence or a sequence of both [23, 24] prevents its use as a trustworthy target for MRD. Since immunophenotypical MRD is possible for the vast majority of patients, the emphasis of this chapter will be on it. We will summarise technical demands of immunophenotypic MRD and opportunities for prognostic purposes and for clinical decision-making. We will discuss pitfalls and possible ways to circumvent these. In that respect, opportunities of leukaemia stem cell (LSC) assessment and impact of molecular and immunophenotype ‘shifts’ will be discussed in particular.

Immunophenotypic MRD 172

It is still assumed that there are two different app­ roa­ ches to perform immunophenotypic MRD, the so-called leukaemia-associated immunophenotype 12

(LAIP) approach and the “different from normal” approach, although they are basically similar. LAIP are defined based on the presence of particular aberrant combinations of markers, as seen in one or more patients with AML, and that are absent, or present at low frequencies, in normal BM. The particular location of an aberrant immunophenotype in a two or more dimensional space in multiparameter FCM plots is in normal BM referred to as an ‘empty space’. It should be emphasised that the ability to successfully define as many LAIPs as possible is based on a thorough knowledge of normal BM differentiation patterns. Aberrant populations are defined at diagnosis then traced back and quantified after/during therapy. The restriction of the classical LAIP approach is that the LAIPs traced in the post-diagnosis BM after therapy are only those defined to be present already at diagnosis. This is unlike the ‘different from normal’ approach, which bases MRD on differentiation patterns that are different from the patterns in normal or regenerating BM. MRD can thus be assessed without knowledge of aberrancies at diagnosis. Under particular conditions, for example if no diagnosis samples are obtained, this would be an advantage. Nevertheless, also in these cases, for accurate quantification of MRD, identification of cell populations with aberrant immunophenotypes within the aberrant differentiation patterns is necessary. In this perspective, the ‘different from normal’ approach is essentially not different from the LAIP approach if the latter is applied in an unbiased search for all aberrancies possible in AML, irrespective of their actual presence at diagnosis. Since most of the known studies have been carried out using the LAIP approach, we will focus on this, with reference to the ‘different from normal’ approach if appropriate, for example, in case of false-negative MRD results.

Aberrant Immunophenotypes Types of Aberrancies Nowadays the following types of marker aberrancies are recognised: 1. Cross-lineage expression, where antigens of the lymphoid lineage are present on myeloid cells, for example, CD56 on CD13+ cells; 2. Asynchronous expression, where antigens associated with more mature differentiation stages are present on primitive cells (e.g. CD14 on CD34+ cells) or vice versa (CD133 on CD34− cells); 19:55:26

Chapter 11: Minimal Residual Disease in Acute Myeloid Leukaemia

3. Antigen overexpression; 4. Antigen underexpression. Since AML in general represents immature stages of differentiation with the most primitive cells likely playing a role in therapy resistance and sustaining the leukaemia, the inclusion of primitive markers (CD34 and/or CD117 and/or CD133) is in general thought to be a prerequisite for assessment of relevant MRD populations. Gating strategies to identify primitive marker-positive blast cells are illustrated in Figure 11.1. Examples of gating strategies for CD34+ AML and CD34− AML and the different types of aberrancies are shown in Figure 11.2. This figure also shows representative examples of normal BM. Usually, at diagnosis, a leukaemia is characterised by more than one aberrancy, although several authors found that around 10% of the patients did not harbour any usable LAIPs [14, 15, 25]. The immunophenotypical heterogeneity of AML is illustrated by the fact that a huge number of LAIPs can be defined, for example, over 100 in two recent studies [14, 26].

Subjectivity in Defining LAIPs and in Estimating Specificity Despite the fact that different authors use the previously described classification of aberrancies, the relative frequencies of the types of aberrancies differ largely between different studies [14, 25, 26]. For example, Cui et al. found the following sequence of frequencies: lack of antigen expression > asynchronous expression > lineage infidelity > underexpression [26]. Terwijn et al. found: lineage infidelity > lack of antigen expression > asynchronous expression > overexpression [14]. Others found overexpression to be higher than in these studies [11, 27]. Although apparently this did not lead to large differences in the roughly 90% coverage, it illustrates the subjective character of defining suitable LAIPs (Figure 11.3). For example, the use of marker overexpression or marker underexpression may depend on the investigator’s trust in the stability of marker expression, which in turn depends, at least partly, on how stringent the calibration of flow cytometers is. The definition of an empty space is also subjective: in FCM research, performed to molecularly or functionally defined subpopulations, there usually is emphasis on defining ‘real populations’, that is, those obeying the principle ‘homogeneous in light scatter properties’. Thus, whereas in the field of MRD research some define empty as empty, others may define an 12

empty space as a space where either no populations are present at all or, alternatively, no ‘real populations’ are present. This hereby implies that the events that are present are in the outer expression ranges of other well-defined cell populations, thereby making them irrelevant for MRD detection. An example is shown in Figure  11.3, panels D and F for CD7. Since quality (specificity, sensitivity) and thereby applicability of LAIPs depends on such ‘expression’ in non-leukaemic BM, this has great impact on how LAIPs are used in practice, that is, what sensitivity/specificity is needed to report a reliable MRD cell frequency. As elaborated upon later, this has led to a rather conservative approach: a single cut-off value, used for all different LAIPs and in all AML cases, is chosen to define MRD positivity and MRD negativity with the aim of enabling easy use in clinical decision-making. However, the choice for such a broadly applicable cut-off value is thus dictated by the least specific LAIP(s). Hereby, the potency of the most specific LAIPs, which allow for a more precise assessment of MRD (far) below such a threshold, is likely underestimated. In such, MRD is in fact detected but with MRD values below the threshold. As a result, with few exceptions, [12, 28] most of the large studies use cut-off levels in the order of 0.04%–0.1% (of WBC) to define a patient as MRD-‘positive’ or MRD‘negative’. [11, 14, 15, 25, 29] As outlined earlier, it would probably be extra informative if we defined samples as MRD-positive if MRD cell frequency is above a level that is defined for each individual LAIP and is based on the specificity of that particular LAIP in non-AML BM or in regenerating AML BM (Figure 11.4). Doing so, some of present MRD-negative cases would shift to MRDpositive, which in fact may turn out to be highly relevant since a considerable number of patients who are at present defined as MRD-negative, experience a relapse [11, 14, 15, 30]. Thus, several factors may be of influence on the contribution of LAIP expression on normal BM cells. This expression may differ per LAIP, per patient and per time point as suggested, for example, by Figure 11.3(B, D) and later on in Figure 11.4(B, D). Figure 11.4 outlines and illustrates how such situations may be incorporated in correct MRD interpretation and assessment. LAIP on normal BM, under normal or regenerating conditions, when expressed as marker-positive cells per white blood cell (WBC), is usually made up of different components: • First, the frequency of the primitive population (as percentage of WBC) itself. In normal BM, CD34 percentages may differ widely (0.1%–3.8%) [31], 19:55:26

173

0

SSC-A LINEAR 1E2

1E3

1E4

1E2

1E3

1E4

1

1E5

5

Lymphocytes Primitive marker positive blast cells 1

1E5

1E1

1E2

1E3

1E4

1E5

CD45 KO:Horizon V500-A

3 Lymphocytes

SSC-A LINEAR

SSC-A LINEAR

1E1

CD45 KO:Horizon V500-A

0 1E1

CD45 KO:Horizon V500-A

Blast cells

0 1

2

0

19:55:26

1

4

White blood cells

Blasts

0

FSC-H LINEAR

50000 100000 150000 200000 250000

12

SSC-A LINEAR

50000 100000 150000 200000 250000

50000 100000 150000 200000 250000

0

1

All events

50000 100000 150000 200000 250000

SSC-A LINEAR

50000 100000 150000 200000 250000

White blood cells

0

SSC-H LINEAR

50000 100000 150000 200000 250000

All events

50000 100000 150000 200000 250000

Chapter 11: Minimal Residual Disease in Acute Myeloid Leukaemia

1E1

1E2

1E3

CD34:PerCP-A

1E4

1

1E5

6

1E1

1E2

1E3

1E4

1E5

CD34:PerCP-A

Primitive maker (CD34) Figure 11.1  Gating strategy of primitive marker-postive blast cells. In upper left and middle panels, gating of the white blood cells (WBCs) is shown. Debris were excluded in a FSC/SSC plot (upper left panel) and CD45− cells were excluded to select the population of WBCs. In the CD45/SSC plot (upper middle panel) gating of the lymphocytes, often used as a reference population, is shown whereby they are characterised by low SSC and high CD45 expression. Blasts were identified in the upper right panel by moderate CD45 expression and low SSC. Within the population of blast cells, primitive marker (CD34, CD117 or CD133)-positive cells are gated (lower row of plots).

174

Chapter 11: Minimal Residual Disease in Acute Myeloid Leukaemia

also due to variations in WBC which in turn under normal conditions may already vary between 4 and 10 × 109/L. • Second, the expression of the aberrant marker(s) on the primitive population (usually CD34+, CD117+ or CD133+). Primitive, aberrant markerpositive cells, when enumerated as percentage of the total primitive population, show considerable variation between different BM samples [32]. Taking CD34+ LAIPs as an example, in normal BM, CD34+LAIP+ (background) expression as percentage of WBC varies significantly due to variations in both CD34 and WBC counts. Similarly, CD34 percentages and with it the normal part of the CD34+ compartment show large variations in the BM of a CD34+ patient after therapy. As a consequence, with high CD34+ counts in such BM, and with a low percentage of neoplastic cells, the background (non-leukaemic) expression of CD34+

LAIP+ cells as percentage of WBC may largely overrule the neoplastic LAIP+ CD34+ percentage. This is illustrated in Figures 11.4B and 11.4D and Table 11.1. Figure  11.4A,B shows an example in which the aberrancy (Figure 11.4A: CD33+CD13−) has very low background of normal CD34+ cells: 1.87%, with a low level as percentage of WBC (0.013%) (Figure 11.4B). The final MRD percentage (0.02% in Figure 11.4A) is very similar compared to 0.013% LAIP+ seen in the normal BM, and therefore 0.02% likely does not reflect an accurate percentage of MRD. Figure 11.4C, 11.4D shows another example with CD34+CD22+: now the MRD percentage of 0.04% (in Figure  11.4C) is even lower than LAIP+ as percentage of total WBC in normal BM (0.06% in Figure 11.4D). However, when normalising the CD34 percentage in normal BM to that in AML (0.08%), the correction for LAIP+ normal cells completely changes the conclusion about MRD positivity.

Blast cells

SSC-A

SSC-A

SSC-A

White blood cells Lymphocytes Primitive blast cells

2

3 Asynchronous antigen expression

5

CD13:BV421

CD11b:APC CD13:BV421

8

Antigen overexpression

CD13:BV421

9

Lack of antigen expression

CD33:APC

6

CD13:BV421

CD13:BV421

CD34:PerCP

CD56:PE CD13:BV421

CD56:PE

4

NBM

CD11b:APC

Cross-lineage antigen expression

AML

LAIP+ primitive blast cells

CD34:PerCP

CD34:PerCP

1

CD45:Horizon V500

7

CD33:APC

FSC-A

CD13:BV421

10

CD13:BV421

11

Figure 11.2  Gating strategy of leukaemia-associated immunophenotypes (LAIPs) in CD34+ and CD34− acute myeloid leukaemia (AML). A. CD34+ AML. Primitive marker (CD34)-positive cells are identified as shown in Figure 11.1. Panels 1 to 3 show gating of CD34+ cells; panels 4 to 7 show examples of different types of aberrancies in AML; panels 8 to 11 show examples of corresponding expression in normal BM.

12

19:55:26

175

Chapter 11: Minimal Residual Disease in Acute Myeloid Leukaemia

Blast cells

SSC-A

SSC-A

SSC-A

White blood cells

Lymphocytes Primitive blast cells

FSC-A

1

2 CD45:Horizon V500

3

Lack of antigen expression

CD22:PE

7

8

CD13:BV421

CD22:PE

NBM

CD13:BV421

CD34:PerCP

6

CD117PC7

5

CD34:PerCP

AML

SSC-A

SSC-A

Cross-lineage antigen expression

CD133:PE

4

LAIP+ primitive blast cells

CD34:PerCP

9

CD13:BV421

CD13:BV421

Figure 11.2  (cont.) B. CD34− AML. Panels 1 to 3 show gating of CD34− cells (no neoplastic CD34+ cells present); panel 4 shows the CD34− cells to be partly CD133+; panel 5 shows the CD34− to be partly CD117+ cells; panel 6 shows all CD133+ cells to be CD34−; panel 8 shows that in normal BM, all CD133+ cells are CD34+. CD133+CD13+CD34− cells in this AML case, thus display a LAIP in CD34− AML cells. Panel 7 shows that part of the CD117+ cells are CD22+; panels 8 and 9 show that in normal BM CD34+ cells are CD13+ and CD22−. CD117+CD13+CD22+ cells thus display a LAIP in CD34+ AML.

176

Moreover, it shows that MRD percentages below 0.1 % (0.02% and 0.04% in the examples of Figure 11.4 and Table 11.1), instead of being non-informative, may in fact be highly accurate. In such cases with a primitive compartment present, it is the deviation from normal in the percentages of aberrant marker in this compartment that gives clear hints about the presence of MRD. The percentage of LAIPs on the primitive cell compartment can be used as a starting point because in almost all 12

normal BM cases LAIP as percentage of the primitive compartment (CD34, CD117, CD133) is below 10%. Above that percentage, aberrancies may be considered as leukaemia-specific. A preliminary study showed that when taking this 10% cut-off to define background expression on the primitive marker, there was clear prognostic impact of primitive marker-positive (>10%) frequencies [33]. In addition, this approach contributed to prognostic impact as measured with the classical MRD approach [33]. It would thus be of 19:55:26

Chapter 11: Minimal Residual Disease in Acute Myeloid Leukaemia

AML follow up

NBM

CD34 3.70% CD56 on CD34 18.0% CD56 on WBC 0.62%

CD56 PE

CD56 PE (A)

CD13 BV421 Follow up

Follow up

Primitive blast cells

CD13 BV421 NBM

NBM

CD34 0.84% CD7 on CD34 7.40% CD7 on WBC 0.06%

CD34 0.84% CD7 on CD34 2.52% CD7 on WBC 0.02%

CD7:FITC CD13 BV421

Follow up

Follow up

CD34 1.12% CD7 on CD34 5.39% CD7 on WBC 0.06%

CD34 1.12% CD7 on CD34 3.77% CD7 on WBC 0.04%

(D)

CD13 BV421

NBM

NBM

CD34 0.70% CD7 on CD34 12.67% CD7 on WBC 0.09%

CD34 0.70% CD7 on CD34 3.16% CD7 on WBC 0.02%

CD7 PE

CD7:FITC (E)

Lymphocytes

LAIP+ primitive blast cells

(B)

CD34 6.18% CD7 on CD34 29.70% CD7 on WBC 1.36%

CD7:FITC

CD34 6.18% CD7 on CD34 5.55% CD7 on WBC 0.25%

(C)

CD34 0.70% CD56 on CD34 0.38% CD56 on WBC 0.003%

CD13:PerCP

(F)

CD13 BV421

Figure 11.3  Specificity problems associated with low minimal residual disease (MRD) cell frequency. Gating of leukaemia-associated immunophenotypes was performed as illustrated in Figures 11.1 and 11.2. A,B. (A) shows CD56 expression on MRD cells in an acute myeloid leukaemia (AML) that was CD56+ at diagnosis (not shown) and which was high, both on the CD34+ cells (18%) and on white blood cells (WBCs) (0.62%). (B) In normal bone marrow (BM), CD56 is expressed at very low levels (0.38%) on CD34+ cells, while CD56+CD34+ cells, as percentage of WBC are very low (0.003%). In this case there is no doubt about MRD positivity. C,D. In normal BM, there are two types of background events: at relatively low CD13 expression, there is always a CD7+ ‘population’ which is already relatively high, in this case 7.40% on CD34+ cells and 0.06% on WBCs, see (D), green rectangle; (F) shows another example with even higher percentages. The other is at higher CD13 expression; here percentages are relatively low, in this case 2.52% on CD34+ cells and 0.02% on WBCs, see (D), red rectangle; F shows another example. The left (green rectangle) location cannot be used to define MRD. (C) shows CD7 expression on an AML that was CD7+ at diagnosis (not shown): expression (shown in the red rectangle) is high, both on CD34+ cells (29.7%) and on WBCs (1.36%). There is no doubt about MRD positivity. E,F show another AML case (E) and another normal BM (F). Clearly, follow-ups with such percentages of CD7+ cells (red rectangle in (E)) fall in the normal BM range and so in this there is no clear MRD positivity in (E).

177

12

19:55:26

Chapter 11: Minimal Residual Disease in Acute Myeloid Leukaemia

Normal bone marrow

1E4 1E2 1E5

0

1E3

(B)

1E4

1E5

CD33:PC7 1E5

CD34: 0.08% CD22 on CD34: 50% CD22 on WBC: 0.04%

CD34: 0.70% CD22 on CD34: 9.19% CD22 on WBC: 0.06%

1E4

CD22PE

1E5

LAIP+ primitive blast cells

0 1E4

1E3

1E4

Primitive blast cells

0

0

1E2

1E3

1E3

CD33:PC7

1E2

CD22PE

Blast cells

1E3

CD13:PerCP

1E3

CD13:PE

1E2 0

0

(A)

0

(C)

CD34: 0.70% CD13–CD33+ on CD34: 1.87% CD13–CD33+ on WBC: 0.013%

1E5

CD34: 0.08% CD13–CD33+ on CD34: 25% CD13–CD33+ on WBC: 0.02%

1E4

1E5

AML follow-up

1E2

1E3

CD13:BV421

1E4

1E5

0

(D)

1E2

1E3

1E4

1E5

CD13:BV421

Figure 11.4  High specificity of markers allows to define low minimal residual disease (MRD) percentages. Gating of primitive marker-positive blasts and leukaemia-associated immunophenotypes (LAIPs) was performed as illustrated in Figures 11.1 and 11.2. Figures 11.4A and 11.4B show that the aberrancy used (CD13−CD33+) has only a 1.5 higher LAIP+ frequency in the patient sample compared to normal BM (0.02% vs 0.013%). However, the percentage of LAIP+ cells in the CD34+ compartment is much higher in the patient sample: factor 13.4 (25%/1.87%), clearly showing a high percentage of neoplastic cells. The reason for this discrepancy with the LAIP+/ white blood cell (WBC) percentage is the much higher CD34 content of the normal BM: factor 8.75 (0.70%/0.08%). This causes an overestimation of normal LAIP+ cells for the patient sample: normal LAIP+ cells present in the patient sample would be a factor 8.75 lower than the presumed 0.013%: 0.0015%, which is a factor 13.4 (0.02%/0.0015%) lower than the actual 0.02% LAIP+ cells seen in the patient sample in (A). The conclusion would then be: clear MRD both when expressed per WBC and per CD34. Figure 11.4C and 11.4D show a similar approach for CD22 in the same patient. LAIP+ as percentage of WBC is now even a factor 1.5 higher in the normal BM compared to the AML. However, again the LAIP+ content as percentage of the CD34+ compartment is a factor 5.4 (50%/9.19%) higher in the patient sample, showing that a considerable part of the CD34 compartment is neoplastic. Again, when correcting the factor 8.75 difference in CD34 percentage, the corrected LAIP+/WBC value for the normal BM would be 0.06/8.75= 0.006. The actual MRD percentage was 0.04 and thus considerably higher than this normalised normal LAIP+/WBC content. The conclusion would then be (similar to Figure 11.4A): clear MRD both when expressed per WBC and per CD34.

178

particular interest to apply this approach specifically to the individual LAIP. One LAIP may then indicate a neoplastic character when covering only 3% of the primitive marker-positive cells, while for another the 10% cut-off is appropriate. Figure 11.4 shows that the differences in percentage aberrancy on primitive cells between AML and normal BM may be much larger than the differences seen in LAIP as percentage 12

of WBC. This illustrates that WBC independent assessment of MRD may sometimes be more specific and at least helpful in defining true MRD (as percentage of WBC). In the examples of Figures 11.3A, B and 11.4A–D, the percentages of LAIP+ cells among CD34+ cells, without the calculations of Table 11.1, would have led to the conclusion that MRD is positive in these cases. 19:55:26

Chapter 11: Minimal Residual Disease in Acute Myeloid Leukaemia

Table 11.1  Role of normal CD34% and LAIP hereon in defining true MRD

Normal BM of Fig 3B

Normal BM with CD34% normalised to the AML BM in Fig. 3A (0.08%)

CD34: 0.70% CD13–CD33+ on CD34: 1.87% CD13–CD33+ on WBC: 0.013% MRD factor 1,54x higher than NBM

CD34: 0.08% CD13–CD33+ on CD34: 1.87% CD13–CD33+ on WBC: 0.0015% MRD factor 13.3x higher than NBM

Conclusion for A: no clear MRD

Conclusion for A: clear MRD

Normal BM of Fig 3D

Normal BM with CD34% normalised to the AML BM in Fig. 3C (0.08%)

CD34: 0.70% CD22 on CD34: 9.19% CD22 on WBC: 0.06% MRD factor 1.5 lower than NBM

CD34: 0.08% CD22 on CD34: 9.19% CD22 on WBC: 0.0069% MRD factor 5.8 higher than NBM

Conclusion for C: no clear MRD

Conclusion for C: clear MRD

Data from Figure 3 were used for calculations.

It should be emphasised here that with the definition of a true empty space, the considerations just mentioned no longer apply: the sensitivity to quantify specific LAIP+ events then merely depends on the number of cells analysed. With a million events acquired, 10 LAIP+ events then mean a sensitivity of 1:100,000 (10−5). Specificity will likely increase with an increasing number of marker parameters used, especially using exclusion markers, that is, markers that enable to exclude LAIP+, but non-neoplastic, events [34]. Therefore, as argued earlier, if we accept background expression on normal CD34+ cells, and normalise these to 10% for all LAIPs [32], and thereby limit specific MRD detection to the 0.1% level, we may largely underestimate the value of true MRD. On the other hand, when not taking background into consideration, we may overestimate true MRD. Such a dilemma may become highly problematic if background levels in normal BM are high, as in some of the cases in the study by Cui et al. (0.03%–0.77%), especially when the LAIP expression on blast cells is low in some of the cases (3%–81% in the same study) [26]. However, it is important to consider the specificity of each individual LAIP: while most of the LAIPS have background levels of up to 10% on the primitive marker used, some have between 1% and 5% (e.g. CD117+HLA-DR−CD15−, CD117+CD56+, CD34+CD15+), and some marker/ marker combinations may even have 2 were observed for more than one lineage. The panel of markers recommended by EGIL (Table 12.1) is still valid. However, the EGIL scoring system for BAL has often been applied inappropriately by accepting scores equal to 2 as high enough for lineage assignment, while in the original publication it was stressed that only scores >2 should be accepted. Thus, in many publications, cases of AML or ALL with aberrant expression of other lineage markers were considered as BAL. In the 2008 edition of the WHO classification of tumours of haematopoietic origin, new criteria for the identification of MPAL were proposed [1]. 07:15:28

Chapter 12: Ambiguous Lineage and Mixed Phenotype Acute Leukaemia

absence of CD10, the presence of another B-lymphocyte marker is considered enough to establish B-lineage. If CD19 expression is of low intensity, the presence of two other B-lineage-associated markers will be necessary. The markers corroborating B-cell lineage can be chosen between cytoplasmic CD79a, CD22, CD24 or (less frequently expressed in MPAL) CD20 or CD21 (Figure 12.2). The expression of CD10 can also be considered in this context as a B-lineage-related marker. The identification of Pax5 in immunochemistry is another contributory marker for B-lineage assignment.

Table 12.1  European Group for the Immunophenotypic ­Characterization of Leukaemias scoring system for ­biphenotypic* acute leukaemias (BAL) [7]

Points

B-Lineage

T-Lineage

Myeloid

2

CD79 Cμ cCD22

CD3 TCR

MPO

1

CD19 CD10 CD20

CD2 CD5 CD8 CD10

CD13 CD33 CDw65 CD117

0.5

TdT CD24

TdT CD1a CD7

CD14 CD15a CD64

*To establish BAL diagnosis, a score >2 has to be established for at least two lineages.

These criteria also took into account potential cytogenetic anomalies and thus defined MPAL with t(9;22), MPAL with translocations involving the MLL (now KMT2A) gene and MPAL not otherwise specified. In the 2016 WHO update, some modifications have been brought to these criteria and are detailed later, in order to better delineate some difficult cases with ambiguous lineage.

Myeloid Lineage The most significant marker of myeloid lineage is cytoplasmic MPO detected by cytochemistry, IHC or FCM. Great care should be taken in interpreting and reporting MPO positivity in blasts since various thresholds have been reported in the literature (reviewed in [11]). MPO positivity in blasts has to be compared with internal negative and positive controls as illustrated in Figure 12.1. In cases with monocytic differentiation, which may lack MPO, the presence of non-specific esterase by cytochemistry, or more than one marker among cytoplasmic lysozyme or surface CD14, CD11c, CD36 or CD64 by FCM can be used. In some cases with low or disputable MPO positivity, the presence of additional myeloid markers such as CD117, CD13bright or CD33bright may help to conclude to a diagnosis of MPAL, especially in MPAL with co-expression of B-lineage markers.

B-Lineage B-lineage assignment is based on CD19 expression. If the CD19 labelling is bright and associated with CD10, B-lineage marker expression can be concluded. In the 13

T-Lineage The strongest marker indicating T-lineage is the cytoplasmic expression of CD3, which must be demonstrated with a bright fluorochrome such as phycoerythrin or allophycocyanin and appear as a strong labelling (Figure 12.3). The presence of other T-cellassociated markers such as CD2, CD5 or CD7 is not lineage-specific since some of these markers can be seen on myeloid cells in AML with aberrant expression but, together with cyCD3, can contribute to consolidate the co-expression of T-lineage markers on the blasts. The best strategy to investigate for T-lineage is to use two CD3 antibodies conjugated with two different fluorochromes, used respectively before and after permeabilisation of the cells (Figure 12.3). This allows to differentiate between a contingent of persisting mature T-lymphocytes, positive for both markers and the blast population lacking surface CD3 and most often displaying a gradient of cytoplasmic CD3 labelling reaching the intensity expressed by mature T-cells. Only weak expression of cyt. CD3 in a minor fraction of blasts does not fulfil requirements for T-lineage assignment [11].

Subtypes of MPAL Based on these criteria, four types of MPAL can be delineated: B-myeloid, T-myeloid, B-T and, exceptionally, myeloid/B/T. The respective frequencies of these different subtypes have been reported in a number of publications (Table 12.3). Of note, weak MPO expression as a sole MPAL criterion in a case that ­otherwise is a clear-cut B-ALL, must be considered with caution [15]. Similarly, cases of early T-ALL (see also Chapter 6) should not be misinterpreted as MPAL. Also, cases with complex karyotypes included in some series would now rather be classified as AML with myelodysplasia-related changes and MPAL phenotype. 07:15:28

193

Chapter 12: Ambiguous Lineage and Mixed Phenotype Acute Leukaemia

1000

103

800

102

SSC

CD3

600

101

400 100 200 0 0

100

(A)

101 CD45

102

103

1000

103

800

102

SSC

CD3

600

100

101 cMPO

102

103

100

101 MPO

102

103

101

400 100 200 0 0 (B)

100

101 CD45

102

103

Figure 12.1  Strategy to evaluate myeloperoxidase (MPO) expression in leukaemic blasts. A. Acute leukaemia with a low expression of MPO in the blasts population (cyan) is shown. CD3+ lymphocytes (magenta) can be used as negative control, while monocytes (green) and mostly granulocytes (red) provide positive controls. B. Acute leukaemia with a bright expression of MPO in the blasts population (cyan) is shown. CD3+ lymphocytes (magenta) can be used as negative control (note also the CD3− MPO− B-cells), while monocytes (green) and mostly granulocytes (red) provide positive controls over which the blasts appear superimposed.

194

Of interest, a comparison of the EGIL and 2008 WHO classifications showed excellent correlation in a publication by the EGIL in 2011 [3]. Out of 100 cases analysed only 9 would have failed being scored BAL according to EGIL and another 9 cases would have been excluded from the MPAL category by WHO (Table 12.2). In a series by Weinberg et al., 16 of 61 patients (25%) classified as BAL according to EGIL would not fulfil the WHO MPAL criteria while 2 of 45 patients classified as MPAL according to WHO would not be classified as BAL by EGIL criteria [17]. Another series of 26 patients who 13

could be classified as BAL by EGIL criteria described by van den Ancker et al. showed high discrepancy with WHO criteria mostly due to a high frequency of AML with myelodysplasia-related features in the cohort [18].

Proposed Panels for MPAL Diagnosis Examples of panels proposed in the literature that comply with the recommendations mentioned earlier are presented in Table 12.3. It is recommended to rely first on gating out mature cells on a CD45/SSC scattergram 07:15:28

Chapter 12: Ambiguous Lineage and Mixed Phenotype Acute Leukaemia

600

cCD79

SSC

800

400

103

102

102

101

100

200 0

100

101

102

101

100

103

100

CD45

(A) (x 103)

103

cMPO

1000

101

102

103

100

cMPO

250

102

103

CD45

105

CD33

200 104

CD33

150

SSC

101

100

103

CD19

200

50 0

CD34

CD20

0

CD22

–200 2

3

10

10

(B)

4

10

105

–400

3

0

4

10

CD45

10

CD10

105

CD19

Figure 12.2  Examples of B/myeloid mixed phenotype acute leukaemia (MPAL). A. Note a co-expression of cytoplasmic CD79a and ­myeloperoxidase (MPO) in the blast population (cyan). Blasts were also CD19+ (not shown). Lymphocytes (magenta) are used on the right panel to assess the threshold of MPO positivity (black line). The strong positive staining in granulocytes serves as positive control. B. Biclonal MPAL is shown with two clearly separated myeloid and lymphoid clones when displaying CD19 versus CD33 but undiscernible on the CD45/ SSC histogram (left). The third histogram is a radar display gated on the ‘bermudes’ blast area, showing the separation between the myeloid blasts (orange) expressing CD33 and CD34 and the B-lymphoid blasts (purple) co-expressing CD20, CD22 and CD10.

1E4 1E3 –100 0 1E2

Cytoplasmic CD3

1E5

Figure 12.3  Strategy to identify blasts ­expressing intracytoplasmic CD3. The sample is first incubated with an anti-CD3 antibody, which will stain the surface of mature T-cells (green). The sample is then permeabilised and stained with another anti-CD3 antibody, conjugated with a different fluorochrome, which will stain the blasts (dark blue), usually with ­heterogeneous expression pattern. Note that mature T-cells are stained by both antibodies.

–100 0 1E2

1E3

1E4

1E5

Surface CD3

195

13

07:15:28

Chapter 12: Ambiguous Lineage and Mixed Phenotype Acute Leukaemia

Table 12.2  Examples of 8- and 10-colour panels for mixed phenotype acute leukaemia diagnosis

Ref

FITC

PE

Van Dongen et al. [19]

cMPO

cCD79a

Porwit and Rajab [20]

nTdT

cMPO

GEIL #10-colour nTdT or MPO

MPO or CD10

GEIL # 7-colour

cCD13

MPO

ECD

PerCPPC5.5 or PC5.5

PECy7

APC

APCH7 or APCAlexaF700

CD34

CD19

CD7

sCD3

CD14

CD33

CD34

cCD79

cCD22

cCD13

CD33

CD34

cCD79a

cCD22

cCD79a

cCD22

APCAlexaF750

PBl

PAcO or KO

cCD3

CD45

CD19

cCD3

CD45

cCD3

sCD3 or CD11b

CD45

CD19

cCD3

CD45

# Groupe dÉtude Immunologique des Leucémies, France, consensus document, Bordeaux, September 2012.

103

1000 800

102

SSC

cCD3

600

101

400 100

200 0

100

101 CD45

102

103

100

101 cMPO

102

103

Figure 12.4  Example of T/myeloid mixed phenotype acute leukaemia. T/myeloid MPAL with co-expression of cytoplasmic CD3 and myeloperoxidase in the blasts population (cyan) is shown. Residual lymphocytes (magenta) and granulocytes (red) serve as positive and negative controls. 103

E1

103 E1

E2

102

CD10

cCD79a

102

101

100

E2

E3

100

E4

100

101

101

102

103

E3

E4

100

cCD3

196

101

102

103

CD19

Figure 12.5  Example of T/B mixed phenotype acute leukaemia. Blasts (cyan) co-express CD79a and cytoplasmic CD3 (left), and are also positive for CD10 and CD19 (right) (modified from Ref. 11 with permission)

13

07:15:28

Chapter 12: Ambiguous Lineage and Mixed Phenotype Acute Leukaemia

Table 12.3  Some series of mixed phenotype acute leukaemia (MPAL) cases published in the literature

13 07:15:28

No. of cases

Age

B/Myeloid

T/myeloid

B/T

B/T/M

Genetics

Outcome

Reference

117

35 (14–81)

64

38

14

1

BCR/ABL 11q23 Complex Normal Other

14 4 22* 33 19

Better outcome after allo-HSCT

Yan et al. [21]

100

Children/adults 32/68

59

35

4

2

BCR/ABL 11q23 Complex Normal Other

15 6 24* 10 21

Median survival (months) Children 139 (undefined) Adults 11.8

Matutes et al. [3]

95

20 (1–68)

55

31

3

6

BCR/ABL 11q23 Normal Poor risk Intermediate

10 10 32 19* 18

All HSCT 3-year overall survival 67% 3-year leukaemia-free survival 56%

Munker et al. [22]

45

28 (1–78)

30

14

1

27

33 (1–82)

24

3

0

18

≤50 13 >50 5

12

0

5

BCR/ABL 11q23 Normal Other

6 2 5 4

Median progression-free survival 12 months

Weinberg et al. [17]

0

BCR/ABL 11q23 Other

4 1 8

Median progression-free survival 13 months

Deffis-Court et al. [23]

1

BCR/ABL 11q23

7 0

5-year overall survival 48.1% 5-year relapse-free survival 39.7%

Shimizu et al. [24]

*Patients with complex karyotype are now classified as acute myeloid leukaemia with myelodysplasia-related changes and MPAL phenotype. D. A. Arber, A. Orazi, R. Hasserjian, et al. The 2016 Revision to the World Health Organization Classification of myeloid neoplasms and acute leukemia, Blood, 127 (2016), 2391–405.

197

Chapter 12: Ambiguous Lineage and Mixed Phenotype Acute Leukaemia

and use further gating procedures (see Chapter  1, Figure 1.6) Critical in the assessment of MPAL is to ensure the visualisation of markers normally expressed to be mutually exclusive, that is, myeloid and B-lineage, myeloid and T-lineage, B- and T-lineage on the same cell population or to prove that there are two or more blast subsets that would each fulfil criteria for AML or ALL. Examples of various subtypes of MPAL are illustrated in Figures 12.2, 12.4 and 12.5.

References 1.

198

S. H. Swerdlow, E. Campo, N. L. Harris, et al. Who Classification of Tumours of Haematopoietic and Lymphoid Tissues (Lyon: IARC, 2008). 2. D. A. Arber, A. Orazi, R. Hasserjian, et al. The 2016 Revision to the World Health Organization Classification of myeloid neoplasms and acute leukemia, Blood, 127 (2016), 2391–405. 3. E. Matutes, W. F. Pickl, M. Van’t Veer, et al. MixedPhenotype Acute Leukemia: Clinical and Laboratory Features and Outcome in 100 Patients Defined According to the Who 2008 Classification, Blood, 117 (2011), 3163–71. 4. R. Shi, and R. Munker. Survival of Patients with Mixed Phenotype Acute Leukemias: A Large Population-Based Study, Leuk Res, 39 (2015), 606–16. 5. O. Wolach, and R. M. Stone. How I Treat MixedPhenotype Acute Leukemia, Blood, 125 (2015), 2477–85. 6. S. Heesch, M. Neumann, S. Schwartz, et al. Acute Leukemias of Ambiguous Lineage in Adults: Molecular and Clinical Characterization, Ann Hematol, 92 (2013), 747–58. 7. M. C. Bene, G. Castoldi, W. Knapp, et al. Proposals for the Immunological Classification of Acute Leukemias. European Group for the Immunological Characterization of Leukemias (EGIL), Leukemia, 9 (1995), 1783–6. 8. M. C. Bene, T. Nebe, P. Bettelheim, et al. Immunophenotyping of Acute Leukemia and Lymphoproliferative Disorders: A Consensus Proposal of the European Leukemianet Work Package 10, Leukemia, 25 (2011), 567–74. 9. M. C. Bene, and A. Porwit, Acute Leukemias of Ambiguous Lineage, Semin Diagn Pathol, 29 (2012), 12–18. 10. M. C. Bene, M. Bernier, R. O. Casasnovas, et al. Acute Myeloid Leukaemia M0: Haematological, Immunophenotypic and Cytogenetic Characteristics and Their Prognostic Significance: An Analysis in 241 Patients, Br J Haematol, 113 (2001), 737–45. 1 1. A. Porwit, and M. C. Bene, Acute Leukemias of Ambiguous Origin, Am J Clin Pathol, 144 (2015), 361–76. 13

12. U. Deotare, K. W. Yee, L. W. Le, et al. Blastic Plasmacytoid Dendritic Cell Neoplasm with Leukemic Presentation: 10-Color Flow Cytometry Diagnosis and Hypercvad Therapy, Am J Hematol, 91 (2016), 283–6. 13. L. Pagano, C. G. Valentini, S. Grammatico, and A. Pulsoni. Blastic Plasmacytoid Dendritic Cell Neoplasm: Diagnostic Criteria and Therapeutical Approaches, Br J Haematol, 174 (2016), 188–202. 14. G. Narayanan, M. T. Sugeeth, and L. V. Soman. Mixed Phenotype Acute Leukemia Presenting as Leukemia Cutis, Case Rep Med, 2016 (2016), 1298375. 15. M. J. Borowitz. Mixed Phenotype Acute Leukemia, Cytometry B Clin Cytom, 86 (2014), 152–3. 16. N. Harris, E. Jaffe, H. V. J. Stein, and J. W. Vardiman. Pathology and Genetics of Tumors of Hematopoietic and Lymphoid Tissues, (Lyon: IARC, 2001). 17. O. K. Weinberg, M. Seetharam, L. Ren, A. Alizadeh, and D. A. Arber. Mixed Phenotype Acute Leukemia: A Study of 61 Cases Using World Health Organization and European Group for the Immunological Classification of Leukaemias Criteria, Am J Clin Pathol, 142 (2014), 803–8. 18. W. van den Ancker, M. Terwijn, T. M. Westers, et al. Acute Leukemias of Ambiguous Lineage: Diagnostic Consequences of the WHO 2008 Classification, Leukemia, 24 (2010), 1392–6. 19. J. J. van Dongen, L. Lhermitte, S. Bottcher, et al. and Consortium EuroFlow, Euroflow Antibody Panels for Standardized N-Dimensional Flow Cytometric Immunophenotyping of Normal, Reactive and Malignant Leukocytes, Leukemia, 26 (2012), 1908–75. 20. A. Porwit, and A. Rajab. Flow Cytometry Immunophenotyping in Integrated Diagnostics of Patients with Newly Diagnosed Cytopenia: One Tube 10-Color 14-Antibody Screening Panel and 3-Tube Extensive Panel for Detection of MDS-Related Features, Int J Lab Hematol, 37 Suppl 1 (2015), 133–43. 21. L. Yan, N. Ping, M. Zhu, et al. Clinical, Immunophenotypic, Cytogenetic, and Molecular Genetic Features in 117 Adult Patients with MixedPhenotype Acute Leukemia Defined by WHO-2008 Classification, Haematologica, 97 (2012), 1708–12. 22. R. Munker, R. Brazauskas, H. L. Wang, et al. Blood Center for International, and Research Marrow Transplant, Allogeneic Hematopoietic Cell Transplantation for Patients with Mixed Phenotype Acute Leukemia, Biol Blood Marrow Transplant, 22 (2016), 1024–9. 23. M. Deffis-Court, M. Alvarado-Ibarra, G. J. RuizArguelles, et al. Diagnosing and Treating Mixed Phenotype Acute Leukemia: A Multicenter 10-Year Experience in Mexico, Ann Hematol, 93 (2014), 595–601. 24. H. Shimizu, T. Saitoh, S. Machida, et al. Allogeneic Hematopoietic Stem Cell Transplantation for Adult Patients with Mixed Phenotype Acute Leukemia: Results of a Matched-Pair Analysis, Eur J Haematol, 95 (2015), 455–60. 07:15:28

Chapter

13

Flow Cytometry in Myelodysplastic Syndromes Theresia M. Westers and Arjan A. van de Loosdrecht

Introduction Myelodysplastic syndromes (MDSs) comprise a heterogeneous set of clonal disorders that are characterised by ineffective haematopoiesis resulting in cytopenia, morphologic dysplasia in one or more cell lineages and a propensity to transform into acute myeloid ­leukaemia (AML) [1]. In MDS, an altered haematopoietic stem cell is generally considered to be the diseaseinitiating cell, either in primary disease or secondary to chemotherapy-induced alterations. Differentiation of this affected stem cell may result in a disturbed ­haematopoiesis, mainly the myeloid, erythroid and megakaryocytic lineages [2–6]. Reports on clonal evolvement of the lymphoid cells are scarce [2, 7]. According to World Health Organization (WHO) guidelines, the diagnostic work-up of MDS includes evaluation of cytopenias, morphologic evaluation of the peripheral blood (PB) and bone marrow (BM), histopathology and cytogenetics including fluorescence in-situ hybridisation. Results should enable an accurate diagnosis and risk stratification [1, 8]. The morphological definition of MDS is based on the percentage of BM and PB blasts, the degree of dysplasia and the presence of ring sideroblasts. The required percentage of cells with dysplastic features to qualify morphologic dysplasia in one or more of the haematopoietic cell lineages is ≥10% [1]. The WHO classification for myeloid neoplasms has recently been revised [9, 10]. Regarding the application of flow cytometry (FCM) for diagnostic purposes, a special role for FCM might be to define subgroups of MDS within the WHO classification. For example, it is discussed whether or not single-lineage dysplasia and multi-lineage dysplasia based on FCM only may be of value. Single-lineage dyspoiesis in either the granulocytic, erythroid or megakaryocytic lineage has been recognised and defined as MDS with single-lineage dysplasia. However, the clinical impact regarding treatment and/or prognosis is not yet defined. Since single-lineage dysplasia MDS might have diagnostic 14

and therapeutic implications not covered by morphologic criteria [11], FCM may also provide additional information in this respect. Recently, published guidelines for the diagnosis of MDS endorse FCM as an informative tool. In addition, molecular techniques such as single-nucleotide polymorphism analysis and next-generation sequencing may refine diagnosis, prognosis and possibly select patients for specific therapies [8]. Reflecting the heterogeneity of MDS, a large genomic variation is present within the MDS patient population. However, only a few genes are mutated in >10% of the patients [12–17]. Moreover, so called driver mutations seen in MDS are also frequently observed in other myeloid neoplasms, that is, de novo AML and myeloproliferative neoplasms (MPNs). Thus, the specificity of these features for the diagnosis of MDS needs to be elucidated. Differentiation of normal haematopoietic cells is a tightly regulated process which leads to highly conserved antigen expression levels at the different maturational stages [18]. Numerous studies have proven the applicability of FCM in diagnostics of MDS with respect to the identification of aberrancies in characteristic expression patterns of several cell surface antigens reflecting dysplastic features. The main focus of FCM in MDS in the past years has been on the evaluation of aberrancies in the myeloid progenitor compartment, the maturing myelomonocytic compartment and the decrease in progenitor B-cells and, to a lesser extent, the evaluation of the erythroid lineage. An accumulation of aberrancies is not only associated with the presence of dysplasia and the diagnosis of MDS but also with worse prognosis. Furthermore, FCM can identify different risk groups within patients who are stratified according to validated risk classification systems (WHO and the international prognostic scoring systems such as International Prognostic Scoring System (IPSS) and International Prognostic Scoring System-Revised (IPSS-R) [19–25]). 19:59:52

199

Chapter 13: Flow Cytometry in Myelodysplastic Syndromes

FCM enables the evaluation of antigen features ‘different from normal’ with respect to altered distribution of cell subsets or altered levels of antigen expression. These include: overexpression or underexpression of antigens, gain or loss of antigen expression (e.g., expression of lineage infidelity markers) and asynchronous expression (antigens associated with mature cells expressed on immature cells or vice versa). All of these can cause abnormal appearance of differentiation patterns. Evaluation of these patterns and the accompanying antigen expression levels warrants knowledge on normal expression patterns and validated procedures. Normal patterns of antigen expression in various haematopoietic cell populations are presented in detail in Chapter 3.

Practical Considerations for FCM in MDS Guidelines for processing BM samples for FCM in suspected MDS have been described by the International and European Leukaemia Net-associated Working Group focusing on standardisation of FCM in MDS

(IMDSFlow group) [26, 27]. Samples should be processed within 24 hours of the aspirate being drawn. Bulk lysis of mature erythrocytes in the BM aspirate can be performed before the cells are incubated with a panel of monoclonal antibodies. Applied panels may differ with respect to the applicability of 4, 5, 6, 8 or more colour instruments and various fluorochromeconjugated antibodies. A backbone of CD45 in every tube is warranted. Other markers are added to enable adequate analysis of differentiation patterns, for example, CD34/CD117/HLA-DR and CD13/CD11b/ CD16 for myeloid progenitor and neutrophil maturation, respectively [27]. An increase in the number of recurring, backbone, markers in every tube facilitates more accurate gating procedures upon analysis of the BM cellular compartments. Several comprehensive panels have been proposed in literature by various groups (Table 13.1) [28–33]. The application of a fixative (e.g., paraformaldehyde or a commercial reagent) may preserve staining upon delayed acquisition of the data. Note that fixation may alter light scatter properties [34]. Upon acquisition of the data, a minimum of 100,000 CD45+ events is advised. At least 250 CD34+

Table 13.1  Examples of seven-to-ten colour panels proposed in the literature for myelodysplastic syndrome flow cytometry assessment

Euroflow, 2013 [30] 1

PacB

PacO

FITC

PE

PerCPCy5.5

PECy7

APC

APCH7

HLA-DR

CD45

CD16

CD13

CD34

CD117

CD11b

CD10

2

HLA-DR

CD45

CD35

CD64

CD34

CD117

CD300e

CD14

3

HLA-DR

CD45

CD36

CD105

CD34

CD117

CD33

CD71

4

HLA-DR

CD45

TdT

CD56

CD34

CD117

CD7

CD19

5

HLA-DR

CD45

CD15

NG2

CD34

CD117

CD22

CD38

6

HLA-DR

CD45

CD42a CD203c and CD61

CD34

CD117

CD123

CD4

7

HLA-DR

CD45

CD41

CD34

CD117

CD42b

CD9

CD25

Porwit and Rajab, 2015 [32] FITC

PE

PC5.5

PC7

APC

APC-A700 APC-A750 PacB

KrO

Screening

CD4 kappa

CD8 CD3 lambda CD14

ECD

CD33

CD20 CD56

CD34

CD19

CD10

CD5

CD45

1

CD65

CD13

CD33

CD34

CD117

CD7

CD11b

CD16

CD45

CD14

2

CD35

CD64

CD56

CD33

CD34

CD123

CD19

CD38

HLA-DR

CD45

3

CD71

CD11c

CD4

CD33

CD34

CD2

CD10

CD235a

CD15

CD45

Bardet et al., 2015 [33]

200

FITC

PE

PC5.5

PC7

APC

APC-700 APC-A750

CD45

CD10

CD19

CD34

CD56

CD7

14

CD5

19:59:52

Chapter 13: Flow Cytometry in Myelodysplastic Syndromes

events should be acquired for reliable analysis of aberrancies in the progenitor subset [26, 27]. Analysis of subpopulations requires a FCM definition of these populations. Light scatter properties (forward scatter and side scatter (SSC), respectively) and CD45 are the basic components of these definitions in combination with more specific markers. Analysis of the myelomonocytic lineage can be divided into the analysis of its components: the myeloid progenitors, the neutrophils and the monocytes. It is of utmost importance to evaluate these subpopulations in relation to each other to unravel altered differentiation and maturation. Marker expression during the normal differentiation from immature myeloid progenitors towards the more mature neutrophils and monocytes is described in detail in Chapter 3. Interpretation of differentiation patterns during analysis is facilitated by application of fixed colours for subpopulations and fixed orientation of the parameters on the X- and Y-axes (Figure 13.1).

Aberrancies in Myeloid Progenitor Cells The percentage of myeloid progenitor cells (MyPCs) is one of the major diagnostic parameters in FCM assessment of a BM and/or PB specimen. However, it should be noted that a blast cell by morphology does not equal the FCM definition of a MyPC. Beside the role in the quantification of the MyPC population, FCM can characterise aberrant features of MyPC independent of their number and may possibly add significantly to identify subgroups of MDS with prognostic implications. In addition, the presence of circulating MyPCs, even at a very low level, with or without aberrant antigen expression as evaluated by FCM, is likely to impact both diagnostic and prognostic assessments. In MDS, identification of subpopulations may be hampered in case of altered antigen expression, especially upon loss of the more immature (e.g., CD34 and HLA-DR) or gain of maturation-associated markers (e.g., CD11b and CD15). Nevertheless, FCM analysis facilitates enumeration of these progenitor cells despite their altered immunophenotypic profile. It must be kept in mind that quantification may also be hampered by the quality of the sample (e.g., haemodilution). However, quantification of myeloid progenitors in MDS is less important than the detection of aberrancies. The presence of a clonal myeloid progenitor population can lead to alterations in the characteristic 14

patterns of antigen expression (Figures 13.1 and 13.2). Aberrancies in the myeloid progenitor population may be abnormally high or low expression of generally expressed antigens (CD45, CD34, CD117, HLA-DR, CD13, CD33, etc.). It has been shown that the loss of heterogeneity in the pattern of CD13/HLA-DR within CD34+ myeloid progenitors was highly associated with the diagnosis of MDS (Figure 13.2) [35]. Asynchronous marker expression (CD11b, CD10 and CD15) and lineage infidelity marker expression (CD2, CD5, CD7, CD19 and CD56) can also be observed. Similar to AML, CD25 may be aberrantly expressed on MDS progenitor cells [36, 37]. Beside aberrant level of expression, markers can also be expressed in a more homogenous or heterogeneous way. Both homogeneity and heterogeneity of marker expression can be evaluated by means of comparing coefficients of variation (CVs, decreased or increased as compared to normal, respectively). Some aberrant profiles are displayed in Figure 13.1 and 13.2. Furthermore, increased SSC may reflect the presence of slightly granular blasts, whereas a decreased SSC of myeloid progenitors may reflect a more immature, stem cell-like stage. Stem cells are known to have low-to-no CD38 expression. Diminished expression of CD38 is frequent in MDS and it was even proposed to be applied as a single diagnostic marker [38, 39]. However, CD38 alone could not reproduce clear separation between MDS and related, pathological controls (IMDSFlow, unpublished data). In normal BM samples, CD15 may be expressed on the most mature CD34+CD38high myeloid progenitors and should only be considered aberrant if expressed on the more immature CD34+CD38low-neg progenitors. In contrast, normal BM samples may show dim expression of CD7 on the more immature CD34+CD13dim progenitors (Figure 13.2). Interestingly, a more immature appearance of the CD34+ myeloid progenitors as demonstrated by a lower SSC, a homogenous (increase in) CD117 expression, an increase in CD7 or a decrease in CD15 expression, are associated with shorter leukaemiafree survival and overall survival [40, 41]. Notably, CD33 expression can be low because of a polymorphism in the gene. This causes decreased expression not only in the myeloid progenitors but also in the neutrophils and monocytes. Potential aberrancies in CD33 expression on MyPCs should be analysed in the context of its expression on monocytes and MyPCs (Figure 13.3). 19:59:52

201

Chapter 13: Flow Cytometry in Myelodysplastic Syndromes

G

G

M

HLA-DR CD11b:APC-A LOGICAL 0 1E3 1E4 1E5

CD117:PC7-A LOGICAL 0 1E3 1E4 1E5

NBM

NBM

0 1E3 1E4 1E5 HLA-DR:Horizon V450-A

MDS-2

MDS-2

MDS-2

0

1E3

1E4

1E5

CD34 Cy5-5:PerCP-A LOGICAL

0 1E3 1E4 1E5 HLA-DR:Horizon V450-A

0 1E3 1E4 1E5 HLA-DR:Horizon V450-A MDS-3

CD11b:APC-A LOGICAL 0 1E3 1E4 1E5

CD117:PC7-A LOGICAL 0 1E3 1E4 1E5

MDS-3

CD11b:APC-A LOGICAL 0 1E3 1E4 1E5

0 1E3 1E4 1E5 HLA-DR:Horizon V450-A CD117:PC7-A LOGICAL 0 1E3 1E4 1E5

CD117:PC7-A LOGICAL 0 1E3 1E4 1E5

HLA-DR

0 1E3 1E4 1E5 CD34 Cy5-5:PerCP-A LOGICAL

CD117:PC7-A LOGICAL 0 1E3 1E4

1E5

NBM

0 1E3 1E4 1E5 CD34 Cy5-5:PerCP-A LOGICAL CD117:PC7-A LOGICAL 0 1E3 1E4 1E5

MyPC

M

G CD34

202

M

CD11b

MyPC

CD117

CD117

MyPC

0

1E3

1E4

1E5

HLA-DR:Horizon V450-A

MDS-3

0

1E3

1E4

1E5

HLA-DR:Horizon V450-A I

Figure 13.1  Differentiation and maturation patterns of antigen expression in the myelomonocytic lineage. The column on the left displays CD34 versus CD117 staining patterns of these subsets. The middle column displays HLA-DR versus CD117; the column on the right displays HLA-DR versus CD11b. The top row schematically represents the pathways along which the myeloid progenitor cells differentiate towards granulocytes and monocytes (indicated by pink and orange arrows, respectively). A normal bone marrow sample and two myelodysplastic syndrome (MDS) cases are displayed. MDS-2 shows aberrant homogenous and increased expression of CD117 on myeloid progenitor cells and an increased amount of more immature granulocytes. MDS-3 shows loss of HLA-DR expression and increased expression of CD117 on part of the myeloid progenitors. 14

19:59:52

1E5

CD34 Cy5-5:PerCP-A LOGICAL

0

1E3

1E4

1E5 1E4

1E4

1E5

1E5

0

1E3

1E4

1E5

1E5

CD13:APC-A LOGICAL

MDS-5

1E3

1E4

1E5

CD13:APC-A LOGICAL

1E3

1E3

1E4

1E3

1E5

MDS-5

0

1E3

CD13:APC-A LOGICAL MDS-4

0

HLA-DR:Horizon V450-A 1E4

1E4

1E5 1E4

1E3

0

CD7:FITC-A LOGICAL

1E5 1E4 1E3

CD7:FITC-A LOGICAL

0

1E3

HLA-DR:Horizon V450-A

1E4 1E5

0

CD13:APC-A LOGICAL

MDS-5

0

0

1E5

1E4

1E5

CD34 Cy5-5:PerCP-A LOGICAL

NBM

0

1E4

1E4

HLA-DR:Horizon V450-A

1E3

1E3

CD13:APC-A LOGICAL MDS-4

1E3

CD7:FITC-A LOGICAL

1E3 0

0

1E3

0

1E5

0

1E4

NBM

0

CD7:FITC-A LOGICAL

1E4 1E3

1E3

CD34 Cy5-5:PerCP-A LOGICAL MDS-4

1E4

1E5

0

CD7:FITC-A LOGICAL

1E5

NBM

0

CD7:FITC-A LOGICAL

1E5

Chapter 13: Flow Cytometry in Myelodysplastic Syndromes

0

1E3

1E4

1E5

CD13:APC-A LOGICAL

Figure 13.2  Patterns of myeloid- and lymphoid-related antigen expression on myeloid progenitor cells. The column on the left represents a CD34 versus CD7 plot. The middle column displays CD13 versus CD7; the column on the right displays CD13 versus HLA-DR. In all plots CD7 expression on the myeloid progenitor cells as defined in the left column is highlighted in red. A normal bone marrow (NBM) sample and two myelodysplastic syndrome (MDS) cases are displayed. In a NBM sample a few CD34+ myeloid progenitor cells may express CD7. These cells are CD13dim and show heterogeneous HLA-DR expression. MDS-4 shows an increased amount of CD7+ myeloid progenitors. In addition, CD13 expression on these cells is different from normal and HLA-DR expression shows loss of heterogeneity. The same holds true for MDS-5, although differently as compared to MDS-4.

Aberrancies in the Granulocytic Compartment The frequency of neutrophils in relation to the lymphocytes can illustrate the presence of a BM disorder. Neutropenia, as observed in aplastic anaemia and in some cases of MDS, is associated with a largely decreased neutrophil:lymphocyte ratio [19]. Hypogranulation, a key feature of granulopoietic cells in MDS, is reflected by a decrease in SSC. Hypogranulation may interfere in the separation of neutrophils and monocytes. In contrast to monocytes, neutrophils are HLA-DR−, CD15high and CD33int. 14

Maturation from the myeloid progenitors towards segmented neutrophils can be distinguished by different expression levels of HLA-DR, CD117, CD13, CD11b and CD16. A disturbance in the differentiation of neutrophils is reflected by patterns that are different from normal as shown in Figure 13.4. CD16 can be (partly) aberrantly lost or lost due to the existence of a paroxysmal nocturnal haemoglobinuria (PNH) clone. This should be confirmed according to the standard guidelines for analysis of PNH and interpreted as such [42]. The loss of CD10, normally expressed at the more mature stage, is another common aberrancy. However, CD10 expression is also lost on apoptotic neutrophils. 19:59:52

203

Chapter 13: Flow Cytometry in Myelodysplastic Syndromes

MDS-6

G

M

G

M CD33

MyPC

MyPC

MDS-6

MDS-7

CD33 G MDS-6

CD33

G MDS-7

CD33 M MDS-6

CD33

CD33

Density

Density

M NBM

CD33

Density

Density

G NBM

Density

MyPC NBM

CD33

Density

M

CD33

Density

Density

CD33

Density

MyPC

MyPC

CD34

CD34

MyPC

MDS-7 CD34

NBM

CD33

M MDS-7

CD33

Figure 13.3  Expression of CD33 on myeloid progenitor cells, granulocytes and monocytes, polymorphism and aberrant expression. The top row schematically represents the pathways along which the myeloid progenitor cells differentiate towards granulocytes and ­monocytes (indicated by pink and orange arrows, respectively). Histograms of CD33 expression in the described subsets are displayed for a normal bone marrow (NBM) sample and two myelodysplastic syndrome (MDS) cases. Ratios of CD33 expression between these subsets should be constant. MDS-6 (middle column) demonstrates decreased expression in all subsets as compared to the NBM sample (grey and green lines, respectively). The latter is not referred to as aberrant, but as CD33 polymorphism (in contrast to aberrant decrease in one or two of the subsets). MDS-7 (right column) displays increased CD33 expression on a part of the myeloid progenitors and on the granulocytes.

204

Presence of apoptotic cells may change maturation pattern as they show lower CD11b and less CD16 expression than normal neutrophils. Increased percentages of eosinophilic granulocytes may also interfere in maturation patterns and SSC evaluation as they are CD45high, SSChigh, CD13+, CD11b+, CD16− and CD10− (but HLA-DRdim) [27, 43]. Neutrophils in MDS may aberrantly express markers such as CD56 and CD71. CD56 expression on neutrophils often coincides with that on monocytes [33]. In MDS with an isolated chromosome 5 abnormality (MDS del (5q)), an increase in the percentage of CD10+ neutrophils, increased CD15 expression and decreased SSC are frequently seen reflecting more mature neutrophils [44]. However, these features are also linked to haemodilution, a shift to the right 14

regarding neutrophil maturation. The FCM observations in MDS with del (5q) confirm that MDS del (5q) is a multi-lineage disease reflecting stem cell ­dyspoiesis [44]. Another marker that may provide additional information regarding the diagnosis of MDS is myeloidderived nuclear antigen (MNDA). This antigen is expressed in maturing myelomonocytic cells, a subset of lymphocytes and some lymphomas but not in other human cells or tissues. The highest expression is shown in mature granulocytes and monocytes. Several studies have shown subpopulations of neutrophils and monocytes with diminished expression in MDS and AML [45–48]. Interestingly, no differences in MNDA expression were found in chronic myelomonocytic leukaemia (CMML) compared to controls [48]. Analysis 19:59:52

0

1E3

1E4

1E5

1E5 1E4 1E3 0 1E5 1E4 1E3

0

1E3

1E3

1E4

1E3

1E4

1E5

1E5

CD11b:APC-A LOGICAL

1E5

MDS-9

1E4

MDS-9

0

1E5

CD16:FITC-A LOGICAL

1E4

1E5 1E4

1E4

MDS-8

1E5

0 1E3

CD16:FITC-A LOGICAL

1E3

0

CD11b:APC-A LOGICAL

1E5

0

CD13:PE-A LOGICAL

1E5 1E4 1E3 0

MDS-8

CD11b:APC-A LOGICAL

MDS-9

0

0

CD16:FITC-A LOGICAL

1E4

1E5

CD16:FITC-A LOGICAL

CD13:PE-A LOGICAL

1E5

1E3

CD13:PE-A LOGICAL 1E4

1E4

0

1E5 1E4 1E3

CD13:PE-A LOGICAL

0

1E3

1E3

CD11b:APC-A LOGICAL

MDS-8

0

CD11b:APC-A LOGICAL

1E4 1E3

1E5

1E3

1E4

NBM

0

1E3

CD16:FITC-A LOGICAL

CD11b:APC-A LOGICAL

0

NBM

0

1E3

1E4

CD13:PE-A LOGICAL

1E5

NBM

0

CD13:PE-A LOGICAL

1E5

Chapter 13: Flow Cytometry in Myelodysplastic Syndromes

0

1E3

1E4

1E5

CD16:FITC-A LOGICAL

Figure 13.4  Patterns of CD13, CD11b and CD16 expression in the maturing myeloid lineage. In the left column, maturation patterns of the selected neutrophil and progenitor populations are shown in CD16 versus CD13 plots. Arrows indicate normal maturation from progenitor cells towards segmented neutrophils. In the middle column, CD11b versus CD13 is plotted; the right column displays CD16 versus CD11b plots. A normal bone marrow sample and two myelodysplastic syndrome (MDS) cases are displayed. MDS-8 displays loss of CD16 (paroxysmal nocturnal haemoglobinuria excluded), decreased CD11b expression on the CD13−CD11b+ metamyelocyte subset. MDS-9 shows a convex rather than concave CD16 versus CD13 pattern and overexpression of CD11b on the CD16 − (and CD13−) metamyelocyte subset.

of MNDA expression may aid evaluation of dyspoiesis when incorporated in FCM panels [49]. Observed aberrancies in neutrophils may point to dysgranulopoiesis and hence, the existence of MDS. Note that MFC aberrancies in neutrophils are also f­ requently observed in CMML and other MDSs/MPNs [50].

Aberrancies in the Monocytic Lineage FCM analysis of the monocytic lineage in MDS and MDS/MPN can be instrumental since dyspoiesis in these cells by morphology may be difficult to identify. 14

Combinations of the antibodies CD11b and HLA-DR, or CD14 or CD300e with CD36 and/or CD64 enable the discrimination of immature and mature stages of monocytic cells. Aberrancies in the monocytic population may involve all of the previously described markers. Altered distribution of immature and mature monocytic cells should be taken into account when interpreting data for expression levels of maturation stage-specific antigens. Notably, CD14 can be (partly) lost due to the existence of a PNH clone [42]. Aberrancies in monocytes may also concern homogenously increased expression of CD13, and 19:59:52

205

Chapter 13: Flow Cytometry in Myelodysplastic Syndromes

expression of lymphocyte-associated markers CD2 and CD56. The latter are often described in CMML, yet they are not unique for CMML [33, 50–53]. Bardet et al. reported CD56 expression on monocytes (cutoff 30%) on more than half of the CMML and other MDS/MPNs whereas the incidence was approximately 20% of MDS patients and 2% of the controls [33]. Of note, CD56 is frequently seen in regenerating BM after chemotherapy or stem cell transplants, on granulocytes and monocytes during granulocyte colony-­ stimulating factor (G-CSF)-primed stem cell collections and during infections [19, 51, 54, 55]. Upon activation, CD56 expression is often seen in combination with increased CD64 and HLA-DR. Some examples are shown in Figure 13.5. High proportions of classical CD14+/CD16− monocytes in PB (>94%) have been reported to be a diagnostic feature of CMML [56].

Aberrancies in the Erythroid Lineage

206

Erythroid cells have not been routinely included in FCM BM assessments. These cells are selected based on CD45neg-to-dim, scatter profile and absence of myeloid markers. Aberrancies in the erythroid lineage that can be found in MDS are an increased number of nucleated erythroid cells, an abnormal proportion of sequential erythroid differentiation stages and altered expression levels of CD36, CD71 and CD105 [22, 57–65]. In addition, anaemic MDS patients show an increased fraction of CD117+ erythroid precursors [66]. A multicentre study within IMDSFlow indicated increased CV of CD36 and CD71, in combination with a decrease in the MFI of CD71 and a decrease or increase in the frequency of immature CD117+ erythroid progenitors as markers of MDS-associated dyserythropoiesis (examples in Figure 13.6) [67]. In another multicentre study, increased CV of CD36 and CD71 were described as relevant in a diagnostic model for MDS in combination with haemoglobin (Hb) level [31]. Other markers may play a role in FMC assessment of the erythroid lineage. CD35 appears to be an early marker of CD34+ erythroid-committed progenitors; it is expressed before CD105 and CD35 expression remains throughout maturation. Slight alterations have been observed between MDS and controls [66]. Furthermore, overexpression of CD44 was reported to be specific for MDS patients. Interestingly, it was observed in patients with and without dyserythropoiesis 14

by cytomorphology [66]. Detected aberrancies in the erythroid lineage cells may not only complement evaluation of FCM aberrancies in the myeloid lineage but also add to the diagnosis of MDS [68].

Dysmegakaryopoiesis Evaluation of dyspoiesis in the megakaryocytic lineage by FCM is limited, because the cells are too infrequent for reliable analysis. Moreover, binding of platelets to non-megakaryocytic cells may result in inaccurate interpretation of antigen expression levels of platelet-associated antigens. CD41a and CD61 have been applied to evaluate quantitative aberrancies, but FCM appeared to be less sensitive than cytomorphology regarding the evaluation of abnormalities in megakaryocytes (59% cases with increased numbers versus 91% of cases with dysmegakaryopoiesis, respectively) [57]. Platelets may serve as an alternative for megakaryocytes. A FCM study on PB platelets of MDS patients and controls revealed that MDS-specific phenotypes can be seen regarding scatter properties and altered expression levels of CD36, CD61 and/or CD42a, and expression of CD34 [69]. Future analysis of large data sets may reveal the true value of FCM analysis of platelets.

Diagnostic Scores in MDS The four-parameter diagnostic score (also called in the literature ‘Ogata score’) was designed to serve as a simple FCM test for MDS and consists of markers with minimal inter-observer variability [70, 71]. This score consists of: 1. the sideward light scatter of neutrophils (defined as ratio to lymphocyte SSC for internal reference); 2. the percentage of CD34+ MyPCs among all nucleated cells; 3. the percentage of CD34+ B-cell precursors among all CD34+ cells and 4. the CD45 expression of MyPCs (as ratio to lymphocytes’ CD45 expression). Deviations from normal reference values (Table 13.2) are scored as 1 point with a maximum score of 4 points; a score of ≥2 points is indicative of MDS. An example is displayed in Figure 13.7. The specificity and sensitivity of this score were 93% and 70%, respectively, in a multicentre cohort of low-grade MDS (1 log and percentage of subset above reference values). Concomitantly shown labeling on lymphocytes (green dots) helps to better interpret the findings by providing internal positive and negative controls.

of this method that may affect the sensitivity of the score may be: • the selection of nucleated cells without a nuclear dye; • the potential different reference ranges in different centres, especially regarding CD45 and SSC (due to different fluorochromes, instruments and instrument set-up); • the challenging separation of subpopulations in case of hypogranulation; • the underestimation of progenitor cells in case of haemodiluted samples (e.g., due to marrow fibrosis) and CD34− progenitors. Furthermore, a decrease in the percentage of B-cell progenitors has been associated with ageing [72], and hypogranulation (reflected in a decreased SSC) can be seen in other conditions such as infection [19, 73, 74]. However, it was recently shown that the impact of blood 14

dilution on the score was marginal [33]. Interestingly, in some MPN patients an adequate B-cell development is maintained while others have low fraction of B-cell precursors as seen in MDS or CMML [52, 75]. Noteworthy, the ‘Ogata score’ is less applicable in paediatric patients with refractory cytopenia. Due to low cell numbers, not all parameters are reliable and available [76]. In a prospective cohort of children with refractory cytopenia, FCM analysis was shown to be of relevance in addition to histopathology. Comprehensive FMC analysis of myelomonocytic, erythroid and lymphoid cells revealed that the number of aberrancies was significantly higher in children with refractory cytopenia than in healthy controls and in children with aplastic anaemia, but lower than in advanced MDS [77]. Analysis of additional markers may increase sensitivity of FMC analysis in general. A French multicentre study demonstrated that the ‘Ogata score’ sensitivity could indeed benefit from analysis of extra markers. 19:59:52

207

500

Density

100

1E3

300

400

NBM

200

1E4

1E5

NBM

0

0

CD105:PE-A LOGICAL

1E3

1E4

1E5

NBM

0

CD117:PC7-A LOGICAL

Chapter 13: Flow Cytometry in Myelodysplastic Syndromes

1E4

1E5

0

1E3

1E4

0

1E5

1E3

1E4

1E5

CD71:APC-H7-A LOGICAL

MDS-13

MDS-13

MDS-13 400 300

Density

200

1E4 0

100

1E3 1E2

1E3

CD105:PE-A LOGICAL

1E5

CD71:APC-H7-A LOGICAL

1E4

1E5

1E3

CD36:FITC-A LOGICAL

0

0

CD117:PC7-A LOGICAL

0

1E4

0

1E5

1E4

0

1E5

MDS-14

1E4

1E5

1E3

1E4

1E5

CD71:APC-H7-A LOGICAL

1E3

MDS-14

0

0

CD105:PE-A LOGICAL

1E3

1E4

1E5

MDS-14

0

CD117:PC7-A LOGICAL

1E3

CD71:APC-H7-A LOGICAL

Density

1E3

CD36:FITC-A LOGICAL

100 200 300 400 500 600 700

0

0

1E3

1E4

CD36:FITC-A LOGICAL

1E5

0

1E3

1E4

1E5

CD71:APC-H7-A LOGICAL

0

1E3

1E4

1E5

CD71:APC-H7-A LOGICAL

Figure 13.6  Patterns of antigen expression in the erythroid lineage are shown. The left column displays CD36 versus CD117 expression and the middle column displays CD71 versus CD105 expression. The arrows indicate the pathway along which the erythroid progenitor cells differentiate towards erythrocytes. The right column shows histograms of CD71 expression. A normal bone marrow (NBM) sample and two myelodysplastic syndrome (MDS) cases are displayed. MDS-13 shows increased CD117 expression. MDS-14 shows increased CD105 expression, CD36 expression and CD71 expression is lower and more heterogeneous (increased CV) in both MDS cases as compared to the NBM.

208

Addition of the presence of CD7 and/or CD56 on MyPCs and/or CD56 expression on monocytes as an extra point resulted in a sensitivity of 75% in confirmed low-grade MDS, 98% in refractory anaemia with excess of blasts 1 and 2 and 89% in MDS/MPN; the original four-parameter diagnostic score resulted in sensitivities of 65%, 73% and 74%, respectively. The specificity remained similar 89% versus 88% [33]. Other scores taking into consideration more comprehensive FCM panels have been proposed. Wells et al. proposed a composite score that correlated with severity of the disease [19]. This score was 14

Table 13.2  Diagnostic score according to Della Porta et al. [71] and Ogata et al. [70]

Parameters

Reference ranges

Percentage CD34+ myeloid of all nucleated cells

5%

CD45 expression of CD34+ myeloid progenitor as ratio to lymphocytes:   Ly/My

4–7.5

Side scatter of neutrophils as ratio to lymphocytes: Gran/Ly

>6

19:59:52

250000 SSC-A Exp-SSC Low 50000 100000

1E5 1E4 1E3

CD34 Cy5-5:PerCP-A

CD34+B 19%

0

0

50000

CD34+ 0.8%

0

NBM

100000

SSC-A Exp-SSC Low

250000

Chapter 13: Flow Cytometry in Myelodysplastic Syndromes

0

50000 100000

(A)

200000

0

(B)

1E3

1E4

1E5

0

(C)

CD45:Horizon V500-A

1E3

1E4

1E5

CD45:Horizon V500-A

1E4

1E4

1E4 1E3

0

1E5

CD34+ 7.3%

1E3

1E4

1E5

CD19:APC-H7-A LOGICAL

(F)

250000

1E5

50000

100000

CD34 Cy5-5:PerCP-A

MDS-1

1E3

CD45:Horizon V500-A LOGICAL

1E3

250000

0

(E)

CD34+B 0%

100000

25000

0

0

SSC-A Exp-SSC Low

100000

SSC-A Exp-SSC Low

50000

SSC-A Exp-SSC Low

0

0

(D)

CD45 7.2

50000

Density

Density

SSC 9.1

0

CD117:PC7-A LOGICAL

1E5

FSC-A LINEAR

0

(G)

50000 100000

200000

0

(H)

1E3

1E4

0

1E5

(I)

CD45:Horizon V500-A LOGICAL

1E3

1E4

1E5

CD45:Horizon V500-A LOGICAL

Density

Density 0

(J)

50000

100000

SSC-A Exp-SSC Low

0

250000

(K)

1E3

1E4

CD45:Horizon V500-A

1E4

CD45 8.9

1E3

SSC 5.2

0

CD117:PC7-A LOGICAL

1E5

FSC-A LINEAR

0

1E5

(L)

1E3

1E4

1E5

CD19:APC-H7-A LOGICAL

Figure 13.7  ‘Ogata score’ in a normal bone marrow and a Myelodysplastic syndrome (MDS) specimen. A fixed colour code indicates flow cytometrically identifiable subsets, mainly based on CD45 and light scatter properties: lymphocytes in green, granulocytes in pink, monocytes in orange, myeloid progenitor cells in dark blue and lymphoid progenitors in pale blue (panels A and G). Mononuclear cells are plotted in a CD45 versus CD34 plot (panels B and H) in which CD34+ cells are selected. CD34+ cells ­progenitor cells are divided in myeloid and lymphoid progenitors based on their CD45 and side scatter (SSC) properties: CD45dim–intSSCintCD34bright and CD45dimSSClowCD34dim, respectively (panels C and I). Additional markers such as CD117 and CD19 may aid in the identification of these subsets (panels F and L) but be aware of possible aberrant marker expression. Two items of the diagnostic score can be derived from this analysis: the percentage of CD34+ myeloid progenitor cells among all nucleated cells and the percentage of CD34+ B-cell precursors among all CD34+ cells. The other two parameters are calculated from histograms: the peak value of SSC of granulocytes (pink line) defined as ratio to lymphocyte SSC for internal reference (green line, panels D and J) and the CD45 expression of myeloid progenitor cells (blue line) as ratio to lymphocytes’ CD45 expression (green line; panels E and K). Deviations from normal reference values are scored as 1 point with a maximum score of 4 points; a score of ≥2 points is indicative of MDS. The normal bone marrow sample scores 0 points, and the MDS case scores 4 points (reference ranges in Table 13.2). 14

19:59:52

209

Chapter 13: Flow Cytometry in Myelodysplastic Syndromes

refined and applied in several clinical studies [20, 21, 23, 37, 68, 78]. Other groups used different types of marker combinations with variable levels of complexity (some including erythroid markers) that also gave diagnostic and/or prognostic information [22, 25, 40, 58, 60, 61, 68, 79, 80]. As already mentioned, the ‘red score’ reported by the French group, integrated Hb value with information on erythroid dyspoiesis [31].

Integrated Diagnostics Including FCM in MDS: Reporting to Clinicians

210

In patients with cytopenias, the ‘Ogata score’ can be applied for screening purposes. High scores (3 or 4) are associated with high probability of MDS. However, it has to be pointed out that score ‘≥2’ can be seen in reactive conditions, and that some MDS BM samples can have low scores (0–1). Average sensitivity and specificity in several independent cohorts ranged from 49%–70% and 88%–93%, respectively) [31, 33, 70, 71]. In patients in whom clinical data provide strong suspicion of MDS, more comprehensive panels are recommended. High numbers of FMC abnormalities in MDS are associated with cytogenetic abnormalities, transfusion dependency, progressive disease and risk of ­transformation to AML. Reporting FCM results in MDS should be done in an integrated diagnostic report, together with morphological, cytogenetic and/or molecular findings. If the FCM analysis is part of an integrated report, an interpretative comment stating whether the results are consistent with MDS, show limited number of changes seen in MDS or do not show that MDS-related features should be added [81]. If the FCM report is released independently of other diagnostic reports, it should be descriptive and final conclusions concerning MDS diagnosis should be avoided. As for morphology and cytogenetics, there is only a limited negative predictive value when FCM is used as a sole diagnostic technique. As any of the diagnostic techniques, FCM may fail to show informative results in individual patients, because of the heterogeneity of MDS or for technical reasons. As stated earlier, the diagnosis should be an integrated process. However, addition of FCM to morphology in minimal samples, or if smears are of poor quality, can support a diagnosis of MDS or suggest clinical follow-up and repeated laboratory tests at a later time. The added value of FCM results in the diagnosis and classification of MDS varies depending on MDS 14

category and other diagnostic results. In cases with minimal morphological dysplasia and no detected cytogenetic/molecular abnormalities, aberrant FCM findings may support MDS diagnosis. Conversely, normal FCM findings should prompt further investigation for other causes of cytopenias, close follow-up and retesting when clinically indicated [82]. In patients with cytological findings suggesting MDS with single-­lineage dysplasia, aberrant FCM findings in the myeloid or monocytic lineage may indicate multi-lineage dysplasia, which is of prognostic significance. It is important, that even small populations of MyPCs with multiple immunophenotypic aberrant features may indicate a higher risk of progression to AML. Whenever possible, the results should be assessed together with smears and BM biopsy evaluation to exclude the impact of possible BM fibrosis.

Guidance of Treatment Strategies and Disease Monitoring Therapy in MDS is based on risk assessment, either to improve symptoms related to BM failure or to alter the natural course of the disease, that is, to change the risk of leukaemic evolution. Goals of therapy are different in lower-risk patients than in the higher-risk group. In lower risk, the goal is to decrease transfusion needs and transformation to higher-risk disease or AML, as well as to improve survival. In higher risk, the goal is to prolong survival. Current available therapies include growth factor support, lenalidomide, hypomethylating agents such as 5-azacitidine, intensive chemotherapy and allogeneic stem cell transplantation [83]. FCM may serve as guidance for risk-adapted therapeutic strategies. Specific FCM profiles may be found in patients with unique chromosomal abnormalities [84]. To illustrate this, a specific profile was described for the well-defined MDS del (5q) category [44]. This might be of use in either disease monitoring or identification of a subgroup of patients with additional chromosomal abnormalities and/or different clinical responses on treatments such as lenalidomide. Studies are ongoing to define the precise role of FCM in these specific disease entities. Next to that, molecular lesions may aid in the selection of currently available agents or serve as new therapeutic targets. It was shown that aberrant expression of lineage-infidelity markers on myeloid progenitors as assessed by FCM was associated with lack of response or short response duration upon standard treatment with erythropoietin/G-CSF 19:59:52

Chapter 13: Flow Cytometry in Myelodysplastic Syndromes

in low- and int-1-risk MDS and upon the hypomethylating agent azacitidine in int-2 and high-risk MDS [37, 85]. Next to therapy guidance, FCM can be applied to monitor treatment responses in MDS, MDS/MPN and CMML similar to its application in the evaluation of minimal residual disease in AML [86]. Analysis of the total myelomonocytic compartment during azacitidine treatment illustrated a decrease or disappearance of aberrancies in responders versus an increase of aberrancies in non-responders in int-2 and high-risk MDS [36, 37]. In line with this, it was shown that 40% of the non-responders (7/18) to hypomethylating agents in CMML had persistent aberrant expression of CD14, whereas responders had normal CD14 expression on monocytes (11/11) [50]. Since most therapies do not target the affected stem cell, thereby not completely eradicating the disease, the focus of disease monitoring may shift from the total myelomonocytic compartment to the myeloid progenitor compartment or even stem cells in the future.

6.

7.

8.

9.

10.

11.

Acknowledgements Authors are involved in the International and European Leukaemia Net Working Party on the Standardization of Flow Cytometry in MDS (IMDSFlow) and the Working Party on Flow cytometry in MDS of the Dutch Society of Cytometry. The members are gratefully acknowledged for their ongoing contribution to the work in this field.

References 1.

2.

3.

4.

5.

R. Brunning, A. Orazi, U. Germing, et al. Myelodysplastic syndromes/neoplasms. In, Swerdlow et al., editor. WHO Classification of Tumours and Haematopoietic and Lymphoid Tissues. (Lyon, IARC, 2008):88–103. L. Nilsson, I. Astrand-Grundstrom, I. Arvidsson, et al. Isolation and characterization of hematopoietic progenitor/stem cells in 5q-deleted myelodysplastic syndromes, evidence for involvement at the hematopoietic stem cell level. Blood, 96 (2000), 2012–21. L. Nilsson, P. Eden, E. Olsson, et al. The molecular signature of MDS stem cells supports a stem-cell origin of 5q myelodysplastic syndromes. Blood, 110 (2007), 3005–14. B. Will, L. Zhou, T.O. Vogler, et al. Stem and progenitor cells in myelodysplastic syndromes show aberrant stage-specific expansion and harbor genetic and epigenetic alterations. Blood, 120 (2012), 2076–86. W.W. Pang, J.V. Pluvinage, E.A. Price, et al. Hematopoietic stem cell and progenitor cell 14

12.

13. 14. 15.

16.

17.

18.

19.

mechanisms in myelodysplastic syndromes. Proc Natl Acad Sci U S A, 110 (2013), 3011–16. P.S. Woll, U. Kjallquist, O. Chowdhury, et al. Myelodysplastic syndromes are propagated by rare and distinct human cancer stem cells in vivo. Cancer Cell, 25 (2014), 794–808. N.J. White, E. Nacheva, F.A. Asimakopoulos et al. Deletion of chromosome 20q in myelodysplasia can occur in a multipotent precursor of both myeloid cells and B cells. Blood, 83 (1994), 2809–16. L. Malcovati, E. Hellstrom-Lindberg, D. Bowen, et al. Diagnosis and treatment of primary myelodysplastic syndromes in adults, recommendations from the European LeukemiaNet. Blood, 122 (2013), 2943–64. D.A. Arber and R.P. Hasserjian. Reclassifying myelodysplastic syndromes, what’s where in the new WHO and why. Hematology Am Soc Hematol Educ Program, 2015 (2015), 294–8. D.A. Arber, A. Orazi, R.P. Hasserjian, et al. The 2016 revision to the World Health Organization classification of myeloid neoplasms and acute leukemia. Blood, 127 (2016), 2391–405. S. Knipp, C. Strupp, N. Gattermann, et al. Presence of peripheral blasts in refractory anemia and refractory cytopenia with multilineage dysplasia predicts an unfavourable outcome. Leuk Res, 32 (2008), 33–7. S.M. Langemeijer, R.P. Kuiper, M. Berends, et al. Acquired mutations in TET2 are common in myelodysplastic syndromes. Nat Genet, 41 (2009), 838–42. R. Bejar, K. Stevenson, O. Abdel-Wahab, et al. Clinical effect of point mutations in myelodysplastic syndromes. N Engl J Med, 364 (2011), 2496–506. E. Papaemmanuil, M. Cazzola, J. Boultwood, et al. Somatic SF3B1 mutation in myelodysplasia with ring sideroblasts. N Engl J Med, 365 (2011), 1384–95. L. Malcovati, E. Papaemmanuil, D.T. Bowen, et al. Clinical significance of SF3B1 mutations in myelodysplastic syndromes and myelodysplastic/ myeloproliferative neoplasms. Blood, 118 (2011), 6239–46. E. Papaemmanuil, M. Gerstung, L. Malcovati, et al. Clinical and biological implications of driver mutations in myelodysplastic syndromes. Blood, 122 (2013), 3616–27. T. Haferlach, Y. Nagata, V. Grossmann, et al. Landscape of genetic lesions in 944 patients with myelodysplastic syndromes. Leukemia, 28 (2014), 241–7. E.G. van Lochem, V.H.J. van der Velden, H.K. Wind, et al. Immunophenotypic differentiation patterns of normal hematopoiesis in human bone marrow, reference patterns for age-related changes and diseaseinduced shifts. Cytometry B Clin Cytom, 60 (2004), 1–13. D.A. Wells, M. Benesch, M.R. Loken, et al. Myeloid and monocytic dyspoiesis as determined by flow 19:59:52

211

Chapter 13: Flow Cytometry in Myelodysplastic Syndromes

20.

21.

22.

23.

24.

25.

26.

27.

28.

29.

212

cytometric scoring in myelodysplastic syndrome correlates with the IPSS and with outcome after hematopoietic stem cell transplantation. Blood, 102 (2003), 394–403. A.A. van de Loosdrecht, T.M. Westers, A.H. Westra, et al. Identification of distinct prognostic subgroups in low- and intermediate-1-risk myelodysplastic syndromes by flow cytometry. Blood, 111 (2008), 1067–77. B.L. Scott, D.A. Wells, M.R. Loken, et al. Validation of a flow cytometric scoring system as a prognostic indicator for posttransplantation outcome in patients with myelodysplastic syndrome. Blood, 112 (2008), 2681–6. S. Matarraz, A. Lopez, S. Barrena, et al. Bone marrow cells from myelodysplastic syndromes show altered immunophenotypic profiles that may contribute to the diagnosis and prognostic stratification of the disease, a pilot study on a series of 56 patients. Cytometry B Clin Cytom, 78 (2010), 154–68. S.C. Chu, T.F. Wang, C.C. Li, et al. Flow cytometric scoring system as a diagnostic and prognostic tool in myelodysplastic syndromes. Leuk Res, 357 (2011), 868–73. C. Alhan, T.M. Westers, E.M.P. Cremers, et al. High flow cytometric scores identify adverse prognostic subgroups within the revised international prognostic scoring system for myelodysplastic syndromes. Br J Haematol, 167 (2014), 100–9. W. Kern, U. Bacher, C. Haferlach, et al. Multiparameter flow cytometry provides independent prognostic information in patients with suspected myelodysplastic syndromes, a study on 804 patients. Cytometry B Clin Cytom, 88 (2015), 154–64. A.A. van de Loosdrecht, C. Alhan, M.C. Bene, et al. Standardization of flow cytometry in myelodysplastic syndromes, report from the first European LeukemiaNet working conference on flow cytometry in myelodysplastic syndromes. Haematologica, 94 (2009), 1124–34. T.M. Westers, R. Ireland, W. Kern, et al. Standardization of flow cytometry in myelodysplastic syndromes, a report from an international consortium and the European LeukemiaNet Working Group. Leukemia, 26 (2012), 1730–41. W. Kern, U. Bacher, S. Schnittger, et al. Multiparameter flow cytometry reveals myelodysplasia-related aberrant antigen expression in myelodysplastic/ myeloproliferative neoplasms. Cytometry B Clin Cytom, 84 (2013), 194–7. T.M. Westers, V.H.J. van der Velden, C. Alhan, et al. Implementation of flow cytometry in the diagnostic work-up of myelodysplastic syndromes in a multicenter approach, report from the Dutch Working Party on Flow Cytometry in MDS. Leuk Res, 36 (2012), 422–30. 14

30. J.J. van Dongen, L. Lhermitte, S. Bottcher, et al. EuroFlow antibody panels for standardized n-dimensional flow cytometric immunophenotyping of normal, reactive and malignant leukocytes. Leukemia, 26 (2012), 1908–75. 31. S. Mathis, N. Chapuis, C. Debord, et al. Flow cytometric detection of dyserythropoiesis, a sensitive and powerful diagnostic tool for myelodysplastic syndromes. Leukemia, 27 (2013), 1981–7. 32. A. Porwit and A. Rajab. Flow cytometry immunophenotyping in integrated diagnostics of patients with newly diagnosed cytopenia, one tube 10-color 14-antibody screening panel and 3-tube extensive panel for detection of MDS-related features. Int J Lab Hematol, 37 Suppl 1 (2015), 133–43. 33. V. Bardet, O. Wagner-Ballon, J. Guy, et al. Multicentric study underlining the interest of adding CD5, CD7 and CD56 expression assessment to the flow cytometric Ogata score in myelodysplastic syndromes and myelodysplastic/myeloproliferative neoplasms. Haematologica, 100 (2015), 472–8. 34. C. Alhan, T.M. Westers, E.M.P. Cremers, et al. Application of flow cytometry for myelodysplastic syndromes, pitfalls and technical considerations. Cytometry B Clin Cytom, 90 (2016), 358–67. 35. D. Jevremovic, M.M. Timm, K.K. Reichard, et al. Loss of blast heterogeneity in myelodysplastic syndrome and other chronic myeloid neoplasms. Am J Clin Pathol, 142 (2014), 292–8. 36. M. Terwijn, N. Feller, A. van Rhenen, et al. Interleukin-2 receptor alpha-chain (CD25) expression on leukaemic blasts is predictive for outcome and level of residual disease in AML. Eur J Cancer, 45 (2009), 1692–9. 37. C. Alhan, T.M. Westers, L.H. van der Helm, et al. Absence of aberrant myeloid progenitors by flow cytometry is associated with favorable response to azacitidine in higher risk myelodysplastic syndromes. Cytometry B Clin Cytom 86 (2014), 207–15. 38. M.B. Monreal, M.L. Pardo, M.A. Pavlovsky, et al. Increased immature hematopoietic progenitor cells CD34+/CD38dim in myelodysplasia. Cytometry B Clin Cytom, 70 (2006), 63–70. 39. N. Goardon, E. Nikolousis, A. Sternberg, et al. Reduced CD38 expression on CD34+ cells as a diagnostic test in myelodysplastic syndromes. Haematologica, 94 (2009), 1160–3. 40. P. Falco, A. Levis, A. Stacchini, et al. Prognostic relevance of cytometric quantitative assessment in patients with myelodysplastic syndromes. Eur J Haematol, 87 (2011), 409–18. 41. K. Ogata, K. Kakumoto, A. Matsuda, et al. Differences in blast immunophenotypes among disease types in myelodysplastic syndromes, a multicenter validation study. Leuk Res, 36 (2012), 1229–36. 42. D.R. Sutherland, A. Illingworth, M. Keeney, et al. High-sensitivity detection of PNH red blood cells, 19:59:52

Chapter 13: Flow Cytometry in Myelodysplastic Syndromes

43.

44.

45. 46.

47.

48.

49.

50.

51.

52.

53.

red cell precursors, and white blood cells. Curr Protoc Cytom, 72 (2015), 6–30. D. Stachurski, B.R. Smith, O. Pozdnyakova, et al. Flow cytometric analysis of myelomonocytic cells by a pattern recognition approach is sensitive and specific in diagnosing myelodysplastic syndrome and related marrow diseases, emphasis on a global evaluation and recognition of diagnostic pitfalls. Leuk Res, 32 (2008), 215–24. U. Oelschlaegel, T.M. Westers, B. Mohr, et al. Myelodysplastic syndromes with a deletion 5q display a characteristic immunophenotypic profile suitable for diagnostics and response monitoring. Haematologica, 100 (2015), e93–6. A. Pradhan, A. Mijovic, K. Mills, et al. Differentially expressed genes in adult familial myelodysplastic syndromes. Leukemia, 18 (2004), 449–59. R.C. Briggs, K.E. Shults, L.A. Flye, et al. Dysregulated human myeloid nuclear differentiation antigen expression in myelodysplastic syndromes, evidence for a role in apoptosis. Cancer Res, 66 (2006), 4645–51. S.A. Clintock-Treep, R.C. Briggs, K.E. Shults, et al. Quantitative assessment of myeloid nuclear differentiation antigen distinguishes myelodysplastic syndrome from normal bone marrow. Am J Clin Pathol, 135 (2011), 380–5. F. Bellos, T. Alpermann, E. Gouberman, et al. Evaluation of flow cytometric assessment of myeloid nuclear differentiation antigen expression as a diagnostic marker for myelodysplastic syndromes in a series of 269 patients. Cytometry B Clin Cytom, 82 (2012), 295–304. F. Bellos and W. Kern. Flow cytometry in the diagnosis of myelodysplastic syndromes (MDS) and the value of myeloid nuclear differentiation antigen (MNDA). Cytometry B Clin Cytom (2015), DOI: 10.1002/ cyto.b.21190. Q. Shen, J. Ouyang, G. Tang, et al. Flow cytometry immunophenotypic findings in chronic myelomonocytic leukemia and its utility in monitoring treatment response. Eur J Haematol, 95 (2015), 168–76. Y. Xu, R.W. McKenna, N.J. Karandikar, et al. Flow cytometric analysis of monocytes as a tool for distinguishing chronic myelomonocytic leukemia from reactive monocytosis. Am J Clin Pathol, 124 (2005), 799–806. D. Subira, P. Font, L. Villalon, et al. Immunophenotype in chronic myelomonocytic leukemia, is it closer to myelodysplastic syndromes or to myeloproliferative disorders? Transl Res, 151 (2008), 240–5. W. Kern, U. Bacher, C. Haferlach, et al. Acute monoblastic/monocytic leukemia and chronic myelomonocytic leukemia share common immunophenotypic features but differ in the extent of aberrantly expressed antigens and amount of 14

54.

55. 56.

57.

58.

59.

60.

61.

62.

63.

64.

65.

66.

granulocytic cells. Leuk Lymphoma, 52 (2011), 92–100. G. Sconocchia, H. Fujiwara, K. Rezvani, et al. G-CSFmobilized CD34+ cells cultured in interleukin-2 and stem cell factor generate a phenotypically novel monocyte. J Leukoc Biol, 76 (2004), 1214–19. B.L. Wood. Myeloid malignancies, myelodysplastic syndromes, myeloproliferative disorders, and acute myeloid leukemia. Clin Lab Med, 27 (2007), 551–75. D. Selimoglu-Buet, O. Wagner-Ballon, V. Saada, et al. Characteristic repartition of monocyte subsets as a diagnostic signature of chronic myelomonocytic leukemia. Blood, 125 (2015), 3618–26. M. Stetler-Stevenson, D.C. Arthur, N. Jabbour, et al. Diagnostic utility of flow cytometric immunophenotyping in myelodysplastic syndrome. Blood, 98 (2001), 979–87. L. Malcovati, M.G. Della Porta, M. Lunghi, et al. Flow cytometry evaluation of erythroid and myeloid dysplasia in patients with myelodysplastic syndrome. Leukemia, 19 (2005), 776–83. M.G. Della Porta, L. Malcovati, R. Invernizzi, et al. Flow cytometry evaluation of erythroid dysplasia in patients with myelodysplastic syndrome. Leukemia, 20 (2006), 549–55. W. Kern, C. Haferlach, S. Schnittger, et al. Clinical utility of multiparameter flow cytometry in the diagnosis of 1013 patients with suspected myelodysplastic syndrome, correlation to cytomorphology, cytogenetics, and clinical data. Cancer, 116 (2010), 4549–63. I. Lorand-Metze, E. Ribeiro, C.S. Lima, et al. Detection of hematopoietic maturation abnormalities by flow cytometry in myelodysplastic syndromes and its utility for the differential diagnosis with non-clonal disorders. Leuk Res, 31 (2007), 147–55. F. Xu, L. Wu, Q. He, et al. Immunophenotypic analysis of erythroid dysplasia and its diagnostic application in myelodysplastic syndromes. Intern Med J, 42 (2012), 401–11. J. Hu, J. Liu, F. Xue, et al. Isolation and functional characterization of human erythroblasts at distinct stages, implications for understanding of normal and disordered erythropoiesis in vivo. Blood, 121 (2013), 3246–53. M. Fajtova, A. Kovarikova, P. Svec, et al. Immunophenotypic profile of nucleated erythroid progenitors during maturation in regenerating bone marrow. Leuk Lymphoma, 54 (2013), 2523–30. L. Eidenschink Brodersen, A.J. Menssen, J.R. Wangen, et al. Assessment of erythroid dysplasia by ‘difference from normal’ in routine clinical flow cytometry work-up. Cytometry B Clin Cytom, 88 (2015), 125–35. P. Laranjeira, R. Rodrigues, T. Carvalheiro, et al. Expression of CD44 and CD35 during normal and myelodysplastic erythropoiesis. Leuk Res, 39 (2014), 361–70. 19:59:52

213

Chapter 13: Flow Cytometry in Myelodysplastic Syndromes

67. T.M. Westers, E.M.P. Cremers, U. Oelschlaegel, et al. Immunophenotypic analysis of erythroid dysplasia in myelodysplastic syndromes. A report from the IMDSFlow Working Group. Haematologica, 102 (2017) 308–19. 68. E.M.P. Cremers, T.M. Westers, C. Alhan, et al. Implementation of erythroid analysis by flow cytometry in diagnostic models for myelodysplastic syndromes. Haematologica, 102 (2017), 320–6. 69. A.F. Sandes, M. Yamamoto, S. Matarraz, et al. Altered immunophenotypic features of peripheral blood platelets in myelodysplastic syndromes. Haematologica, 97 (2012), 895–902. 70. K. Ogata, M.G. Della Porta, L. Malcovati, et al. Diagnostic utility of flow cytometry in low-grade myelodysplastic syndromes, a prospective validation study. Haematologica, 94 (2009), 1066–74. 71. M.G. Della Porta, C. Picone, C. Pascutto, et al. Multicenter validation of a reproducible flow cytometric score for the diagnosis of low-grade myelodysplastic syndromes, results of a European LeukemiaNET study. Haematologica, 97 (2012), 1209–17. 72. K. Kuranda, J. Vargaftig, P. de la Rochere, et al. Age-related changes in human hematopoietic stem/ progenitor cells. Aging Cell, 10 (2011), 542–6. 73. S. Parmentier, J. Schetelig, K. Lorenz, et al. Assessment of dysplastic hematopoiesis, lessons from healthy bone marrow donors. Haematologica, 97 (2012), 723–30. 74. M.G. Della Porta, E. Travaglino, E. Boveri, et al. Minimal morphological criteria for defining bone marrow dysplasia, a basis for clinical implementation of WHO classification of myelodysplastic syndromes. Leukemia, 29 (2015), 66–75. 75. D. Primo, M.L. Sanchez, A.B. Espinosa, et al. Lineage involvement in chronic myeloid leukaemia, comparison between MBCR/ABL and mBCR/ABL cases. Br J Haematol, 132 (2006), 736–9. 76. A.M. Aalbers, M.M. van den Heuvel-Eibrink, V. de Haas, et al. Applicability of a reproducible flow cytometry scoring system in the diagnosis of refractory cytopenia of childhood. Leukemia, 27 (2013), 1923–5. 77. A.M. Aalbers, M.M. van den Heuvel-Eibrink, I. Baumann, et al. Bone marrow immunophenotyping

78.

79.

80.

81.

82.

83. 84.

85.

86.

by flow cytometry in refractory cytopenia of childhood. Haematologica, 100 (2015), 315–23. E.M.P. Cremers, T.M. Westers, C. Alhan, et al. Multiparameter flow cytometry is instrumental to distinguish myelodysplastic syndromes from nonneoplastic cytopenias. Eur J Cancer, 54 (2016), 49–56. F. Xu, J. Guo, L.Y. Wu, et al. Diagnostic application and clinical significance of FCM progress scoring system based on immunophenotyping in CD34+ blasts in myelodysplastic syndromes. Cytometry B Clin Cytom, 84 (2013), 267–78. A.F. Sandes, D.M. Kerbauy, S. Matarraz, et al. Combined flow cytometric assessment of CD45, HLA-DR, CD34, and CD117 expression is a useful approach for reliable quantification of blast cells in myelodysplastic syndromes. Cytometry B Clin Cytom, 84 (2013), 157–66. A. Porwit, A.A. van de Loosdrecht, P. Bettelheim, et al. Revisiting guidelines for integration of flow cytometry results in the WHO classification of myelodysplastic syndromes-proposal from the International/European LeukemiaNet Working Group for Flow Cytometry in MDS. Leukemia, 28 (2014), 1793–8. W. Kern, C. Haferlach, S. Schnittger, et al. Serial assessment of suspected myelodysplastic syndromes, significance of flow cytometric findings validated by cytomorphology, cytogenetics, and molecular genetics. Haematologica, 98 (2013), 201–7. G. Garcia-Manero. Myelodysplastic syndromes, 2014 update on diagnosis, risk-stratification, and management. Am J Hematol, 89 (2014), 97–108. J.A. Cutler, D.A. Wells, A.A. van de Loosdrecht, et al. Phenotypic abnormalities strongly reflect genotype in patients with unexplained cytopenias. Cytometry B Clin Cytom, 80 (2011), 150–7. T.M. Westers, C. Alhan, M.E.D. Chamuleau, et al. Aberrant immunophenotype of blasts in myelodysplastic syndromes is a clinically relevant biomarker in predicting response to growth factor treatment. Blood, 115 (2010), 1779–84. G.J. Ossenkoppele, A.A. van de Loosdrecht and G.J. Schuurhuis. Review of the relevance of aberrant antigen expression by flow cytometry in myeloid neoplasms. Br J Haematol, 153 (2011), 421–36.

214

14

19:59:52

Chapter

14

Future Applications of Flow Cytometry and Related Techniques Marie-Christine Béné and Francis Lacombe

Introduction Flow cytometry was invented first for cell-sorting, efficient separation of erythrocytes from tumour cells or of  viable Chinese hamster ovaries (CHO) being reported in Science by Mack Fulwyler [1] and shortly after by the group of Leonard A Herzenberg [2]. The technology we know to be associated with the measurement of fluorescent particles (without sorting) will soon celebrate its 60th anniversary, since the first patent was filed in 1968 by Wolfgang Göhde from the University of Münster. The first applications mostly dealt with DNA assessment for the exploration of cell proliferation and sensitivity to chemotherapy in leukaemia [3]. As indicated in the latter publication, the name initially chosen does describe the invention as ‘pulse cytophotometry’. It was only 40 years ago, in 1976 in Pensacola (Florida, USA), that participants of the Fifth American Engineering Foundation Conference on Automated Cytology agreed by consensus on the  term ‘flow cytometry’. Since then, much progress has been achieved and the large and cumbersome instruments that needed gallons of cooling water and constant finetuning have given way to compact table-top units incredibly empty when opened up. Meanwhile, sister ­technologies have emerged, which still have to make their way towards potential clinical daily applications. This chapter gives an overview of innovation in ­multiparameter flow cytometry (FCM) and related methodologies in terms of instruments, software and new applications. In all cases, flow remains the basis of the method, that is, large numbers of cells/particles are evaluated in a short time, driven by a controlled liquid flux aligning them. This notion is very important, since (except for imaging flow cytometry technology) there is no ­morphology/ cytology in this methodology which complements microscopy examinations.

15

Innovative Instrumentation ImageStream®, a Flow Cytometer with Eyes Amnis® ImageStream Mark II Imaging Flow Cytometer is an instrument combining the electronic acquisition of flow cytometry parameters and imaging [4, 5]. The concept of this flow cytometer is basically that of any instrument, except that cameras capture all the parameters of each cell passing in front of each laser to provide huge galleries of all the events analysed. It also uses dark-field and bright-field morphological evaluation of the cells. The galleries displayed allow the evaluation of the shape and granularity of each cell, making many as recognisable as they would be in phase contrast. Morphological images are comparable to 40× microscopy with approximately 1 μm resolution. In parallel, each fluorescent channel is displayed, allowing for assessment of marker expression or not. Incidentally, since compensations are performed a posteriori, this is a great way to understand spillover between fluorescence channels. Consequent to this sophisticated acquisition, the speed or acquisition cannot be more than 5,000 cells per second, which is anyway much higher than what can be achieved optically. The instrument is equipped with highly sophisticated software which allows for a multitude of analyses of the parameters collected. The Image Data Exploration and Analysis Software (IDEAS®) is similar to any multiparameter FCM analysis software in that it allows to display fluorescence intensities in many combinations but, in addition, provides the ability to select cells based on morphological features [6]. Thus, for each cell, features such as cell shape, nuclear area, nuclearto-cytoplasmic ratio, subcellular characteristics such as texture as well as fluorescence intensity and localisation of specific nuclear and cytoplasmic markers can be determined, recorded and combined [4, 5]. Image analysis is based on the definition of two-dimensional 20:01:35

215

Chapter 14: Future Applications of Flow Cytometry and Related Techniques

areas within images called ‘masks’ [7]. The basic mask simply defines any area that differs from background, while additional morphology masks (e.g., nuclear size and shape) can be defined based on changes in signal intensity. Multiple features can be combined using Boolean logic to define various populations with different morphological characteristics. Of note, clicking on any spot or graph in the software allows to bring up the specific gallery of any individual cell. This technique has been used in research to study, for example, microparticles, apoptosis, autophagy and nuclear translocation of proteins [5, 8]. In haematopathology, the ImageStream® technology has to date mainly been used for the identification and characterisation of malignant cells [9], while reports describing its application in the evaluation of haematopoietic cells have been based mainly on murine bone marrow and concerned erythropoietic and megakaryocytic lineage [10]. An example of ImageStream® application in ­studies of megakaryocytes is shown in Figure 14.1. The technique can also be applied for functional studies. For example, because images are there, it is possible to appreciate the translocation of a cytoplasmic transcription factor such as NFk-B to the nucleus [8]. Then the proportion of cells that have indeed translocated NFk-B can be evaluated in a given subset, identified by multiparameter surface labelling [8]. Immunolocalisation of promyelocytic leukaemia (PML) protein in nuclear PML bodies or disrupted in the cytoplasm can also be visualised [9]. Another exciting application is the possibility of performing fluorescence in-situ hybridisation (FISH) analyses [12], knowing exactly which cells are involved, for instance demonstrating an anomaly both in the blasts and in immature progenitors likely thus to be leukaemic stem cells.

Spectral Flow Cytometry

216

Spectral flow cytometry takes advantage of analysing emitted light over the whole spectrum [13], that is, from 420 to 800 nm. This possibility has been explored since the beginning of the technology, but the extremely short time during which light must be acquired made analysis complicated. Basically, spectroscopy has been commonly used by astronomers to analyse the light spectrum of stars. Application of this type of spectrum analysis eliminates the need for dichroic mirrors and band-pass filters. Rather, either prisms or grated plates are used to collect and refract the whole light emitted by each cell as it hits the laser beam (Figure 14.2).

15

The identified light is turned into electronic signals with specific detectors providing high accuracy and enhanced specificity. Compensation matrices also are no longer needed. Essentially, the instrument records the spectral fingerprints of each cell. Another advantage of spectral flow cytometry is its ability to detect the spectral characteristics of cells’ autofluorescence and remove it from the spectra of fluorochromes, that is, labels. An interesting application of these properties is accessible on Sony’s website where Wahlstrom et al. [14] examined leukaemic cell lines phosphorylation upon incubation with different drugs. In terms of acquisition rates, and although this depends on the experiment, usually at least 1,000 events/s can be achieved [13]. At the moment, only Sony proposes two commercially available spectral flow cytometers, the SP6800® and SP3800®.

Cytometry by Time of Flight, Isotopes Instead of Fluorochromes and Mass Spectrometry instead of PMT Current flow cytometers in clinical laboratories allow to use 8 to 10 fluorochromes, as mentioned in many chapters of this book. Instruments still mostly devoted to research applications go as far as 14 to 19, limited by the possibilities of discriminating as many fluorescence signals. In some instances, and examples are reported in this book, several antibodies identifying mutually exclusive antigens can be used with the same fluorochrome in a panel to extend analysing capacities (see examples in Chapter 7). Another approach has focused on developing completely different instruments, no longer relying on photomultipliers gathering emitted light, but on mass spectrometry, hence the name CyTOF for cytometry by time of flight (Figure 14.3) [15, 16]. Antigens, either on the surface or inside the cells are labelled classically with monoclonal antibodies, yet the latter are conjugated to isotopes of elements at the bottom of the classification table. These heavy metals, mostly lanthanides, are virtually absent from living tissues and therefore have no interference or background [17]. Up to 45 different antibodies can be used at the same time currently, but theoretically over 100 isotopes are available for mass cytometry [18]. A typical flow cytometry hydrofocalisation technique is used to align cells and nebulise them in individual droplets. Then, as it exits from the flow cell, each droplet is vaporised, 20:01:35

Chapter 14: Future Applications of Flow Cytometry and Related Techniques

(A)

(B)

Intensity DNA in Halo vs CD41Masks

Median Pixel CD41

100

CD41+

1e3 50

500 0

BF

CD41

4

0.5

5

DNA in Mk 1e4

1e5

DNA

1e5 1e6 1e4 Intensity DNA in CD41Mask

kit

CD41

1

6

DNA

7

2

Masks:

8

CD41 Halo

CD41

Masks:

(C)

Not Plt

1e3

9 x

0.8 0.6

x

0.4

10

Kit-

800 600 400

Prog

200

x

1

0

300

600

900

1.2e3

0

Area_CD41 Mask

30

60

90

600

300

Median Pixel Kit_M07

300

11

10

Oddly shaped

10

0.6

13

12

20

Aspect ratio_ CD41 Mask Oddly shaped

DNA

9

30

600

Circulatrity_M07 Erode7

CD41

Round

40

Area_DNA Object in CD41 Mask

Round + Reclaimed = Mk BF

50

0 0

120

Halo

60

Intact DNA

0

0

0.2

CD41

Intact DNA

KitMedian Pixel DNA_ DNA object in CD41 mask

1

Area_CD41 Mask

Aspect ratio_CD41 Mask

Not Plt

Circulatrity_CD41 Mask

DNA in Mk 1.2

DNA

1

0

0 –1e3 01e3 1e4 1e5 1e6 1e7 Intensity CD41

DNA

3

1.5

150

2e3 1.5e3

CD41

BF

2.5e3

14

20 Reclaimed

15 10 5 0

300 600 900 1.2e3 Area_ CD41 Mask

0

(E)

2

BF

600

CD41

DNA

2N 300

4N

120 60 90 Area_ CD41 Mask

30

BF

CD41

150

1e6 16N

1.5

32N

2N 1

4N

64N

8N

0.5

1e5 1e4 1e3 0

0

8N

ExMk

0

GP1bβ

Intensity GP1bβ

Mk

Normralized Frequency

Area_Cytoplasm Mask

(D)

1e4

1e5

0

Intensity DNA_CD41 Mask

200

400

600

800

1e3

Area_ CD41 Maks

60%

16N

DNA

40%

32N 32N

20% 0%

64N 64N

2N

4N

8N

16N

32N

64N

Figure 14.1  Example of ImageStream (Amnis) technology (with permission from McGrath et al. [11], see the original publication for details) illustrating the gating strategy for analysis of megakaryopoiesis of live bone marrow cells. 5 × 106 cells per sample were stained with CD117, CD41, GP1bβ and DRAQ5. Cells were analysed on an ISX imaging flow cytometer (Amnis, ­Merck) and data were processed with IDEAS 6.1 software. (A) Initial gating of CD41+ cells. (B) Megakaryocytes (Mks) are gated based on ­having 2N or greater DNA intensity within a CD41 mask. (C) Continued gating of Mks by size and aspect ratio of the CD41 mask, distinguishing small kit+ ­progenitors from Mk, removing fragments of Mk with dim large DNA, selection of round cells and reclaimed large asymmetric Mk based on circularity of an eroded default mask of CD41. (D) Gating of ExMk from the Mk population based on low cytoplasm relative to cell size. ­Examples of ExMk are shown with ploidy indicated. (E) Analysis of identified Mk. Images of Mk at specific ploidy are on the left including signal for the maturational marker GP1bβ. Gating of Mk by ploidy is shown on the histogram and the average ± SEM of three independent ­experiments is presented in the graph below. A graph of other Mk maturation characteristics (GP1bβ intensity and cell size) is shown in the upper right. Mks of different ploidy (not including ExMk) are plotted using the same colours for populations as in the graph below.

15

20:01:35

217

Chapter 14: Future Applications of Flow Cytometry and Related Techniques

106

CPSA

CPSA

CPSA All events

106

No FP

106

CPSA

iRFP 670

iRFP 682

106

105 105

105

105

104

LD638_H

E Cok 87.69 %

103

LD638_H

LD638_H

SSC_A

104 104

104

102 103

103

103

101 100 –101

102

103

104

105

106

102 500

600

FSC_A

700

CPSA

LD638_H

LD638_H

20:01:35

104

3

800

6

600 700 Wavelength

800

6

105

104

104

103

102 500

600 700 Wavelength

800

500

600 700 Wavelength

Figure 14.2  Spectral flow. (From David Buschke, Sony Biotechnology’s SP6800 spectral analyser, freely available from https://flowcytometry.gwu.edu/spectral-flow-cytometry-fluorescent-proteins.) Shown here are scatter and spectral plots of each near-infrared fluorescent protein (iRFP) as well as the spectrum plot of a mixture of five iRFPs (circled in red).

218

800

Mix of Single Color Controls

10

102 500

700

CPSA

iRFP 720

103

102

102

600 Wavelength

105

LD638_H

15

10

10

103

102 500

800

CPSA

105

600 700 Wavelength

700

CPSA

105

500

600 Wavelength

iRFP 713

106

104

102 500

Wavelength

iRFP 702

106

800

LD638_H

–102

800

Chapter 14: Future Applications of Flow Cytometry and Related Techniques

Nebulize Single-Cell Droplets Antibodies Labelled with Elemental Isotopes

Ba La

Nd

Sm

Pr

Eu

Gd

Tb

Dy

Ho

Er Tm

Yb Lu

Intensity

100 90 80 70 60 50 40 30 20 10 0

ICP-MS Elemental Analysis

138

143

148

153 158 163 Stable Isotope Mass

168

173

178

Figure 14.3  Cytometry by time of flight. (From Evan Newell, freely available at www.a-star.edu.sg/sign/Resources/CyTOF.aspx.) The mass-cytometry concept. Very much like traditional fluorescence flow cytometry, mass cytometry works by using molecularly tagged antibodies to detect and quantify specific cellular antigens. In this case, the antibodies are tagged with heavy metal isotopes instead of fluorophores to stain the cells. Then, the cells are sprayed one by one (up to 500 cells/s) into a plasma torch that breaks all chemical bonds and ionises the contents of the cells. The elemental content of all atoms with a molecular weight between 100 and 200 atomic mass units is quantified for each cell. The benefit of this approach is that many more parameters can be assessed per cell (now >>40 parameters are being acquired per cell) and the crosstalk between the channels is minimal. That is because the time-of-flight mass spectrometer can easily discriminate single atomic mass unit differences in atomic mass (lower panel).

atomised and ionised in the plasma flame at 16,000° and the ions of whichever conjugate had attached to the cell are detected by the mass spectrometer. One of the drawbacks of this method is the rather low acquisition speed, at best of 1,000 cells/s but usually more around 300 to 500/s. This technique provides a vast amount of information for each cell, and specific analysis tools have been developed to sort it (see later). The most commonly used is spanning-tree progression analysis of density normalised events (SPADE), which allows to hierarchise and characterise cell subsets [17, 18]. Among the many current and growing applications of CyTOF, this method has been used, for instance, to reconstruct the steps of haematopoiesis [19–21], study the effects of stimulation with cytokines, mitogens or engagement of the antigen receptors BCR or TCR [22], explore T-cells specificity [18] and analyse tumour cells and their environment [22–24]. Application to the dissection of acute leukaemia in the presence of various potential drugs [25] has allowed to identify chemosensitive and chemoresistant subclones already present

15

at diagnosis [20]. For this type of study, the single cell network profiling strategy (SCNP), already developed for less-sophisticated FCM studies [26] has been fully implemented.

Compact Dedicated Instruments Moving closer to automation is one of the challenges of flow cytometry. Several solutions are available for partial or complete automation reducing the risk of errors (i.e., pipetting) and alleviating time-consuming processes. Several instruments devoted to rapid field use were developed [27] to face the monitoring of HIV-infected patients in emerging countries, such as BD Biosciences FACSCount, Partec CyFlow®, Apogeee Auto40® and Guava EasyCD4®. The non-flow-based technology of Alere Pima® was also introduced at the end of the twentieth century with the advantage of directly collecting capillary blood on the dedicated cartridge. These technologies are devoted to a limited number of assays but provide the expected output allowing to monitor the efficiency of tritherapy in the patients. 20:01:35

219

Chapter 14: Future Applications of Flow Cytometry and Related Techniques

Near-automation has been proposed more recently by BD Biosciences with the FACSPresto® working with ready-to-use cartridges just requiring to add the blood sample [28]. FACPresto provides CD4 percentage and absolute count and also assesses haemoglobin levels. It was also designed as a compact field-use instrument for emerging countries.

Closed Hands-Off Instruments Full automation is already in use for T-cell subsets and HIV patient monitoring with the Beckman Coulter Aquios® instrument, which just requires the operator to place the capped blood sample tube in the adequate slot [29]. The Aquios®, which is not designed for field work, provides a wider range of parameters with currently two four-colour panels allowing respectively for the assessment of CD3, CD4 and CD8 together with CD45, or for the evaluation of B-, T- and natural killer (NK)-cells (CD19, CD3, CD16+56, CD45). All parameters are provided as percentages and absolute counts. For the latter, fluidics are used rather than calibrated beads. Such instruments are not yet specified to be used for haematological malignancies although the B/T/NK combination of the Aquios® could provide orientation towards lymphoproliferative disorders.

The Revolution of Microfluidics

220

For almost 15 years now, innovation in microfluidics has dwelled on the technologies of electronic ­microchips to develop new tools. The mere concept of ‘lab-on-chip’ has really emerged in the twenty-first century [30]. Several concepts had to be revised in transposing classical cell counting and flow cytometry standards down to the field of micro/nano technology. Fluidics at that level obeys different rules that need to be circumvented or, contrarily, taken advantage of [31]. Computer-assisted design greatly helped/helps in the conception of complex mazes directing cell subsets towards differentiating channels [32]. The material of choice appears to be polydimethylsiloxane, easily worked to prepare grooves, inlets and outlets [33, 34]. Its transparency is valuable for combining fluorescent cell analysis and cell sorting. Applications of m ­ icrofluidic devices seem to rapidly grow in the pre-industrialised field with devices for complete blood cell counts, identification and sorting of rare leucocyte or malignant cells subsets [35]. Sorting techniques rely in this miniaturized different conundrum not only on magnetic fields or fluorescent markers

15

expression but also on fluidic behaviour or response to acoustic waves [36]. Much is to be expected if the necessary miniaturization of analysis tools follows that of yet almost ready single-use microfluidics chips [37].

Innovative Cell Functions Measurements Cytokines and Other Proteins Analysing the peculiar behaviour of malignant cells or of cells of their microenvironment remains an exciting challenge. Among such applications is the identification of cytokines production which can be successfully done in multiparameter FCM, even appreciating the polyfunctionality of some cell subsets. Basically, cells need to be stimulated in vitro in conditions leading to the production or not of cytokines [38]. After blocking secretion with such compounds as brefeldin A or monensin, the cells are permeabilised to allow for the intracytoplasmic targeting of accumulated cytokines. This has found applications in studying anti-tumour immune responses [39, 40] and has begun to be applied to the selection of the most efficient genetically engineered chimaeric antigen receptor (CAR) T-cells [41]. The detection of fusion transcript proteins inside cancer cells is another application of this strategy. It can be achieved as described earlier, directly within the cell [9]. Such strategies have the advantage of allowing, by prior surface labelling, to precisely identify the cell subsets involved in protein synthesis. The detection of fusion proteins can also be achieved after protein extraction using specific capture beads and fluorescent conjugates [42].

Phospho-Flow This application of classical flow cytometry has recently attracted more interest. While most FCM techniques target stable epitopes of surface differentiation antigens, phospho-flow will investigate specific activation states [43]. The antibodies used in phospho-flow are directed to the transient state of activation of a given cytoplasmic or nuclear protein characterised by its phosphorylation. This can be used to address spontaneous phosphorylation which can be displayed by some leukaemic cells [44] or to study activation pathways in response to specific signals (as illustrated in Figure 14.4). It even led to the detection of an unreported mutation of Flt3 in a case of acute myeloblastic leukaemia (AML) [45]. 20:01:35

Chapter 14: Future Applications of Flow Cytometry and Related Techniques

ETP-ALL

(A)

T-ALL

ALL18

ETP1

ALL81 ALL98

ETP5

ALL1

ETP8

ALL13

ETP12

ALL23

ETP13

ALL16

ETP14

ALL5 ALL8

ETP37

ALL33

ETP39 0

102

103

104

ALL77 0

105

pSTAT5

103

pSTAT5

105

Delta pSTAT5

(C) 6000

4000

4000

4000

2000

l 7 sa IL Ba ETP-ALL

pSTAT5MFI

6000

pSTAT5MFI

6000

0

104

pSTAT5

(B)

pSTAT5MFI

102

2000

0

l

B

a as

P = 0.012

2000

0

7

L

IL

T-ALL

E

L -A TP

L

AL T-

Figure 14.4  Phospho-flow (with permission from Maude et al. [46]). ETP-ALL hypersensitivity to IL-7 stimulation. (A) Phospho-flow cytometry analysis of levels of pSTAT5 in xenografted non-early T-precursor (ETP) T-ALL and ETP-ALL leukaemic blasts in the basal state (shaded) and in response to stimulation with 100 ng/mL IL-7 for 15 minutes (unshaded). (B) Quantitation of mean fluorescence intensity (MFI) in the basal state and in response to IL-7 exposure. (C) Change in pSTAT5 MFI with IL-7 exposure compared with vehicle treatment of each xenograft. The non-parametric Mann-Whitney U test was used to evaluate statistical significance.

The signals collected in phospho-flow are delicate to obtain when the target proteins are close to the inner side of the plasma membrane and also because they are likely to be engaged in interactions with other proteins involved in the pathway investigated.

15

Moreover, signals are prone to vary from one experiment to the other for several reasons and comparisons of different cell types or activation conditions are better run on the same day. The cells must be fixed to stop phosphorylation/dephosphorylation reactions 20:01:35

221

Chapter 14: Future Applications of Flow Cytometry and Related Techniques

and permeabilised so that specific antibodies can gain access to the phosphoproteins. These preparations can be frozen before staining, and thawed on the day of the labelling experiment. Of course, concomitant surface labelling will allow to better target the cell subsets involved in such activation while running the FCM analysis. Inhibitory phosphopeptides, such as those used for the generation of the specific antibodies will impair cell staining and can be used as controls. It is also possible to treat the cells with phosphatases before staining. Gary Nolan’s group [43, 47] has been the first to extensively study the best conditions for performing phospho-flow experiments. They initially used cell lines and compared their results with those of western blot assays. They determined that fixation in paraformaldehyde and permeabilisation with methanol provided the best results for nuclear translocated phosphoproteins, while a milder permeabilisation with saponin was optimal for cytoplasmic staining. It was also observed that attachment of the antibodies stabilised the phosphoproteins. The use of small fluorochromes such as FITC, Alexa Fluor 488® or Alexa Fluor 647® was shown to facilitate cytoplasmic access of the labelled antibodies. Moreover, these fluorophores are more likely to resist to the conditions of permeabilisation than others which may lose their fluorescence in methanol. An important point, common to all permeabilisation techniques, is that the side scatter (SSC) properties of the cells are likely to be changed and the use of surface staining to properly identify subsets is crucial in such settings. Although not all improvements have been published, a rather large array of conjugates, fixation and permeabilisation reagents are now commercially available with robust protocols and recommendations about the most suitable surface and phosphoproteinspecific clones. More progress is also coming from the use of new techniques such as mass cytometry [48, 49].

Flow Epigenetics

222

Several methods, usually relying on beads as particles to evaluate in FCM, have been developed to perform high-throughput analyses of histone modifications such as acetylation or methylation, DNA methylation and transcription factors binding. The MagPIE (Magnetic Protein Immobilization in Enhancer DNA) proposal [50] amplifies genomic regions likely to bind transcription factors, using a pair

15

of primers, respectively biotinylated and fluorescent (Figure 14.5). The biotinylated amplified DNA is captured on magnetic beads coated with streptavidin. The beads are incubated with a crude nuclear protein lysate of the cells to analyse. Finally, fluorescent antibodies directed to transcription factors, cofactors or histones are used to identify which nuclear proteins have bound to the amplified DNA. This rapid and sensitive method allows for the examination of multiple events occurring at the DNA level, including histone-binding and both DNA and histone methylation during the test, if this had been activated in the cell lysate tested. Chromatin immunoprecipitation (ChIP)-on-beads aims at evaluating the products of ChIP in a more rapid and more sensitive way than quantitative polymerase chain reaction (QPCR) [51]. ChIP is performed classically, by using antibodies to the proteins targeted, such as acetylated or methylated histones. After precipitation, the proteins are digested and the DNA part of the immunoprecipitate is amplified by PCR. For ChIP-onbeads detection, the primers are respectively tagged, one by biotin and the other by a fluorochrome. After the PCR, the amplicons are collected on streptavidincoated beads, and the fluorescence of bead suspensions is measured. ChIP-on-beads provide more rapid and more sensitive data, with the potential of performing a large number of different assays (changing stimulation conditions, diseases, proteins precipitated, etc.) more quickly than in QPCR.

Innovative Software Automated Analysis A number of proposals have been published in the literature for a comprehensive analysis of leucocytes subsets with a single combination of antibodies [52–54]. Most of them indeed provide a good delineation of up to 17 cell subpopulations in peripheral blood. For one of these combinations, software has been developed for an automated analysis through predefined gates. The Hematoflow® solution uses a six antibodies/ five-colour panel to identify 13 cell populations [55] after acquisition of 20,000 cells in a short time. Results are automatically provided after running a sample automatically prepared in a lysis no-wash method. The operator only has to place the sample in the preparatory and then run the stained/lysed sample. Over a few minutes all subsets are quantified in percentages (Figure 14.6). More recently [56], a series of flags have 20:01:35

Chapter 14: Future Applications of Flow Cytometry and Related Techniques

(B)

Lysate Lysate Count

+RG108

Anti-Methyl-CpG

(E)

H3K4Me1+ region

(C) 2 1.8 1.6 1.4 1.2 1 0.8 0.6 0.4 0.2 0

mES H3K4Me1+ mES H3K4Me1(D) 200 bp (SoxOct) 226 bp

p