Molecular imaging in multiple myeloma 9783030190194, 3030190196

Citation preview

Molecular Imaging in Multiple Myeloma Cristina Nanni Stefano Fanti Lucia Zanoni Editors

123

Molecular Imaging in Multiple Myeloma

Cristina Nanni • Stefano Fanti • Lucia Zanoni Editors

Molecular Imaging in Multiple Myeloma

Editors Cristina Nanni Department of Metropolitan Nuclear Medicine Policlinico S.Orsola-Malpighi Bologna Italy

Stefano Fanti Department of Metropolitan Nuclear Medicine Policlinico S.Orsola-Malpighi Bologna Italy

Lucia Zanoni Department of Metropolitan Nuclear Medicine Policlinico S.Orsola-Malpighi Bologna Italy

ISBN 978-3-030-19018-7    ISBN 978-3-030-19019-4 (eBook) https://doi.org/10.1007/978-3-030-19019-4 © Springer Nature Switzerland AG 2019 This work is subject to copyright. All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed. The use of general descriptive names, registered names, trademarks, service marks, etc. in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use. The publisher, the authors, and the editors are safe to assume that the advice and information in this book are believed to be true and accurate at the date of publication. Neither the publisher nor the authors or the editors give a warranty, expressed or implied, with respect to the material contained herein or for any errors or omissions that may have been made. The publisher remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. This Springer imprint is published by the registered company Springer Nature Switzerland AG. The registered company address is: Gewerbestrasse 11, 6330 Cham, Switzerland

Contents

Multiple Myeloma: Clinical Aspects��������������������������������������������������������������    1 Paola Tacchetti and Michele Cavo What Does a Clinician Need from New Imaging Procedures?��������������������   15 Elena Zamagni FDG PET in Multiple Myeloma ��������������������������������������������������������������������   27 Bastien Jamet, Clément Bailly, Thomas Carlier, Anne-Victoire Michaud, Cyrille Touzeau, Philippe Moreau, Caroline Bodet-Milin, and Françoise Kraeber-Bodéré Role of Standard Magnetic Resonance Imaging ������������������������������������������   39 Eugenio Salizzoni, Alberto Conficoni, and Manuela Coe Whole Body Diffusion-Weighted Magnetic Resonance Imaging: A New Era for Whole Body Imaging in Myeloma?��������������������������������������   73 Christina Messiou and Dow-Mu Koh CXCR4 Imaging in Multiple Myeloma����������������������������������������������������������   87 M. I. Morales, C. Lapa, and A. K. Buck PET/CT with Standard Non-FDG Tracers in Multiple Myeloma��������������   93 Cristina Nanni The Issue of Interpretation������������������������������������������������������������������������������   99 Cristina Nanni Clinical Teaching Cases: FDG PET/CT��������������������������������������������������������  105 Cristina Nanni, Lucia Zanoni, and Stefano Fanti

v

Multiple Myeloma: Clinical Aspects Paola Tacchetti and Michele Cavo

Definition Multiple Myeloma (MM) is a plasma cell (PC) disorder characterized by clonal proliferation of malignant PCs (Fig. 1) in the bone marrow (BM) or, more rarely, in extramedullary tissues. Neoplastic PCs typically synthesize monoclonal proteins (M-protein), which can be either intact immunoglobulins (Ig) or free light chains (FLC).

Fig. 1  Multiple myeloma. Cytology from aspirate. It showed an accumulation of plasma cells, characterized by the presence of an abundant basophilic cytoplasm and a small eccentric circular core, with chromatin plates and arranged radially, surrounded on the cytoplasmic side by a less basophilic area (plasma arc)

P. Tacchetti · M. Cavo (*) Seràgnoli Institute of Hematology, S.Orsola-Malpighi University Hospital of Bologna, University of Bologna, Bologna, Italy e-mail: [email protected]; [email protected] © Springer Nature Switzerland AG 2019 C. Nanni et al. (eds.), Molecular Imaging in Multiple Myeloma, https://doi.org/10.1007/978-3-030-19019-4_1

1

2

P. Tacchetti and M. Cavo

Pathophysiology and Pathogenesis The original cell of MM has not been yet recognised. The target of the neoplastic transformation is likely a B-lineage cell and the presence in peripheral blood of B lymphocytes clonally related to the transformed PCs strongly suggests that the two types of cells might have a common origin [1]. Both are derived during antigen-­ dependent maturation in the follicular germinative centre, which takes place in secondary lymphoid organs (lymph nodes). Indeed, a common Ig gene somatic hypermutation pattern can be observed in the two kinds of cell types. Neoplastic B-lymphocytes subsequently migrate from lymph nodes to BM, where they directly interact with both stromal cells and extracellular matrix. The presence of adhesion molecules on B-lymphocytes surface enhances their interaction with receptors on stromal cells, thus exposing tumor cells to cytokines released from stromal cells and extracellular-matrix. The most crucial cytokine involved in MM growth – both in vivo and in vitro – is IL-6. It exerts both a proliferative and an anti-apoptotic activity, and activates osteocla stogenesis. IL-6 production by BM stromal cells is increased as a consequence of the direct contact with PCs, which, in turn, are stimulated by stromal cells to produce other cytokines such as IL-1β, TNF-α and β and M-CSF.  These cytokine signals activate stromal cells, other accessory cells and osteoclasts. Neoplastic PCs stimulate BM angiogenesis as well, throughout the production of VEGF (Vascular Endothelial Growth Factor) and FGF (Fibroblast Growth Factor). Karyotyping reveals cytogenetic abnormalities in 20–30% of patients, those being mainly numerical abnormalities. The introduction of more sensitive techniques as FISH (Fluorescent in Situ Hybridization), CNAs (Copy Number Alterations) analysis by SNPs (Single Nucleotide Polymorphisms) array and NGS (Next Generation Sequencing) have improved the resolution of genomic analysis allowing for the detection of genomic aberrations in virtually all newly diagnosed patients. As a result, MM is now recognized to be a much more heterogeneous and complex disease then previously thought, placing the malignancy at the boundary between solid and haematological tumours. Primary events are chromosome translocations involving the Ig heavy chain (IgH) locus, and hyperdiploidy with multiple copies of odd-numbered chromosomes. IgH translocations are observed in 40% of cases. Frequently involved partner chromosomes/loci are 4p16 (FGFR3/MMSET) (12–15%), 11q13 (CCND1) (15–20%), 16q23 (MAF) (3%), 6p21 (CCND3) (5%), and 20q11 (MAFB) (1%). Hyperdiploidy occurs in 50% of newly diagnosed MM. Trisomies and IgH translocations are considered primary cytogenetic abnormalities. Secondary cytogenetic abnormalities arise along the course of MM, and include gain(1q) (CKS1B, ANP32E) (40%), del(1p) (CDKN2C, FAF1, FAM46C) (30%), del(17p) (TP53) (7%), del (13) (RB1, DIS3) (44%), RAS mutations, and secondary translocations involving MYC. Both primary and secondary cytogenetic abnormalities can influence disease course, response to therapy, and prognosis. The International Myeloma Working Group (IMWG) [2] recommends the use of FISH to define the cytogenetic

Multiple Myeloma: Clinical Aspects

3

risk. According the IMWG consensus statement t(4;14), t(14;16), t(14;20), and del(17/17p) and any nonhyperdiploid karyotype are high risk cytogenetic abnormalities; moreover, gain(1q) is associated with del(1p) carrying poor risk. Whereas FISH analysis continues to have an important diagnostic role in MM due to its widespread availability, newer technologies are primarily employed in clinical research. High throughput sequencing can detect a wider spectrum of genomic aberrations, including point mutations, gene expression deregulation and CNAs. Nonetheless, the prognostic significance of these aberrations is still not well understood, even if it is likely they will eventually be included in genomic markers of higher-risk disease. One of the more striking observations highlighted as a consequence of the use of high throughput sequencing is the presence of genomically variable subclones within the same patient within any given disease phase. This intra-clonal heterogeneity tends to change during the disease course, leading to the emergence of new clones, which might be different in different phases [3]. These modifications follow an evolutionary logic, with therapy acting as selective pressure, leading to the selection of the fitter clone, as compared to the less adapted, which is eliminated. This clonal dynamic has been described also in several solid tumours, as well as in other haematological malignancies and may be an important factor underlying therapy resistance and mechanisms for disease progression. Therefore, the analysis of samples collected in different disease phases is mandatory and provides a more complete picture of the patient’s genomic landscape during disease progression.

Epidemiology and Etiology MM accounts for approximately 1% of neoplastic diseases and 13% of hematologic cancers, accounting for 0.9% of all cancer deaths [4, 5]. In Western countries, the annual age-adjusted incidence is 5 cases per 100,000 persons. Lifetime risk of being diagnosed with MM is about 0.7%. The frequency of MM increases with age and reaches a peak in the 6th–seventh life decades. The median age of the population affected is about 70 years old, and less than 10% of all patients come to the diagnosis between the second and fourth decades of life. The prevalence is higher in males as compared to females (1.3:1.0) and in black race as compared to the white (ratio 2.0: 1.0). The incidence in the latter population is higher than that observed among Asians who live in the same geographic areas. Both genetic and environmental factors are hypothesized to explain racial differences in the incidence of the disease. The main known risk factors are occupational exposure to pesticides, petroleum and ionizing radiation. MM may develop de novo or, most commonly, represents the progression of a preceding monoclonal gammopathy of undetermined significance (MGUS). This last evolutionary model is virtually the basis of almost all MM cases, as evidenced by studies conducted on large series of healthy individuals and with long followup [6].

4

P. Tacchetti and M. Cavo

Clinical Features MM onset is clinically asymptomatic in 10–20% of cases and is detected by chance during routine laboratory examinations. In patients with a prior history of MGUS, the progression is characterized by the increse in serum and/or urinary M-protein, and medullary plasmacytosis, with consequent transformation in MM.  In the remaining 80–90%, the most common symptoms, for frequency and severity, are skeletal involvement, renal failure, infectious morbidity, myeloid failure, hypercalcemia, neurological complications, hyperviscosity syndrome and amyloidosis. Regardless of the presence of organ damage, MM is defined as active or symptomatic when at least one of the following conditions (defined as CRAB criteria) (Table  1) are present: hyperCalcemia, Renal failure, Anemia, or Bone lesions. Recent guidelines identify as myeloma defining events, also biomarkers of malignancy (as markers of early evolution to organ damage, i.e. more than 80% of risk within 2 years) defined as: high medullary plasmacytosis (≥60%), and/or high serum FLC (sFLC) ratio (involved/uninvolved ≥100), and/or presence of more than one focal lesion identifiable with nuclear magnetic resonance (MRI) [7]. MM cahracterized by the presence of at least one of CRAB criteria or biomarkers of malignancy, requires the start of treatment. Conversaly, if the diagnostic criteria of MM is present without CRAB or biomarkers of malignancy, the disease is called smoldering MM. Although smoldering MM represents a disease in neoplastic phase, it does not require immediate therapy.

Skeletal Involvement Up to 80% of MM patients present with osteolytic bone lesions at diagnosis and have an increased risk of skeletal-related events associated with increased morbidity and mortality. Approximately 60% of myeloma patients will develop a fracture during the disease course. The most commonly affected sites are those rich in BM, including the spine, ribs, pelvis, skull, and long bones. Radiologically, the classical aspect of osteolysis is that of a round lesion with sharp margins, well-defined, in the absence of surrounding signs of new bone formation and/or periosteal reaction. In Table 1  CRAB criteria C Hypercalcemia (serum calcium >0.25 mmol/L (>1 mg/dL) higher than the upper limit of normal or >2.75 mmol/L (>11 mg/dL) R Renal impairment (creatinine clearance 177 μmol/L (>2 mg/dL))a B Osteolytic lesions, demonstrated with one of the imaging methods available (whole body X ray, PET/CT, MRI, WBLDCT) Necessary at least one criterion for the definition of active MM, deserving treatment aExcluding other causes of kidney failure not related with gammopathy, such as diabetic or cardiovascular nephropathy, or others. If a diagnostic dilemma sussists, a kidney biopsy is recommended

Multiple Myeloma: Clinical Aspects

5

the vertebrae one can often find crushing aspects and wedging of the body (vertebral fractures). Bone disease is the major cause of morbidity in patients. The basis of the pathogenesis of myeloma-related bone disease is the uncoupling of the bone-­ remodeling process. The interaction between myeloma cells and the bone microenvironment ultimately leads to the activation of osteoclasts and suppression of osteoblasts, resulting in bone loss [8]. Studies in transgenic mice have shown that the enhanced osteoclastic activity is caused by an alteration in the balance between the production of RANK-L and osteoprotegerin (OPG). RANK (Receptor Activator of Nuclear factor KB) is a transmembrane receptor, belonging to the superfamily of TNF receptors, predominantly expressed on osteoclast precursors and mature osteoclasts. Its ligand, RANK-L, is produced by several cells in the bone marrow microenvironment and lymphoid cells. The RANK-RANK-L bond promotes the differentiation maturation, proliferation, activation of osteoclasts and may inhibit apoptosis of mature osteoclasts. The activity of RANK is inhibited by OPG, which acts as a “bait” receptor for RANK, preventing its activation by RANK-L and consequently inhibiting the proliferation and maturation of osteoclasts, as well as their activity. Many cytokines, growth factors and hormones, including parathyroid hormone, influence the level of RANK-L and OPG, to regulate the activity and differentiation of osteoclasts. In MM, the balance between OPG and RANK-L is altered by several cytokines, including IL-1ß, IL-6, TNF-α, TGF-ß, which increase the production of RANK-L and reduce the production of OPG, inducing osteoclastogenesis. Notch signaling pathway is actively implicated in MM-induced osteoclastogenesis. The net effect of Notch activation is the production of the osteoclastogenic factor RANKL by MM cells. At the same time, the osteoblastic activity is suppressed by inhibition of the differentiation of precursors into functional mature cells, via inhibition of the transmission of Runx2/Cbfa1 signaling and down-­ regulation of genes that encode for the WNT signaling, which is essential for osteoblast maturation. Numerous cytokines (IL-3, IL-7, IL-6) and soluble factors (Dickkopf-1 and secreted frizzle related proteins), produced by the interaction of PCs with the BM microenvironment, are responsible for inhibition of differentiation and maturation of osteoblasts.

Kidney Involvement Renal impairment is the second most common and most severe complication of MM. It presents in approximately 20% of patients at the time of diagnosis, while another 20–30% of patients develop renal impairment during the course of their disease [9]. The pathogenesis of renal failure in MM is multifactorial. First, and most importantly, it is due to urinary excretion of monoclonal light chains (Bence Jones proteinuria), that causes damage to tubules and/or glomeruli by three different mechanisms a) intra-tubular precipitation; b) direct damage of the tubular epithelium by monoclonal light chains or lysosomal enzymes; c) deposition of monoclonal light chains along the basement membrane of the glomeruli and tubules.

6

P. Tacchetti and M. Cavo

Consequently, the most important morphological and functional signs are myeloma cast nephropaty and light chain deposition disease, as well as deposition of amyloid fibrils. In addition to urinary excretion of monoclonal light chains, other factors that can contribute to the worsening of renal injury especially at the onset of the disease include: hypercalcemia, dehydration, use of nephrotoxic drugs (first of all non-­ steroidal anti-inflammatory), infections and hyperuricemia. Despite modern treatment strategies have significantly improved the prognosis of patients with renal failure, the presence of this complication, especially if acute, increases the risk of early mortality within 60 days of MM diagnosis.

Other Symptoms Another important clinical feature of MM is infectious morbidity, which is also the leading cause of death. Inceased susceptibility to infection in patients with MM is multifactorial. The main contributing factor is the suppression of humoral immunity via altered monocyte-macrophage function and T lymphocytes signaling and increased catabolism of immunoglobulins and suppression of normal B lymphocytopoiesis. Several drugs used for the treatment of MM, such as steroids and chemotherapeutic agents, enhance underlying immunosuppression. Other possible manifestations of the disease are hypercalcemic syndrome (secondary to bone destruction), neurological manifestations due to extrinsic nerves or nerve roots compression or sensori-motor polyneuropathy and hyperviscosity syndrome (more rare, present especially when the M-protein is very high, especially if IgM type).

Laboratory With the use of standard serum and urine electrophoresis plus immunofixation, MM can be classified in Ig-secreting MM (77%), Bence Jones MM (20%), and non-­secretory MM (3%) [10]. Ig-secreting MM are characterized by the production of clonal complete Ig, identical for isotype and idiotype. The most frequent isotype is IgG, followed by IgA, and much more rarely IgM or IgD. The M-protein alters the electrophoretic graph, with the typical appearance of “cathedral spire” (monoclonal hyper-gammaglobulinemia). Bence Jones MM are characterized by the production of monoclonal light chains unbound to heavy chains, excreted in the urine and resulting in the monoclonal Bence Jones proteinuria. In this case the serum electrophoretic graph can demonstrate hypogammaglobulinemia. In addition to standard electrophoresis and immunofixation, the sFLC assay is an automated nephelometric assay able to identify and measure the concentrations of free kappa and lambda light chains unbound to heavy chains in the serum. An abnormal sFLC ratio is a marker of clonality. With the use of sFLC assay, an abnormal sFLC

Multiple Myeloma: Clinical Aspects

7

ratio has been reported in almost all patients with Bence Jones MM, approximately 90% of Ig-secreting MM, and 70% of patients otherwise defined non-secretory MM [11]. In all patients with suspected MM, a serum electrophoresis, serum protein dosage, serum immunofixation, sFLC assay, 24  h proteinuria analysis with urinary electrophoresis and urinary immunofixation tests should be performed. In addition, a BM aspirate/biopsy is warranted to document an excess of PCs which is usually greater than 10% of the total mononuclear cells and can reach a total replacement of normal myeloid parenchyma. In qualitative terms, the neoplastic population may be morphologically indistinguishable from normal PCs or, more often, can present numerous cytologic atypia, up to the acquisition of clear anaplastic characteristics. Cytogenetic-molecular analysis of malignant PCs is important for prognostic staging. Blood counts can reveal anemia and/or leukopenia and/or thrombocytopenia. Anemia, generally normocromic normocitic-type, is the most common laboratory abnormality. At the observation of the peripheral blood smear, red blood cells are usually stacked, with an increase of erythrocyte sedimentation, due to the hypergammaglobulinemia. Rarely, excess of circulant PCs is present and, if in number greater than 2000/mm3, identifies the leukemic variant of MM (plasma cell leukemia). Other important laboratory parameters to consider during the course of MM are renal function and serum calcium. Relevant for prognostic purposes is, finally, the determination of the concentration of serum β2-micro globulin and albumin.

Diagnosis According the revised IMWG diagnostic criteria [7], MM and smoldering MM are defined as follow (Table 2).

Table 2  Clinical and laboratory features of monoclonal gammopathy of undetermined significance (MGUS), smoldering MM (SMM) or active MM Bone marrow PCs (%) M component (g/L) Bence Jones (g/24 h) Myeloma defining events

MGUS SMM 0.25 mmol/L (>1 mg/dL) higher than the upper limit of normal or >2.75 mmol/L (>11 mg/dL) –– Renal insufficiency: creatinine clearance 177 μmol/L (>2 mg/dL) –– Anaemia: haemoglobin value of >20 g/L below the lower limit of normal, or a haemoglobin value 1 focal lesions on MRI studies (each focal lesion must be 5 mm or more in size).

Definition of Smoldering MM Both criteria must be met: • -Serum M-protein (IgG or IgA) ≥30 g/L or urinary M-protein ≥500 mg per 24 h and/or clonal BM PCs 10–60% • -Absence of myeloma defining events or amyloidosis

Multiple Myeloma: Clinical Aspects

9

Differential Diagnosis The differential diagnosis of MM includes the following diseases: a) lympho-­ immunoproliferative syndromes, with production of M-protein (MGUS, solitary plasmacytoma, Waldenstrom’s macroglobulinemia, heavy chains deposition disease, POEMS syndrome, AL amyloidosis, cryoglobulinemia); or b) diseases with osteolytic skeletal involvement (metastatic carcinomas; hyperparathyroidism), in which, however, PC infiltration of the bone is not present, as well as M-protein; alkaline phosphatase may also be substantially increased, differently than MM.

Prognosis The median survival of patients with MM ranges from few months to more than 10 years, and is influenced by several clinical and laboratory parameters, some of which correlate directly with the tumor burden, others are an expression of the inherent malignancy of the tumor clone and others eventually depend on the patient’s response to the therapy (Table 3). It is generally agreed that a combination of the International Staging System (ISS) and cytogenetic abnormalities allows risk stratification. The ISS [12] identified three subgroups of patients with different prognosis on the basis of two laboratory parameters, albumin and β2-microglobulin. Moreover, cyotgenetic-molecular abnormalities split the patients in a high-risk group - if are present t(4;14) and/or t(14;16) and/or t(14;20) and/or del(17/17p) and/or gain(1q) by FISH, and/or nonhyperdiploid karyotype, and/or del [13] by conventional karyotype, and/o high-risk signature by GEP - and standard risk (all others) [2]. More recently, the revised ISS (R-ISS) was defined, incorporating high risk FISH by t(4;14), t(14;16), and del(17p) with ISS and LDH (Table 4) [13]. Table 3  Prognostic factors of MM Related to the tumor mass β2-microglobulin Clinical stage Degree and type of BM infiltration Renal function Number of FLs at PET/CT or MRI

Related to the biology of Related to the tumor Related to the patient therapy Cytogenetic abnormalities Age Quality of response Albumin Performance status C reactive protein Presence of co-morbidities Circulating PCs Extramedullary disease LDH

10

P. Tacchetti and M. Cavo

Table 4  Revised ISS classification incorporating ISS, cytogenetic abnormalities (CA) and LDH Prognostic factor Criteria ISS stage I Serum β2-microglobulin 5.5 mg/L CA by High risk Presence of del(17p) and/or translocation t(4;14) and/or iFISH translocation t(14;16) Standard No high-risk CA risk LDH Normal Serum LDH  upper limit of normal New model for risk stratification for MM R-ISS stage I ISS stage I, standard-risk CA by iFISH and normal LDH II Not R-ISS stage I or III III ISS stage III and either high-risk CA by iFISH or high LDH

Therapy Because only one prospective clinical trial has demonstrated a survival advantage with early treatment of MM during the smoldering phase [14], current recommendations are to start therapy only in the presence of myeloma events, according to the CRAB criteria, or to the new markers of early evolution. It is likely that in the future treatment will be started in earlier phases with the goal of disease eradication as new drugs are introduced with improved specificity for the neoplastic clone and better characterization of tumor biomarkers. The treatment landscape of MM has dramatically evolved over the last decades, with an increasing availability of highly active new therapies which have considerably improved patient outcomes. The first major change was the introduction of autologous stem cell transplantation (ASCT). Studies performed in the mid-1990s demonstrated that, in patients under the age of 60  years, ASCT was superior, ­compared to conventional therapy, in terms of percentage and depth of responses, as well as duration of survival outcomes. The ASCT procedure, widely used today and extended to patients up to 65–70 years of age, is composed of an induction phase, performed with combinations of new drugs, peripheral blood stem cells collection (PBSC), often preceded by mobilization therapy with cyclophosphamide 2–4 g/m2, with the addition of granulocyte colony stem cell factor (G-CSF), and, finally, the reinfusion of PBSC (transplant), preceded by treatment with high-dose melphalan (200 mg/m2). The improved results obtained with ASCT along with limited toxicity, have led different institutions to experiment the use of two sequential procedures of ASCT (double ASCT) with further intensification of chemotherapy that resulted in prolonged disease control [15]. Actually, up to 60% of patients treated at diagnosis with single or double ASCT will survive longer than 10 years, 30% without disease relapse. Allogeneic stem cell transplantation, from HLA-identical family or registry donor, may have the benefits of an uncontaminated graft and potential graft versus myeloma effect, however is associated with more frequent and serious side effects

Multiple Myeloma: Clinical Aspects

11

without clear evidence of therapeutic benefit. Thus, its use is generally limited to young patients with high risk disease at first relaspe. Moreover, the understanding of the complex system of interactions between myeloma cells and the BM microenvironment has further expanded the treatment options, through the introduction of new drugs targeting the MM clone in the BM. The first of this new class of agents was the immunomodulatory (IMiD) drug thalidomide. From thalidomide, other second and third generation IMiDs, such as lenalidomide and pomalidomide, were developed and are now widely used, showing a more potent efficacy than thalidomide and a better toxicity profile. The mechanisms of action of IMiDs are multiple, including an action on the BM microenvironment, with inhibition of cytokines that support PCs (IL-6, TNF-α, IL-12), potent anti-angiogenic activity, and direct anti-proliferative effects on PCs mediated through inhibition of cyclin-dependent kinase and down-regulation of anti-apoptotic molecules, and an immunomodulatory action, with induction of Th1 response, NK cell activation and production of IL-2 and IFN-γ. A second important category of new drugs used in MM is represented by proteasome inhibitors (PIs), which target the proteasome, an intracellular multicatalitic structure responsible for the degradation of most of the intracellular protein, that is essential for the normal functioning of the PCs and hyper-expressed in MM. PIs, whose progenitor is bortezomib, followed by second generation inhibitors such as carfilzomib and ixazomib, bind to the enzymatic sites of the proteasome, blocking its function. The antineoplastic effect of this class of drugs is mediated through several different mechanisms that include inhibition of cell proliferation signals, the inhibition of the expression of adhesion molecules, and the induction of apoptosis. Actually, standard treatment [16] for transplant eligible patients includes a three-­ drugs bortezomib-based combinations induction therapy, ASCT, and a maintenance phase with lenalidomide. Double ASCT may be of use, particularly in high-risk patients. In the setting of non-transplant eligible patients, a fixed duration treatment with bortezomib-melphalan-prednisone (VMP) or the all alkylator-free, continuous treatment with lenalidomide-dexamethsone (Rd) are currently recommended. Moreover, recent data have shown the benefits of bortezomib-lenalidomide-­ dexamethasone (VRd) over Rd., probably becoming a new standard of care in the near future. Next generation IMiDs and PIs, and new novel agents with different mechanism of action as the monoclonal antibodies, anti-CD38 or anti SLAM F7, have beed approved in the relapsed setting and are being explored in newly diagnosed MM in the contest of clinical trials. New novel agents such as inhibitors of histone deacetylases, check-point inhibitors, bcl-2 inhibitors and nuclear export protein XPO1 inhibitors are under development.

Goals of Treatment Primary goals of treatment are to maximize the depth of response, to minimize the burden of residual tumor cells and to prevent or delay relapse as a way to prolong progression free survival (PFS) and untimely overall survival (OS). Next generation

12

P. Tacchetti and M. Cavo

flow (NGF) and sequencing (NGS) methods to quantify minimal residual disease (MRD) in the BM with sensitivities in the range of 10−5–10−6 cells are currently in use. These technologies may be combined with functional imaging techniques to detect MRD outside of BM.  The relevance of MRD in MM was officially sanctioned in 2016 by the IMWG, with the introduction of the new evaluation criteria for the response [17]. The category of NGF MRD negativity is defined as the absence of phenotypically aberrant PCs on BM evaluated by NGF with a method sensitivity ≥10−5. The category of NGS MRD negativity is defined as the absence of clonal PCs evaluated on BM by NGS with a sensitivity ≥10−5. These categories aimed at defining the medullary MRD were then added to the Imaging-MRD category, which requires, in addition to the normalization of MRD by NGF or NGS, the absence of areas of pathological uptake by PET/CT, in order to explore both the intra- and extra-medullary segments. A condition of MRD sustained negativity is finally defined as BM and imaging negativity, maintained for at least 1 year.

References 1. Hallek M, Bergsagel PL, Anderson KC, et al. Multiple myeloma: increasing evidence for a multistep transformation process. Blood. 1998;91(1):1–16. 2. Sonneveld P, Avet-Loiseau H, Lonial S, et  al. Treatment of multiple myeloma with high-­ risk cytogenetics: a consensus of the international myeloma working group. Blood. 2016;127(24):2955–62. 3. Keats JJ, Chesi M, Egan JB, et al. Clonal competition with alternating dominance in multiple myeloma. Blood. 2012;120(5):1067–76. 4. Altekruse SF, Kosary CL, Krapcho M, et  al. SEER cancer statistics review, 1975–2007. Bethesda, MD: National Cancer Institute. http://seer.cancer.gov/csr/1975_2007/index.html. 5. Becker N. Epidemiology of multiple myeloma. Recent Results Cancer Res. 2011;183:25–35. 6. Landgren O, Kyle RA, Pfeiffer RM, Katzmann JA, Caporaso NE, Hayes RB, Dispenzieri A, Kumar S, Clark RJ, Baris D, Hoover R, Rajkumar SV.  Monoclonal gammopathy of ­undetermined significance (MGUS) consistently precedes multiple myeloma: a prospective study. Blood. 2009;113(22):5412–7. 7. Rajkumar SV, Dimopolous MA, Palumbo A, et  al. International myeloma working group updated criteria for the diagnosis of multiple myeloma. Lancet Oncol. 2014;15:538–48. 8. Terpos E, Ntanasis-Stathopoulos I, Gavriatopoulou M, Dimopoulos MA. Pathogenesis of bone disease in multiple myeloma: from bench to bedside. Blood Cancer J. 2018;8(1):7. 9. Dimopoulos MA, Terpos E, Chanan-Khan A, et al. Renal impairment in patients with multiple myeloma: a consensus statement on behalf of the international myeloma working group. J Clin Oncol. 2010;28(33):4976–84. 10. Kyle RA, Gertz MA, Witzig TE, et al. Review of 1027 patients with newly diagnosed multiple myeloma. Mayo Clin Proc. 2003;78(1):21–33. 11. Mead GP, Carr-Smith HD, Drayson MT, Morgan GJ, Child JA, Bradwell AR. Serum free light chains for monitoring multiple myeloma. Br J Haematol. 2004;126(3):348–54. 12. Greipp PR, San Miguel J, Durie BG, et al. International staging system for multiple myeloma. J Clin Oncol. 2005;23(15):3412–20. 13. Palumbo A, Avet-Loiseau H, Oliva S, et  al. Revised international staging system for multiple myeloma: a report from international myeloma working group. J Clin Oncol. 2015;33(26):2863–9.

Multiple Myeloma: Clinical Aspects

13

14. Mateos MV, Hernández MT, Giraldo P, et al. Lenalidomide plus dexamethasone for high-risk smoldering multiple myeloma. N Engl J Med. 2013;369(5):438–47. 15. Cavo M, Rajkumar SV, Palumbo A, et al. International myeloma working group consensus approach to the treatment of multiple myeloma patients who are candidates for autologous stem cell transplantation. Blood. 2011;117(23):6063–73. 16. Moreau P, San Miguel J, Sonneveld P, et al. ESMO guidelines committee. Multiple myeloma: ESMO clinical practice guidelines for diagnosis, treatment and follow-up. Ann Oncol. 2017;28:iv52–61. 17. Kumar S, Paiva B, Anderson KC, et al. International myeloma working group consensus criteria for response and minimal residual disease assessment in multiple myeloma. Lancet Oncol. 2016;17(8):e328–46.

What Does a Clinician Need from New Imaging Procedures? Elena Zamagni

 he Role of Imaging in Multiple Myeloma and the  T “Death” of Conventional Radiography Bone disease is the most frequent feature of multiple myeloma (MM), occurring in approximately two thirds of patients at diagnosis and in nearly all patients during their disease [1]. Despite remarkable advances in MM therapy over the last decade, the consequences of skeletal involvement still remain clinically relevant. Bone disease impairs patients’ quality of life and represents a major cause of morbidity and mortality. For this reason, imaging plays a very important role in the management of MM [2] First of all, it is necessary for detection of lytic bone lesions, which represent a marker of disease-related end-organ damage, traditionally used to diagnose MM and to establish the need to immediate start of therapy [3]. Additionally, imaging could identify sites of extramedullary disease (EMD), that represent an unfavorable prognostic feature, and helps to accurately differentiate between solitary plasmacytoma (SP) and MM, as well as to predict the risk of early progression from smoldering MM (SMM) to active disease. During the course of MM, imaging is also essential to establish the diagnosis of relapse and eventually to detect sites of bone damage at potential risk of pathological fractures or neurological complications. Lastly, functional imaging techniques enable more careful assessment of the depth of response to treatment, in particular in patients with non-secretory MM and normal serum free light chain ratio [2], and more generally contribute to the definition of negative minimal residual disease (MRD). Although conventional radiography has historically been the standard imaging technique for many years, it has several limitations. For a lytic lesion to become apparent, it involves losing more than 30% of trabecular bone. Other limitations E. Zamagni (*) “Seràgnoli” Institute of Hematology, Bologna University School of Medicine, Bologna, Italy e-mail: [email protected] © Springer Nature Switzerland AG 2019 C. Nanni et al. (eds.), Molecular Imaging in Multiple Myeloma, https://doi.org/10.1007/978-3-030-19019-4_2

15

16

E. Zamagni

include the prolonged study time, difficulty in assessing certain areas, like the pelvis and spine, inability in distinguishing vertebral fractures secondary to benign osteoporosis from those related to the underlying MM clone and limitation in the assessment of response to treatment, due to the low sensitivity of the technique for the limited bone healing. Recently, two retrospective trials on a large number of patients with suspected SMM, who received skeletal survey and either positron emission tomography/computed tomography (PET/CT) [4] or whole-body low dose CT (WBLDCT) [5] as part of their diagnostic work-up, demonstrated that the use of conventional radiography would have under-estimated the presence of bone disease in approximately 25–40% of cases. Limitations of conventional radiography led to increasing use of more advanced imaging modalities. In 2014 the International Myeloma Working Group (IMWG) updated the diagnosis of MM and established that (1) one or more lytic lesion seen on CT or WBLDCT or PET/CT, regardless of the detection or not on skeletal radiography, and (2) more than 1 unequivocal (≥5  mm in size) focal bone marrow lesion (FL) on MRI, fulfill the criteria for MM-related bone disease [6]. This innovation was introduced after the clear demonstration that the novel imaging techniques have a higher detection rate than skeletal survey [7] and led to their use in routine clinical practice. The advantages, indications of use, and applications of the main novel imaging techniques is related to the different MM phases, as well as the clinical questions that need to be addressed.

 he Role of New Imaging Techniques for the Diagnostic T Work-Up of MM and SMM The diagnostic work-up is a very critical step for MM patients, both at diagnosis and in subsequent relapse phases, as it defines the presence of an active disease, needing the start of treatment, and forms the basis for the comparison with the result of therapy. Over the past years, the development of novel and more sensitive imaging methods for the identification of osteolyses has progressively led to the substitution of skeletal survey.

Whole Body Low Dose Computed Tomography WBLDCT was introduced to detect osteolytic lesions in the whole skeleton, with high accuracy, no need for contrast agents and two- to three-fold lower radiation dose exposure compared to standard CT [8, 9]. In several studies, WBLDCT was found to be superior to WBXR in detecting osteolytic lesions, affording higher sensitivity and a higher detection rate, in particular in the spine and pelvis, resulting in higher accuracy [4, 5, 8–11]. In addition, CT provides important information

What Does a Clinician Need from New Imaging Procedures?

17

regarding the potential instability and fracture risk of the vertebrae, as well as being a guide for radiotherapy and surgery. It is generally suggested one perform a whole-­ body low dose multi-detector CT with 3.2–4.8 milli Sievert (mSv) as radiation doses (skeletal survey referral: usually 1.2–4.8) [12]. While the main role of WBLDCT is identification of bone destruction, bone marrow plasma cell (PC) infiltration can be detected in adults in the long bones, which are generally substituted by fatty marrow. The detection of nodular or diffuse infiltration of long bones has proved of prognostic significance [13].

18F-FDG-Positron Emission Tomography/CT (PET/TC) PET/CT, usually with 18F-fluorodeoxyglucose (FDG) as the radiopharmaceutical, is a dual technique that blends the ability to identify bone destruction and lytic lesions with assessment of tumor burden and disease activity, in different areas of the BM and of the cancellous and cortical bone [2]. This is of particular interest as BMPC infiltration in MM is not homogeneous. PET/CT can be used for the diagnostic work-up of the disease, as several studies have reported a sensitivity and specificity for detection of bone lesions in the range between 80 and 100% [14–18]. The combination of functional imaging with PET plus morphological assessment with CT makes this technique the most effective in identifying potential sites of EMD [19]. In addition to the presence of EMD, also the number and metabolism of FLs prior to treatment have been identified as predictors for clinical outcomes in several prospective or retrospective studies. Finally, as discussed more in depth in the next paragraph, performing PET/CT at baseline allows to assess the metabolic response to therapy, by comparing pre- with post-treatment images. The CT part of a PET/CT may be considered broadly comparable to a WBLDCT, according to the minimum technical requirements established by the IMWG [20].

Magnetic Resonance Imaging (MRI) MRI has been established as a valuable technique for diagnosing bone involvement in MM [21]. MRI is based on examining the composition of the tissue regarding water and fat content and has the highest sensitivity when it comes to detecting BM infiltration by myeloma cells, without radiation exposure. Diffusion weighted imaging (DWI) is an additional MRI protocol, measuring the movement of water molecules in the tissue [22], underlining the presence of high cellularity (limited movements) as opposed to low cellularity and/or higher microcirculation (increased movements), without the need for a contrast agent.

18

E. Zamagni

Several studies have shown that MRI, both axial or whole body, is more sensitive than WBXR for the detection of bone involvement in MM, affording higher ­diagnostic precision [23, 24]. Studies that have compared MRI with PET/CT have shown that the two techniques are equally effective in detecting FLs [17]. Studies comparing MRI with WBLDCT have suggested excellent agreement in terms of lesion detection, pattern and BM involvement [25]. MRI-FLs correlate with standard known prognostic factors, in particular cytogenetics, and with clinical outcomes [23, 26]. On the contrary, the prognostic meaning of a diffuse pattern is less clear; addition of the DWI technique as an adjunct may clarify this issue [27].

 hat Does a Clinician Need from Newer Imaging Techniques W for the Diagnostic Work-Up? There is considerable heterogeneity in clinical practice regarding incorporation of the different imaging modalities in patient management. The clinical use of different techniques is often influenced by their availability, local expertise, affordability and national guidelines for reimbursement. Likewise, international societies, such as the IMWG, the European Myeloma Network (EMN) [28] and the European Society for Medical Oncology (ESMO) [29] have provided different recommendations. When making a choice, one should also remember that findings at baseline impact on interpretation of the same findings after therapy. Since the detection of lytic bone lesions, eventually at risk of developing possible complications, is the mainstay for starting anti-MM treatment and managing correctly MM patients, WBLDCT is considered in most guidelines, and routinely used in clinical practice, as the standard technique for assessment of myeloma bone disease. Skeletal survey should only be applied if nothing else is available. If no signs of bone destruction are detected on CT, MRI, preferably whole body or at least of the axial skeleton, should be performed to establish the presence of more than one FLs, which are now myeloma defining events, according to the new diagnostic criteria [6]. PET/CT is recommended if there is a strong suspicion of EMD, in cases of oligo-non-secretory MM with normal serum free light chain (sFLC) ratio or in clinical trials where systematic MRD assessment is being applied, to create a baseline for response assessment. In patients with suspected relapse, the diagnosis requires direct indicators of tumor growth (at least one of the clone-related biomarkers) and/or the presence of organ damage, including bone. For this purpose, WBLDCT should be the first choice, as it’s recommended at diagnosis. If a biochemical progression is present, or a disease with low tumor burden, MRI or PET may be preferable, for consistency with previous recommendations in SMM. Table 1 summarizes the advantages of the three main imaging techniques for diagnostic work-up.

What Does a Clinician Need from New Imaging Procedures?

19

Table 1  Comparison of novel imaging techniques as part of the diagnostic work-up regarding most relevant topics WBLDCT Ease of use •  Patient-friendly (fast scanning time, 1 FL on MRI (reported in 16 and 28% of cases) had a median

20

E. Zamagni

time to progression to symptomatic disease of 13–15 months, and a 2-year probability of progression of approximately 70–80% [30, 31]. Patients with a single FL or >1 small ( 4.2 and the presence of EMD were associated with a shorter OS. The prognostic value of FDG-PET was also studied in a smaller series of patients with MM (55) and 6 with SP [6] [25]. A correlation was found between the most intense EMD, medullary uptake (p = 0.027) and the International Staging System (ISS) score (p = 0.048). Medullary SUVmax was correlated with the ISS (p = 0.013). The 44 patients with a positive FDG-PET had an estimated 5-year OS (61%) less than the 11 all-alive negative FDG-PET patients (p = 0.01). By multivariate analysis, only the intense EMD had a prognostic value for OS (p = 0.03). The Imajem study [13] confirmed the poor prognostic value of the EMD at baseline diagnosis on PFS (p 210  cm3 at baseline significantly decreased PFS and OS after adjustment for known prognostic factors. Combined with the gene expression profiling prognostic score (GEP70), a TLGWB >205 g identified a high-risk subgroup and separated ISS II patients into two subgroups, with a similar outcome to ISS I and ISS III patients. Finally, as described by Carlier et al. for 62 patients in the French Imajem study, intra-tumoral textural features (e.g. reflecting of tumor heterogeneity), and especially entropy, also seem to be of prognostic value (independent prognostic value of entropy on PFS) [29]. More work is in progress on this subject.

Potential Use of FDG-PET in Therapeutic Evaluation Obtaining a complete metabolic response (CMR) by FDG-PET during intermediate evaluation, prior to or after ASCT, is associated with better survival. FDG-PET is considered as the reference imaging technique for therapeutic evaluation in MM, allowing evaluation of the response earlier than MRI, with a strong independent prognostic value [5]. FDG-PET, coupled with a biological technique for the detection of minimal residual disease (MRD), makes it possible to improve the definition of complete response [30]. Bartel et al. showed in 2009 that normalization of FDG uptake of FLs after chemotherapy induction cycles, and before the transplant procedure, was associated with better EFS and OS [23]. In the context of genetic profiles, pre-autologous CMR conferred better OS in low-risk patients and better EFS in high-risk patients. In 2013, the same team reported that for patients treated according to the same intensive protocol (Total Therapy 3) the prognostic value of the early FDG-PET performed at Day 7 of induction in a series of 302 patients, 277 of them were also the object of a gene expression profile study [31]. FDG-PET was compared with radiographs and MRI. Multivariate analysis revealed that three FLs on FDG-PET were associated with lower PFS and OS, even in the high-risk group, and in relation to genetic profiling, FDG-PET could be considered as a tool for early therapeutic adaptation.

34

B. Jamet et al.

The Bologna group showed that after therapeutic induction, a SUV >4.2 was associated with a reduced PFS [24]. Three months after ASCT, CMR was achieved in 65% of patients, with PFS and OS at 4 years higher than those in PET-positive patients. Interestingly, 23% of patients achieving CR in accordance with conventional criteria were considered PET-positive. Multivariate analysis showed that post ASCT PET status was an independent prognostic factor of PFS. In 2015, the same group confirmed these results in 282 patients with symptomatic MM undergoing front line treatment between 2002 and 2012 [32]. The median follow-up was 67 months. After treatment, a CMR was obtained in 70% of the patients, whereas the conventional biological methods concluded at 53% of CR. The FDG-PET negativity affected the PFS and the OS positively. In 12% of relapsed patients, bone progression was only detected by FDG-PET in the systematic follow-up and multivariate analysis showed that a SUVmax >4.2 after frontline treatment was an independent predictor of progression. The Imajem prospective study recently confirmed the major benefit of FDG-PET in therapeutic evaluation [13]. Although normalization of MRI after 3  cycles of Lenalidomide (R), Bortezomib (V) and Dexamethasone (D) before maintenance did not significantly affect either PFS or OS, FDG-PET normalization before maintenance was strongly associated with better PFS and OS. The PFS and OS of PET-­ negative patients were better than those of PET-positive patients (24-months PFS by 72% vs. 56.8%: p = 0.01; OS at 24 months of 94.2% vs. 72.9%: p = 0.03). In addition, multivariate analysis revealed that normalization of pre-maintenance FDG-­ PET was independently associated with longer PFS, such as the absence of EMD at diagnosis and at least a very good partial biological response after three cycles of RVD. Moreover, for patients with FDG-avid MM included in this Imajem cohort, the prognostic value of FDG-PET after three cycles of RVD was also reported [33]. Indeed, in the multivariate analysis, only ΔSUVmax (p −25 vs.  ≤−25%) which identified patients with improved median PFS. The benefit of post-ASCT FDG-PET was also reported in 2013 in a prospective series of 77 patients assessed by FDG-PET 3 months after ASCT, and then every 6–12 months during follow-up [34]. The duration of the response was longer when the PET scan was negative (27.6 months) than when it was positive (18 months, p = 0.05), whereas in patients with positive PET, SUVmax was inversely correlated with the duration of the response (P  50%

Focal lesion

PCI

T1-w FS-T2-w Fat cells

++

++++

+++

++

+

0

Hematopoietic marrow

++++

+++

+++

++

+

0

Vessels / Perfusion

+++

+

++

+++

++++

+++++

Interstitial space

++++

+

+

+++

++

+

Interstitial water

++++

+

+

++

+++

+++

BM = Bone Marrow, yrs = years, PCI = Plasma Cells Infiltration, T1-w = T1-weighted images, FS-T2-w = fat suppressed T2-weighted images

The different imaging characteristics of normal and infiltrated bone marrow are presented as signal intensities in grey-scale, while the changes in amount of fat cells, hematopoietic cells, interstitial space and vascularization are described in a semi-quantitative manner. Modified with permission from Dutoit et al. 2016, ref. [46], Creative Commons Attribution 4.0 International License (http:// creativecommons.org/licenses/by/4.0/) BM bone marrow, yrs years, PCI plasma cells infiltration, T1-w T1-weighted images, FS-T2-w fat suppressed T2-weighted images

Role of Standard Magnetic Resonance Imaging Fig. 5  Normal bone marrow MRI pattern in a newly-diagnosed 77-year-old myeloma patient. T1-w (a) and post-gadolinium fat-­ suppressed T1-w (b) sagittal images of the entire spine are reported

51

a

b

ally associated with a low tumor burden, with only a slight plasma cells infiltration (50% plasma cells on bone marrow biopsy), because the signal intensity is nearly equal to or lower than the signal of the intervertebral disc on T1-w images [58, 81] (Fig. 7). The signal reduction of intermediate-grade

52 Fig. 6  Multiple areas of bone marrow infiltration in a newly-diagnosed 75-year-old myeloma patient with a focal infiltration MRI pattern. T1-w (a) and post-­ gadolinium fat-suppressed T1-w (b) sagittal images of the spine are reported

Fig. 7  Diffuse infiltration MRI pattern in a newly-­ diagnosed 66-year-old myeloma patient. Dixon “in-phase” T1-w (a) and post-contrast Dixon “water-only” T1-w (b) sagittal images of the lumbar spine are reported

E. Salizzoni et al.

a

b

a

b

involvement (20–50% plasma cells on bone marrow biopsy) may be subtler and often hard to distinguish even from the normal bone marrow appearance [84, 86]: in such cases the qualitative evaluation of MRI images is often subjective and difficult to reproduce, making the diagnosis a very challenging process [87]. A combined diffuse and focal pattern is found in 10% of patients and it is characterized by a diffusely decreased signal intensity on T1-w images interspersed with additional hypointense foci [16, 81]. Those foci are often better demarcated on FS– T2-w/STIR images [46] (Fig.  8). In about 1–5% of patients, MRI of the bone marrow is characterized by a very inhomogeneous patchy micronodular pattern (the so-called ­“variegated” or “salt and pepper” pattern), with innumerable small bone marrow focal lesions and interposition of fat islands [14, 16, 46] (Fig. 9). In histologic specimens, this pattern corresponds to bone marrow with circumscribed fat islands beside normal bone marrow and a minor plasma cells infiltration (25 focal lesions on whole-­body MRI or >7 lesions on axial MRI) and high-risk cytogenetics identified a subgroup of patients with median progression-free survival (PFS) and OS of 23 and 56 months, respectively. In another study [90] conducted on 126 patients with newly diagnosed symptomatic myeloma eligible for ASCT, researchers found that diffuse and variegated MRI patterns had an independent predictive value for disease progression. In a more recent study conducted on 206 patients with newly diagnosed MM in order to assess the association between whole-body MRI findings and baseline disease features, Mai and colleagues [91] confirmed that a moderate to severe diffuse infiltration pattern and a continuously increasing number of focal lesions (cut-off number: 23) adversely affected survival, while a variegated pattern had a favorable prognosis despite its association with unfavorable cytogenetic abnormalities and adverse disease features. Despite the prognostic significance of MRI findings, last IMWG consensus statement advocates that additional prospective clinical studies are needed to define if these patients have to be treated in a different way [14]. In fact, currently accepted prognostic scores predicting PFS and OS of newly diagnosed MM patients are exclusively based on the evaluation of tumor burden with the International Staging System (ISS), both in alternative or in combination with disease biology (i.e. cytogenetic analyses and lactate dehydrogenase levels) [1]. Nevertheless, MRI pattern recognition and assignment do not represent a mere exercise for its own sake. Radiologists should be aware of the prognostic significance of these findings and should report the bone marrow involvement pattern on MRI for prognostic assessment [94].

Role of Standard Magnetic Resonance Imaging

55

The Role of Conventional MRI in Smouldering Myeloma Smouldering multiple myeloma represents an advanced premalignant stage clinically positioned between MGUS and MM. Patients affected by SMM have a risk of progression to symptomatic disease of approximately 10% per year over the first 5 years following diagnosis [1, 7]. The current definition of SMM comprises a heterogeneous group of patients with different clinical features as well as different risk profiles. Within this group, “high-­ risk” patients are affected by a 25% per year risk of progression in the first 2 years and they are characterized by the presence of one or more clinical, cytogenetic and laboratory specific features [95]. Among these criteria, MRI is of great value both in definition and prognostic assessment of patients with SMM and actually it is included in the standard work-up for SMM by the IMWG [2, 14]. Abnormal MRI features in newly diagnosed SMM have been associated with increased risk of progression to MM [27, 28, 30]. In the work of Moulopoulos and colleagues [30], an abnormal MR study of the spine identified asymptomatic patients who were likely to require treatment after a median of 16 months versus 43 months for those with normal MRI (p 5  mm on MRI are now considered as having symptomatic disease requiring treatment compared with previous recommendations which required evidence of cortical lytic destruction on plain film or CT [1, 2]. In the United Kingdom, whole body (WB) MRI has been recommended as first line imaging in all patients with a suspected diagnosis of myeloma (NICE). These recommendations were influenced by the high sensitivity of MRI and the capabilities for detecting mechanical complications in addition to extramedullary disease. The risk of progression to symptomatic myeloma for patients with SMM is about 8% per year after diagnosis [3]. The study by Mateos et  al. [4] demonstrated a potential benefit of early therapy for high risk SMM patients and evidence suggests that MRI findings can be prognostic [5–7]. The SWOG S0120 study reported that detection of multiple focal lesions on MRI conferred an increased risk of progression [8] and abnormal signal on MRI has been shown to be associated with very high risk of SMM progression and with development of lytic bone lesions [3]. When MRI spine has been used for patients with SMM [9], around 16% will have focal lesions on MRI spine [3].

C. Messiou (*) · D.-M. Koh Department of Radiology, The Royal Marsden and The Institute of Cancer Research, London, UK e-mail: [email protected]; [email protected] © Springer Nature Switzerland AG 2019 C. Nanni et al. (eds.), Molecular Imaging in Multiple Myeloma, https://doi.org/10.1007/978-3-030-19019-4_5

73

74

C. Messiou and D.-M. Koh

Different patterns of myeloma marrow infiltration are encountered on MRI. The presence of focal lesions and a number greater than one have been shown to be the strongest adverse prognostic factors for progression followed by diffuse infiltration [6, 10]. Whole body coverage is highly desirable as 50% of lesions would be missed by imaging the spine alone [11]. Patients with myeloma and normal MRI appearances respond better to therapy with longer survival [12]. Micronodular or “salt and pepper” pattern of marrow infiltration is also recognized, which is associated with a better prognosis than focal or diffuse disease [13]. Between 3 and 8% of patients are also thought to have extramedullary sites of disease although these figures are likely to increase with the growing trend towards use of cross sectional imaging [14, 15]. Extramedullary disease detection is relevant as an independent prognostic factor influencing progression-free survival [16].

Whole Body MRI Protocols A core WB-MRI can be acquired within about 45 min of scanning time and contemporary protocols now comprise anatomical and functional MRI sequences. Although it has been shown that WB-MRI is generally well tolerated [17], it must be recognised that 45 min of scanning time is not always practical when the patient is acutely unwell or has significant pain, and in these circumstances an abbreviated protocol can be used to acquire critical information such as imaging of the spinal column. Pain and claustrophobia may be mitigated by arranging analgesia or light sedation respectively and good communication with the clinical team is essential. Both 1.5 Tesla and 3 Tesla scanners can be used but magnetic field heterogeneity in the latter makes 1.5 Tesla the field strength of choice [18, 19]. More recent 3 Tesla scanners with image-based shimming technology can significantly improve the quality of WB-MRI studies. Sagittal T1- and T2-weighted sequences of the spine allow disease detection, as well as assessment of mechanical complications. Axial (and/or coronal) gradient-­ echo DIXON T1-weighted images supply additional anatomical information for disease detection and localisation but also allow semi-quantitative relative water and fat fraction maps to be derived, which contribute to disease characterisation and response assessments although detailed discussion is beyond the scope of this chapter [20]. The introduction of whole body diffusion-weighted MRI (WB DW MRI) [21] has revolutionised WB-MRI and it is now a core protocol component. This is usually performed in the axial plane, extending from the skull vertex to the knees, although coronal imaging is preferred at some institutions. The speed relative to existing MRI sequences, coverage, high sensitivity and the capability to quantify burden of disease and response to treatment has led to increasing adoption at leading centres worldwide for imaging malignant marrow disease. These capabilities are achieved without need for intravenous access or radiation exposures.

Whole Body Diffusion-Weighted Magnetic Resonance Imaging: A New Era for Whole…

75

Basic Principles of Diffusion-Weighted MRI DW-MRI produces images where the contrast between tissues is based on differences in the motion of water at a cellular level. Water motion in tissues occurs within different compartments  – intracellular, transmembrane, extracellular and intravascular. The choice of diffusion weighting (b value) influences the sensitivity to water diffusion in the different compartments. Low b values are sensitive to large diffusion distances, such as flow in blood vessels. Large b values are sensitive to small diffusion distances postulated to be related to extracellular space distances and therefore thought to be directly related to cell packing/cellularity. Theoretically, with very large b values (> 4000 s/mm2) it is possible to probe the movement of intracellular water protons, but this is extremely challenging. Imaging with more than one b value allows automated calculation of the Apparent Diffusion Coefficient (ADC) for each voxel in the image and a quantitative map can be produced. Tumours which consist of tightly packed cells therefore appear as areas of highly impeded water diffusion -high signal on source diffusion images and low values on ADC [22]. Following treatment reduced cellularity results in diminishing signal at high b value DW MRI and high values on ADC (Fig. 1). For patients with myeloma it has been shown that b values of around 1400 smm−2 are optimal for maximizing contrast between normal and infiltrated marrow [23]. However, there are technical challenges to achieving such high b values. Therefore, a b value of 900 smm−2 is often chosen as a compromise between robust data and WB coverage achievable in a reasonable time frame. The reported inverse correlation between cell density in soft tissues and ADC [24–27] is further supported by evidence that choline, a marker of cell turnover, is inversely correlated with ADC in glioma [28]. However, the relationship between cellularity and ADC has not been so impressive in all types [24, 28]. Other factors may influence ADC, for example cellular architecture, cell size and size variability within tissue, viscosity of cytoplasm, bulk flow in capillaries and active transport have been implicated. It is likely therefore that ADC is a complex function of tissue microarchitecture influenced by several components. In response to treatment, the increased extracellular spaces within a tumour produces increases in distances of water motion and an increase in ADC [29]. This has been demonstrated in several tumour types including brain, breast, prostate and liver metastases [29–32], where it has been used to predict treatment response ahead of dimension changes and serum markers [33, 34]. Some studies have used DW-MRI to detect the emergence of drug resistance during the course of therapy [35]. However, the presence of fat in marrow necessitates a slightly adapted approach. In normal adult marrow, fat predominates and there is a paucity of free water. The motion of the small amount of water which is present is restricted by fat and hence normal adult marrow has very little signal on diffusion-weighted MRI. As cellularity in marrow increases secondary either to disease or increased hematopoietic tissue, the amount of free water increases and so does ADC [36, 37]. It is thought that the increased vascularity associated with plasma cell infiltration is also influential [37].

76

C. Messiou and D.-M. Koh

a

e

b

f

c

g

d

h

Fig. 1  Interpreting diffusion-weighted MRI. Axial DW MRI upper thorax of a 58 year-old male with a focal myeloma deposit in the medial right clavicle (arrows) before (a–d) and after (e–h) treatment. Baseline images show that the b 50 smm−2 DW MRI and b 900 smm−2 DW MRI both demonstrate high signal from the focal deposit. These images are automatically used to construct a corresponding quantitative ADC map (c). The mean ADC of the focal clavicular deposit was 705 μm2/s. A maximum intensity projection image (MIP, e) can also be produced from the b 900 smm−2 DW MRI and this shows an additional deposit in the right pelvis (dashed arrow). MIP images provide a useful overview of sites of abnormal signal but should never be interpreted in isolation. Following treatment (e–h) although the right clavicle deposit did not change significantly in size, there was a marked rise in ADC from 705 to 2387 μm2/s which is equivalent to fluid. Despite stable dimensions, DW MRI is able to demonstrate that the densely cellular deposit has become hypocellular/acellular following treatment. It is interesting to note that there is persistence of some signal on the axial and MIP b 900 smm−2 images. This is due to “T2 shine through” effects and it is therefore essential that every focal lesion is interrogated on the corresponding ADC map in order to assess disease status

Whole Body Diffusion-Weighted Magnetic Resonance Imaging: A New Era for Whole…

77

Hence in adults the different microarchitecture of plasma cell infiltrated marrow results in markedly different signal on DW MRI and contrasts against normal marrow which returns little signal, leading to their enhanced detection.

Clinical Applications of WB DW MRI Disease Detection The excellent image contrast between normal and diseased marrow on DW MRI results in superior lesion conspicuity compared to conventional STIR and contrast enhanced MRI sequences [38–40]. Furthermore, the ADC value of myeloma infiltrated marrow is significantly different to normal adult marrow with little overlap [10, 23]. This means that qualitative differences in image contrast translate into quantitative differences and the ADC value can be used to separate myelomatous from normal marrow with a sensitivity of 90% and specificity of 93%. The ADC of marrow of patients in remission and with monoclonal gammopathy of undetermined significance (MGUS) are not significantly different to normal age matched volunteers making WB DW MRI a promising surveillance tool [23, 41]. We should note however that the new IMWG guidelines stipulate focal lesions only as an indication to treat in patients with asymptomatic myeloma despite the poor prognosis associated with diffuse infiltration. This is perhaps because diagnosis of diffuse infiltration on conventional anatomical MRI is challenging and is most often a subjective diagnosis based on comparison of marrow signal with intervertebral discs. Thus, there is potential for functional DW MRI and ADC measurements to reduce this subjectivity and this should be a priority for future studies (Fig. 2). Although WB DW-MRI is emerging as one of the most sensitive tools for imaging bone marrow [38, 42, 43], some debate remains as to its specificity. Lecouvet et al. [42] presented data to suggest high specificity for detection of metastatic bone disease (98–100%), but a recent meta-analysis showed a lower pooled specificity of 86.1% [43]. The paucity of myeloma-specific prospective studies and marked heterogeneity in reference standards make current judgments on the specificity of MRI challenging, especially since biopsy of all imaging-detected lesions is not feasible. The approach offered by the IMWG of 3–6 month follow-up of equivocal solitary small lesions is a pragmatic solution. The skull and ribs have historically been difficult sites for interrogation using MRI; however diffusion-weighted imaging has shown increased sensitivity for lesion detection in the ribs compared to skeletal survey [44, 45]. However, the sensitivity to lesion detection in the skull was reduced compared to skeletal survey in both studies. This is possibly because the small volume of marrow in the skull is challenging to interrogate against adjacent high diffusion signal in the brain. However false positives on plain film of the skull secondary to venous lakes and granulations are also possible and difficult to confirm.

78

C. Messiou and D.-M. Koh

a

b

c

d

e

f

Fig. 2  Diffusion-weighted MRI is sensitive for detection of diffuse disease and can quantify treatment response in this disease phenotype. MRI imaging in a 67 year-old female before (a–c) and following (d–f) treatment. At baseline, sagittal T1W MRI of the lumbosacral spine (a) shows marrow signal at the lower range of normal but this does not meet generally accepted imaging criteria for diffuse disease which requires marrow signal to be lower than intervertebral discs. However axial b 900 smm−2 DW MRI shows marrow signal greater than background muscle indicating a hypercellular bone marrow which is confirmed on the corresponding ADC map (c). Following treatment, sagittal T1W MRI of the lumbosacral spine (d) demonstrated increasing marrow signal which indicates returning normal marrow fat. Reduction in signal on b 900 smm−2 DW MRI confirms restoration of normal marrow architecture. The ADC map enables quantification of response with a reduction in ADC from 857 to 548 μm2/s. This reduction in ADC value is due to repopulation by fat containing yellow marrow

Assessment of Disease Status and Response to Treatment The capability of WB DW-MRI to demonstrate both focal and diffuse marrow infiltration throughout the whole skeleton makes this extremely promising as a tool for monitoring disease status and assessment of response. Focal lesion changes in size and number can be easily assessed and for diffuse infiltration, signal changes are evident. The use of DW MRI is essential for assessing response to treatment as focal lesions often persist on conventional MR sequences following treatment but DW MRI may reveal an ADC equivalent to water indicating a hypocellular treated site. Changes in ADC can also predate changes in size and allow for early assessment of disease status. MRI studies performed prior to the advent of DW MRI suffer from false positive findings following treatment and should be interpreted with caution (Figs. 2 and 3). ADC measurements offer the capability to quantify disease throughout the skeleton. Advances in data informatics have made semi-automated skeletal ­segmentation a reality which enables histogram quantification of a patient’s whole marrow. This

Whole Body Diffusion-Weighted Magnetic Resonance Imaging: A New Era for Whole…

a

b

c

d

79

Fig. 3  False positive T1-W MRI following treatment is resolved by DW MRI. Axial T1-W (a), b 50 smm−2 DW MRI (b), b 900 smm−2 DW MRI (c) and corresponding ADC map (d) of the bony pelvis of a patient with myeloma following autograft. T1-W MRI demonstrates two focal lesions (arrows) which can easily be misinterpreted as active focal lesions. The focal lesions are also evident on corresponding DW MRI performed at b50 (b) and b900 smm−2 (c). System software automatically generates an ADC map (d) which shows that both focal lesions (arrows) have a very high ADC almost equivalent to water in small bowel (dashed arrow). The lesions are therefore acellular and inactive sites of disease. This functional assessment of disease status is only possible with interrogation of the DW MRI. T1-W MRI alone can be misleading. A combination of anatomical and functional imaging is required for accurate assessment of disease status. The persistence of high signal on b 900 smm−2 DW MRI (C) is due to “T2 shine through” effects

has been demonstrated by Giles et al. who used these techniques to segment patients’ bone marrow on diffusion-weighted MRI to quantify response to treatment [46]. Reassuringly the measurement repeatability was excellent with a coefficient of variation of 2.8%. Mean ADC increased in 95% of responding patients and decreased in all non-responders (p