The Milan System for Reporting Salivary Gland Cytopathology [2 ed.] 3031266617, 9783031266614

Citation preview

The Milan System for Reporting Salivary Gland Cytopathology Second Edition William C. Faquin Esther Diana Rossi Editors Zubair Baloch Güliz A. Barkan Maria Pia Foschini Daniel F.I. Kurtycz Marc P. Pusztaszeri Philippe Vielh Associate Editors

123

The Milan System for Reporting Salivary Gland Cytopathology

William C. Faquin  •  Esther Diana Rossi Editors

Zubair Baloch  •  Güliz A. Barkan Maria Pia Foschini  •  Daniel F.I. Kurtycz Marc P. Pusztaszeri  •  Philippe Vielh Associate Editors

The Milan System for Reporting Salivary Gland Cytopathology Second Edition

Editors William C. Faquin Department of Pathology Massachusetts General Hospital Boston, MA, USA

Esther Diana Rossi Division of Anatomic Pathology and Histology Catholic University of the Sacred Heart Rome, Italy

Associate Editors Zubair Baloch Department of Pathology Hospital of the University of Pennsylvan Philadelphia, PA, USA

Güliz A. Barkan Rm2257, Bldg 110 Loyola University Medical Center Maywood, IL, USA

Maria Pia Foschini Biomedical and Neuromotor Sciences University of Bologna Bologna, Italy

Daniel F.I. Kurtycz Pathology and Laboratory Medicine, University of Wisconsin–Madison Madison, WI, USA

Marc P. Pusztaszeri Department of Pathology Jewish General Hospital, McGill University Montreal, QC, Canada

Philippe Vielh Department of Pathology Medipath and American Hospital of Paris Paris, France

ISBN 978-3-031-26661-4    ISBN 978-3-031-26662-1 (eBook) https://doi.org/10.1007/978-3-031-26662-1 © American Society of Cytopathology 2023 This work is subject to copyright. All rights are solely and exclusively licensed 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

—William C. Faquin To my devoted husband and favorite poet, Judson, and in loving memory of my mother, Mary Clay, who was always a constant support and inspiration. Also, to Aliza and Eti who embody the true essence of a loving and caring spirit. “How can we know the dancer from the dance?” WB Yeats —Esther D. Rossi To my father, Antonio, the first Rossi in my family who became a professor and taught me the sense of duties and the real meaning of the words “perseverance, resiliency and believe in yourself.” To my lovely mom, Francesca, and my sister, Marta, the gems and passion of my life. They are always the safe harbor and the bright lighthouse of my life and the deepest and warmest joy of my days. Chapter 7 of this Second Edition Atlas is dedicated by the editors and co-authors to the loving memory of Dr. Stefan E. Pambuccian who was instrumental in formulating and writing Chapter 7 in the First Edition.

Foreword

As the author of this foreword, I have a terrible confession to get out of the way: I am not a cytopathologist. As a head and neck surgical pathologist, however, my work is closely tied to that of very talented cytopathologists who are routinely tasked with triaging various masses in the head and neck region for appropriate management. My association and affinity for cytopathology began with Dr. Syed Ali during our time together at The Johns Hopkins Hospital and has continued to bring me into close working relationships with treasured colleagues across the world like Drs. Rossi and Faquin. Naturally, I was thrilled when they asked me to write a foreword for their new book. This is an incredibly exciting time to be studying salivary gland neoplasms. We are in a revolution of understanding of these enigmatic tumors, largely due to the increasing accessibility of molecular tools like RNA sequencing and fluorescence in situ hybridization. With the awareness of certain genetic gold standards, pathologists are now just beginning to be able to appreciate fully the entire, frankly bizarre, histologic spectra of established tumor types like mucoepidermoid carcinoma and adenoid cystic carcinoma. For example, we now know that mucoepidermoid carcinoma can have cilia or even lack “epidermoid” cells altogether. Moreover, we now have the tools to define entirely novel tumor types, allowing for precise refinement of classification schemes. It was possible, for example, to introduce microsecretory adenocarcinoma with just five cases and no follow-up—unthinkable in prior eras— because of the existence of a unique histologic appearance combined with a novel MEF2C::SS18 fusion. Despite the cutting-edge advances in salivary gland tumor classification and diagnosis, it remains a fact that perhaps its single most effective diagnostic tool is fine needle aspiration. This time-honored technique remains incredibly powerful in triaging patients to their appropriate care. In this book, Drs. Rossi, Faquin, and colleagues build on the success of the 2018 Milan System for Reporting Salivary Gland Cytopathology (MSRSGC), which was in turn designed to emulate the triumphs of the Bethesda systems for reporting cervical cytology and thyroid cytopathology. The MSRSGC applied much-needed order to the formerly haphazard process of reporting salivary gland cytopathology by guiding pathologists to well-defined, vii

viii

Foreword

reproducible diagnostic categories which are tied to risks for malignancy and next therapeutic steps. Countless studies have demonstrated that MSRSGC has met its goals and is an unqualified success. The second edition of MSRSGC refines the system and incorporates newly understood concepts in classification and molecular analysis. I have no doubt of its continued adoption across institutions, and I am confident MSRSGC will realize its full potential of standardizing terminology to improve the care of patients with salivary gland tumors. I celebrate the editors and authors of this endeavor, and I look forward to a continued close collaborative relationship with my partners in diagnostic salivary gland pathology. University of Texas Southwestern Medical Center Dallas, TX, USA

Justin Bishop, MD

Preface

The Second Edition of The Milan System for Reporting Salivary Gland Cytopathology, like the First Edition, represents a collaborative effort by a multidisciplinary group of cytopathologists, surgical pathologists, molecular pathologists, radiologists, and head and neck surgeons. This international group shares the goal of creating a practical and uniform reporting system for salivary gland fine needle aspiration (FNA). The initial idea for the Milan System was codified in 2015 by a taskforce, coordinated by Drs. Faquin and Rossi, during the European Cytology Congress in Milan, Italy, hence the system’s name. The atlas is supported by the American Society of Cytopathology, the International Academy of Cytology, and the European Federation of Cytology Societies. The Second Edition of the Milan System Atlas has been updated based upon numerous discussions and subgroup meetings among co-authors and associate editors, careful review, and analysis of the latest salivary gland cytology publications, as well as a 2021 online survey organized by Dr. Daniel Kurtycz. The atlas includes an international group of 58 co-authors from approximately 20 countries. Like the First Edition, the atlas is organized into six general diagnostic categories: “Non-­ Diagnostic,” “Non-Neoplastic,” “Atypia of Undetermined Significance (AUS),” “Neoplasm: Benign,” “Neoplasm: Salivary Gland Neoplasm of Uncertain Malignant Potential (SUMP),” “Suspicious for Malignancy,” and “Malignant.” Dr. Zubair Baloch was instrumental in calculating and updating risks of malignancy (ROM) associated with each diagnostic category based on evidence in the current literature. Since the publication of the First Edition in 2018, there have been over 100 publications focusing on specific aspects of the Milan System, and it has been translated into Japanese and Chinese. The Second Edition includes definitions, morphologic criteria, and explanations for each of the diagnostic categories. Specific chapters are dedicated to the application of the latest available ancillary studies, radiologic features of salivary gland lesions, clinical management, and histological considerations including updates from the most recent fifth edition WHO blue book. The Milan System for Reporting Salivary Gland Cytopathology represents an essential step towards increasing the overall effectiveness of salivary gland FNA and fostering better communication between clinicians and between institutions to ix

x

Preface

improve overall patient care. All the contributors to this Second Edition trust that it will offer a practical and useful reporting system, meeting the needs of the international cytology community and bettering the lives of the patients we serve. Boston, MA, USA Rome, Italy 

William C. Faquin Esther Diana Rossi

Contents

1

The Milan System for Reporting Salivary Gland Cytopathology������������������������������������������������������������������������������    1 Zubair Baloch, Daniel Lubin, Nora Katabi, Bruce M. Wenig, and Eva M. Wojcik

2

Non-Diagnostic ����������������������������������������������������������������������������������������   15 Maria Pia Foschini, Esther Diana Rossi, Kayoko Higuchi, Ivana Kholova, Nirag Jhala, Makoto Urano, Laszlo Vass, Philippe Vielh, Daniel Johnson, and Valentina Robila

3

Non-Neoplastic ����������������������������������������������������������������������������������������   27 William C. Faquin, Massimo Bongiovanni, Fabiano Callegari, Tarik M. Elsheikh, Daniel F.I. Kurtycz, Oscar Lin, Ravi Mehrotra, Marc P. Pusztaszeri, Judy Qiuying Shi, and Zhihui Zhang

4

 Atypia of Undetermined Significance (AUS) ����������������������������������������   49 Marc P. Pusztaszeri, Zubair Baloch, William C. Faquin, Christopher C. Griffith, Zahra Maleki, Celeste N. Powers, Esther Diana Rossi, and Z. Laura Tabatabai

5

Neoplasm��������������������������������������������������������������������������������������������������   71 Zubair Baloch, Guido Fadda, Christopher C. Griffith, Jerzy Klijanienko, Jeffrey F. Krane, Lester Layfield, Zahra Maleki, Ritu Nayar, and Marc P. Pusztaszeri

6

Suspicious for Malignancy����������������������������������������������������������������������  105 Esther Diana Rossi, Syed Ali, Claude Bigorgne, Ashish Chandra, Yun Gong, Daniel Lubin, Renata B. Perak, Bo Ping, He Wang, and Matthew Zarka

xi

xii

Contents

7

Malignant��������������������������������������������������������������������������������������������������  117 Swati Mehrotra, Mousa Al-Abbadi, Claude Bigorgne, Jalal Jalalay, Jeffrey F. Krane, Renata B. Perak, Philippe Vielh, Paul E. Wakely Jr, He Wang, Eva M. Wojcik, and Güliz A. Barkan

8

 Ancillary Studies for Salivary Gland Cytology������������������������������������  165 Marc P. Pusztaszeri, Vickie Y. Jo, Jeffrey F. Krane, Xiaoyin “Sara” Jiang, and Philippe Vielh

9

 Imaging of the Salivary Glands��������������������������������������������������������������  191 Jillian W. Lazor and Joanie M. Garratt

10 Clinical Management������������������������������������������������������������������������������  217 Mandeep Bajwa, Piero Nicolai, Sidharth V. Puram, Joseph Zenga, and Mark Varvares 11 Histologic  Considerations and Salivary Gland Tumor Classification in Surgical Pathology������������������������������  229 Bruce M. Wenig, Vickie Y. Jo, Masato Nakaguro, and Alena Skálová Index������������������������������������������������������������������������������������������������������������������  237

Contributors

Mousa  Al-Abbadi  Pathology and Cytopathology, Jordan University Hospital, Amman, Jordan Histopathology, Microbiology and Forensic Medicine, University of Jordan– College of Medicine, Amman, Jordan Syed  Ali  Department Baltimore, MD, USA

of

Pathology,

The

Johns

Hopkins

Hospital,

Mandeep  Bajwa  Regional Maxillofacial Unit, Aintree University Hospital, Liverpool, UK Zubair Baloch  Pathology and Laboratory Medicine, University of Pennsylvania Medical Center, Philadelphia, PA, USA Güliz  A.  Barkan  Department of Pathology and Laboratory Medicine, Loyola University Hospital, Maywood, IL, USA Claude Bigorgne  Cytopathology and Imaging Center, Paris, France Massimo Bongiovanni  Synlab Pathology SA, Lausanne, Switzerland Fabiano  Callegari  Department of Pathology, São Paulo Hospital, Federal University of São Paulo (EPM-UNIFESP), São Paulo, SP, Brazil Ashish  Chandra  Cellular Pathology, Guy’s and St Thomas’ NHS Foundation Trust, London, UK Tarik  M.  Elsheikh  Department of Pathology, Cleveland Clinic, Pathology and Laboratory Medicine Institute, Cleveland, OH, USA Guido  Fadda  Department of Human Pathology of the Adulthood and of the Developing Age “Gaetano Barresi”, University of Messina Faculty of Medicine and Surgery, Messina, Italy William  C.  Faquin  Head and Neck Pathology, Massachusetts Eye and Ear Infirmary, Boston, MA, USA xiii

xiv

Contributors

Anatomic and Molecular Pathology, Massachusetts General Hospital, Harvard Medical School, Boston, MA, USA Maria  Pia  Foschini  Department of Biomedical and Neuromotor Sciences, University of Bologna, Anatomic Pathology, Bologna, Italy Joanie M. Garratt  Department of Radiology, The Hospital of the University of Pennsylvania, Philadelphia, PA, USA Yun  Gong  Department of Pathology, University of Texas MD Anderson Cancer Center, Houston, TX, USA Christopher C. Griffith  Anatomic Pathology, Robert J. Tomsich Pathology and Laboratory Medicine Institute, Cleveland Clinic, Cleveland, OH, USA Kayoko  Higuchi  Section Matsumoto, Japan

of

Anatomic

Pathology,

Aizawa

Hospital,

Jalal B. Jalaly  Head and Neck and Cytopathology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA Nirag  Jhala  Department of Pathology and Laboratory Medicine, Temple University Hospital, Philadelphia, PA, USA Xiaoyin “Sara” Jiang  Department of Pathology, Duke Health, Durham, NC, USA Daniel  Johnson  Department of Pathology and Laboratory Medicine, OSF HealthCare Little Company of Mary Medical Center, Evergreen Park, IL, USA Vickie Y. Jo  Department of Pathology, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA, USA Nora Katabi  Department of Pathology, Memorial Sloan Kettering Cancer Center, New York, NY, USA Ivana Kholova  Fimlab Laboratories, Department of Pathology, Tampere University Hospital and Tampere University, Tampere, Finland Jerzy Klijanienko  Department of Pathology, Institut Curie, Paris, France Jeffrey  F.  Krane  Department of Pathology and Laboratory Medicine, David Geffen School of Medicine at UCLA, Los Angeles, CA, USA Daniel  F.I.  Kurtycz  Department of Pathology and Laboratory Medicine and Wisconsin State Laboratory of Hygiene, University of Wisconsin Hospitals and Clinics, Madison, WI, USA Lester Layfield  Department of Pathology and Anatomical Science, University of Missouri, Columbia, MO, USA Jillian  W.  Lazor  Department of Radiology, The Hospital of the University of Pennsylvania, Philadelphia, PA, USA

Contributors

xv

Oscar  Lin  Department of Pathology, Memorial Sloan-Kettering Cancer Center, New York, NY, USA Daniel  Lubin  Department of Pathology and Laboratory Medicine, Emory University Hospital Midtown, Atlanta, GA, USA Zahra  Maleki  Division of Cytopathology, Department of Pathology, The Johns Hopkins Hospital, Baltimore, MD, USA Ravi  Mehrotra  Diagnostic Cytopathology, Centre for Health Innovation and Policy, Noida, India Swati  Mehrotra  Department of Pathology and Laboratory Medicine, Loyola University Hospital, Maywood, IL, USA Masato  Nakaguro  Department of Pathology and Laboratory Medicine, Nagoya University Graduate School of Medicine, Nagoya, Japan Ritu Nayar  Department of Pathology, Northwestern University Feinberg School of Medicine and Northwestern Memorial Hospital, Chicago, IL, USA Piero  Nicolai  Otorhinolaryngology–Head and Neck Surgery, University of Brescia, Brescia, Italy Renata  B.  Perak  Department of Pathology, Forensic Medicine and Cytology, University School of Medicine Split, University Hospital Center Split, Split, Croatia Bo Ping  Department of Pathology, Shanghai Cancer Center, Shanghai, China Celeste  N.  Powers  Division of Anatomic Pathology, Department of Pathology, VCU Health System, Medical College of Virginia Hospitals, Richmond, VA, USA Sid  Puram  Department of Otolaryngology—Head and Neck Surgery and Department of Genetics, Washington University School of Medicine, St. Louis, MO, USA Marc P. Pusztaszeri  Department of Pathology, Jewish General Hospital, Montréal, QC, Canada Department of Pathology, McGill University, Montréal, QC, Canada Valentina Robila  Department of Pathology, Virginia Commonwealth University, Richmond, VA, USA Esther Diana Rossi  Unita’ Operativa Istopatologia e Citodiagnostica, Fondazione Policlinico Universitario A. Gemelli, Rome, Italy Judy Qiuying Shi  Department of Pathology, Emory University Hospital Midtown, Atlanta, GA, USA Alena Skálová  Department of Pathology, Faculty of Medicine in Pilsen, Charles University, Pilsen, Czech Republic

xvi

Contributors

Z.  Laura  Tabatabai  Department of Pathology, University of California at San Francisco, San Francisco, CA, USA Vanda  Torous  Anatomic and Molecular Pathology, Massachusetts General Hospital and Harvard Medical School, Boston, MA, USA Makoto  Urano  Diagnostic Pathology, Fujita Health University, Toyoake, Aichi, Japan Mark  Varvares  Department of Otolaryngology, Massachusetts Eye and Ear, Harvard Medical School, Boston, MA, USA Philippe  Vielh  Department of Pathology, Medipath and American Hospital of Paris, Paris, France Paul  E.  Wakely Jr  The Ohio State U.  Wexner Medical Center, James Cancer Hospital and Solove Research Institute, Columbus, OH, USA He Wang  Department of Pathology, Yale University School of Medicine, Yale New Haven Health System St. Raphael Campus, New Haven, CT, USA Bruce M. Wenig  Department of Pathology, Moffitt Cancer Center, Tampa, FL, USA Eva  M.  Wojcik  Department of Pathology and Laboratory Medicine, Loyola University Hospital, Maywood, IL, USA Matthew Zarka  Department of Laboratory Medicine and Pathology, Mayo Clinic Arizona, Scottsdale, AZ, USA Joseph  Zenga  Department of Otolaryngology, Medical College of Wisconsin, Milwaukee, WI, USA Zhihui  Zhang  Department of Pathology, National Cancer Center, Chinese Academy of Medical Sciences, Peking Union Medical College, Cancer Hospital, Beijing, China Christina  Zioga  Department of Cytopathology, “G.  Papanicolaou” General Hospital of Thessaloniki, Thessaloniki, Greece

Abbreviations

F-FDG PET/CT F-18 fluorodeoxyglucose positron emission tomography/computed tomography ACC Acinic cell carcinoma ADC Apparent diffusion coefficient AdCC Adenoid cystic carcinoma AUS Atypia of Undetermined Significance BCA Basal cell adenoma BCAd Basal cell adenocarcinoma CA-ex-PA Carcinoma ex pleomorphic adenoma CAMSG Cribriform adenocarcinoma of the minor salivary gland CCC Clear cell carcinoma CNB Core needle biopsy CT Computed tomography EBER Epstein-Barr encoded small RNAs EBV Epstein-Barr virus ECD Extra-capsular dissection EMC Epithelial-myoepithelial carcinoma EMZL Extranodal marginal zone lymphoma FFPE Formalin fixed paraffin embedded FISH Fluorescent in situ hybridization FNA Fine needle aspiration IC Immunochemistry IDC Intraductal carcinoma LEC Lymphoepithelial carcinoma LESA Lymphoepithelial sialadenitis MEC Mucoepidermoid carcinoma MRI Magnetic resonance imaging MSRSGC Milan System for Reporting Salivary Gland Cytopathology NGS Next-generation sequencing NPP Nerve-preserving parotidectomy PA Pleomorphic adenoma 18

xvii

xviii

Abbreviations

PAC Polymorphous adenocarcinoma PAS Periodic acid-Schiff RON Risk of neoplasm ROSE Rapid on-site evaluation RT-PCR Reverse transcription polymerase chain reaction SC Secretory carcinoma SCC Squamous cell carcinoma SDC Salivary duct carcinoma SUMP Salivary Gland Neoplasm of Uncertain Malignant Potential TALP Tumor-associated lymphoid proliferation US Ultrasound USG Ultrasound guidance WT Warthin tumor

Chapter 1

The Milan System for Reporting Salivary Gland Cytopathology Zubair Baloch, Daniel Lubin, Nora Katabi, Bruce M. Wenig, and Eva M. Wojcik

Introduction Fine-needle aspiration (FNA) is now widely accepted as an efficient first-line diagnostic test for the management of salivary gland lesions. It can differentiate between neoplastic and non-neoplastic salivary gland lesions, and in cases of a neoplasm, FNA can diagnose a majority of common benign tumors [1–8]. In most cases, FNA can also differentiate between low- and high-grade carcinomas. Neoplastic salivary gland lesions are usually managed surgically, while non-neoplastic ones are managed conservatively, usually without surgical intervention. Knowing whether a

Z. Baloch (*) Pathology and Laboratory Medicine, University of Pennsylvania Medical Center, Philadelphia, PA, USA e-mail: [email protected] D. Lubin Department of Pathology and Laboratory Medicine, Emory University Hospital Midtown, Atlanta, GA, USA e-mail: [email protected] N. Katabi Department of Pathology, Memorial Sloan Kettering Cancer Center, New York, NY, USA e-mail: [email protected] B. M. Wenig Department of Pathology, Moffitt Cancer Center, Tampa, FL, USA e-mail: [email protected] E. M. Wojcik Department of Pathology and Laboratory Medicine, Loyola University Hospital, Maywood, IL, USA e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 W. C. Faquin et al. (eds.), The Milan System for Reporting Salivary Gland Cytopathology, https://doi.org/10.1007/978-3-031-26662-1_1

1

2

Z. Baloch et al.

carcinoma is low- or high-grade can also help in determining the extent of surgery, including decisions on preservation of the facial nerve in the case of parotid tumors, and indications for neck dissection. In a subset of benign neoplasms, such as pleomorphic adenoma (PA) and Warthin tumor (WT), the specific FNA diagnosis allows the option for managing the tumor nonsurgically by clinical follow-up and imaging, depending upon patient wishes and health status [8–14]. The risk of malignancy (ROM) prior to FNA for a salivary gland mass varies depending upon its size and location, from 20–25% in the parotid gland to 40–50% in the submandibular gland, and 50–81% in the sublingual and minor salivary glands [2, 4–7, 13, 15–17]. Salivary gland FNA test performance shows a range of sensitivities and specificities depending upon a variety of factors including technical experience of the operator performing the FNA, quality of the cytologic preparations, experience of the cytopathologist, morphologic heterogeneity of the lesion, and presence of a cystic component. The reported overall sensitivity and specificity of salivary gland FNA in most series ranges from 91 to 93.94% and 94.92 to 97.48%, respectively [7, 13, 15–62]. False-negative and false-positive diagnoses are uncommon. The reported ranges of sensitivity and specificity to differentiate neoplastic from non-neoplastic salivary gland lesions are 79–100% and 71–100%, respectively while the accuracy of FNA in distinguishing benign from malignant salivary gland lesions ranges from 81 to 100% [1–8, 12]. In contrast, the accuracy of salivary gland FNA when used to specifically subtype a neoplasm shows a wider range, varying from 48 to 94% [7, 13, 15–62]. The diagnostic challenges posed during the evaluation of salivary gland FNA are mainly due to the inherent morphologic complexity of salivary gland lesions. However, these are further complicated by the lack of a standardized, tiered diagnostic framework by which salivary gland FNA can be reported. The MSRSGC has shown to improve the overall effectiveness of salivary gland FNA. It emphasizes risk stratification rather than specific diagnoses, providing a ROM for each ascending risk category rather than a binary benign or malignant assessment for each individual case [3, 5, 7, 13, 15–70]. The objective of the MSRSGC is to foster better communication between clinicians and between institutions in order to improve overall patient care. Similar to its first iteration, the MSRSGC consists of six diagnostic categories including a non-neoplastic category, and a neoplastic category split into Benign and Salivary Gland Neoplasm of Uncertain Malignant Potential (SUMP) (Table 1.1). It is an evidence-based system derived from the literature which correlates diagnostic categories with ROM and clinical management strategies (Table 1.2). The ROM for each diagnostic category has been refined and reflects studies published after the first edition of MSRSGC [7, 13, 15–62, 71–74].

1  The Milan System for Reporting Salivary Gland Cytopathology

3

Table 1.1  The Milan System for Reporting Salivary Gland Cytopathology (MSRSGC): diagnostic categories, definitions, and explanatory notes Diagnostic category and definitions •  I. Non-Diagnostic Insufficient cellular material for a cytologic diagnosis •  II. Non-Neoplastic Benign entities such as chronic sialadenitis, reactive lymph node, granulomas, and infection

•  III. Atypia of Undetermined Significance (AUS) (≤10% of all salivary gland FNA samples); containing limited atypia; indefinite for a neoplasm •  IV. Neoplasm • A) Benign Reserved for benign neoplasms diagnosed based on established cytologic criteria • B) Salivary Gland Neoplasm of Uncertain Malignant Potential (SUMP) Reserved for FNA samples which are diagnostic of a neoplasm; however, a diagnosis of a specific entity cannot be made •  V. Suspicious for Malignancy This category is for FNA samples showing features that are highly suggestive of, but not unequivocal for malignancy •  VI. Malignant This category is for FNA specimens which are diagnostic of malignancy

Explanatory notes • This diagnostic category should only be used after all the material has been processed and examined • Exceptions include matrix material and mucinous cyst contents • The ROM for this category would be expected to be low if strict inclusion criteria are applied • Specimens will include those lacking cytomorphologic evidence of a neoplastic process • Inflammatory, metaplastic, and reactive changes • Specimens showing evidence of reactive lymphoid tissue (flow cytometry is recommended based on clinical and morphologic suspicion) • Samples are indeterminate for a neoplasm; a neoplastic process cannot be excluded after examination of all the cellular material • A majority of these FNAs will represent reactive atypia or poorly sampled neoplasms • This category will include classic cases of pleomorphic adenoma, Warthin tumor, lipoma, etc.

• This diagnosis should be used for cases where a malignant neoplasm cannot be excluded • A majority of these cases will include cellular benign neoplasms, neoplasms with atypical features, and low-grade carcinomas

• The FNA report should state which type of malignant tumor is suspected or provide a differential diagnosis • A majority of specimens in this category will be high-grade carcinomas • An attempt should be made to sub-classify the neoplasm into specific types and grades of carcinoma: e.g., low-grade (low-grade mucoepidermoid carcinoma) vs. high-grade (salivary duct carcinoma) • “Other” malignancies such as lymphomas, metastases, and sarcomas are also included in this category and should be specifically designated

4

Z. Baloch et al.

Table 1.2  The Milan System for Reporting Salivary Gland Cytopathology (MSRSGC): implied risk of malignancy and recommended clinical management Diagnostic category •  I. Non-Diagnosticc •  II. Non-Neoplastic •  III. Atypia of Undetermined Significance (AUS) •  IV. Neoplasm • IVA. Neoplasm: Benign • IVB. Neoplasm: Salivary Gland Neoplasm of Uncertain Malignant Potential (SUMP)e •  V. Suspicious for Malignancy •  VI. Malignant

% ROMa Managementb 15% Clinical and radiologic correlation/repeat FNA 11% Clinical follow-up and radiologic correlation 30% Repeat FNA or surgery 40 years old), even when there are no obvious features of malignancy. A wide variety of other less common scenarios where a neoplasm cannot be ruled out after examination of all the cellular material can be encountered and may be appropriate for classification within the AUS category [12–15].

66

M. P. Pusztaszeri et al.

Fig. 4.18  Atypia of Undetermined Significance vs suspicious for malignancy. This image showing a collection of cytologically bland keratinizing squamous cells raises a differential diagnosis of metastatic squamous cell carcinoma versus reactive squamous atypia in a benign squamous cyst. Clinical context and quality of the FNA sample will influence the cytologic classification (smear, Papanicolaou stain)

Prevalence and Risk of Malignancy As in other cytology reporting systems, the AUS category is a last resort diagnosis that should be used judiciously. Cytopathologists should make every attempt to classify specimens using other more specific categories whenever possible. It is recommended that no more than 10% of salivary gland FNAs be interpreted as AUS.  In retrospective studies, the proportion of FNAs diagnosed as AUS was approximately 4% with a wide range (0–73%), highlighting the vast differences in salivary gland FNA practices across different institutions [14]. There may be a role for intra-laboratory monitoring of the AUS rate to avoid overuse of this category. It is recommended that the entire FNA specimen be processed for cytomorphologic interpretation before rendering a diagnosis of AUS.  AUS rates may be further reduced by the use of ROSE and/or ancillary studies, especially flow cytometry for lymphoid lesions [13]. The risk of malignancy (ROM) for the AUS category is expected to be in between the ROMs of the Non-Neoplastic and SUMP categories, which was estimated to be 10–35% (approximately 20%) in the first edition of the MSRSGC, based on a limited number of studies that classified cases either as atypical or inconclusive. Based on subsequent retrospective studies using the MSRSGC criteria, the ROM for the AUS category has been refined and is estimated to be 30% on average, with a wide range and some overlap with the SUMP category (see Chap. 1) [12–17]. However, the risk of neoplasia (RON) for the AUS category is significantly lower than for the SUMP category (approximately 63% vs >90%) [12–17]. Different scenarios for AUS may be associated with different RONs and ROMs but further studies are required [12, 13, 15, 17]. For example, the reported ROM (i.e., low-grade

4  Atypia of Undetermined Significance (AUS)

67

mucoepidermoid carcinoma) for mucinous cyst contents may be as high as 60–80% for cases with histologic follow-up [17], but this is likely attributable to the fact that only cases with a high suspicion for malignancy undergo surgical intervention (selection bias).

Management A diagnosis of AUS should lead to careful correlation with clinical and radiologic findings, and management should be individualized for each patient. The use of a multidisciplinary team discussion (tumor board) should be considered. Depending upon the overall risk assessment, the nature of the lesion, and institutional practices, it may result in clinical follow-up, repeat FNA, core-needle biopsy, open biopsy, or surgical excision with or without intraoperative frozen section. For cystic lesions, if the lesion disappears completely after evacuation of the fluid, it is most likely non-neoplastic and clinical follow-up is recommended. If there is any residual mass, re-aspiration using ultrasound guidance can help to achieve a more specific cytologic diagnosis. Surgical excision is usually indicated if the cyst does not resolve by aspiration or if it recurs since over one-third of cystic salivary gland lesions are neoplastic. In aspirates with an atypical lymphoid population, flow cytometry, immunochemistry, and/or tissue biopsy should be considered to rule out a lymphoproliferative disorder and allow specific classification.

Sample Reports Example 7.1 Sample report of mucinous cyst contents. Evaluation limited by scant cellularity. ATYPIA OF UNDETERMINED SIGNIFICANCE. Histiocytes +/− scant epithelial cells in a background of abundant mucin. See Note. Note: The differential diagnosis of mucin-containing cysts includes mucocele, mucus retention cysts, and low-grade mucoepidermoid carcinoma. Clinical and radiological correlation is recommended. Aspiration of a residual mass, if present, may help to achieve a more specific diagnosis. If the lesion persists or recurs, excision should be considered. Example 7.2 Sample report of scant basaloid cells. Evaluation limited by scant cellularity. ATYPIA OF UNDETERMINED SIGNIFICANCE. Few clusters of basaloid cells with mild atypia. See note.

68

M. P. Pusztaszeri et al.

Note: While the aspirate may represent chronic sialadenitis with metaplasia and reactive changes, a salivary gland neoplasm with basaloid features cannot be completely excluded. Recommend clinical and radiologic correlations with consideration for additional sampling if clinically indicated. Example 7.3 Sample report of scant oncocytic cells. Evaluation limited by scant cellularity. ATYPIA OF UNDETERMINED SIGNIFICANCE. Rare oncocytic cells with cytological and/or architectural atypia. See note. Note: While the aspirate may represent oncocytic metaplasia or oncocytic hyperplasia of the salivary gland, a neoplastic process with oncocytic features or oncocytic metaplasia cannot be entirely excluded. Recommend clinical and radiologic correlations and additional sampling if clinically indicated. Example 7.4 Sample report of a lymphoid-rich aspirate. Satisfactory for evaluation. ATYPIA OF UNDETERMINED SIGNIFICANCE. Abundant mixed population of lymphocytes with occasional atypical forms. See note. Note: The aspirate is suggestive of a reactive lymph node, but in the absence of flow cytometry, a low-grade lymphoproliferative disorder cannot be entirely excluded. Clinical and radiological correlations are recommended with consideration for repeat sampling for flow cytometry studies. Example 7.5 Sample report of a lymphoid-rich aspirate with scant epithelial cells. Satisfactory for evaluation. ATYPIA OF UNDETERMINED SIGNIFICANCE. Abundant mixed population of lymphocytes +/− rare epithelial cells. See note. Note: The main differential diagnosis includes chronic sialadenitis and a tumor-­associated lymphoid proliferation which can be present in both benign and malignant salivary gland tumors. Clinical and radiological correlations are recommended. Example 7.6 Sample report of a poorly sampled neoplasm. Evaluation limited by scant cellularity. ATYPIA OF UNDETERMINED SIGNIFICANCE. Stromal fragments and few lesional epithelial cells. See note. Note: The cytologic findings raise the possibility of a matrix-producing salivary gland neoplasm but fall quantitatively short for a definitive diagnosis. Recommend clinical and radiologic correlations and additional sampling if clinically indicated.

4  Atypia of Undetermined Significance (AUS)

69

References 1. Brennan PA, Davies B, Poller D, et al. Fine needle aspiration cytology (FNAC) of salivary gland tumours: repeat aspiration provides further information in cases with an unclear initial cytological diagnosis. Br J Oral Maxillofac Surg. 2010;48:26–9. 2. Hughes JH, Volk EE, Wilbur DC.  Pitfalls in salivary gland fine-needle aspiration cytology: lessons from the College of American Pathologists Interlaboratory Comparison Program in Nongynecologic Cytology. Arch Pathol Lab Med. 2005;129:26–31. 3. Rossi ED, Wong LQ, Bizzarro T, et al. The impact of fine needle aspiration cytology in the management of salivary gland lesions: institutional experiences leading to a risk based classification scheme. Cancer Cytopathol. 2016;124:388–96. 4. Wei S, Layfield LJ, LiVolsi VA, et al. Reporting of fine needle aspiration (FNA) specimens of salivary gland lesions: a comprehensive review. Diagn Cytopathol. 2017;45:820–7. 5. Wong DS, Li GK. The role of fine-needle aspiration cytology in the management of parotid tumors: a critical clinical appraisal. Head Neck. 2000;22:469–73. 6. Pusztaszeri M, Rossi ED, Baloch ZW, Faquin WC. Salivary gland fine needle aspiration and introduction of the Milan reporting system. Adv Anat Pathol. 2019;26:84–92. 7. Schmidt RL, Hall BJ, Wilson AR, Layfield LJ. A systematic review and meta-analysis of the diagnostic accuracy of fine-needle aspiration cytology for parotid gland lesions. Am J Clin Pathol. 2011;136:45–59. 8. Jain E, Gupta R, Kudesia M, Singh S. Fine needle aspiration cytology in diagnosis of salivary gland lesions: a study with histological comparison. Cytojournal. 2013;10:5. 9. Mairembam P, Jay A, Beale T, Morley S, Vaz F, Kalavrezos N, Kocjan G. Salivary gland FNA cytology: role as a triage tool and an approach to pitfalls in cytomorphology. Cytopathology. 2016;27:91–6. 10. Tyagi R, Dey P.  Diagnostic problems of salivary gland tumors. Diagn Cytopathol. 2015;43:495–509. 11. Wang H, Fundakowski C, Khurana JS, Jhala N. Fine-needle aspiration biopsy of salivary gland lesions. Arch Pathol Lab Med. 2015;139:1491–7. 12. Wangsiricharoen S, Maleki Z. Risk stratification and clinical outcome in the atypia of undetermined significance category in the Milan system for reporting salivary gland cytopathology. Cancer Cytopathol. 2021;129:132–9. 13. Alruwaii F, Hang JF, Zeng BR, et al. Risk of malignancy in “atypia of undetermined significance” category of salivary gland fine-needle aspiration: a bi-institutional experience. Diagn Cytopathol. 2020;48:138–43. 14. Farahani SJ, Baloch Z. Retrospective assessment of the effectiveness of the Milan system for reporting salivary gland cytology: a systematic review and meta-analysis of published literature. Diagn Cytopathol. 2019;47:67–87. 15. Wang H, Malik A, Maleki Z, et al. “Atypical” salivary gland fine needle aspiration: risk of malignancy and interinstitutional variability. Diagn Cytopathol. 2017;45:1088–94. 16. Song SJ, Shafique K, Wong LQ, LiVolsi VA, Montone KT, Baloch Z. The utility of the Milan system as a risk stratification tool for salivary gland fine needle aspiration cytology specimens. Cytopathology. 2019;30:91–8. 17. Lubin D, Buonocore D, Wei XJ, Cohen JM, Lin O.  The Milan system at memorial Sloan Kettering: utility of the categorization system for in-house salivary gland fine-needle aspiration cytology at a comprehensive cancer center. Diagn Cytopathol. 2020;48:183–90. 18. Hollyfield JM, O'Connor SM, Maygarden SJ, Greene KG, Scanga LR, Tang S, Dodd LG, Wobker SE. Northern Italy in the American south: assessing interobserver reliability within the Milan system for reporting salivary gland cytopathology. Cancer Cytopathol. 2018;126:390–6.

70

M. P. Pusztaszeri et al.

19. Layfield LJ, Esebua M, Yang Z, Vatzmitsel M, Giorgadze T, Schmidt R. The Milan system for reporting salivary gland cytopathology: a study of inter-observer reproducibility. Diagn Cytopathol. 2019;47:765–8. 20. Van Zante A, Ha P, Pusztaszeri MP. The Milan system for reporting salivary gland cytopathology: benefits and cautions. AJSP Rev Rep. 2020;25:235–42. 21. Viswanathan K, Beg S, He B, Zhang T, Cantley R, Lubin DJ, Shi Q, Maleki Z, Asiry S, Rao R, Katabi N, Nakaguro M, Faquin WC, Sadow PM, Siddiqui MT, Scognamiglio T. NR4A3 immunostain is a highly sensitive and specific marker for acinic cell carcinoma in cytologic and surgical specimens. Am J Clin Pathol. 2021;11:aqab099. https://doi.org/10.1093/ajcp/ aqab099. 22. Chhieng DC, Cohen JM, Cangiarella JF. Fine-needle aspiration of spindle cell and mesenchymal lesions of the salivary glands. Diagn Cytopathol. 2000;23:253–9. 23. Dei P, Amir T, Al Jassar A, et al. Combined applications of fine needle aspiration cytology and flow cytometric immunophenotyping for diagnosis and classification of non Hodgkin lymphoma. Cytojournal. 2006;27(3):24–5. 24. Chhieng DC, Cangiarella JF, Cohen JM. Fine-needle aspiration cytology of lymphoproliferative lesions involving the major salivary glands. Am J Clin Pathol. 2000;113:563–71. 25. Allison DB, McCuiston AM, Kawamoto S, Eisele DW, Bishop JA, Maleki Z. Cystic salivary gland lesions: utilizing fine needle aspiration to optimize the clinical management of a broad and diverse differential diagnoses. Diagn Cytopathol. 2017;45:800–7. 26. Pantanowitz L, Thompson LDR, Rossi ED. Diagnostic approach to fine needle aspirations of cystic lesions of the salivary gland. Head Neck Pathol. 2018;12:548–61. 27. Maleki Z, Allison DB, Butcher M, Kawamoto S, Eisele DW, Pantanowitz L. Application of the Milan system for reporting salivary gland cytopathology to cystic salivary gland lesions. Cancer Cytopathol. 2021;129:214–25.

Chapter 5

Neoplasm Zubair Baloch, Guido Fadda, Christopher C. Griffith, Jerzy Klijanienko, Jeffrey F. Krane, Lester Layfield, Zahra Maleki, Ritu Nayar, and Marc P. Pusztaszeri

Z. Baloch (*) Pathology and Laboratory Medicine, University of Pennsylvania Medical Center, Philadelphia, PA, USA e-mail: [email protected] G. Fadda Department of Human Pathology of the Adulthood and of the Developing Age “Gaetano Barresi”, University of Messina Faculty of Medicine and Surgery, Messina, Italy e-mail: [email protected] C. C. Griffith Anatomic Pathology, Robert J. Tomsich Pathology and Laboratory Medicine Institute, Cleveland Clinic, Cleveland, OH, USA e-mail: [email protected] J. Klijanienko Department of Pathology, Institut Curie, Paris, France e-mail: [email protected] J. F. Krane Department of Pathology and Laboratory Medicine, David Geffen School of Medicine at UCLA, Los Angeles, CA, USA e-mail: [email protected] L. Layfield Department of Pathology and Anatomical Science, University of Missouri, Columbia, MO, USA e-mail: [email protected] Z. Maleki Division of Cytopathology, Department of Pathology, The Johns Hopkins Hospital, Baltimore, MD, USA e-mail: [email protected] R. Nayar Department of Pathology, Northwestern University Feinberg School of Medicine and Northwestern Memorial Hospital, Chicago, IL, USA e-mail: [email protected] M. P. Pusztaszeri Department of Pathology, Jewish General Hospital, Montréal, QC, Canada Department of Pathology, McGill University, Montréal, QC, Canada e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 W. C. Faquin et al. (eds.), The Milan System for Reporting Salivary Gland Cytopathology, https://doi.org/10.1007/978-3-031-26662-1_5

71

72

Z. Baloch et al.

General Background Salivary gland neoplasms are rare, more commonly occur in the parotid gland, and comprise approximately 6% of all head and neck tumors [1–5]. Up to 80% of salivary gland neoplasms (SGN) arising in the parotid gland are benign, as compared to a significantly increased incidence of malignant tumors in the other major and all minor salivary glands. Salivary gland malignancies represent 0.3% of cancers of all body sites [6, 7]. In adults, pleomorphic adenomas (PA) account for about 50% of all salivary gland neoplasms; Warthin tumor (WT) is the second most common benign tumor. Most studies cite mucoepidermoid carcinoma (MEC) as the most common malignant tumor in both children and adults; however, this can vary depending upon the anatomic site and patient cohort [3–7]. Fine-needle aspiration (FNA) is widely utilized in the management of SGNs. It can effectively distinguish with high specificity (97–98%) a non-neoplastic lesion from a neoplasm and a benign neoplasm from a malignant neoplasm [8–11]. In general, FNA can diagnose the most common benign tumors of the salivary gland, PA and WT, with high specificity (>98%). However, FNA is generally less effective in providing a specific diagnosis for other epithelial neoplasms of the salivary gland. The main cited reason for this limitation is the morphologic overlap and diversity among the many different types of salivary gland neoplasms, sometimes even within the same tumor [1, 9, 10]. Therefore, differentiation between a benign and low-­grade malignant neoplasm based on pure cellular and cytoarchitectural features can be challenging in an FNA specimen, particularly when material for ancillary studies is not available. Consequently, such specimens are commonly designated as either a “salivary gland neoplasm” or “suspicious for a neoplasm” with a broad differential diagnosis including both a cellular benign neoplasm and a low-grade carcinoma [1, 11–14]. Based on the cited literature and published meta-analyses, the FNA diagnosis of a salivary gland neoplasm can be consolidated into the following two general diagnostic categories (Table 5.1). 1. Benign Neoplasm. 2. Salivary Gland Neoplasm of Uncertain Malignant Potential (SUMP). Table 5.1  Definitions and entities included in the diagnostic category—Neoplasm [5–25] Neoplasm 1.  Benign Neoplasm FNA specimens showing cytomorphologic features of a benign epithelial or mesenchymal neoplasm (a)  Epithelial origina    1. Pleomorphic adenoma   2. Warthin tumor   3. Oncocytomab (b)  Mesenchymal origin   1.  Lipoma   2.  Schwannoma   3.  Lymphangioma   4.  Hemangioma

5 Neoplasm

73

Table 5.1 (continued) Neoplasm 2.  Salivary Gland Neoplasm of Uncertain Malignant Potential (SUMP) FNA specimens showing cytomorphologic features diagnostic of a neoplastic process, but a malignant neoplasm cannot be excluded  (a)  Cellular basaloid neoplasm  (b)  Cellular oncocytic/oncocytoid neoplasm  (c)  Cellular neoplasm with clear cell features  (d)  Cellular neoplasm with mixed features (an admixture of abovementioned features)  Due to overlapping cytomorphologic features with malignant tumors, most cases of benign neoplasm classified as basal cell adenoma, myoepitheliomas, and cystadenoma on histopathologic examination will be diagnosed as SUMP on FNA (under the subheading of cellular basaloid neoplasm and cellular neoplasm with clear cell features) (see Tables 5.2 and 5.3) b  In some cases, oncocytoma will be classified as SUMP a

General Definitions (a) Benign Neoplasm: This diagnosis is only made when an FNA specimen shows characteristic cytomorphologic features of a specific benign epithelial or mesenchymal neoplasm of the salivary gland; most commonly being PA and WT. (b) Salivary Gland Neoplasm of Uncertain Malignant Potential (SUMP): This diagnosis is reserved for FNA specimens where the cytomorphologic features are diagnostic of a neoplastic process, but the cytologic findings cannot ­effectively distinguish between a benign and malignant neoplasm. Most malignant tumors included in this diagnostic category will be low-grade carcinomas.

Benign Neoplasms The following benign neoplasms of epithelial and mesenchymal origin can be diagnosed by FNA based upon established cytomorphologic features.

Pleomorphic Adenoma (PA) [4, 9, 15–17] PA, also known as benign mixed tumor, is a benign biphasic neoplasm characterized by a variable admixture of ductal epithelial cells, myoepithelial cells, and chondromyxoid matrix (Table 5.1).

74

Z. Baloch et al.

Cytologic Criteria • Distinctive chondromyxoid matrix: Best appreciated using Romanowsky stains (Diff-Quik®, Giemsa), as a bright magenta matrix with a distinct fibrillary/feathery quality (troll-hair pattern); grey to translucent green in Papanicolaou-stained preparations (Fig. 5.1). • Biphasic cell population consisting of myoepithelial (non-luminal) and ductal (luminal) cells. • -Myoepithelial cells: Variety of shapes (polygonal, plasmacytoid, round, spindled, stellate, and clear), bland nuclear features, often the predominant cell type (Fig. 5.2). • Ductal epithelial cells: bland nuclear features, small cohesive groups recapitulating ductal structures. • The iconic PA is modestly cellular with a readily identifiable, abundant fibrillar matrix with myoepithelial cells and few bland ductal cells (Fig. 5.3). a

b

c

Fig. 5.1  Neoplasm: Benign. (a, b) Pleomorphic adenoma. Intense metachromatic fibrillary matrix with myoepithelial cells embedded within. (a—smear, Diff-Quik® stain, b—smear, Papanicolaou stain). (c) FNA of pleomorphic adenoma showing metachromatic fibrillary matrix with embedded myoepithelial cells. Notice the stroma individually surrounds each cell and the so-called “Troll-Hair” appearance of the stroma (smear, Diff-Quik® stain)

5 Neoplasm

75

a

b

c

Fig. 5.2  Neoplasm, SUMP. (a, b) FNA of pleomorphic adenoma showing a highly cellular, matrix-poor tumor with a predominance of plasmacytoid myoepithelial cells. (a—smear, Diff-­ Quik® stain, b—high power, smear, Papanicolaou stain). (c) This pleomorphic adenoma is a cellular, matrix-poor specimen with spindled and epithelioid myoepithelial cells (ThinPrep® preparation, Papanicolaou stain)

a

b

Fig. 5.3  Neoplasm: Benign. (a, b) Pleomorphic adenoma showing myoepithelial cells and very delicate, pale-staining matrix (a, b—smears, Papanicolaou stain)

76

Z. Baloch et al.

Explanatory Notes PA is one of several matrix-producing tumors that also includes adenoid cystic carcinoma (AdCC), basal cell adenoma/adenocarcinoma, and epithelial-myoepithelial carcinoma in the differential diagnosis. The hallmark and most distinguishing feature of PA is the presence of a characteristic chondromyxoid matrix, typically containing embedded myoepithelial cells. Romanowsky and Papanicolaou stains are complementary with the matrix component best appreciated using Romanowsky stains (Diff-Quik®, Giemsa) while Papanicolaou stains highlight the bland nuclear features of the ductal and myoepithelial components. FNA specimens with classic features of PA are diagnosed readily as “Neoplasm: Benign.” The biphasic nature of PA, combined with the variable ratio of epithelial and myoepithelial cells with chondromyxoid matrix, yields a spectrum of cytomorphologic patterns on aspiration, and as a result, there can be an overlap with other salivary gland neoplasms. When classic features of PA are not present or when additional “atypical” features are identified, then the FNA can be classified as “Neoplasm: SUMP.” It is most important to exclude AdCC from the differential diagnosis when faced with a matrix-producing tumor since management and prognosis will differ substantially from PA. This becomes challenging when highly cellular aspirates are obtained with scant to absent matrix (Fig. 5.2). The differential diagnosis may include the solid variant of AdCC, myoepithelioma, or other basaloid neoplasms. Such cases would be classified as “Neoplasm: SUMP.” Occasionally a case of PA can reveal matrix arranged in tubules or globules mimicking the hyaline globules of AdCC (Fig. 5.4). Another potential pitfall occurs when the matrix of PA is thin and mucoid in character (Fig. 5.5). Bland epithelial cells with abundant matrix that has a mucoid rather than fibrillar quality can mimic low-grade MEC.  This is particularly challenging in the presence of squamous or mucinous metaplastic changes; the former is more common and is seen as foci of metaplastic squamous cells with keratin (Fig. 5.6) or even clusters of anucleate squamous cells; the latter as goblet cells, often with variable amounts of a delicate mucoid matrix in the background. Ancillary studies using immunochemistry (IC) to demonstrate myoepithelial differentiation combined with PLAG1 or HMGA2 positivity can be helpful (see Chap. 8). Fig. 5.4 Neoplasm: SUMP. FNA of pleomorphic adenoma having adenoid cystic carcinoma-like areas (smears, Diff-Quik® and Papanicolaou stains)

5 Neoplasm

77

Fig. 5.5 Pleomorphic adenoma. The stroma lacks the usual fibrillary character and mimics thick mucin (smear, Diff-Quik® stain)

Fig. 5.6 Neoplasm: SUMP. FNA of pleomorphic adenoma showing squamous metaplastic features (smear, Papanicolaou stain)

When myoepithelial cells predominate, their morphology and cellularity will dictate the differential diagnosis. While myoepithelioma and cellular PA are consistently in the differential diagnosis, when myoepithelial cells have clear cytoplasm, the diagnostic considerations can include epithelial-myoepithelial carcinoma, sebaceous adenoma/carcinoma, myoepithelial carcinoma, and even metastases such as renal cell carcinoma. Myoepithelial cells with spindled morphology (Fig. 5.7) and even palisading will have bland nuclear features with a differential diagnosis that includes schwannoma, possibly hemangioma, or even nodular fasciitis, but typically sarcomas and spindle cell carcinomas are excluded based on the absence of significant nuclear pleomorphism and mitoses. While atypical myoepithelial cells can occasionally be encountered in PA (Fig. 5.8), the presence of numerous atypical cells (nuclear pleomorphism, distinct nucleoli, mitoses) and/or necrosis is a worrisome sign for malignancy. If the clinical history includes rapid enlargement of a previously stable PA or development of a tumor in a patient with a history of PA, then carcinoma ex PA enters the differential diagnosis.

78

a

Z. Baloch et al.

b

Fig. 5.7  Neoplasm: SUMP. (a, b) This aspirate of a pleomorphic adenoma has a predominance of spindled myoepithelial cells mimicking a salivary gland neoplasm of mesenchymal origin (smears, Papanicolaou stain) Fig. 5.8 Neoplasm: SUMP. This pleomorphic adenoma has nuclear atypia of the myoepithelial cells; in such cases, malignant transformation needs to be excluded (smear, Papanicolaou stain)

Warthin Tumor (WT) [4, 14, 15, 17, 18] Warthin tumor (WT) is the second most common neoplasm of the parotid gland. The majority occur in the sixth to seventh decades of life and patients usually have a significant history of smoking. Patients present with a doughy painless mass that may fluctuate in size.

Cytologic Criteria • Tripartite appearance with dirty proteinaceous background, small lymphocytes, and sheets of uniform oncocytes (Fig. 5.9). • Oncocytes: Abundant homogeneous granular cytoplasm (orange on Papanicolaou stain) with well-defined borders (Figs. 5.5, 5.6, 5.7, 5.8, 5.9 and 5.10).

5 Neoplasm

79

a

b

c

d

Fig. 5.9  Neoplasm: Benign. (a–c) FNA of Warthin tumor showing classic cytomorphologic features consisting of background lymphocytes, granular background material, and groups of oncocytic epithelial cells (smears, Diff-­Quik® stain); (d) This case of Warthin tumor only shows oncocytic cells arranged in papillary groups. Notice the lack of lymphocytes, such cases may be classified as SUMP: oncocytic/oncocytoid neoplasm (ThinPrep® preparation, Papanicolaou stain) Fig. 5.10 Neoplasm: Benign. This classic aspirate of Warthin tumor consists of oncocytic cells with abundant eosinophilic cytoplasm and welldefined borders in the background of lymphocytes (smear, Papanicolaou stain)

• Epithelial cell nuclei: Centrally placed and round, with prominent nucleolus. • Lymphocytes: Mixed population dominated by small mature-appearing cells.

80

Z. Baloch et al.

Explanatory Notes WT occurs almost exclusively in the region of the parotid gland and the tripartite appearance is essentially diagnostic. Aspirates with classic features should be diagnosed as “Neoplasm: Benign.” Distinction should be made from intra-parotid lymph nodes, lymphoepithelial sialadenitis (LESA), oncocytoma, and lymphoepithelial cyst. Intra-parotid lymph nodes and LESA lack the oncocytic epithelium and dirty cyst debris characteristic of WT. Oncocytomas consist of epithelial cells only and lack the dirty cystic background and lymphocytes of WT. The accuracy of cytologic examination for the diagnosis of Warthin tumor is 95%. Rarely, WTs undergo infarction, and the nodule will often rapidly increase in size after infarction, raising the possibility of a salivary gland malignancy and severely complicating the cytologic interpretation. Material aspirated from an infarcted WT may contain necrotic debris and atypical squamous cells. These cells may raise the differential diagnosis of squamous cell carcinoma (SCC). Distinction is aided by the recognition of scattered necrotic columnar cell ghosts and the small number of atypical squamous elements characteristic of infarcted WT.  SCC will have larger numbers of atypical squamous cells than are characteristically seen in WT as well as more severe atypia and scattered mitoses. The very rapid enlargement of a preexistent nodule also favors infarction of WT. Lymphoepithelial cysts or HIV-­ associated benign lymphoepithelial cysts are characterized by unilocular or multilocular cysts with a glandular or squamous lining and hyperplastic lymphoid tissue. Aspirates contain a mixed lymphoid population distributed among rare glandular or more commonly squamous cells. The background may be proteinaceous, but the sheets of oncocytic epithelium characteristic of WT are not seen in aspirates of lymphoepithelial cysts or HIV-associated lymphoepithelial lesions. MEC, acinic cell carcinoma (ACC), and secretory carcinoma (SC) may also have prominent associated lymphoid infiltrates and undergo cystic changes similar to WT. A Warthin-like variant and an oncocytic variant of MEC have been described [19, 20]. In most cases, the distinct epithelial component(s) of these tumors will allow distinction between these entities but given the cytomorphologic overlap, a SUMP interpretation is prudent in difficult cases where ancillary studies are not feasible (see below).

Oncocytoma [21–24] Nearly 90% of oncocytomas occur within the major salivary glands, but they comprise only 1–2% of parotid gland neoplasms. Most cases occur in the sixth to eighth decades of life. Rarely, oncocytomas may arise in a background of nodular oncocytic hyperplasia (nodular oncocytosis).

5 Neoplasm

a

c

81

b

d

Fig. 5.11  Neoplasm: Benign versus SUMP. (a–d) FNA of oncocytoma showing various patterns of a monotonous population of oncocytic cells with abundant granular cytoplasm and well-defined borders arranged in cohesive groups. In some cases, aspirates of oncocytoma may be classified as SUMP [smears, Diff-Quik® (a) and Papanicolaou stain (b–d)]

Cytologic Criteria • Irregular sheets and clusters of large polygonal cells with abundant homogeneous granular cytoplasm (Fig. 5.11). • Oncocytes have well-defined cytoplasmic borders. • Nuclei are enlarged, round, and have a distinct nucleolus. • Background is clean or contains red blood cells. • Nuclear pleomorphism and mitotic figures are absent.

Explanatory Notes The differential diagnosis of oncocytoma includes WT, nodular oncocytosis, sclerosing polycystic adenoma (SPA), oncocytic variant of MEC, ACC, and SCC. In cases without material for ancillary studies and/or lacking good clinical and radiologic information, aspirates of oncocytoma will be classified as SUMP. Aspirates of oncocytoma

82

Z. Baloch et al.

and nodular oncocytosis are virtually identical; however, oncocytoma presents clinically and radiologically as a distinct circumscribed mass while nodular oncocytosis is characterized by multiple oncocytic nodules ranging in size from 0.2 to 2.5 cm in one or both parotid glands. A large dominant mass may constitute an oncocytoma arising within nodular oncocytosis, especially if the mass is well-circumscribed and partially encapsulated. WT contains occasional groups of oncocytes but differs from oncocytoma by also having a dirty proteinaceous background, and a mixed lymphoid population. ACC and SC contain a mixture of polygonal cells with delicate vacuolated cytoplasm. In contrast, oncocytomas lack the cytoplasmic vacuoles of ACC and SC; a Romanowsky stain can be used to highlight this subtle distinction. ACC and SC can be readily distinguished from oncocytomas by ancillary studies, with the use of a panel of immunostains that includes the following antibodies: SOX-10, DOG-1, Mammaglobin, pan-TRK, NR4A3; NR4A3, DOG-1, and SOX10 are positive in ACC but negative in oncocytomas; mammaglobin and pan-TRK are positive in SC (see Chap. 8). Oncocytic carcinoma of the salivary gland is extremely rare, and some cases previously classified as oncocytic carcinoma have been subtypes of salivary duct carcinoma or oncocytic variants of other high-grade carcinomas. Clinical and radiologic findings may be useful to provide evidence of an infiltrative neoplasm. Nuclear atypia, mitotic activity, or necrosis are not features of oncocytoma, and when present are suggestive of malignancy.

Tumors of Mesenchymal Origin Pure mesenchymal lesions of the salivary glands are uncommon with mesenchymal components being more commonly expressed as a constituent of pleomorphic adenomas, malignant mixed tumors, and sarcomatoid variants of salivary duct carcinoma. Most mesenchymal neoplasms are benign, but malignant mesenchymal neoplasms have been reported [25]. Lipomas represent over 50% of all benign mesenchymal lesions of the salivary glands with hemangiomas, schwannomas, neurofibromas, lymphangiomas, and solitary fibrous tumors composing the remainder [25]. Benign mesenchymal neoplasms appear equally in men and women and are most frequent in middle-aged individuals. The majority occurs within the parotid gland. Sarcomas make up approximately 7.5% of all mesenchymal neoplasms of the salivary glands with most being malignant peripheral nerve sheath tumors, malignant solitary fibrous tumors, synovial sarcomas, leiomyosarcomas, liposarcomas, and desmoplastic small round cell tumors [26, 27].

Lipoma [25, 28–30] Lipomas are uncommon neoplasms of the salivary gland. They represent approximately 0.5% of salivary gland tumors and three-quarters of them arise in the parotid gland. They most often present as palpable soft nodules. The cytologic appearance

5 Neoplasm

83

of variants of lipoma has been reported with spindle cell lipoma being the most common and occurring most often in the posterior neck [25, 30]

Cytologic Criteria (Classic Lipoma) • Lace-like sheets and clusters of very low nuclear cytoplasmic (N:C) ratio cells with optically clear cytoplasm (Fig. 5.12). • Individual cells are of large size with a single large clear vacuole occupying the entire cytoplasmic volume (Fig. 5.13). • Nuclei are small hyperchromatic and frequently displaced to the margin of the cell. • The background contains many droplets of lipid (best seen on Romanowsky stains). Fig. 5.12 Neoplasm: Benign. FNA of lipoma showing a lace-like group of mature adipocytes with abundant clear cytoplasm and small dark nuclei, (smear, Diff-Quik® stain)

Fig. 5.13 Neoplasm: Benign. This FNA contains a group of adipocytes from a lipoma characterized by large cells with abundant clear cytoplasm. The small dark nuclei are often displaced to the edge of the cell (smear, Diff-Quik® stain)

84

Z. Baloch et al.

Explanatory Notes It may be difficult to distinguish fatty change of the salivary gland from a lipoma by cytologic evaluation alone. While the fatty change will contain normal acinar and ductal elements, lipomas are composed purely of adipose tissue, although rare examples of lipomas with entrapped normal serous acini have been reported. Clinical correlation is needed. Spindle cell lipomas can closely resemble lipomatous pleomorphic adenomas [29]. Aspirates of these neoplasms demonstrate a mixture of mature adipose tissue and bland spindle cells in a myxoid background. No nuclear atypia is observed, and the spindle cell component is positive for CD34 while the fatty component is positive for S-100 protein. Chondroid lipomas may be confused cytologically with pleomorphic adenomas [29–31]. Conversely, PA can sometimes undergo extensive adipose metaplasia; depending on the sampling, it may be difficult to distinguish these from lipomas on cytology [32, 33].

Schwannoma [34–37] Schwannomas represent one of the two most common benign mesenchymal neoplasms of the salivary glands. FNA of schwannoma is often associated with pain of a radicular type.

Cytologic Criteria • • • • •

Scant to moderately cellular aspirates. Spindle-shaped cells with wispy bipolar cytoplasmic processes (Fig. 5.14). Cells form cohesive groups and clusters sometimes with palisading. Cytoplasm is pale, ill-defined, and often fibrillary in appearance. Nuclei are small, dark, and bland with elongated/spindled form; nuclei may be bent, curved, or S-shaped. • Occasional large but bland nuclei can be seen (ancient change). • Nucleoli are small or absent. • Background occasionally has a myxoid appearance.

Explanatory Notes The most common differential diagnosis is the distinction of schwannomas from pleomorphic adenomas and myoepitheliomas [38]. The abundance of bland spindle cells within these three neoplasms may lead to diagnostic confusion. Moreover,

5 Neoplasm

85

Fig. 5.14 Neoplasm: Benign. This aspirate of schwannoma shows a group of bland spindle cells with wispy cytoplasm. The cytoplasmic borders are indistinct. Nuclei are spindle-shaped and display bends or curves. Ancillary studies are needed for definitive classification (smear, Diff-­Quik® stain)

glandular forms of schwannoma exist heightening the problems of differential diagnosis. Melanocytic schwannomas have been reported in the submandibular gland requiring separation from metastatic melanoma. Differential diagnosis is aided by ancillary studies. Schwannomas are strongly and diffusely positive for S-100 protein and SOX10, akin to myoepithelial cells, but are negative for keratins, MelanA (except melanocytic schwannomas) [39], PLAG-1, and other more specific myoepithelial markers such as p63, p40, SMA, or calponin. Other differential diagnostic considerations include sarcomas which are very rare in the submandibular glands, but malignant peripheral nerve sheath tumors have been reported. The possibility of sarcoma should be considered when aspirates are highly cellular and display significant nuclear atypia often with scattered mitoses and apoptosis.

Lymphangioma [25, 40, 41] Lymphangiomas occurring in the salivary glands are rare with most occurring in children. They present as slowly growing fluctuant masses. Most arise in the parotid gland.

Cytologic Criteria • • • •

Hypocellular smears with a watery background. Occasional red blood cells present. Scattered mature-appearing lymphocytes. Rare background clusters of non-neoplastic salivary gland acinar tissue may be present.

86

Z. Baloch et al.

Explanatory Notes Aspirates obtained from salivary gland lymphangiomas are frequently non-diagnostic and are composed purely of watery fluid containing scattered mature-appearing lymphocytes and occasional groups of non-neoplastic acinar tissue. Endothelial cells are generally absent from aspirated material. The diagnosis usually requires careful clinical and radiographic correlation. The differential diagnosis of a cystic parotid gland lesion includes Warthin tumor, lymphoma, benign lymphoepithelial lesion, brachial cleft cyst, and cystic low-grade mucoepidermoid carcinoma. Careful attention to minor cell populations is important in excluding these lesions.

Hemangioma [25, 42–44] Hemangiomas are one of the three most common mesenchymal neoplasms of the salivary gland with most examples arising within the parotid gland. Most hemangiomas occur in the first decade of life, especially the first year. The latter neoplasms are of the so-called “juvenile type” and may spontaneously regress. Reported juvenile hemangiomas can be highly cellular.

Cytologic Criteria • Aspirate dominated by red blood cells. • Rare groups of bland spindle-shaped or polygonal-shaped endothelial cells which may form elongated cord-like structures (Fig. 5.15). • Individual endothelial cells are oval to spindle-shaped. Fig. 5.15 Neoplasm: Benign. Smears obtained from hemangiomas are characteristically bloody but may contain small aggregates of bland spindle-shaped endothelial cells. Rarely, sheet-like structures composed of oval or spindle-shaped endothelial cells will be present. Clinical and radiologic correlation is needed in the evaluation (smear, Diff-Quik® stain)

5 Neoplasm

87

• Nuclei are small, bland, and oval shaped. • Nucleoli generally absent. • Scattered histiocytes may be present in the background.

Explanatory Notes Aspirates of hemangioma are dominated by blood to an extreme degree making recognition of the endothelial cell component difficult. This component is often overlooked. In reported cases of hemangiomas of the parotid gland, authors have been unable to make the diagnosis cytologically due to the absence of a recognizable endothelial component. A careful search for groups of bland spindle or oval-­ shaped endothelial cells is required to establish the diagnosis. In some cases, the aspirate will be classified as Non-Diagnostic.

Rare Salivary Gland Mesenchymal Tumors Rare reports of other mesenchymal lesions appear within the literature generally as isolated case reports. These mesenchymal lesions have usually displayed similar morphology as described in other body sites. In most cases, aspirates of these uncommon lesions will be classified as SUMP or AUS.

Explanatory Notes Examples of nodular fasciitis, inflammatory pseudotumor, and solitary fibrous tumor have been reported [45, 46]. While generally benign in behavior, some have been capable of recurrence and even metastasis. Ancillary studies including immunocytochemistry and molecular analysis are helpful in their recognition. The rare examples of malignant mesenchymal tumors are discussed in Chap. 7.

 alivary Gland Neoplasm of Uncertain Malignant S Potential (SUMP) SUMP is a category reserved for FNA specimens which are diagnostic of a neoplasm; however, classification of a specific entity cannot be made. This category should be reserved for aspirates that are clearly neoplastic but where the differential includes both benign and malignant entities. Alternatively, some aspirates may have

88

Z. Baloch et al.

the overall appearance of a benign neoplasm such as PA, but a readily identifiable atypical cellular component or clinical/radiologic finding may raise concern for a malignant process, and SUMP would be an appropriate categorization. This is a heterogeneous category and includes three main cytomorphologic patterns useful for differential diagnostic considerations. The most common patterns of SUMP are “cellular basaloid neoplasms” and “oncocytic/oncocytoid neoplasms” followed by “clear cell neoplasms.” Each pattern can be further subdivided based on additional background and cellular features as discussed below. Some cases may demonstrate an admixture of cellular patterns, such as the presence of oncocytic/oncocytoid and clear cells. However, one must be cognizant of the fact that there are many other potential cytomorphologic scenarios of cases classified as SUMP, which are beyond the scope of this monograph [24, 47–49]. Studies have shown variable rates of malignancy between the basaloid and oncocytoid subgroups of SUMP but without consistent differences in terms of which group has the greatest risk of malignancy [12, 24, 50, 51]. A majority of salivary gland aspirates classified as SUMP will be benign neoplasms or low-grade carcinomas on histologic follow-up.

Cellular Basaloid Neoplasm [24, 51–54] The SUMP category with a subcategorization of “cellular basaloid neoplasm” should be reserved only for tumors in which a specific diagnosis is not possible, and the differential diagnosis includes both benign and malignant tumors. Cellular basaloid neoplasms are characterized by a predominant population of cells with scant cytoplasm conferring an immature (“basaloid”) cytomorphology. Basaloid salivary gland neoplasms commonly produce an extracellular matrix with variable quantity and quality (i.e., fibrillary versus non-fibrillary) and some have suggested further subcategorization of cellular basaloid neoplasms based on these matrix characteristics (Fig. 5.16). Basaloid neoplasms categorized as SUMP will generally lack the characteristic abundant chondromyxoid fibrillary stroma of PA precluding a specific diagnosis of a benign neoplasm. In some cases, the amount of matrix may be too limited or completely absent to be confident of the diagnosis of PA. Alternatively, some cases may either demonstrate hyalinized stroma or lack an easily identifiable chondromyxoid matrix.

Cytologic Criteria • Typically, cellular and composed of cells with scant cytoplasm and high nuclear-­ to-­cytoplasmic ratios. • Matrix may be present or absent but characteristic fibrillary matrix of PA is not identified.

5 Neoplasm

89

a

c

b

d

Fig. 5.16  Neoplasm: SUMP. (a, b) FNA of cellular basaloid neoplasms showing a predominant population of cells with scant cytoplasm with hyaline stroma (smears, Diff-Quik® stain); (c, d) This FNA contains a monotonous population of basaloid cells arranged in cohesive groups with scant hyaline stroma (smears, Papanicolaou stain)

• Evidence of malignancy such as increased mitotic activity, tumor necrosis, and/ or high-grade nuclear atypia is absent. • Matrix patterns include absent to scant fibrillary matrix, hyaline stroma, mixed/ other stroma, and minimal to no matrix (i.e., hyaline, mixed, or other) which may inform differential diagnostic considerations as in Table 5.2.

Explanatory Notes The morphologic overlap in cytologic samples of basaloid salivary gland neoplasms lacking characteristic features of PA is considerable and such aspirates are best classified as SUMP. The diagnosis of some cellular PA and most other low-grade basaloid neoplasms (including benign neoplasms and low-grade malignancies) often requires histologic examination. Cellular basaloid neoplasms categorized as SUMP have a nearly 100% rate of being neoplastic on histologic follow-up with a risk of malignancy that has been reported from 22 to 47.8% in various retrospective and

90

Z. Baloch et al.

Table 5.2  Morphologic scenarios and differential diagnosis of cases classified as “basaloid neoplasm” [12, 24, 52, 55–57] Cytomorphologic featuresa 1. Cellular basaloid neoplasm with scant fibrillary matrix

Differential diagnosisb • Cellular pleomorphic adenoma • Epithelial-myoepithelial carcinoma • Basal cell adenoma/adenocarcinoma • Carcinoma ex pleomorphic adenoma 2. Cellular basaloid neoplasm with hyaline • Basal cell adenoma/adenocarcinoma stroma • Adenoid cystic carcinoma • Epithelial-myoepithelial carcinoma • Carcinoma ex pleomorphic adenoma 3. Cellular basaloid neoplasm with mixed/other • Adenoid cystic carcinoma matrix • Polymorphous adenocarcinomac • Cellular pleomorphic adenoma • Carcinoma ex pleomorphic adenoma 4. Cellular basaloid neoplasm with minimal to • Cellular pleomorphic adenoma no matrix • Canalicular adenoma • Myoepithelioma • Myoepithelial carcinoma • Adenoid cystic carcinoma • Carcinoma ex pleomorphic adenoma  Highly dependent on cytologic preparations  Provided as a guide—may or may not be included in the diagnostic report c  Rarely encountered in major salivary glands a

b

prospective studies [9, 24, 51, 56]. On histologic follow-up, the most common benign neoplasm is PA (particularly cellular PA) but basal cell adenoma and myoepithelioma are also frequently categorized as SUMP. The most common malignancies are AdCC and epithelial-myoepithelial carcinoma with basal cell adenocarcinoma and myoepithelial carcinoma being less common [56]. Rarely, tumor types more frequently thought of as having abundant oncocytoid cytoplasm, such as MEC, may also be more basaloid on cytologic samples. A deep cutaneous neoplasm can occasionally mimic a superficial salivary gland tumor thus cutaneous basal cell carcinoma, pilomatrixoma, basaloid squamous cell carcinoma, and highgrade neuroendocrine carcinoma (Merkel cell carcinoma or distant metastasis) should also be considered in the setting of a cellular basaloid neoplasm. The differential diagnosis of cellular basaloid neoplasms also includes benign neoplasms with a malignant counterpart that cannot be definitively diagnosed on FNA since the cytomorphologic features are nearly identical. These include basal cell adenoma and basal cell adenocarcinoma, myoepithelioma and myoepithelial carcinoma, and PA and epithelial-myoepithelial carcinoma. Histologic evaluation to exclude invasive growth and lymphovascular/perineural invasion is often needed to definitively distinguish between these benign and malignant tumors. Identifying differences in matrix quality and quantity is helpful in the differential diagnosis of cellular basaloid neoplasms and therefore the absence of any matrix limits one’s ability to narrow differential diagnostic considerations. Romanowsky

5 Neoplasm

91

stains on air-dried smears are helpful for visualizing and evaluating matrix characteristics. In general, the presence of fibrillary matrix, even when scant, is suggestive of a benign neoplasm with most being PA on follow-up. In these cases, the limited amount of the matrix or other atypical features may preclude a confident cytologic diagnosis of PA; however, the diagnosis of PA could be favored in some cases, especially if clinical and radiologic features support this interpretation. Various studies have suggested a lower risk of malignancy in aspirates with fibrillary matrix, but most fail to show a statistically significant difference in this risk compared to aspirates with other matrix features [24, 51]. In contrast, other types of matrix raise differential diagnostic considerations other than PA including malignant entities, and thus have a slightly higher rate of malignancy. Hyaline matrix spheres or globules are often considered characteristic of AdCC but can be seen in a variety of other benign or malignant, salivary or non-salivary neoplasms [57, 58]. Ancillary testing with IC and/or molecular techniques can be helpful in some cases to achieve a specific diagnosis in the setting of a cellular basaloid neoplasm (see Chap. 8). In particular, several entities in the differential diagnosis of a cellular basaloid neoplasm have defining recurrent translocations including PLAG1 or HMGA2 rearrangements in PA and MYB rearrangements in AdCC. By definition, categorization of an aspirate as SUMP is a result of the inability to provide a specific diagnosis for a particular tumor. The differential diagnosis typically includes tumors with a range of clinical behavior including benign tumors (cellular PA, basal cell adenoma, canalicular adenoma, myoepithelioma), low-grade malignancies (basal cell adenocarcinoma, epithelial-myoepithelial carcinoma, polymorphous adenocarcinoma) and intermediate to high-grade malignancies (AdCC) (Figs. 5.17 and 5.18). Surgical excision is generally indicated for a tumor categorized as SUMP and enables definitive classification. If the extent of surgery is a

b

Fig. 5.17  Neoplasm: SUMP. (a) This FNA shows basaloid tumor cells associated with well-­ demarcated hyaline stroma. Depending upon the cellularity and cytomorphologic features combined with clinical findings, the diagnosis of cases such as this can range from SUMP-Basaloid neoplasm to suspicious for adenoid cystic carcinoma (Smear, Papanicolaou stain). (b) This aspirate shows basaloid tumor cells arranged in a three-dimensional cohesive group with nuclear crowding and minimal to no matrix (ThinPrep® preparation, Papanicolaou stain)

92

Z. Baloch et al.

a

b

c Fig. 5.18  Neoplasm: SUMP. (a–c) In this FNA of a cellular basaloid neoplasm there is a cohesive group of basaloid cells with no matrix. On histologic follow-up this case was diagnosed as solid variant of adenoid cystic carcinoma. (a, b - Smears, Papanicolaou stain, c - ThinPrep®preparation)

critical, a repeat aspiration may be helpful in some cases, particularly if additional ancillary studies such as IC or molecular studies offer the potential for a more specific diagnosis. Alternatively, frozen section examination at the time of surgery can frequently provide further useful information to guide the overall extent of surgery [8, 59, 60].

Cellular Oncocytic/Oncocytoid Neoplasm [9, 22–24, 53, 56] Neoplasms showing oncocytic or oncocytic-like (i.e., oncocytoid) features are common in the salivary glands. While oncocytic features are the main characteristic of some SGNs such as WT and oncocytoma, it can also be an accompanying finding in several other salivary gland neoplasms including PA, myoepithelioma, and MEC. In addition, some non-oncocytic neoplasms like ACC, SC, and metastatic renal cell carcinoma can cytomorphologically mimic true oncocytic tumors. In most of these cases, it is possible to give an accurate diagnosis if characteristic cytomorphologic features are present (see separate related chapters), and/or if the diagnostic pitfalls are carefully assessed and if ancillary tests are performed. However, there remains

5 Neoplasm

93

a subset of salivary gland neoplasms with oncocytic features which cannot be confidently subtyped, and it is appropriate to classify these as SUMP-oncocytic/oncocytoid neoplasms.

Cytologic Criteria (Table 5.3) Aspirates of salivary neoplasms classified as “Neoplasm: SUMP” with oncocytic/ oncocytoid features have the following characteristics: • Cellular aspirate. • Neoplastic cells with oncocytic or oncocytoid features which cannot be classified further (Figs. 5.19 and 5.20). –– Moderate amounts of oncocytic granular cytoplasm. –– Round to oval nucleus +/− distinct nucleolus. • Oncocytic/oncocytoid neoplastic cells lack high-grade cellular features such as marked nuclear atypia, high mitotic activity, and necrosis. Table 5.3  Morphologic scenarios and differential diagnosis of cases classified as “SUMP: cellular oncocytic/oncocytoid” [9, 23, 24, 63] Cytomorphologic features Cellular oncocytic/oncocytoid neoplasm with 1. Cystic background (histiocytes, proteinaceous debris, +/− inflammatory cells)

2. Mucinous background

3. Granular (usually coarse)/ vacuolated cytoplasm 4. Appreciable focal nuclear atypiad

Differential diagnosis

• Warthin tumora • Sclerosing polycystic adenoma • Cystadenoma, oncocytic • Acinic cell carcinoma • Mucoepidermoid carcinoma, oncocytic variant • Mucoepidermoid carcinoma, oncocytic variant • Rare case of Warthin tumor with focal mucinous metaplastic changeb • Acinic cell carcinomac • Secretory carcinomac • Metastatic renal cell carcinoma • Salivary duct carcinomac • High-grade mucoepidermoid carcinoma • Oncocytic carcinoma • High-grade, oncocytic epithelial-­myoepithelial carcinoma • Metastatic carcinoma

 Tumor usually shows lymphocytes in the background and is intimately associated with tumor cell groups b  Diagnosis requires exclusion of oncocytic MEC c  Rare cases may show prominent oncocytic change d  Cases with multifocal or diffuse presence of nuclear atypia should be classified as suspicious for carcinoma or malignant a

94

a

Z. Baloch et al.

b

Fig. 5.19  Neoplasm: SUMP. (a, b) This aspirate shows neoplastic cells with oncocytic/oncocytoid cytoplasm arranged in cohesive cluster with associated crystalline material. The cells demonstrate eccentrically placed nuclei (plasmacytoid appearance). On histologic follow-up, this case was diagnosed as myoepithelioma (smear, Diff-Quik® stain)

Fig. 5.20 Neoplasm: SUMP. FNA of cellular oncocytic/oncocytoid neoplasm showing loose groups and dispersed neoplastic cells with bland oncocytic features. On histologic follow-up, this case was classified as acinic cell carcinoma (smear, Diff-Quik® stain)

Explanatory Notes The SUMP-oncocytic/oncocytoid subcategory should be reserved for cases which include both primary salivary gland oncocytic neoplasms and their mimics, mainly low-grade carcinomas, in the differential diagnosis. In addition, these are tumors where a definitive interpretation is not possible. The most common oncocytic neoplasm of salivary glands is WT which is accurately diagnosed by cytology in most cases. However, in a small subset of cases, all of the usual diagnostic features of WT may not be readily discernible, or the tumor may show mucinous or squamous

5 Neoplasm

95

metaplasia leading to diagnostic difficulties. In some cases, oncocytes are present in a cystic background without accompanying lymphocytes, even though a lymphocyte-­ poor WT is favored. An aspirate of a salivary gland neoplasm showing oncocytic neoplastic cells with focal intra-cytoplasmic mucin in a mucinous background should raise concern for an oncocytic MEC, and depending upon the overall cytomorphologic features present, a diagnosis of “SUMP-oncocytic/oncocytoid neoplasm” would be warranted. Clinical and radiologic findings can be useful to provide evidence of an infiltrative cancer. Nuclear atypia, mitotic activity, or necrosis when present are suggestive of malignancy. A salivary gland neoplasm with oncocytic features can be difficult to differentiate from ACC due to the shared features of low nuclear grade and abundant oncocytic cytoplasm (Fig.  5.20). Aspirates of ACC usually display cells with delicate vacuolated or pale cytoplasm, indistinct cytoplasmic borders, and nuclei that are sometimes larger than those in oncocytes. The background usually contains many stripped nuclei (especially in smear preparations), and some cases may contain background lymphocytes. Ancillary studies can be very helpful to arrive at a definitive diagnosis of ACC (i.e., Malignant); however, in cases with limited cellularity and/or lacking material for ancillary studies, a diagnosis of SUMP-oncocytic/oncocytoid neoplasm can be made with a comment that ACC is in the differential diagnosis. SC can exhibit oncocytic features in FNA specimens. SC usually consists of an admixture of cells with granular and eosinophilic cytoplasm, cells with multi-­ vacuolated cytoplasm, and some cells with intracellular mucin. These features are frequently misinterpreted as ACC or oncocytic MEC. PAs and myoepitheliomas may show oncocytic metaplasia, but generally other specific characteristic features of these neoplasms such as the presence of fibrillary metachromatic matrix lead to the correct interpretation. Rarely, metastatic carcinomas with eosinophilic cytoplasm can also mimic primary oncocytic neoplasms. These can be readily distinguished using ancillary studies, especially IC, correlated with clinical findings. Sclerosing polycystic adenoma (SPA) is a rare benign neoplasm of the salivary gland, usually seen in the parotid gland and resembles fibrocystic disease of the breast clinically and morphologically. On cytology, SPA shows a varied morphological picture and is frequently misdiagnosed. There are only a few case reports in the literature which were misdiagnosed as WT, oncocytoma, PA, cystadenoma/cystadenocarcinoma, ACC, low-grade MEC, and salivary duct carcinoma [61, 62].

Cellular Neoplasm with Clear Cell Features [64–68] SGNs with clear cell features are uncommon. These tumors include a broad range of benign and malignant neoplasms with overlapping cytomorphologic features. Lesional cells with clear or vacuolated cytoplasm are the key diagnostic feature (Figs.  5.21, 5.22, 5.23, 5.24 and 5.25). Because they are uncommon, clear cell

96

a

Z. Baloch et al.

b

Fig. 5.21  Neoplasm: SUMP. (a, b) FNA of a cellular neoplasm with clear cell to oncocytoid features showing sheets of epithelial cells with finely vacuolated cytoplasm. Nuclei are enlarged but retain smooth nuclear membranes. The histologic follow-up of this case was acinic cell carcinoma (smear, Papanicolaou stain)

a

b

Fig. 5.22  Neoplasm: SUMP. (a, b) This aspirate contains loosely cohesive groups of cells with indistinct finely vacuolated, pale-staining cytoplasm imparting a clear quality. Nuclei are small to medium-sized nuclei with even chromatin. No nuclear pleomorphism is seen. The histologic follow-­up was acinic cell carcinoma [(smear, (a) Papanicolaou stain and (b) Diff-Quik stain]

neoplasms represent a minor subgroup of the SUMP category. Since most of the neoplasms predicted to be placed into this subcategory are low-grade malignancies, the ROM may be at the higher end of the ROM for SUMP (20–40%); however, the risk of high-grade malignancy for this diagnostic subcategory is expected to be low.

5 Neoplasm

a

97

b

Fig. 5.23  Neoplasm: SUMP. (a, b) This FNA shows a neoplastic proliferation of cells with delicate pale cytoplasm containing variably sized clear vacuoles and round nuclei with inconspicuous nucleoli and minimal nuclear pleomorphism [smears, Diff-Quik® (a) and Papanicolaou stain (b)]

a

b

Fig. 5.24  Neoplasm: SUMP. (a, b) FNA of a neoplasm with variably oncocytoid to clear cell features. A monotonous population of neoplastic cells arranged in cohesive groups with finely granular cytoplasm. The background shows thin mucin and clear histiocytic-type cells that raise a differential diagnosis of mucoepidermoid carcinoma. (a) smear Diff-Quik® stain. (b) smear, Papanicolaou stain) Fig. 5.25 Neoplasm: SUMP. FNA of a neoplasm with scattered large cells with finely vacuolated clear to pale cytoplasm and occasional smaller epithelial cells (smear, Papanicolaou stain)

98

Z. Baloch et al.

Cytologic Criteria • Cellular aspirate diagnostic of a neoplasm but characteristic cytomorphologic features of a specific tumor entity (see specific tumors in other Chapters) are absent. • Neoplastic cells with clear cell features: clear, foamy, granular, or vacuolated cytoplasm, or any combination thereof, but features are not characteristic of true oncocytes (Figs. 5.21, 5.22, 5.23, 5.24 and 5.25). • Nuclear cytologic grade is low to moderate. • Absence of high-grade features (e.g., necrosis, marked nuclear atypia, mitotic activity). • Ancillary studies, if performed, do not allow classification into another diagnostic category (e.g., benign or malignant).

Explanatory Notes Salivary gland aspirates comprised of a prominent population of cells with clear cytoplasm should be evaluated with caution since the differential diagnosis is broad and the distinction between the various clear cell tumors can be challenging. Cytoplasmic “clearing” represents a non-specific change resulting from one or a combination of cellular alterations including (1) intra-cytoplasmic lipid, mucin, or glycogen; (2) intracellular edema; (3) paucity of intracellular organelles. Depending upon the type of cytologic preparation and the type of neoplasm, the cytoplasm of the neoplastic cells can range from coarsely granular to foamy (Figs. 5.21 and 5.23), vacuolated, optically clear, or a combination thereof. The clear cytoplasmic change is best appreciated on Papanicolaou and H&E (cell block) staining, while MGG (Romanowsky) staining will usually impart a non-specific pale blue hue to the cytoplasm, with the exception of mucin or lipid. There exists a significant overlap between tumors with clear cell features. In the evaluation of an aspirate containing clear cell features and for which definitive tumor classification cannot be made by standard cytologic evaluation, additional material should be obtained, if possible, for ancillary marker studies (see Chap. 8). Defining the nature of the clear cell change (e.g., lipid, mucin, or glycogen) using histochemical stains (PAS and PASDiastase, Mucicarmine) can be helpful to limit the differential diagnosis of clear cell neoplasms followed by a focused IC panel. If a specific tumor classification is not possible based upon quantitative or qualitative cytomorphologic features and any ancillary studies performed, the aspirate can be classified as SUMP, Suspicious for Malignancy, or Malignant depending upon the degree of nuclear atypia, differential diagnostic considerations, and the estimated ROM.

5 Neoplasm

99

Clinical Management FNA cases classified as “Neoplasm: Benign” should have MRI or CT studies performed to assess the extent of the lesion prior to complete excision of the benign lesion with nerve preservation (see Chap. 9). A subset of patients who are not surgical candidates or who are unable to accept the risk of potential nerve injury might be clinically followed without surgical management, particularly for tumors like WT where there is negligible risk of malignant transformation. The management of FNA cases classified as “Neoplasm: SUMP” is similar, but includes a greater degree of clinical decision-making. Preoperative imaging by MRI or CT should be performed on this group of patients to evaluate the extent of the tumor as well as assess the neck. Nerve-sparing surgical resection is indicated unless the patient is not a surgical candidate. Intraoperative frozen section in cases diagnosed as SUMP may better define the histologic classification and margin status, and help determine if a neck dissection is indicated.

Sample Reports Example 5.1 Satisfactory for evaluation. NEOPLASM: BENIGN. Pleomorphic Adenoma. Example 5.2 Satisfactory for evaluation. NEOPLASM—SALIVARY GLAND NEOPLASM OF UNCERTAIN MALIGNANT POTENTIAL (SUMP). Cellular Basaloid Neoplasm. See note. Note: The specimen shows a monomorphic population of basaloid cells with minimal nuclear atypia associated with the fibrillary matrix. No mitoses or tumor necrosis is seen. The findings are suggestive of a cellular pleomorphic adenoma; however, other matrix-producing basaloid tumors such as basal cell adenoma, basal cell adenocarcinoma, and epithelial-myoepithelial carcinoma cannot be completely excluded. Example 5.3 Evaluation limited by scant cellularity. NEOPLASM—SALIVARY GLAND NEOPLASM OF UNCERTAIN MALIGNANT POTENTIAL (SUMP). Cellular Neoplasm with Oncocytoid features. See note.

100

Z. Baloch et al.

Note: The specimen shows a monomorphic population of oncocytoid cells with minimal nuclear atypia in a mucinous background. No mitoses or tumor necrosis is seen. The evaluation is limited by scant cellularity precluding the use of ancillary studies and a definitive diagnosis. The differential diagnosis includes Warthin tumor with focal mucinous metaplastic change, sclerosing polycystic adenoma, oncocytic variant of mucoepidermoid carcinoma, and secretory carcinoma. A repeat FNA could be useful to clarify the diagnosis if clinically indicated. Example 5.4 Satisfactory for evaluation. NEOPLASM—SALIVARY GLAND NEOPLASM OF UNCERTAIN MALIGNANT POTENTIAL (SUMP). Cellular Neoplasm with Clear Cell Features. See note. Note: The specimen shows a low-grade biphasic neoplasm with clear cell features. The differential diagnosis includes pleomorphic adenoma and myoepithelioma; however, epithelial-myoepithelial carcinoma cannot be completely excluded.

References 1. Layfield LJ, Glasgow BJ. Diagnosis of salivary gland tumors by fine-needle aspiration cytology: a review of clinical utility and pitfalls. Diagn Cytopathol. 1991;7(3):267–72. 2. Schmidt RL, Hall BJ, Wilson AR, Layfield LJ. A systematic review and meta-analysis of the diagnostic accuracy of fine-needle aspiration cytology for parotid gland lesions. Am J Clin Pathol. 2011;136(1):45–59. 3. Ellis GL, Auclair PL. In: Armed Forces Institute of Pathology, editor. Tumors of the salivary glands. Northfield: American Registry of Pathology; 2008. 4. IARC. WHO classification of head and neck tumors. 4th ed. Lyon: IARC; 2017. 5. Slootweg PJ, El-Naggar AK. World Health Organization 4th edition of head and neck tumor classification: insight into the consequential modifications. Virchows Arch. 2018;472(3):311–3. 6. Ferlay J, Colombet M, Soerjomataram I, Parkin DM, Pineros M, Znaor A, Bray F.  Cancer statistics for the year 2020: an overview. Int J Cancer. 2021;2021:33588. 7. Sung H, Ferlay J, Siegel RL, Laversanne M, Soerjomataram I, Jemal A, Bray F. Global cancer statistics 2020: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2021;71(3):209–49. 8. Seethala RR, LiVolsi VA, Baloch ZW. Relative accuracy of fine-needle aspiration and frozen section in the diagnosis of lesions of the parotid gland. Head Neck. 2005;27(3):217–23. 9. Griffith CC, Pai RK, Schneider F, Duvvuri U, Ferris RL, Johnson JT, Seethala RR. Salivary gland tumor fine-needle aspiration cytology: a proposal for a risk stratification classification. Am J Clin Pathol. 2015;143(6):839–53. 10. Rossi ED, Faquin WC. Experience from the world: the accuracy of salivary gland FNA and reliability of the Milan system for reporting salivary gland cytopathology in a large study from The Netherlands. Cancer Cytopathol. 2021;129(9):675–6. 11. Farahani SJ, Baloch Z. Retrospective assessment of the effectiveness of the Milan system for reporting salivary gland cytology: a systematic review and meta-analysis of published literature. Diagn Cytopathol. 2019;47(2):67–87.

5 Neoplasm

101

12. Cantley RL.  Fine-needle aspiration cytology of cellular basaloid neoplasms of the salivary gland. Arch Pathol Lab Med. 2019;143(11):1338–45. 13. Cardillo MR. Salivary gland masses: the diagnostic value of fine-needle aspiration cytology. Arch Anat Cytol Pathol. 1990;38(1–2):26–32. 14. Colella G, Cannavale R, Flamminio F, Foschini MP. Fine-needle aspiration cytology of salivary gland lesions: a systematic review. J Oral Maxillofac Surg. 2010;68(9):2146–53. 15. Ahn S, Kim Y, Oh YL.  Fine needle aspiration cytology of benign salivary gland tumors with myoepithelial cell participation: an institutional experience of 575 cases. Acta Cytol. 2013;57(6):567–74. 16. Das DK, Anim JT. Pleomorphic adenoma of salivary gland: to what extent does fine needle aspiration cytology reflect histopathological features? Cytopathology. 2005;16(2):65–70. 17. Allison DB, Smith AP, An D, Miller JA, Shafique K, Song S, Viswanathan K, Eykman E, Rao RA, Wiles A, Barkan GA, Nayar R, Fadda G, Powers CN, Rossi ED, Siddiqui MT, Ali SZ, Kholova I, Layfield LJ, Field A, Baloch Z, Maleki Z. Assessing the diagnostic accuracy for pleomorphic adenoma and Warthin tumor by employing the Milan system for reporting salivary gland cytopathology: an international, multi-institutional study. Cancer Cytopathol. 2021;129(1):43–52. 18. Rossi ED, Faquin WC, Baloch Z, Barkan GA, Foschini MP, Pusztaszeri M, Vielh P, Kurtycz DFI. The Milan system for reporting salivary gland cytopathology: analysis and suggestions of initial survey. Cancer Cytopathol. 2017;125(10):757–66. 19. Zhang D, Liao X, Tang Y, Meyer RG, Vand DL, Liu X, Islam MN, Lai J. Warthin-like mucoepidermoid carcinoma of the parotid gland: unusual morphology and diagnostic pitfalls. Anticancer Res. 2019;39(6):3213–7. 20. Hang JF, Shum CH, Ali SZ, Bishop JA.  Cytological features of the Warthin-like variant of salivary mucoepidermoid carcinoma. Diagn Cytopathol. 2017;45(12):1132–6. 21. Carluccio F, Amendola S.  Parapharyngeal oncocytoma of the parotid gland. Acta Otorhinolaryngol Ital. 1996;16(1):57–61. 22. Rainwater L, Farrow G, Hay I, Lieber M. Oncocytic tumours of the salivary gland, kidney, and thyroid: nuclear DNA patterns studied by flow cytometry. Br J Cancer. 1949;53:799–804. 23. Lubin D, Song S, Zafar HM, Baloch Z. The key radiologic and cytomorphologic features of oncocytic and oncocytoid lesions of the salivary gland. Diagn Cytopathol. 2019;47(6):617–36. 24. Manucha V, Gonzalez MF, Akhtar I. Analysis of the risk of malignancy associated with the basaloid and oncocytic subtypes of the salivary gland neoplasm of unknown malignant potential (SUMP) category in the Milan system. Cancer Cytopathol. 2021;129(8):603–11. 25. Torres JMV, Duarte EM, Diaz-Perez JA, Leibowitz J, Weed DT, Thomas G, Sargi Z, Civantos FJ, Arnold DJ, Gomez-Fernandez C, Montgomery EA, Rosenberg AE.  Mesenchymal neoplasms of salivary glands: a clinicopathologic study of 68 cases. Head Neck Pathol. 2021;16(2):353–65. 26. Lozano MD, Landa A, Tobar LG, De Andrea C, Larrache J, Echeveste JI, Paricio JJ, Sanchez B, Medina A, Paisan A. A comprehensive diagnosis of a desmoplastic small round cell tumor of unusual location based on fine-needle aspiration cytology: report of a case arising in the parotid gland and review of the literature. Diagn Cytopathol. 2020;48(9):827–32. 27. Ninchritz-Becerra E, Gonzalez-Garcia J, Garcia-Iza L, Estomba CMC Sr. Desmoplastic small round cell tumor: a rare location in the parotid gland. Cureus. 2020;12(8):e10068. 28. Husain N, Bandhauer F, Kurrer M, Kollias SS.  Lipoma of the parotid gland. Neuroradiol J. 2008;21(1):81–6. 29. Kaur M, Chatterjee D, Kundu R, Punia RS, Saini V. Spindle cell lipoma mimicking pleomorphic adenoma: a diagnostic challenge on cytology. Diagn Cytopathol. 2019;47(4):341–4. 30. Tong KN, Seltzer S, Castle JT.  Lipoma of the parotid gland. Head Neck Pathol. 2020;14(1):220–3. 31. Gandhi A, Mahajan G, Sharma S.  Chondroid lipoma mimicking pleomorphic adenoma on cytology. Diagn Cytopathol. 2013;41(11):1012–3.

102

Z. Baloch et al.

32. Kondo T.  A case of lipomatous pleomorphic adenoma in the parotid gland: a case report. Diagn Pathol. 2009;4:16. 33. Agaimy A. Fat-containing salivary gland tumors: a review. Head Neck Pathol. 2013;7(Suppl 1):S90–6. 34. Yu GH, Sack MJ, Baloch Z, Gupta PK. Difficulties in the fine needle aspiration (FNA) diagnosis of schwannoma. Cytopathology. 1999;10(3):186–94. 35. Guzzo M, Ferraro L, Ibba T, Quattrone P, Bianchi R, Rezzonico S, Scaramellini G. Schwannoma in the parotid gland. Experience at our institute and review of the literature. Tumori. 2009;95(6):846–51. 36. Chandra SR, Das S, Wolf A. Parotid neurogenic tumors: MPNST sarcoma to schwannoma-­ review of literature and guidelines in management. J Maxillofac Oral Surg. 2021;20(3):356–63. 37. Bhaker P, Chatterjee D, Gochhait D, Radotra BD, Dey P. Schwannoma of the parotid gland: diagnosis by fine-needle aspiration cytology. J Cytol. 2014;31(4):196–8. 38. Tille JC, Reychler H, Hamoir M, Schmitz S, Weynand B. Schwannoma-like pleomorphic adenoma of the parotid. Rare Tumors. 2011;3(4):e40. 39. Killeen RM, Davy CL, Bauserman SC. Melanocytic schwannoma. Cancer. 1988;62(1):174–83. 40. Henke AC, Cooley ML, Hughes JH, Timmerman TG.  Fine-needle aspiration cytology of lymphangioma of the parotid gland in an adult. Diagn Cytopathol. 2001;24(2):126–8. 41. Gutmann EJ.  Lymphangioma presenting as a primary parotid neoplasm in an adult. Report of a case with the diagnosis suggested by fine needle aspiration biopsy. Acta Cytol. 1994;38(5):747–50. 42. Hilborne LH, Glasgow BJ, Layfield LJ. Fine-needle aspiration cytology of juvenile hemangioma of the parotid gland: a case report. Diagn Cytopathol. 1987;3(2):152–5. 43. Khurana KK, Mortelliti AJ.  The role of fine-needle aspiration biopsy in the diagnosis and management of juvenile hemangioma of the parotid gland and cheek. Arch Pathol Lab Med. 2001;125(10):1340–3. 44. Qureshi NA, Singhal J, Sharma J.  Parotid gland hemangioma. Indian Pediatr. 2013;50(11):1071–2. 45. Allison DB, VandenBussche CJ, Rooper LM, Wakely PE, Rossi ED, Faquin WC, Ali SZ. Nodular fasciitis of the parotid gland: a challenging diagnosis on FNA. Cancer Cytopathol. 2018;126(10):872–80. 46. Peng WX, Kudo M, Yamamoto T, Inai S, Fujii T, Teduka K, Kawahara K, Naito Z. Nodular fasciitis in the parotid gland: a case report and review of the literature. Diagn Cytopathol. 2013;41(9):829–33. 47. Hang JF, Lee JJL, Nga ME, Higushi K, Hirata Y, et  al. Multi-institutional validation of a modified scheme for subcategorizing salivary gland neoplasm of uncertain malignant potential (SUMP). Cancer Cytopathol. 2022;130(7):511–22. 48. Layfield LJ, Esebua M, Pantanowitz L, Maleki Z, Vazmitsel M, Baloch Z, Cantley RL, Schmidt R. Salivary gland neoplasms with basaloid features in the era of the Milan system for reporting salivary gland cytology: classification and interobserver agreement. Diagn Cytopathol. 2022;50(7):341–9. 49. Wu HH, Alruwaii F, Zeng BR, Cramer HM, Lai CR, Hang JF. Application of the Milan system for reporting salivary gland cytopathology: a retrospective 12-year bi-institutional study. Am J Clin Pathol. 2019;151(6):613–21. 50. Jalaly JB, Farahani SJ, Baloch ZW. The Milan system for reporting salivary gland cytopathology: a comprehensive review of the literature. Diagn Cytopathol. 2020;48(10):880–9. 51. Hosseini SM, Resta IT, Baloch ZW. Diagnostic performance of Milan system for reporting salivary gland cytopathology: a prospective study. Diagn Cytopathol. 2021;49(7):822–31. 52. Molnar SL, Zarka MA, De Las Casas LE. Going beyond “basaloid neoplasm”: fine needle aspiration cytology of epithelial-myoepithelial carcinoma of the parotid gland. Diagn Cytopathol. 2016;44(5):422–5.

5 Neoplasm

103

53. Liu H, Ljungren C, Lin F, Zarka MA, Chen L. Analysis of histologic follow-up and risk of malignancy for salivary gland neoplasm of uncertain malignant potential proposed by the Milan system for reporting salivary gland cytopathology. Cancer Cytopathol. 2018;126(7):490–7. 54. Tawfik O, Tsue T, Pantazis C, Nuckols D, Younes S, Webb P. Salivary gland neoplasms with basaloid cell features: report of two cases diagnosed by fine-needle aspiration cytology. Diagn Cytopathol. 1999;21(1):46–50. 55. Faquin WC. Diagnosis and grading of basaloid salivary gland tumors using the Milan system for reporting salivary gland cytopathology. Cancer Cytopathol. 2020;128(2):87–8. 56. Gargano SM, Sebastiano C, Solomides CC, Griffith CC, HooKim K.  Cytohistologic correlation of basaloid salivary gland neoplasms: can cytomorphologic classification be used to diagnose and grade these tumors? Cancer Cytopathol. 2020;128(2):92–9. 57. Saeed OA, Whaley RD, Segura S, Wu HH, Cramer H. The cytology of salivary gland neoplasms with globules of extracellular matrix: case-based review of adenoid cystic carcinoma and its potential mimics. Diagn Cytopathol. 2021;49(6):E195–202. 58. Aisagbonhi OA, Tulecke MA, Wilbur DC, Goldar-Najafi A, Iqbal S, Sadow PM, Faquin WC. Fine-needle aspiration of epithelial-myoepithelial carcinoma of the parotid gland with prominent adenoid cystic carcinoma-like cribriform features: avoiding a diagnostic pitfall. Am J Clin Pathol. 2016;146(6):741–6. 59. Layfield LJ, Tan P, Glasgow BJ. Fine-needle aspiration of salivary gland lesions. Comparison with frozen sections and histologic findings. Arch Pathol Lab Med. 1987;111(4):346–53. 60. Mostaan LV, Yazdani N, Madani SZ, Borghei H, Mortazavi S, Ojani L, Mokhtari Z. Frozen section as a diagnostic test for major salivary gland tumors. Acta Med Iran. 2012;50(7):459–62. 61. Fulciniti F, Losito NS, Ionna F, Longo F, Aversa C, Botti G, Foschini MP. Sclerosing polycystic adenosis of the parotid gland: report of one case diagnosed by fine-needle cytology with in situ malignant transformation. Diagn Cytopathol. 2010;38(5):368–73. 62. Shilpi, Ahmad Ansari F, Bahadur S, Katyal A, Narula A, Nargotra N, Singh S. Sclerosing polycystic adenosis: a rare tumor misdiagnosed as retention cyst on fine needle aspiration cytology. Diagn Cytopathol. 2017;45(7):640–4. 63. Wade TV, Livolsi VA, Montone KT, Baloch ZW. A cytohistologic correlation of mucoepidermoid carcinoma: emphasizing the rare oncocytic variant. Pathol Res Int. 2011;2011:135796. 64. Layfield LJ, Glasgow BJ. Aspiration cytology of clear-cell lesions of the parotid gland: morphologic features and differential diagnosis. Diagn Cytopathol. 1993;9(6):705–11. 65. Reis-Filho JS. Primary salivary clear cell tumors. Arch Pathol Lab Med. 2003;127(3):278–9. 66. Woods T, Fitzpatrick S, Cohen D, Islam M, Bhattacharyya I. Clear cell changes in salivary gland neoplasms: a 20-year retrospective study. Med Oral Patol Oral Cir Bucal. 2017;22(3):e276–81. 67. Balakrishnan M, George SA, Haji BE, Francis IM. Fine-needle aspiration cytology of mammary analog secretory carcinoma of the parotid gland: a diagnostic conundrum. J Cytol. 2019;36(4):215–6. 68. Samulski TD, LiVolsi VA, Baloch Z. The cytopathologic features of mammary analog secretory carcinoma and its mimics. Cytojournal. 2014;11:24.

Chapter 6

Suspicious for Malignancy Esther Diana Rossi, Syed Ali, Claude Bigorgne, Ashish Chandra, Yun Gong, Daniel Lubin, Renata B. Perak, Bo Ping, He Wang, and Matthew Zarka

E. D. Rossi (*) Unita’ Operativa Istopatologia e Citodiagnostica, Fondazione Policlinico Universitario A. Gemelli, Rome, Italy e-mail: [email protected] S. Ali Department of Pathology, The Johns Hopkins Hospital, Baltimore, MD, USA e-mail: [email protected] C. Bigorgne Cytopathology and Imaging Center, Paris, France A. Chandra Cellular Pathology, Guy’s and St Thomas’ NHS Foundation Trust, London, UK e-mail: [email protected] Y. Gong Department of Pathology, University of Texas MD Anderson Cancer Center, Houston, TX, USA e-mail: [email protected] D. Lubin Department of Pathology and Laboratory Medicine, Emory University Hospital Midtown, Atlanta, GA, USA e-mail: [email protected] R. B. Perak Department of Pathology, Forensic Medicine and Cytology, University School of Medicine Split, University Hospital Center Split, Split, Croatia B. Ping Department of Pathology, Shanghai Cancer Center, Shanghai, China H. Wang Department of Pathology, Yale University School of Medicine, Yale New Haven Health System St. Raphael Campus New Haven, New Haven, CT, USA e-mail: [email protected] M. Zarka Department of Laboratory Medicine and Pathology, Mayo Clinic Arizona, Scottsdale, AZ, USA e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 W. C. Faquin et al. (eds.), The Milan System for Reporting Salivary Gland Cytopathology, https://doi.org/10.1007/978-3-031-26662-1_6

105

106

E. D. Rossi et al.

General Background The categories “Atypia of Undetermined Significance” (AUS), “Neoplasm: Salivary Gland Neoplasm of Uncertain Malignant Potential” (SUMP), and “Suspicious for Malignancy” (SM) represent indeterminate diagnostic categories in the Milan System [1]. They are used to stratify the risk of malignancy (ROM) and to inform the treating clinician that a particular specimen cannot be placed into a more specific benign or malignant diagnostic category due to diagnostic limitations such as sparse cellularity or various specimen artifacts (see also Chaps. 4 and 5). The SM category is a traditional diagnostic category used in nearly all cytology reporting systems, and, as such, its characteristics are well-known to practicing cytologists [2–7]. The purpose of separating SM from the malignant category is to preserve the high positive predictive value (PPV) of an FNA classified as Malignant while at the same time offering a diagnostic option with a relatively high ROM for those FNAs where the cytomorphologic criteria fall short in quantity and/or quality for a Malignant diagnosis [8–21]. In the second edition of the Milan System, the ROM for the SM category is 83% which is related to published series since 2018 demonstrating an increased ROM of 80–85%. With the growing availability of immunohistochemical and molecular markers for salivary gland tumors (see Chap. 8, Ancillary Studies), a subset of FNAs classified as SM may benefit from the application of ancillary testing (see Chap. 8) to yield a more specific interpretation.

Definition A salivary gland FNA is classified as SM when some, but not all, of the criteria for a specific diagnosis of malignancy are present, and the overall cytologic features are suggestive of malignancy.

Cytologic Criteria When making a diagnosis of SM, the FNA should be subcategorized if possible. For example, a SM specimen may be described as suspicious for a primary salivary gland malignancy, suspicious for a metastasis, or suspicious for lymphoma [8–12]. A significant proportion of SM cases will be suboptimal samples of a high-grade malignancy. Aspects of a salivary gland FNA leading to an interpretation of SM include: • Presence of markedly atypical cells with poor smear preparation, cell preservation, fixation artifact, or obscuring inflammation and blood (Figs.  6.1a, b and 6.2a, b).

6  Suspicious for Malignancy

a

107

b

Fig. 6.1  Suspicious for Malignancy. (a, b) The images show rare markedly atypical cells suggestive of carcinoma, but the classification is limited by scant cellularity and partial air-drying artifact (smears, Papanicolaou stain)

a

b

Fig. 6.2  Suspicious for Malignancy. (a, b) The aspirates contain markedly atypical cells suspicious for high-grade carcinoma, but with obscuring blood limiting the assessment (smears, Diff-­ Quik stain)

• Presence of limited cytologic features of a specific malignant lesion (e.g., adenoid cystic carcinoma, mucoepidermoid carcinoma, acinic cell carcinoma) in an otherwise sparsely cellular aspirate (Figs. 6.3a, b, 6.4 and 6.5a, b). • Presence of markedly atypical and/or suspicious cytologic features but with scant cellularity (Fig. 6.6). Atypical features can include prominent nucleoli or macronucleoli, anisonucleosis, increased nuclear to cytoplasmic ratio, nuclear molding, prominent nuclear pleomorphism, atypical mitoses, and clumped, coarse chromatin (Fig. 6.7). • Scant sample with atypical features suggestive of a neuroendocrine neoplasm (Fig. 6.8). • Scant sample with atypical features suggestive of a metastatic lesion (combined with the patient history and clinical evidence of a primary malignancy).

108

a

E. D. Rossi et al.

b

Fig. 6.3  Suspicious for Malignancy. (a, b) The aspirates show a group of epithelial cells suggestive of acinic cell carcinoma, but hypocellularity and background blood in the absence of ancillary studies limit the evaluation (smear, Papanicolaou stain)

Fig. 6.4  Suspicious for Malignancy. This specimen is comprised of basaloid cells and abundant matrix spheres with a pattern suspicious for adenoid cystic carcinoma (smear, Papanicolaou stain)

a

b

Fig. 6.5  Suspicious for Malignancy. (a, b) The smears consist of epithelial cells with an epidermoid appearance, suggestive of mucoepidermoid carcinoma (smears, Diff-Quik stain)

6  Suspicious for Malignancy

109

Fig. 6.6  Suspicious for Malignancy. The slide shows the presence of markedly atypical (upper left) cytologic features in a subset of cells but admixed with features of pleomorphic adenoma (smear, Papanicolaou stain)

a

b

c

Fig. 6.7  Suspicious for Malignancy. (a–c) This aspirate is hypocellular but contains occasional small groups of markedly atypical cells suspicious for carcinoma. The corresponding resection showed a high-grade mucoepidermoid carcinoma (smears, Papanicolaou stain)

A differential diagnosis of lymphoma is usually considered in salivary gland aspirates with a prominent population of lymphocytes with background lymphoglandular bodies [17, 18]. Immunophenotyping, usually by flow cytometry, is key to making a diagnosis of lymphomas in cytologic specimens. Thorough clinical correlation is also essential. Successful subclassification of the lymphoma may require histological evaluation, as well as ancillary immunohistochemical and molecular

110

E. D. Rossi et al.

Fig. 6.8  Suspicious for Malignancy. This slide shows neoplastic cells with “salt and pepper” chromatin suggestive of neuroendocrine differentiation (smear, Papanicolaou stain)

a

b

Fig. 6.9  Suspicious for Malignancy. (a, b) This slide shows a population of enlarged atypical lymphocytes suspicious for a large cell lymphoma (smear, Papanicolaou stain)

studies. Many of the aspirates of lymphoma classified as SM lack sufficient material for the performance of these ancillary studies [17, 18]. A detailed cytology review of a lymphoma diagnosis is beyond the scope of this atlas, but some of the cytomorphologic features suggestive of lymphoma include: • A population of enlarged atypical lymphocytes as seen in large cell lymphomas (Fig. 6.9a, b). • A monomorphic lymphoid population. This may be made up of small/intermediate lymphocytes as in intermediate-grade follicular lymphoma (Fig.  6.10) or showing monotonous medium-sized lymphocytes suggesting mantle cell ­lymphoma, or small mature-appearing lymphocytes with round nuclei and coarse chromatin suggesting small lymphocytic lymphoma.

6  Suspicious for Malignancy

111

Fig. 6.10  Suspicious for Malignancy. This aspirate shows a monotonous population of intermediate-­ size lymphocytes that based upon cytomorphology alone are highly suspicious for lymphoma. Additional ancillary studies including immunophenotyping are needed for classification (smear, Papanicolaou stain)

Fig. 6.11  Suspicious for Malignancy. This aspirate shows a mixed pattern with a predominance of intermediate-size lymphocytes as can be seen in extranodal marginal zone lymphomas. A tingible body macrophage is present. Ancillary studies are needed for further classification (smear, Papanicolaou stain)

• A heterogeneous lymphoid population with atypical forms (Fig. 6.11). Extranodal marginal zone lymphomas (EMZL) are especially characterized by a heterogeneous cell population including small to intermediate-size centrocyte-like cells, and a smaller number of larger lymphoid cells, plasmacytoid cells, tingible body macrophages, dendritic cells, and plasma cells. As patients with Sjogren’s syndrome are at increased risk for these lymphomas, a high index of suspicion should be employed in these patients, particularly in those with concerning clinical (lymphadenopathy, salivary gland enlargement, palpable purpura) or laboratory (rheumatoid factor positivity, C4 hypocomplementemia) findings [22].

112

E. D. Rossi et al.

Explanatory Notes A diagnosis of SM can be made in aspirates showing focal marked cellular atypia in a less than optimal specimen. Once significant atypia, suggesting malignancy, is identified in an FNA, which has scant cellularity or is poorly prepared, the case is no longer insufficient or non-diagnostic. SM also usually indicates that the cytology is characterized by a higher degree of atypia than that seen in the AUS and SUMP categories thus highly suggestive of a malignant lesion. SM should not be used for cases where the cytological features are only mildly to moderately atypical and better classified as AUS or SUMP.  The latter is associated with significantly lower ROMs than cases classified as SM. The cytomorphological stratification of AUS, SUMP, and SM can be subtle and, in some cases, subjective, but careful scrutiny of the cytomorphological features and proper application of ancillary techniques will aid in accurate classification. In some cases, the diagnosis of SM may be upgraded to “Malignant” once the results of any ancillary studies become available. Where ROSE is offered, it can be used to improve the quality and quantity of the FNA specimen and assist in triaging material for additional diagnostic studies. A majority of salivary gland FNAs classified as SM will be samples of high-­grade cancers that have some limiting factor precluding a definitive diagnosis of malignancy. A subset of cases will be lower grade salivary gland cancers that exhibit many of the characteristic cytologic features of a particular salivary gland cancer, but for qualitative or quantitative reasons, are not sufficient to be diagnostic (Fig.  6.12). Most commonly, aspirates of low-grade mucoepidermoid carcinoma, acinic cell carcinoma, and adenoid cystic carcinoma will fall into the latter subgroup. Other tumors such as certain neuroendocrine carcinomas, which are rare in the salivary gland, are usually diagnostic of malignancy and can be subtyped provided that adequate material is available for ancillary studies. The most common form of neuroendocrine carcinoma in the salivary gland is poorly differentiated neuroendocrine carcinoma Fig. 6.12  Suspicious for Malignancy. This aspirate shows cytologic features that are highly suspicious for adenoid cystic carcinoma, but the specimen is limited to a single Papanicolaoustained smear (smear, Papanicolaou stain)

6  Suspicious for Malignancy

113

with Merkel cell-like features, and, by cytomorphology alone, it would typically be interpreted as malignant unless it were a compromised specimen. Salivary gland aspirates containing a prominent lymphoid population will require ancillary studies for a definitive diagnosis of lymphoma. Otherwise, the classification of the aspirate as SM can be used for cases where there is a cytologic pattern suggesting lymphoma, such as the presence of large atypical lymphocytes or a monomorphic lymphoid population. Most often, there will be a heterogeneous lymphoid population, and the differential diagnosis will include a benign process such as reactive lymphoid hyperplasia, chronic sialadenitis, or Sjogren’s syndrome. Occasionally, such cases can exhibit sufficiently atypical cytomorphologic and clinical features as to be suspicious for lymphoma, but flow cytometry or other methods of immunophenotypic analysis are essential to ultimately rule in or rule out lymphoma. If the FNA has not been submitted for flow cytometry, repeat FNA with flow is the best approach. For proper subclassification and grading of lymphoma, obtaining a core needle biopsy or excisional biopsy for histological evaluation is often needed. While rarely involving the salivary glands or intra-parotid lymph nodes, classic Hodgkin lymphoma has distinctive cytomorphologic features, which would lead to a diagnosis of at least “suspicious for Hodgkin lymphoma” in many cases. Flow cytometry would generally not be useful for confirming the diagnosis of Hodgkin lymphoma, but material for other ancillary studies would be indicated; excisional biopsy would usually be needed for a definitive diagnosis.

Clinical Management The cytologic diagnosis of SM is not equivalent to “Malignant” even though it is suggestive of a malignant lesion and the risk of malignancy is high. It cannot be used alone as a basis for radical surgery, chemotherapy, or radiotherapy (see Chap. 9). In response to a diagnosis of SM, consideration should be given as to whether obtaining additional material by repeat FNA, core biopsy, open biopsy, or surgical excision would be useful. For cases with repeat FNA, every effort should be made to obtain adequate material for any ancillary studies that would be indicated. Clinical and radiologic correlations are of course important, and when surgery is performed, intraoperative frozen section can be considered in appropriate cases.

Sample Reports Example 6.1 Satisfactory for evaluation. SUSPICIOUS FOR MALIGNANCY. Rare markedly atypical cells, suspicious for high-grade carcinoma.

114

E. D. Rossi et al.

Example 6.2 Satisfactory for evaluation. SUSPICIOUS FOR MALIGNANCY. Suspicious for high-grade mucoepidermoid carcinoma/adenoid cystic carcinoma/salivary duct carcinoma. Example 6.3 Evaluation limited by scant cellularity. SUSPICIOUS FOR MALIGNANCY. Atypical cells in a mucinous background, suspicious for low-grade mucoepidermoid carcinoma. Example 6.4 Satisfactory for evaluation. SUSPICIOUS FOR MALIGNANCY. Rare large atypical lymphocytes, suspicious for non-Hodgkin lymphoma. See note. Note: Further evaluation using immunophenotyping studies such as flow cytometry or immunochemistry in a repeat FNA or other tissue biopsy is recommended. Example 6.5 Satisfactory for evaluation. SUSPICIOUS FOR MALIGNANCY. Monomorphic population of small lymphocytes, suspicious for non-Hodgkin Lymphoma. See note. Note: Additional tissue sampling either by repeat FNA or other tissue biopsy is recommended for further evaluation with ancillary studies including flow cytometry. Example 6.6 Evaluation limited by scant well-preserved cells. SUSPICIOUS FOR MALIGNANCY. Cyst contents with occasional atypical squamous cells and dyskeratotic cells, suspicious for metastatic keratinizing squamous cell carcinoma.

References 1. Rossi ED, Faquin WC, Baloch Z, Barkan GA, Foschini MP, Pusztaszeri M, Vielh P, Kurtycz DFI. The Milan system for Reporting salivary gland cytopathology: analysis and suggestions of initial survey. Cancer Cytopathol. 2017;125:757–66. 2. Contucci AM, Corina L, Sergi B, Fadda G, Paludetti G.  Correlation between fine needle aspiration biopsy and histologic findings in parotid masses. Personal experience. Acta Otorhinolaryngol Ital. 2003;23:314–8. 3. Griffith CC, Reetesh KP, Schneider F, et al. Salivary gland tumor fine needle aspiration cytology. A proposal for a risk stratification classification. Am J Clin Pathol. 2015;143:839–53.

6  Suspicious for Malignancy

115

4. Hughes JH, Volk EE, Wilbur DC, et al. Pitfalls in salivary gland fine needle aspiration cytology: lessons from the college of American pathologists interlaboratory comparison program in nongynaecologic cytology. Arch Pathol Lab Med. 2005;129:26–31. 5. Jain E, Gupta R, Kudesia M, Singh S. Fine needle aspiration cytology in diagnosis of salivary gland lesions: a study with histological comparison. Cytojournal. 2013;10:5. 6. Mairembam P, Jay A, Beale T, Morley S, Vaz F, Kalavrezos N, Kocjan G. Salivary gland FNA cytology: role as a triage tool and an approach to pitfalls in cytomorphology. Cytopathology. 2016;27:91–6. 7. Rossi ED, Wong LQ, Bizzarro T, et al. The impact of fine needle aspiration cytology in the management of salivary gland lesions: institutional experiences leading to a risk based classification scheme. Cancer Cytopathol. 2016;124:388–96. 8. Brennan PA, Davies B, Poller D, et al. Fine needle aspiration cytology (FNAC) of salivary gland tumours: repeat aspiration provides further information in cases with an unclear initial cytological diagnosis. Br J Oral Maxillofac Surg. 2010;48:26–9. 9. Wei S, Layfield LJ, LiVolsi VA, et al. Reporting of fine needle aspiration (FNA) specimens of salivary gland lesions: a comprehensive review. Diagn Cytopathol. 2017;45:91–6. 10. Colella G, Cannavale R, Flamminio F, Foschini MP. Fine-needle aspiration cytology of salivary gland lesions: a systematic review. J Oral Maxillofac Surg. 2010;68(9):2146–53. 11. Tyagi R, Dey P.  Diagnostic problems of salivary gland tumors. Diagn Cytopathol. 2015;43(6):495–509. 12. Wang H, Fundakowski C, Khurana JS, Jhala N. Fine-needle aspiration biopsy of salivary gland lesions. Arch Pathol Lab Med. 2015;139(12):1491–7. 13. Darvishian F, Lin O.  Myoepithelial cell-rich neoplasms: cytological features of benign and malignant lesions. Cancer Cytopathol. 2004;102:355–61. 14. Chen L. Cytopathologic analysis of stroma-poor salivary gland epithelia/myoepithelial neoplasms on fine needle aspiration cytology. Acta Cytol. 2012;56:25–33. 15. Layfield LJ, Glasgow BJ. Diagnosis of salivary gland tumors by fine needle aspiration cytology: a review of clinical utility and pitfalls. Diagn Cytopathol. 1991;7:267–72. 16. Nanda KDS, Mehta A, Nanda J. Fine needle aspiration cytology: a reliable tool in the diagnosis of salivary gland lesions. J Oral Pathol Med. 2012;41:106–12. 17. Turner MD. Salivary gland disease in Sjögren’s syndrome: sialoadenitis to lymphoma. Oral Maxillofac Surg Clin N Am. 2014;26(1):75–81. 18. Dei P, Amir T, Al Jassar A, et al. Combined applications of fine needle aspiration cytology and flow cytometric immunophenotyping for diagnosis and classification of non Hodgkin lymphoma. Cytojournal. 2006;27(3):24–5. 19. Maleki Z, Miller JA, Arab SE, et al. Suspicious salivary gland FNA: risk of malignancy and interinstitutional variability. Cancer Cytopathol. 2018;126:94–100. 20. Wang H, Malik A, Maleki Z, Rossi ED, Ping B, Chandra A, Ali SZ, Fadda G, Wang J, Arab SE, Zhao H, Jhala N. Atypical salivary gland fine needle aspiration: risk of malignancy and interinstitutional variability. Diagn Cytopathol. 2017;45:1088–94. 21. Rohilla M, Singh P, Rajwanshi A, Gupta N, Srinivasan R, Dey P, Vashishta RK. Three-year cytohistological correlation of salivary gland FNA cytology at a tertiary center with the application of the Milan system for risk stratification. Cancer Cytopathol. 2017;125:767–75. 22. Fragkioudaki S, Mavragani CP, Moutsopoulos H. Predicting the risk for lymphoma development in Sjogren syndrome: an easy tool for clinical use. Medicine. 2016;95:1–8.

Chapter 7

Malignant Swati Mehrotra, Mousa Al-Abbadi, Claude Bigorgne, Jalal Jalalay, Jeffrey F. Krane, Renata B. Perak, Philippe Vielh, Paul E. Wakely Jr, He Wang, Eva M. Wojcik, and Güliz A. Barkan

S. Mehrotra (*) · E. M. Wojcik · G. A. Barkan (*) Department of Pathology and Laboratory Medicine, Loyola University Hospital, Maywood, IL, USA e-mail: [email protected], [email protected]; [email protected]; [email protected] M. Al-Abbadi Pathology and Cytopathology, Jordan University Hospital, Amman, Jordan Histopathology, Microbiology and Forensic Medicine, University of Jordan-College of Medicine, Amman, Jordan C. Bigorgne Cytopathology and Imaging Center, Paris, France J. Jalalay Head and Neck and Cytopathology, Perelman School of Medicine, University of Pennsylvania, Philadelphia, PA, USA e-mail: [email protected] J. F. Krane Department of Pathology and Laboratory Medicine, David Geffen School of Medicine at UCLA, Los Angeles, CA, USA e-mail: [email protected] R. B. Perak Department of Pathology, Forensic Medicine and Cytology, University School of Medicine Split, University Hospital Center Split, Split, Croatia P. Vielh Department of Pathology, Medipath and American Hospital of Paris, Paris, France P. E. Wakely Jr The Ohio State U. Wexner Medical Center, James Cancer Hospital and Solove Research Institute, Columbus, OH, USA e-mail: [email protected] H. Wang Department of Pathology, Yale University School of Medicine, Yale New Haven Health System St. Raphael Campus New Haven, New Haven, CT, USA e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 W. C. Faquin et al. (eds.), The Milan System for Reporting Salivary Gland Cytopathology, https://doi.org/10.1007/978-3-031-26662-1_7

117

118

S. Mehrotra et al.

General Background Malignant salivary gland tumors include a diverse group of primary neoplasms involving both the major and minor salivary glands [1–4]. In addition, various secondary neoplasms (e.g., metastatic cutaneous squamous cell carcinoma) can also involve the salivary glands or lymph nodes within or closely associated with salivary glands. A recent study indicates that parotid surgery for malignant lesions is more likely to be performed for secondary tumors, particularly metastatic cutaneous squamous cell carcinomas [5]. A majority of malignant tumors occur in the parotid and submandibular glands and account for most of the salivary gland fine needle aspirations (FNA) [1–4, 6–9]. This chapter will discuss tumors that commonly involve the major salivary glands, and that can be diagnosed by FNA.  A brief description of minor salivary gland carcinomas which may at times be sampled by FNA is also included. While some low-grade salivary gland cancers overlap cytomorphologically with their benign counterparts, many of the more common primary cancers in well-sampled and well-prepared FNAs with adequate material for ancillary studies will exhibit features that are sufficiently distinctive to be classified as Malignant. Once the diagnostic threshold has been reached for classifying a salivary gland FNA as Malignant, especially those with epithelial differentiation, an attempt should be made when possible to grade the cancer given the potential impact on clinical management (see Chap. 10, Clinical Management).

Definition Salivary gland aspirates classified as Malignant contain a combination of cytomorphologic features which either alone or in combination with ancillary studies are diagnostic of malignancy. When possible, an attempt should be made to provide the grade of the neoplasm as well as the specific tumor type (e.g., low-grade mucoepidermoid carcinoma). If a specific grade cannot be readily assigned, the tumor should be designated as “indeterminate grade” [10].

Low-Grade Carcinomas Acinic Cell Carcinoma (ACC) Acinic cell carcinoma (ACC) comprises approximately 10–15% of all salivary gland epithelial malignancies and overall is the second most common malignant salivary gland tumor after mucoepidermoid carcinoma (MEC). In the pediatric age group, it constitutes about a third of salivary gland carcinomas [7, 9]. It shows a slight female predilection (1.5:1) and a wide age distribution; the mean age at diagnosis is 50  years. ACC occurs most commonly in the parotid gland while many

7 Malignant

119

minor salivary gland tumors previously diagnosed as ACC have been reclassified as secretory carcinoma. Most ACC present as mobile, soft to firm, well-circumscribed masses ranging from 1 to 4 cm in size. The tumors are usually asymptomatic and slow-growing. Pain, fixation to surrounding tissues, and facial nerve involvement are considered poor prognostic features and may indicate high-grade transformation. ACC can metastasize to cervical lymph nodes and local tumor recurrences can be seen in up to 35% of cases. Distant metastases are rare; however, they have been reported in the liver and lung.

Cytologic Criteria ACC is a malignant epithelial neoplasm in which at least some of the neoplastic cells exhibit serous acinar differentiation, characterized by the presence of Periodic Acid Schiff (PAS)-positive diastase-resistant cytoplasmic zymogen granules. Most FNA specimens of ACC show the following characteristics [11, 12]: • Cellular with uniform population of epithelial cells (Fig. 7.1). • Polygonal tumor cells with low N:C ratio and abundant delicate vacuolated cytoplasm with basophilic quality (Fig. 7.2). • Variable numbers of cytoplasmic zymogen granules. These granules are PAS-­ positive and diastase-resistant (Fig. 7.3). • Predominantly dispersed or loosely cohesive cell population (Fig. 7.1). • Capillary meshwork with loosely adherent cells or well-developed papillary formations can be seen. • Uniform, round eccentric nuclei with distinct nucleoli (Fig. 7.4). • Minimal nuclear pleomorphism. • No mitotic activity or necrosis. Fig. 7.1  Malignant. Acinic cell carcinoma. Cellular smears with loosely cohesive groups of acinar cells adherent to delicate capillary meshwork. Note the presence of naked nuclei in a delicate frothy background; and the conspicuous absence of ductal cells (Diff-Quik stain)

120

S. Mehrotra et al.

Fig. 7.2  Malignant. Acinic cell carcinoma. Discohesive well-preserved tumor cells with granular cytoplasm. The cells are polygonal with low N/C ratio and conspicuous nucleoli (Diff-Quik stain)

a

b

Fig. 7.3 (a, b) Malignant. Acinic cell carcinoma. Aspirate showing a sheet of cells with abundant granular cytoplasm with scattered small coarse granules [Papanicolaou (a) and Diff-Quik stains] Fig. 7.4  Malignant. Acinic cell carcinoma. Threedimensional clusters of acinar cells with abundant granular cytoplasm, low N:C ratio, uniform, round, eccentric nuclei, and conspicuous nucleoli (Papanicolaou stain)

• Clean or frothy background; stripped nuclei. • Background lymphocytes are present in a subset of cases. • Rare psammoma bodies.

7 Malignant

121

Explanatory Notes While ACC usually has a dispersed cell pattern, small, crowded groups of cells or papillary clusters around a rich capillary meshwork can sometimes be seen. The tumor cells are large and polygonal to oval with indistinct cell borders, and abundant delicate vacuolated cytoplasm with a subtle basophilic quality. Cytoplasmic zymogen granules, which are indicative of serous acinar differentiation, are usually coarse, and basophilic in Papanicolaou-stained preparations, but are best seen in Romanowsky-type stains in which they appear red or magenta. Unfortunately, zymogen granules are often sparse and/or difficult to detect on routinely stained cytologic preparations. In addition to serous acinar cells, aspirates can also show clear cells, intercalated duct-like cells, and non-specific glandular cells. Intercalated duct-like cells are smaller, cuboidal, have a higher N:C ratio with centrally placed nuclei, and lack the classic cytoplasmic zymogen granules. Non-specific glandular cells are frequently seen; they resemble the intercalated duct-like cells but are larger and rounder (Fig. 7.5). Most ACC have minimal to no nuclear pleomorphism and lack necrosis and mitoses. Numerous stripped nuclei may be present in the aspirate and can be difficult to distinguish from lymphocytes. Material should be collected whenever possible for ancillary studies. Demonstration of PAS-positive diastase-­ resistant cytoplasmic zymogen granules is helpful. In contrast to normal salivary gland acini, amylase is not regularly expressed in ACC, and myoepithelial markers (e.g., smooth muscle actin, p63, keratin 5/6, calponin, and S-100) are generally negative. The most useful ancillary markers of ACC are DOG1 and SOX10 (Figs. 7.6a–b) (See Table 7.1). A recent study concluded that moderate to strong immunohistochemical NR4A3 staining is diagnostic of ACC [13]. (See Chap. 8, Ancillary Markers). ACC are often deceptively bland tumors, and hence can occasionally be confused with non-neoplastic salivary gland elements or sialadenosis. The latter two entities maintain the characteristic grape-like arrangement of normal acinar cells with associated ductal cells whereas aspirates of ACC have a more monotonous Fig. 7.5  Malignant. Acinic cell carcinoma. Aspirates may show nonspecific glandular cells which resemble intercalated duct-like cells (Papanicolaou stain)

122

S. Mehrotra et al.

a

b

Fig. 7.6  Acinic cell carcinoma, cell block. (a) Cytoplasmic zymogen granules are variably present in most neoplastic cells (H&E stain). (b) DOG1 immunochemistry. Cytoplasmic and membranous staining is readily apparent Table 7.1  Immunohistochemical stains in the differential diagnosis of low-grade salivary gland malignancies

ACC

p63/ p40 −

AdCC + (ME) MEC + (LG) (SQ) SC − EMC + (ME) MC +

SMA, SMMHC, calponin − + (ME) − − + (ME) +

S-100 CK8/18 CK5/6 CD117 MUCIN PAS-D − + − − − + (GRANULES) + + (EP) + + − − (ME) − + + − + + (MUC) (SQ) +++ + − − − +/− + + + − − − (ME) (ME) + +/− + − − −

DOG1 + − − −/+ + −

ACC acinic cell carcinoma, AdCC adenoid cystic carcinoma, EMC epithelial-myoepithelial carcinoma, EP epithelial (luminal) cells, LG low-grade, MC myoepithelial carcinoma, ME myoepithelial cells, MEC mucoepidermoid carcinoma, MUC mucin-secreting cells, SC secretory carcinoma, SMA smooth muscle actin, SMMHC smooth muscle myosin heavy chain SQ squamoid (epidermoid) cells

population of acinar cells with extensive cellular dyshesion and lack the grape-like cytomorphology of non-neoplastic acinar cells. When the cells are more vacuolated or clear, it can lead to confusion with low-grade MEC, which is positive with mucin stains. Low-grade MEC also shows an admixture of three cell types including intermediate, epidermoid, and mucinous cells. Similarly, vacuolated acinar cells may lead to confusion with sebaceous tumors, which have abundant lipid-rich cytoplasm that is negative for PAS-D, and with epithelial-myoepithelial carcinomas, which express myoepithelial markers and are PAS-D negative. ACCs share many cytomorphologic features with secretory carcinoma, which before its recognition as a separate entity, was classified as ACC. However, secretory carcinoma lacks cytoplasmic zymogen granules, is negative for DOG1, and expresses S-100, GATA3, and mammaglobin diffusely. Molecular studies can be

7 Malignant

123

performed on the cytology material for the ETV6::NTRK3 translocation which is specific for secretory carcinoma. Other entities in the differential diagnosis of ACC due to its oncocytoid qualities are oncocytoma and Warthin tumor. True oncocytes as seen in Warthin tumor and oncocytoma have dense non-vacuolated granular cytoplasm and are histochemically positive for PTAH.  ACC with a predominance of intercalated duct-like cells and non-specific glandular cells is among the most difficult ACC to recognize cytologically and may be classified as Neoplasm: SUMP (Chap. 5) or as SM (Chap. 6). Metastatic renal cell carcinomas can resemble ACC and are best differentiated from ACC with the help of immunohistochemistry, clinical history, and supporting imaging studies. In rare cases, ACC can undergo high-­ grade transformation (“dedifferentiation”) and would be classified by FNA as a high-grade carcinoma.

Secretory Carcinoma (SC) SC is analogous to secretory breast cancer. SC expresses S-100 protein, GATA-3, mammaglobin, vimentin, and harbors a t(12;15)(p13;q25) translocation which leads to the ETV6::NTRK3 fusion product. The tumor is found most commonly in the parotid gland, followed by the intraoral minor salivary glands, and submandibular gland. Most SCs occur in adults and show an equal gender distribution; the mean age is 47 years (range 14–78 years) [14, 15]. The tumors range in size from 1 to 4 cm. The clinical course of SC is characterized by an indolent behavior, with a moderate risk of local recurrence (15%), lymph node metastasis (20%), and a low risk of distant metastasis (5%) [7, 14–16]. High-grade transformation, akin to that seen in other low-grade salivary gland malignancies has been described but is infrequent in SC [17, 18].

Cytologic Criteria SC is composed of microcystic, tubular, and solid structures with eosinophilic colloid-­like background secretory material (Fig.  7.7a–c). Cells have low-grade vesicular nuclei with finely granular chromatin and distinctive centrally located nucleoli (Fig.  7.8). Moderate to abundant pale vacuolated cytoplasm is present (Fig. 7.9a–b). Marked nuclear atypia, mitoses, or necrosis is absent or rare. • Cellular aspirate. • Cells present singly and in tubular, follicular, and papillary groups and no lobular arrangement of normal acinar cells. • Bland cuboidal, polygonal low N:C cells with indistinct cell borders. • Abundant vacuolated pale cytoplasm. • Absence of cytoplasmic zymogen granules. • Uniform round eccentrically located nuclei with smooth nuclear contours, fine chromatin, and distinct nucleolus. • Background mucoproteinaceous material.

124

a

S. Mehrotra et al.

b

c

Fig. 7.7 (a–c) Malignant. Secretory carcinoma. Aspirates show different architectural patterns of microcystic, tubular, microfollicular, and solid sheet of malignant cells with eosinophilic colloid-­ like background secretory material (Diff-Quik and Papanicolaou stains) Fig. 7.8 Malignant. Secretory carcinoma. Aspirate consists of cells with low-grade vesicular nuclei with finely granular chromatin in some of them and distinctive centrally located nucleoli (ThinPrep, Papanicolaou stain)

7 Malignant

a

125

b

Fig. 7.9 (a, b) Malignant. Secretory carcinoma. Aspirate shows cells with moderate to abundant pale to pink vacuolated or granular cytoplasms (Diff-Quik stain)

Explanatory Notes The most common differential diagnostic consideration for SC is ACC, and many cases previously diagnosed as papillary cystic ACC [16] have been reinterpreted as SC after performing the appropriate molecular studies. SC should be suspected in aspirates with cytomorphologic features resembling ACC but lacking the characteristic cytoplasmic basophilic PAS-positive diastase-resistant zymogen granules. In addition, SC tends to have more pronounced cytoplasmic vacuoles than ACC. Other clues to SC include the presence of papillary structures, particularly for tumors with these features from non-parotid sites. The cytologic differential diagnosis of SC also includes other salivary gland tumors with mucin such as low-grade MEC, or that are characterized by large eosinophilic cells such as Warthin tumor, oncocytoma, and oncocytic cystadenoma. The multivacuolated cells seen in SC (Fig. 7.9a, b) are not characteristic of any of these tumors and are among the most useful differentiating features. Similarly, the lack of squamous, intermediate cell, and goblet-­ type mucinous cells seen in low-grade MEC supports SC. Appropriate immunohistochemical and molecular studies can be used to confirm the diagnosis of SC (see Chap. 8). SC is immunohistochemically positive for S-100 protein, mammaglobin, and GATA-3 (Fig.  7.10a–c). SC is usually negative or at most focally positive for DOG1, and lacks reactivity for the myoepithelial markers calponin, CK5/6, and p63. Mucicarmine can be used to demonstrate intra- and extracellular mucin. In general, the cytomorphologic features of SC are not unique, and it is therefore important to ensure the availability of adequate material for ancillary studies if the diagnosis of SC is suspected.

126

a

S. Mehrotra et al.

b

c

Fig. 7.10 (a–c) Malignant. Secretory carcinoma. Cell block. Vacuolated pink cytoplasm is evident. H&E stain (a). Mammoglobin (b) and S-100 (c) show diffuse positive staining in tumor cells

Epithelial-Myoepithelial Carcinoma (EMC) Epithelial-myoepithelial carcinoma (EMC) is an uncommon low-grade malignancy. It accounts for 1% of salivary gland tumors and 2% of salivary gland malignancies [7, 19]. Approximately 75% of EMCs arise in the parotid gland, while the rest are equally distributed between the submandibular gland and the minor salivary glands. EMC is a disease of older individuals in the sixth to seventh decade, with a slight female predilection. Patients usually present with a localized slow-growing painless tumor [20]. EMC is a biphasic tumor with an inner layer of cuboidal ductal cells and an outer layer of larger clear myoepithelial cells. The ratio of myoepithelial to ductal cells shows marked variation but is usually 2:1 to 3:1. EMC is typically an indolent tumor with conservative surgery being the treatment of choice.

Cytologic Criteria EMC is comprised of variable proportions of ductal and myoepithelial cells, but the latter predominate (Fig. 7.11). Aspirates of EMC show the following: • Cellular aspirate • Arrangement of bland cells in pseudopapillary groups, sheets, and three-­ dimensional groups (Figs. 7.12 and 7.13)

7 Malignant Fig. 7.11 Malignant. Epithelial-myoepithelial carcinoma. The aspirate shows a biphasic tumor with inner cuboidal ductal cells and prominent outer myoepithelial cells (Papanicolaou stain)

Fig. 7.12 Malignant. Epithelial-myoepithelial carcinoma. Aspirate of epithelial-myoepithelial carcinoma showing biphasic cells organized in pseudopapillary tubules and sheets (Papanicolaou stain)

Fig. 7.13 Malignant. Epithelial-myoepithelial carcinoma. This aspirate of epithelial-­myoepithelial carcinoma has a prominent biphasic pattern of ductal cells and abundant pale myoepithelial cells as well as focal proteinaceous material (Diff-Quik stain)

127

128

S. Mehrotra et al.

Fig. 7.14 Malignant. Epithelial-myoepithelial carcinoma. This epithelialmyoepithelial carcinoma has prominent concentrically laminated proteinaceous secretions that should be distinguished from the matrix material of adenoid cystic carcinoma (Papanicolaou stain)

• • • •

Laminated, acellular stromal cores (Fig. 7.14) Predominant population of clear myoepithelial cells Minor population of cohesive ductal cells with scant cytoplasm Background-stripped nuclei

Explanatory Notes The predominant cell in aspirates of EMC is the myoepithelial cell which has bland nuclei with open chromatin and an unusually abundant clear or pale cytoplasm. Because of the bland nuclear features, EMC will often be classified as “Neoplasm: SUMP” or as “SM.” The delicate nature of the glycogen-rich cytoplasm results in fragile myoepithelial cells and frequently stripped nuclei in the background. The cuboidal ductal component is sometimes more difficult to identify. Concentrically laminated acellular stromal spheres often stain pink with Diff-Quik and green with Papanicolaou. The material should be collected for ancillary studies to highlight the biphasic nature of the tumor. Epithelial markers (Cytokeratins, EMA, CD117) highlight the ductal elements and myoepithelial markers (P63, smooth muscle actin, calponin, S-100) stain the myoepithelial cells. Several entities are included in the differential diagnosis with EMC such as AdCC, myoepithelioma, myoepithelial carcinoma, and cellular PA. However, none of the tumors in the differential diagnosis has the prominent population of large clear myoepithelial cells seen in EMC. Unlike EMC, AdCC is a basaloid neoplasm with matrix material lacking the laminated features seen in EMC. Of note, EMC commonly harbors HRAS mutations which are helpful for making a definitive diagnosis [21]. Myoepithelial carcinoma and myoepithelioma lack the biphasic pattern of EMC, and the cells are smaller with less abundant pale cytoplasm. EMC can be difficult to distinguish from a cellular PA; however, PA lacks the prominent biphasic pattern, large clear myoepithelial cells, atypia, and unique laminated matrix cores (see Chap. 5).

7 Malignant

129

Given the abundant clear cells in EMC, other tumors with clear cell features such as metastatic renal cell carcinoma (RCC) could also be considered; however, RCC lacks the biphasic pattern of EMC and has a distinct immunoprofile. Similarly, salivary gland clear cell carcinoma which occurs primarily in the minor salivary glands also lacks the biphasic pattern of EMC.  Careful correlation with clinical history coupled with judicious use of ancillary studies is necessary to arrive at a diagnosis of EMC.

High-Grade Carcinomas Salivary Duct Carcinoma (SDC) SDC is a high-grade malignant salivary gland tumor, initially described in 1968 by Kleinsasser, Klein, and Hubner as a tumor analogous to ductal carcinoma of the breast [7, 22–25]. SDC can arise de novo, but up to 80% of cases represent malignant transformation of an existing pleomorphic adenoma (CA-ex-PA). SDC constitutes approximately 10% of all malignant salivary gland tumors, occurs in older individuals with a peak incidence in the seventh decade, and is much more common in men. The parotid gland is the most common primary site (80%). SDC presents as a rapidly growing mass frequently with symptoms of facial nerve involvement. Tumors are usually large and have an infiltrative growth pattern with foci of intraductal carcinoma and necrosis. Several histologic variants have been reported, including papillary, micropapillary, mucin-rich, sarcomatoid, and oncocytic; however, these morphologies are typically associated with areas of classic SDC showing an apocrine carcinoma-like pattern [7]. Regional or distant metastases may already be present at the time of diagnosis, contributing to the poor prognosis of this tumor. The standard management for resectable tumors is radical surgery with ipsilateral neck dissection, followed by postoperative adjuvant radiotherapy.

Cytologic Criteria SDC is a high-grade malignant neoplasm resembling mammary ductal carcinoma and has the following cytologic characteristics: • Cellular aspirate. • Single cells, flat sheets, and three-dimensional crowded clusters with occasional papillary and cribriform patterns and with overtly malignant cytologic features (Fig. 7.15). • Medium to large polygonal cells with well-defined cell borders and abundant finely granular cytoplasm that may have oncocytic or apocrine features (Fig. 7.16a, b).

130

S. Mehrotra et al.

Fig. 7.15 Malignant. Salivary duct carcinoma. Aspirate is highly cellular with three-dimensional clusters and single discohesive cells in a dirty/ necrotic background (Diff-Quik stain)

a

b

Fig. 7.16 (a, b) Malignant. Salivary duct carcinoma. This aspirate of salivary duct carcinoma contains groups of high-grade malignant cells with abundant cytoplasm, nuclear pleomorphism, prominent nucleoli, and glandular features (Diff-Quik stain)

• Enlarged round to oval, pleomorphic nuclei with anisonucleosis, hyperchromasia, and prominent nucleolus (Figs. 7.17 and 7.18). • Mitoses are common. • Necrotic background; stripped enlarged nuclei may be present (Fig. 7.18).

Explanatory Notes SDC are easily recognized cytologically as high-grade carcinomas, but more specific classification will often require ancillary studies (Fig. 7.19a–d). SDCs are positive for androgen receptor (AR), whereas expression of ER or PR is rare [26]. In addition, GATA-3 is also positive in SDC, and >80% are also positive for GCDFP-15. Her2/neu expression is frequently seen, but diffuse strong membranous staining or HER2 amplification demonstrated by FISH is only seen in about 25% of cases. SDC typically has a high proliferation index of over 25% with Ki67/MIB1 staining.

7 Malignant

131

Fig. 7.17 Malignant. Salivary duct carcinoma. Aspirate shows a syncytial cluster of pleomorphic cells with prominent nucleoli and abundant granular “oncocytic appearing” cytoplasm (Papanicolaou stain)

Fig. 7.18 Malignant. Salivary duct carcinoma. The polygonal cells in this FNA of salivary duct carcinoma have large pleomorphic nuclei with distinct nucleoli and moderate amounts of cytoplasm (Papanicolaou stain)

The main differential diagnostic considerations for SDC include high-grade MEC, oncocytic carcinoma, and metastatic carcinoma from the breast, prostate, or lung. High-grade adenocarcinoma, NOS also enters the differential diagnosis, but this is a diagnosis of exclusion, which should only be made after other diagnostic considerations are excluded. Both MEC and SDC are composed of large pleomorphic epithelial cells, but SDC lacks the squamoid features of MEC, and intracellular mucin is uncommon in SDC. Keratinization is absent in SDC and would favor metastatic squamous carcinoma. Oncocytic carcinomas lack the necrosis and ductal features seen in SDC, and they are immunophenotypically distinct. Metastatic carcinoma from the breast or prostate enters the differential diagnosis, particularly in a patient with a known history of these cancers. Clinical correlation and interpretation of the cytologic findings in the appropriate clinical context is essential for the diagnosis of high-grade primary and secondary salivary gland cancers. A focused panel of immunohistochemical markers can usually resolve difficult cases where the cytomorphologic findings are not definitive (See Table 7.2).

132

a

c

S. Mehrotra et al.

b

d

Fig. 7.19 (a–d) Malignant. Salivary duct carcinoma. (a) Cell block section shows a cluster of tumor cells with nuclear pleomorphism, defined cellular borders, relatively abundant granular cytoplasm, and nuclei with prominent nucleoli. Note the mitotic figure in the upper right corner. (b) Tumor cells show diffuse immunoreactivity with GCDFP-15. (c) Her2neu immunostain showing strong membranous staining in tumor cells. (d) AR exhibits strong diffuse positive nuclear staining (H&E stain) Table 7.2  Immunohistochemical stains in the differential diagnosis of high-grade salivary gland malignancies SMA, p63/ SMMHC, p40 calponin MEC (HG) + − SQCCa + − SDC − − PDC, NE −/+ − METASTATIC − −

SYN, CHROMO, CD56, CK8/18 CK5/6 CK20 MUCIN AR CD57 Focal + + − + − − −/+ + − − − − + − − − + − + (dot) − + − − + + −/+ −/+ −/+ − −

Site-­ specific − − − − +b

AR Androgen receptor, CHROMO chromogranin, MEC (HG) Mucoepidermoid carcinoma, high grade, PDC, NE Poorly differentiated carcinoma neuroendocrine type, SMA smooth muscle actin, SMMHC smooth muscle myosin heavy chain, SDC Salivary Duct Carcinoma, SQCC Squamous cell carcinoma, SYN synaptophysin a  SQCC includes primary and metastatic squamous cell carcinoma, as well as lymphoepithelial carcinoma b  TTF1 for lung/thyroid primaries; CDX2 for colorectal primaries, PAX8 for renal primaries; HMB45/MART1 for melanomas

7 Malignant

133

Lymphoepithelial Carcinoma (LEC) LEC is a rare salivary gland tumor which comprises 50%. The most common differential diagnosis is with metastatic small cell carcinoma, either cutaneous Merkel cell carcinoma or small cell carcinoma from the lung or other anatomic sites. Less often, the differential diagnosis will include other high-grade carcinomas with basaloid features or small round blue cell tumors. A combination of ancillary marker studies and clinical correlation is usually sufficient to resolve the differential diagnosis. Primary large cell neuroendocrine carcinomas are rare tumors predominantly affecting the parotid glands with cytomorphologic features similar to those described in the lungs (Fig. 7.27). Most published cases are based on single case reports and small case series. Diagnosis requires adequate sampling with ancillary studies [35].

138

S. Mehrotra et al.

Fig. 7.27 Malignant. Large cell neuroendocrine carcinoma. Rosette-like arrangement of tumor cells with fine chromatin and prominent nucleoli. Rare nuclear streak is seen (Papanicolaou stain)

Cancers with Indeterminate or Multiple Grades Mucoepidermoid Carcinoma (MEC) Mucoepidermoid carcinoma (MEC) is the most common primary salivary gland malignancy in both adults and children, with a peak incidence in the second decade of life [7, 36, 37]. MEC occur most often in the parotid gland, followed by intraoral minor salivary glands, especially those of the palate. Low-grade tumors vary in size from 1 to 3  cm and are usually circumscribed. MEC is variably solid and cystic depending upon histologic grade. MEC is graded according to a three-tiered system as low, intermediate, and high-­ grade. The histopathologic grading systems in current use rely on some features that are difficult to appreciate in cytologic samples, such as perineural invasion, lymphovascular invasion, and pattern of invasion, but also include features that can be assessed in cytologic preparations, such as the proportion of solid vs. cystic (mucinous) components, presence of necrosis, anaplasia, and mitoses. Using the relative amounts of tumor cells and mucin, as well as the presence or absence of high-grade cytologic features including necrosis, mitoses, and nuclear pleomorphism, MEC can often be graded as low-grade versus high-grade in FNA samples. While lowand intermediate-grade MEC can often be adequately treated by complete surgical excision, high-grade MEC may require adjuvant therapy in addition to surgery. The 10-year survival rates for low, intermediate, and high-grade tumors is approximately 90%, 70%, and 25%, respectively [7].

Cytologic Criteria MEC is a malignant epithelial neoplasm characterized by epidermoid, intermediate, and goblet-type mucus cells, which vary in proportion depending upon the histologic grade. In addition, MEC can have columnar, clear, and oncocytic features.

7 Malignant

139

• Cellularity is variable depending upon the grade of the tumor. • Admixture of goblet-type mucus cells, intermediate, and epidermoid cells (Fig. 7.28). • Low-grade tumors contain more mucinous cells (Figs. 7.28, 7.29, 7.30 and 7.31) and high-grade tumors have a predominance of epidermoid cells (Figs.  7.32 and 7.33). • Variable nuclear atypia from mild (low-grade) to markedly atypical (high-grade). • Variable presence of oncocytic cells, clear cells, and columnar cells. • Cystic background with abundant extracellular mucin in low- and intermediate-­ grade tumors. • Background lymphocytes are present in approximately 20% of cases. • Keratinization is not a feature of MEC. Fig. 7.28 Malignant. Mucoepidermoid carcinoma, low-grade. The aspirate contains abundant mucin in the background and lose sheets of bland epidermoid and mucinous cells (smear, Papanicolaou stain). (Courtesy of William Geddie, MD, Laboratory Medicine & Pathobiology, University of Toronto, Toronto, Canada)

Fig. 7.29 Malignant. Mucoepidermoid carcinoma, low-grade. This aspirate of low-grade mucoepidermoid carcinoma contains bland epidermoid cells with moderate amounts of dense cytoplasm and well-defined cell borders, while mucus cells contain abundant delicate pink mucinous cytoplasm (smear, Papanicolaou stain). (Courtesy of William Geddie, MD, Laboratory Medicine & Pathobiology, University of Toronto, Toronto, Canada)

140 Fig. 7.30 Malignant. Mucoepidermoid carcinoma, low-grade. This aspirate of low-grade mucoepidermoid carcinoma has occasional mucus cells with a large cytoplasmic vacuole indenting the nucleus and occupying the majority of cytoplasm with a central pink mucin droplet (smear, Papanicolaou stain)

Fig. 7.31 Malignant. Mucoepidermoid carcinoma, low- to intermediate-grade. The aspirate shows a sheet of tumor cells, predominantly composed of intermediate cells with rare, interspersed mucus/goblet cells (Papanicolaou stain)

Fig. 7.32 Malignant. Mucoepidermoid carcinoma, high-grade. Aspirates show a cluster of highly pleomorphic cells with dense cytoplasm and rare interspersed glandular cells with intracytoplasmic mucin at the lower right corner. Pink material farthest to the right of the image likely represents thick mucin (Diff-Quik stain)

S. Mehrotra et al.

7 Malignant

141

Explanatory Notes The cytologic features of MEC are dependent on the grade of the tumor. Low-grade MEC usually shows abundant background mucin, cyst debris, and a few scattered bland epidermoid cells. Low-grade MEC is among the most common causes of a false negative salivary gland FNA with some cases being diagnosed as a mucocele or benign cyst contents. If only cyst fluid is obtained, the aspirate should be classified as Non-Diagnostic (see Chap. 2); however, the presence of mucinous cyst contents should prompt the diagnosis of AUS (see Chap. 4). An effort should be made to aspirate any residual solid areas of a cystic salivary gland lesion. The epidermoid cells of low- and intermediate-grade MEC occur in bland cohesive but crowded sheets with well-defined intercellular borders and dense cytoplasm. Intermediate cells are columnar to polygonal, occur in flat cohesive sheets, and have a higher N:C ratio than epidermoid cells. Goblet-type mucus cells have abundant vacuolated cytoplasm, low N:C ratio, indented eccentrically placed nuclei, and can occur singly, admixed within a sheet of epidermoid cells, or in clusters. The presence of other cell types including oncocytic cells, clear cells, and columnar cells can create a diagnostic challenge. Obtaining an adequate sample combined with ancillary studies can help. Keratinization is not a feature of MEC and if present should raise suspicion of metastatic squamous cell carcinoma or an adenosquamous carcinoma. Approximately 20% of MECs have abundant background lymphocytes, which together with oncocytic cells and cystic debris can be misinterpreted as Warthin tumor. This is particularly challenging since a subset of Warthin tumors can exhibit squamous metaplasia and/or have a mucoid background. High-grade MEC yields cellular aspirates with a predominance of markedly atypical epidermoid cells in crowded sheets and clusters. Overt malignant nuclear features are present resembling a high-grade squamous cell carcinoma. A careful search for rare, interspersed goblet cells is a clue to the diagnosis of high-grade MEC (Figs. 7.32 and 7.33). The differential diagnosis includes other high-grade carcinomas such as Fig. 7.33 Malignant. Mucoepidermoid carcinoma, high-grade. FNA of high-grade mucoepidermoid carcinoma with markedly atypical epidermoid cells and occasional interspersed mucus cells (Diff-Quik stain)

142

S. Mehrotra et al.

SDC, CA-ex-PA, primary squamous cell carcinoma, and metastatic carcinomas. SDC can be distinguished immunohistochemically using a panel that includes AR, GATA-3, and p63/p40. MEC is positive for p63/p40 and negative for AR and GATA-3. The most common differential diagnosis of high-grade MEC is squamous cell carcinoma. Primary squamous cell carcinoma of the salivary gland is rare; most cases of salivary gland squamous cell carcinoma represent metastases to intra- or peri-glandular lymph nodes from a cutaneous head and neck primary. A preceding history of head and neck cutaneous squamous cell carcinoma and the lack of mucin-positive epithelial cells can help in the differential diagnosis. The rare adenosquamous carcinoma cannot be reliably differentiated from high-grade MEC based on cytologic criteria alone. However, most adenosquamous carcinomas arise in minor glands of the upper aerodigestive tract and do not affect the major salivary glands. Most MECs are characterized by a t(11;19) translocation with CRTC1::MAML2 gene fusion with a smaller percentage harboring a t(11;15) translocation with CRTC3::MAML2 fusion [7]. PCR-based or FISH studies can be very useful for definitive classification (see Chap. 8 for a detailed discussion).

Adenoid Cystic Carcinoma (AdCC) Adenoid cystic carcinoma (AdCC) is a primary salivary gland malignancy, accounting for 30% solid area). Similar to ACC, AdCC can undergo high-grade transformation [37, 38].

Cytologic Criteria AdCC is a malignant matrix-producing basaloid tumor consisting of epithelial and myoepithelial cells in various morphologic configurations including tubular, cribriform, and solid patterns (Fig.  7.34a–c) and a propensity for perineural invasion. Aspirates of AdCC are characterized by: • Variably cellular aspirate. • Cohesive groups of basaloid cells arranged in small syncytial sheets with irregular borders, sometimes showing microcystic sieve-like spaces, clusters, “cylinders,” and tubules (Fig. 7.34a–c).

7 Malignant

a

143

b

c

Fig. 7.34 (a–c) Malignant. Adenoid cystic carcinoma. Aspirates show small high N:C ratio basaloid tumor cells surrounding acellular matrix with different architectural patterns (a) tubular pattern (b) cribriform pattern and (c) solid patterns (Diff-Quik stains)

• • • •

Uniform, small basaloid tumor cells with a high N:C ratio (Fig. 7.35). Scant, indistinct cytoplasm. Bland, oval to angulated hyperchromatic nuclei with indistinct nucleoli. Mitoses and necrosis are uncommon in the absence of high-grade transformation. • Acellular homogenous matrix with sharp borders, best seen on Romanowsky-­ type stains (magenta); matrix is translucent and therefore less well visualized in Papanicolaou-stained smears, and may be absent or sparse in the solid variant (Figs. 7.36a–b and 7.37).

Explanatory Notes The cribriform subtype of AdCC is the most common and easiest to recognize in salivary gland aspirates. The cells are monotonous, small, and basaloid and are arranged most often in a sheet or tubular pattern. Nuclei are dark and angulated, and the cells have scant indistinct cytoplasm. Mitoses, necrosis, and significant

144

S. Mehrotra et al.

Fig. 7.35 Malignant. Adenoid cystic carcinoma. Aspirate shows monotonous basaloid tumor cells, with high N:C ratio, some of which are surrounding basement membrane-like material (Papanicolaou stain)

a

b

Fig. 7.36 (a, b) Malignant. Adenoid cystic carcinoma. This FNA of Adenoid cystic carcinoma shows abundant acellular homogenous matrix with sharp borders, basaloid tumor cells often form a syncytial smear surrounding the matrix material (smear Diff-Quik and cell block H&E stain) Fig. 7.37 Malignant. Adenoid cystic carcinoma. This Papanicolaou-stained smear shows cribriform architecture with pale translucent matrix material

7 Malignant

145

pleomorphism are typically absent. The most important cytologic feature of AdCC is its characteristic homogenous, acellular, non-fibrillary, and intensely metachromatic matrix, that appears magenta in Romanowsky-type stains. The matrix takes the form of variably sized spheres, cylinders, and branching tubules with sharp edges. The matrix is pale green and translucent and is often difficult to visualize on Papanicolaou-stained preparations. The differential diagnosis of AdCC includes several benign and malignant entities. In contrast to PA, in which the neoplastic cells are embedded in the matrix, the basaloid cells of AdCC surround the matrix with a sharp interface between the cells and matrix. In addition, the matrix of AdCC typically lacks the fibrillary qualities and frayed edges seen in PA. The solid subtype of AdCC is the most difficult to recognize in cytologic specimens. It is composed of sheets of basaloid cells with little or no matrix (Figs. 7.38a, b). This subtype may have larger and less monotonous nuclei, overlapping nuclei, visible nucleoli, occasional mitoses, apoptotic bodies, and focal necrosis. These features make the specific diagnosis of the solid subtype of AdCC very challenging. Ancillary studies and a careful search for telltale signs of acellular matrix globules can sometimes be helpful. Hyaline globules are not specific for AdCC and can be encountered in several other matrix-producing tumors including polymorphous adenocarcinoma, basal cell adenoma, basal cell adenocarcinoma, epithelial-myoepithelial carcinoma, and even basaloid squamous cell carcinoma. AdCC lacks squamous differentiation, which is usually present at least focally in basaloid SCC.  In addition, basaloid SCC characteristically has higher grade features than AdCC including apoptotic bodies, mitotic figures, conspicuous necrosis, and severe cytologic atypia. Polymorphous adenocarcinoma enters into the differential diagnosis for aspirates of minor salivary gland lesions, especially in the palate. In contrast to AdCC, the cells of polymorphous adenocarcinoma are not basaloid. They are a uniform population of polygonal and mediumsize cells with moderate amounts of cytoplasm, nuclei with open chromatin, and small distinct nucleoli. The differential diagnosis of AdCC with basal cell adenoma and adenocarcinoma is among the most challenging in salivary gland cytology (see also Chap. 5). a

b

Fig. 7.38 (a, b) Malignant. Adenoid cystic carcinoma. FNA of solid subtype of adenoid cystic carcinoma containing sheets of basaloid tumor cells with no matrix and large and monotonous nuclei with scant cytoplasm (Papanicolaou stain)

146

S. Mehrotra et al.

The arrangement of the extracellular basement membrane-like material in some basal cell tumors can be an important clue to distinguishing them from AdCC. This is especially true for the membranous subtype of basal cell tumor. However, the overlapping cytomorphologic features in some cases will lead to a diagnosis of “Neplasm: SUMP” or “SM.” Epithelial-myoepithelial carcinoma is recognized based primarily on its abundance of large clear myoepithelial cells which is not a feature seen in AdCC [19]. Obtaining material for ancillary studies can be very helpful for confirming the diagnosis of AdCC. A majority of AdCC is strongly positive for CD117 (c-kit, cytoplasmic), but no mutation in the c-kit gene has been identified. In addition, most AdCC show nuclear overexpression of MYB and cytoplasmic expression of NOTCH (especially in the solid subtype). A majority of AdCC have a signature chromosomal translocation t(6;9)(q22–23;p23–24) resulting in a fusion involving the v-myb myeloblastosis viral oncogene homolog (MYB) and the transcription factor gene NFIB which can be detected by FISH analysis or next-generation sequencing (see Chap. 8).

Myoepithelial Carcinoma (MC) MC is rare, accounting for 2× the size of normal lymphocyte) that has a diffuse growth pattern. Aspirates show the following: a

b

Fig. 7.47 (a, b) Malignant. Extranodal marginal zone B cell lymphoma. The aspirate contains a dispersed population of small to intermediate-sized lymphocytes with small amounts of preserved cytoplasm, coarse chromatin, and round to irregular nuclei. Scattered large centroblasts are also seen (Papanicolaou stain)

7 Malignant

153

• Cellular aspirate • Large atypical lymphoid cells (2–3 times the size of mature lymphocytes) (Fig. 7.48) • Distinct to large nucleolus often present • Lymphoglandular bodies in the background • Tingible body macrophages may be present • Immunocytochemistry shows CD19+, CD20+, CD45+, CD22+, CD79a+, Ki67-high • Flow cytometry: CD5−/CD19+, CD19+/FMC7−, CD19+/CD23−, CD10−/+, κ or λ light chain restriction

Explanatory Notes The differential diagnosis of lymphoproliferative lesions includes both reactive and neoplastic conditions of salivary glands accompanied by a prominent lymphoid reaction. The list of reactive conditions includes chronic sialadenitis, lymphoepithelial sialadenitis (LESA), HIV-associated lymphoepithelial cysts, and, most importantly, reactive lymph nodes. Chronic sialadenitis usually yields a paucicellular aspirate with rare groups of ductal cells and a few small matureappearing B and T lymphocytes which are polyclonal by flow cytometry. In contrast, aspirates of lymphomas are typically cellular and include an abundance of background lymphoglandular bodies. In addition to EMZL and DLBCL, peri- and intra-parotid lymph nodes can occasionally be involved by other lymphomas such as mantle cell lymphoma and follicular lymphoma (Fig. 7.49a–b). Aspirates of the latter are usually suggestive of lymphoma, but ancillary studies are required for accurate subclassification. One of the most difficult diagnostic problems when evaluating a lymphoid lesion of the salivary gland is the cytologic distinction Fig. 7.48 Malignant. Diffuse large B cell lymphoma. Aspirate containing a dispersed population of large lymphoid cells >3 times the size of a small mature lymphocyte (Diff-Quik stain)

154

a

S. Mehrotra et al.

b

Fig. 7.49 (a, b) Malignant. Mantle cell lymphoma (a) and follicular lymphoma (b) show very atypical cytomorphologic features suggestive of lymphoma but ancillary studies are needed for accurate subclassification of the lymphomas (smear, Diff-Quik stain). (Courtesy of William Geddie, MD, Laboratory Medicine & Pathobiology, University of Toronto, Toronto, Canada)

between LESA and a low-­grade lymphoma such as EMZL (see also Chap. 3). The two entities have overlapping cytomorphologic features that include a heterogeneous population of cells composed of abundant mixed but predominantly small lymphocytes, tingible body macrophages, follicular dendritic cells, plasma cells, and lymphohistiocytic aggregates. Cytomorphology alone cannot reliably distinguish between these two entities, and flow cytometry or some other means of immunophenotypic analysis is necessary to make the distinction and would be needed prior to classifying an aspirate as lymphoma. It is therefore essential to collect material for ancillary studies including flow cytometry. Consultation with a pathologist having subspecialty experience in hematopathology is also recommended. Aspirates of DLBCL are usually readily recognized cytomorphologically as malignant, but in some cases, the differential diagnosis will also include other malignant neoplasms composed of small cells such as melanoma, small cell carcinoma, and certain sarcomas. The recognition of lymphoglandular bodies in the background can provide a clue to the diagnosis. Obtaining material for ancillary studies including a directed immunocytochemical panel for cytokeratin, CD45, CD20, and S-100 among others can be used in difficult cases. Flow cytometry can be informative as well; however, caution is warranted in the interpretation of flow cytometry since a significant subset of DLBCLs can yield a benign flow cytometry result.

Selected Tumors of Minor Salivary Glands In contrast to tumors of the major salivary glands, tumors of minor salivary glands (MISG) are subjected to FNA much less often. Core biopsies are sometimes used, but definitive diagnosis often requires resection and histologic evaluation.

7 Malignant

155

Polymorphous adenocarcinoma (PAC) comprises 41% of MISG carcinomas. The recently described entity cribriform adenocarcinoma of minor salivary glands (CAMSG) is included under PAC in the recent WHO classification of salivary gland tumors. The cytologic experience with the tumor is limited with few reported series [7].

Polymorphous Adenocarcinoma (PAC) PAC is defined as a malignant epithelial tumor characterized by cytologic uniformity, morphologic diversity, and infiltrative growth. PAC is the second most common intraoral malignancy accounting for 26% of all carcinomas at this site. PAC predominantly affects females (M:F is 1:2) with a mean age of 59  years (range 16–94 years). The palate is the most frequent intraoral site involved. Tumors are usually firm, painless, and of variable duration with an average size of 2.1  cm. Complete surgical resection with or without lymph node dissection is the treatment of choice. The overall 5-year survival is good with local recurrence in 10–33% of cases. Regional lymph node metastasis varies from 9 to 16%. Though rare, high-­ grade transformation (HGT) has been described [7, 45].

Cytologic Criteria PAC exhibits a ductal phenotype with cellular monotony and architectural heterogeneity [46–48] (Fig. 7.50). • Cellular smears. • Cohesive three-dimensional clusters with irregular shapes, branching papillary clusters, acinar arrangement, and flat sheets (Fig. 7.51). Fig. 7.50 Malignant. Polymorphous adenocarcinoma. Cellular smear with proliferation of uniform cells in short slightly twisted cords. Bare nuclei are common (Papanicolaou stain)

156

S. Mehrotra et al.

• Scant to moderate metachromatic matrix dispersed as defined stromal globules. • Stromal globules are small, usually 10–50 μm, regular and embedded in epithelial aggregates (Fig. 7.52). • Tumor cells are small with scant to moderate cytoplasm and round to ovoid monotonous nuclei (Fig. 7.53). • Papanicolaou-stained smears show vesicular chromatin with inconspicuous nucleoli. • Occasional nuclear grooves or even intranuclear cytoplasmic inclusions may be seen (Fig. 7.54). • Mitoses and necrosis are absent. Fig. 7.51 Malignant. Polymorphous adenocarcinoma. This aspirate smear exhibits a branching papillary architecture with stromal material dispersed as small defined globules. Note the cellular monotony (Diff-Quik stain)

Fig. 7.52 Malignant. Polymorphous adenocarcinoma. This aspirate of polymorphous adenocarcinoma shows cohesive three-dimensional clusters of cells with stromal material embedded in epithelial clusters (Papanicolaou stain)

7 Malignant

157

Fig. 7.53 Malignant. Polymorphous adenocarcinoma. This aspirate of polymorphous adenocarcinoma shows small tumor cells with scant cytoplasm and round to oval relatively monotonous nuclei (Papanicolaou stain)

Fig. 7.54 Malignant. Polymorphous adenocarcinoma. Nuclei of polymorphous adenocarcinoma are ovoid with vesicular chromatin and inconspicuous nucleoli. Nuclear grooves may be seen (Papanicolaou stain)

Explanatory Notes PAC exhibits consistent cytologic features which lack specificity for a definitive diagnosis. Ancillary studies including immunocytochemistry may be helpful in the presence of ample cellular material in a cell block. PAC shows consistent positive staining for S-100 with a discordant p63+/p40− staining pattern. Most PACs exhibit diffuse p63 expression with p40 being negative. Specific mutation analysis for PRKD1 gene may be performed when available. Protein kinase D1 E710D hotspot mutation is reportedly highly specific for polymorphous adenocarcinoma [49] (see Chap. 8).

158

S. Mehrotra et al.

Several benign and malignant entities enter the differential diagnosis with PAC including AdCC, basal cell adenoma/adenocarcinoma, and cellular PA. The spherules of AdCC are quite large and have a hyaline aspect compared to PAC which harbors smaller spherules of uniform size. The stromal spherules of PAC are usually embedded within cellular clusters and unlikely to be present as separate fragments. The individual cells of PAC have moderate appreciable cytoplasm and nuclei with open chromatin. In contrast, the cells of AdCC are distinctly basaloid with minimal cytoplasm and hyperchromatic nuclei. Most AdCC show consistent positive staining with smooth muscle actin and exhibit concordant p63/p40 staining. Immunocytochemical staining with MYB or FISH for the myb rearrangement may help in a definitive diagnosis. The differential diagnosis between PAC and basal cell adenoma/adenocarcinoma is even more difficult, requiring ancillary studies. Of note, basal cell adenoma/ adenocarcinoma is usually seen in the major salivary glands. Cytologic smears of cellular PA may show a classic fibrillary chondromyxoid matrix with associated myoepithelial cells which is not seen in PAC. Immunocytochemical staining for PLAG1 or HMGA2, if available can be used to support a diagnosis of PA. Given the significant cytomorphologic overlap between PAC and other matrix-producing minor salivary gland tumors, they will most often be classified as “Neoplasm: SUMP” or “SM.”

Secondary Malignant Tumors Metastatic tumors constitute approximately 7.5% of all non-hematolymphoid malignant salivary gland neoplasms, and most cases present as a solitary salivary gland mass [50]. In the majority of cases, there is a known history of non-salivary gland primary cancer. The parotid gland, in particular intra-parotid and peri-parotid lymph nodes, is involved 20 times more often than the submandibular gland in metastatic disease. The peak incidence of a secondary malignant salivary gland tumor is in the seventh to eigth decade with almost 70% occurring in men. Eighty percent of the metastatic tumors to the parotid gland are from head and neck sites, especially cutaneous carcinomas of the face and anterior scalp, while 85% of metastatic tumors in the submandibular gland are from the oral cavity or distant sites [27]. Cutaneous squamous cell carcinoma is the most commonly diagnosed secondary tumor of the parotid gland followed by melanoma. Secondary salivary gland tumors from distant sites include those from lung, breast, and kidney.

Cytologic Criteria Aspirates of secondary cancers involving the salivary gland exhibit the following cytologic features • Cellular aspirate • Usually high-grade nuclear features

7 Malignant

159

• The cytomorphologic characteristics depend upon the tumor type; most common metastases are squamous cell carcinoma, melanoma, Merkel cell carcinoma, or cancers from distant sites (lung, breast, and kidney) (Figs. 7.55a–b, 7.56 and 7.57)

Explanatory Notes Squamous cell carcinoma is the most common tumor metastatic to salivary glands. Aspirates are cellular and include atypical squamous cells, keratin debris, and necrosis. In some cases, there may be a cystic background. In contrast to MEC, metastatic squamous cell carcinomas lack evidence of intracellular mucin and are usually keratinizing. Most cases occur in older patients with a known history of a cutaneous squamous cell carcinoma; primary squamous cell carcinoma of the salivary glands is very rare. a

b

Fig. 7.55 (a, b) Malignant. Metastatic melanoma. The aspirate smears of metastatic melanoma show the characteristic dyshesive pattern of pleomorphic cells as well as background melanophages containing melanin pigment which appears blue-black in Romanowsky stains and dusky brown in Papanicolaou stains (smear, Diff-Quik, and Papanicolaou stains) Fig. 7.56 Malignant. Metastatic squamous cell carcinoma. The cellular aspirate shows high N:C ratio cells as well as dyskeratotic orangeophilic cells in a necrotic background (Papanicolaou stain)

160

S. Mehrotra et al.

Fig. 7.57 Malignant. Metastatic Merkel cell carcinoma. A monotonous malignant cell population of “small” cells with rounded/angulated nuclei arranged in solid sheets with nuclear streaking artifact, individual cell necrosis, and coarsely granular nucleoplasm. This image is indistinguishable from that of small cell neuroendocrine carcinoma (Papanicolaou stain)

Aspirates of metastatic melanoma have a very wide range of cytomorphologic appearances. The classic FNA of metastatic melanoma shows a population of dyshesive pleomorphic cells with eccentric nuclei, prominent nucleoli, and fine-to-­ coarsely granular cytoplasmic melanin pigment (blue-green in Romanowsky-type stains and brown in Papanicolaou stain). Intranuclear inclusions and mirror-image binucleation are also a common finding in metastatic melanoma. Amelanotic melanoma or spindle cell melanomas in the absence of a prior history can be misinterpreted as poorly differentiated carcinoma or sarcoma. Material for ancillary studies is key for any case where the cytomorphology does not match that of a primary salivary gland tumor, or for cases where there is a history of a non-salivary gland primary cancer. This is especially true for cases where the patient has a history of melanoma.

Malignant Mesenchymal Tumors Primary salivary gland soft tissue tumors are rare, with benign tumors being more common than malignant ones. A wide variety of soft tissue tumors can involve the parotid gland with benign vascular neoplasms (hemangiomas) being the most frequent [2, 7]. The reader is referred to other sources for a detailed description of soft tissue tumor cytology.

Clinical Management A definitive classification of a specific malignant salivary gland tumor including grade provides important information for clinical decision-making (see Chap. 10). The grade of cancer will often be useful to the clinician in determining the extent of

7 Malignant

161

surgery including the need to perform a neck dissection, and the potential need to sacrifice a large nerve. For high-grade salivary gland cancers involving the deep lobe of the parotid gland, a total parotidectomy would be necessary. In addition, identifying a cancer as primary versus metastatic would also have implications for the treating clinician. Patients with metastatic disease to parotid gland lymph nodes often require a concurrent neck dissection. If a lesion is metastatic from a non-­ cutaneous source, PET-CT may be indicated to locate a primary site of origin.

Sample Reports Example 7.1 Satisfactory for evaluation. MALIGNANT. Keratinizing squamous cell carcinoma. See note. Note: Since primary squamous cell carcinomas of salivary glands are exceedingly rare, a comprehensive clinical examination including a detailed history and skin examination should be performed; a metastasis from a cutaneous or mucosal head and neck site is favored. Example 7.2 Satisfactory for evaluation. MALIGNANT. High-grade carcinoma, consistent with salivary duct carcinoma. See note. Note: The aspirate is cellular and shows high-grade pleomorphic cells arranged in cribriform and papillary groupings with prominent nucleoli and background necrosis. Immunocytochemical stains performed on the corresponding cell block material are positive for androgen receptor (AR), GATA-3, and Her2/neu. Example 7.3 Satisfactory for evaluation. MALIGNANT. High-grade carcinoma. See note. Note: The aspirate is cellular and shows pleomorphic cells arranged in cribriform and papillary groupings with prominent nucleoli and background necrosis. The cytomorphologic findings are suggestive of salivary duct carcinoma; however, ancillary testing could not be performed due to a lack of adequate cellularity in the corresponding cell block sections. Example 7.4 Satisfactory for evaluation. MALIGNANT. Adenoid cystic carcinoma. See note. Note: The aspirate is cellular and shows basaloid cells with scant cytoplasm, and angulated dark nuclei arranged around homogenous, magenta-colored matrix

162

S. Mehrotra et al.

spheres. A FISH study confirmed a MYB (6q23) gene rearrangement, supporting the diagnosis of adenoid cystic carcinoma. Example 7.5 Satisfactory for evaluation. MALIGNANT. High-grade carcinoma ex pleomorphic adenoma. See note. Note: The aspirate shows high-grade pleomorphic tumor cells with prominent nucleoli, anisonucleosis, and rare mitotic figures. Separate foci of bland cells embedded within chondromyxoid matrix are also present. The patient’s history of pleomorphic adenoma at the same site is noted. The overall findings are consistent with a high-grade carcinoma ex pleomorphic adenoma. Example 7.6 Satisfactory for evaluation. MALIGNANT. High-grade carcinoma. See note. Note: The aspirate shows high-grade pleomorphic tumor cells with prominent nucleoli, anisonucleosis, and rare mitotic figures with scant chondromyxoid matrix in one slide. In the context of the patient’s history of a long-standing mass with recent rapid increase in size, a diagnosis of carcinoma ex pleomorphic adenoma is favored. Example 7.7 Satisfactory for evaluation. MALIGNANT. Low-grade mucoepidermoid carcinoma. See note. Note: The aspirate smears and cell block sections show abundant background mucin with clusters of epithelioid and goblet-type mucinous cells, consistent with low-grade mucoepidermoid carcinoma.

References 1. Al-Abbadi MA. Salivary gland cytology: a color atlas. New York: Wiley-Blackwell; 2011. 2. Faquin WC, Powers CN. Salivary gland cytopathology. In: Essentials in cytopathology, vol. 5. New York: Springer Science+ Business Media; 2008. 3. Klijanienko J, Vielh P. Salivary gland tumours. In: Monographs in clinical cytology, vol. 15. Basel: Karger; 2000. 4. Wang H, Fundakowski C, Khurana JS, Jhala N. Fine-needle aspiration biopsy of salivary gland lesions. Arch Pathol Lab Med. 2015;139(12):1491–7. 5. Mayer M, Thoelken R, Jering M, Märkl B, Zenk J. Metastases of cutaneous squamous cell carcinoma seem to be the most frequent malignancies in the parotid gland: a hospital-based study from a salivary gland center. Head Neck Pathol. 2021;15(3):843–51. 6. Boerner SL. Patterns and pitfalls in fine needle aspiration of salivary gland. Curr Diagn Pathol. 2003;9:339–54. 7. El-Naggar AK, Chan JKC, Grandis JR, Takata T, Slootweg PJ. WHO classification of head and neck tumours. Lyon: IARC; 2017.

7 Malignant

163

8. Seethala RR.  An update on grading of salivary gland carcinomas. Head Neck Pathol. 2009;3(1):69–77. 9. Vander Poorten V, Triantafyllou A, Thompson LD, et al. Salivary acinic cell carcinoma: reappraisal and update. Eur Arch Otorhinolaryngol. 2016;273(11):3511–31. 10. Johnson DN, Onenerk M, Krane JF, et al. Cytologic grading of primary malignant salivary gland tumors: a blinded review by an international panel. Cancer Cytopathol. 2020;128(6):392–402. 11. Klijanienko J, Vielh P. Fine-needle sample of salivary gland lesions. V: Cytology of 22 cases of acinic cell carcinoma with histologic correlation. Diagn Cytopathol. 1997;17(5):347–52. 12. Wakely PE Jr, Lott-Limbach AA. Cytopathology of acinic cell carcinoma: a study of 50 cases, including 9 with high-grade transformation. Cancer Cytopathol. 2021;129(12):973–83. 13. Wong KS, Mariño-Enriquez A, Hornick JL, Jo VY. NR4A3 immunohistochemistry reliably discriminates acinic cell carcinoma from mimics. Head Neck Pathol. 2021;15(2):425–32. 14. Skalova A, Michal M, Simpson RH.  Newly described salivary gland tumors. Mod Pathol. 2017;30(s1):S27–43. 15. Bishop JA. Mammary analog secretory carcinoma of salivary glands: review of a new entity with an emphasis on differential diagnosis. Pathol Case Rev. 2015;20(1):7–12. 16. Bishop JA, Yonescu R, Batista DA, Westra WH, Ali SZ. Cytopathologic features of mammary analogue secretory carcinoma. Cancer Cytopathol. 2013;121(5):228–33. 17. Petersson F. High-grade transformation (“dedifferentiation”)—malignant progression of salivary gland neoplasms, including carcinoma ex pleomorphic adenoma: a review. Pathol Case Rev. 2015;20(1):27–37. 18. Nagao T. “Dedifferentiation” and high-grade transformation in salivary gland carcinomas. Head Neck Pathol. 2013;7(Suppl 1):S37–47. 19. Seethala RR, Barnes EL, Hunt JL. Epithelial-myoepithelial carcinoma: a review of the clinicopathologic spectrum and immunophenotypic characteristics in 61 tumors of the salivary glands and upper aerodigestive tract. Am J Surg Pathol. 2007;31(1):44–57. 20. Gore MR. Epithelial-myoepithelial carcinoma: a population-based survival analysis. BMC Ear Nose Throat Disord. 2018;18:15. 21. Urano M, Nakaguro M, Yamamoto Y, et  al. Diagnostic significance of HRAS mutations in epithelial-myoepithelial carcinomas exhibiting a broad histopathologic spectrum. Am J Surg Pathol. 2019;43(7):984–94. 22. Kleinsasser O, Klein HJ, Hübner G.  Salivary duct carcinoma. A group of salivary gland tumors analogous to mammary duct carcinoma. Arch Klin Exp Ohren Nasen Kehlkopfheilkd. 1968;192(1):100–5. 23. Chiosea SI, Thompson LD, Weinreb I, et al. Subsets of salivary duct carcinoma defined by morphologic evidence of pleomorphic adenoma, PLAG1 or HMGA2 rearrangements, and common genetic alterations. Cancer. 2016;122(20):3136–44. 24. Williams L, Thompson LD, Seethala RR, et al. Salivary duct carcinoma: the predominance of apocrine morphology, prevalence of histologic variants, and androgen receptor expression. Am J Surg Pathol. 2015;39(5):705–13. 25. Elsheikh TM.  Cytologic diagnosis of salivary duct carcinoma. Pathol Case Rev. 2004;9(6):236–41. 26. Jalaly JB, Sanati S, Chernock RD, Dibe DG, El-Mofty SK. Salivary duct carcinoma and invasive ductal carcinoma of the breast: a comparative immunohistochemical study. Head Neck Pathol. 2018;12(4):488–92. 27. Wang H, Hoda RS, Faquin W, et  al. FNA biopsy of secondary nonlymphomatous malignancies in salivary glands: a multi-institutional study of 184 cases. Cancer Cytopathol. 2017;125(2):91–103. 28. Schneider M, Rizzardi C. Lymphoepithelial carcinoma of the parotid glands and its relationship with benign lymphoepithelial lesions. Arch Pathol Lab Med. 2008;132(2):278–82. 29. Hipp JA, Jing X, Zarka MA, et al. Cytomorphologic characteristics and differential diagnoses of lymphoepithelial carcinoma of the parotid. J Am Soc Cytopathol. 2016;5(2):93–9. 30. Skalova A, Leivo I, Hellquist H, et  al. High-grade transformation/dedifferentiation in salivary gland carcinomas: occurrence across subtypes and clinical significance. Adv Anat Pathol. 2021;28(3):107–18.

164

S. Mehrotra et al.

31. Nakaguro M, Faquin WC, Baloch ZW, et al. Fine needle aspiration of salivary gland carcinomas with high-grade transformation: a multi-institutional study of 22 cases and review of the literature. Cancer Cytopathol. 2021;129(4):318–25. 32. Gnepp DR, Corio RL, Brannon RB. Small cell carcinoma of the major salivary glands. Cancer. 1986;58(3):705–14. 33. Chan JK, Suster S, Wenig BM, Tsang WY, Chan JB, Lau AL. Cytokeratin 20 immunoreactivity distinguishes Merkel cell (primary cutaneous neuroendocrine) carcinomas and salivary gland small cell carcinomas from small cell carcinomas of various sites. Am J Surg Pathol. 1997;21(2):226–34. 34. Chernock RD, Duncavage EJ, Gnepp DR, El-Mofty SK, Lewis JS Jr. Absence of Merkel cell polyomavirus in primary parotid high-grade neuroendocrine carcinomas regardless of cytokeratin 20 immunophenotype. Am J Surg Pathol. 2011;35(12):1806–11. 35. Chernock RD, Duncavage EJ. Salivary neuroendocrine carcinoma—an overview of a rare disease with an emphasis on determining tumor origin. Head Neck Pathol. 2018;12(1):13–21. 36. Coca-Pelaz A, Rodrigo JP, Triantafyllou A, et al. Salivary mucoepidermoid carcinoma revisited. Eur Arch Otorhinolaryngol. 2015;272(4):799–819. 37. Krane JF, Farquin WC. Salivary gland. In: Cibas ES, Ducatman B, editors. Cytology: diagnostic principles and clinical correlates. 5th ed. Elsevier; 2021. 38. Coca-Pelaz A, Rodrigo JP, Bradley PJ, et al. Adenoid cystic carcinoma of the head and neck— an update. Oral Oncol. 2015;51(7):652–61. 39. Savera AT, Sloman A, Huvos AG, Klimstra DS.  Myoepithelial carcinoma of the salivary glands: a clinicopathologic study of 25 patients. Am J Surg Pathol. 2000;24(6):761–74. 40. Skálová A, Agaimy A, Vanecek T, et al. Molecular profiling of clear cell myoepithelial carcinoma of salivary glands with EWSR1 rearrangement identifies frequent PLAG1 gene fusions but no EWSR1 fusion transcripts. Am J Surg Pathol. 2021;45(1):1–13. 41. Khanna D, Chaubal T, Bapat R, Abdulla AM, Philip ST, Arora S. Carcinoma ex pleomorphic adenoma: a case report and review of literature. Afr Health Sci. 2019;19(4):3253–63. 42. Hu Y, Xia L, Zhang C, Xia R, Tian Z, Li J. Clinicopathologic features and prognostic factors of widely invasive carcinoma ex pleomorphic adenoma of parotid gland: a clinicopathologic analysis of 126 cases in a Chinese population. J Oral Maxillofac Surg. 2020;78(12):2247–57. 43. Chhieng DC, Cangiarella JF, Cohen JM. Fine-needle aspiration cytology of lymphoproliferative lesions involving the major salivary glands. Am J Clin Pathol. 2000;113(4):563–71. 44. Al-Abbadi MA, Barroca H, Bode-Lesniewska B, et al. A proposal for the performance, classification, and reporting of lymph node fine-needle aspiration cytopathology: the Sydney system. Acta Cytol. 2020;64(4):306–22. 45. Wakely PE Jr, Lott-Limbach AA. Classic polymorphous adenocarcinoma: fine needle aspiration cytopathology of eight cases. Cytopathology. 2021;32(6):789–94. 46. Gibbons D, Saboorian MH, Vuitch F, Gokaslan ST, Ashfaq R.  Fine-needle aspiration findings in patients with polymorphous low grade adenocarcinoma of the salivary glands. Cancer. 1999;87(1):31–6. 47. Jiménez-Heffernan JA, Rodríguez-García AM, González-Peramato P, et al. Fine needle aspiration cytology of polymorphous adenocarcinoma of the salivary glands: a report of 11 patients and review of the literature. Diagn Cytopathol. 2020;48(11):1013–20. 48. Vander Poorten V, Triantafyllou A, Skálová A, et  al. Polymorphous adenocarcinoma of the salivary glands: reappraisal and update. Eur Arch Otorhinolaryngol. 2018;275(7):1681–95. 49. Andreasen S, Melchior LC, Kiss K, et al. The PRKD1 E710D hotspot mutation is highly specific in separating polymorphous adenocarcinoma of the palate from adenoid cystic carcinoma and pleomorphic adenoma on FNA. Cancer Cytopathol. 2018;126(4):275–81. 50. Franzen A, Buchali A, Lieder A. The rising incidence of parotid metastases: our experience from four decades of parotid gland surgery. Acta Otorhinolaryngol Ital. 2017;37(4):264–9.

Chapter 8

Ancillary Studies for Salivary Gland Cytology Marc P. Pusztaszeri, Vickie Y. Jo, Jeffrey F. Krane, Xiaoyin “Sara” Jiang, and Philippe Vielh

General Background and Role of Ancillary Studies The primary goal of salivary gland FNA is facilitating appropriate clinical management, which in many cases does not require a precise diagnosis. The MSRSGC risk-stratifies salivary gland FNAs into six diagnostic categories with subcategories which are generally sufficient to guide management including the type and extent of surgery (see Table 1.1). Ancillary studies for salivary gland FNA are needed only in a minority of cases since many of the commonly encountered lesions including PA and WT can be accurately diagnosed by cytomorphology alone. Therefore, ancillary studies should be considered essential for aspirates in which the results would

M. P. Pusztaszeri (*) Department of Pathology, Jewish General Hospital, Montréal, QC, Canada Department of Pathology, McGill University, Montréal, QC, Canada e-mail: [email protected] V. Y. Jo Department of Pathology, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA, USA e-mail: [email protected] J. F. Krane Department of Pathology and Laboratory Medicine, David Geffen School of Medicine at UCLA, Los Angeles, CA, USA e-mail: [email protected] X. “S.” Jiang Department of Pathology, Duke Health, Durham, NC, USA e-mail: [email protected] P. Vielh Department of Pathology, Medipath and American Hospital of Paris, Paris, France © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 W. C. Faquin et al. (eds.), The Milan System for Reporting Salivary Gland Cytopathology, https://doi.org/10.1007/978-3-031-26662-1_8

165

166

M. P. Pusztaszeri et al.

modify clinical management and/or clinical risk within the MSRSGC.  For cases that are definitely neoplastic but in which there is uncertainty regarding the specific diagnosis, the role of ancillary studies should be dictated by the specific diagnostic concerns. If only benign tumors (e.g., oncocytoma vs WT), low-grade malignancies (e.g., ACC vs SC), or high-grade malignancies (e.g., SDC vs carcinoma ex-PA) are in the differential diagnosis, then ancillary studies are not essential to resolve the diagnostic uncertainty since the clinical management will be similar. For some aspirates in the indeterminate categories of SUMP or Suspicious for Malignancy, ancillary studies can help to refine the differential diagnosis and to potentially place the specimen into a more definitive diagnostic category with or without a specific diagnosis. The application of ancillary studies for aspirates classified as AUS in the MSRSGC to distinguish neoplastic from non-neoplastic entities is often limited by the scant cellularity of these specimens (e.g., cystic lesions), and a repeat FNA +/− rapid on-site evaluation may be more helpful for these cases. Over the last decade, in addition to the traditional nonspecific epithelial/myoepithelial immunomarkers, several new immunomarkers have appeared which can be useful to narrow the differential diagnoses or to favor a specific entity when cytomorphology alone is not sufficient [1–12]. This is especially true in the SUMP subcategories of basaloid neoplasms and oncocytic/oncocytoid/clear cell neoplasms (see Chap. 5). A subset of salivary gland tumors has been characterized cytogenetically by the presence of specific molecular/genetic alterations including mainly translocations (see Chap. 11) [1, 2]. These translocations and their resulting fusion oncogenes and oncoproteins can be used as diagnostic markers in salivary gland FNA [1–12]. Currently, several antibodies are available to identify protein surrogates of specific genetic alterations which are overexpressed in a subset of salivary gland tumors, including MYB secondary to MYB::NFIB fusion in AdCC (Fig. 8.1), PLAG-1, and HMGA2 secondary to rearrangement of the corresponding genetic loci in PA and carcinomas ex-PA (Figs. 8.2 and 8.3), and more recently pan-TRK and NR4A3 expression for SC and ACC, respectively (Figs. 8.4 and 8.5) [3–14]. These “molecular” immunomarkers show variable sensitivity and specificity for

Fig. 8.1  Adenoid cystic carcinoma. MYB immunostaining showing strong nuclear expression in the tumor cells in a cytologic smear (Left) and in a cell block (Right)

8  Ancillary Studies for Salivary Gland Cytology Fig. 8.2 Pleomorphic adenoma. PLAG1 immunostaining showing strong nuclear expression in the tumor cells in a cell block

Fig. 8.3 Pleomorphic adenoma. HMGA-2 immunostaining showing moderate nuclear expression in the tumor cells in a cell block

Fig. 8.4 Secretory carcinoma. Pan-TRK immunostaining showing strong membranous expression in the tumor cells in a cell block

167

168

M. P. Pusztaszeri et al.

Fig. 8.5  Acinic cell carcinoma. NR4A3 immunostaining showing strong nuclear expression in the tumor cells in a cell block

Fig. 8.6 Secretory carcinoma. Fluorescent in situ hybridization (FISH) showing rearrangement of the ETV6 locus (separation of red and green signals) (Courtesy of Dr. Joaquin Garcia, Mayo Clinic, Rochester, MN)

salivary gland tumors. In general, they are more sensitive but less specific than the corresponding genetic alterations which can be demonstrated by various methods such as FISH or NGS (Fig. 8.6) [10, 15]. With the continuing discovery of specific genetic alterations in salivary gland tumors and the increasing availability of diagnostic markers that can be applied to FNA material, the accuracy of salivary gland FNA can be significantly improved, leading to better patient management. In this chapter, ancillary techniques and several currently available ancillary markers for salivary gland FNA are described with a practical approach covering the most common diagnostically challenging scenarios.

8  Ancillary Studies for Salivary Gland Cytology

169

Ancillary Techniques for Cytology Different methods including special histochemical stains, IC, FISH, RT-PCR, NGS, and flow cytometry can be successfully applied to FNA material to improve the diagnostic accuracy for many salivary gland tumors [3–17]. Most of these methods can be readily integrated into the diagnostic workflow, particularly as they become more widely available, cost-effective, and efficient with shorter turnaround times [3, 4]. An unresolved question is to what extent the ancillary tests discussed in this chapter represent the clinical standard of care. In resource-poor centers with infrequent salivary gland FNAs, it is unreasonable to expect access to most of these tests including some IC markers. Conversely, in larger academic centers with a significant volume of salivary gland tumors on cytology and/or surgical pathology, many of the antibodies and FISH probes should be routinely available. Smaller practice settings should have a low threshold for referring cases to academic centers or reference laboratories for those cases in which such testing might significantly alter patient management. While many of the immunocytochemical and molecular techniques can be applied to a variety of cytologic preparations including alcohol-fixed or air-dried smears, cytospins, and liquid-based preparations, their application to FFPE cell block material is considered the most reliable and is the preferred one among cytopathologists [3, 4, 8, 18]. This is because most biomarkers have been validated using FFPE tissue blocks, and most of the primary studies involving the use of cytological samples for molecular analysis have relied on FFPE cell blocks. Cell blocks have the advantage of being analogous to paraffin tissue blocks, with minimal need for standardization and reliable results. In addition, cell blocks have the advantage over other cytologic preparations of being able to produce a number of nearly identical samples for cases where a panel of IC stains will be anticipated. In contrast, cytologic preparations are generally superior to FFPE cell block sections or FFPE tissue sections for FISH because the issue of section-based nuclear truncation that can lead to inaccuracy in signal visualization and counting is avoided by having the probe hybridize directly to intact cells on a glass slide. For practical purposes, FISH or NGS is most useful for confirming a specific diagnosis that is strongly favored by the clinical, cytomorphologic, and/or immunophenotypic features of the tumor, but negative results will not definitively exclude a diagnosis in many cases [3, 4, 9, 10]. When positive, FISH or NGS analysis can confirm a diagnosis. Even on samples with few cells, a suspected malignancy from an FNA can be confirmed using FISH or NGS to assess for a specific gene rearrangement (see Sample Reports). The overexpression of translocation-associated proteins and/or downstream target proteins can be assessed using IC and can serve as a diagnostic surrogate for the molecular alterations discussed above [3–8, 11– 14]. Since IC for the fusion protein is usually more sensitive but less specific than FISH/NGS analysis, it can be used as a triage tool before FISH/NGS testing (see Sample Reports).

170

M. P. Pusztaszeri et al.

Special Stains Histochemical stains are often used to highlight stromal or cytoplasmic components. PAS and PAS with diastase (d-PAS) can be used to highlight the zymogen granules in the granular cytoplasm of ACC (Fig. 8.7) or intracytoplasmic glycogen in CCC. These special stains also detect intracytoplasmic mucin which can be found in a variety of SGT, most notably MEC and SC. Other mucin stains include mucicarmine for neutral mucin (Fig.  8.8) and Alcian blue pH 2.5 for acid mucin. Alternatively, specific types of intracellular mucin can be demonstrated by IC (e.g., MUC4 for SC) [19]. Fat stains such as Oil Red O or Sudan IV can confirm sebaceous differentiation by highlighting lipid droplets in unfixed cells (as lipid is dissolved during standard processing), although IC for adipophilin or perilipin may achieve the same purpose on FFPE cell blocks.

a

c

b

d

Fig. 8.7  Acinic cell carcinoma. Tumor cells have abundant, basophilic, granular cytoplasm (zymogen granules) and round nuclei (a: FFPE cell block, H&E). The PAS-diastase special stain (b) highlights cytoplasmic zymogen granules, which are diagnostic of acinic cell carcinoma. DOG-1 immunostain (c) is positive in the tumor cells with a characteristic membranous and canalicular staining pattern. The resection specimen (d) showed a well-demarcated acinic cell carcinoma

8  Ancillary Studies for Salivary Gland Cytology

171

Fig. 8.8 Mucoepidermoid carcinoma. Mucicarmine histochemical stain highlighting a mucin-­ positive goblet cell

Immunochemistry IC can be used to narrow the differential diagnosis in challenging cases or to confirm the diagnosis of a suspected salivary gland tumor. Importantly, published experience with salivary gland FNA material, including cell blocks, is still limited. Therefore, IC results should be interpreted with caution and in the context of cytomorphological features. In addition, a panel approach to the use of IC in salivary gland cytology is recommended as opposed to using a single immunostain (Fig. 8.9). Knowledge of the staining patterns of each IC marker in both non-neoplastic and neoplastic tissues is required to ensure a correct interpretation of positive or negative expression.

IC for Basaloid Neoplasms Aspirates of basaloid neoplasms present a very broad and challenging differential diagnosis, and many of these cases are classified as SUMP (see Chap. 5). IC can be useful in narrowing the differential diagnostic possibilities and may allow for a specific diagnosis in a subset of cases. Table 8.1 summarizes the most common IC profiles for selected basaloid salivary gland tumors. Among them, the distinction between PA and AdCC is probably the most critical given the significant clinical and prognostic implications. Conversely, the distinction between PA, BCA, BCAdc,

172

a

M. P. Pusztaszeri et al.

b

c

d

e

f

Fig. 8.9  Secretory carcinoma. Tumor cells have abundant, eosinophilic, granular cytoplasm and round nuclei (a: FFPE cell block, H&E). The main differential diagnosis includes a zymogen-poor acinic cell carcinoma and several oncocytic neoplasms, including an oncocytoma. Without ancillary studies, this case would likely be placed into the SUMP category. By IC, tumor cells in the cell block material are positive for S-100 (b), GATA-3 (c), SOX-10 (d), mammaglobin (e), and pan-­ TRK (f), supporting the diagnosis of secretory carcinoma and placing the specimen into a definitive Malignant category, without the need to demonstrate the specific ETV6 gene rearrangement by FISH or NGS

and EMC is less crucial since the management of these low-grade neoplasms is generally similar (i.e., surgical resection with clear margins). While myoepithelial cell markers are not specific for a particular diagnosis, they can be useful in a variety

8  Ancillary Studies for Salivary Gland Cytology

173

Table 8.1  Most common immunoprofiles for selected basaloid salivary gland tumors Immunostains P63

P40

SMA

calponin

SOX10

S100

C-kit (CD117)

Betacatenin

LEF1

PLAG1

HMGA2

Pleomorphic adenoma

+

+

+

+

+

+

±

-

±

+

±

Basal cell adenoma/Basal cell adenocarcinoma

+

+

+

+

+

-

±

+

+

-

-

-

Adenoid cystic carcinoma

±

±

+

+

+

+

+

-

-

-

-

+

Myoepithelioma/Myoepithelial carcinoma

+

+

+

+

+

+

-

-

-

±

±

Epithelial myoepithelial carcinoma

+

+

+

+

+

+

-

-

-

±

±

-

Polymorphous adenocarcinoma Biphasic staining pattern (abluminal cells) Stromal cells can be S100 positive Positive in a subset of cases Nuclear expression mainly in the stromal and basal cells

+

-

-

-

+

+

±

-

-

±

-

-

Diagnosis

MYB -

-

Fig. 8.10 Pleomorphic adenoma. Dual immunostaining showing p63 nuclear expression (brown) in the myoepithelial cells and CK7 cytoplasmic expression (red) in the luminal cells of pleomorphic adenoma in a cell block

of circumstances to demonstrate the presence of a minor or predominant myoepithelial component in a salivary gland tumor. Salivary gland neoplasms containing a myoepithelial component include several benign and malignant tumors such as PA, myoepithelioma, myoepithelial carcinoma, BCA, BCAdc, AdCC, EMC, IDC, and to a limited extent PAC. Due to the high variability of expression profiles of neoplastic myoepithelial cells, a panel of immunostains is recommended to confirm myoepithelial differentiation including p63, p40, CK5/6, GFAP, SOX-10, and S-100, as well as more specific myoid markers such as SMA and calponin. They can be used alone or in conjunction with epithelial luminal/ductal markers (e.g., CK, EMA, CEA) to demonstrate the biphasic staining pattern (mixed ductal/myoepithelial cell pattern) of some benign and malignant salivary gland tumors (Figs. 8.10, 8.11, and 8.12). AdCC, BCA/BCAdc, EMC, and PA all contain a p40/p63-­positive basal/myoepithelial cell population (Figs. 8.10, 8.11, and 8.12). In contrast, PAC, which is usually restricted to minor salivary glands (see below), is the only tumor in this group that has a p63-positive single-cell population in which p40 is not co-­ expressed (Table  8.1) [20]. While SOX10 can also be used as a myoepithelial marker, SOX10 has been shown to be expressed in most salivary gland tumors including basaloid neoplasms and may not aid in their differential diagnosis [21]. However, SOX10 is negative in CCC, SCC, SDC, oncocytic tumors, and most MECs (see below). Therefore, SOX10 could prove helpful in a specific context,

174

M. P. Pusztaszeri et al.

a

b

c

d

Fig. 8.11  Basal cell adenoma. The aspirate shows a basaloid neoplasm with a tubulo-trabecular growth pattern and cellular stroma (a: FFPE cell block, H&E). The main differential diagnosis includes basal cell adenoma and basal cell adenocarcinoma, which cannot be accurately distinguished on cytology. As a result, these cases are classified as SUMP, basaloid neoplasm, according to the MSRSGC; (b) Dual immunostain of p63 (brown nuclear staining) and CK7 (red cytoplasmic staining) confirms the presence of a biphasic (basal/luminal) phenotype. The surgical follow-up shows an encapsulated noninvasive cellular basaloid neoplasm consistent with a basal cell adenoma; (c) On higher magnification; (d) basaloid cells show peripheral palisading, typical of basal cell adenoma Fig. 8.12 Epithelialmyoepithelial carcinoma. Pancytokeratin immunostain showing the biphasic pattern of the tumor

especially if negative. Several more specific immunomarkers are useful for addressing the differential diagnosis of basaloid salivary gland tumors (Table 8.1). In order to increase the sensitivity and specificity in distinguishing various basaloid neoplasms, an IC panel consisting of MYB, CD117 (c-KIT), PLAG1, HMGA2,

8  Ancillary Studies for Salivary Gland Cytology

175

β-catenin, and LEF-1 can be helpful (Fig. 8.1, 8.2, 8.3, 8.13, and 8.14) [3, 4, 6, 8, 11, 12]. Most PAs with or without a PLAG1 gene rearrangement are immunoreactive for PLAG1 (Fig. 8.2) [8]. PLAG1 is also positive in myoepitheliomas, which is considered a myoepithelial-predominant variant of PA by some authors, in a subset of PAC, and in many carcinomas ex-PA including myoepithelial carcinomas and EMC [8]. In contrast, PLAG1 is usually negative in other salivary gland tumors including AdCC, MEC, and ACC. HMGA2 is positive in about 20% of PA (Fig. 8.3) and is usually negative in other salivary gland tumors, although myoepitheliomas and CA-ex-PA may be positive [22]. The combined use of PLAG1 and HMGA2 can identify about 70% of PAs and about 50% of CA-ex-PA. Most AdCCs show strong immunoreactivity for MYB (Fig. 8.1), predominantly in the nuclei of basal/myoepithelial cells, while other basaloid neoplasms (BCA/BCAdc, PAC, PA) are usually negative or only focally positive [8, 11, 12]. However, a large percentage of basaloid SCCs are positive for MYB. Recently, MYB RNA chromogenic ISH was found to provide superior specificity compared with MYB IC and superior sensitivity for the diagnosis of AdCC compared with MYB FISH [23]. This assay could provide a useful tool for rapidly confirming the diagnosis of AdCC in cell block specimens. In addition to MYB, over 90% of AdCC are strongly and diffusely positive for CD117 (c-KIT) (Fig.  8.13) [24]. While CD117 has generally been regarded as a marker of AdCC, it should be interpreted with caution, as it is also expressed in several other salivary gland tumors, including BCA/BCAdc and PAC [24]. The main discriminating feature for BCA and BCAdc is the presence or absence of invasion as detected by histologic evaluation (Fig.  8.11). Therefore, it is usually not possible to distinguish these two entities by FNA.  In addition, the immunohistochemical phenotypes for BCA and BCAdc are relatively similar [6, 25]. BCA/ BCAdc may demonstrate nuclear expression of β-catenin, usually in the abluminal cells and stromal cells, while other salivary gland tumors generally show only membranous expression (Fig. 8.14). In surgical resection specimens, nuclear β-catenin overexpression has moderate sensitivity (82%) but high specificity (96%) for BCA Fig. 8.13  Adenoid cystic carcinoma. CD117 immunostaining showing strong cytoplasmic expression in the tumor cells in a cytologic smear

176

M. P. Pusztaszeri et al.

Fig. 8.14  Basal cell adenoma. Beta-catenin immunostaining showing strong nuclear expression in the tumor cells in a cell block

in comparison with PA and AdCC [25]. Although experience is limited, to date β-catenin appears to have limited diagnostic utility in cell blocks, likely due in part to the characteristic heterogeneous staining pattern that may be difficult to appreciate in limited specimens. In addition to β-catenin, LEF-1, a transcription factor in the Wnt signaling pathway that interacts with β-catenin (i.e., β-catenin coactivator), has been evaluated as a diagnostic marker. LEF-1 has been shown to be expressed in most BCAs, often with co-expression of β-catenin, but this marker is not specific in isolation and can also be positive in PA, AdCC, EMC, and CCC. In a study examining LEF-1 in cell block material, 60% of BCA expressed LEF-1, although staining was also seen in more than half of PA. In contrast, the vast majority (>90%) of AdCC were negative [26].

IC for Oncocytic/Oncocytoid Neoplasms Oncocytic salivary gland tumors detected by FNA have a broad differential diagnosis (see Chap. 5). The use of IC can be used to narrow the diagnostic possibilities and may allow for a specific diagnosis in a subset of cases (Fig. 8.9). Table 8.2 summarizes the most common IC profiles in salivary gland tumors with oncocytic features. A limited IC panel consisting of DOG-1, SOX10, and p63 is often sufficient for separating ACC from WT, MEC, and oncocytoma [26–28]. DOG1 and SOX10 are markers of acinar and intercalated duct differentiation in salivary gland neoplasms, and both are characteristically strongly and diffusely positive in ACC (Figs. 8.15 and 8.16), with DOG-1 often demonstrating a canalicular staining pattern (Fig. 8.7) [27, 28]. In isolation, SOX10 is also a marker of myoepithelial cells and is expressed in many salivary gland tumors (see above) [21, 28]. DOG-1 and SOX10 are predominantly negative in WT, oncocytoma, oncocytic carcinoma, SC, and MEC. Conversely, p63 typically shows diffuse expression in MEC, including its

8  Ancillary Studies for Salivary Gland Cytology

177

Table 8.2  Most common immunoprofiles in salivary gland tumors with oncocytic features Immunostains P63

P40

S100

Mammaglobin

SOX10

DOG1

GATA3

Androgen receptor

PLAG1

HMGA2

Warthin Tumor/Oncocytoma

+

+

-

-

-

-

-

-

-

-

-

Acinic cell carcinoma

-

-

-

-

+

+

-

-

-

-

+

-

Secretory carcinoma

-

-

+

+

+

-

+

-

-

-

-

+

Mucoepidermoid carcinoma

+

+

-

-

±

±

-

-

-

-

-

-

Salivary duct carcinoma both P63 and P40 show peripheral basal cell staining positive in salivary duct carcinoma ex pleomorphic adenoma

-

±

-

-

+

+

±

±

-

-

Diagnosis

NR4A3

pan-Trk -

Fig. 8.15  Acinic cell carcinoma. DOG-1 immunostaining showing membranous and canalicular expression in the tumor cells in a cell block

Fig. 8.16  Acinic cell carcinoma. SOX-10 immunostaining showing strong nuclear expression in the tumor cells in a cell block

oncocytic variant, and is negative in ACC. While WT and MEC both show reactivity for p63 and p40, the distribution of positive cells differs; a single basal layer of cells in WT and a more diffuse pattern in MEC. S100, GATA-3, and mammaglobin are useful to support the diagnosis of SC since other oncocytic neoplasms in the differential diagnosis are usually negative for those IC markers (Fig. 8.9). However, distinguishing between SC and IDC (intercalated duct-type IDC), an emerging entity in salivary glands, can be challenging, as these entities share histological,

178

M. P. Pusztaszeri et al.

Fig. 8.17 Secretory carcinoma. MUC-4 immunostaining showing strong cytoplasmic expression in the tumor cells in a cell block

cytological, and IC (expression of S100, SOX10, and mammaglobin) features [17]. Pan-TRK IC is emerging as a surrogate marker for NTRK fusion-positive SC, but data are still limited. The presence of any nuclear or cytoplasmic staining for pan-­ TRK IC has been shown to be sensitive for SC (Fig. 8.4), while consideration of nuclear staining alone shows diminished sensitivity (70%) but high specificity (close to 100%), including in cytologic material [13]. MUC4 is another valuable IC marker for distinguishing SC from its main mimics (ACC, PAC, and IDC), with a high sensitivity and specificity (Fig.  8.17) [17]. However, MUC4 is frequently expressed with a membranous and cytoplasmic pattern in all cell types (epidermoid, intermediate, mucinous, clear, and columnar) of MEC, and MUC4 is uncommonly and usually weakly expressed in PA. NR4A3 translocation is present in most ACC (see below) and NR4A3 IC is an effective surrogate marker for the translocation (Fig. 8.5) [14]. NR4A3 IC for ACC has high sensitivity and specificity in cytologic material and performs better than both DOG-1 IC and NR4A3 FISH [14]. Since NR4A3 is not expressed in normal acinar cells, it can also be used to distinguish ACC from normal acinar cells and to support a diagnosis of ACC in poorly differentiated/dedifferentiated cases. Demonstration of specific translocations through molecular testing remains the gold standard for confirming the diagnosis of SC and MEC (Fig. 8.6), especially if the morphology and IC findings are equivocal.

ICC Markers in Clear Cell Neoplasms Several salivary gland tumors, in particular several with oncocytic features as mentioned above, can also have clear cell morphology (see Chap. 5). However, some of these tumors such as CCC are almost always restricted to minor salivary glands (see below). The use of IC with the same panel of markers described for oncocytic

179

8  Ancillary Studies for Salivary Gland Cytology Table 8.3  Most common immunoprofiles in salivary gland tumors with clear cell features Diagnosis Myoepithelioma/myoepithelial carcinoma Epithelial-myoepithelial carcinoma Acinic cell carcinoma Mucoepidermoid carcinoma Hyalinizing clear cell carcinoma

Immunostains P63 P40 + + + + − − + + + +

S100 + + − − −

SOX10 + + + ± −

DOG1 − − + ± n/a

salivary gland tumors can be useful; Table 8.3 summarizes the most common IC results in SGT with clear cell features. EMC is characterized by a predominant population of myoepithelial cells displaying an unusually large amount of clear cytoplasm. A panel of IC stains to demonstrate the myoepithelial nature of the clear cells combined with a marker for ductal cells such as keratin AE1.3 or EMA is helpful to demonstrate the biphasic pattern of the tumor (Fig. 8.12). Although experience is limited, IC for RAS Q61R has been shown to be highly sensitive and specific for the diagnosis of EMC with the HRAS Q61R mutation (see below) [29].

IC for Primary Vs Secondary Salivary Gland Tumors High-grade salivary gland carcinomas are easily recognized as malignant; however, the distinction between primary and secondary malignancy can occasionally be problematic yet clinically important. Most patients with a secondary malignancy of the salivary gland have a clinical history. SCC is the most common secondary malignancy in the salivary glands, followed by melanoma, and carcinomas from distant sites (lung, breast, kidney, colon). A limited panel of immunostains can be very helpful to distinguish a primary salivary gland malignancy from a metastasis (Table 8.4). Since primary SCC of the parotid gland is extremely rare, any SCC involving a major salivary gland should be considered to represent either metastasis or direct extension from a head and neck cutaneous (e.g., scalp, auricle, face) or mucosal site. While HPV-associated SCC does not typically involve intra-parotid lymph nodes, for cases involving level II cervical lymph nodes near the tail of the parotid gland, high-risk HPV testing should be considered for metastatic SCC of unknown primary origin. CAP guidelines consider p16 positivity in surgical pathology material suggestive of an HPV-associated oropharyngeal SCC when there is a metastatic SCC of unknown primary in upper or middle jugular lymph nodes (levels II and III) with typical non-keratinizing morphology [30] although HPV-specific testing, when possible, is recommended for confirmation. Importantly, p16 IC can also be positive in branchial cleft cysts and in approximately 30% of cutaneous SCC. The appropriate threshold for p16 positivity in cytologic material has not been established and HPV-specific methods of testing for high-risk HPV are recommended for cytology specimens if appropriately validated. Finally, in cases of metastatic

180

M. P. Pusztaszeri et al.

Table 8.4  Common immunomarkers to suggest a site of origin for salivary gland metastatic carcinomas Immunostain CDX-2, SATB2 TTF-1 Napsin A ER, PR PAX-8 CD10, RCC PSA, NKX3.1 Thyroglobulin HepPar1, Arginase1 GATA-3 P63, p40, CK5/6 GCDFP15, mammaglobin

Probable site of origina Enteric Lung, thyroid, small cell carcinoma (any site) Lung Breast, müllerianb Kidney, müllerian, thyroid Kidney Prostatec Thyroid Hepatocellular Breast, urothelial, salivary gland, othersd Squamous or urothelial Breaste

Transcription factors are in bold (nuclear staining) TTF-1 thyroid transcription factor-1, ER estrogen receptor, PR progesterone receptor, RCC renal cell carcinoma, PSA prostate specific antigen, GATA3 GATA binding protein 3, GCDFP15 gross cystic disease fluid protein 15, MGB mammaglobin a  Immunomarkers are best used as a panel and in conjunction with clinicoradiological data; several of these markers can also be expressed in primary salivary gland carcinomas b  Can be expressed in a wide variety of other carcinomas c  A subset of salivary duct carcinomas and oncocytomas are also PSA positive d  A subset of carcinomas from the skin are also positive e  A subset of primary salivary gland carcinomas (e.g., secretory carcinoma) are also positive

SCC of unknown primary to upper jugular chain lymph nodes, EBV-associated nasopharyngeal carcinoma is also an important consideration, and EBER testing should be considered either concurrently with HPV testing or in cases that prove to be HPV negative. Although rarely necessary, cutaneous origin of SCC in intraparotid and peri-parotid lymph nodes may be confirmed by the detection of ultraviolet signature mutations. Histochemical staining for intracellular mucin is helpful to distinguish metastatic SCC from MEC which is positive. In addition, AR, GATA-3, and p63/p40/CK5–6 are very helpful to distinguish SDC from its cytomorphologic mimics, especially metastatic non-keratinizing SCC. More than 95% of SDC are positive for AR and GATA-3, but negative for p63/p40/CK5–6 (see Chap. 7 and Table 7.2).

Molecular Genetic Alterations in Salivary Gland Tumors Salivary gland tumors currently known to harbor recurrent molecular genetic alterations are summarized in Chap. 11. With ongoing advances in molecular diagnostics, other salivary gland neoplasms as well as additional molecular alterations are likely to join this list. Although some of these molecular genetic alterations can be found in various tumors including salivary gland tumor analogs in other organs, they are

8  Ancillary Studies for Salivary Gland Cytology

181

highly specific in the spectrum of primary neoplasms of the salivary glands, representing powerful diagnostic markers in histological specimens as well as in FNA material [1–4, 15, 31]. As outlined above, there are now many available IC markers that can serve as surrogates for underlying molecular alterations, contributing to accurate diagnosis. The absence of a given genetic alteration however may not exclude a particular tumor as its prevalence may vary significantly between different salivary gland neoplasms. In addition to their diagnostic role, molecular genetic alterations can also represent prognostic markers and therapeutic targets in some cases (e.g., NTRK, ALK, and RET) [1, 2, 15, 31].

Salivary Gland Tumors with Specific Molecular Features Approximately 70% of PA harbor recurrent translocations. The majority are rearrangements involving PLAG1 (encoded on 8q12), with CTNNB1 being the most common fusion partner [1, 2, 31]. Rearrangements of HMGA2 (encoded on 12q14–15) occur in a smaller subset of PA (10–20%) [21, 31]. CA-ex-PA, regardless of morphology, also harbors PLAG1 and HMGA2 rearrangements, including SDC-ex-PA and cases of EMC arising ex-PA [1, 2, 31]. De novo EMC show frequent HRAS Q61R mutations [29, 31]. MEC are characterized by MAML2 rearrangements, most commonly CRTC1::MAML2 fusions secondary to the translocation t(11;19)(q21;p13), and rarely a CRTC3::MAML2 fusion secondary to translocation t(11;15)(q21;q26) [1, 2, 31]. MAML2 rearrangements have been detected in nearly 75% of low- to intermediate-­grade MEC but less frequently in high-grade MEC. While previously “translocation-negative” high-grade MEC were considered to have poorer prognosis compared to MAML2-rearranged high-grade MEC, most “translocation-­ negative” MEC likely represent a heterogeneous group of other tumor types such as adenosquamous carcinoma. Detection of MAML2 rearrangement by FISH or NGS in cytologic samples and small biopsies has high utility in confirming the diagnosis of MEC and is particularly helpful for oncocytic and clear cell MEC variants [7, 15]. However, one limitation is that low-grade MECs may yield hypocellular FNA specimens that lack sufficient cellularity for molecular testing. AdCC is characterized by the translocation t(6;9)(q21–24;p13–23), involving MYB and NFIB genes in approximately half of all cases [1, 2, 31]. However, MYB activation appears to be a common pathogenetic event as MYB overexpression is detected in up to 90% of all AdCC, including fusion-negative tumors [31]. Variant alterations are seen in a small subset of AdCC, including MYBL1::NFIB fusion and 5′-NFIB fusion (without MYB/MYBL1) [31]. Activating mutations of NOTCH1 also occur in about 10% of AdCC and are associated with the solid subtype and poorer outcomes [31]. ETV6::NTRK3 fusion is the hallmark of SC [1, 2, 31]. This fusion gene occurs secondary to the translocation t(12;15)(p13;q25), identical to its counterpart in the breast (Fig.  8.6). Rare variant fusion genes have also been reported, including ETV6::RET, ETV6::MET, NCOA4::RET, TRIM27Z::RET, and VIM::RET. [15, 31]

182

M. P. Pusztaszeri et al.

ACC is now known to be characterized by the upregulation of NR4A3 (encoded on 4q13) [14, 15]. Tumors harbor the recurrent translocation t(4;9)(q13;q31), which results in constitutive upregulation of NR4A3 through enhancer hijacking. Notably, NR4A3 IC expression does not perfectly correlate with NR4A3 rearrangement. Up to 15% of ACC harbor alternative HTN3::MSANTD3 fusions, all showing NR4A3 upregulation. NR4A3 IC is thus likely the most sensitive diagnostic marker for ACC, given that NR4A3 overexpression occurs regardless of the underlying genomic mechanism [14]. At the molecular level, 30–80% of BCA have CTNNB1 mutations, while BCAdc shows a different and sometimes more complex genomic profile, including activating mutations in PIK3CA, usually without CTNNB1 mutation despite β-catenin expression [24]. The uncommon membranous subtype of BCA/BCAdc, which is associated with CYLD1 gene alterations, and may be associated with Brooke-­ Spiegler syndrome, is also less likely to show β-catenin or LEF-1 expression. In addition, CYLD1 gene mutation has not been shown to correlate with CYLD1 immunoexpression. The molecular genetic profile of SDC includes TP53 mutations, HRAS mutations, BRAF mutations, PIK3CA mutations, and PTEN deletions [31]. Approximately 50% of SDCs arise in a preexisting PA, and SDC-ex-PA often harbors either PLAG1 or HMGA2 rearrangements [31]. HER2 (ERBB2) amplification occurs in up to 35% of SDC where it functions as a potential therapeutic target, but it is infrequent in de novo tumors. AR gene activation, including splice variants, has also been reported. IDC is an emerging entity which may be sampled by FNA, although histologic evaluation is necessary to confirm the predominantly intraductal growth. Most cases fall into four main groups which show some genotypic-phenotypic associations: intercalated duct IDC, apocrine IDC, oncocytic IDC, and mixed-type IDC. RET fusions occur in intercalated duct IDC (with rare alternate fusions including ALK) and mixed IDC [32]. Oncocytic IDC has RET fusions or BRAF V600E mutations [32]. Apocrine IDC tends to show molecular profiles mirroring SDC, including HRAS, PIK3CA, and TP53 mutations (as well as AR IC-positivity) [32].

 arely Encountered Neoplasms of the Minor Salivary Glands R with Specific Molecular Features Some uncommon salivary gland tumors such as CCC, PAC, and CAMSG are very rarely encountered in salivary gland cytology since they are essentially restricted to the minor salivary glands, are infrequently sampled by FNA, and are generally not diagnostic considerations in aspirates from the major salivary glands. They significantly overlap morphologically with other entities and therefore should always be included in the differential diagnosis of aspirates from minor salivary gland sites. Furthermore, ancillary testing is useful for their diagnosis. CCC is a rare low-grade salivary gland carcinoma which can be difficult to classify by FNA.  Tumors are challenging to distinguish from other salivary gland

8  Ancillary Studies for Salivary Gland Cytology

183

neoplasms with clear cell features, especially myoepithelioma/myoepithelial carcinoma and MEC with clear cell features (see Table 8.3). However, correct classification and distinction from other more aggressive salivary gland tumors is important. CCC is often positive for squamoid markers CK5/6, p40, and p63, but it lacks myoepithelial differentiation. Thus, SOX10 and other myoepithelial markers are useful in the differential diagnosis between CCC and myoepithelioma/myoepithelial carcinoma with clear cell features. CCC is characterized by the specific translocation t(12;22)(q13;q12) generating an EWSR1::ATF1 fusion gene, which is present in approximately 85% of cases [1, 2, 31]. In cases with this diagnostic suspicion, a conclusive diagnosis of CCC can be established with the assessment for the specific EWSR1 rearrangement, which is not present in other salivary gland tumors with clear cell features, except for a subset (35%) of clear cell myoepithelial carcinomas and rare EMC (9%) that can be usually distinguished by their different IC profiles [31]. There is no IC correlation for the EWSR1 rearrangement. PAC is another rare minor salivary gland cancer that primarily arises in the oral cavity, particularly the palate. Because of its typically low-grade behavior, it is important to distinguish it from AdCC.  The majority of PAC harbor a PRKD1 E710D hotspot mutation [31]. Mutational analysis for E710D is feasible on cytologic material and allows distinction of PAC from AdCC [33]. The presence of the PRKD1 mutation has been significantly associated with metastasis-free survival. CAMSG (currently categorized by the World Health Organization as a variant of PAC) demonstrates some propensity for lymph node metastases and may occasionally present as a metastatic carcinoma of unknown primary. Cytologically, CAMSG appears as a low-grade neoplasm often with nuclear features mimicking papillary thyroid carcinoma or SC. Clinical history and IC are usually sufficient to distinguish these tumors. In addition, CAMSG has PRKD1 rearrangements that can be identified by FISH or NGS. Rearrangements of PRKD family genes (PRKD1, PRKD2, or PRKD3) are not entirely mutually exclusive between CAMSG and PAC [31]. The recently recognized microsecretory adenocarcinoma of the salivary gland typically arises in the oral cavity and shows recurrent MEF2C::SS18 fusions [34]. FISH or NGS for SS18 is diagnostic, which may be helpful given its nonspecific immunophenotypic features, including positivity for S-100, p63, and SOX10 and negativity for p40 and mammaglobin [15, 34, 35].

Fluorescent In Situ Hybridization (FISH) In contrast to other molecular methods, in situ-based detection of nucleic acids has the advantage of providing useful diagnostic information within the context of the cytomorphology rather than histology. Currently, there are two major ways of assessing DNA copy number/rearrangement status in situ—fluorescent-based methods (FISH) and bright-field-based methods—chromogenic in situ hybridization (CISH). In salivary gland tumors, most clinically relevant genetic alterations are rearrangements generating gene fusions, and FISH has been shown to be superior to

184

M. P. Pusztaszeri et al.

other ISH techniques for demonstrating rearrangements. Break-apart FISH probes are designed to detect specific translocations. They are designed to flank on either side of a gene so that in the presence of a translocation, the two colors will lead to a split signal (Fig.  8.6). As previously mentioned, cytologic smears and cytospin preparations have the advantage of evaluating signals in whole nuclei without the truncation artifact from histologic sectioning; however, cell blocks can also be used with the same adaptations and validations used for histological FFPE material [7]. It is recommended to use dual-observer scoring to circumvent intra- and inter-­ observer variability. When positive, FISH analysis can confirm a diagnosis, even in salivary gland FNA samples with a limited number of lesional cells [7, 10]. At present no set requirements have been established regarding the minimum cell number required for interpreting FISH studies in salivary gland FNA specimens [7, 10].

Polymerase Chain Reaction (PCR) The core principle of PCR is the amplification of a DNA region of interest. Material from different cytological preparations is an excellent source for PCR analysis, and 50–100 cells are adequate to obtain good PCR results. One of the most used applications of PCR is the study of gene expression, including the production of fusion transcripts based on the ability of PCR to amplify RNA (RT-PCR). PCR is much more sensitive than FISH for detecting known translocations; however, it is not able to detect unknown molecular variants which can be detected by FISH analysis.

Next-Generation Sequencing (NGS) NGS is a high-throughput molecular platform that allows sequencing of multiple gene sequences in parallel and interrogating various genetic alterations for multiple patients in a single run. In addition to the evaluation of DNA alterations (e.g., point mutations, insertions/deletions), the comprehensive molecular assessment of cancer biomarkers also requires extended molecular testing for gene fusions and rearrangements, especially in the context of salivary gland tumors (See Chap. 11). Although routine cytological samples are generally suitable for NGS analysis, care must be taken to validate protocols developed for non-FFPE specimens, such as smears and LBC samples [36]. Different commercial NGS platforms including the Ion Torrent sequencer (Thermo Fisher Scientific, Waltham, MA) or Illumina (Illumina, San Diego, CA) can be used for RNA sequencing. The main steps of next-generation RNA sequencing are cDNA library preparation, template preparation, sequencing, and data analysis [36]. Next-generation RNA sequencing has immense potential in terms of sensitivity and throughput. The main disadvantages of next-generation RNA sequencing however are its high cost and high bioinformatics requirements for data analysis and interpretation [36]. However, gene alterations specifically found

8  Ancillary Studies for Salivary Gland Cytology

185

in many salivary gland tumor subtypes are uncommon fusion-type changes, and many of them are not covered by these commercially available NGS panels [15]. As a result, there is currently a limited routine diagnostic role for NGS in salivary gland cytology. However, this is a rapidly evolving field and there may be future roles for salivary gland-specific panel-based testing and NGS, especially in the diagnosis of uncommon tumors and/or for targeted therapies. Some institutions have already developed their own comprehensive NGS panels that may be enhanced and customized to include all relevant genes to detect specific alterations, including mutations, fusions, and gene expression levels on the RNA level in order to facilitate the diagnosis and classification of salivary gland neoplasms [15]. For example, SalvGlandDx is an all-in-one RNA-based NGS panel that detects alterations in 27 genes (including NR4A3) found in salivary gland tumors [15]. The main advantage of this approach is the simultaneous capturing of aberrant expression, gene mutations, and fusions with only the need for RNA extraction. This promising approach covers most of the common molecular alterations of SGT in one test and can be reliably performed on FFPE cell block specimens [15]. Akin to molecular testing for thyroid cytologic samples, it is likely in the near future that NGS for cytologic samples of salivary gland aspirates will not only help in establishing a specific diagnosis but also provide prognostic and therapeutic information to guide clinical management.

Flow Cytometry (FC) FC is a technique that measures the physical and immunological properties of intact cells in suspension. In salivary gland FNA, FC is primarily used to characterize lymphoproliferative lesions by immunophenotyping (see Chaps. 2, 3, 6, and 7) [16]. Dedicated passes should be made, and the needle rinsed immediately in the appropriate fluid recommended by the hematology laboratory. Usually, RPMI solution is preferred but sterile saline is also appropriate so long as it is rapidly processed for FC, ideally on the same day. A morphological assessment of the FNA material before processing it for FC, using a cytospin, for example, is highly desirable. If the number of cells is limited, tailored antibody panels should be designed based on clinical features, patient history, and specimen source. Given that the diagnosis of lymphoid lesions in aspirates of the salivary gland has significant limitations using cytomorphology alone, FC can be extremely useful in distinguishing reactive conditions from lymphoma [16]. For B-cell lymphomas, the demonstration of a clonal population based upon the presence of kappa or lambda light chain restriction as well as expression of Bcl2 is diagnostic. The presence of an altered T-cell immunophenotype also can be used to suggest a possible T-cell lymphoma. In a series of 61 cases, Stacchini et al. showed that a combination of cytology and FC could diagnose and classify lymphoproliferative lesions in salivary gland FNAs with a sensitivity of 100% and specificity of 83% [16]. FC is also able to detect the presence of non-lymphoid neoplastic cells in an FNA such as CD56 positive neuroendocrine cells.

186

M. P. Pusztaszeri et al.

Summary Ancillary techniques are needed only in a limited number of salivary gland FNA cases in order to overcome cytomorphological limitations, especially in the SUMP and Suspicious for Malignancy categories of the MSRSGC, especially when it is expected that the results will impact clinical management. IC is often helpful as a first step and may be sufficient in the evaluation of some challenging salivary gland aspirate samples. A variety of antibodies, including several newer antibodies associated with specific molecular/genetic alterations, are increasingly useful in this setting, especially when used in panels and in conjunction with cytomorphologic features and clinico-radiological findings. Additionally, the detection of specific molecular/genetic alterations of salivary gland tumors through FISH or NGS can be useful to confirm the diagnosis of a particular salivary gland neoplasm, especially if the morphology and IC findings are equivocal. When used judiciously in the framework of the MSRSGC, the combined use of these ancillary techniques can improve the diagnostic accuracy of salivary gland cytology, which may be helpful in clinical management and decision-making.

Sample Reports Example 8.1 Satisfactory for evaluation. NEOPLASM: SUMP Basaloid neoplasm most consistent with cellular pleomorphic adenoma. See Note. Note: By immunohistochemistry, the tumor cells are positive for PLAG-1 and negative for beta-catenin, MYB, and CD117 (focal weak). This immunoprofile combined with the cytomorphologic findings favors a cellular pleomorphic adenoma. Conservative excision with clear margins is recommended. Example 8.2 Satisfactory for evaluation SUSPICIOUS FOR MALIGNANCY Basaloid neoplasm suspicious for adenoid cystic carcinoma. See Note. Note: By immunohistochemistry, the tumor cells are positive for MYB and CD117, while negative for PLAG1 and beta-catenin. Combined with the cytomorphologic findings, these results are suspicious for adenoid cystic carcinoma. Correlation with clinical and radiological findings is advised. Molecular testing for MYB rearrangement could be helpful to confirm the diagnosis, if clinically indicated. Example 8.3 Satisfactory for evaluation. MALIGNANT Acinic cell carcinoma. See Note.

8  Ancillary Studies for Salivary Gland Cytology

187

Note: By immunohistochemistry, the tumor cells are positive for NR4A3, DOG-1, and SOX-10, while negative for pan-TRK, mammaglobin, S-100, and p63. Combined with the cytomorphologic findings, the overall features are diagnostic of acinic cell carcinoma. Example 8.4 Satisfactory for evaluation MALIGNANT Secretory carcinoma. See Note. Note: The presence of the specific t(12;15) translocation, demonstrated by FISH analysis, supports the diagnosis of secretory carcinoma. Example 8.5 Satisfactory for evaluation NON-NEOPLASTIC Reactive lymph node. See Note. Note: The combined cytomorphologic findings and corresponding benign flow cytometry favor a reactive lymph node. If lymphadenopathy persists, repeat sampling would be indicated for further evaluation.

References 1. Andersson MK, Stenman G. The landscape of gene fusions and somatic mutations in salivary gland neoplasms—implications for diagnosis and therapy. Oral Oncol. 2016;57:63–9. 2. Weinreb I.  Translocation-associated salivary gland tumors: a review and update. Adv Anat Pathol. 2013;20:367–77. 3. Pusztaszeri MP, García JJ, Faquin WC. Salivary gland FNA: new markers and new opportunities for improved diagnosis. Cancer Cytopathol. 2016;124:307–16. 4. Pusztaszeri MP, Faquin WC.  Update in salivary gland cytopathology: recent molecular advances and diagnostic applications. Semin Diagn Pathol. 2015;32:264–74. 5. Griffith CC, Schmitt AC, Little JL, Magliocca KR.  New developments in salivary gland pathology: clinically useful ancillary testing and new potentially targetable molecular alterations. Arch Pathol Lab Med. 2017;141:381–95. 6. Griffith CC, Siddiqui MT, Schmitt AC. Ancillary testing strategies in salivary gland aspiration cytology: a practical pattern-based approach. Diagn Cytopathol. 2017;45:808–19. 7. Darras N, Mooney KL, Long SR. Diagnostic utility of fluorescence in situ hybridization testing on cytology cell blocks for the definitive classification of salivary gland neoplasms. J Am Soc Cytopathol. 2019;8:157–64. 8. Foo WC, Jo VY, Krane JF.  Usefulness of translocation-associated immunohistochemical stains in the fine-needle aspiration diagnosis of salivary gland neoplasms. Cancer Cytopathol. 2016;124:397–405. 9. Evrard SM, Meilleroux J, Daniel G, Basset C, Lacoste-Collin L, Vergez S, Uro-Coste E, Courtade-Saidi M. Use of fluorescent in-situ hybridisation in salivary gland cytology: a powerful diagnostic tool. Cytopathology. 2017;28:312–20. 10. Hudson JB, Collins BT.  MYB gene abnormalities t(6;9) in adenoid cystic carcinoma fine-­ needle aspiration biopsy using fluorescence in situ hybridization. Arch Pathol Lab Med. 2014;138:403–9.

188

M. P. Pusztaszeri et al.

11. Pusztaszeri MP, Sadow PM, Ushiku A, et al. MYB immunostaining is a useful ancillary test for distinguishing adenoid cystic carcinoma from pleomorphic adenoma in fine-needle aspiration biopsy specimens. Cancer Cytopathol. 2014;122:257–65. 12. Moon A, Cohen C, Siddiqui MT. MYB expression: potential role in separating adenoid cystic carcinoma (ACC) from pleomorphic adenoma (PA). Diagn Cytopathol. 2016;44:799–804. 13. Xu B, Haroon Al Rasheed MR, Antonescu CR, Alex D, Frosina D, Ghossein R, Jungbluth AA, Katabi N.  Pan-Trk immunohistochemistry is a sensitive and specific ancillary tool for diagnosing secretory carcinoma of the salivary gland and detecting ETV6-NTRK3 fusion. Histopathology. 2020;76:375–82. 14. Skaugen JM, Seethala RR, Chiosea SI, Landau MS. Evaluation of NR4A3 immunohistochemistry (IHC) and fluorescence in situ hybridization and comparison with DOG1 IHC for FNA diagnosis of acinic cell carcinoma. Cancer Cytopathol. 2021;129:104–13. 15. Freiberger SN, Brada M, Fritz C, Höller S, Vogetseder A, Horcic M, Bihl M, Michal M, Lanzer M, Wartenberg M, Borner U, Bode PK, Broglie MA, Rordorf T, Morand GB, Rupp NJ. SalvGlandDx—a comprehensive salivary gland neoplasm specific next generation sequencing panel to facilitate diagnosis and identify therapeutic targets. Neoplasia. 2021;23:473–87. 16. Stacchini A, Aliberti S, Pacchioni D, Demurtas A, Isolato G, Gazzera C, Veltri A, Maletta F, Molinaro L, Novero D.  Flow cytometry significantly improves the diagnostic value of fine needle aspiration cytology of lymphoproliferative lesions of salivary glands. Cytopathology. 2014;25:231–40. 17. Taverna C, Baněčková M, Lorenzon M, Palomba A, Franchi A, Skalova A, Agaimy A. MUC4 is a valuable marker for distinguishing secretory carcinoma of the salivary glands from its mimics. Histopathology. 2021;79:315–24. 18. Morand GB, Alsayegh R, Mlynarek AM, Plourde M, Mach T, Mascarella MA, Hier MP, Florianova L, Pusztaszeri MP. Application of the Milan system for reporting salivary gland cytopathology using cell blocks. Virchows Arch. 2022;481(4):575–83. https://doi.org/10.1007/ s00428-022-03364-x. Epub 2022 Jun 24. PMID: 35750873. 19. Rooper L, Sharma R, Bishop JA. Polymorphous low grade adenocarcinoma has a consistent p63+/p40− immunophenotype that helps distinguish it from adenoid cystic carcinoma and cellular pleomorphic adenoma. Head Neck Pathol. 2015;9:79–84. 20. Hsieh MS, Lee YH, Chang YL. SOX10-positive salivary gland tumors: a growing list, including mammary analogue secretory carcinoma of the salivary gland, sialoblastoma, low-grade salivary duct carcinoma, basal cell adenoma/adenocarcinoma, and a subgroup of mucoepidermoid carcinoma. Hum Pathol. 2016;56:134–42. 21. Mito JK, Jo VY, Chiosea SI, Dal Cin P, Krane JF. HMGA2 is a specific immunohistochemical marker for pleomorphic adenoma and carcinoma ex-pleomorphic adenoma. Histopathology. 2017;71:511–21. 22. Rooper LM, Lombardo KA, Oliai BR, Ha PK, Bishop JA. MYB RNA in situ hybridization facilitates sensitive and specific diagnosis of adenoid cystic carcinoma regardless of translocation status. Am J Surg Pathol. 2021;45:488–97. 23. Mino M, Pilch BZ, Faquin WC. Expression of KIT (CD117) in neoplasms of the head and neck: an ancillary marker for adenoid cystic carcinoma. Mod Pathol. 2003;16:1224–31. 24. Jo VY, Sholl LM, Krane JF. Distinctive patterns of CTNNB1 (β-catenin) alterations in salivary gland basal cell adenoma and basal cell adenocarcinoma. Am J Surg Pathol. 2016;40:1143–50. 25. Schmitt AC, Griffith CC, Cohen C, Siddiqui MT. LEF-1: diagnostic utility in distinguishing basaloid neoplasms of the salivary gland. Diagn Cytopathol. 2017;45:1078–83. 26. Schmitt AC, McCormick R, Cohen C, Siddiqui MT. DOG1, p63, and S100 protein: a novel immunohistochemical panel in the differential diagnosis of oncocytic salivary gland neoplasms in fine-needle aspiration cell blocks. J Am Soc Cytopathol. 2014;3:303–8. 27. Schmitt AC, Cohen C, Siddiqui MT. Expression of SOX10 in salivary gland oncocytic neoplasms: a review and a comparative analysis with other immunohistochemical markers. Acta Cytol. 2015;59:384–90.

8  Ancillary Studies for Salivary Gland Cytology

189

28. Canberk S, Onenerk M, Sayman E, Goret CC, Erkan M, Atasoy T, Kilicoglu GZ. Is DOG1 really useful in the diagnosis of salivary gland acinic cell carcinoma?—a DOG1 (clone K9) analysis in fine needle aspiration cell blocks and the review of the literature. Cytojournal. 2015;12:18. 29. Nakaguro M, Tanigawa M, Hirai H, Yamamoto Y, Urano M, Takahashi RH, Sukeda A, Okumura Y, Honda S, Tasaki K, Shimizu A, Tsukahara K, Tada Y, Matsubayashi J, Faquin WC, Sadow PM, Nagao T. The diagnostic utility of RAS Q61R mutation-specific immunohistochemistry in epithelial-myoepithelial carcinoma. Am J Surg Pathol. 2021;45:885–94. 30. Lewis JS Jr, Beadle B, Bishop JA, et al. Human papillomavirus testing in head and neck carcinomas. Arch Pathol Lab Med. 2018;142:559–97. 31. Skálová A, Stenman G, Simpson RHW, Hellquist H, Slouka D, Svoboda T, Bishop JA, Hunt JL, Nibu KI, Rinaldo A, Vander Poorten V, Devaney KO, Steiner P, Ferlito A.  The role of molecular testing in the differential diagnosis of salivary gland carcinomas. Am J Surg Pathol. 2018;42:e11–27. 32. Todorovic E, Weinreb I.  Intraductal carcinomas of the salivary gland. Surg Pathol Clin. 2021;14:1–15. 33. Andreasen S, Melchior LC, Kiss K, Bishop JA, Høgdall E, Grauslund M, Wessel I, Homøe P, Agander TK. The PRKD1 E710D hotspot mutation is highly specific in separating polymorphous adenocarcinoma of the palate from adenoid cystic carcinoma and pleomorphic adenoma on FNA. Cancer Cytopathol. 2018;126:275–81. 34. Bishop JA, Sajed DP, Weinreb I, Dickson BC, Bilodeau EA, Agaimy A, Franchi A, Khurram SA, Da Forno P, Robledo J, Kalmar JR, Aguirre S, Krane JF, Tapia JL, Kiss K, Cordell K, Rosebush M, Barrett AW, Oda D, Assaad A, Nagao T, Kawakami F, Nakaguro M, Zahir I, Wakeman K, Ihrler S, Chenevert J, Lin YL, Westra WH, Gagan J, Rooper LM. Microsecretory adenocarcinoma of salivary glands: an expanded series of 24 cases. Head Neck Pathol. 2021;15:1192–201. 35. Bishop JA, Koduru P, Veremis BM, Oliai BR, Weinreb I, Rooper LM, Dickson BC, Demicco EG. SS18 break-apart fluorescence in situ hybridization is a practical and effective method for diagnosing microsecretory adenocarcinoma of salivary glands. Head Neck Pathol. 2021;15:723–6. 36. Pisapia P, Pepe F, Sgariglia R, Nacchio M, Russo G, Conticelli F, Girolami I, Eccher A, Bellevicine C, Vigliar E, Malapelle U, Troncone G. Next generation sequencing in cytology. Cytopathology. 2021;32:588–95.

Chapter 9

Imaging of the Salivary Glands Jillian W. Lazor and Joanie M. Garratt

 verview of Salivary Gland Imaging: Indications O and Modalities Using conventional imaging, there can be considerable overlap in the features of benign and malignant salivary masses [1, 2]. However, while imaging may not establish a definitive histologic diagnosis for a salivary lesion, imaging plays an important role in patients with suspected salivary gland disease. Imaging can be used to: • Confirm the salivary gland as the site of the mass • Narrow the list of diagnostic considerations • Delineate the local anatomic extent of the pathological process and, in cases of malignancy, evaluate for regional and distant metastases • Direct tissue sampling • Guide surgical and/or radiation planning • Assess for recurrence in the setting of treated malignancy Computed tomography (CT) and magnetic resonance imaging (MRI) are the primary modalities for salivary gland imaging. However, ultrasound (US) and nuclear and molecular imaging techniques may also prove helpful in certain clinical scenarios [3].

Jillian W. Lazor and Joanie M. Garratt are co-first authors. J. W. Lazor (*) · J. M. Garratt (*) Department of Radiology, The Hospital of the University of Pennsylvania, Philadelphia, PA, USA e-mail: [email protected]; [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 W. C. Faquin et al. (eds.), The Milan System for Reporting Salivary Gland Cytopathology, https://doi.org/10.1007/978-3-031-26662-1_9

191

192

J. W. Lazor and J. M. Garratt

Computed Tomography CT is the preferred radiologic modality in the setting of suspected inflammatory and infectious salivary gland conditions given the technique’s widespread availability, rapid speed of image acquisition, large field of view, high spatial resolution, and relative insensitivity to motion artifact [4]. CT is also more sensitive than MRI for the detection of calcified sialoliths [5]. In cases of a suspected neoplasm, CT is inferior to MRI for characterizing masses and defining their extent [5]. However, CT will provide a better assessment than MRI of subtle cortical erosion due to neoplastic involvement at the skull base [6], so CT may be used in a complementary manner with MRI in patients with malignancy abutting the skull base. Disadvantages of CT scanning include exposure to ionizing radiation and potential obscuration of portions of the salivary glands by artifact from dental amalgam. Unless the patient has a contrast allergy or renal insufficiency, CT scanning of salivary gland lesions should be performed with intravenous contrast; contrast will permit distinction between phlegmon and abscess and may help characterize some masses [5, 7]. Calcified sialoliths are not obscured by intravenous contrast. However, if the clinical indication is solely to determine the presence or absence of sialoliths or cortical neoplastic erosion, contrast may be omitted [3]. While the field of view could be limited to only the maxillofacial structures, imaging from the skull base to the thoracic inlet is preferred to assess the full extent of an infectious or inflammatory process and to evaluate for regional metastases in cases of malignancy [3]. CT imaging should be performed with at most 3 mm slice thickness. Axial, coronal, and sagittal reformations should be provided with both soft tissue and bone reconstruction algorithms.

Magnetic Resonance Imaging MRI is the first-line imaging modality for a suspected salivary neoplasm. MRI offers improved soft tissue contrast over CT—better defining the margins of a tumor and its pattern of infiltration—and is somewhat better at distinguishing benignity from malignancy. MRI is also superior for the identification of perineural spread, marrow involvement, and intracranial extension of tumor [5, 7]. MR sialography, a technique that delineates the ductal system of the salivary glands using high resolution, highly fluid-sensitive sequences, may be helpful in the evaluation of chronic sialadenitis through the depiction of ductal stenoses, irregularities, diverticula, and cysts [7]. Disadvantages of MRI include the relatively high cost, long acquisition time, sensitivity to patient motion, poor tolerability in claustrophobic patients, and contraindication in patients with some metallic foreign bodies and implanted metallic medical devices. MRI for a suspected neoplasm should be performed with intravenous contrast, though some information regarding tumor extent and characterization can be

9  Imaging of the Salivary Glands

193

obtained by MRI without contrast [3]. Characterization of the neoplasm and evaluation of its local extent can be assessed with an MRI face or MRI neck, and an MRI neck should be performed if local nodal staging is desired. Imaging can be performed at 1.5T or 3T. When imaging the face, the field of view should cover the top of the frontal sinus to the chin and the nose to the anterior pons with a maximum slice thickness of 4 mm; sequences should include pre-contrast axial and coronal T1-weighted images, axial and coronal T2-weighted images with fat saturation, axial diffusion-weighted images, and axial and coronal postcontrast T1-weighted images with fat saturation. When imaging the neck, the field of view should cover the top of the frontal sinus to the thoracic inlet, with a maximum slice thickness of 5 mm; sequences should include sagittal and axial pre-contrast T1-weighted images, axial T2-weighted images, axial T2-weighted images with fat saturation, axial diffusion-­weighted images, axial postcontrast T1-weighted images, and coronal postcontrast T1-weighted images with fat saturation.

Ultrasound US is a useful technique for imaging of the salivary glands in pediatric patients, since it does not require ionizing radiation or sedation, and can also be helpful in adults in some circumstances. US can localize a disease process to the salivary gland, differentiate focal from diffuse disease, distinguish solid from cystic masses, assess the vascularity of lesions, and evaluate for the involvement of cervical lymph nodes. US is especially useful in guiding tissue sampling either by FNA or core biopsy [8, 9]. A notable disadvantage of US is the inability to adequately visualize large, trans-spatial masses involving deep spaces of the neck, particularly areas deep to bone such as the deep lobe of the parotid gland and parapharyngeal space [3]. US of the salivary glands is performed using high-frequency (5–12 MHz) linear probes. The gland should be scanned in at least two planes and bilateral glands should be assessed. Color Doppler may be used in the setting of focal mass or inflammatory disease. When there is a concern for malignancy, the remainder of the neck can be evaluated for lymph node involvement [9, 10].

Nuclear Scintigraphy and Positron Emission Tomography Technetium-99m pertechnetate scintigraphy can be used to assess salivary gland function in the setting of chronic sialadenitis, such as that seen in Sjogren’s syndrome, chronic obstructive sialadenitis, and following radiation therapy [11, 12]. Technetium-99m pertechnetate scintigraphy may also be used to characterize certain tumors such as Warthin tumor or oncocytoma as these tumors demonstrate avid radiotracer uptake [5, 11].

194

J. W. Lazor and J. M. Garratt

F-18 fluorodeoxyglucose positron emission tomography/computed tomography (18F-FDG PET/CT) has little utility for the characterization of salivary gland masses, since benign masses such as pleomorphic adenoma (PA) and Warthin tumor (WT) can demonstrate high 18F-FDG uptake, and some malignancies such as adenoid cystic carcinoma (AdCC), mucoepidermoid carcinoma (MEC), and malignancies of small size may exhibit low 18F-FDG uptake similar to that of the background salivary gland parenchyma [7, 13, 14]. However, 18F-FDG PET/CT is useful for the identification of cervical nodal and distant metastases in a newly diagnosed salivary gland malignancy and for the detection of locoregional and distant recurrence in treated patients [13].

Normal Imaging Anatomy of the Major Salivary Glands While each of the major salivary glands is distinctive with respect to function, histology, and anatomy, there is a similarity in imaging appearance. The parotid and submandibular glands occupy much of the parotid and submandibular areas of the suprahyoid space, and the sublingual glands lie in the sublingual space [15]. Each of these three pairs of major salivary glands should appear symmetric [16].

Parotid Gland The largest of the salivary glands, the parotid gland, is encapsulated and has several lobulations or recesses, including an inferior projection known as the “tail”; additionally, unilateral or bilateral accessory parotid tissue is common [15, 16]. The parotid gland is mostly superficial to the masseter muscle and mandible. The excretory (Stensen’s) duct courses anteriorly from the parotid gland, overlies the masseter muscle and enters the oral cavity at the level of the second molar [8, 15, 17]. There is no true anatomic division of the parotid gland; however, the plane of the facial nerve divides the gland into superficial and deep lobes, and the retromandibular vein or posterior margin of the mandible may be used as an imaging landmark to approximate this plane on imaging [15–17]. The deep lobe enters the parapharyngeal space, posterior and deep to the mandible [16]. By US, the normal parotid gland appears homogenous and, secondary to intraglandular fatty tissue, hyperechoic relative to the adjacent musculature [15, 17, 18] (Fig.  9.1). The degree of intraglandular fat determines echogenicity [18]. Intra-­ parotid ducts and lymph nodes may be seen within a normal parotid gland on US. The location of the parotid gland allows high-resolution sonographic imaging of the superficial lobe; however, shadowing from the overlying mandibular ramus and suppression of sound waves by high-fat content precludes visualization of the deep lobe [15, 17]. Cross-sectional imaging with CT or MRI is necessary for accurate evaluation of the deep parotid lobes [17].

9  Imaging of the Salivary Glands

195

a

b

c

d

e

f

Fig. 9.1  Normal appearance of the parotid glands (arrows) by US (a), axial (f) and coronal (c) contrast enhanced-CT, and axial T1-weighted (d) and T2-weighted (e) MRI. Normal appearance of the submandibular gland by US (b)

CT and MRI also demonstrate the fatty nature of the parotid gland. On CT, the attenuation of the parotid gland is approximately 15–50 Hounsfield units (HU), which is greater than the attenuation of fat (−125 to 50 HU) but less than that of muscle (35–60 HU) [16, 19]. The parotid gland appears slightly heterogeneous on CT secondary to thin hyper-attenuating strands throughout the gland, which correspond to normal interstitium (Fig.  9.1). CT does not allow visualization of fine anatomic detail, including the facial nerve and normal intraglandular ducts; abnormally dilated ducts may be seen by CT. Furthermore, CT cannot distinguish between a dense normal gland and a diffusely infiltrating process [16]. The intra-parotid retromandibular vein and external carotid artery however can be visualized by CT [17]. By MRI, because of its fat content, a normal parotid gland is typically low to intermediate signal intensity on T2-weighted imaging and intermediate to high signal intensity on T1-weighted imaging (Fig. 9.1). Interestingly, the signal intensity on T1-weighted imaging increases with age, which is not true of the other major salivary glands. Similar to CT, the parotid gland is normally heterogeneous on T1-weighted MRI, with intervening low signal intensity interstitial tissue and ducts in an otherwise intermediate to high signal intensity gland [16].

196

J. W. Lazor and J. M. Garratt

Submandibular Gland The submandibular gland is the second largest of the salivary glands. It is positioned posteriorly in the submandibular space, medial to the mandible, and wraps around the dorsal free edge of the mylohyoid muscle, the primary muscular support of the floor of the mouth. The excretory (Wharton’s) duct extends anteriorly, cranially, and medially in the sublingual space along the sublingual gland to the floor of the mouth. The anterior facial vein runs over the lateral aspect of the submandibular gland and separates it from the adjacent soft tissues and level IB lymph nodes, serving as an important marker for differentiating the origin of masses in this region [8, 15, 16]. The anterior facial artery, by comparison, runs through the submandibular gland [20]. On US, the submandibular gland is triangular in shape, homogeneous, and hyperechoic relative to adjacent musculature, although less echogenic than the parotid gland [15] (Fig. 9.1). Less fatty than the parotid gland, the submandibular gland is higher in attenuation on CT, and, therefore, appears homogeneous. On MRI, the submandibular gland appears slightly higher in signal on T2-weighted imaging and slightly lower in signal on T1-weighted imaging relative to the parotid gland [16].

Sublingual Gland The smallest of the major salivary glands, the sublingual gland is anteriorly located in the sublingual space [15, 16]. This gland is sandwiched between the mylohoid muscle and the muscles of the tongue root, the genioglossus, and geniohyoid muscles. It is without a single draining duct; instead, numerous ducts directly open into the floor of the mouth [16, 19]. On US, the sublingual gland is the shape of a flattened almond, homogenous, and hyperechoic, similar in echogenicity to the submandibular glands [15, 16]. On CT, the sublingual glands are homogeneous. By T1-weighted imaging, the sublingual glands are similar to the other major salivary glands, demonstrating higher signal intensity relative to muscle but lower signal intensity relative to surrounding fat; on T2-weighted imaging, the sublingual glands are high in signal intensity [16].

Imaging Features of Non-neoplastic Conditions Obstructive or inflammatory processes account for most non-neoplastic conditions of the major salivary glands [8]. Clinical history is imperative as there is ample overlap of imaging features. Imaging is most beneficial in excluding an underlying focal lesion.

9  Imaging of the Salivary Glands

a

b

197

c

Fig. 9.2  Sialolithiasis on US (a and b) and CT (c). Sagittal (a) and transverse (b) US images of the left parotid gland an echogenic calculus (white arrow) within the parotid duct with associated acoustic shadowing (dashed white arrow) and upstream dilation of Stensen’s duct (asterisk). Note the associated abnormal heterogeneity of the parotid gland parenchyma. Axial CT image (c) shows calculus in the parotid duct (white arrow) with associated duct dilation (black arrow)

Sialolithiasis Sialolithiasis is one of the most common diseases of the major salivary glands, particularly in the submandibular gland [5, 8]. Sialoliths can be visualized by all imaging modalities that are sensitive to calculi, including radiography, US, and CT. MRI is less sensitive in detecting calculi but allows better visualization of the ductal system and potential complications such as mucoceles [5]. By US, sialoliths appear as markedly hyperechoic points or lines with associated posterior acoustic shadowing [18]. Dilated excretory ducts may be visible if there is associated duct occlusion, typically seen if the patient is symptomatic [9, 18]. Sialoliths appear as dense calcifications on CT (Fig. 9.2).

Acute Sialadenitis Sonographically, acute inflammation of a salivary gland is characterized by glandular enlargement and hypoechogenicity. Diffuse heterogeneity, increased blood flow, and reactive lymph nodes may also be seen [18] (Fig.  9.3). If present, focal hypoechoic foci in the parotid gland typically correspond to enlarged lymph nodes [17, 18]. CT and MRI will also show glandular enlargement with concomitant inflammatory stranding, hyperenhancement, and abnormal attenuation or intensity (Fig. 9.4). Accompanying prominent intraglandular or adjacent lymph nodes that are high signal intensity on T2-weighed imaging suggests an inflammatory etiology [5]. In severe cases of acute sialadenitis, abscesses can develop. Imaging is especially helpful in both the diagnosis, as abscesses are difficult to diagnose clinically, and drainage of abscesses [18]. If an abscess is identified, an underlying stone must be

198

a

c

J. W. Lazor and J. M. Garratt

b

d

Fig. 9.3  Acute sialadenitis with reactive lymphadenopathy on CT (a) and US (b and c). Axial contrast-enhanced CT image (a) shows an oval hypoattenuating focus in the left parotid gland (arrow). Follow-up US shows an enlarged intra-parotid lymph node with thickened cortex (b) and heterogeneity of the parotid parenchyma (c), in keeping with sialadenitis with reactive lymph node. These findings resolved on US performed 1 month later. A sonographically normal lymph node is shown for comparison (d)

excluded, for which imaging is also useful [5]. The sonographic appearance of an abscess is a hypo- to anechoic structure with posterior acoustic enhancement and ill-defined borders; central liquefaction with avascular, mobile, echogenic debris may be seen. Additional potential findings include small internal echogenic foci from internal gas and a peripheral hyperechoic halo [18]. On CT, an abscess will appear as a rim-enhancing organized structure with central hypoattenuation (Fig. 9.5).

9  Imaging of the Salivary Glands

a

199

b

Fig. 9.4  Acute sialadenitis on MRI (a and b). Axial T1 (a) and T2 STIR (b) images on MRI of the neck show diffuse heterogeneity and enlargement of the left parotid gland (arrow) with inflammatory changes of the overlying soft tissues (dashed arrow), consistent with acute sialadenitis of the left parotid gland and associated cellulitis

a

b

Fig. 9.5  Acute sialadenitis with abscess on CT (a). Axial image from a contrast-enhanced CT (a) shows a peripherally enhancing, centrally hypoattenuating lesion in the right parotid gland (arrow), in keeping with a parotid abscess. Axial T2 STIR image (b) from follow-up MR performed 2 months later shows resolution (dashed arrow showing normal right parotid gland)

J. W. Lazor and J. M. Garratt

200

Chronic Sialadenitis Recurrent and chronic inflammation and edema of the salivary glands leads to parenchymal degeneration and fibrosis, presenting as gland heterogeneity on imaging [18]. Although gland size is variable, an atrophic gland may be seen [9, 18]. The gland will typically appear diffusely hypoechoic or have multi-hypoechoic foci on US [5, 18]. Intraglandular calcifications are seen in many cases, but acalculus chronic sialadenitis does occur. On CT and MRI, the gland can be more fatty-­ replaced and small in size in some cases (Fig. 9.6); in others, it can appear more dense with loss or coarsening of normal architecture [5] (Fig. 9.7). The lesion known as Kuttner tumor, a pseudotumor in the setting of chronic sclerosing sialadenitis, is classically seen in the submandibular gland. On ultrasound, findings may be focal, with a solitary heterogenous hypoechoic focus in an otherwise normal gland, or diffuse, with heterogeneity throughout the gland secondary to scattered hypoechoic foci [18].

Granulomatous Sialadenitis The imaging findings from granulomatous sialadenitis are non-specific. On US, the gland may be normal in size or enlarged, either with diffuse hypoechogenicity or multiple small hypoechoic foci. Hyperemia may be present [5, 18]. Involvement of a

b

Fig. 9.6  Chronic sialadenitis on MRI (a and b). Axial T2 MR images show involution of the submandibular glands (a) (arrow at expected location of the glands) and atrophy of the parotid glands (b) (dashed arrow) in a patient with chronic sialadenitis

9  Imaging of the Salivary Glands

a

201

b

Fig. 9.7  Chronic sialadenitis. Heterogenous bilateral parotid glands with a coarsened appearance on pre-contrast T1 (a) and T2 (b) weighted MR images in a patient with chronic sialadenitis

the parotid gland by sarcoidosis is frequently bilateral and may be the initial and only finding of the disease [21]. The sonographic imaging presentation is typically non-specific parotitis [5]. On CT and MRI, sarcoidosis manifests as many non-­ cavitating, benign-appearing masses throughout an enlarged salivary gland. Patients with sarcoidosis are more susceptible to glandular and nodal calcifications [5, 9]. Several granulomatous diseases mimic salivary gland neoplasms on imaging [18]. Mycobacterial infection, for instance, may present sonographically as a heterogeneous, hypoechoic lesion with ill-defined margins and adjacent lymphadenopathy. The parenchymal form of tuberculosis manifests as very hypoechoic to anechoic glandular foci that may have central cavitation with no flow on color Doppler imaging, with the latter feature distinguishing them from neoplasms [18, 21]. Actinomycosis infection also has an infiltrative sonographic appearance with ill-defined margins, most often involving the submandibular gland [9, 18].

Reactive Lymph Node Sonographic features of lymph nodes on B mode (grayscale) in combination with color Doppler imaging is fairly accurate in diagnosing reactive lymph nodes [22]. On US, traits of reactive lymph nodes include oval or reniform shape with an echogenic hilum, unsharp border, central vascularity as well as absence of necrosis, matting, and adjacent soft tissue edema [22] (Fig. 9.3). In acute inflammation, reactive lymph nodes may show increased central blood flow [18]. Abnormal lymph nodes, such as lymphomatous nodes, tend to be rounded, heterogeneously hypoechoic with sharp borders, with an absent hilum and mixed vascularity [22].

J. W. Lazor and J. M. Garratt

202

Benign Lymphoepithelial Lesion/Lymphoepithelial Sialadenitis In the setting of HIV, parotid involvement may be seen as one of two sonographic patterns. Most patients have numerous small anechoic or hypoechoic areas without posterior acoustic enhancement. Nearly one-third of patients will demonstrate multiple anechoic foci with posterior acoustic enhancement, likely related to characteristic small lymphoepithelial cysts; these lesions have classic sonographic features of a cyst, including anechoic contents, posterior acoustic enhancement, well-defined margins, and no internal flow on color Doppler imaging. On CT and MRI, numerous solid and cystic masses are seen within bilateral enlarged parotid glands, often with associated lymphadenopathy [21] (Fig. 9.8). a

d

b

e

c

Fig. 9.8  Autoimmune sialadenitis on US (a–c) and CT (d and e). Diffuse, symmetric, heterogeneity, and nodular enlargement of the parotid glands on US (a, c) and CT (d, e) in a patient with bilateral facial swelling and pain. No findings of ductal dilation. Color Doppler imaging shows increased vascularity of the parotid gland (b), in keeping with hyperemia. A lymphoepithelial cyst is seen (white arrow) on US and CT (c, e)

9  Imaging of the Salivary Glands

203

In Sjögren’s syndrome, all of the salivary glands are typically involved [9]. Initially, the glands may appear normal across all modalities [21]; however, imaging findings will evolve with the stage of the disease. In later stages, the glands enlarge and develop lobulations. By US, there is diffuse heterogeneity with numerous small, rounded, or oval, anechoic to hypoechoic areas that are well-defined and have increased parenchymal blood flow [18]; these areas are thought to represent lymphocytic infiltrate, dilated ducts, and damaged glandular tissue [18, 21]. This diffuse heterogeneity of the salivary gland on US is indicative of an advanced stage of Sjögren’s syndrome. Using CT, the salivary glands in Sjögren’s syndrome may appear enlarged, heterogeneous, and hyper-attenuating [21]. In advanced Sjögren’s syndrome, the salivary glands develop a characteristic “honeycomb” or “salt and pepper” appearance on MRI, in which there are multiple discrete foci of high signal intensity intermixed with areas of low signal intensity on T1- and T2-weighted imaging (Fig. 9.9). It is a

b

c

d

Fig. 9.9  Chronic sialadenitis on CT (a and b) and MRI (c and d). Axial (a) and coronal (b) CT images show diffuse enlargement and heterogeneity of both parotid glands (dashed white arrows). Multiple small mixed solid and cystic masses are better delineated on axial T2 STIR (c) and unenhanced T1 (d) MR images; the largest mass measures 2.2 cm and is located in the left parotid gland (white arrow). Bilateral parotid calcifications are evident on CT (black arrow). Findings are suggestive of Sjogren syndrome or other chronic sialadenitis with bilateral lymphoepithelial lesions. No focal mass of the submandibular glands (black asterisks)

204

J. W. Lazor and J. M. Garratt

thought that this appearance is secondary to fat deposits, small cysts, and sialectatic changes [21, 23]. MR sialography shows changes in ductal morphology from Sjögren’s syndrome [21, 24]. It is important to note that there is a higher incidence of lymphoma in patients with Sjögren’s syndrome; however, the background gland heterogeneity on imaging of these patients precludes optimal assessment for small lesions [18].

Sialadenosis Most often affecting the parotid glands, sialadenosis manifests on ultrasound as a homogeneous gland with diffuse hyperechogenicity. There is no focal lesion or increased blood flow [18, 25]. The glands may be enlarged [9].

Overview of Imaging for Salivary Gland Neoplasms There is considerable overlap in the imaging appearance of benign and malignant tumors on CT, MRI, US, and 18F-FDG PET/CT [1, 2, 5, 10, 13, 14]. Nevertheless, some imaging features may be useful for differentiation: • Ill-defined margins; invasion of adjacent fat, muscle, or bone; and perineural spread of tumor are suggestive of malignancy [5, 26]. However, infectious and inflammatory pathologies may demonstrate ill-defined borders [26], and small malignancies may feature smooth borders [10]. • Malignant neoplasms are often low or intermediate in T2 intensity, while benign tumors are frequently T2 hyperintense. Of note, low-grade MEC and some AdCCs may be T2 hyperintense, and WTs are often intermediate or low in T2 signal [5, 26]. • Intra-tumoral cysts are more common in benign salivary neoplasms than malignant, though a central location of the cyst is more frequent in malignant tumors. Intrinsic T1 hyperintensity of the intra-tumoral cyst contents is more often seen in benign tumors [27]. Thick and irregular wall enhancement is more commonly seen in malignant neoplasms than benign [28]. • Benign tumors often have a higher apparent diffusion coefficient (ADC values) than malignant tumors on diffusion-weighted MRI, and low or extremely low ADC values often involve greater than 60% of the tumor area in cases of malignancy. However, WTs can be indistinguishable from malignant tumors on diffusion-­weighted MRI [29]. • Malignant salivary tumors often exhibit higher 18F-FDG uptake at 18F-FDG PET/ CT than benign tumors. Nevertheless, some benign masses such as PAs and WTs

9  Imaging of the Salivary Glands

205

may demonstrate high 18F-FDG uptake, and some malignancies such as AdCC, MEC, and malignancies of small size may exhibit low 18F-FDG uptake near that of background salivary parenchyma [7, 13, 14]. • High-grade malignancies tend to show higher 18F-FDG uptake at 18F-FDG PET/ CT than low- and intermediate-grade malignancies, though there is substantial overlap [13]. • Pathologic lymph nodes in the drainage territory of a mass are suggestive of malignancy [18]. MR and CT perfusion imaging, dual-energy CT, and MR spectroscopy may have a role in distinguishing benign and malignant tumors, though these techniques require further research validation and are not routinely used in the clinical setting [7, 30].

Imaging Features of Selected Benign Neoplasms Pleomorphic Adenoma By US, PA appear as hypoechoic masses with posterior acoustic enhancement and well-circumscribed smooth or lobulated borders. Power Doppler ultrasound will demonstrate mild peripheral vascularity in a PA, with a relatively hypovascular center [18]. PA appear as low- or soft tissue density masses on CT with circumscribed, lobulated borders and heterogeneous enhancement [27]. On MRI, PA characteristically exhibit well-defined margins, a lobulated contour, striking T2 hyperintensity, heterogeneous nodular enhancement, and a T2 hypointense rim; the presence of all of these features is 95% specific for PA, but less than half of PA demonstrate all features [31] (Fig. 9.10). Less commonly, PA may feature areas of cystic degeneration (anechoic areas on US, areas of low density and nonenhancement on CT, and areas of T2 hyperintensity and nonenhancement on MRI), calcification (foci of hyperechogenicity on US and hyperdensity on CT), or lipometaplasia (regions of fat density on CT and T1 hyperintensity on MRI) [27].

Warthin Tumor WTs occur almost exclusively in the parotid gland/peri-parotid region, and most often in the parotid tail [5]. On US, WTs are round or oval, well-circumscribed, and hypoechoic, sometimes with anechoic cystic regions. WTs feature prominent vascularity on power Doppler US, which is often seen centrally or centrally and peripherally [32]. WTs demonstrate well-circumscribed margins at CT and appear as solid

206

J. W. Lazor and J. M. Garratt

a

b

c

d

Fig. 9.10  Pleomorphic adenoma of the left parotid gland as seen on MRI (a–c) and US (d). T2 (a) and STIR (b) MR images show a well-circumscribed, lobulated T2 hyperintense left parotid pleomorphic adenoma with a T2 hypointense rim (yellow arrow). Postcontrast T1-weighted MRI (c) shows nodular enhancement within the mass. US (d) demonstrates a well-­circumscribed, lobulated hypoechoic mass with posterior acoustic enhancement (red arrow)

or cystic masses. Regions of intrinsic T1 hyperintensity and T2/STIR hypo-intensity within a mass are suggestive of WT, correlating to proteinaceous cystic components. Low ADC values within WT are also characteristic [33]. WTs demonstrate intense radiotracer uptake on 18F-FDG PET/CT and technetium-99m pertechnetate scintigraphy [14] (Fig. 9.11).

Oncocytoma On US, oncocytomas appear as round or ovoid, well-circumscribed, hypoechoic masses, sometimes with anechoic cysts or hyperechoic regions of intralesional fat. Doppler US may demonstrate intra-tumoral vessels, which can be oriented in a

9  Imaging of the Salivary Glands

207

a

b

c

d

e

Fig. 9.11  Bilateral Warthin tumors demonstrated on MRI (a–c) and a single Warthin tumor depicted on US (d) and fused 18F-FDG PET/CT image (e). Unenhanced T1-weighted MRI (a) shows regions of intrinsic T1 hyperintensity within bilateral parotid masses (yellow arrows), and some of these correlate with areas of low T2 signal intensity (red arrows) on T2-weighted MRI (b), reflecting proteinaceous cystic components. Contrast-enhanced T1-weighted images show superimposed enhancement in the bilateral well-delineated masses (white arrows). US (d) demonstrates a well-circumscribed hypoechoic mass with small internal cystic areas (yellow arrows) and internal vascularity (red arrow). The Warthin tumor (white arrow) demonstrates avid radiotracer uptake on 18F-FDG PET/CT (e)

spoke-wheel configuration [34]. At CT, oncocytomas demonstrate well-defined margins and relatively homogeneous enhancement, though when large, a non-­ enhancing curvilinear cleft or cyst may be apparent [35]. The MRI appearance may be most helpful in distinguishing these masses: oncocytomas appear hypointense to the remainder of the gland on T1-weighted images, but are isointense to the gland

208

a

J. W. Lazor and J. M. Garratt

b

c

Fig. 9.12  Right parotid oncocytoma depicted on MRI (a–c). T1-weighted MR image (a) demonstrates a well-circumscribed mass in the right parotid tail (red arrow). This mass is inconspicuous on the STIR (b) and postcontrast T1-weighted MR images (c) (yellow arrows), reflecting the “vanishing” appearance of oncocytoma on MRI

on fat-saturated T2 and postcontrast T1-weighted images, earning them the nickname of “vanishing tumors” [36] (Fig.  9.12). Oncocytomas retain radiotracer on 18 F-FDG PET/CT and technetium-99m pertechnetate scintigraphy [14].

Lipoma A lipoma, on US, is oval and hypoechoic compared to the remainder of the gland, with circumscribed margins. Linear hyperechoic striations are often evident throughout the lesion [18]. Lipomas will appear as circumscribed masses on CT and MRI, with density similar to subcutaneous fat (around −100 Hounsfield units) on CT and will follow the signal intensity of subcutaneous fat on all sequences at MRI [37] (Fig. 9.13).

Imaging Features of Selected Malignant Neoplasms Mucoepidermoid Carcinoma (MEC) The imaging appearance of MEC varies by grade. Low-grade MEC and even some small MEC may mimic the imaging appearance of a benign tumor, with well-­ circumscribed margins, high T2 signal intensity on MRI, and internal cystic areas which may be proteinaceous/T1 hyperintense on MRI. Higher grade MEC tends to demonstrate poorly defined borders, more solid appearance, and low to intermediate T2 signal intensity on MRI.  At US, both low-grade and higher grade MEC can demonstrate heterogeneous hypoechoic echotexture and absence of posterior acoustic enhancement [18, 38–40] (Fig. 9.14).

9  Imaging of the Salivary Glands

a

209

b

c

d

Fig. 9.13  Left parotid lipoma on CT (a) and MRI (b–d). The lipoma demonstrates the same density as subcutaneous fat on CT (a; yellow arrow) and follows the signal intensity of subcutaneous fat on T1 (b), STIR (c), and T2-weighted (d) MR images (white arrows)

Adenoid Cystic Carcinoma (AdCC) AdCC can also mimic benign tumors on imaging. Low-grade neoplasms tend to be T2 hyperintense on MRI, while higher grade tumors demonstrate T2 hypo-intensity. Margins on imaging are often ill-defined (Fig.  9.15). High-grade tumors may be associated with adjacent osseous destruction, manifesting as osseous erosion on CT and marrow signal abnormality on MRI.  Perineural spread may also be evident, depicted as thickening and enhancement of the involved nerve on MRI [18, 39–41].

210

J. W. Lazor and J. M. Garratt

a

b

c

d

Fig. 9.14  Intermediate-grade mucoepidermoid carcinoma of the left parotid gland presents as a well-circumscribed, thick-walled cystic mass (yellow arrows) on CT (a) and US (b). An intermediate-­grade mucoepidermoid carcinoma of the right submandibular gland in another patient (white arrows) manifests as a solid mass on CT (c) and an irregular, hypoechoic, solid mass on US (d)

Acinic Cell Carcinoma (ACC) The imaging of ACC often mimics that of benign tumors and other low-grade malignant tumors. ACC tends to be well-circumscribed on US, CT, and MRI, though some tumors demonstrate infiltrative margins. The carcinoma may be solid, cystic, or cystic with a mural nodule. Solid tumors exhibit variable degrees of enhancement and often demonstrate small regions of hypo-enhancement on CT and MRI, which correlate to microcysts or necrosis. The tumors are hypoechoic, heterogeneous, and relatively avascular on Doppler US [42, 43].

9  Imaging of the Salivary Glands

211

a

b

c

d

Fig. 9.15  An adenoid cystic carcinoma of the left parotid gland (yellow arrows) depicted on MRI (a–d). The mass demonstrates infiltrative margins on T1-weighted MRI (a), appears relatively T2 hypointense on T2-weighted and STIR MRI (b and c), and exhibits homogeneous mild enhancement on postcontrast T1-weighted MRI (d)

Secretory Carcinoma (SC) SC shares many imaging characteristics with ACC. However, SC is more frequently cystic than ACC [44]. Layering T2 hypo-intensity within cystic components, reflecting layering hemorrhage, helps distinguish SC from ACC [45].

Carcinoma Ex Pleomorphic Adenoma Carcinoma ex pleomorphic adenoma (CA-ex-PA) is difficult to distinguish on imaging from PA and other benign tumors in the absence of invasion beyond the tumor capsule [46–48]. Invasive CA-ex-PA tends to exhibit an irregular shape, a poorly defined border, and loss of posterior acoustic enhancement at ultrasound [46]. Similarly, invasive CA-ex-PA may demonstrate partially or circumferentially poorly defined margins at CT and MRI. Intermediate and low ADC values may be suggestive of CA-ex-PA with or without invasion [47, 48].

J. W. Lazor and J. M. Garratt

212

Lymphoma At CT and MRI, lymphoma involvement may manifest as diffuse infiltration and enlargement of a salivary gland or as single or multiple enlarged lymph nodes within the gland [49]. In cases with diffuse infiltration of the gland, lymphoma on US will appear as single or multiple hypoechoic and homogeneous or inhomogeneous masses with irregular borders and increased vascularity, an appearance which can mimic other neoplasms and inflammatory conditions. Lymph nodes involved with lymphoma on US may be circumscribed and anechoic with posterior acoustic enhancement on ultrasound, which can mimic a simple cyst [18, 50, 51] (Fig. 9.16). aa

c c

b

b

d d

Fig. 9.16  Lymphoma involving the parotid gland in two different patients shown on CT (a), US (b), and MRI (c and d). Lymphoma presents as an infiltrative mass involving the right parotid gland and peri-parotid soft tissues (yellow arrow) in one patient on unenhanced CT (a). Lymphoma in a second patient appears as a pathologic lymph node with a thickened cortex and posterior acoustic enhancement (white arrow) on US (b) and as a well-circumscribed, T2 hyperintense, enhancing nodule (red arrows) on STIR (c) and postcontrast T1-weighted (d) MR images

9  Imaging of the Salivary Glands

213

Fig. 9.17  A centrally necrotic left parotid lymph node metastasis on CT (yellow arrow) in a patient who had recently undergone Mohs resection of a left cheek squamous cell carcinoma (postsurgical changes seen at white arrow)

Metastases The parotid gland contains lymphatic channels and lymph nodes so that neoplasms from other head and neck sites and also from distant organs can metastasize to the parotid gland. These metastases will present on imaging as single or multiple enlarged or enlarging lymph nodes within the gland. Some nodal metastases may demonstrate central necrosis and/or infiltrative borders [49] (Fig. 9.17).

Image-Guided FNA and Core Biopsy Considerations As described earlier in Chap. 1, image guidance may be a useful adjunct for FNA and core biopsy of the salivary glands [20, 52, 53]. US guidance is particularly advantageous, given its low cost, portability, versatility, multiplanar scanning, and high-resolution imaging of the salivary glands. For instance, sonographic visualization of the intra-parotid retromandibular vein allows the operator to avoid the adjacent facial nerve for parotid lesion FNAs and core biopsies [20]. Also, US can discern whether a lesion is within or adjacent to the salivary gland, as well as differentiate cystic from solid lesions [54]. Three-dimensional knowledge of anatomy, review of other imaging modalities, and prebiopsy US evaluation are essential to accurately identify the appropriate lesion prior to tissue sampling. Pre-procedural US also allows the operator to choose

214

J. W. Lazor and J. M. Garratt

an US transducer that will attain the best resolution imaging, fit the patient’s neck contour, and, if applicable, access a tight acoustic window. The proceduralist must adjust the imaging parameters for optimal visibility and assess for vascularity in the region of interest with color Doppler. The proceduralist should plan the biopsy approach even before entering the skin—including the FNA/core biopsy route to avoid other anatomic structures, needle length, angle of entry of the needle, and maximum depth to avoid overshooting [20]. US-guided FNA of salivary gland masses and adjacent lymph nodes is diagnostically effective; collaboration with on-site cytopathology for ROSE further improves diagnostic accuracy. Furthermore, this technique is minimally invasive with some studies reporting no intraprocedural or postprocedural complications [54]. US however is not suitable for image-guided FNA/core biopsy of lesions deep into bone, calcifications, or metallic hardware given acoustic shadowing from these structures. Also, US is limited by the acoustic window; lesions deeper than the field of view for the chosen transducer are not amenable to US-guided FNA/core biopsy. In these cases, CT-guided biopsy may be an option [20, 52, 53].

References 1. Berg HM, Jacobs JB, Kaufman D, Reede DL. Correlation of fine needle aspiration biopsy and CT scanning of parotid masses. Laryngoscope. 1986;96(12):1357–62. 2. Freling NJ, Molenaar WM, Vermey A, Mooyaart EL, Panders AK, Annyas AA, et  al. Malignant parotid tumors: clinical use of MR imaging and histologic correlation. Radiology. 1992;185(3):691–6. 3. Aulino JM, Kirsch CFE, Burns J, Busse PM, Chakraborty S, Choudhri AF, et al. ACR appropriateness criteria® neck mass-adenopathy. J Am Coll Radiol. 2019;16(5):S150–60. 4. Brucker JL, Gentry LR.  Imaging of head and neck emergencies. Radiol Clin North Am. 2015;53(1):215–52. 5. Yousem DM, Kraut MA, Chalian AA.  Major salivary gland imaging. Radiology. 2000;216(1):19–29. 6. Horowitz SW, Leonetti JP, Azar-Kia B, Fine M, Izquierdo R.  CT and MR of temporal bone malignancies primary and secondary to parotid carcinoma. AJNR Am J Neuroradiol. 1994;15(4):755–62. 7. Atkinson C, Fuller J, Huang B. Cross-sectional imaging techniques and normal anatomy of the salivary glands. Neuroimaging Clin N Am. 2018;28(2):137–58. 8. Abdullah A, Rivas FFR, Srinivasan A.  Imaging of the salivary glands. Semin Roentgenol. 2013;48(1):65–74. 9. Gritzmann N, Rettenbacher T, Hollerweger A, Macheiner P, Hübner E.  Sonography of the salivary glands. Eur Radiol. 2003;13(5):964–75. 10. Zengel P, Schrötzlmair F, Reichel C, Paprottka P, Clevert DA. Sonography: the leading diagnostic tool for diseases of the salivary glands. Semin Ultrasound CT MR. 2013;34(3):196–203. 11. Grove AS, Di Chiro G.  Salivary gland scanning with technetium 99m pertechnetate. Am J Roentgenol Radium Ther Nucl Med. 1968;102(1):109–16. 12. Wu C-B, Xi H, Zhou Q, Zhang L-M. The diagnostic value of technetium 99m Pertechnetate salivary gland scintigraphy in patients with certain salivary gland diseases. J Oral Maxillofac Surg. 2015;73(3):443–50.

9  Imaging of the Salivary Glands

215

13. Roh J-L, Ryu CH, Choi S-H, Kim JS, Lee JH, Cho K-J, et al. Clinical utility of 18F-FDG PET for patients with salivary gland malignancies. J Nucl Med. 2007;48(2):240–6. 14. Uchida Y, Minoshima S, Kawata T, Motoori K, Nakano K, Kazama T, et  al. Diagnostic value of FDG PET and salivary gland scintigraphy for parotid tumors. Clin Nucl Med. 2005;30(3):170–6. 15. Radhakrishnan R, Kline-Fath BM. The pediatric head and neck. In: Diagnostic ultrasound. 5th ed. Elsevier; 2011. p. 1628–1671. 16. Som PM, Brandwein-Gensler MS. Anatomy and pathology of the salivary glands. In: Head and neck imaging. 5th ed. Elsevier; p. 2449–609. 17. Burke CJ, Thomas RH, Howlett D. Imaging the major salivary glands. Br J Oral Maxillofac Surg. 2011;49(4):261–9. 18. Bialek EJ, Jakubowski W, Zajkowski P, Szopinski KT, Osmolski A. US of the major salivary glands: anatomy and spatial relationships, pathologic conditions, and pitfalls. Radiographics. 2006;26(3):745–63. 19. Bryan R, Miller R, Ferreyro R, Sessions R.  Computed tomography of the major salivary glands. AJR Am J Roentgenol. 1982;139:547. 20. Learned KO, Lev-Toaff AS, Brake BJ, Wu RI, Langer JE, Loevner LA.  US-guided biopsy of neck lesions: the head and neck neuroradiologist’s perspective. Radiographics. 2016;36(1):226–43. 21. Lowe LH, Stokes LS, Johnson JE, Heller RM, Royal SA, Wushensky C, et  al. Swelling at the angle of the mandible: imaging of the pediatric parotid gland and periparotid region. Radiographics. 2001;21(5):1211–27. 22. Pattanayak S, Chatterjee S, Ravikumar R, Nijhawan VS, Sharma V, Debnath J.  Ultrasound evaluation of cervical lymphadenopathy: can it reduce the need of histopathology/cytopathology? Med J Armed Forces India. 2018;74(3):227–34. 23. Izumi M, Eguchi K, Ohki M, Uetani M, Hayashi K, Kita M, et al. MR imaging of the parotid gland in Sjögren’s syndrome: a proposal for new diagnostic criteria. Am J Roentgenol. 1996;166(6):1483–7. 24. Rzymska-Grala I, Stopa Z, Grala B, Gołębiowski M, Wanyura H, Zuchowska A, et al. Salivary gland calculi—contemporary methods of imaging. Pol J Radiol. 2010;75(3):25–37. 25. Patel CM, Hogarth K, Richards PS. Imaging of salivary glands 2013;16. 26. Christe A, Waldherr C, Hallett R, Zbaeren P, Thoeny H. MR imaging of parotid tumors: typical lesion characteristics in MR imaging improve discrimination between benign and malignant disease. Am J Neuroradiol. 2011;32(7):1202–7. 27. Kato H, Kanematsu M, Watanabe H, Mizuta K, Aoki M. Salivary gland tumors of the parotid gland: CT and MR imaging findings with emphasis on intratumoral cystic components. Neuroradiology. 2014;56(9):789–95. 28. Sakamoto M, Iikubo M, Kojima I, Sasano T, Mugikura S, Murata T, et al. Diagnostic value of capsule-like rim enhancement on magnetic resonance imaging for distinguishing malignant from benign parotid tumours. Int J Oral Maxillofac Surg. 2014;43(9):1035–41. 29. Eida S, Sumi M, Sakihama N, Takahashi H, Nakamura T. Apparent diffusion coefficient mapping of salivary gland tumors: prediction of the benignancy and malignancy. AJNR Am J Neuroradiol. 2007;28(1):116–21. 30. Abdel Razek AAK, Mukherji SK.  State-of-the-art imaging of salivary gland tumors. Neuroimaging Clin N Am. 2018;28(2):303–17. 31. Zaghi S, Hendizadeh L, Hung T, Farahvar S, Abemayor E, Sepahdari AR. MRI criteria for the diagnosis of pleomorphic adenoma: a validation study. Am J Otolaryngol. 2014;35(6):713–8. 32. Hamilton BE, Salzman KL, Wiggins RH, Harnsberger HR. Earring lesions of the parotid tail. AJNR Am J Neuroradiol. 2003;24(9):1757–64. 33. Ikeda M, Motoori K, Hanazawa T, Nagai Y, Yamamoto S, Ueda T, et al. Warthin tumor of the parotid gland: diagnostic value of MR imaging with histopathologic correlation. AJNR Am J Neuroradiol. 2004;25(7):1256–62.

216

J. W. Lazor and J. M. Garratt

34. Corvino A, Caruso M, Varelli C, Di Gennaro F, Pignata S, Corvino F, et al. Diagnostic imaging of parotid gland oncocytoma: a pictorial review with emphasis on ultrasound assessment. J Ultrasound. 2021;24(3):241–7. 35. Tan TJ, Tan TY. CT features of parotid gland oncocytomas: a study of 10 cases and literature review. AJNR Am J Neuroradiol. 2010;31(8):1413–7. 36. Patel ND, van Zante A, Eisele DW, Harnsberger HR, Glastonbury CM. Oncocytoma: the vanishing parotid mass. AJNR Am J Neuroradiol. 2011;32(9):1703–6. 37. Ethunandan M, Vura G, Umar T, Anand R, Pratt CA, Macpherson DW, et  al. Lipomatous lesions of the parotid gland. J Oral Maxillofac Surg. 2006;64(11):1583–6. 38. Gong X, Xiong P, Liu S, Xu Q, Chen Y. Ultrasonographic appearances of mucoepidermoid carcinoma of the salivary glands. Oral Surg Oral Med Oral Pathol Oral Radiol. 2012;114(3):382–7. 39. Kato H, Kanematsu M, Makita H, Kato K, Hatakeyama D, Shibata T, et al. CT and MR imaging findings of palatal tumors. Eur J Radiol. 2014;83(3):e137–46. 40. Shah GV. MR imaging of salivary glands. Magn Reson Imaging Clin N Am. 2002;10(4):631–62. 41. Sigal R, Monnet O, de Baere T, Micheau C, Shapeero LG, Julieron M, et al. Adenoid cystic carcinoma of the head and neck: evaluation with MR imaging and clinical-pathologic correlation in 27 patients. Radiology. 1992;184(1):95–101. 42. Li J, Gong X, Xiong P, Xu Q, Liu Y, Chen Y, et al. Ultrasound and computed tomography features of primary acinic cell carcinoma in the parotid gland: a retrospective study. Eur J Radiol. 2014;83(7):1152–6. 43. Suh S, Seol HY, Kim T-K, Lee NJ, Kim JH, Kim KA, et al. Acinic cell carcinoma of the head and neck: radiologic-pathologic correlation. J Comput Assist Tomogr. 2005;29(1):121–6. 44. Ding A, Zuo J, Lv H, Zhou L, Ding S, Yu Q, et al. Imaging identification of primary mammary analogue secretory carcinoma and acinic cell carcinoma in major salivary glands. Oral Dis. 2021;28:1861. 45. Kuwabara H, Yamamoto K, Terada T, Kawata R, Nagao T, Hirose Y.  Hemorrhage of MRI and Immunohistochemical panels distinguish secretory carcinoma from Acinic cell carcinoma. Laryngoscope Investig Otolaryngol. 2018;3(4):268–74. 46. Angang D, Jia L, Xia G, Ping X, Jiang L. Gray scale and doppler ultrasonography features of the carcinoma ex pleomorphic adenoma. Dento Maxillo Facial Radiol. 2018;47(4):20170268. 47. Wang C, Yu Q, Li S, Sun J, Zhu L, Wang P. Carcinoma ex pleomorphic adenoma of major salivary glands: CT and MR imaging findings. Dento Maxillo Facial Radiol. 2021;50(7):20200485. 48. Seok J, Hyun SJ, Jeong W-J, Ahn S-H, Kim H, Jung YH. The difference in the clinical features between carcinoma ex pleomorphic adenoma and pleomorphic adenoma. Ear Nose Throat J. 2019;98(8):504–9. 49. Lobo R, Hawk J, Srinivasan A.  A review of salivary gland malignancies: common histologic types, anatomic considerations, and imaging strategies. Neuroimaging Clin N Am. 2018;28(2):171–82. 50. Yasumoto M, Yoshimura R, Sunaba K, Shibuya H.  Sonographic appearances of malignant lymphoma of the salivary glands. J Clin Ultrasound JCU. 2001;29(9):491–8. 51. Mantsopoulos K, Koch M, Fauck V, Schinz K, Schapher M, Constantinidis J, et al. Primary parotid gland lymphoma: pitfalls in the use of ultrasound imaging by a great pretender. Int J Oral Maxillofac Surg. 2021;50(5):573–8. 52. McKnight CD, Glastonbury CM, Ibrahim M, Rivas-Rodriguez F, Srinivasan A.  Techniques and approaches for safe, high-yield CT-guided suprahyoid head and neck biopsies. Am J Roentgenol. 2017;208(1):76–83. 53. Wan Y-L, Chan S-C, Chen Y-L, Cheung Y-C, Lui K-W, Wong H-F, et al. Ultrasonography-­ guided core-needle biopsy of parotid gland masses. 2004;5:S90. 54. Sharma G, Jung AS, Maceri DR, Rice DH, Martin SE, Grant EG. US-guided fine-needle aspiration of major salivary gland masses and adjacent lymph nodes: accuracy and impact on clinical decision making. Radiology. 2011;259(2):471–8.

Chapter 10

Clinical Management Mandeep Bajwa, Piero Nicolai, Sidharth V. Puram, Joseph Zenga, and Mark Varvares

General Background The heterogeneity of salivary gland diseases presents unique challenges for the pathologist, radiologist, and treating clinician in their pursuit of optimal patient care. Clinical history, physical exam, and information provided by imaging studies such as ultrasound, contrast-enhanced computed tomography (CT) (Fig. 10.1), or magnetic resonance imaging (MRI) with contrast as well as fine needle aspiration (FNA) all contribute to the development of a management plan that can range from observation to limited or extensive surgical resection and possible adjuvant therapy [1]. FNA has a well-established role in salivary gland diagnostics and is

M. Bajwa Regional Maxillofacial Unit, Aintree University Hospital, Liverpool, UK e-mail: [email protected]; [email protected] P. Nicolai Department of Otorhinolaryngology–Head and Neck Surgery, University of Brescia, Brescia, Italy e-mail: [email protected] S. V. Puram Department of Otolaryngology-Head and Neck Surgery and Department of Genetics, Washington University School of Medicine, St. Louis, MO, USA e-mail: [email protected] J. Zenga Department of Otolaryngology, Medical College of Wisconsin, Milwaukee, WI, USA e-mail: [email protected] M. Varvares (*) Department of Otolaryngology, Massachusetts Eye and Ear, Harvard Medical School, Boston, MA, USA e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 W. C. Faquin et al. (eds.), The Milan System for Reporting Salivary Gland Cytopathology, https://doi.org/10.1007/978-3-031-26662-1_10

217

218

M. Bajwa et al.

Fig. 10.1  Axial CT with intravenous contrast of a superficial left parotid gland tumor. The mass measures 1.2 cm, has sharp margins, and shows slight enhancement. FNA of the mass showed a pleomorphic adenoma. (Contributed from Salivary Gland Cytopathology, 2008, Springer)

recommended by ASCO. Cytomorphology can provide valuable information regarding the nature of the salivary gland lesion. FNA is quick, and well tolerated with very few complications. It also lends itself to rapid onsite evaluation (ROSE) when used in conjunction with clinical assessment and imaging studies and can significantly improve triage of the patient for definitive therapy [2]. An understanding of the diagnostic challenges that cytopathologists face when assessing a salivary gland FNA can be extrapolated from the WHO classification of salivary gland neoplasms which has over 40 different entities based on histological features [3]. Because of the significant morphologic overlap of some entities, it is unavoidable that at times only a morphological description of the FNA will be provided to the treating clinician without a specific diagnosis [2]. This mandates that a clear line of communication exists between the cytopathologist and the treating clinician to ensure that the patient receives the correct management. It is in this context that a uniform reporting system for salivary gland cytology is most beneficial. The clinical utility of the Milan System for Reporting Salivary Gland Cytopathology (MSRSGC) to surgical specialists can be summarized as follows: • • • •

Standardizes reporting and clarity of communication. Correlates and stratifies the cytologic diagnosis with a risk of malignancy (ROM). Facilitates the use of a management algorithm. Is relevant, transferable, and practical for institutions with variable experience and expertise in salivary gland cytology. • Facilitates quality assurance review and clinical audits by setting standards (e.g., the proportion of inadequate samples less than 10%) as well as providing a potential outcome measure for further research.

10  Clinical Management

219

 linical Management Considerations: Overview of the Parotid C and Submandibular Glands There are several key questions that the clinician should address when developing a clinical management strategy for salivary gland lesions: • Do I need any additional information, clarification, or staging radiologic scans prior to formulating a definitive treatment plan? For masses involving the parotid gland, nearly all patients should have cross-­ sectional imaging performed preoperatively (CT or MRI with contrast). This is done to determine the extent of the lesion (superficial and/or deep lobe involvement) and the probability of complete resection of the primary tumor with facial nerve preservation in cases where this is possible. In a few patients with small (1 cm or less), well-defined lesions that are lateral in the parotid gland and with a benign cytologic diagnosis (i.e., Neoplastic: Benign), cross-sectional imaging may not be necessary. Patients with clinical scenarios that indicate the possibility of nerve involvement by tumor should undergo specific assessment for cranial nerve involvement (using MRI and/or CT). Patients with malignant disease should also have imaging that assesses the regional lymph node levels (CT or MRI with contrast), and the most likely sites of distant metastasis should be studied (CT of chest with contrast or skull base to mid-thigh PET/CT or PET/ MRI would both be acceptable). • Does this case need to be discussed in a multidisciplinary setting with early involvement of the medical or radiation oncologist for treatment planning? In both small and large institutions, the use of a multidisciplinary discussion should be considered for any salivary gland lesion that is not unequivocally benign. • Does the lesion need to be surgically removed or can it be safely monitored clinically? In certain scenarios, asymptomatic benign lesions with a low risk of malignant transformation, such as a Warthin tumor or a deep lobe pleomorphic adenoma in an elderly patient, may be managed by clinical observation. This can include selected cases when the patient wishes to avoid the possible risk of facial nerve injury. Particularly in the case of pleomorphic adenoma, these clinical decisions should be individualized based on shared patient and physician decision-making. The risk of malignant transformation increases with time, and evolution to a carcinoma ex pleomorphic adenoma is associated with significant morbidity and mortality [4]. • If I am considering monitoring the lesion, do I need any further investigations to be sure this is a safe option? Some lesions may require serial imaging or repeat FNA.  This will vary depending on the individual patient scenario. Tumors in  locations not easily assessed on the physical exam could be imaged serially until a “growth rate” is determined at which time the interval between studies may be lengthened.

220

M. Bajwa et al.

Tumors with ­indeterminate cytology (e.g., Neoplasm: SUMP) that appear to be benign based on their clinical presentation could undergo repeat FNA after a period of observation. Lastly, benign, or indeterminate tumors under observation that show a change in their clinical status such as rapid growth after a period of stability or the onset of new symptoms such as pain or facial nerve weakness should undergo repeat FNA to help further define the evolving tumor. • When surgical intervention is indicated, what is the minimal necessary procedure needed to adequately manage the tumor? The presurgical evaluation should address the possibility of postoperative facial nerve dysfunction and contour defect that may be required to completely remove the tumor and leave the patient with the smallest possible risk of recurrence. In the case of parotid neoplasms, the procedure may span the spectrum from extracapsular dissection for definitively benign lesions through superficial parotidectomy to subtotal or total parotidectomy. In all cases, the facial nerve is preserved unless it is impossible to separate it from the tumor without leaving gross disease behind. In most scenarios, the decision can be made to incompletely resect benign disease rather than sacrifice the main nerve or a major nerve trunk. In cases of malignancy, when considering nerve sacrifice a balance must be reached between the morbidity of resection and the possibility of eventual therapeutic failure and patient mortality if gross disease is left behind to be controlled with adjuvant radiation or chemoradiation. • Do I need to consent the patient for an increased risk of nerve injury or sacrifice and the donor site morbidity of a nerve graft? This topic is the centerpiece of the process of informed consent. For patients with large but clearly benign tumors, the low risk of permanent and significant nerve injury should be discussed. In any patient with the possibility of malignancy, the potential of nerve sacrifice, graft harvest, nerve defect reconstruction, and nerve transfer should be discussed with the patient. The possibility of eyelid procedures as well as static procedures to maintain midface tone should also be discussed. • Is a neck dissection indicated? Patients with clinical evidence of cervical lymph node involvement will require therapeutic neck dissection in nearly all cases. Patients without known neck disease may undergo elective neck dissection depending upon either the preoperative FNA evaluation, the findings of intraoperative frozen section, or both. The authors accept that the use of frozen section is highly variable internationally and needs to be interpreted by an expert pathologist; however, it can facilitate the management decision. The best time to perform a neck dissection is at the time of primary site surgery. Alternatively, the decision of how to manage the neck and the treatment modality (neck dissection versus radiation) may be made after a formal histological assessment of the primary lesion. Patients that do not have a diagnosis of malignancy prior to surgery, either due to an inaccessible site of lesion for FNA or an equivocal cytological diagnosis may have the decision made to proceed with neck dissection based upon the intraoperative frozen section diagnosis rendered on the primary parotid lesion. Patients with

10  Clinical Management

221

low-grade malignancies such as low-grade mucoepidermoid carcinoma can be followed clinically without neck dissection if the clinical and radiological evaluations both indicate that the neck is free of metastatic disease. Patients with higher grade pathology (e.g., salivary duct carcinoma or high-grade mucoepidermoid carcinoma) are candidates for elective neck dissection. Additionally, for patients with intermediate-grade lesions, elective neck dissection may play a role in pathologically staging patients and thereby guide the decision to proceed with postoperative adjuvant radiation versus active surveillance. • Will I require the use of intraoperative frozen section to address prior indeterminate cytology such as Atypia of Undetermined Significance (AUS), Neoplasm: (SUMP), SM, or Non-Diagnostic FNA? In some institutions, intraoperative frozen section is used as an important adjunct to the preoperative cytological diagnosis. This involves sending a partial parotidectomy specimen containing the entire tumor to an expert pathologist. It is important not to breach the capsule by performing an incisional biopsy as this risks tumor spillage and the associated increased risk of recurrence. When used, frozen section has a role in the assessment of the completeness of surgical resection margins and clearance of nerve margins in cases with nerve invasion. Frozen section can be helpful in clarifying what may have been an equivocal cytological diagnosis by defining the histologic classification, tumor grade, and extent of invasion. It can be used to assess the completeness of surgical resection and clearance of nerve margins in cases with nerve invasion such as adenoid cystic carcinoma. Clinicians are cautioned that frozen sections have their own sets of artifacts and limitations to consider. The impact of the results of the frozen section on decision-making on neck management was addressed in the prior section.

Management Options by Milan System Diagnostic Category Category I: Non-Diagnostic This category constitutes approximately 10% of salivary gland FNAs and has a reported mean Risk of Malignancy (ROM) of 15% [5]. The relatively high ROM in this category should stimulate clinicians to undertake further work-up of these lesions prior to formulating a definitive treatment plan. Recommended Management • Repeat FNA. If the first FNA was by palpation, then consider ultrasound guidance (USG). • If the second FNA is also Non-Diagnostic despite USG and adequate sample preparation, consider alternative investigations. First, perform cross-sectional imaging with contrast-enhanced CT or MRI if not already obtained. Second, if the MRI, CT, or clinical picture shows features concerning malignancy or if there is still doubt as to the nature of the lesion, consider US-guided Core Needle

222

M. Bajwa et al.

Biopsy (CNB), open biopsy (controversial due to the inherent risk of tumor spillage), or formal surgical excision. • If the sample is “cyst contents only,” completely aspirate the cyst contents under USG. If a solid component remains, it should be resampled. If the lesion disappears completely then repeat US  ±  FNA in 3–4  months. The FNA would be repeated in cases where US shows a recurrent lesion. Category II: Non-Neoplastic This category constitutes approximately 10% of salivary gland FNAs and has a reported mean ROM of 11% [5]. The majority of Non-Neoplastic lesions are managed nonsurgically. The management strategies outlined below attempt to identify the minority of lesions within this category that may be malignant thereby preventing undertreatment. Recommended Management • USG for the FNA is important for “Non-Neoplastic” cases to help avoid sampling errors which are not uncommon in this diagnostic category. If the FNA findings do not provide sufficient diagnostic information to explain clinical and radiologic findings, repeat FNA, the possible use of CNB, open biopsy, or surgical resection could be considered. • MRI or CT is useful to assess the lesion serially and to assess regional lymph nodes. • Lesions that are clearly “Non-Neoplastic” on FNA may be followed with either serial physical examinations, cross-sectional imaging, or a combination of both to assure stability. Any change in either the clinical exam or imaging could warrant repeat sampling to confirm no change in cytological status. Category III: Atypia of Undetermined Significance (AUS) This category constitutes approximately 7% of salivary gland FNAs and has a reported mean ROM of 30% [5]. The high ROM in this category needs to be considered when formulating a treatment plan for lesions in this category. Recommended Management • Repeat FNA. If the first FNA was performed by palpation, then consider US-­ guided FNA. • Cross-sectional imaging with contrast-enhanced MRI or CT. • CNB, open biopsy, or surgical resection should be considered for this lesion when the clinical presentation is concerning for neoplasia and/or malignancy. Suspicion for malignancy should be raised when there is a painful mass that lacks signs of inflammation, a concurrent facial nerve weakness, or paralysis, or a prior history of cutaneous malignancy.

10  Clinical Management

223

• If no evidence of neoplasia and/or malignancy is identified, consider regular clinical follow-up with duration interval to be determined based upon clinical suspicion; every 3–6 months is a general rule of thumb. Category IV: Neoplasm Category IVA: Benign This category constitutes approximately 42% of salivary gland FNAs and has a reported mean ROM of 98% [5]. General Considerations • Presurgical staging scans (ideally contrast-enhanced MRI of the neck and CT of the chest) is paramount for this category. • In the clinical management of clearly malignant salivary gland lesions, a definitive classification of a specific malignant histologic tumor type including grade

226

M. Bajwa et al.

(low versus high-grade) provides important information for clinical decision-­ making. When a definitive classification is not possible, information about tumor grade is still useful. Low versus intermediate versus high-grade classification may be useful to the clinician in determining the extent of surgery required at the primary site and the likelihood that a neck dissection would be needed. • If the grade of the lesion is not provided, elective neck dissection may be based on features such as tumor size >4 cm, extraglandular extension, and neurologic deficit [1, 7]. • For high-grade malignancies involving the deep lobe, a total parotidectomy is necessary. For lateral lesions, controversy exists regarding the extent of surgery with some surgeons electing to perform a total parotidectomy to optimize surgical clearance and others performing a superficial parotidectomy with the knowledge that the patient will be receiving postoperative radiotherapy. • A sub-category of “metastatic” would also be informative for the managing clinician. Parotid gland lymph nodes are a common site for metastases from head and neck cutaneous primaries, and these patients often require a concurrent neck dissection. If a lesion is metastatic from a non-cutaneous source, PET-CT may be indicated to locate a primary site of origin. Considerations for Malignant (V) Lesions of the Parotid Gland • For low-grade lesions with no clinical or radiographic evidence of pathological neck nodes and no other indicators for neck dissection (as mentioned above), a NPP is recommended aiming for complete excision. • For intermediate or high-grade lesions and no pathological neck nodes, a NPP and elective neck dissection are recommended. • For any category V lesion with evidence of pathological neck nodes, a NPP and therapeutic neck dissection is recommended. • Consent patient for the increased risk of nerve dysfunction and the possibility that nerve cannot be separated from tumor. The surgeon may choose to use intraoperative frozen section to confirm malignancy before sacrificing the facial nerve. • Consent patient that nerve may need to be sacrificed and reanimation procedures performed. Considerations for Malignant (V) Lesions of the Submandibular Gland • For low-grade lesions with no clinical or radiographic evidence of pathological neck nodes and no other indicators for neck dissection, a level IB clearance is recommended. • For intermediate or high-grade lesions without clinical or radiographic evidence of pathological neck nodes, an elective neck dissection including a minimum of levels I–III is recommended. • For intermediate or high-grade lesions with clinical or radiological evidence of pathological neck nodes, a therapeutic neck dissection including a minimum of levels I–III is recommended. Some surgeons may choose to perform a modified radical neck dissection; however, it is beyond the scope of this chapter to mandate the exact levels a surgeon should dissect.

10  Clinical Management

227

Consideration for Malignant (V) Lesions Suspected of Being Metastatic • Management should be based on the primary site of the tumor and involvement of the relevant multidisciplinary team. • For cutaneous squamous cell carcinoma, a NPP with elective neck dissection is recommended. • For cases of unknown primary, manage within the head and neck multidisciplinary team and consider PET-CT. If identified, management would be based on specific aspects of primary cancer. If no primary site is identified and the salivary gland lesion is isolated, it can be managed as a high-grade primary lesion in order to avoid issues related to uncontrolled head and neck malignancy. In such a setting, avoiding facial nerve injury is a priority. Key features related to (1) Indications for clinical observation, (2) Indications for neck dissections, (3) Degree of parotidectomy required, and (4) Management of the facial nerve are summarized in Tables 10.1, 10.2, 10.3, and 10.4. Table 10.1  Indications for clinical observation versus operative management 1. Unequivocal diagnosis of category IVA tumor with very low ROM (4 cm, high-grade features on frozen section of the primary site, extraglandular extension on imaging or noted intraoperatively, or preoperative facial weakness), should undergo elective neck dissection Table 10.3  Degree of parotidectomy required 1. Category IVA tumor: Nerve preserving tumor resection with a small cuff of normal parotid tissue may be less than complete lateral lobectomy or superficial parotidectomy. ECD may be appropriate for category IVA tumors 2. Category III and IVB lesions: Nerve preserving tumor resection with cuff of normal parotid tissue. Consider use of frozen sections to facilitate intraoperative decisionmaking. If the frozen section is consistent with a low-grade malignancy, consider immediate completion of superficial parotidectomy to encompass adjacent intraparotid lymph nodes. For high-grade malignancy consider neck dissection 3. Category V tumors: Superficial parotidectomy for low-grade lesions, total or subtotal parotidectomy for higher grade lesions, both with facial nerve preservation whenever possible

228

M. Bajwa et al.

Table 10.4  Management of the facial nerve 1. Never sacrifice a major nerve branch when removing benign disease unless the nerve branch is completely encased and even in that circumstance consider debulking 2. Do not sacrifice a functioning nerve without first establishing a diagnosis of malignancy (unequivocal cytology or frozen section) and determining that the nerve cannot be separated from tumor with only microscopic residual disease. Loss of function (stimulability) of an abnormal appearing nerve intraoperatively is consistent with a nerve that is invaded with cancer and should be resected after frozen section confirms malignancy 3. A nonfunctional nerve in the setting of proven malignancy should be resected and rehabilitated with the appropriate method based on available donor and recipient nerve for grafts and transfers and by static techniques

References 1. Armstrong JG, Harrison LB, Thaler HT, Friedlander-Klar H, Fass DE, Zelefsky MJ, et al. The indications for elective treatment of the neck in cancer of the major salivary glands. Cancer. 1992;69(3):615–9. 2. Bajwa MS, Rose SJ, Mairembam P, Nash R, Hotchen D, Godden D, et al. Feasibility of a novel classification for parotid gland cytology: a retrospective review of 512 cytology reports taken from 4 United Kingdom general hospitals. Head Neck. 2016;38:1596. 3. WHO Classification of Head and Neck Tumors. WHO/IARC classification of tumors, vol. 9. 4th ed. Lyon: World Health Organization/International Agency for Research on Cancer; 2017. 4. Tarakji B, Baroudi K, Hanouneh S, Kharma MY, Nassani MZ, Azzeghaiby SN.  Extensive review in the detection of the malignant transformation of pleomorphic adenoma. Gulf J Oncolog. 2013;1(13):67–82. PMID: 23339983. 5. Jalaly JB, Farahani SJ, Baloch ZW. The Milan system for reporting salivary gland cytopathology: a comprehensive review of the literature. Diagn Cytopathol. 2020;48(10):880–9. https:// doi.org/10.1002/dc.24536. Epub 2020 Jul 8. 6. McGurk M, Thomas BL, Renehan AG. Extracapsular dissection for clinically benign parotid lumps: reduced morbidity without oncological compromise. Br J Cancer. 2003;89(9):1610–3. 7. Frankenthaler RA, Byers RM, Luna MA, Callender DL, Wolf P, Goepfert H.  Predicting occult lymph node metastasis in parotid cancer. Arch Otolaryngol Head Neck Surg. 1993;119(5):517–20.

Chapter 11

Histologic Considerations and Salivary Gland Tumor Classification in Surgical Pathology Bruce M. Wenig, Vickie Y. Jo, Masato Nakaguro, and Alena Skálová

The classification of salivary gland neoplasms is dynamic and continues to evolve (Table 11.1) as reflected in the World Health Organization Classification of Head and Neck Tumours [1]. This classification includes tumor types that owing to recent evidence of unique molecular alterations have redefined their classification to include various subtypes [2]. Such tumors include polymorphous adenocarcinoma subtyped into classic and cribriform (also referred to as cribriform adenocarcinoma of salivary glands) subtypes [3], and intraductal carcinoma subtyped into intercalated duct, apocrine, oncocytic, and mixed subtypes [4]. Nomenclature changes made in the 2017 WHO Classification are retained in the 2022 WHO Classification including polymorphous adenocarcinoma for polymorphous low-grade adenocarcinoma (PLGA), and secretory carcinoma for mammary analog secretory carcinoma (MASC). An exception in the retained nomenclature from the fourth to fifth editions of the WHO Classification of Salivary Gland tumours is the recognition of sclerosing polycystic adenoma as a neoplasm based on the identification of monoclonality

B. M. Wenig (*) Department of Pathology, Moffitt Cancer Center, Tampa, FL, USA e-mail: [email protected] V. Y. Jo Department of Pathology, Brigham and Women’s Hospital, Harvard Medical School, Boston, MA, USA e-mail: [email protected] M. Nakaguro Department of Pathology and Laboratory Medicine, Nagoya University Graduate School of Medicine, Nagoya, Japan A. Skálová Department of Pathology, Faculty of Medicine in Pilsen, Charles University, Pilsen, Czech Republic e-mail: [email protected] © The Author(s), under exclusive license to Springer Nature Switzerland AG 2023 W. C. Faquin et al. (eds.), The Milan System for Reporting Salivary Gland Cytopathology, https://doi.org/10.1007/978-3-031-26662-1_11

229

230 Table 11.1 WHO classification of salivary gland neoplasms, fifth edition [1]

B. M. Wenig et al. Non-neoplastic epithelial lesions  Nodular oncocytic hyperplasia  Lymphoepithelial sialadenitis Benign epithelial tumors  Pleomorphic adenoma  Basal cell adenoma  Warthin tumor  Oncocytoma  Salivary gland myoepithelioma  Canalicular adenoma  Ductal papillomas  Sialadenoma papilliferum  Lymphadenoma  Sebaceous adenoma  Intercalated duct adenoma and hyperplasia  Striated duct adenoma  Sclerosing polycystic adenoma  Keratocystoma Malignant epithelial tumors  Mucoepidermoid carcinoma  Adenoid cystic carcinoma  Acinic cell carcinoma  Secretory carcinoma  Microsecretory carcinoma  Polymorphous carcinoma  Carcinoma ex pleomorphic adenoma  Hyalinizing clear cell carcinoma  Basal cell adenocarcinoma  Intraductal carcinoma  Salivary duct carcinoma  Myoepithelial carcinoma  Epithelial-myoepithelial carcinoma  Mucinous adenocarcinoma  Sclerosing microcystic adenocarcinoma  Carcinoma ex pleomorphic adenoma  Carcinosarcoma of salivary glands  Sebaceous adenocarcinoma  Lymphoepithelial carcinoma  Squamous cell carcinoma  Sialoblastoma Mesenchymal tumors specific to the salivary glands  Sialolipoma

by polymorphism of the human androgen receptor assay [2, 5, 6] rather than a nonneoplastic lesion termed sclerosing polycystic adenosis. Several new benign and malignant salivary gland tumors have been added to the fifth edition including

11  Histologic Considerations and Salivary Gland Tumor Classification in Surgical…

231

keratocystoma, intercalated duct adenoma, striated duct adenoma, microsecretory adenocarcinoma, and sclerosing microcystic adenocarcinoma [2]. Another new addition to malignant salivary gland tumors is mucinous adenocarcinoma which may be related to salivary intraductal papillary mucinous neoplasm [2] that includes the morphologic subtypes of papillary, colloid, signet ring, and mixed, all of which harbor recurrent AKT1 E17K mutations [2, 7]. Noteworthy too is the elimination in the fifth edition of the WHO Classification of previously defined tumor types including oncocytic carcinoma. Molecular studies have shown that many tumors dominated by or exclusively composed of oncocytes represent oncocytic variants of other salivary carcinomas [1] (e.g., mucoepidermoid carcinoma, salivary duct carcinoma). However, whether elimination of oncocytic carcinoma is or is not justified remains controversial and, at present, it is unclear if oncocytic carcinoma exists as an independent entity [1]. For this reason, the fifth edition of the WHO Classification included a chapter entitled “salivary carcinoma, not otherwise specified and emerging entities” for those tumor types not believed to represent independent entities and/or have yet to be more specifically defined on the basis of their morphology, immunohistochemical profile, and/or molecular signature thereby represent diagnoses of exclusion [8]. While the WHO classification separates neoplastic entities primarily on tumor morphology, immunohistochemical (IHC) markers [9] (Table 11.2) and a growing number of specific molecular alterations in salivary gland tumors support the morphologic-based classification [1, 10] (Table 11.3). Table 11.2  Selective immunohistochemistrya of selective salivary gland neoplasms [3] Tumor PA BCA/BCAdC MYO MEC ACC SC AdCC PAC SDC EMC CCC

PanK + + + + + + + + + + +

p40/p63 +/+ +/+ +/+ vb −/− −/− +/+ −/+ − +/+ +/+

S100 + + + − − + − + − + −

SOX10 (n) + + + − − + − + − + −

DOG1 − − − − +c − − − − − −

NR4A3 − − − − + (n) − − − − − −

MGB − − − − − +d − − − − −

AR − − − − − − − − + (n) − −

CD117 v v v v − − + (lum) v − − −

PA Pleomorphic adenoma, BCA Basal cell adenoma, BCAdC Basal cell adenocarcinoma, MYO Myoepithelioma, MEC Mucoepidermoid carcinoma, ACC Acinic cell carcinoma, SC Secretory carcinoma, AdCC Adenoid Cystic Carcinoma, PAC Polymorphous Adenocarcinoma, SDC Salivary duct carcinoma, EMC Epithelial-myoepithelial carcinoma, CCC Clear cell carcinoma including hyalinizing type, PanK pancytokeratin (e.g., AE1/AE3; CAM5.2), DOG1 Discovered on GIST1, MGB Mammaglobin, AR Androgen Receptor, n nuclear, lum strong staining luminal cells, v variably positive a Staining characteristics vary widely among tumor types and even within the same tumor type. This table details ideal staining characteristics per tumor type and while these staining patterns generally remain consistent, any given tumor listed may defy “convention” and show reactivity for a marker usually not associated with that tumor or may lack a marker associated with that tumor Specific staining characteristics: b MEC: show a basal/myoepithelial patterns of staining for p40 and p63; c DOG1: should be an admixture of strong apical membranous, cytoplasmic, and complete membranous staining; d Mammaglobin: should be strong and diffuse cytoplasmic staining

232

B. M. Wenig et al.

Table 11.3  Salivary gland neoplasms: chromosomal alterations [1, 4] Tumor Pleomorphic adenoma Basal cell adenoma

Myoepithelioma, oncocytic Sialadenoma papilliferum Sclerosing polycystic adenoma

Mucoepidermoid carcinoma

Adenoid cystic carcinoma

Secretory carcinoma

Acinic cell carcinoma

PAC, classic PAC, cribriform

Salivary duct carcinoma

Carcinoma ex pleomorphic adenoma Hyalinizing clear cell carcinoma Myoepithelial carcinoma

Chromosomal alteration 8q12 12q13–15 3p22.1 16q12.1 Q6p13.3 5q22.2 8q12 7q34 11p15.5 3q26.3 14q32.33 10q23.31 t(11;19)(q21;p13) t(11;15)(q21;q26) 9p21.3 6q22–23 8q13 9q23.3 t(12;15)(p13q25) t(12;10) (p13;q11) t(12;7) (p13;q31) t(12;4) (p13;q31) t(10;10)(p13;q11) t(5q14.3) 9q31 19q31.1 14q12 14q12 19q13.2 2p22.2 17q21.1 8p11.23 17p13.1 3q26.32 11p15.5 Xq12 10q23.31 9p21.3 8q12 12q13–15 17p13.1 t(12;22)(q21;q12) 8q2

Fusions, amplification, mutations PLAG1 HMGA2 CTNNB1 CYLD AXIN1 APC PLAG1 BRAF v600E HRAS PIK3CA AKT1 PTEN CRTC1::MAML2 CRTC3::MAML2 CDKN2A deletion MYB MYBL1 NOTCH mutations ETV6::NTKR3 ETV6::RET ETV6::MET ETV6::MAML3 VIM::RET

Percentage of cases >50% 10–20% 37–80% 36% 9% 3% 40% 50–100% 10–30%

NR4A3 MSANTD3 fusion/ amplification PRKD1 mutations PRKD1 fusions PRKD2 fusions PRKD3 fusions Her2 FGFR1 TP53 PIK3CA HRAS AR copy gain PTEN loss CDKN2A loss PLAG1 HMGA2 TP53 EWSR::ATF1 EWSR::CREM PLAG

86% 4%

40–90% 6% or less 25% 80–90% Approx 10% 14% >90% 2–5%