Methods in molecular biology vol 1606 molecular profiling methods and protocols 2nd edition

Chapters 1–3 introduce the reader to clinical medicine through a primer on tumor staging and grading, ethics in medicine, and clinical trial design.. 1606, DOI 10.1007/978-1-4939-6990-6_

Molecular Profi ling

Virginia Espina Editor

Methods and Protocols

Second Edition

Methods in

Molecular Biology 1606

Series Editor

John M Walker School of Life and Medical Sciences University of Hertfordshire Hatfield, Hertfordshire, AL10 9AB, UK

For further volumes:

http://www.springer.com/series/7651

ISSN 1064-3745 ISSN 1940-6029 (electronic)

Methods in Molecular Biology

ISBN 978-1-4939-6989-0 ISBN 978-1-4939-6990-6 (eBook)

DOI 10.1007/978-1-4939-6990-6

Library of Congress Control Number: 2017937315

© Springer Science+Business Media LLC 2017

This work is subject to copyright All rights are reserved by the Publisher, whether the whole or part of the material is concerned, specifically the rights of translation, reprinting, reuse of illustrations, recitation, broadcasting, reproduction

on microfilms or in any other physical way, and transmission or information storage and retrieval, electronic adaptation, computer software, or by similar or dissimilar methodology now known or hereafter developed.

The use of general descriptive names, registered names, trademarks, service marks, etc in this publication does not imply, even in the absence of a specific statement, that such names are exempt from the relevant protective laws and regulations and therefore free for general use.

The publisher, the authors and the editors are safe to assume that the advice and information in this book are believed to

be true and accurate at the date of publication Neither the publisher nor the authors or the editors give a warranty, express 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 Printed on acid-free paper

This Humana Press imprint is published by Springer Nature

The registered company is Springer Science+Business Media LLC

The registered company address is: 233 Spring Street, New York, NY 10013, U.S.A.

Virginia Espina

Center for Applied Proteomics and Molecular Medicine

George Mason University

Manassas, VA, USA

Dedication

This book is proudly dedicated to Mary Anne and Len Schiff, for their initial faith in my future scientific career, and to my husband Tito; my children Ben, Paul, and Emily; and my favorite future scientist Olivia, for always listening to my ideas

Virginia Espina

The first edition of Molecular Profiling (published in 2012) was at the forefront of the

per-sonalized medicine movement The first edition included reviews of genomics and genomic profiling, technologies which in the intervening years have rapidly evolved into routine clinical assays for detecting mutations Mass spectrometry for protein profiling has also evolved into sensitive and specific multiple reaction monitoring assays, enabling quantifica-tion of proteins without antibody-based methods, while tumor staging and grading and tissue preservation continue to be important aspects of molecular profiling As you can see from these examples, staying current in molecular profiling requires lifelong learning and incorporating “routine” laboratory analyses with cutting-edge technologies Hence, when

Professor John Walker, editor for the Methods in Molecular Biology series, inquired as to my

interest in editing an updated version of this book, I was honored that the readers found the first edition informative and that there was sufficient, continuing interest in molecular profiling for an updated version However, I also knew that the second edition would require many updates to the protocol chapters to reflect the current state of the art in molecular profiling

The purpose of this revised volume is to provide both an update on technology and an accelerated tutorial to assist students, entrepreneurs, new investigators, and established investigators who want to quickly become versed in, and immersed in, the entire process from discovery to clinical trial validation and commercial public benefit The aims of the first and second edition are the same: to span the full spectrum of molecular profiling from tumor staging and grading through biomarker discovery to commercialization The practical guides are not limited to experimental methods I have included tutorials on tumor staging, ethics, patents and intellectual property, product development, innovative clinical trial designs, and grant writing tips for investigators seeking funding in translational research

Molecular Profiling, second edition, includes 17 new chapters and 9 revised/updated

chapters The new chapters cover some “tried-and-true” laboratory methods such as PCR and scanning electron microscopy The second edition also includes updated versions of antibody validation and Western blotting I had two reasons for including these “standard,” often kit-based, laboratory methods: (1) understanding the science behind the “kits” can help solve many problems encountered in assay development, and (2) the book is intended for a wide audience, including students and physician-scientists The 17 new chapters cover laboratory methods relevant to human disease: microsatellite analysis, somatic mutation analysis, proteomic bioinformatics, microscopic imaging, preservation of bone tissue for molecular profiling, glycomics, metabolomics, immunohistochemistry, FISH, ELISA development, and multiple reaction monitoring mass spectrometry

Chapters 1–3 introduce the reader to clinical medicine through a primer on tumor staging and grading, ethics in medicine, and clinical trial design These chapters have been updated to address the current relevant information and issues For example, the chapter

on clinical trials discusses examples of innovative trial design in which data generated during the clinical trial can be used to modify therapies administered to the patients as the trial is accruing patients

Preface

A set of core chapters (4–23) covering genomics, proteomics, imaging, and matics illustrate current laboratory protocols for generating data relevant to molecular medicine Each of these disciplines is complementary, and the grouping simply provides a means for differentiating the classes of molecular analytes An emphasis is placed on tissue- based molecular profiling, which is the core of personalized medicine Although many of the techniques discussed in this volume use commercially available reagents and instrumen-tation, it is imperative for the user/reader to understand the principles and nuances of these techniques, because they are designed for use with irreplaceable human tissue specimens.The three topics covered in Chapters 24–26 are a unique aspect of this volume of the

bioinfor-Methods in Molecular Biology series These latter chapters discuss, in a narrative or tutorial style,

real-world needs in personalized molecular medicine The narrative chapters are designed to provide the reader with a well-rounded discussion of intellectual property issues in biotechnol-ogy, human subjects research requirements, tips for grant writing in translational research, and

an overview of technology transfer (patent) issues As with the protocol chapters, important points are highlighted in the Notes section for each of the narrative chapters

I hope that the readers of this second edition of Molecular Profiling will use it as a

prac-tical guide at the lab bench as well as in the classroom The intended readership spans the range of scientists, pathologists, oncologists, residents, biotechnologists, medical students, and nurses involved in clinical trial research

I would like to express my sincere gratitude to my editorial assistant, Emily Espina, who provided excellent grammar editing I truly appreciate, and thank, all my authors for their time and effort in compiling and submitting new and updated chapters Their collective contributions and input have greatly expanded the scope and depth of the book I thank Lance Liotta, my co-editor on the first edition, who supported me with the utmost respect and trust, while I pursued this solo editing endeavor

I anticipate that this revised volume will attract new investigators, and invigorate rienced researchers, who can apply their creative talents to realize the promise of individual-ized molecular medicine I hope you find this revised edition a useful and informative guide for your molecular profiling adventures

Contents

Preface vii Contributors xi

1 Tumor Staging and Grading: A Primer 1

Stacy M Telloni

2 Innovations in Clinical Trial Design in the Era of Molecular Profiling 19

Julia D Wulfkuhle, Alexander Spira, Kirsten H Edmiston,

and Emanuel F Petricoin III

3 Personalized Medicine: Ethical Aspects 37

G Terry Sharrer

4 Antibody Validation by Western Blotting 51

Michele Signore, Valeria Manganelli, and Alex Hodge

5 Scanning Electron Microscopy Sample Preparation and Imaging 71

Jenny Ngoc Tran Nguyen and Amanda M Harbison

6 One-Step Preservation and Decalcification of Bony Tissue

for Molecular Profiling 85

Claudius Mueller, Michael G Harpole, and Virginia Espina

7 Application of Hydrogel Nanoparticles for the Capture, Concentration,

and Preservation of Low-Abundance Biomarkers 103

Ruben Magni and Alessandra Luchini

8 Using Laser Capture Microdissection to Isolate Cortical Laminae

in Nonhuman Primate Brain 115

Brian A Corgiat and Claudius Mueller

9 Western Blot Techniques 133

Brianna Kim

10 ELISA for Monitoring Nerve Growth Factor 141

Justin B Davis

11 Reverse Phase Protein Microarrays 149

Elisa Baldelli, Valerie Calvert, Alex Hodge, Amy VanMeter,

Emanuel F Petricoin III, and Mariaelena Pierobon

12 Clustering and Network Analysis of Reverse Phase Protein Array Data 171

Adam Byron

13 PCR: Identification of Genetic Polymorphisms 193

Amanda M Harbison and Jenny Ngoc Tran Nguyen

14 Microsatellite Analysis for Identification of Individuals Using Bone

from the Extinct Steller’s Sea Cow (Hydrodamalis gigas) 205

Jeffery F Warner, Michael G Harpole, and Lorelei D Crerar

15 Somatic DNA Mutation Analysis 219

Anthony O’Grady and Robert Cummins

16 Optimization of Immunostaining for Prospective Image Analysis 235

Scott M Lawrence and Yelena G Golubeva

17 Fluorescence In Situ Hybridization of Cells, Chromosomes,

and Formalin-Fixed Paraffin-Embedded Tissues 265

Ahmad Alamri, Jun Yeb Nam, and Jan K Blancato

18 High-Resolution Image Stitching as a Tool to Assess Tissue-Level

Protein Distribution and Localization 281

Bryan A Millis and Matthew J Tyska

19 Mass Spectrometry-Based Biomarker Discovery 297

Weidong Zhou, Emanuel F Petricoin III, and Caterina Longo

20 Quantitative Mass Spectrometry by Isotope Dilution and Multiple

Reaction Monitoring (MRM) 313

Paul Russo, Brian L Hood, Nicholas W Bateman, and Thomas P Conrads

21 LC-Mass Spectrometry for Metabolomics 333

Allyson L Dailey

22 Metabolomic Bioinformatic Analysis 341

Allyson L Dailey

23 Stable Isotope Quantitative N-Glycan Analysis by Liquid Separation

Techniques and Mass Spectrometry 353

Stefan Mittermayr, Simone Albrecht, Csaba Váradi,

Silvia Millán-Martín, and Jonathan Bones

24 Grant Writing Tips for Translational Research 367

Lindsay Wescott, Michael Laskofski, Donna Senator, and Carly Curran

25 Inventions and Patents: A Practical Tutorial 379

Hina Mehta, Lille Tidwell, and Lance A Liotta

26 Product Development and Commercialization of Diagnostic or Life Science

Products for Scientists and Researchers 399

Meghan M Alonso

Index 409

AhmAd AlAmri • Lombardi Comprehensive Cancer Center, Georgetown University

Medical Center, Washington, DC, USA; Department of Clinical Laboratories Sciences, College of Applied Medical Sciences, King Khalid University, Abha, Saudi Arabia

Simone Albrecht • NIBRT—The National Institute for Bioprocessing Research & Training, Dublin, Ireland

meghAn m AlonSo • IMUA Services, Medical Invention and Device Development Consulting, Carlsbad, CA, USA

eliSA bAldelli • Center for Applied Proteomics and Molecular Medicine, George Mason University, Manassas, VA, USA

nicholAS W bAtemAn • DOD Gynecologic Cancer Center of Excellence, Annandale, VA, USA

Medical Center, Washington, DC, USA; Georgetown Lombardi Comprehensive Cancer Center, Fisher Center for Hereditary Cancer and Clinical Genomics Research,

Georgetown University, Washington, DC, USA

JonAthAn boneS • NIBRT—The National Institute for Bioprocessing Research &

Training, Dublin, Ireland

and Molecular Medicine, University of Edinburgh, Edinburgh, UK

VAlerie cAlVert • Center for Applied Proteomics and Molecular Medicine, George Mason University, Manassas, VA, USA

thomAS P conrAdS • Inova Dwight and Martha Schar Cancer Institute, Falls Church,

VA, USA; Gynecologic Cancer Center of Excellence, Women’s Health Integrated Research Center, Annandale, VA, USA

Mason University, Manassas, VA, USA

lorelei d crerAr • Department of Biology, George Mason University, Fairfax, VA, USA

robert cumminS • Department of Pathology, RCSI Education & Research Center, Royal College of Surgeons in Ireland, Beaumont Hospital, Dublin, Ireland

cArly currAn • Office of Sponsored Programs, George Mason University, Fairfax, VA, USA

AllySon l dAiley • Department of Chemistry and Biochemistry, George Mason

University, Manassas, VA, USA

JuStin b dAViS • Department of Chemistry and Biochemistry, George Mason University, Manassas, VA, USA

KirSten h edmiSton • Departement of Surgery, Inova Fairfax Hospital Cancer Center, Falls Church, VA, USA

VirginiA eSPinA • Center for Applied Proteomics and Molecular Medicine, George Mason University, Manassas, VA, USA

yelenA golubeVA • Medimmune, Gaithersburg, MD, USA

AmAndA m hArbiSon • Northern Virginia Community College, Manassas, VA, USA

Contributors

michAel g hArPole • Center for Applied Proteomics and Molecular Medicine, George Mason University, Manassas, VA, USA

University, Manassas, VA, USA

briAnnA Kim • Ceres Nanosciences, Manassas, VA, USA

michAel lASKofSKi • Office of Sponsored Programs, George Mason University, Fairfax,

VA, USA

Biomedical Research, Inc , Frederick, MD, USA

Univesity, Manassas, VA, USA

cAterinA longo • Dermatology and Skin Cancer Unit, Arcispedale S Maria Nuova IRCCS, Reggio Emilia, Italy

AleSSAndrA luchini • Center for Applied Proteomics and Molecular Medicine, George Mason University, Manassas, VA, USA

University, Manassas, VA, USA

VAleriA mAngAnelli • Department of Experimental Medicine, Sapienza University of Rome, Rome, Italy

SilViA millán-mArtín • NIBRT—The National Institute for Bioprocessing Research & Training, Dublin, Ireland

Jenny ngoc trAn nguyen • Northern Virginia Community College, Manassas, VA, USA

Anthony o’grAdy • Department of Pathology, RCSI Education & Research Centre, Royal College of Surgeons in Ireland, Beaumont Hospital, Dublin, Ireland

emAnuel f Petricoin iii • Center for Applied Proteomics and Molecular Medicine, George Mason University, Manassas, VA, USA

mAriAelenA Pierobon • Center for Applied Proteomics and Molecular Medicine, George Mason University, Manassas, VA, USA

Univesity, Manassas, VA, USA

donnA SenAtor • Office of Sponsored Programs, George Mason University, Fairfax, VA, USA

g terry ShArrer • Retired Curator of Health Sciences Smithsonian Institution,

Washington, DC, USA

michele Signore • Istituto Superiore di Sanità, Rome, Italy

AlexAnder SPirA • Virginia Cancer Specialists, Falls Church, VA, USA

lille tidWell • Tidwell Medical Technologies, LLC, Durham, NC, USA

mAttheW J tySKA • Department of Cell and Developmental Biology, Vanderbilt

University, Nashville, TN, USA

University, Manassas, VA, USA

cSAbA VárAdi • NIBRT—The National Institute for Bioprocessing Research & Training, Dublin, Ireland

Jeffery f WArner • Department of Biology, George Mason University, Fairfax, VA, USA

lindSAy WeScott • Office of Sponsored Programs, George Mason University, Fairfax,

VA, USA

George Mason University, Manassas, VA, USA

Weidong Zhou • Center for Applied Proteomics and Molecular Medicine, George Mason Univesity, Manassas, VA, USA

Virginia Espina (ed.), Molecular Profiling: Methods and Protocols, Methods in Molecular Biology, vol 1606,

DOI 10.1007/978-1-4939-6990-6_1, © Springer Science+Business Media LLC 2017

Key words Breast cancer, Grade, Lung cancer, Lymph node, Metastasis, Prostate cancer, Ovarian

cancer, Stage, Tumor

1 Introduction

Tumor staging and grading are critical for the practice of clinical oncology because these classifications serve as the starting point for patient care During the staging/grading process, malignancies are categorized according to anatomic location and pathologic character-istics The most recent seventh edition of the TMN staging system was adopted from the American Joint Committee on Cancer in 2010.Cancer stage refers to the anatomic extent of the disease spread Stage I through III diseases are considered curable using surgery, radiation, chemotherapy, and hormonal/biologic therapies Stage

IV disease is considered incurable The internationally accepted criterion for establishing stage is the tumor-node-metastasis (TNM) system, which includes (a) tumor size and local growth (T), (b) extent of lymph node metastases (N), and (c) occurrence

of distant metastases (M) Cancers are categorized as primary tumor size between T0 and T4, nodes between N0 and N3, and

metastases between M0 and M1 Generally, as the size of the mary untreated cancer (T) increases, regional lymph node involve-ment (N) and distant metastasis (M) become more frequent Common sites for solid tumor metastases include lymph nodes, the lung, bone, liver, brain, and bone marrow [1] The following is a very basic TNM schema [1 2]:

pri-Primary Tumor (T)

TX: Tumor cannot be assessedT0: No evidence of primary tumorTis: Carcinoma in situ

T1, T2, T3, T4: Increasing size and/or local extent of tumor

Regional Lymph Nodes (N)

NX: Regional lymph nodes cannot be assessedN0: No evidence of disease in lymph nodes N1, N2, N3: Increasing disease involvement of regional lymph nodes

Distant Metastasis (M)

MX: Distant metastasis cannot be assessedM0: No distant metastasis

M1: Distant metastasisSolid tumor malignancies are staged only once, at the time of initial diagnosis Cancers can be assigned both a clinical and patho-logic stage Clinical stage is established before initiation of therapy and is based on physical examination, laboratory findings, and imaging studies Pathologic stage is determined using tumor tissue procured through surgical exploration of disease [2] Pathologic stage is particularly significant for cancers which are not easily clas-sified in a clinical setting, such as ovarian carcinoma [3] Both clini-cal and pathologic stages should be documented in every patient’s permanent medical record

Stage I: Tumor limited to organ of origin, without nodular or cular spread

vas-Stage II: Local spread of tumor into surrounding tissue and regional lymph nodes The lesion is resectable, but there is sometimes uncertainty about completeness of removal due to tumor microinvasion into surrounding tissue

2.1 Tumor Stages

Stage III: Extensive primary tumor with invasion into deeper structures and lymph nodes The lesion is operable, but often-times gross disease is left behind.

Stage IV: Evidence of distant metastasis beyond tumor organ of origin

Tumor grade must be assigned by certified pathologists and is based on histology and architecture By definition, malignant tumors invade the basement membrane and extracellular matrix to invade surrounding tissue with indistinct borders [4] Additional microscopic evidence of abnormal, or malignant, behavior includes giant tumor cells, high numbers of mitoses, nucleoli and chroma-tin morphology, unusual mitoses, aneuploidy, and nuclear pleo-morphism [1 4]

In general, low-grade cancers are well differentiated, resembling healthy cellular counterparts, and high-grade cancers are anaplastic and disorderly The most poorly differentiated part of the tumor determines overall tumor grade with the exception of prostate can-cers [3] In general, high-grade cancers are more clinically aggressive than low-grade cancers Most grading systems divide tumors into three or four grades according to cellular differentiation [2]:

GX: Grade cannot be evaluatedG1: Well differentiated

G2: Moderately differentiatedG3–G4: Poorly differentiatedUsing cancer grading and staging in addition to other clinical data, clinicians can construct nomograms to predict treatment out-comes, cure rates, and disease-free survival times Following is a discussion of specific cancer staging and grading for lung, prostate, breast, and ovarian cancers Clinical staging information is from the seventh (2010) edition of the American Joint Committee on Cancer’s (AJCC) Staging Manual [2]

3 Cancer Classification Examples

Lung cancer is one of the most common malignancies in the Western hemisphere and the leading cause of cancer death in men and women [2 5] Stage of lung carcinomas at diagnosis remains, in general, the most important prognostic factor for patients [2 6] Patients with clinically suspected lung carcinoma should receive detailed history and physical exam, complete blood count, chemistry profile, staging positron emission tomography (PET)/CT scan, and magnetic reso-nance imaging (MRI) of the brain for stage II disease or higher PET/CT scans are used to show pattern of disease spread and also

2.2 Tumor Grades

3.1 Lung Cancer

Clinical

Staging Workup

demonstrate tumor metabolic activity by uptake of fludeoxyglucose Suspicious lesions at distant sites may be biopsied [5] Lung cancer spreads locally into other mediastinal structures and also to intratho-racic, scalene, and supraclavicular lymph nodes Distant sites of lung cancer metastases include the liver, adrenal glands, contralateral lung, and brain [2].

Primary tumor tissue must be procured for confirmation of pathology and definition of histology Tissue may be collected either through bronchoscopy for central lesions or CT-guided needle biopsy for peripheral lesions Thoracentesis should also be per-formed in patients with pleural effusions to determine whether the effusion cells are malignant or paramalignant and exudative with negative cytology Scalene and intrathoracic lymph nodes that appear irregular or enlarged on CT scan could also be sampled using medi-astinoscopy This regional lymph node sampling is critical for con-struction of the best sequence of treatment for patients, which includes surgery, radiation, chemotherapy, and sometimes targeted therapies [5 7 8] (see Table 1 for specific TMN staging criteria).Small-cell lung carcinoma (SCLC), a common subtype of lung cancer, is frequently described using a two-stage system rather than TNM staging [5 6 9] SCLC tends to be dissemi-nated at the time of diagnosis, with only 25% of patients present-ing with “limited” disease [5] SCLC is considered “limited” when it is confined to an area which can be safely treated with definitive radiation doses In general, “limited” SCLC corre-sponds to stages I through III in the TNM system, and “exten-sive” SCLC corresponds to stage IV disease [2] All patients diagnosed with SCLC must have brain and bone imaging because disease most commonly metastasizes to these sites [5]

Lung cancers are classified using light microscopy with routinely stained (hematoxylin and eosin) preparations Immunostains (IHC) are used to distinguish NSCLC subtypes, most generally adenocarcinoma and squamous cell carcinoma Adenocarcinomas stain positive on IHC for thyroid transcription factor-1 (TTF-1) and squamous cell carcinomas stain negative for TTF-1, positive for p63 New biomarkers, or genetic mutations, have arisen as an important classification for advanced NSCLC as well These mark-ers, which are both prognostic and predictive, include the anaplastic lymphoma kinase (ALK) fusion oncogene and the epidermal growth factor receptor (EGFR) mutation Both are targetable with currently available tyrosine kinase inhibitor therapy

Adenocarcinomas are glandular tumors which produce mucin and are usually located at the periphery of the lung These tumors are graded according to number and appearance of glandular structures (Fig 1a) Well-differentiated tumors consist of distinctive gland struc-tures throughout 90% of the tumor mass The glands resemble a healthy lung tissue, with tall columnar or mucinous epithelium,

3.1.1 Histology

and Grading

Table 1

TNM classification and stage grouping for NSCLC

N1: Metastasis to ipsilateral peribronchial and/or ipsilateral hilar lymph nodes and intrapulmonary nodes including involvement

by direct extension of the primary tumor

Stage IIB T2b, N1, M0

T3, N0, M0 T3: Tumor >7 cm or invades any one of the following: chest wall, diaphragm, phrenic nerve, mediastinal pleura, parietal

pericardium, or tumor in the main bronchus <2 cm distal to the carina but without involvement of the carina, or associated atelectasis or obstructive pneumonitis of the entire lung or separate tumor nodule(s) in the same lobe

Stage IIIA T1a, N2, M0

T1b, N2, M0 T2a, N2, M0 T2b, N2, M0 T3, N1, M0 T3, N2, M0 T4, N0, M0 T4, N1, M0

N2: Metastasis to ipsilateral mediastinal and/or subcarinal lymph node(s)

Stage IIIB T1a, N3, M0

T1b, N3, M0 T2a, N3, M0 T2b, N3, M0 T3, N3, M0 T4, N2, M0 T4, N3, M0

N3: Metastasis to contralateral mediastinal, contralateral hilar, ipsilateral or contralateral scalene, or supraclavicular lymph node(s)

T4: Tumor of any size that invades any of the following:

mediastinum, heart, great vessels, trachea, esophagus, vertebral body, carina or separate tumor nodules in different ipsilateral lobes

Stage IV Any T, any N,

M1a or M1b M1a: Separate tumor nodules in a contralateral lobe; tumor with pleural nodules or malignant pleural (or pericardial) effusion

M1b: Distant metastases Used with permission of the American Joint Committee on Cancer (AJCC), Chicago, Illinois The original and primary

source for this information is the AJCC Cancer Staging Manual, seventh edition (2010) published by Springer

Science + Business Media

eosinophilic cytoplasm, basal nuclei, and prominent nucleoli A key variant of well-differentiated adenocarcinoma is the invasive mucinous type (formerly bronchioalveolar carcinoma), characterized by bland-appearing tumor cells growing continuously along alveolar walls [9] Poorly differentiated adenocarcinomas have glandular or acinar struc-tures throughout at least 50% of the tumor mass and usually contain solid stromal areas with atypical mucinous cells [6 9]

Squamous cell cancers commonly arise from epithelial cells in the proximal tracheobronchial tree and may therefore present with

signs of airway obstruction [5 6] These carcinomas are characterized microscopically by the presence of intercellular des-mosomes, or “intercellular bridges,” and also keratin production (Fig 1b) [6 9] Grade 1 or well-differentiated tumors have sheets

of cells with ample eosinophilic cytoplasm, round nuclei, nent nucleoli, and well-defined cellular borders with intercellular bridges These well-differentiated tumors may contain concentric laminated deposits of amorphous, keratinous material called “squa-mous pearls” [6 9] Areas of comedo-like necrosis characterize grade 2 tumors Grade 3 tumors are poorly differentiated and cells tend to grow in confluent sheets Cells are characterized by bizarre nuclei, cytological atypia, increased mitotic figures, and areas of necrosis and/or hemorrhage [6]

promi-Neuroendocrine carcinomas stain positive for synaptophysin and chromogranin A and are further subdivided according to cel-lular differentiation Neuroendocrine tumor types include carci-noid tumors and SCLC, which is the most poorly differentiated variant [6 9] Microscopically, SCLC cells are primitive appearing with scant cytoplasm, granular chromatin, and high mitotic activ-ity Encrustation, with basophilic deposition of DNA within blood vessel walls, is a distinctive histological feature of SCLC (Fig 2).Initial diagnosis of prostate cancer is usually based on abnormal digital rectal exam (DRE) or an elevated prostate-specific antigen (PSA) blood level Prostate biopsy, often obtained transrectally with ultrasound guidance, is necessary for a definitive diagnosis of prostate adenocarcinoma Approximately ten-core needle biopsies are procured to evaluate all potentially affected lobes of the pros-tate [10, 11]

Staging prostate cancer is based on primary tumor size (clinical

T stage), serum PSA level, Gleason score, and extent of disease

3.2 Prostate Cancer

Clinical

Staging Workup

Fig 1 NSCLC histology (a) Adenocarcinomas are usually located at the periphery of the lung and graded according

to the number and appearance of glandular structures (b) Area of a normal lung epithelium adjacent to squamous

cell carcinoma in situ Squamous cell carcinomas are graded based on proportion of intercellular bridges and other characteristics of keratinization Courtesy of William Funkhouser, MD (with permission of Springer)

spread The risk of local prostate cancer progressing over the short term is low [10, 11]; therefore, many patients with low-risk lesions choose watchful waiting over aggressive staging and treatment When more advanced disease is suspected, staging studies should include transrectal ultrasound and/or pelvic magnetic resonance imaging (MRI) For evaluation of distant metastases, PET scans and radionuclide bone scan may be used Prostate adenocarcinoma most commonly metastasizes to distant lymph nodes and the bone, but lung and liver metastases are common in late-stage disease [2]

(see Table 2 for specific TMN staging criteria)

The vast majority of prostate cancers are epithelial adenocarcinomas, although variants include neuroendocrine tumors, stromal tumors, and mesenchymal tumors such as leiomyosarcoma or sarcomatoid carcinoma [12, 13] Prostatic intraepithelial neoplasia [14] is con-sidered a premalignant lesion [13] Low-grade PIN is characterized

by a slight increase in cellularity with irregular spacing of epithelial cells High-grade PIN displays a marked increase in cellularity with nuclear enlargement and hyperchromasia Both low-and high-grade PIN demonstrate preservation of the basal cell layer [11, 13].Prostate adenocarcinomas are graded using the Gleason system, which classifies specimens between one and five based on glandular architecture and cellular cytomorphology [10–12] Higher grades of prostate cancer have the most aberrant glandular morphology Several clinical trials have validated the prognostic value of the Gleason system, with higher scores predicting widespread disease and worse prognosis [11, 13] Pathologists report both primary and secondary scores, with the primary score representing the most

3.2.1 Histology and

Grading

Fig 2 Small-cell lung carcinoma (SCLC) This histology demonstrates poor

dif-ferentiation SCLC cells are primitive appearing with scant cytoplasm, granular chromatin, and high mitotic activity Courtesy of William Funkhouser, MD (with permission of Springer)

Table 2

TNM classification and stage grouping for prostate adenocarcinoma

I T1a-c or T2a, N0, M0, PSA <10,

Gleason ≤6

T1–2a, N0, M0, PSA X, Gleason X

T1a: Tumor incidental histologic finding in 5% or less of prostate tissue resected

T1b: Tumor incidental histologic finding in more than 5% of tissue resected

T1c: Tumor identified by needle biopsy T2a: Tumor involves one half of one lobe or less N0: No regional lymph node metastasis M0: No distant metastasis

IIA T1a-c or T2a, N0, M0, PSA <20,

III T3, N0, M0, any PSA, any Gleason T3: Tumor extends through the prostate capsule

IV T4, N0, M0, any PSA, any Gleason

Any T, N1, M0, any PSA, any Gleason

Any T, any N, M1, any PSA, any

Gleason

T4: Tumor is fixed or invades adjacent structures except seminal vesicles: bladder neck, external sphincter, rectum, levator muscles, and/or pelvic wall

N1: Metastasis in regional lymph node(s) M1: Distant metastasis

Used with permission of the American Joint Committee on Cancer (AJCC), Chicago, Illinois The original and primary

source for this information is the AJCC Cancer Staging Manual, seventh edition (2010) published by Springer

Science + Business Media

common histological grade in the specimen and the secondary score reflecting the second most common grade The primary and sec-ondary scores are added to yield overall Gleason score Thus, Gleason scores between 1 and 3 represent well-differentiated adeno-carcinomas, and scores between 8 and 10 represent poorly differen-tiated cancers [11, 12]

Generally, Gleason pattern one tissue contains simple round glands with uniform size, shape, and spacing The nuclei and nucleoli are markedly enlarged Gleason pattern two tumors show more variation in glandular size and shape and appear incompletely circumscribed Haphazardly separated glands among bands of fibrous stroma characterize pattern three, the most common microscopic pattern of prostate adenocarcinoma Pattern four tumor cells are organized into closely packed or fused glands, which invade the stroma with ragged infiltrative edges Gleason pattern five tumors contain solid sheets of anaplastic cells with comedo-like necrosis in cribriform nests [12, 13]

Breast cancer is usually discovered either through screening mammography or detection of a breast lump [2 15] Abnormal mammogram findings include breast masses, microcalcifications, asymmetries between the breasts, and architectural distortions Malignant breast lumps typically present in women over 30 years old as asymptomatic, painless masses which are fixed to surround-ing tissue [16] Patients with an abnormal mammogram and/or suspicious breast mass must undergo large-core needle biopsies for pathologic diagnosis Approximately ten-core biopsies are pre-ferred, each with diameter between 14 and 18 gauge and length between 1 and 3 cm For women without palpable masses, mam-mogram or ultrasound guidance is used to precisely localize the lesion [15, 16].

Extensive use of screening mammography has led to increased diagnoses of noninvasive breast carcinoma or ductal carcinoma in situ (DCIS) [14, 15] DCIS encompasses a wide spectrum of dis-eases with multiple staging and treatment options In general, DCIS has low metastatic potential but must be completely excised with either radical mastectomy or lumpectomy to prevent local recurrence [14, 17–19] The most important prognostic factors influencing local recurrence of DCIS include lesion size, adequacy

of resection, histologic grade, and patient age [15] It is therefore important for surgeons to obtain a wide surgical margin, preferably

10 mm in each dimension In addition, pathologists must examine biopsy tissue for areas of microscopic stromal invasion or microin-

vasion The AJCC Cancer Staging Manual classifies microinvasion

as T1mic, a subset of T1 breast cancer [2] Sentinel lymph node biopsy and axillary lymph node dissections are not necessary with DCIS unless the patient has high-grade disease or documentation

of microinvasion [17–19]

Invasive breast cancers require complete operative excision plus sentinel lymph node biopsy and/or axillary lymph node dis-section (ALND) Important operative findings for staging include the size of the primary tumor and presence of chest wall invasion

If the primary tumor is invasive and sentinel lymph node biopsy is positive, ALND should be considered to evaluate for metastases Breast lymphatics drain by way of three major routes: axillary, transpectoral, and internal mammary Any other lymph node metastases, with the exception of supraclavicular spread, are con-sidered metastatic disease (M1) [2]

Additional workup for suspected breast carcinoma might include supplementary breast imaging, chest imaging, and laboratory work with complete blood count and liver function tests Breast ultra-sound is useful to assess primary lesions in women with dense breasts, precisely locate breast masses, and evaluate ipsilateral axillary lymph nodes Breast MRI can be used to evaluate for occult disease, either

in the ipsilateral or contralateral breast, and to screen for nous breast lesions [15, 16] Women with advanced- stage cancer

synchro-3.3 Breast Cancer

Clinical

Staging Workup

should have a CT scan of the chest, abdomen, and pelvis and possibly radionuclide bone scan [15] The most common sites of breast can-cer metastasis are the bone, brain, liver, and lung [2] (see Table 3 for specific TMN staging criteria).

Adenocarcinoma, which may be either noninvasive or invasive, is the most common histologic type of breast cancer The noninva-sive adenocarcinomas include DCIS and lobular carcinoma in situ (LCIS) [15] The most common histologic type of invasive breast adenocarcinoma is ductal carcinoma, NOS (not otherwise

3.3.1 Histology and

Grading

Table 3

TNM classification and stage grouping for breast adenocarcinoma

0 Tis, N0, M0 Tis: Ductal or lobular carcinoma in situ

N0: No regional lymph node metastasis M0: No distant metastasis

IA T1, N0, M0 T1: Tumor 2 cm or less in greatest dimension

T4, N1, M0

T4, N2, M0

T4: Tumor of any size with direct extension to the chest wall or skin

IIIC Any T, N3, M0 N3: Metastases in ten or more axillary lymph nodes; or in

infraclavicular lymph nodes; or in clinically detected ipsilateral internal mammary lymph nodes in the presence of one or more positive level I or II axillary nodes; or in more than three axillary lymph nodes with micrometastases or macrometastases

by sentinel lymph node biopsy but not clinically detected; or in ipsilateral supraclavicular lymph nodes

IV Any T, Any N, M1 M1: Distant metastasis

Used with permission of the American Joint Committee on Cancer (AJCC), Chicago, Illinois The original and primary

source for this information is the AJCC Cancer Staging Manual, seventh edition (2010) published by Springer

Science + Business Media

specified); however, other subtypes of invasive disease include infiltrating lobular, mucinous, medullary, and papillary carcinoma Familiar subtypes of breast cancer that do not represent special pathologic categories include Paget’s disease and inflammatory carcinoma Paget’s disease of the nipple is a variant of high-grade DCIS in subareolar breast ducts but can also be associated with invasive carcinoma [15] Inflammatory carcinoma is a clinical diagnosis and considered an aggressive variant of infiltrating duc-tal carcinoma, NOS [14].

Breast cancer grading is applicable for DCIS and all invasive carcinomas [2 20] DCIS is graded on a three-tiered system based

on nuclear characteristics Grade 1 or low-grade DCIS cells tain small, round, and uniform nuclei with evenly dispersed chromatin (Fig 3a) Cribriform and micropapillary architectures are common, and neoplastic cells form geometric bulbous projec-tions around which the cells are polarized [14] Grade 3 or high-grade DCIS tumor cells are pleomorphic with high nuclear-cytoplasmic ratio, coarse chromatin, and large nucleoli (Fig 3b) Mitoses are frequent, and necrosis often occurs in the center of ducts surrounded by a solid pattern of neoplastic cells [14] The presence of necrosis within DCIS automatically qualifies the specimen as grade 2 or 3 [20]

con-Invasive breast carcinoma (Fig 4) is graded based on three tologic components: (1) extent of gland and tubule formation, (2) degree of nuclear pleomorphism, and (3) number of mitotic figures Pathologists assign between one and three points in each of these dimensions, with one point for the most differentiated histology and three points for the least differentiated histology In the tubule/gland formation category, one point is given for tubule formation in more than 75% of the tumor mass and three points are given for tubule formation in less than 10% of the tumor mass [14, 20]

his-Fig 3 Breast ductal carcinoma in situ (DCIS) (a) Grade 1 solid breast DCIS cells contain small, round nuclei

with evenly dispersed chromatin (b) Grade 3 breast DCIS cells show nuclear pleomorphism and coarse

chro-matin Courtesy of Chad Livasy, MD (with permission of Springer)

Likewise, nuclear pleomorphism is assessed with one point for small, regular, and uniform nuclei and three points for marked variation among nuclei Low numbers of mitotic figures receive one point and high numbers receive three points The final Nottingham grade for invasive breast carcinoma is determined by totaling points, with low-grade carcinomas between three and five total points and high-grade carcinomas between eight and nine total points [20].

Ovarian cancer is notoriously asymptomatic until it metastasizes, but there is no reliable screening method for this disease Therefore, the majority of patients present with symptoms of advanced disease such as bloating, abdominal or pelvic pain, and early satiety [21–

24] Routine pelvic exams may sometimes reveal earlier stage cinomas presenting as solid, irregular, and fixed adnexal masses in postmenopausal women Suspected ovarian cancer cases should have a transvaginal ultrasound to more accurately determine tumor size and consistency, solid versus cystic [25] Additional workup for metastatic ovarian cancer includes chest imaging, serum chem-istries, and liver function tests [25] Women with suspected ovarian cancer should also have serum glycoprotein CA-125 measurement CA-125 levels are useful in evaluating the success of future treat-ments including surgery, radiation, or chemotherapy Abdominopelvic CT or MRI may also demonstrate sites of meta-static disease The most common sites for ovarian spread include the peritoneum, diaphragmatic, and liver surfaces; however, peri-toneal ovarian lesions are not classified as distant metastases M1 disease is characterized by metastases to the liver, lung, skeleton parenchyma, supraclavicular nodes, and axillary nodes [2]

car-3.4 Ovarian Cancer

Clinical

Staging Workup

Fig 4 Poorly differentiated, infiltrating ductal carcinoma of the breast showing

cells with high mitotic rates and high-grade nuclei Courtesy of Chad Livasy, MD (with permission of Springer)

Surgery is necessary for ovarian cancer cytoreduction and also plays a role in definitive staging [21–24] Surgical staging should include the following: (1) aspiration and cytologic evaluation of ascites; (2) inspection of the upper abdomen, bowel surfaces, omentum, appendix, and pelvic organs with resection of suspicious lymph nodes; (3) pelvic and para-aortic lymph node dissection for patients with suspicious nodules outside the pelvis; and (4) total abdominal hysterectomy, bilateral salpingo-oophorectomy, and partial omentectomy [21, 25] An experienced gynecologic oncol-ogist should perform this extensive surgery, with a pathologist available for frozen specimen interpretation.

Epithelial ovarian cancer may spread using any of three mary pathways First, the tumor can penetrate the ovarian capsule and directly invade adjacent structures such as the uterus, bladder, rectum, or pelvic peritoneum Second, tumors can spread via lym-phatics to the pelvic or para-aortic lymph nodes Finally, ovarian tumor cells can escape into the peritoneal cavity and spread through the abdomen using peristalsis and the diaphragm’s respiratory motions [21] (see Table 4 for specific TMN staging criteria).Ovarian tumor histology and grading are especially valuable for pre-dicting prognosis and planning treatment of early-stage ovarian can-cers [25] Histopathologic subtypes of ovarian cancer include epithelial, malignant germ cell tumors, and sex cord-stromal tumors Epithelial tumors are the most common ovarian malignancies Sex cord-stromal tumors, including granulosa cell tumors, are extremely rare and arise from the ovarian stroma and/or sex cord derivatives [26] While epithelial tumors usually occur in postmenopausal women, many types of germ cell and sex cord-stromal tumors occur

pri-in younger women Different types of ovarian cancer present with unique symptoms depending on histology; however, tumor grading criteria are usually applied to epithelial tumors [26, 27]

Epithelial tumors arise as adenocarcinomas from the mation of ovarian coelomic epithelium and surrounding stroma [21–24] Histologic subtypes of epithelial tumors include high- grade serous, mucinous, endometrioid, clear-cell, and low-grade serous carcinomas [22–24] Serous carcinoma, the most common subtype of epithelial ovarian tumor, has a mixture of cystic, papil-lary, or solid growth patterns which infiltrate surrounding fibrotic stroma (Fig 5) Psammoma bodies, or small areas of calcification around products of cellular breakdown, are common in serous car-cinomas Mucinous carcinomas are large, multilocular cystic tumors composed of columnar cells with stratified nuclei and coarse chro-matin Endometrioid carcinomas have glandular or papillary archi-tecture and resemble endometrial adenocarcinomas Clear-cell carcinomas are characterized by cells that are cuboidal or polygonal with abundant cytoplasmic glycogen, and central vesicular nuclei characterize clear-cell carcinomas Although all malignant epithelial

transfor-3.4.1 Histology and

Grading

tumors have metastatic potential, high-grade serous and clear-cell tumors carry a worse prognosis than the other subtypes [26, 27].All epithelial ovarian tumors are assigned both nuclear and architectural grades on a scale from 1 to 3 [27] This grading can inform the decision for adjuvant chemotherapy for stage IA and IB epithelial ovarian cancer Nuclear grade 1 cells have mildly enlarged, uniform nuclei with evenly dispersed chromatin Nuclear grade 3 cells have enlarged and pleomorphic nuclei with irregular coarse chromatin and prominent nuclei Architectural grade 1 ovarian

Table 4

TNM classification and stage grouping for epithelial ovarian carcinoma

I T1, N0, M0 T1: Tumor limited to ovaries

N0: No regional lymph node metastasis

IA T1a, N0, M0 T1a: Tumor limited to one ovary; capsule intact, no tumor on

ovarian surface No malignant cells in ascites or peritoneal washings

IB T1b, N0, M0 T1b: Tumor limited to both ovaries; capsules intact, no tumor

on ovarian surface No malignant cells in ascites or peritoneal washings

IC T1c, N0, M0 T1c: Tumor limited to one or both ovaries with any of the

following: capsule ruptured, tumor on ovarian surface, malignant cells in ascites, or peritoneal washings

II T2, N0, M0 T2: Tumor involves one or both ovaries with pelvic extension

and/or implants IIA T2a, N0, M0 T2a: Extension and/or implants on the uterus and/or tube(s)

No malignant cells in ascites or peritoneal washings IIB T2b, N0, M0 T2b: Extension to and/or implants on other pelvic tissues No

malignant cells in ascites or peritoneal washings IIC T2c, N0, M0 T2c: Pelvic extension and/or implants (T2a or T2b) with

malignant cells in ascites or peritoneal washings III T3, N0, M0 T3: Tumor involves one or both ovaries with microscopically

confirmed peritoneal metastasis outside the pelvis IIIA T3a, N0, M0 T3a: Microscopic peritoneal metastasis beyond the pelvis (no

macroscopic tumor) IIIB T3b, N0, M0 T3b: Macroscopic peritoneal metastasis beyond the pelvis 2 cm

or less in greatest dimension

Any T, N1, M0 T3c: Peritoneal metastasis beyond the pelvis more than 2 cm in greatest dimension and/or regional lymph node metastasis

N1: Regional lymph node metastasis

IV Any T, any N, M1 M1: Distant metastasis (excludes peritoneal metastases)

Used with permission of the American Joint Committee on Cancer (AJCC), Chicago, Illinois The original and primary

source for this information is the AJCC Cancer Staging Manual, seventh edition (2010) published by Springer

Science+Business Media

Fig 5 Epithelial cell ovarian carcinoma of serous type Epithelial cells display

high-grade nuclei and complex papillary growth with treelike pattern Courtesy

of Chad Livasy, MD (with permission of Springer)

carcinomas are well differentiated and composed predominantly of glands Architectural grade 3 cells are poorly differentiated and com-posed mostly of solid areas rather than glandular structures [27].The TNM classification system has evolved to accommodate increas-ing knowledge about cancer biology Efforts are ongoing to keep the system both synchronized with the most sophisticated cancer tech-nology and simple for ease of clinician/patient use One improve-ment in precision of the TMN seventh edition, for example, was inclusion of breast cancer lymph node micrometastatic disease.New molecular technologies, such as genomic and proteomic profiling of tumors, and microRNA profiling are likely to be inte-grated into future TNM staging systems These technologies are allowing clinicians to subclassify cancers based on their molecular signatures in addition to anatomic extent of disease spread We have already identified many prognostic and targetable genetic variations in cancer cells One example of this, as discussed above,

is the EGFR genetic mutation in some lung cancers Continued efforts to identify these types of mutations have the potential to dramatically improve cancer survivorship

Acknowledgment

Special thanks to William Funkhouser, M.D., and Chad Livasy, M.D., pathologists at the University of North Carolina’s School of Medicine, for including their images in this chapter

3.5 Future Directions

for Cancer

Classification

1 Rubin E, Rubin R, Aaronson S (2005)

Neoplasia In: Rubin E (ed) Rubin’s

pathol-ogy: clinicopathologic foundations of

medi-cine, 4th edn Lippincott Williams & Wilkins,

Baltimore, pp 165–213

2 Greene F, Page D, Fleming I (2002) AJCC

cancer staging manual, 6th edn Springer,

New York

3 Rubin P, Williams JP, Okunieff P, Rosenblatt

JD, Sitzmann JV (2001) Statement of the

clinical oncologic problem In: Rubin P

(ed) Clinical oncology: a multidisciplinary

approach for physicians and students, 8th edn

W.B Saunders, Philadelphia, pp 1–31

4 Spitalnik PF, Santagnese PA (2001) The

pathol-ogy of cancer In: Rubin P (ed) Clinical

oncol-ogy: a multidisciplinary approach for physicians

and students W.B Saunders, Philadelphia,

pp 47–61

5 Houtte P, McDonald S, Chang A (2001) Lung

cancer In: Rubin P (ed) Clinical oncology: a

multidisciplinary approach for physicians and

students, 8th edn W.B Saunders, Philadelphia,

pp 823–844

6 Moran C, Suster S (2007) Tumors of the lung

and pleura In: Fletcher C (ed) Diagnostic

his-topathology of tumors Elsevier, Philadelphia,

pp 181–214

7 Khan OA, Fitzgerald JJ, Field ML, Soomro

I, Beggs FD, Morgan WE, Duffy JP (2004)

Histological determinants of survival in

completely resected T1-2N1M0

nons-mall cell cancer of the lung Ann Thorac

Surg 77(4):1173–1178 doi: 10.1016/j.

athoracsur.2003.08.080

8 Sayar A, Turna A, Kilicgun A, Solak O, Urer

N, Gurses A (2004) Prognostic significance of

surgical-pathologic multiple-station N1

dis-ease in non-small cell carcinoma of the lung

Eur J Cardiothorac Surg 25(3):434–438

doi: 10.1016/j.ejcts.2003.12.005

9 Suster S, Cesar M (2007) Tumors of the lungs

and pleura In: Damjanov I (ed) Cancer

grad-ing manual Sprgrad-inger, New York, pp 23–30

10 Roach M, Small E, Reese D, Carroll P (2001)

Urologic and male genital cancers In: Rubin

P (ed) Clinical oncology: a multidisciplinary

approach for physicians and students, 8th edn

W.B Saunders, Philadelphia, pp 523–564

11 Routh JC, Leibovich BC (2005)

Adenocarcinoma of the prostate:

epidemiolog-ical trends, screening, diagnosis, and surgepidemiolog-ical

management of localized disease Mayo Clin

Proc 80(7):899–907 doi: 10.4065/80.7.899

12 Damjanov I, Mikuz G (2007) Tumors of the kidney and the male urogenital system In: Damjanov I (ed) Tumor grading manual Springer, New York, pp 55–63

13 Ro J, Amin M, Kim K, Ayala A (2007) Tumors

of the male genital tract: prostate and nal vesicles In: Fletcher C (ed) Diagnostic histopathology of tumors, vol 1.3 Elsevier, Philadelphia, pp 749–811

14 Ellis I, Pinder S, Lee A (2007) Tumors of the breast In: Fletcher C (ed) Diagnostic histopathology of tumors, vol 1.3 Elsevier, Philadelphia, pp 903–970

15 Prosnitz L, Iglehart J, Winer E (2001) Breast cancer In: Rubin P (ed) Clinical oncology: a multidisciplinary approach for physicians and students, 8th edn W.B Saunders, Philadelphia,

pp 252–266

16 Osuch JR, Reeves MJ, Pathak DR, Kinchelow

T (2003) BREASTAID: clinical results from early development of a clinical deci- sion rule for palpable solid breast masses Ann Surg 238(5):728–737 doi: 10.1097/01 sla.0000094446.78844.ae

17 Lambert LA, Ayers GD, Hwang RF, Hunt

KK, Ross MI, Kuerer HM, Singletary SE, Babiera GV, Ames FC, Feig B, Lucci A, Krishnamurthy S, Meric-Bernstam F (2006) Validation of a breast cancer nomogram for predicting nonsentinel lymph node metasta- ses after a positive sentinel node biopsy Ann Surg Oncol 13(3):310–320 doi: 10.1245/ ASO.2006.03.078

18 Peintinger F, Kuerer HM, Anderson K, Boughey JC, Meric-Bernstam F, Singletary

SE, Hunt KK, Whitman GJ, Stephens T, Buzdar AU, Green MC, Symmans WF (2006) Accuracy of the combination of mammography and sonography in predicting tumor response

in breast cancer patients after neoadjuvant chemotherapy Ann Surg Oncol 13(11):1443–

1449 doi: 10.1245/s10434-006-9086-9

19 Sgroi DC (2010) Preinvasive breast cancer Annu Rev Pathol 5:193–221 doi: 10.1146/ annurev.pathol.4.110807.092306

20 Fan F, Thomas P (2007) Tumors of the breast In: Damjanov I (ed) Tumor grading manual Springer, New York, pp 75–81

21 Bhoola S, Hoskins WJ (2006) Diagnosis and management of epithelial ovarian cancer Obstet Gynecol 107(6):1399–1410 doi: 10.1097/01 AOG.0000220516.34053.48

22 Cho KR (2009) Ovarian cancer update: sons from morphology, molecules, and mice

les-Arch Pathol Lab Med 133(11):1775–1781

doi: 10.1043/1543-2165-133.11.1775

23 Cho KR, Shih Ie M (2009) Ovarian cancer

Annu Rev Pathol 4:287–313 doi: 10.1146/

annurev.pathol.4.110807.092246

24 Kurman RJ, Shih Ie M (2010) The

ori-gin and pathogenesis of epithelial

ovar-ian cancer: a proposed unifying theory Am

J Surg Pathol 34(3):433–443 doi: 10.1097/

PAS.0b013e3181cf3d79

25 Perez C, Grigsby P, Mutch D, Chao K, Basil

J (2001) Gynecologic tumors In: Rubin P

(ed) Clinical oncology: a multidisciplinary approach for physicians and students, 8th edn W.B Saunders, Philadelphia, pp 462–522

26 Zaloudek C (2007) Tumors of the female genital tract: ovary, fallopian tube and broad and round ligaments In: Fletcher C (ed) Diagnostic histopathology of tumors, vol 1 Elsevier, Philadelphia, pp 567–651

27 Fan F, Damjanov I (2007) Tumors of the female genital organs In: Damjanov I (ed) Tumor grading manual Springer, New York,

pp 64–74

Virginia Espina (ed.), Molecular Profiling: Methods and Protocols, Methods in Molecular Biology, vol 1606,

DOI 10.1007/978-1-4939-6990-6_2, © Springer Science+Business Media LLC 2017

Chapter 2

Innovations in Clinical Trial Design in the Era

of Molecular Profiling

Julia D Wulfkuhle, Alexander Spira, Kirsten H Edmiston,

and Emanuel F Petricoin III

Key words Biomarkers, Clinical trial, Design, Endpoints, Neoadjuvant therapy, Tumor

1 Introduction

In the field of oncology, the traditional process for determining which drugs will benefit patients is both long and expensive It is estimated to take 10–15 years and cost over two billion dollars for

a promising therapy to reach the marketplace [1 2] This process

is composed of a series of clinical trials beginning with the ment of drug safety and dosing (Phase I), establishment of efficacy and indications (Phase II), and testing for superiority over current standard of care treatment regimens (Phase III) Historically, can-cer has been studied and therapeutic agents have been evaluated

assess-based on organ site and clinical staging Despite the fact that our best opportunity to cure cancer occurs in its earliest stages, most early oncology drug development begins in the metastatic setting However, therapeutic benefit in the metastatic setting doesn’t nec-essarily translate to benefit in the adjuvant setting for primary tumors, and a lack of benefit in the metastatic setting could also miss efficacy potential in earlier stages of disease [1].

Traditional clinical trial structure usually addresses a single therapeutic question in a large population of patients with disease

in a defined organ site and focuses on reducing false-positive and false-negative results These trials are usually comparative and look for superiority of an experimental therapy over a concur-rently accrued control group with a primary endpoint of overall survival Enhanced efficacy in any patient subgroup is determined only retrospectively, and survival data take many years to accrue

In the past, these types of trials were acceptable because there were limited therapies available, and cancer within an organ site was thought to be a homogeneous disease However, the last two decades have brought a literal explosion in our knowledge of can-cer at the cellular and molecular level such that numerous sub-types are being described based on biomarker expression and genetic mutations that transcend traditional classifications of the disease [3–5] Drug development has experienced a concomitant revolution in response to this knowledge Focus has shifted from general cytotoxic agents to the development of targeted agents that interfere with cancer cell growth and survival more selectively and potentially protect normal tissues to a greater extent While improved molecular characterization of cancer presents great prospects for a future of truly individualized therapy, these advances are dividing cancer into smaller and smaller subpopula-tions and create situations where patient selection is based on underpinning molecular alterations rather than organ site, which present significant challenges for evaluating new therapies The abundance of newly identified molecular targets and therapeutic agents means we no longer have the luxury of focusing on a lim-ited number of therapies to evaluate going forward but must instead replace the traditional, large phase II and phase III adju-vant trial models with smaller, shorter, and more focused trials in the neoadjuvant, adjuvant, and metastatic setting These trials need to be more efficient and adaptive in order to quickly assess the efficacy of new agents and develop new companion diagnos-tics when appropriate without the burden of the current regula-tory structure of multiple sequential trials and long evaluation periods The goal of these trials should be to deliver effective ther-apies to patients with narrowly defined disease subtypes We are now seeing a substantial shift from the traditional multiphase trial model to an increase in phase II neoadjuvant trials in earlier-stage disease incorporating surrogate endpoints for long- term survival

to assess efficacy of therapeutic agents In this chapter, we will describe recent innovations in clinical trial design and endpoint assessment that are moving the field of oncology toward achieving these goals.

2 Innovations in Clinical Trial Endpoints

Establishing the efficacy of a therapeutic agent is the most tant goal of oncology drug development Traditional gold stan-dard endpoints are overall survival (OS), i.e., the time from diagnosis to death, and progression-free survival (PFS), a shorter endpoint related more directly to quality of life [6] (Table 1) For years, response rate (RR) as determined by tumor volume changes from imaging studies, and physical exams have been what investi-gators and patients have used to measure efficacy of a drug For a variety of reasons, tumor RR is not the best endpoint, and it is not generally accepted in obtaining regulatory approval for a therapy Instead, the more stringent measurement of overall survival (OS)

impor-as a “defining” endpoint reflects what is considered most indicative

of clinical benefit (Table 1)

2.1 Traditional

Clinical Trial Endpoints

Table 1

Overall survival (OS) Time from randomization until death from any cause

Disease-free survival (DFS) Time from randomization until recurrence of tumor or

death from any cause Objective response rate (ORR) Proportion of patients with tumor size reduction (sum of

partial + complete responses) Time to progression (TTP) Time from randomization until objective tumor

progression Progression-free survival (PFS) Time from randomization until objective tumor

progression or death from any cause Event-free survival (EFS) Time from randomization to (a) progression of disease that

precludes surgery, (b) local or distant recurrence, or (c) death from any cause

Pathological complete response (pCR) The absence of residual invasive cancer on hematoxylin and

eosin evaluation of the complete resected (breast) specimen and all sampled regional lymph nodes following completion of neoadjuvant systemic therapy Residual cancer burden (RCB) A continuous variable measurement derived from primary

tumor dimensions, cellularity of the tumor bed, and axillary nodal burden at the time of surgery

While OS is frequently utilized by the US Food and Drug Administration (FDA) as a major endpoint in the drug approval process, it presents significant problems for assessment of targeted therapeutics [8] Phase II and III studies can often take one or more years to accrue adequately large numbers of patients Additional years must pass, usually three or more, before the appropriate numbers of patients live out the natural course of their disease This prolonged time frame required to assess OS is one reason that it may not be the most practical primary endpoint for phase II studies OS is also affected by the use of subsequent lines

of therapies as well as improvements in supportive care that may obscure the effects of a particular therapeutic agent during the assessment period Moreover, prolonged time courses add to the cost of clinical trials as one gets further along in drug development With the tremendous costs of bringing novel agents to market, pharmaceutical manufacturers have reduced incentive to evaluate novel risky therapies, particularly for targeted agents where only a small portion of the patient population may be responsive and bio-markers to identify those patients may not be well defined Finally, judging a therapy by OS alone ignores reduction of symptoms and improvements in the quality of life

Based on all the drawbacks of using the traditional OS point, clinicians also utilize alternative measurement endpoints that are reached earlier in the treatment course and may be more meaningful to the patient Newer clinical trials are incorporating quality-of-life (QOL) measurements such as the “Lung Cancer Symptom Scale” which is a subjective QOL questionnaire filled out by patients and nurses that reports subjective symptoms, in addition to the standard measurements of response rate and overall survival

end-Endpoints such as time to progression (TTP), which is the time from randomization to the time of progressive disease, may

be used as a surrogate for OS, particularly in clinical trials including patients that have received prior treatment or have inoperable or metastatic disease TTP and progression-free survival (PFS) (Table 1) have been correlated with OS in patients with rectal can-cer [9] These endpoints offer several advantages over traditional

OS in clinical trial design Both TTP and PFS permit smaller ple sizes and shorter study durations – (e.g., months as compared

sam-to years) TTP and PFS do not require demonstration of tumor mass shrinkage and thus are useful in trials designed to evaluate cytostatic agents that arrest growth but do not shrink the tumor Lastly, TTP and PFS can be measured in real time after a single line

of therapy, and when these measures are used, the designation of a response is not confounded by subsequent events The disadvan-tage of TTP and PFS, as compared to OS, in clinical trials is the requirement for costly, frequent, and careful imaging assessment for progression [10, 11]

Since 2000, the Response Evaluation Criteria in Solid Tumors (RECIST) criteria have been used to assess therapeutic response [12] (Table 2) These criteria are based on the premise that tumor shrinkage reflects a positive outcome of antineoplastic therapy Utilizing two-dimensional CT and/or MRI imaging data, tumor response rate is categorized as complete response, partial response, stable, or progressive disease based on the extent and duration of tumor shrinkage (Table 2) As an example, the drug sorafenib, a multiple tyrosine kinase inhibitor, was approved for use in patients with advanced hepatocellular carcinoma (HCC) [14, 15] The use

of sorafenib was found to prolong survival in patients with HCC

by 3 months and was associated with a 31% increase in OS at

1 year Nevertheless, the RECIST response rate was only 2% [14] The same drug has been studied in patients with advanced renal cell cancer [16] The response rate was only 10%, but this consti-tuted a significant prolongation in survival that led to FDA approval of this agent in advanced renal cell cancer

There are a number of reasons why RECIST-defined response rates are an inadequate endpoint for efficacy in many clinical tri-als First, tumor cytotoxicity may not result in rapid shrinkage, especially in tumors that induce large amounts of stroma (rather than cellular elements that may “die” with chemotherapy) Second, RECIST categories are broad and thus somewhat impre-cise: a tumor that shrinks 29% and a tumor that grows 19% are both considered stable disease, while these two responses are clearly different [13] Third, despite imaging advances there are issues with scanning variability and resolution, particularly with lesions <1 cm, that can significantly affect accuracy Lastly, RECIST criteria cannot be used in tumors that localize or metas-tasize to the bone since these lesions do not shrink, making it difficult in the assessment of diseases such as prostate cancer, as well as in many hematologic malignancies that cannot be mea-sured with tumor size

sustained for at least 4 weeks Progressive disease (PD) At least 20% increase in tumor size with

no CR, PR or SD documented before the increase of disease Stable disease (SD) Neither PR or PD criteria are met

The concept of achieving greater therapeutic benefit from ducing agents earlier in the disease process has led to an increase in the number of phase II trials in the neoadjuvant setting in early-stage cancers A critical principle of neoadjuvant clinical trials is that tumor response serves as a surrogate endpoint for, and is strongly correlated with long-term patient survival.

intro-One measure of tumor response that is uniquely available in the neoadjuvant trial setting is pathologic(al) complete response (pCR) pCR is defined as the absence of residual invasive cancer

on hematoxylin and eosin evaluation of the complete resected specimen and all sampled regional lymph nodes following com-pletion of neoadjuvant systemic therapy [7] (Table 1) A number

of studies and meta-analyses have shown that pCR is associated with long- term survival, and it has been adopted as a primary end-point in neoadjuvant trials, particularly in breast cancer [17–25] Some of these studies have shown that pCR predicts long-term outcome more effectively in various breast cancer subtypes rather than the disease as a whole [24, 25] However, there is some con-troversy as to whether sufficient evidence exists to confidently use pCR as a surrogate endpoint to recommend or reject a treatment for general clinical use [26] Some of these concerns have been based on inconsistent definitions of pCR regarding the presence

of nodal metastasis, residual in situ carcinoma, and residual larity [22, 23], as well as some seemingly contradictory results for therapeutic regimens tested in the adjuvant vs neoadjuvant set-ting [1] While these concerns are legitimate, more precise criteria for pCR have been accepted as the current standard, and many studies do show promising correlations between pCR and EFS and OS on the individual patient level Future studies with more homogeneous tumor subtypes where there are potentially greater differences in pCR rates between treatment arms may help to determine more definitively the strength of the relationship between pCR and long-term outcomes [27]

cellu-A second endpoint that has been more recently developed for assessment of response to neoadjuvant therapy is residual cancer burden (RCB) RCB is a pathology-based, continuous variable measurement derived from primary tumor dimensions, cellularity

of the tumor bed, and axillary nodal burden in breast cancer [28]

It was developed out of a need to assess more effectively the trum of residual disease (RD) found in surgical specimens, which can range from pCR to therapeutic resistance Each of the compo-nents of the RCB index has prognostic significance, and as a result, the RCB endpoint is strongly prognostic and represents the range

spec-of residual disease seen in a treated population RCB can also be divided into four classes (RCB 0-RCB III) that associate with prog-nosis Extensive RD (RCB III) associates with poor prognosis,

whereas patients in category 0 or I (pCR or minimal RD) have very similar 5-year prognosis and represent a pool of patients that benefit from neoadjuvant therapy [28].

The component measurements of RCB were found to be highly reproducible among pathologists over a large number of specimens [29], and the methods for assessment of RCB compo-nents could readily be incorporated into routine pathologic review with no increase in patient care cost An RCB calculation program

is freely available online: (

rou-tine analysis parameter for neoadjuvant studies, it is an endpoint for the I-SPY 2 TRIAL described in Subheading 3

3 Innovations in Clinical Trial Design in the Era of Molecular Profiling

In 2012, the US Food and Drug Administration published

“Guidance for Industry Pathological Complete Response in Neoadjuvant Treatment of High-Risk Early-Stage Breast Cancer: Use as an Endpoint to Support Accelerated Approval” [7] These guidelines provide two pathways for drug approval in the neoad-juvant setting for breast cancer The first is to perform a single randomized trial where all patients are treated neoadjuvantly, but

a sufficient number of patients are accrued to demonstrate agent superiority for both the pCR and longer-term event-free survival (EFS) endpoints When all patients have been accrued and the pCR endpoint has been successfully achieved, the pharmaceutical company can file for accelerated approval Full approval would fol-low at the 3-year EFS endpoint if superiority of the new agent is demonstrated If superior EFS is not observed, accelerated approval will be withdrawn A second pathway is to conduct a neoadjuvant trial for the pCR assessment and simultaneously begin a confirmatory adjuvant trial for long-term survival If a suc-cessful pCR endpoint is achieved in the neoadjuvant trial, acceler-ated approval is considered only when accrual is complete for the confirmatory trial [30]

Pertuzumab, a humanized monoclonal antibody that binds the dimerization domain of the HER2 receptor, was awarded accelerated approval in September of 2013 and was the first targeted agent to achieve this landmark approval under the FDA’s 2012 draft guid-ance Approval of the drug for neoadjuvant use was granted based on several lines of evidence The first was that pertuzumab had gained approval for use in advanced breast cancer based on results of the phase III CLEOPATRA trial CLEOPATRA was a randomized con-trol trial for metastatic breast cancer patients that demonstrated sig-nificant improvement in PFS and OS in patients treated with pertuzumab added to trastuzumab/docetaxel over a placebo/trastu-zumab/docetaxel control arm [31] Additional evidence came out of

two phase II neoadjuvant trials, NeoSphere, and TRYPHAENA The NeoSphere trial accrued just over 400 patients with Stage II/III HER2+ breast cancer The patients were randomized to receive one

of four regimens: pertuzumab/trastuzumab/docetaxel, zumab/docetaxel, pertuzumab/trastuzumab, or pertuzumab/docetaxel Significantly higher pCR rates were seen in the pertu-zumab/trastuzumab/docetaxel arm over the trastuzumab/docetaxel arm (39.3% vs 21.5%, respectively) [32] The TRYPHAENA trial was designed mainly as a cardiac safety study incorporating pertu-zumab into various standard of care regimens for Stage II/III HER2+ breast cancers [33] Though there was no control arm in this trial, the pCR rates achieved in three treatment arms of this study exceeded those of NeoSphere [30] The final critical step toward meeting the requirements for accelerated approval of pertuzumab was achieved when the phase III adjuvant trial, APHINITY, com-pleted enrollment of 3800 patients in August 2013 The APHINITY trial is a double-blind, placebo-controlled study in which patients are allowed to receive any standard chemotherapy regimen and are then randomized to receive either pertuzumab or placebo The primary endpoint is invasive DFS Results are not yet known but will move toward a definitive determination of whether DFS is improved with the addition of adjuvant pertuzumab and possibly lead to full approval for its use in the neoadjuvant setting [30]

trastu-The goal of neoadjuvant therapy trials is to increase the pace at which potentially beneficial agents become available for use Accelerated approval can provide access to therapies with solid evi-dence of improved response, while long-term survival data are awaited It is not necessarily meant to change the standard of care for patients, particularly in the absence of evidence for long-term outcome improvement, but it may be challenging for some oncol-ogists to pass on an opportunity to provide a beneficial agent such

as pertuzumab to their high-risk HER2+ patients Others may choose not to provide a drug approved under these guidelines sim-ply because long-term outcome data are missing [30]

Regardless, the accelerated approval of pertuzumab establishes

a new paradigm for a pathway to make promising new drugs able to patients more quickly and cheaply This new pipeline opens the door for many new additional trials and drugs to be approved using this approach, and not just in breast cancer [30] In the last

avail-5 years, several new clinical trials with innovative designs have been implemented to work toward the goals of neoadjuvant therapy in the era of molecular profiling: assessment of multiple drugs for activity in various molecular subtypes of disease in an efficient and cost-effective manner

In many respects, the concepts of patient stratification and biomarker- based design in clinical trials have been in use for many years (reviewed in [34, 35]) These early designs are gradually

3.2 New Clinical Trial

Designs

being replaced by newer designs that include multiple therapeutic agents, multiple molecular groups, and even multiple disease types [35] Among these new designs are the basket trial and platform (or umbrella) trials Some of these trials also incorporate adaptive elements based on Bayesian statistics that make them even more innovative and efficient [5].

The concept of the basket trial is based on the hypothesis that the presence of a molecular aberration in a tumor predicts response to

a particular targeted therapy independent of other factors such as tissue of origin, disease stage, and histopathology Basket trials generally include multiple tumor types, and a master protocol is implemented to screen patients for selected molecular alterations that are actionable therapeutic targets involved in tumorigenesis, tumor growth, or metastasis Patients are then assigned to the appropriate sub-protocol corresponding to an identified molecular alteration in their tumor (Fig 1a) This type of trial is becoming more attractive because it provides an opportunity to look at very rare mutations and rare disease types as well as mutations that are difficult to study in the context of a single disease [36]

3.2.1 Basket Trials

Fig 1 Diagrams for basket trials and platform trial formats (a) Basket trial; (b) platform trial

The CUSTOM (Molecular Profiling and Targeted Therapy for Advanced Thoracic Malignancies) trial was the first completed basket trial that evaluated response to multiple targeted therapies against specific molecular aberrations for multiple tumor histologic types simultaneously It was designed to identify biomarkers in, and to eval-uate five targeted therapies in patients grouped by molecular markers and tumor type in advanced non-small cell lung cancer (NSCLC), small cell lung cancer (SCLC), and thymic malignancies [37] The drugs included in the trial were erlotinib (EGFR mutations); MK2206, an AKT inhibitor (PIK3CA, AKT1, and PTEN muta-tions); the MEK inhibitor selumetinib (KRAS, HRAS, NRAS, and BRAF mutations); lapatinib (HER2 mutations); and sunitinib (KIT and PDGFRA mutations) The intent was to evaluate each drug in all three tumor types for a total of 15 arms, but the trial ultimately only enrolled 45 of 257 patients with actionable mutations, with over half

of those being NSCLC tumors in the erlotinib and selumetinib arms This was largely due to the low incidence of the targeted mutations

in specific tumor types The trial did surpass their response goal of 40% in the erlotinib arm of the trial with an ORR of 60% (9/15 PR), but the ORR in selumetinib arm was only 11% (1/9 PR) These results were consistent with other clinical trial data for these agents and demonstrated the potential capability to identify compounds of high and low activity in small cohorts of patients with selected molec-ular alterations using this type of trial design [37]

Two currently accruing basket trials, the ASCO-sponsored Targeted Agent and Profiling Utilization Registry (TAPUR) trial and the National Cancer Institute-Molecular Analysis for Therapy Choice (NCI-MATCH) trial, greatly expand the scope and size of the CUSTOM trial design The TAPUR trial began accruing patients in March 2016 and has a screening goal of over 1000 patients in a 3-year time frame It aims to address two common issues in precision medicine: lack of access to FDA-approved drugs for off-label use and the lack of safety and efficacy data for these agents TAPUR will address these issues by making these FDA- approved drugs available

to patients at no charge and creating a registry to record and make key clinical outcomes freely available [38] The trial is open to patients with advanced solid tumors, multiple myeloma, and non-Hodgkin’s lymphoma and will evaluate 10–15 drugs in cohorts of up to 35 patients that will be defined by tumor type, genomic alteration, and drug The TAPUR trial allows for therapy matching when the pre-scribed molecular tests used for patient selection were completed in any CAP-/CLIA- accredited laboratory The primary endpoint is OR

or stable disease for 16 weeks

The NCI-MATCH trial is complementary to the TAPUR trial

in that they both aim to elucidate beneficial uses for targeted pies The trial opened in August 2015 to patients with advanced solid tumors and lymphoma with a goal that 25% of patients enrolled have rare tumor types Ten treatment arms were initially

thera-available and included drugs that were FDA-approved for a cancer indication or were being tested experimentally and had evidence for effectiveness against tumors with a specific molecular alteration The trial design allows for arms to be added or dropped over time Next-generation sequencing is used to screen tumors for action-able genetic abnormalities, and patients are then assigned to an appropriate treatment arm if a drug is available [39] Unlike the TAPUR trial, NCI-MATCH requires testing to be performed by one of four predetermined CAP/CLIA laboratories In November

2015, the NCI-MATCH trial paused for a planned interim analysis upon reaching the benchmark of enrolling 500 patients for muta-tion screening in just 3 months This enrollment rate was about five times higher than expected and reflects a need for these types

of avenues to provide access and opportunities to treat patients with limited options The interim analysis revealed that the match rate of mutations to a targeted therapy was approximately 9%, which is close to the 10% rate projected in the original design However, it was also noted that the actual mutation prevalence rates found were much lower than expected based on estimates from TCGA and other databases [39, 40]

Over half (19/33) of patients enrolled to treatment arms of the trial had rare tumor types It was found that not all matched patients were enrolled in a treatment arm of the trial for a variety

of reasons One is that the turnaround time for sequencing results was up to 48 days in some cases, and this often left the door open for changes in health status that affected eligibility Another issue that is impacting NCI-MATCH is the requirement that all patients must have molecular profiling done at specific molecular labs and requires prospective molecular analysis With these requirements, there have been issues with adequate biopsy sample quantity and quality [41] The trial reopened in May 2016 with 24 treatment arms available for cohorts of up to 35 patients and a fourfold increase in laboratory capacity that is anticipated to deliver sequenc-ing results in less than 2 weeks after biopsy [39] It should be noted that while there is some minor duplication in the evaluable actionable mutations between NCI-MATCH and TAPUR, there is very little overlap in the targeted agents matched to them In the NCI- MATCH trial, patients are allowed to switch to a second treatment arm if their first treatment proves ineffective and they have a second actionable mutation; those who fail to match to the complement of open trial arms at the time of screening can be notified if an appropriate therapeutic agent for any identified muta-tions is brought into the trial

These trials are an important step in challenging the traditional dogma of treating cancers based on organ site, stage, and tumor histology and are beginning to focus more on patient- and tumor- specific biology They are generating significant interest in the medical community by bringing precision medicine into clinical