Clinical Sequencing Contributes to a BRCA-Associated Cancer Rediagnosis That Guides an Effective Therapeutic Course

Authors:
Jocelyn S. Chapman From the Departments of Obstetrics, Gynecology, and Reproductive Sciences, Medicine, Bioengineering and Therapeutic Sciences, Radiology and Biomedical Imaging, Pathology, Laboratory Medicine, and Epidemiology and Biostatistics, Helen Diller Family Comprehensive Cancer Center, University of California, San Francisco, San Francisco, California; and (Present address) Human Oncology and Pathogenesis Program, Memorial Sloan Kettering Cancer Center, New York, New York.
From the Departments of Obstetrics, Gynecology, and Reproductive Sciences, Medicine, Bioengineering and Therapeutic Sciences, Radiology and Biomedical Imaging, Pathology, Laboratory Medicine, and Epidemiology and Biostatistics, Helen Diller Family Comprehensive Cancer Center, University of California, San Francisco, San Francisco, California; and (Present address) Human Oncology and Pathogenesis Program, Memorial Sloan Kettering Cancer Center, New York, New York.

Search for other papers by Jocelyn S. Chapman in
Current site
Google Scholar
PubMed Close
 MD
,
Saurabh Asthana From the Departments of Obstetrics, Gynecology, and Reproductive Sciences, Medicine, Bioengineering and Therapeutic Sciences, Radiology and Biomedical Imaging, Pathology, Laboratory Medicine, and Epidemiology and Biostatistics, Helen Diller Family Comprehensive Cancer Center, University of California, San Francisco, San Francisco, California; and (Present address) Human Oncology and Pathogenesis Program, Memorial Sloan Kettering Cancer Center, New York, New York.

Search for other papers by Saurabh Asthana in
Current site
Google Scholar
PubMed Close
 PhD
,
Lindsay Cade From the Departments of Obstetrics, Gynecology, and Reproductive Sciences, Medicine, Bioengineering and Therapeutic Sciences, Radiology and Biomedical Imaging, Pathology, Laboratory Medicine, and Epidemiology and Biostatistics, Helen Diller Family Comprehensive Cancer Center, University of California, San Francisco, San Francisco, California; and (Present address) Human Oncology and Pathogenesis Program, Memorial Sloan Kettering Cancer Center, New York, New York.

Search for other papers by Lindsay Cade in
Current site
Google Scholar
PubMed Close
,
Matthew T. Chang From the Departments of Obstetrics, Gynecology, and Reproductive Sciences, Medicine, Bioengineering and Therapeutic Sciences, Radiology and Biomedical Imaging, Pathology, Laboratory Medicine, and Epidemiology and Biostatistics, Helen Diller Family Comprehensive Cancer Center, University of California, San Francisco, San Francisco, California; and (Present address) Human Oncology and Pathogenesis Program, Memorial Sloan Kettering Cancer Center, New York, New York.

Search for other papers by Matthew T. Chang in
Current site
Google Scholar
PubMed Close
 PharmD
,
Zhen Wang From the Departments of Obstetrics, Gynecology, and Reproductive Sciences, Medicine, Bioengineering and Therapeutic Sciences, Radiology and Biomedical Imaging, Pathology, Laboratory Medicine, and Epidemiology and Biostatistics, Helen Diller Family Comprehensive Cancer Center, University of California, San Francisco, San Francisco, California; and (Present address) Human Oncology and Pathogenesis Program, Memorial Sloan Kettering Cancer Center, New York, New York.

Search for other papers by Zhen Wang in
Current site
Google Scholar
PubMed Close
 MD
,
Charles J. Zaloudek From the Departments of Obstetrics, Gynecology, and Reproductive Sciences, Medicine, Bioengineering and Therapeutic Sciences, Radiology and Biomedical Imaging, Pathology, Laboratory Medicine, and Epidemiology and Biostatistics, Helen Diller Family Comprehensive Cancer Center, University of California, San Francisco, San Francisco, California; and (Present address) Human Oncology and Pathogenesis Program, Memorial Sloan Kettering Cancer Center, New York, New York.
From the Departments of Obstetrics, Gynecology, and Reproductive Sciences, Medicine, Bioengineering and Therapeutic Sciences, Radiology and Biomedical Imaging, Pathology, Laboratory Medicine, and Epidemiology and Biostatistics, Helen Diller Family Comprehensive Cancer Center, University of California, San Francisco, San Francisco, California; and (Present address) Human Oncology and Pathogenesis Program, Memorial Sloan Kettering Cancer Center, New York, New York.

Search for other papers by Charles J. Zaloudek in
Current site
Google Scholar
PubMed Close
 MD
,
Stefanie Ueda From the Departments of Obstetrics, Gynecology, and Reproductive Sciences, Medicine, Bioengineering and Therapeutic Sciences, Radiology and Biomedical Imaging, Pathology, Laboratory Medicine, and Epidemiology and Biostatistics, Helen Diller Family Comprehensive Cancer Center, University of California, San Francisco, San Francisco, California; and (Present address) Human Oncology and Pathogenesis Program, Memorial Sloan Kettering Cancer Center, New York, New York.
From the Departments of Obstetrics, Gynecology, and Reproductive Sciences, Medicine, Bioengineering and Therapeutic Sciences, Radiology and Biomedical Imaging, Pathology, Laboratory Medicine, and Epidemiology and Biostatistics, Helen Diller Family Comprehensive Cancer Center, University of California, San Francisco, San Francisco, California; and (Present address) Human Oncology and Pathogenesis Program, Memorial Sloan Kettering Cancer Center, New York, New York.

Search for other papers by Stefanie Ueda in
Current site
Google Scholar
PubMed Close
 MD
,
Eric A. Collisson From the Departments of Obstetrics, Gynecology, and Reproductive Sciences, Medicine, Bioengineering and Therapeutic Sciences, Radiology and Biomedical Imaging, Pathology, Laboratory Medicine, and Epidemiology and Biostatistics, Helen Diller Family Comprehensive Cancer Center, University of California, San Francisco, San Francisco, California; and (Present address) Human Oncology and Pathogenesis Program, Memorial Sloan Kettering Cancer Center, New York, New York.

Search for other papers by Eric A. Collisson in
Current site
Google Scholar
PubMed Close
 MD
, and
Barry S. Taylor From the Departments of Obstetrics, Gynecology, and Reproductive Sciences, Medicine, Bioengineering and Therapeutic Sciences, Radiology and Biomedical Imaging, Pathology, Laboratory Medicine, and Epidemiology and Biostatistics, Helen Diller Family Comprehensive Cancer Center, University of California, San Francisco, San Francisco, California; and (Present address) Human Oncology and Pathogenesis Program, Memorial Sloan Kettering Cancer Center, New York, New York.
From the Departments of Obstetrics, Gynecology, and Reproductive Sciences, Medicine, Bioengineering and Therapeutic Sciences, Radiology and Biomedical Imaging, Pathology, Laboratory Medicine, and Epidemiology and Biostatistics, Helen Diller Family Comprehensive Cancer Center, University of California, San Francisco, San Francisco, California; and (Present address) Human Oncology and Pathogenesis Program, Memorial Sloan Kettering Cancer Center, New York, New York.
From the Departments of Obstetrics, Gynecology, and Reproductive Sciences, Medicine, Bioengineering and Therapeutic Sciences, Radiology and Biomedical Imaging, Pathology, Laboratory Medicine, and Epidemiology and Biostatistics, Helen Diller Family Comprehensive Cancer Center, University of California, San Francisco, San Francisco, California; and (Present address) Human Oncology and Pathogenesis Program, Memorial Sloan Kettering Cancer Center, New York, New York.
From the Departments of Obstetrics, Gynecology, and Reproductive Sciences, Medicine, Bioengineering and Therapeutic Sciences, Radiology and Biomedical Imaging, Pathology, Laboratory Medicine, and Epidemiology and Biostatistics, Helen Diller Family Comprehensive Cancer Center, University of California, San Francisco, San Francisco, California; and (Present address) Human Oncology and Pathogenesis Program, Memorial Sloan Kettering Cancer Center, New York, New York.

Search for other papers by Barry S. Taylor in
Current site
Google Scholar
PubMed Close
 PhD
Volume/Issue:
Volume 13: Issue 7
Online Publication Date:
Jul 2015
DOI:
https://doi.org/10.6004/jnccn.2015.0101
Full access

Cancer is currently classified and treated using an approach based on tissue of origin. Ambiguous or incorrect diagnoses, however, are common and often go unnoticed. Clinical cancer sequencing can provide diagnostic precision, therapeutic direction, and hereditary cancer risk assessment. This report presents a patient with an initial diagnosis of metastatic pancreatic adenocarcinoma (PDA), a disease with a dismal prognosis. Tumor sequencing revealed genomic abnormalities inconsistent with PDA, instead suggesting serous ovarian cancer. This molecular rediagnosis was further refined by the identification of a BRCA2 truncating mutation in the tumor, subsequently confirmed to be a germline event. These findings prompted the initiation of platinum-based chemotherapy, which produced a life-altering response, and referral to genetic counseling for her offspring. These results suggest that clinical tumor sequencing can simultaneously clarify diagnoses, guide therapy, and inform familial risk, even in patients with end-stage metastatic disease, making the case for the development of specific strategies to deploy sequencing coupled with big data in oncology to improve clinical cancer management.

NCCN: Continuing Education

Accreditation Statement

This activity has been designated to meet the educational needs of physicians and nurses involved in the management of patients with cancer. There is no fee for this article. No commercial support was received for this article. The National Comprehensive Cancer Network (NCCN) is accredited by the ACCME to provide continuing medical education for physicians.

NCCN designates this journal-based CME activity for a maximum of 1.0 AMA PRA Category 1 Credit™. Physicians should claim only the credit commensurate with the extent of their participation in the activity.

NCCN is accredited as a provider of continuing nursing education by the American Nurses Credentialing Center‘s Commission on Accreditation.

NCCN designates this educational activity for a maximum of 1.0 contact hours. Accreditation as a provider refers to recognition of educational activities only; accredited status does not imply endorsement by NCCN or ANCC of any commercial products discussed/displayed in conjunction with the educational activity. Kristina M. Gregory, RN, MSN, OCN, is our nurse planner for this educational activity.

All clinicians completing this activity will be issued a certificate of participation. To participate in this journal CE activity: 1) review the learning objectives and author disclosures; 2) study the education content; 3) take the posttest with a 66% minimum passing score and complete the evaluation at http://education.nccn.org/node/70209; and 4) view/print certificate.

Release date: July 9, 2015; Expiration date: July 9, 2016

Learning Objectives

Upon completion of this activity, participants will be able to:

  • Describe the potential for clinical cancer sequencing in diagnosis, treatment, and hereditary risk assessment

  • Identify strategies for the application of sequencing to clinical cancer management

The practice of genomically driven cancer medicine is accelerating in oncology clinics. Such precision oncology depends on the timely identification of one or more molecular abnormalities for which tumor growth and progression depend, and then rationally targeting these pharmacologically. Consequently, treatments targeting specific genetic lesions that span many tumor types are being tested independent of the organ of origin for individual cancers.1 Nevertheless, systemic therapies based on an organ of origin classification remain the mainstay of cancer management. The accuracy of pathologic diagnosis, therefore, is critical to ensure timely delivery of life-saving and potentially curative primary treatment to the estimated 1.6 million patients with cancer diagnosed annually in the United States.2 However, conventional pathologic approaches are limited in their diagnostic sensitivity.3,4 The economics, feasibility, and availability of clinical cancer genome sequencing are improving and will complement pathologic review to improve diagnostic accuracy. Moreover, such molecular characterization during the standard diagnostic process will not only simultaneously facilitate the choice of therapy, targeting somatic aberrations of significance, but also identify the 5% to 10% of patients with inherited cancer-causing mutations,5,6 impacting the care of index patients and their families.

This report presents a patient initially diagnosed with stage IV pancreatic adenocarcinoma (PDA) based on conventional imaging and pathologic assessment. Clinical tumor sequencing was performed to identify somatic and germline mutations and focal DNA copy number alterations that, when analyzed in the context of population-scale cancer genome data, refined the original diagnosis and identified a lesion of clinical significance to inform both individualized therapy and family counseling.

Case and Results

A 54-year-old woman presented to a community hospital with jaundice, anorexia, and 20-pound weight loss. An abdominal and pelvic CT showed a 5-cm mass at the head of the pancreas with intrahepatic and extrahepatic biliary dilatation and involvement of the celiac plexus with extensive retroperitoneal lymphadenopathy (Figure 1A and B). Results of a CT-guided biopsy of the pancreatic mass showed a poorly differentiated adenocarcinoma. Initial review of the specimen with immunohistochemistry indicated that it was positive for CK7 and CK19 and negative for CK20, TTF-1, ER, CGA, and synaptophysin, with these results interpreted as PDA. On external review, a tertiary medical center pathologist concurred with this diagnosis. The patient underwent decompressive bile duct stenting and was referred to our institution with a diagnosis of stage IV PDA.

On presentation, because of the patient's relatively young age at diagnosis of advanced-stage cancer, she was referred for genetic counseling, which revealed no family history of hereditable cancer. Plasma was sent for cell-free DNA (cfDNA) sequencing.7 A baseline staging CT scan was obtained before the planned initiation of gemcitabine and nanoparticle albumin-bound paclitaxel therapy.8 Results of this CT showed evolving findings, including ascites with peritoneal thickening, a complex 6.1-cm right adnexal lesion, and the absence of a discrete pancreatic mass, which was reinterpreted as a lymph node conglomerate (Figure 1B and C). Extensive celiac plexus retroperitoneal lymphadenopathy with bile duct compression remained. An examination of a repeat biopsy again showed poorly differentiated adenocarcinoma. However, immunohistochemistry staining at our institution suggested a pelvic serous carcinoma (Figure 1C, inset). These new CT results and disparate pathologic conclusions suggested the possibility that this patient was incorrectly diagnosed with PDA and instead had advanced serous ovarian cancer. However, before the biopsy results were available, the patient's clinical course rapidly deteriorated because of untreated progressive disease, pulmonary emboli, pleural effusions, and renal failure, and she required respiratory support in the intensive care unit. She then developed sepsis and organ failure, progressing to a moribund clinical state that precluded therapeutic intervention until her clinical status could be improved. A detailed clinical timeline of this case is available in Appendix 1 (available online, in this article, at JNCCN.org).

Due to the uncertainty in diagnosis from conventional imaging and marker studies, targeted capture and deep sequencing of tumor DNA for 236 genes and related loci of interest was performed.9 This uncovered a complex combination of genomic alterations among key cancer-associated genes in the tumor genome, including focal genomic amplifications of MYC (8q24.21), CCND3 (6p21.1), and SOX2 (3q26.33). We also identified mutations in TP53 G245D, DNMT3A Y197*, and BRCA2 I605fs*11 (Figure 2A).

Figure 1
Figure 1

Imaging and pathology reveal ambiguous diagnosis. (A) Coronal CT of the abdomen shows confluent retroperitoneal and porta hepatis lymphadenopathy (top arrows) resulting in moderate obstruction of the common bile duct (left and bottom arrows). A mass is inseparable from the pancreas (right arrow). (B) The status of key markers determined from immunohistochemistry at the time of initial diagnosis (left) and after referral to our institution (right) is shown with the expected percentage of patients with positive and negative results in the indicated tumor types. Immunohistochemistry results correspond to indicated imaging (left and right, respectively). (C) Results of an abdominal CT show a normal-enhancing pancreas (left, top arrows) without focal masses or pancreatic ductal dilation. Leftmost arrow indicates enlarged retroperitoneal lymph nodes located close to the pancreas, with ascites apparent around the liver and spleen. Right image of the pelvis shows an enhancing mass in the left adnexa (right arrow) and a predominantly cystic mass in the right adnexa (other arrows). Inset image shows that tumor cells are present separately and in small clusters with pleomorphic atypical nuclei, dark chromatin, and prominent nucleoli, and that the cytoplasm is eosinophilic and vacuolated (hematoxylin-eosin, original magnification x400).

Abbreviations: adeno, adenocarcinoma; NA, not assessed.

Citation: Journal of the National Comprehensive Cancer Network J Natl Compr Canc Netw 13, 7; 10.6004/jnccn.2015.0101

Although KRAS was sequenced to an average depth of 446-fold coverage, no mutations were present. Sequencing of cfDNA also indicated no KRAS mutations in the circulation, despite abundant TP53 G245D (5% in cfDNA), suggesting approximately 10% of cfD-NA was circulating tumor DNA. Subsequent germline testing confirmed the BRCA2 frameshift mutation as an inherited truncating mutation introducing a premature stop codon (Appendix 2; available online, in this article, at JNCCN.org). This allele (G602fs; adjacent to the same polyA tract) is identical to a previously identified BRCA2 germline mutation in a patient diagnosed with an MYC-amplified, TP53-mutant ovarian carcinoma, who experienced a 32-month complete response to platinum therapy.10 Moreover, this is a low-incidence variant; only a single report of this likely pathogenic allele in the germline of patients with breast or ovarian cancer has been found in relevant clinical variant databases.11,12 Although the presence of mutant TP53 (G245) is generally not discriminatory for a specific cancer type (Appendix 3; available online, in this article, at JNCCN.org), the absence of a KRAS mutation in a putative PDA diagnosis was unexpected because of its presence in 93% of these tumors.13,14 These findings, in addition to the presence of the other amplified and mutated genes, suggested that this patient's disease was unlikely pancreatic in origin.

To determine whether a diagnosis of PDA was consistent with these genomic findings, we assembled a compendium of mutational and DNA copy number alteration data from 9122 samples across 30 tumor types (see Appendix 4, available online, in this article, at JNCCN.org).

Although the overall frequency in PDA for any one of the genomic alterations in the genes observed in our index patient (excluding TP53) was 21.4%, these gene aberrations are much more commonly observed in high-grade serous ovarian carcinomas, affecting 60.2% of patients (Figure 2B). Moreover, if we assume the PDA diagnosis was true, then the probability that we would encounter a patient with PDA with this ensemble of lesions in their genome is vanishingly small (<10-8; see “Methods”). Alternatively, because these lesions are common in serous ovarian cancers (Figure 2B), we determined the relative probability of observing this combination of lesions in a patient diagnosed with high-grade ovarian cancer compared with PDA (see “Methods”). We estimated this tumor was 1280 times more likely to have a molecular diagnosis of serous ovarian carcinoma based solely on the genomic abnormalities observed in the tumor genome. These mutational findings became available only during the patient's hospitalization and confirmed the repeat imaging and histologic findings suggesting ovarian cancer (Appendix 1; available online, in this article, at JNCCN.org).

Although the patient's clinical status did not improve sufficiently to allow for the administration of cytotoxic chemotherapy15 (Appendix 1, available at JNCCN.org), the presence of a sensitizing BRCA2 lesion suggested that tumor-directed therapy could be considered. Based partly on these genomic analyses and after consultation with a multidisciplinary ethics committee, the patient was administered 150 mg of carboplatin (area under the curve of 5, based on an estimated glomerular filtration rate of 0 mg/mL/hr), which is a platinum-based treatment with maximal therapeutic activity in ovarian cancers with germline or somatic BRCA1 and BRCA2 mutations,16,17 such as the index patient. A week after the administration of this low dose of platinum chemotherapy, the patient's organ failure reversed, enabling her to participate in genetic counseling with her children. She was eventually discharged and has since had optimal interval cytoreduction to no visible disease and with minimal residual disease seen at surgery after completing 6 cycles of neoadjuvant chemotherapy. Although her pretreatment CA-125 was 4495 U/mL, this decreased to 37 U/mL after therapy and before

Figure 2
Figure 2

Clinical sequencing of a tumor genome refines a molecular diagnosis and treatment course with systemic therapy. (A) Clinical next-generation sequencing reveals a tumor genome with a high burden of both focal (candidate target identified in red) and broad DNA copy number alterations, with 3 mutant genes observed (top; copy number segmentation is shown across the autosomes). (B) The frequency of the ensemble of genomic aberrations identified by sequencing (panel A) was assessed in a large compendium of genomic data across >9000 tumors. Although the frequency of alteration in these genes was 21.4% in pancreatic cancers (asterisk in center of figure), they affect 60.2% of all serous ovarian cancers (asterisk at top of figure) with significant heterogeneity across tumor types and organ systems. (C) Coronal abdominal CT after the third cycle of chemotherapy demonstrates near-complete resolution of the previously noted confluent retroperitoneal and porta hepatis lymphadenopathy. A biliary stent (left arrows) and a normal-appearing pancreas (right arrow) are noted. There is resolution of ascites. The adnexal mass (not shown) is decreased in size compared with the pretreatment scan. Black arrow into Panel D indicates the time point in the course of treatment at which the image was obtained. (D) The evolving response to platinum treatment was apparent in the rapid reduction of both pretreatment CA-125 (left axis, light blue) and the frequency of the TP53 G245D mutation in tumor-derived circulating cell-free DNA (cfDNA; right axis, dark blue). Treatment (red dotted lines) and the dates of sampling (x-axis) are indicated.

Citation: Journal of the National Comprehensive Cancer Network J Natl Compr Canc Netw 13, 7; 10.6004/jnccn.2015.0101

surgery (Figure 2C and D). To confirm this objective clinical response molecularly, we sequenced cfDNA from peripheral whole blood obtained after 2 cycles of single-agent carboplatin therapy. The TP53 G245D mutation, present at a frequency of 5% before treatment, was now undetectable (limit of detection, 0.003%). Together, these data reaffirm the deep molecular remission experienced by the patient as a result of platinum therapy.

Discussion

This report presents a patient with a molecular rediagnosis enabled by clinical sequencing and population-scale tumor genome data, which also directed a life-saving therapeutic course and informed the patient and her family of a hereditary cancer risk. The clinical ramifications of this sequencing are multifaceted, informing several aspects of the patient's cancer management. Although the correct diagnosis for this particularly ambiguous case was determined by repeated imaging and specialty pathology review, this required 3 independent pathology consultations across 3 institutions, which resulted in considerable costs in both time and significant morbidity. This report demonstrates how integration of targeted sequencing and population-scale genomic data across many cancer types can provide a level of diagnostic accuracy that is difficult to achieve via histopathology alone. The importance of this molecular reclassification is validated by the long-established observation that solid tumors originating in the abdominopelvic compartment have markedly different prognoses and site-specific treatment regimens that improve clinical outcome.18 Additionally, the patient was negative for CA19-9, the commonly used tumor marker to test for PDA. However, because of the sensitivity and specificity of CA19-9,19 its absence is diagnostically uninformative. Therefore, rediagnoses through a combination of tumor sequencing with population-scale genomic data, such as in the present case, is complementary to the work of others who are using sequencing to refine the origins of cancers of an unknown primary.18,20

The genomic aberrations identified in this tumor genome informed a therapeutic intervention. This patient experienced significant and ongoing sensitivity to platinum-based chemotherapy, a response predicted from prior work that demonstrated both longer platinum-free intervals and overall survival in patients with loss-of-function BRCA1 and BRCA2 mutations compared with their BRCA wild-type counterparts.16,17 Although cisplatin-based chemotherapy is indicated for select patients with PDA who have germline BRCA mutations, the lack of a family history across 3 generations did not suggest a familial syndrome; therefore, germline testing was not initially performed.21 Given the patient's declining functional status, the medical oncologist initially treating the patient for PDA recommended gemcitabine and nab-paclitaxel8 in lieu of oxaliplatin, irinotecan, leucovorin, and fluorouracil (FOLFIRINOX).22 As a result, the patient's poor functional status would have prevented platinum treatment in the absence of a molecularly refined diagnosis confirming ovarian cancer with a sensitizing BRCA2 mutation. Lastly, the identification of this pathogenic germline BRCA2 mutation triggered screening and counseling for a hereditary cancer risk among the patient's family members.

The broader impact of large-scale clinical sequencing on patient outcomes in oncology has yet to be established. However, as with similar reports of sequencing showing exceptional treatment response in individual patients, specific strategies can be proposed and formally tested with the accrual of sufficient data. First, our data indicate that, notwithstanding the present focus on somatic mutation detection with molecular testing, the concurrent sequencing of patient-matched normal specimens allow for simultaneous somatic mutation and germline pathogenic allele detection in the same test, despite its associated costs, logistical complexities, and reporting ambiguities. This would have provided the present patient a hereditary cancer diagnosis in a timely fashion while also reducing costs and inefficiencies associated with a separate reflexive germline test. Nevertheless, few commercial or academic centers require such paired tumor and matched normal samples for clinical implementation of sequencing assays. Our results also provide a demonstration, albeit limited, of the potential for such testing in patients with advanced-stage disease in whom treatment may otherwise appear futile. Although the benefits of establishing a lesion of therapeutic significance in advanced metastatic disease are unknown, this case highlights the ability of such testing to contribute to a rediagnosis that leads to successful alternative systemic therapy. Moreover, even though the correct diagnosis was eventually achieved with conventional means, it came too late to affect care in a rapidly declining patient without compelling evidence for a sensitizing mutation. In the absence of molecular evidence to support platinum sensitivity, the provision of chemotherapy to a dying patient would have been otherwise difficult to justify.15,23–26 Because of the shorter treatment window for patients with advanced-stage disease, these results allude to the potential for initiating prospective sequencing of patients with cancer earlier in the diagnostic process, as would likely have benefited this patient. Together, these strategies indicate the opportunity for improving the care of molecularly defined patients across the spectrum of their cancer management.

Methods

Study Approval

The collection and genomic analysis of this patient sample was performed in accordance with a protocol approved by the Committee on Human Research at the University of California San Francisco (approval numbers 13-12574 & 13-13986). Informed consent for genomic analysis was obtained from the patient at the time of sample collection.

Sequencing and Bioinformatics

Tumor sequencing was performed as previously described,9 as was the profiling of cfDNA.7 Individual genomic lesions were subsequently inspected manually in our laboratory using primary sequencing data, as were DNA copy number alterations of interest genome-wide.

Large-Scale Cancer Genome Data

All genomic data presented here are available in the cBioPortal for Cancer Genomics27 and were inferred from either whole-exome sequencing or high-resolution single-nucleotide polymorphism arrays. These data included 426 serous ovarian carcinomas acquired from The Cancer Genome Atlas (TCGA) provisional and published data28 and 224 patients with PDA from TCGA and the International Cancer Genome Consortium.13,28,29 All mutations and DNA copy number alterations of interest were manually reviewed.

The Relative Probability of Cancer Diagnoses

To determine the relative probability of an ovarian versus PCA diagnosis given the ensemble of genomic aberrations observed from tumor sequencing, we calculated the probability of data given a PDA diagnosis assuming independence among alterations, which is approximately correct as the product of the probabilities of any 2 lesions was estimated to be greater than or equal to the rate at which they co-occur (data not shown). This probability is therefore equal to the product of the probabilities of each individual lesion among the 6 observed (Figure 2A) to which we added an additional event to represent the probability of being KRAS wild-type. The probability of any given lesion in a particular tumor type was estimated as its frequency in that population determined from our 9122-tumor compendium of genomic data. Due to the nature of the reported clinical sequencing results, we assumed that for the 229 genes that were sequenced but in which no alteration was reported, that these were either wild-type or of ambiguous functional impact or clinical relevance, and therefore assigned a probability equal to one. We performed a similar calculation for both ovarian and pancreas cancers, determined the overall probability of each diagnosis from their frequencies in the population (1.37% and 1.48%, respectively),30 and calculated the relative probability as the sum of the ratios of the probabilities of ovarian or pancreas cancer given the genomic aberrations and the overall probabilities of each diagnosis.

Dr. Collisson has disclosed that he receives support from the Doris Duke Charitable Foundation and the V Foundation Rombauer Fellowship; he also receives consulting fees from Guardant Health. Dr. Taylor has disclosed that his work is supported in part by the Prostate Cancer Foundation and the Sontag Foundation. The remaining authors have disclosed that they have no financial interests, arrangements, affiliations, or commercial interests with the manufacturers of any products discussed in this article or their competitors.

EDITOR

Kerrin M. Green, MA, Assistant Managing Editor, JNCCN—Journal of the National Comprehensive Cancer Network

Ms. Green has disclosed that she has no relevant financial relationships.

CE AUTHORS

Deborah J. Moonan, RN, BSN, Director, Continuing Education, has disclosed that she has no relevant financial relationships.

Ann Gianola, MA, Manager, Continuing Education Accreditation & Program Operations, has disclosed that she has no relevant financial relationships.

Kristina M. Gregory, RN, MSN, OCN, Vice President, Clinical Information Operations, has disclosed that she has no relevant financial relationships.

Rashmi Kumar, PhD, Senior Manager, Clinical Content, has disclosed that she has no relevant financial relationships.

Susan D. Darlow, PhD, Oncology Scientist/Medical Writer, has disclosed that she has no relevant financial relationships.

Acknowledgments

The authors are grateful to D. Rodriguez-Jaquez for her assistance throughout this work; J. Chmielecki, D. Lipson, and A. Talasaz for data sharing and access; and N. Schultz for critical reading of this manuscript.

Appendix 1

Appendix 1

Detailed Clinical Timeline.
Appendix 1

Appendix 2

Appendix 2
Appendix 2

Testing confirmed a germline BRCA2 II605fs*11 frameshift mutation. Blue indicates insertion of adenine in the polyA tract. The normal sequence is GGAAAAAAA{dupA}TACCG (the duplicated base is noted in the brackets). The adenine insertion causes a frameshift, and creates a premature stop codon rendering this allele nonfunctional. The mutation in this patient is identical to a previously identified BRCA2 germline mutation in a patient diagnosed with an MYC-amplified, TP53-mutant ovarian carcinoma (G602fs, adjacent to the same polyA tract).10

Citation: Journal of the National Comprehensive Cancer Network J Natl Compr Canc Netw 13, 7; 10.6004/jnccn.2015.0101

Appendix 3

Appendix 3
Appendix 3

TP53 mutations in cancer. The x-axis indicates cohorts used for comparative genomic analysis. The y-axis depicts the frequency of TP53 mutations in the cohort. TP53 mutations are ubiquitous and are therefore not discriminatory between most individual cancer types.

Data from the cBioPortal Web site. Available at: www.cbioportal.org. Accessed June 11, 2014.

Citation: Journal of the National Comprehensive Cancer Network J Natl Compr Canc Netw 13, 7; 10.6004/jnccn.2015.0101

Appendix 4

Appendix 4

Compendium of tumors used for comparative genomic analysis. For Figure 2B we used publically available data from 9122 primary tumors to assess the frequency of genetic abnormalities seen in our index case across cancer. Much of these data have been published and abbreviated references are included next to the cohort's name, where available. All cohorts who had at least one tumor with one of our patient's 5 genetic abnormalities (focal genomic amplifications of MYC, CCND3, and SOX2 or mutations in DNMT3A, and BRCA2) were included.

Data from the cBioPortal Web site. Available at: www.cbioportal.org. Accessed June 11, 2014.
Appendix 4

References

  • 1.↑

    McDermott U, Settleman J. Personalized cancer therapy with selective kinase inhibitors: an emerging paradigm in medical oncology. J Clin Oncol 2009;27:5650–5659.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 2.↑

    Siegel R, Ma J, Zou Z, Jemal A. Cancer statistics, 2014. CA Cancer J Clin 2014;64:9–29.

  • 3.↑

    Grilley-Olson JE, Hayes DN, Moore DT et al.. Validation of interobserver agreement in lung cancer assessment: hematoxylin-eosin diagnostic reproducibility for non-small cell lung cancer: the 2004 World Health Organization classification and therapeutically relevant subsets. Arch Pathol Lab Med 2013;137:32–40.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 4.↑

    Kamel HM. Trends and challenges in pathology practice: choices and necessities. Sultan Qaboos Univ Med J 2011;11:38–44.

  • 5.↑

    Bolton KL, Chenevix-Trench G, Goh C et al.. Association between BRCA1 and BRCA2 mutations and survival in women with invasive epithelial ovarian cancer. JAMA 2012;307:382–390.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 6.↑

    Yang D, Khan S, Sun Y et al.. Association of BRCA1 and BRCA2 mutations with survival, chemotherapy sensitivity, and gene mutator phenotype in patients with ovarian cancer. JAMA 2011;306:1557–1565.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 7.↑

    Talasaz A, Mortimer S, Sebisanovic D et al.. Use of the GUARDANT360 noninvasive tumor sequencing assay on 300 patients across colorectal, melanoma, lung, breast, and prostate cancers and its clinical utility [abstract]. J Clin Oncol 2014;32(Suppl):Abstract e22041.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 8.↑

    Von Hoff DD, Ervin T, Arena FP et al.. Increased survival in pancreatic cancer with nab-paclitaxel plus gemcitabine. N Engl J Med 2013;369:1691–1703.

  • 9.↑

    Frampton GM, Fichtenholtz A, Otto GA et al.. Development and validation of a clinical cancer genomic profiling test based on massively parallel DNA sequencing. Nat Biotechnol 2013;31:1023–1031.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 10.↑

    Cancer Genome Atlas Research Network. Integrated genomic analyses of ovarian carcinoma. Nature 2011;474:609–615.

  • 11.↑

    NM_000059.3(BRCA2):c.1813delA (p.Ile605Tyrfs) AND Neoplastic Syndromes, Hereditary - ClinVar - NCBI [Internet]. Available at: http://www.ncbi.nlm.nih.gov/clinvar/RCV000132177.1/. Accessed January 27, 2015.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 12.↑

    Landrum MJ, Lee JM, Riley GR et al.. ClinVar: public archive of relationships among sequence variation and human phenotype. Nucleic Acids Res 2014;42:D980–985.

  • 13.↑

    Biankin AV, Waddell N, Kassahn KS et al.. Pancreatic cancer genomes reveal aberrations in axon guidance pathway genes. Nature 2012;491:399–405.

  • 14.↑

    Waddell N, Pajic M, Patch AM et al.. Whole genomes redefine the mutational landscape of pancreatic cancer. Nature 2015;518:495–501.

  • 15.↑

    Levy MH, Adolph MD, Back A et al.. Palliative care. J Natl Compr Canc Netw 2012;10:1284–1309.

  • 16.↑

    Alsop K, Fereday S, Meldrum C et al.. BRCA mutation frequency and patterns of treatment response in BRCA mutation-positive women with ovarian cancer: a report from the Australian Ovarian Cancer Study Group. J Clin Oncol 2012;30:2654–2663.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 17.↑

    Farmer H, McCabe N, Lord CJ et al.. Targeting the DNA repair defect in BRCA mutant cells as a therapeutic strategy. Nature 2005;434:917–921.

  • 18.↑

    Tothill RW, Li J, Mileshkin L et al.. Massively-parallel sequencing assists the diagnosis and guided treatment of cancers of unknown primary. J Pathol 2013;231:413–423.

  • 19.↑

    Pleskow DK, Berger HJ, Gyves J et al.. Evaluation of a serologic marker, CA19-9, in the diagnosis of pancreatic cancer. Ann Intern Med 1989;110:704–709.

  • 20.↑

    Greco FA, Lennington WJ, Spigel DR, Hainsworth JD. Molecular profiling diagnosis in unknown primary cancer: accuracy and ability to complement standard pathology. J Natl Cancer Inst 2013;105:782–790.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 21.↑

    Tempero MA, Malafa MP, Behrman SW et al.. Pancreatic adenocarcinoma, version 2.2014: featured updates to the NCCN guidelines. J Natl Compr Canc Netw 2014;12:1083–1093.

  • 22.↑

    Conroy T, Desseigne F, Ychou M et al.. FOLFIRINOX versus gemcitabine for metastatic pancreatic cancer. New Engl J Med 2011;364:1817–1825.

  • 23.↑

    Kao S, Shafiq J, Vardy J, Adams D. Use of chemotherapy at end of life in oncology patients. Ann Oncol 2009;20:1555–1559.

  • 24.

    Keam B, Oh DY, Lee SH et al.. Aggressiveness of cancer-care near the end-of-life in Korea. Jpn J Clin Oncol 2008;38:381–386.

  • 25.

    Matsuyama R, Reddy S, Smith TJ. Why do patients choose chemotherapy near the end of life? A review of the perspective of those facing death from cancer. J Clin Oncol 2006;24:3490–3496.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 26.↑

    O'Brien ME, Borthwick A, Rigg A et al.. Mortality within 30 days of chemotherapy: a clinical governance benchmarking issue for oncology patients. Br J Cancer 2006;95:1632–1636.

    • PubMed
    • Search Google Scholar
    • Export Citation
  • 27.↑

    Cerami E, Gao J, Dogrusoz U et al.. The cBio cancer genomics portal: an open platform for exploring multidimensional cancer genomics data. Cancer Discov 2012;2:401–404.

  • 28.↑

    The Cancer Genome Atlas - Cancer Genome - TCGA [Internet]. Available at: http://cancergenome.nih.gov/. Accessed January 27, 2015.

  • 29.↑

    Zhang J, Baran J, Cros A et al.. International Cancer Genome Consortium data portal—a one-stop shop for cancer genomics data. Database (Oxford) 2011;2011:bar026.

  • 30.↑

    Siegel R, DeSantis C, Virgo K et al.. Cancer treatment and survivorship statistics, 2012. CA Cancer J Clin 2012;62:220–241.

Correspondence: Eric A. Collisson, MD, Department of Medicine, Division of Hematology and Oncology, Helen Diller Family Comprehensive Cancer Center, 1450 3rd Street, San Francisco, CA 94158. E-mail: collissonlab@gmail.com; Barry S. Taylor, PhD, Human Oncology and Pathogenesis Program, Memorial Sloan Kettering Cancer Center, 1275 York Avenue, Box 20, New York, NY 10065. E-mail: taylorb@mskcc.org

Supplementary Materials

  • Collapse
  • Expand
  • View in gallery

    Imaging and pathology reveal ambiguous diagnosis. (A) Coronal CT of the abdomen shows confluent retroperitoneal and porta hepatis lymphadenopathy (top arrows) resulting in moderate obstruction of the common bile duct (left and bottom arrows). A mass is inseparable from the pancreas (right arrow). (B) The status of key markers determined from immunohistochemistry at the time of initial diagnosis (left) and after referral to our institution (right) is shown with the expected percentage of patients with positive and negative results in the indicated tumor types. Immunohistochemistry results correspond to indicated imaging (left and right, respectively). (C) Results of an abdominal CT show a normal-enhancing pancreas (left, top arrows) without focal masses or pancreatic ductal dilation. Leftmost arrow indicates enlarged retroperitoneal lymph nodes located close to the pancreas, with ascites apparent around the liver and spleen. Right image of the pelvis shows an enhancing mass in the left adnexa (right arrow) and a predominantly cystic mass in the right adnexa (other arrows). Inset image shows that tumor cells are present separately and in small clusters with pleomorphic atypical nuclei, dark chromatin, and prominent nucleoli, and that the cytoplasm is eosinophilic and vacuolated (hematoxylin-eosin, original magnification x400).

    Abbreviations: adeno, adenocarcinoma; NA, not assessed.

  • View in gallery

    Clinical sequencing of a tumor genome refines a molecular diagnosis and treatment course with systemic therapy. (A) Clinical next-generation sequencing reveals a tumor genome with a high burden of both focal (candidate target identified in red) and broad DNA copy number alterations, with 3 mutant genes observed (top; copy number segmentation is shown across the autosomes). (B) The frequency of the ensemble of genomic aberrations identified by sequencing (panel A) was assessed in a large compendium of genomic data across >9000 tumors. Although the frequency of alteration in these genes was 21.4% in pancreatic cancers (asterisk in center of figure), they affect 60.2% of all serous ovarian cancers (asterisk at top of figure) with significant heterogeneity across tumor types and organ systems. (C) Coronal abdominal CT after the third cycle of chemotherapy demonstrates near-complete resolution of the previously noted confluent retroperitoneal and porta hepatis lymphadenopathy. A biliary stent (left arrows) and a normal-appearing pancreas (right arrow) are noted. There is resolution of ascites. The adnexal mass (not shown) is decreased in size compared with the pretreatment scan. Black arrow into Panel D indicates the time point in the course of treatment at which the image was obtained. (D) The evolving response to platinum treatment was apparent in the rapid reduction of both pretreatment CA-125 (left axis, light blue) and the frequency of the TP53 G245D mutation in tumor-derived circulating cell-free DNA (cfDNA; right axis, dark blue). Treatment (red dotted lines) and the dates of sampling (x-axis) are indicated.

  • View in gallery

    Testing confirmed a germline BRCA2 II605fs*11 frameshift mutation. Blue indicates insertion of adenine in the polyA tract. The normal sequence is GGAAAAAAA{dupA}TACCG (the duplicated base is noted in the brackets). The adenine insertion causes a frameshift, and creates a premature stop codon rendering this allele nonfunctional. The mutation in this patient is identical to a previously identified BRCA2 germline mutation in a patient diagnosed with an MYC-amplified, TP53-mutant ovarian carcinoma (G602fs, adjacent to the same polyA tract).10

  • View in gallery

    TP53 mutations in cancer. The x-axis indicates cohorts used for comparative genomic analysis. The y-axis depicts the frequency of TP53 mutations in the cohort. TP53 mutations are ubiquitous and are therefore not discriminatory between most individual cancer types.

    Data from the cBioPortal Web site. Available at: www.cbioportal.org. Accessed June 11, 2014.

Imaging and pathology reveal ambiguous diagnosis. (A) Coronal CT of the abdomen shows confluent retroperitoneal and porta hepatis lymphadenopathy (top arrows) resulting in moderate obstruction of the common bile duct (left and bottom arrows). A mass is inseparable from the pancreas (right arrow). (B) The status of key markers determined from immunohistochemistry at the time of initial diagnosis (left) and after referral to our institution (right) is shown with the expected percentage of patients with positive and negative results in the indicated tumor types. Immunohistochemistry results correspond to indicated imaging (left and right, respectively). (C) Results of an abdominal CT show a normal-enhancing pancreas (left, top arrows) without focal masses or pancreatic ductal dilation. Leftmost arrow indicates enlarged retroperitoneal lymph nodes located close to the pancreas, with ascites apparent around the liver and spleen. Right image of the pelvis shows an enhancing mass in the left adnexa (right arrow) and a predominantly cystic mass in the right adnexa (other arrows). Inset image shows that tumor cells are present separately and in small clusters with pleomorphic atypical nuclei, dark chromatin, and prominent nucleoli, and that the cytoplasm is eosinophilic and vacuolated (hematoxylin-eosin, original magnification x400).

Abbreviations: adeno, adenocarcinoma; NA, not assessed.

Clinical sequencing of a tumor genome refines a molecular diagnosis and treatment course with systemic therapy. (A) Clinical next-generation sequencing reveals a tumor genome with a high burden of both focal (candidate target identified in red) and broad DNA copy number alterations, with 3 mutant genes observed (top; copy number segmentation is shown across the autosomes). (B) The frequency of the ensemble of genomic aberrations identified by sequencing (panel A) was assessed in a large compendium of genomic data across >9000 tumors. Although the frequency of alteration in these genes was 21.4% in pancreatic cancers (asterisk in center of figure), they affect 60.2% of all serous ovarian cancers (asterisk at top of figure) with significant heterogeneity across tumor types and organ systems. (C) Coronal abdominal CT after the third cycle of chemotherapy demonstrates near-complete resolution of the previously noted confluent retroperitoneal and porta hepatis lymphadenopathy. A biliary stent (left arrows) and a normal-appearing pancreas (right arrow) are noted. There is resolution of ascites. The adnexal mass (not shown) is decreased in size compared with the pretreatment scan. Black arrow into Panel D indicates the time point in the course of treatment at which the image was obtained. (D) The evolving response to platinum treatment was apparent in the rapid reduction of both pretreatment CA-125 (left axis, light blue) and the frequency of the TP53 G245D mutation in tumor-derived circulating cell-free DNA (cfDNA; right axis, dark blue). Treatment (red dotted lines) and the dates of sampling (x-axis) are indicated.

Testing confirmed a germline BRCA2 II605fs*11 frameshift mutation. Blue indicates insertion of adenine in the polyA tract. The normal sequence is GGAAAAAAA{dupA}TACCG (the duplicated base is noted in the brackets). The adenine insertion causes a frameshift, and creates a premature stop codon rendering this allele nonfunctional. The mutation in this patient is identical to a previously identified BRCA2 germline mutation in a patient diagnosed with an MYC-amplified, TP53-mutant ovarian carcinoma (G602fs, adjacent to the same polyA tract).10

TP53 mutations in cancer. The x-axis indicates cohorts used for comparative genomic analysis. The y-axis depicts the frequency of TP53 mutations in the cohort. TP53 mutations are ubiquitous and are therefore not discriminatory between most individual cancer types.

Data from the cBioPortal Web site. Available at: www.cbioportal.org. Accessed June 11, 2014.
Copyright:
Copyright © 2015 by the National Comprehensive Cancer Network 2015
Page Numbers:
11
Article Category:
Research Article
Received:
03 Oct 2014
Accepted:
27 Apr 2015
Print Publication Date:
01 Jul 2015
Online Publication Date:
Jul 2015
All Time Past Year Past 30 Days
Abstract Views 0 0 0
Full Text Views 551 64 9
PDF Downloads 258 46 7
EPUB Downloads 0 0 0