Abstract
PURPOSE
Breast and ovarian tumors in germline BRCA1/2 carriers undergo allele-specific loss of heterozygosity, resulting in homologous recombination deficiency (HRD) and sensitivity to poly-ADP-ribose polymerase (PARP) inhibitors. This study investigated whether biallelic loss and HRD also occur in primary nonbreast/ovarian tumors that arise in germline BRCA1/2 carriers.
METHODS
A clinically ascertained cohort of BRCA1/2 carriers with a primary nonbreast/ovarian cancer was identified, including canonical (prostate and pancreatic cancers) and noncanonical (all other) tumor types. Whole-exome sequencing or clinical sequencing results (n = 45) were analyzed. A pan-cancer analysis of nonbreast/ovarian primary tumors from germline BRCA1/2 carriers from The Cancer Genome Atlas (TCGA, n = 73) was used as a validation cohort.
RESULTS
Ages of nonbreast/ovarian cancer diagnosis in germline BRCA1/2 carriers were similar to controls for the majority of cancer types. Nine of 45 (20%) primary nonbreast/ovarian tumors from germline BRCA1/2 carriers had biallelic loss of BRCA1/2 in the clinical cohort, and 23 of 73 (32%) in the TCGA cohort. In the combined cohort, 35% and 27% of primary canonical and noncanonical BRCA tumor types, respectively, had biallelic loss. High HRD scores (HRDex > 42) were detected in 81% of tumors with biallelic BRCA loss compared with 22% (P < .001) of tumors without biallelic BRCA loss. No differences in genomic profile, including mutational signatures, mutation spectrum, tumor mutational burden, or microsatellite instability, were found in primary nonbreast/ovarian tumors with or without biallelic BRCA1/2 loss.
CONCLUSION
A proportion of noncanonical primary tumors have biallelic loss and evidence of HRD. Our data suggest that assessment of biallelic loss and HRD could supplement identification of germline BRCA1/2 mutations in selection of patients for platinum or PARP inhibitor therapy.
Only 20% of noncanonical primary tumors from germline BRCA1/2 carriers have homologous recombination deficiency.
INTRODUCTION
Heterozygous carriers of germline likely pathogenic/pathogenic variants (PGVs) in BRCA1/2 are predisposed to a high risk for breast and ovarian cancers.1,2 BRCA1/2 PGVs also confer increased risks to develop pancreatic and prostate cancer.2 However, a wide variety of other tumor types have been reported in germline BRCA1/2 carriers.1,3-6 In patients with a heterozygous BRCA1/2 PGV, loss of the wild-type allele is predicted to occur in a process called allele-specific loss of heterozygosity (LOH), leading to complete loss of BRCA1/2 function and subsequent tumor formation.7 BRCA1 and BRCA2 are tumor suppressor genes necessary for homologous recombination (HR), an error-free DNA repair process that uses the undamaged sister chromatid as a template to repair DNA double-strand breaks (DSBs) accumulating during replication.8 Thus, tumors with biallelic loss of BRCA1/2 are predicted to have HR deficiency (HRD). In the event of DNA DSBs in BRCA1/2-deficient tumors, HR is inactive and replaced by error-prone DNA repair pathways, such as base excision repair or nonhomologous end joining, resulting in genomic instability and subsequent cancer development.9
CONTEXT
Key Objective
In this study, we analyzed clinical characteristics and whole-exome sequencing of tumor-normal pairs from patients with primary nonbreast/ovarian tumors and germline BRCA1/2 mutations from two cohorts.
Knowledge Generated
Nonbreast/ovarian tumors occur at similar ages in germline BRCA1/2 carriers as mutation-negative controls. A minority (20%-30%) of both canonical (prostate and pancreatic) and noncanonical (all other) primary tumors in germline BRCA1/2 carriers have biallelic loss and genomic indicators of homologous recombination deficiency (HRD).
Relevance
Identification of a germline BRCA1/2 mutation alone in primary tumors is insufficient to predict HRD and possible response to poly-ADP-ribose polymerase (PARP) inhibition. Biallelic loss status from paired tumor/normal sequencing along with genomic measures of HRD should be studied as biomarkers for PARP inhibitor response.
In the past 2 decades, therapies targeting defects in HRD9 have been developed. In particular, platinum-containing chemotherapies, including cisplatin or carboplatin, have been widely used to treat BRCA1/2-associated breast and ovarian cancers.9 Inhibitors of poly-ADP-ribose polymerase (PARPi) are US Food and Drug Administration (FDA)–approved for BRCA1/2-associated breast, ovarian, prostate, and pancreatic cancer.10 The FDA approvals for PARPi rely variably on germline or somatic mutational status alone,10 with the assumption that all tumors in individuals with BRCA1/2 PGVs undergo allele-specific LOH or another mechanism of biallelic loss. Analysis of metastatic tumors has shown that approximately 75% of tumors known to be associated with BRCA1/2 PGVs (canonical tumors) have biallelic loss compared with only 26% of metastatic noncanonical tumors.11
Genomic profiles, such as HRD scores or mutational signatures, may also be used to determine BRCAness12 to predict PARPi sensitivity. HRD score is calculated on the basis of DNA copy-number profiles via a combination of metrics, including genome-wide LOH, nontelomeric allelic imbalance (TAI), and large-scale state transition,13,14 and has shown clinical utility for predicting BRCA1/2 mutation status, prognosis, and treatment response.15 However, HRD scores are not routinely used in clinical settings to supplement the findings of germline or somatic BRCA1/2 mutational status.
As PARPis are currently being used or tested in the primary nonmetastatic setting, we determined the frequency of biallelic BRCA1/2 loss and HRD in primary nonmetastatic, nonbreast/ovarian tumors from carriers of BRCA1/2 PGVs. Understanding the underlying genomic drivers of these tumors furthers understanding of the mechanisms in which nonbreast/ovarian cancers arise in association with BRCA1/2 PGVs. Furthermore, these data are critical to determining appropriate biomarkers for HRD-directed therapies.
METHODS
Clinical Cohort Ascertainment
The BRCA1/2 clinical cohort was ascertained from the Basser Center for BRCA Cancer Risk Evaluation Program registry. Acquisition of the patient samples was approved by the Institutional Review Board of the University of Pennsylvania, and written informed consent was obtained from each participant for use of their samples and clinical data in genetic studies. Inclusion criteria included a clinically confirmed BRCA1/2 PGV and a primary cancer other than breast and ovarian cancer, excluding nonmelanoma skin cancers. Chart review was performed via the associated research database and the Penn Medicine electronic health record for clinical characteristics, tumor information, and treatment information.
The BRCA1/2-negative clinical cohort was previously described.16 Briefly, this cohort was created from the Penn Medicine Biobank,17,18 and patients were included if they had (1) undergone germline whole-exome sequencing (WES) and had no identified PGV in BRCA1/2 or any of 44 other autosomal dominant cancer susceptibility genes (Data Supplement); and (2) cancer clinical data present in the Penn Medicine Cancer Registry.
A validation cohort was created by identifying carriers of BRCA1/2 PGVs in the pan-cancer analysis of The Cancer Genome Atlas (TCGA) as reported,19 excluding breast and ovarian cancers (Data Supplement). Level 1 tumor and matched normal DNA binary alignment map (BAM) files were downloaded from the Genomic Data Commons using a National Center for Biotechnology Information Genotypes and Phenotypes Database–approved protocol and underwent quality control to ensure the BRCA1/2 PGV was found at the expected heterozygous frequency in the normal BAM file.
Sample Acquisition and Preparation of Clinical BRCA1/2 Tumor Sequencing Data
Of 358 nonbreast/nonovarian tumors in 322 patients with BRCA1/2 PGVs in the clinical registry, 71 had available archived blood DNA and identified formalin-fixed paraffin-embedded (FFPE) tumor blocks; blocks were available and of appropriate quality for 39 tumors. FFPE tumor blocks were sectioned and stained with hematoxylin and eosin to ensure sections of over 70% invasive tumor were used for DNA extraction. Selected tumor areas of slides were macrodissected and DNA prepared.20 All DNA samples were quantitated with a Qubit. Germline DNA from blood or saliva was extracted using standard protocols. Library preparation of tumor and matched germline DNA was as described.20 Libraries were subjected to WES using the SureSelect All-Exon Kit v7 (Agilent, Santa Clara, CA). Tumors were sequenced on an Illumina Hi-Seq 4000 (Illumina, San Diego, CA). If samples were not available for WES, clinical sequencing was reviewed where available (n = 6). Forty-five tumor samples had available sequencing data for analysis (Data Supplement).
Bioinformatics Analysis
Bioinformatics analysis was as previously described.20 Briefly, tumor and normal BAM files were aligned to the GRCh37 reference genome using Burrows-Wheeler Aligner for short-read alignment. Variants underwent initial quality control filtering according to Genome Analysis ToolKit (GATK) best practices. Somatic variant calling was performed using a consensus of GATK Mutect and VarScan2. Somatic variants were annotated with both Annovar and OncoKB21 to determine oncogenicity. Tumor purity and copy-number status were calculated for each sample using Sequenza v3.0.0.22 Variants were kept for further analysis if alternate allele depth >5 and alternate allele frequency >10%.
The presence of locus specific LOH in BRCA1/2 was determined for each sample using a combination of the somatic variant allele frequency of the mutation as determined by Varscan2, tumor purity as determined by Sequenza, and copy number at the genomic locus as determined by Sequenza. To correct for tumor purity, a minimum expected LOH variant allele frequency was calculated as a function of the tumor purity: E = P + (1 − P)/2, where E = expected LOH variant allele fraction of 50% and P = tumor purity. If the observed somatic variant allele frequency exceeded the expected LOH variant allele frequency and/or the B allele copy number was zero, the variant was classified as having allele-specific LOH. The presence of second somatic hits in BRCA1/2, depending on germline PGV, was investigated in the Mutect and Varscan2 somatic calls.
Tumor mutational burden (TMB) was calculated as the number of nonsynonymous somatic variants, as determined by Mutect2, per megabase of exome captured. Microsatellite instability (MSI) score was calculated using MSI Sensor.23 HRD was calculated using custom scripts to calculate the sum of TAI, large state transitions, and genomic LOH events using custom R scripts (HRDex).20,24 Mutational signatures were determined using deconstructSigs.25
RESULTS
We identified 322 carriers of a BRCA1/2 PGV with 358 primary nonbreast/ovarian cancers within a clinical high-risk research registry. The majority of the subjects were females (67%), Caucasian (95%), and alive at last follow-up (65%; Table 1). Of the 322, 143 (44%), 177 (55%), and two (0.6%) subjects had PGVs in BRCA1, BRCA2, or both BRCA1/2, respectively. Of 257 patients with follow-up data available, 68% did not develop local or metastatic recurrence (Table 1). The 322 carriers had 358 malignant tumors; pancreatic (13%), colorectal (12%), and thyroid cancers (11%) were the most common tumor types in both sexes (Data Supplement). Of 105 male carriers, 40% had prostate cancer, and 11% of 217 female carriers had endometrial cancer. Age of diagnosis for carriers with BRCA1/2 PGVs in the clinical cohort was compared with sequencing-confirmed nongenetic mutation carriers with the same tumor types from an institutional biobank. Median age of diagnosis was younger for germline BRCA1/2 carriers for lung cancer only, using a Bonferroni-corrected P = value cutoff (P < .001; Data Supplement).
TABLE 1.
Clinical Characteristics of BRCA1/2 Carriers With Nonbreast/Nonovarian Tumors
We analyzed 73 carriers of BRCA1/2 PGVs with a primary nonbreast/ovarian tumor in the TCGA.19 The majority (64%) had a BRCA2 mutation (Table 1) similar to the clinical cohort, but these subjects were more commonly male (59%; P < .0001). As with the clinical cohort, 62% did not develop recurrence (v 69%; P = .324), and 67% of patients were alive (v 65%; P = .786). The spectrum of tumor types differed in the unselected TCGA cohort compared with the clinical cohort. Of the 73 malignant tumors, lung squamous cell carcinoma (eight), cervical cancer (seven), and stomach cancers (six) were the most common (11%, 10%, and 8%, respectively; Data Supplement). Comparing carriers of BRCA1/2 PGVs with a tumor type with noncarriers of the same tumor type in TCGA, no significant differences were seen for age of diagnosis compared with noncarriers with the same tumor type (Data Supplement).
We next determined biallelic BRCA1/2 loss in 73 TCGA and 45 clinical cohort tumors. In the clinical cohort overall, nine of 45 (20%) tumors had biallelic loss of BRCA1/2 (Fig 1A, Table 2), including two prostate (n = 4) and one each of pancreatic (n = 7), lung adenocarcinoma (n = 4), mesothelioma (n = 2), colon (n = 2), endometrial (n = 2), renal cell carcinoma (n = 1), and thyroid cancer (n = 9; Data Supplement). Similarly, in the TCGA cohort, 23 of 73 (32%; P = .205) primary tumors had biallelic loss of BRCA1/2 (Fig 1A, Table 2), including four cervical (n = 7), four stomach (n = 6), three uterine (n = 3), two pancreatic (n = 5), two prostate (n = 4), two lung squamous (n = 8), two bladder (n = 5), two sarcomas (n = 3), one colorectal (n = 3), and one esophageal (n = 2; Data Supplement). Overall, 27% and 44% of BRCA-canonical tumors (prostate and pancreatic) had biallelic loss in the clinical and TCGA cohorts, respectively (Table 2), in comparison with 83% biallelic loss rate in our published BRCA1/2 primary breast and ovarian tumors.20 For noncanonical BRCA tumors, 18% and 30% had biallelic BRCA1/2 loss in the clinical and TCGA cohorts, respectively (Table 2). Notably, certain tumor types had no tumors with biallelic BRCA1/2 loss, including melanoma, glioblastoma, and pheochromocytoma (Fig 1A).
FIG 1.
Genomic profiles in germline BRCA1/2 primary nonbreast/ovarian tumors. (A) Whole-exome and targeted sequencing data were analyzed for 118 primary nonbreast/ovarian tumors from carriers of a BRCA1/2 PGV, including canonical BRCA tumors (prostate cancer [PRAD] and pancreatic cancer [PAAD]) and noncanonical BRCA tumors (endometrial cancer [UCEC], melanoma [SKCM], thyroid [THCA], lung squamous cell carcinoma [LUSC], lung adenocarcinoma [LUAD], bladder cancer [BLCA], cervical cancer [CESC], stomach cancer [STAD], colon cancer [COAD], glioblastoma [GBM], pheochromocytoma [PCPG], sarcoma [SARC], head and neck squamous cell carcinoma [HNSC], papillary renal cell carcinoma [KIRP], and hepatocellular carcinoma [LIHC]), and miscellaneous tumor types with only one or two tumors analyzed—in order from top down: esophageal (n = 2), mesothelioma (n = 2), testicular germ cell tumor (n = 2), diffuse large B-cell lymphoma (n = 2), cholangiocarcinoma, laryngeal cancer, leukemia, anal cancer, mature B-cell lymphoma, gastrointestinal stromal tumor, and duodenal cancer. Biallelic BRCA1/2 loss was defined as evidence of loss of heterozygosity or a second somatic hit in the corresponding gene. TP53+ indicates at least one likely pathogenic or pathogenic somatic mutation in TP53. An oncogenic driver was defined as the presence of at least one likely pathogenic or pathogenic somatic mutation in a gene known to cause oncogenesis in that tumor type. (B) HRD scores calculated by HRDex for tumors with biallelic loss of BRCA1/2 (LOHpos) and tumors without biallelic loss (LOHneg). (C) TMB scores for tumors with biallelic loss of BRCA1/2 (LOHpos) and tumors without biallelic loss (LOHneg). (D) MSI scores calculated by MSIsensor for tumors with biallelic loss of BRCA1/2 (LOHpos) and tumors without biallelic loss (LOHneg). HRD, homologous recombination deficiency; MSI, microsatellite instability; PGV, pathogenic/pathogenic variant; TMB, tumor mutational burden.
TABLE 2.
Fraction of Germline BRCA1/2 Tumors With Biallelic BRCA1/2 Loss
Comparing published data from metastatic germline BRCA1/2 tumors and those that later developed metastatic disease11 with our primary tumor analysis, we observed a significant enrichment of biallelic loss in canonical BRCA metastatic tumors compared with primary tumors (74% v 29%; P = .0002; Table 2). No enrichment of biallelic BRCA1/2 loss was seen in BRCA noncanonical tumor types (26% v 27%; P = 1.000).
We next analyzed HRD using a composite HRD score (HRDex). Overall, the HRD score was significantly higher in tumors with biallelic BRCA1/2 loss (55 v 29; P < .001; Fig 1B), and the fraction of tumors with a high HRDex score (>42) was higher in tumors with biallelic BRCA1/2 loss (81% v 22%; P < .0001; Table 3). Of the six tumors with BRCA1/2 biallelic loss but an HRDex score <40 (pancreatic, n = 2; prostate, n = 2; colon, n = 1; cervical, n = 1), all were from germline BRCA2 carriers. The 18 tumors that did not have BRCA1/2 biallelic loss but had HRDex > 40 were equally distributed among BRCA1 and BRCA2 carriers, and were mostly platinum-sensitive tumor types, including lung squamous cell carcinoma (n = 5), lung adenocarcinomas (n = 4), bladder cancers (n = 4), cervical cancers (n = 2), and stomach cancers (n = 2).
TABLE 3.
Genomic Profiles and Mutational Status in Germline BRCA1/2 Tumors With and Without Biallelic Loss
To determine if BRCA1/2 germline atypical tumors shared other genomic characteristics associated with biallelic loss status, we analyzed mutational signatures, mutational profiles, TMB, and MSI. Combining both cohorts of tumors, despite different spectrum of malignancies, 21 of 32 (66%) germline BRCA1/2 tumors with biallelic BRCA1/2 loss had TP53 mutations compared with 18 of 86 (21%) without biallelic BRCA1/2 loss (P < .0001; Table 3, Data Supplement). Neither TMB nor MSISensor score was significantly different in germline BRCA1/2 tumors with versus without biallelic loss (Figs 1C and 1D). The fraction of tumors with a lineage-specific oncogenic driver and the mutational signature profiles, including the BRCA-associated signature 3, were also similar in germline BRCA1/2 tumors with or without biallelic loss (Table 3, Data Supplement). For individual cancers with at least five patients, neither age of diagnosis nor rate of disease recurrence was associated with biallelic loss status (Data Supplement). When cancers associated with older age of onset (bladder, colorectal, endometrial, lung, pancreatic, prostate, and stomach) were combined, the presence of biallelic loss was associated with slightly younger age of diagnosis (60% v 66%; P = .04) although similar rates of recurrence (62% v 48%; P = .27).
DISCUSSION
This study analyzed primary nonbreast/ovarian tumors in carriers of BRCA1/2 PGVs for markers of HRD, the presence of which would suggest response to platinum chemotherapy and PARPi. We find most nonbreast/ovarian tumors in germline BRCA1/2 carriers are not associated with biallelic loss of BRCA and did not have HRD. For those tumors with high HRD, this biomarker is associated with biallelic loss of BRCA1 or BRCA2. Often nonbreast/ovarian tumors in germline BRCA1/2 carriers had somatic mutational profiles similar to non-BRCA tumors of the same lineage. These results have significant implications for the expected response of nonbreast/ovarian tumors in germline BRCA1/2 carriers to HRD-directed therapies such as platinum chemotherapy and PARPi.
Comparing primary with metastatic tumors, the rate of biallelic BRCA1/2 loss was similarly high between primary and metastatic tumors in breast and ovarian cancer and similarly low between primary and metastatic tumors in BRCA noncanonical tumors, similar to a previous institutional study.11 In our study, however, we show a lower rate of biallelic loss of BRCA1/2 in primary prostate and pancreatic tumors, compared with metastatic tumors.11,26 These data suggest that biallelic loss of BRCA1/2 may be a driving factor of the development of metastatic disease in prostate and pancreatic tumors that develop in germline BRCA1/2 carriers, as has been suggested to occur in a recent study of matched primary-recurrent breast and ovarian cancers.27
Tissue-specific oncogenic driver pathway alterations were consistently prevalent among tumors with or without biallelic loss. For example, most melanomas and thyroid cancers did not have biallelic BRCA1/2 loss but had canonical BRAF mutations. These data suggest that noncanonical tumors in carriers of BRCA1/2 PGVs likely develop because of typical tissue-specific processes involved in tumor development versus being driven by loss of BRCA1/2 function. It is possible, in addition, that noncanonical tissues do not gain a survival advantage by loss of BRCA1/2 function, explaining the lack of enrichment of biallelic loss in noncanonical metastatic tumors compared with primaries.
A study from Foundation Medicine using a mixed primary/metastatic cohort, tumor-only sequencing, and inferred germline status similarly showed a high proportion (over 70%) biallelic loss of BRCA1/2 in breast, ovarian, prostate, pancreatic, and hepatobiliary cancers, whereas BRCA1/2 noncanonical cancer types had significantly lower biallelic mutation rates.28,29 High genomic LOH scores in this study were associated with an inferred germline mutation and presence of biallelic loss. These studies together suggest that genomic measures of HRD may be a better biomarker for PARPi sensitivity than germline BRCA1/2 status alone in patients with noncanonical tumors.30 Indeed, a recent pan-cancer tumor-agnostic clinical trial of talazoparib and avelumab in advanced tumors selected only by the presence of a germline or somatic BRCA1/2 alteration showed a 32% overall response rate (ORR) in BRCA-associated tumors and only an 8% ORR in non–BRCA-associated tumors.31
Our study has limitations, namely the small number of each noncanonical tumor type, and by lumping tumor types, we may be missing some tumor-type–specific differences in rates of allele-specific LOH and HRD, for example, in endometrial and gastric cancer, where the majority of tumors did have biallelic loss in BRCA1/2 and HRD. In addition, the HRD score cutoffs using HRDex that predict platinum and PARPi responsiveness in noncanonical tumor types is unknown and likely to differ by tissue type.32 Therefore, some low-HRD tumors in this cohort, particularly those with biallelic BRCA1/2 loss, may still respond to targeted therapy; conversely some high-HRD tumors, perhaps those without biallelic BRCA1/2 loss may not respond. In addition, we used WES compared with whole-genome sequencing, which limits analysis of mutational signatures. Promoter methylation as a mechanism of biallelic loss for BRCA1 is theoretically possible, although previous work demonstrated BRCA1 promoter hypermethylation was not a cause of biallelic loss in pancreatic cancer.33 In addition, we have shown that the presence of BRCA1 promoter methylation in the absence of allele-specific LOH was associated with retained BRCA1 protein.20 Finally, we were unable to correlate genomic profiles with treatment response.
In conclusion, our study found a low rate of biallelic loss of BRCA1/2 PGVs in noncanonical BRCA tumor types. In addition, we found a lower rate of biallelic loss of BRCA1/2 in primary BRCA prostate and pancreatic tumors than has been previously reported in metastatic tumors. Our finding of a low rate of biallelic BRCA1/2 loss in nonbreast/ovarian cancers has significant clinical implications for PARPi clinical trials. These findings will be particularly relevant in the adjuvant setting as biallelic loss is lower in primary versus metastatic tumors. Our data suggest that additional measurements of BRCA function (ie, HRD or biallelic loss) will be crucial to supplement germline sequencing, especially in nonbreast/ovarian cancers, to identify patients most likely to respond to HRD-directed therapy.
ACKNOWLEDGMENT
The authors acknowledge the technical contributions of Liza Dorfman, who passed away in February 2019. See author contributions in Appendix 1.
APPENDIX 1. AUTHOR LIST
Penn Medicine BioBank Banner Author List and Contribution Statements
Penn Medicine BioBank Leadership Team
Daniel J. Rader, MD; Marylyn D. Ritchie, PhD; Michael D. Feldman, MD
Contribution: All authors contributed to securing funding, study design, and oversight. All authors reviewed the final version of the manuscript.
Patient Recruitment and Regulatory Oversight
JoEllen Weaver; Afiya Poindexter; Ashlei Brock; Khadijah Hu-Sain; Yi-An Ko
Contributions: J.W. managed patient recruitment and regulatory oversight of study. A.P., A.B., K.H.-S., and Y.-A.K. managed recruitment and enrollment of study participants.
Laboratory Operations
JoEllen Weaver; Meghan Livingstone; Fred Vadivieso; Ashley Kloter; Stephanie DerOhannessian; Teo Tran; Linda Morrel; Ned Haubein; Joseph Dunn
Contribution: J.W., M.L., F.V., and S.D. are responsible for oversight of laboratory operations. M.L., F.V., A.K., S.D., T.T., and L.M. performed sample processing. N.H. and J.D. are responsible for sample tracking and the laboratory information management system.
Clinical Informatics
Anurag Verma, PhD; Colleen Morse, MS; Marjorie Risman, MS; Renae Judy, BS
Contribution: all authors contributed to the development and validation of clinical phenotypes used to identify study subjects and (when applicable) controls.
Genome Informatics
Anurag Verma, PhD; Shefali S. Verma, PhD; Yuki Bradford, MS; Scott Dudek, MS; Theodore Drivas, MD, PhD
Contribution: A.V. and S.S.V. are responsible for the analysis, design, and infrastructure needed quality control genotype and exome data. Y.B. performed the analysis. T.D. and A.V. provided variant and gene annotations and their functional interpretation of variants.
Regeneron Genetics Center Banner Author List and Contribution Statements
RGC Management and Leadership Team
Goncalo Abecasis, PhD; Aris Baras, MD; Michael Cantor, MD; Giovanni Coppola, MD; Andrew Deubler; Aris Economides, PhD; Luca A. Lotta, MD, PhD; John D. Overton, PhD; Jeffrey G. Reid, PhD; Katherine Siminovitch, MD; Alan Shuldiner, MD
Sequencing and Laboratory Operations
Christina Beechert; Caitlin Forsythe, MS; Erin D. Fuller; Zhenhua Gu, MS; Michael Lattari; Alexander Lopez, MS; John D. Overton, PhD; Maria Sotiropoulos Padilla, MS; Manasi Pradhan, MS; Kia Manoochehri, BS; Thomas D. Schleicher, MS; Louis Widom; Sarah E. Wolf, MS; Ricardo H. Ulloa, BS
Clinical Informatics
Amelia Averitt, PhD; Nilanjana Banerjee, PhD; Michael Cantor, MD; Dadong Li, PhD; Sameer Malhotra, MD; Deepika Sharma, MHI; Jeffrey C. Staples, PhD
Genome Informatics
Xiaodong Bai, PhD; Suganthi Balasubramanian, PhD; Suying Bao, PhD; Boris Boutkov, PhD; Siying Chen, PhD; Gisu Eom, BS; Lukas Habegger, PhD; Alicia Hawes, BS; Shareef Khalid; Olga Krasheninina, MS; Rouel Lanche, BS; Adam J. Mansfield, BA; Evan K. Maxwell, PhD; George Mitra, BA; Mona Nafde, MS; Sean O'Keeffe, PhD; Max Orelus, BBA; Razvan Panea, PhD; Tommy Polanco, BA; Ayesha Rasool, MS; Jeffrey G. Reid, PhD; William Salerno, PhD; Jeffrey C. Staples, PhD; Kathie Sun, PhD
Analytical Genomics and Data Science
Goncalo Abecasis, DPhil; Joshua Backman, PhD; Amy Damask, PhD; Lee Dobbyn, PhD; Manuel Allen Revez Ferreira, PhD; Arkopravo Ghosh, MS; Christopher Gillies, PhD; Lauren Gurski, BS; Eric Jorgenson, PhD; Hyun Min Kang, PhD; Michael Kessler, PhD; Jack Kosmicki, PhD; Alexander Li, PhD; Nan Lin, PhD; Daren Liu, MS; Adam Locke, PhD; Jonathan Marchini, PhD; Anthony Marcketta, MS; Joelle Mbatchou, PhD; Arden Moscati, PhD; Charles Paulding, PhD; Carlo Sidore, PhD; Eli Stahl, PhD; Kyoko Watanabe, PhD; Bin Ye, PhD; Blair Zhang, PhD; Andrey Ziyatdinov, PhD
Therapeutic Area Genetics
Ariane Ayer, BS; Aysegul Guvenek, PhD; George Hindy, PhD; Giovanni Coppola, MD; Jan Freudenberg, MD; Jonas Bovijn, MD; Katherine Siminovitch, MD; Kavita Praveen, PhD; Luca A. Lotta, MD, PhD; Manav Kapoor, PhD; Mary Haas, PhD; Moeen Riaz, PhD; Niek Verweij, PhD; Olukayode Sosina, PhD; Parsa Akbari, PhD; Priyanka Nakka, PhD; Sahar Gelfman, PhD; Sujit Gokhale, BE; Tanima De, PhD; Veera Rajagopal, PhD; Alan Shuldiner, MD; Bin Ye, PhD; Gannie Tzoneva, PhD; Juan Rodriguez-Flores, PhD
Research Program Management and Strategic Initiatives
Esteban Chen, MS; Marcus B. Jones, PhD; Michelle G. LeBlanc, PhD; Jason Mighty, PhD; Lyndon J. Mitnaul, PhD; Nirupama Nishtala, PhD; Nadia Rana, PhD; Jaimee Hernandez
Emanuel Barrett
Employment: HCA Healthcare
Speakers' Bureau: HCA Healthcare
Expert Testimony: HCA Healthcare
Travel, Accommodations, Expenses: HCA Healthcare
Other Relationship: HCA Healthcare
Abigail Doucette
Employment: Children's Hospital of Philadelphia
Kim A. Reiss
Honoraria: MJH Life Sciences
Consulting or Advisory Role: AstraZeneca, Carisma Therapeutics, Bristol Myers Squibb, Foundation Medicine
Research Funding: Lilly (Inst), Clovis Oncology, Bristol Myers Squibb (Inst), Tesaro (Inst), GlaxoSmithKline (Inst)
Michael Feldman
Research Funding: Scopio Inc (Inst)
Susan M. Domchek
Honoraria: AstraZeneca, GlaxoSmithKline
Research Funding: AstraZeneca (Inst), Clovis Oncology (Inst)
Open Payments Link: https://openpaymentsdata.cms.gov/physician/917904
Katherine L. Nathanson
Consulting or Advisory Role: Merck
No other potential conflicts of interest were reported.
SUPPORT
Supported by the National Cancer Institute (K08CA215312, K.N.M.), the Burroughs Wellcome Foundation (#1017184, K.N.M.), Basser Center for BRCA (K.N.M., K.A.R., S.M.D., K.L.N.), V Foundation for Cancer Research (K.L.N.), Gray Foundation (S.M.D., K.L.N.) the Konner Family Foundation (K.N.M.), and Breast Cancer Research Foundation (S.M.D., K.L.N.). The authors acknowledge the Penn Medicine BioBank (PMBB) for providing data and thank the patient-participants of Penn Medicine who consented to participate in this research program. The authors would also like to thank the PMBB team and Regeneron Genetics Center for providing genetic variant data for analysis. The PMBB is approved under IRB protocol# 813913 and supported by Perelman School of Medicine at University of Pennsylvania, a gift from the Smilow family, and the National Center for Advancing Translational Sciences of the National Institutes of Health under CTSA award number UL1TR001878.
D.W. and A.N.L. contributed equally to this work.
Contributor Information
Collaborators: Adam J. Mansfield, Adam Locke, Afiya Poindexter, Alan Shuldiner, Alexander Li, Alexander Lopez, Alicia Hawes, Amelia Averitt, Amy Damask, Andrew Deubler, Andrey Ziyatdinov, Anthony Marcketta, Anurag Verma, Arden Moscati, Ariane Ayer, Aris Baras, Aris Economides, Arkopravo Ghosh, Ashlei Brock, Ashley Kloter, Ayesha Rasool, Aysegul Guvenek, Bin Ye, Blair Zhang, Boris Boutkov, Caitlin Forsythe, Carlo Sidore, Charles Paulding, Christina Beechert, Christopher Gillies, Colleen Morse, Dadong Li, Daniel J. Rader, Daren Liu, Deepika Sharma, Eli Stahl, Eric Jorgenson, Erin D. Fuller, Esteban Chen, Evan K. Maxwell, Fred Vadivieso, Gannie Tzoneva, George Hindy, George Mitra, Giovanni Coppola, Gisu Eom, Goncalo Abecasis, Hyun Min Kang, Jack Kosmicki, Jaimee Hernandez, Jan Freudenberg, Jason Mighty, Jeffrey C. Staples, Jeffrey G. Reid, Joelle Mbatchou, JoEllen Weaver, John D. Overton, Jonas Bovijn, Jonathan Marchini, Joseph Dunn, Joshua Backman, Juan Rodriguez-Flores, Katherine Siminovitch, Kathie Sun, Kavita Praveen, Khadijah Hu-Sain, Kia Manoochehri, Kyoko Watanabe, Lauren Gurski, Lee Dobbyn, Linda Morrel, Louis Widom, Luca A. Lotta, Lukas Habegger, Lyndon J. Mitnaul, Manasi Pradhan, Manav Kapoor, Manuel Allen Revez Ferreira, Marcus B. Jones, Maria Sotiropoulos Padilla, Marjorie Risman, Mary Haas, Marylyn D. Ritchie, Max Orelus, Meghan Livingstone, Michael Cantor, Michael D. Feldman, Michael Kessler, Michael Lattari, Michelle G. LeBlanc, Moeen Riaz, Mona Nafde, Nadia Rana, Nan Lin, Ned Haubein, Niek Verweij, Nilanjana Banerjee, Nirupama Nishtala, Olga Krasheninina, Olukayode Sosina, Parsa Akbari, Priyanka Nakka, Razvan Panea, Renae Judy, Ricardo H. Ulloa, Rouel Lanche, Sahar Gelfman, Sameer Malhotra, Sarah E. Wolf, Scott Dudek, Sean O'Keeffe, Shareef Khalid, Shefali S. Verma, Siying Chen, Stephanie DerOhannessian, Suganthi Balasubramanian, Sujit Gokhale, Suying Bao, Tanima De, Teo Tran, Theodore Drivas, Thomas D. Schleicher, Tommy Polanco, Veera Rajagopal, William Salerno, Xiaodong Bai, Yi-An Ko, Yuki Bradford, and Zhenhua Gu
DATA SHARING STATEMENT
The raw WES sequencing data generated as part of this study have been deposited in the NCBI SRA database under accession code phs003348.v1. The raw germline WES data are protected because of lack of patient consent to deposit in a public repository. The germline WES will be made available upon request from the corresponding author and will be made available under a Data Transfer Agreement (DTA) and transferred via FTP when the DTA is complete. Most data transfers will be completed within a month's time. TCGA data are available under controlled-use conditions; data use limitations and the instructions for applying for access are available through dbGaP. The raw TGCA data are accessible via the Genomic Data Commons from the National Center Institute once access is made available through dbGAP application. All data needed to evaluate the conclusions in the paper are present in the paper and/or the Data Supplement. Source Data are provided with this paper.
Code for calculating HRD and aneuploidy scores is available at https://github.com/maxwell-lab/HRDex.
AUTHOR CONTRIBUTIONS
Conception and design: Emanuel Barrett, Anupma Nayak, Susan M. Domchek, Katherine L. Nathanson, Kara N. Maxwell
Financial support: Katherine L. Nathanson, Kara N. Maxwell
Provision of study materials or patients: Anupma Nayak, Katherine L. Nathanson, Kara N. Maxwell
Collection and assembly of data: Dylane Wineland, Anh N. Le, Gregory Kelly, Paul Bastian, Heather Symecko, Kurt D'Andrea, Abigail Doucette, Peter Gabriel, Anupma Nayak, Susan M. Domchek, Katherine L. Nathanson, Kara N. Maxwell
Data analysis and interpretation: Dylane Wineland, Anh N. Le, Ryan Hausler, Heena Desai, Bradley Wubbenhorst, John Pluta, Kim A. Reiss, Anupma Nayak, Michael Feldman, Susan M. Domchek, Katherine L. Nathanson, Kara N. Maxwell
Manuscript writing: All authors
Final approval of manuscript: All authors
Accountable for all aspects of the work: All authors
AUTHORS' DISCLOSURES OF POTENTIAL CONFLICTS OF INTEREST
The following represents disclosure information provided by authors of this manuscript. All relationships are considered compensated unless otherwise noted. Relationships are self-held unless noted. I = Immediate Family Member, Inst = My Institution. Relationships may not relate to the subject matter of this manuscript. For more information about ASCO's conflict of interest policy, please refer to www.asco.org/rwc or ascopubs.org/po/author-center.
Open Payments is a public database containing information reported by companies about payments made to US-licensed physicians (Open Payments).
Emanuel Barrett
Employment: HCA Healthcare
Speakers' Bureau: HCA Healthcare
Expert Testimony: HCA Healthcare
Travel, Accommodations, Expenses: HCA Healthcare
Other Relationship: HCA Healthcare
Abigail Doucette
Employment: Children's Hospital of Philadelphia
Kim A. Reiss
Honoraria: MJH Life Sciences
Consulting or Advisory Role: AstraZeneca, Carisma Therapeutics, Bristol Myers Squibb, Foundation Medicine
Research Funding: Lilly (Inst), Clovis Oncology, Bristol Myers Squibb (Inst), Tesaro (Inst), GlaxoSmithKline (Inst)
Michael Feldman
Research Funding: Scopio Inc (Inst)
Susan M. Domchek
Honoraria: AstraZeneca, GlaxoSmithKline
Research Funding: AstraZeneca (Inst), Clovis Oncology (Inst)
Open Payments Link: https://openpaymentsdata.cms.gov/physician/917904
Katherine L. Nathanson
Consulting or Advisory Role: Merck
No other potential conflicts of interest were reported.
REFERENCES
- 1. Wu H, Wu X, Liang Z. Impact of germline and somatic BRCA1/2 mutations: Tumor spectrum and detection platforms. Gene Ther. 2017;24:601–609. doi: 10.1038/gt.2017.73. [DOI] [PubMed] [Google Scholar]
- 2. Levy-Lahad E, Friedman E. Cancer risks among BRCA1 and BRCA2 mutation carriers. Br J Cancer. 2007;96:11–15. doi: 10.1038/sj.bjc.6603535. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 3. Li S, Silvestri V, Leslie G, et al. Cancer risks associated with BRCA1 and BRCA2 pathogenic variants. J Clin Oncol. 2022;40:1529–1541. doi: 10.1200/JCO.21.02112. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4. Buckley KH, Niccum BA, Maxwell KN, et al. Gastric cancer risk and pathogenesis in BRCA1 and BRCA2 carriers. Cancers (Basel) 2022;14((23)):5953. doi: 10.3390/cancers14235953. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Grinshpun A, Halpern N, Granit RZ, et al. Phenotypic characteristics of colorectal cancer in BRCA1/2 mutation carriers. Eur J Hum Genet. 2018;26:382–386. doi: 10.1038/s41431-017-0067-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6. Sanchis-Borja M, Fallet V, Fabre E, et al. Characterization of lung cancers in patients with BRCA germline variants: A multicenter series. Lung Cancer. 2022;173:67–70. doi: 10.1016/j.lungcan.2022.09.002. [DOI] [PubMed] [Google Scholar]
- 7. Venkitaraman AR. How do mutations affecting the breast cancer genes BRCA1 and BRCA2 cause cancer susceptibility? DNA Repair (Amst) 2019;81:102668. doi: 10.1016/j.dnarep.2019.102668. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Powell SN, Kachnic LA. Roles of BRCA1 and BRCA2 in homologous recombination, DNA replication fidelity and the cellular response to ionizing radiation. Oncogene. 2003;22:5784–5791. doi: 10.1038/sj.onc.1206678. [DOI] [PubMed] [Google Scholar]
- 9. Lord CJ, Ashworth A. PARP inhibitors: Synthetic lethality in the clinic. Science. 2017;355:1152–1158. doi: 10.1126/science.aam7344. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Zhou P, Wang J, Mishail D, et al. Recent advancements in PARP inhibitors-based targeted cancer therapy. Precis Clin Med. 2020;3:187–201. doi: 10.1093/pcmedi/pbaa030. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Jonsson P, Bandlamudi C, Cheng ML, et al. Tumour lineage shapes BRCA-mediated phenotypes. Nature. 2019;571:576–579. doi: 10.1038/s41586-019-1382-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12. Lord CJ, Ashworth A. BRCAness revisited. Nat Rev Cancer. 2016;16:110. doi: 10.1038/nrc.2015.21. [DOI] [PubMed] [Google Scholar]
- 13. Telli ML, Timms KM, Reid J, et al. Homologous recombination deficiency (HRD) score predicts response to platinum-containing neoadjuvant chemotherapy in patients with triple-negative breast cancer. Clin Cancer Res. 2016;22:3764–3773. doi: 10.1158/1078-0432.CCR-15-2477. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Timms KM, Abkevich V, Hughes E, et al. Association of BRCA1/2defects with genomic scores predictive of DNA damage repair deficiency among breast cancer subtypes. Breast Cancer Res. 2014;16:475. doi: 10.1186/s13058-014-0475-x. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15. de Luca XM, Newell F, Kazakoff SH, et al. Using whole-genome sequencing data to derive the homologous recombination deficiency scores. NPJ Breast Cancer. 2020;6:33. doi: 10.1038/s41523-020-0172-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16. Lee DJ, Hausler R, Le AN, et al. Association of inherited mutations in DNA repair genes with localized prostate cancer. Eur Urol. 2022;81:559–567. doi: 10.1016/j.eururo.2021.09.029. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17. Park J, Levin MG, Haggerty CM, et al. A genome-first approach to aggregating rare genetic variants in LMNA for association with electronic health record phenotypes. Genet Med. 2020;22:102–111. doi: 10.1038/s41436-019-0625-8. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18. Verma A, Damrauer SM, Naseer N, et al. The Penn Medicine BioBank: Towards a genomics-enabled learning healthcare system to accelerate precision medicine in a diverse population. J Pers Med. 2022;12((12)):197. doi: 10.3390/jpm12121974. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Huang KL, Mashl RJ, Wu Y, et al. Pathogenic germline variants in 10,389 adult cancers. Cell. 2018;173:355–370.e14. doi: 10.1016/j.cell.2018.03.039. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 20. Maxwell KN, Wubbenhorst B, Wenz BM, et al. BRCA locus-specific loss of heterozygosity in germline BRCA1 and BRCA2 carriers. Nat Commun. 2017;8:1–11. doi: 10.1038/s41467-017-00388-9. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21. Chakravarty D, Gao J, Phillips SM, et al. OncoKB: A precision oncology knowledge base. JCO Precis Oncol. 2017 doi: 10.1200/PO.17.00011. 10.1200/PO.17.00011 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22. Favero F, Joshi T, Marquard AM, et al. Sequenza: Allele-specific copy number and mutation profiles from tumor sequencing data. Ann Oncol. 2015;26:64–70. doi: 10.1093/annonc/mdu479. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Niu B, Ye K, Zhang Q, et al. MSIsensor: Microsatellite instability detection using paired tumor-normal sequence data. Bioinformatics. 2014;30:1015–1016. doi: 10.1093/bioinformatics/btt755. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24. Pluta J, Hausler R, Wubbenhorst B, et al. HRDex: A tool for deriving homologous recombination deficiency (HRD) scores from whole exome sequencing data. bioRxiv. 10.1101/2022.09.08.506670 [Google Scholar]
- 25. Rosenthal R, McGranahan N, Herrero J, et al. DeconstructSigs: Delineating mutational processes in single tumors distinguishes DNA repair deficiencies and patterns of carcinoma evolution. Genome Biol. 2016;17:31. doi: 10.1186/s13059-016-0893-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 26.Golan T, O'Kane GM, Denroche RE, et al. Genomic features and classification of homologous recombination deficient pancreatic ductal adenocarcinoma Gastroenterology 1602119–2132.e9., 2021 [DOI] [PubMed] [Google Scholar]
- 27. Shah JB, Pueschl D, Wubbenhorst B, et al. Analysis of matched primary and recurrent BRCA1/2 mutation-associated tumors identifies recurrence-specific drivers. Nat Commun. 2022;13:6728. doi: 10.1038/s41467-022-34523-y. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28. Sokol ES, Pavlick D, Khiabanian H, et al. Pan-cancer analysis of BRCA1 and BRCA2 genomic alterations and their association with genomic instability as measured by genome-wide loss of heterozygosity. JCO Precis Oncol. 2020 doi: 10.1200/PO.19.00345. 10.1200/PO.19.00345 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Westphalen CB, Fine AD, Andre F, et al. Pan-cancer analysis of homologous recombination repair-associated gene alterations and genome-wide loss-of-heterozygosity score. Clin Cancer Res. 2022;28:1412–1421. doi: 10.1158/1078-0432.CCR-21-2096. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 30. Nguyen L, Martens WMM, Van Hoeck A, et al. Pan-cancer landscape of homologous recombination deficiency. Nat Commun. 2020;11:5584. doi: 10.1038/s41467-020-19406-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 31. Schram AM, Colombo N, Arrowsmith E, et al. Avelumab plus talazoparib in patients with BRCA1/2- or ATM-altered advanced solid tumors: Results from JAVELIN BRCA/ATM, an open-label, multicenter, phase 2b, tumor-agnostic trial. JAMA Oncol. 2022;9((1)):29–39. doi: 10.1001/jamaoncol.2022.5218. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Stewart MD, Merino Vega D, Arend RC, et al. Homologous recombination deficiency: Concepts, definitions, and assays. Oncologist. 2022;27:167–174. doi: 10.1093/oncolo/oyab053. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 33. Zheng-Lin B, Rainone M, Varghese AM, et al. Methylation analyses reveal promoter hypermethylation as a rare cause of "second hit" in germline BRCA1-associated pancreatic ductal adenocarcinoma. Mol Diagn Ther. 2022;26:645–653. doi: 10.1007/s40291-022-00614-1. [DOI] [PMC free article] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
The raw WES sequencing data generated as part of this study have been deposited in the NCBI SRA database under accession code phs003348.v1. The raw germline WES data are protected because of lack of patient consent to deposit in a public repository. The germline WES will be made available upon request from the corresponding author and will be made available under a Data Transfer Agreement (DTA) and transferred via FTP when the DTA is complete. Most data transfers will be completed within a month's time. TCGA data are available under controlled-use conditions; data use limitations and the instructions for applying for access are available through dbGaP. The raw TGCA data are accessible via the Genomic Data Commons from the National Center Institute once access is made available through dbGAP application. All data needed to evaluate the conclusions in the paper are present in the paper and/or the Data Supplement. Source Data are provided with this paper.
Code for calculating HRD and aneuploidy scores is available at https://github.com/maxwell-lab/HRDex.