Functional Ex Vivo Assay Reveals Homologous Recombination Deﬁciency in Breast Cancer Beyond BRCA Gene Defects Titia G

Published OnlineFirst August 23, 2018; DOI: 10.1158/1078-0432.CCR-18-0063

Personalized Medicine and Imaging Clinical Cancer Research Functional Ex Vivo Assay Reveals Homologous Recombination Deﬁciency in Breast Cancer Beyond BRCA Gene Defects Titia G. Meijer1, Nicole S.Verkaik1, Anieta M. Sieuwerts2, Job van Riet3,4, Kishan A.T. Naipal1, Carolien H.M. van Deurzen5, Michael A. den Bakker6, Hein F.B.M. Sleddens5, Hendrikus-Jan Dubbink5, T. Dorine den Toom5, Winand N.M. Dinjens5, Esther Lips7, Petra M. Nederlof7, Marcel Smid2, Harmen J. G. van de Werken3,4, Roland Kanaar1, John W. M. Martens2, Agnes Jager2, and Dik C. van Gent1

Abstract

Purpose: Tumors of germline BRCA1/2 mutated carriers Results: RECAP was completed successfully in 148 of 170 show homologous recombination (HR) deficiency (HRD), samples (87%). Twenty-four tumors showed HRD (16%), resulting in impaired DNA double-strand break (DSB) repair whereas six tumors were HR intermediate (HRi; 4%). HRD and high sensitivity to PARP inhibitors. Although this therapy was explained by BRCA deficiencies (mutations, promoter is expected to be effective beyond germline BRCA1/2 mutated hypermethylation, deletions) in 16 cases, whereas seven HRD carriers, a robust validated test to detect HRD tumors is lacking. tumors were non-BRCA related. HRD tumors showed an In this study, we therefore evaluated a functional HR assay increased incidence of high TIL counts (P ¼ 0.023) compared exploiting the formation of RAD51 foci in proliferating cells with HR proficient (HRP) tumors and MSI was more frequent- after ex vivo irradiation of fresh breast cancer tissue: the recom- ly observed in the HRD group (2/20, 10%) than expected in bination REpair CAPacity (RECAP) test. breast cancer (1%; P ¼ 0.017). Experimental Design: Fresh samples of 170 primary breast Conclusions: RECAP is a robust functional HR assay detect- cancer were analyzed using the RECAP test. The molecular ing both BRCA1/2-deficient and BRCA1/2-proficient HRD explanation for the HRD phenotype was investigated by explor- tumors. Functional assessment of HR in a pseudo-diagnostic ing BRCA deficiencies, mutational signatures, tumor-infiltrat- setting is achievable and produces robust and interpretable ing lymphocytes (TIL), and microsatellite instability (MSI). results. Clin Cancer Res; 1–11. 2018 AACR.

Introduction the error-free DNA double-strand break (DSB) repair pathway that operates during the S- and G -phase of the cell cycle. HR Breast cancer is the most common malignancy in women with 2 deficiency (HRD) leading to impaired DNA DSB repair is fre- the second highest cancer-related mortality rate (1). Approximate- quently caused by, but not limited to, defects in BRCA1/2 (4). ly 3% of all breast cancer cases are due to germline mutations in Therapies specifically targeting tumor cells with impaired HR BRCA1/2 (2), and in triple-negative breast cancers (TNBC) this capacity are PARP inhibitors (PARPi), as well as classical che- percentage is even 10% to 20% (3). The BRCA proteins play an motherapies such as platinum-derivates and alkylating agents (5). important role in the homologous recombination (HR) pathway, PARPi causes persistence of single-strand DNA breaks (SSB) by trapping PARP1 on DNA, whereas platinum-derivates cause DNA interstrand crosslinks. Both types of lesions result in replication 1Department of Molecular Genetics and Oncode Institute, Erasmus MC University 2 fork stalling and/or collapse, frequently leading to DSBs that need Medical Center Rotterdam, the Netherlands. Department of Medical Oncology, HR for their repair (5). The targeted approach of PARPi kills tumor Erasmus MC Cancer Institute, Erasmus MC University Medical Center Rotterdam, the Netherlands. 3Erasmus MC University Medical Center Rotterdam, Cancer cells lacking HR, whereas normal cells remain unharmed, due to Computational Biology Center, Rotterdam, the Netherlands. 4Department of their normal DSB repair capacity, a phenomenon often referred to Urology, Erasmus MC University Medical Center Rotterdam, the Netherlands. as synthetic lethality. Recently, FDA approval was granted for the 5Department of Pathology, Erasmus MC University Medical Center Rotterdam, use of Olaparib in germline BRCA mutated breast cancer based on 6 7 the Netherlands. Maasstad ziekenhuis, Rotterdam, the Netherlands. Depart- the results of the Olympiad trial (6). ment of Pathology, Netherlands Cancer Institute, Amsterdam, the Netherlands. Although evidence is emerging that the use of PARPi could be Note: Supplementary data for this article are available at Clinical Cancer extended beyond germline BRCA1/2 mutated cancers to sporadic Research Online (http://clincancerres.aacrjournals.org/). cancers with BRCA-like features, a gold standard test for predicting Corresponding Author: Dik C. van Gent, Erasmus MC University Medical response to treatments targeting HR is not yet available (7). Center Rotterdam and Oncode Institute, P.O. Box 2040, Rotterdam 3000 CA, Several different HRD tests exist, mostly based on genomic the Netherlands. Phone: 31 10 704 39 32; Fax: 31 10 704 10 03; E-mail: patterns or transcriptional predictors of BRCAness (8–12). [email protected] These genomic tests measure the accumulation of mutations doi: 10.1158/1078-0432.CCR-18-0063 and chromosomal aberrations over time, but not necessarily 2018 American Association for Cancer Research. reflect the real-time HR status. Beyond mutational status, several

www.aacrjournals.org OF1

Meijer et al.

residual tumor material for research purposes were not included Translational Relevance in this study. Patients with ductal carcinoma in situ (DCIS) only or The functional RECAP test assesses HR capacity in fresh patients receiving neo-adjuvant chemotherapy were excluded. tissue samples. This is a very accurate method for diagnosing HRD tumors, because the read-out is a dynamic process rather RECAP test than a static genomic status. The main clinical implication of Obtained tissue samples were immediately transferred into this study is that functional assessment of HR in a pseudo- customized breast tissue culture medium, as described in Naipal diagnostic setting is achievable and produces robust and et al. (14). Processing of samples was performed within 4 hours interpretable results. Selection of tumors based on the HRD after the tissue was resected. Microscopic analysis of hematoxylin phenotype, instead of germline BRCA mutations, can identify and eosin (HE) stained sections was performed to determine 50% more HRD tumors. Breast cancers with HRD phenotype presence of invasive carcinoma. The RECAP test, a functional showed an increased incidence of high TILs and microsatellite assay exploiting the formation of RAD51 foci in proliferating instability (MSI). This observation provides a basis to study cells after ex vivo irradiation of fresh breast cancer tissue, was whether these specific subgroups of patients with breast cancer performed and results were analyzed as described previously fi would bene t not only from PARP inhibitors (PARPi) but also (14). In brief, presence of RAD51 foci was determined in S–G2 immunotherapy. Clinical trials have been initiated to cells only, which stain positive for Geminin. At least 30 Geminin evaluate the predictive value of the RECAP test for in vivo expressing cells were counted per tumor sample. A cell was response to PARPi. considered RAD51 positive when at least five RAD51 foci could be detected. Based on previous experiments with patient-derived xenograft (PDX) models with known BRCA status, tumors were classified as HR proficient (HRP), HR deficient (HRD), or inter- other factors influence tumor behavior and therapy response, mediate (HRi) when more than 50%, less than 20% or between such as epigenetic changes and the microenvironment of the 20% and 50% of geminin positive cells showed 5 RAD51 foci, tumor cells. Independent of the underlying cause, the down- respectively. stream effect of HR impairment (phenotype) can be assessed fl functionally. A functional diagnostic assay therefore has the Work ow of molecular characterization of the HRD potential for more precisely detecting patients who may benefit phenotype from PARPi than genomic assays. To unravel the possible molecular mechanism underlying the A functional HRD assay was firstdescribedbyGraeseretal., HRD phenotype, several molecular tests were performed retro- assessing RAD51 focus formation, a marker of HR competence, spectively (Fig. 2; Supplementary Fig. S1). As no DNA could be in tumor biopsies obtained 24 hours after in vivo anthracycline obtained for one HRD sample, we conducted the analyses for 23 BRCA BRCA1 treatment (13). This provided the first evidence that RAD51 HRD and six HRi samples. First, sequencing and focus formation can serve as a predictive biomarker. To promoter methylation analysis was performed in HRD and HRi n ¼ þ n ¼ enhance clinical utility of this biomarker, test outcomes should samples, as well as in all TNBC ( 5), ER/PR HER2 ( 2), þ n ¼ be available before start of treatment. Therefore, we developed and 21 ER/PR HRP tumors (total 28; Supplementary Fig. the homologous recombination REpair CAPacity (RECAP) test S1). The HRD and HRi tumors without molecular explanation for BRCA1 BRCA2 exploiting the formation of RAD51 foci in proliferating cells their phenotype were subjected to and MLPA after ex vivo irradiation of fresh breast cancer tissue, providing a analysis to identify large genomic rearrangements (LGR), as LGRs fi real-time HR status of the tumor (14). The aim of this study was are not usually identi ed by targeted sequencing. In addition to to validate the RECAP test in an extensive cohort of primary this targeted approach, morphologic examination of TILs and breast cancers and provide evidence that this functional test is whole exome sequencing (WES) was performed on a selection of achievable in a pseudo-clinical setting. Additionally, thorough tumors to further explore molecular aspects connected to the HRD molecular characterization of the HRD phenotype is per- phenotype. formed, proving that HRD tumors encompass more than only BRCA deficiencies. DNA isolation Isolation of DNA from 30 mm fresh frozen tissue section samples was performed using the NucleoSpin Tissue Kit Materials and Methods (Macherey-Nagel) according to the manufacturer's instructions. Primary breast cancer specimens Quantity and quality checks of isolated DNA were performed Residual fresh breast cancer tissue was prospectively collected using the MultiNA microchip electrophoresis system (Shimadzu's from lumpectomy of the breast or mastectomy specimens in the Hertogenbosch), Nanodrop 2000-v.1 (Thermo Fisher Scientific), Erasmus MC Cancer Institute, Haven hospital, and Maasstad and Qubit (Thermo Fisher Scientific). hospital in Rotterdam, the Netherlands, between 1 October 2011 and 1 September 2016. The first 41 patients were also BRCA1/2 analyses included in our previous cohort (14). After macroscopic evalu- Ion semiconductor sequencing on the Ion Torrent Personal ation of the surgical specimen by trained pathologists, residual S5XL was performed according to manufacturer's instructions tumor tissue was collected for our research purposes according to (Thermo Fisher Scientific). Adapter-ligated libraries were con- the "Code of proper secondary use of human tissue in the Nether- structed using the AmpliSeq Library Kit 2.0 with amplicons lands" established by the Dutch Federation of Medical Scientific designed targeting BRCA1/2 and TP53. Generation of sequence Societies and approved by the local Medical Ethical committees reads, trimming adapter sequences, filtering, and removal of (MEC-11-098). Patients who had objected to secondary use of poor signal-profile reads was performed via the Ion Torrent

OF2 Clin Cancer Res; 2018 Clinical Cancer Research

Functional Assay Reveals HRD in Breast Cancer Beyond BRCA

platform-specific pipeline software Torrent Suite v5.2.2. Initial and colleagues, (matrix Sij; i ¼ 96; number of trinucleotide motifs; variant calling was performed by comparison to the reference j ¼ number of signatures) were downloaded from COSMIC (as genome hg19 (build 37) using the "Torrent Variant Caller visited on 8-11-2017; ref. 24). Per sample, a constrained linear v5.2.0.3400 plug-in from the Torrent Suite Software. All BRCA2 combination of the 30 validated mutational signatures was variants were validated by Sanger sequencing and pathogenicity constructed, which reconstructs the sample-specificmutational was evaluated using interactive Biosoftware Alamut Visual spectrum, using nonnegative least squares regression imple- v.2.7.2. BRCA1 promoter methylation was analyzed as previ- mented in the R package pracma (v1.9.3). Signatures with ously described (15). Multiplex ligation-dependent probe lower relative contribution than 3% were summarized into a amplification (MLPA) analysis of BRCA1 and BRCA2 was "Filtered" category. undertaken to identify large rearrangements using the SALSA MLPA Kit P002B, and for confirmation of observed abnormal- MSI analysis, MMR protein IHC, and MLH1 promotor ities, the SALSA MLPA Kit P087 was used (MRC Holland). methylation assay Analyses were performed according to the manufacturer's These analyses were performed as previously described (25). instruction; products were run on an ABI automated sequencer fi (ABI 3730XL), and the data were analyzed by Genemarker Tumor-in ltrating lymphocytes fi version 2.7.0 (Softgenetics). Tumor-in ltrating lymphocytes (TIL) were scored on HE stained sections, according to the consensus by the International In situ detection of BRCA1 RNA TILs Working Group 2014 (26). In situ detection of BRCA1 mRNA was performed using RNA- Scope (Advanced Cell Diagnostics) on the automated Ventana Statistical analysis Discovery Ultra system (Ventana Medical Systems). BRCA1 and Statistical analyses were all two-sided and performed using IBM fi positive control peptidylprolyl isomerase B (PPIB) probes (prod- SPSS statistics v21. Signi cance was calculated by Fisher exact test – uct codes: 485479 and 313909) were purchased from the same for categorical data, by Mann Whitney test for continuous data, P company. RNAscope analysis was performed according to man- and by exact binomial test for the incidence of MSI. -values of < fi ufacturer's instructions using the reagent kit (VS Reagent Kit 0.05 were considered signi cant. 320600; Advanced Cell Diagnostics) on proteinase K (0.1%, 5 min at 37 C)-treated paraffin sections (4 mm). Results Ex vivo functional RECAP test Exome sequencing A total of 170 samples were subjected to the RECAP test DNA libraries for Illumina sequencing were generated using (Supplementary Fig. S1; ref. 14). In 148 of 170 (87%) primary standard protocols (Illumina) and subsequently sequenced in an breast cancer tissues RECAP was completed successfully (Fig. 1). Illumina HiSeq 2500 system by GATC-biotech. Exome-targeting In all cases, the reason for failure (n ¼ 22) was lack of proliferating was performed using the Sureselect v5 (v6 for tumors M077, tumor cells. No differences in clinicopathologic characteristics M209, M211) methods using standard protocols (Agilent Tech- between tumors that yielded successful versus nonsuccessful tests nologies). DNA libraries were whole-exome sequenced (2 were observed (Supplementary Table S1). The first 41 patients 125bp) using the HiSeq v4 paired-end sequencing protocol to were also included in our previous cohort. Here, we show that a minimum depth base coverage of 90 for tumor samples and execution of a functional assay in a pseudo-diagnostic setting is 60 for matched normal. Sequence reads were mapped against achievable and validate the findings from the earlier cohort (11% human reference genome GRCh37 using Burrows–Wheeler HRD in cohort 1 vs. 19% HRD in cohort 2, P ¼ 0.339). In total, we Aligner (v0.7.12) with default settings (16). Sequence reads identified 118 (80%) HRP, 24 (16%) HRD, and 6 (4%) HRi originating from multiple lanes were merged after alignment samples (Fig. 1). Both HRi and HRD tumors were more frequently using Samtools (v1.5) prior to further analysis (17). Sequence TN (P < 0.001) and Bloom and Richardson (B&R) grade 3 duplicates were marked using PicardTools (v1.129; ref. 18). (P < 0.001) than HRP tumors. Also, HRD tumors had a larger Somatic variant calling was performed by Mutect2 (v3.7) using size (P ¼ 0.050) and were never B&R grade 1 (Supplementary a matched-normal design while utilizing the dbSNP (v149, hg19) Table S2). and COSMIC (v80, hg19) databases and using default settings (19–21). Variant annotation was performed by ANNOVAR (22). Identification of BRCA defects in HRD breast cancers Heuristic filtering removed variants not passing all standard Pathogenic BRCA2 mutations were found in six (6/23 ¼ 26%) Mutect2 post-calling filters. Sequence data have been deposited HRD samples, but not in the tested HRP samples (Table 1). In six at the European Genome-phenome Archive (EGA, http://www. tumors (two HRD, one HRi, and three HRP), BRCA2 variants were ebi.ac.uk/ega/), which is hosted by the EBI, under accession detected that were classified as benign. We did not identify any number EGAD00001003929. BRCA1 point mutations. Next, BRCA1 promoter hypermethylation was detected in six (6/23 ¼ 26%) HRD tumors and in one Mutational signatures HRi tumor, but not in tested HRP samples (Table 1). Interestingly, For each somatic variant, its trinucleotide context was derived all six HRD tumors displaying BRCA1 promoter hypermethyla- from the human reference genome GRCh37 and enumerated into tion were TNBCs (Table 1; Supplementary Table S3). Vice versa,of a mutational spectrum matrix Mij (i ¼ 96; number of trinucleotide the six HRD tumors harboring a pathogenic BRCA2 mutation, five contexts; j ¼ number of samples) using the MutationalPatterns R showed hormone receptor positivity. package (v1.4.0) in the R statistical platform (23). Multi-allelic Thus, 12/23 HRD and 1/6 HRi samples were explained by and InDel variants were not included in this analysis. The 30 a BRCA2 mutation or BRCA1 promoter hypermethylation. Sub- consensus mutational signatures, as established by Alexandrov sequently, we proceeded with an MLPA analysis for BRCA1 and

www.aacrjournals.org Clin Cancer Res; 2018 OF3

Meijer et al.

A N = 24 N = 6 16% 4%

Primary breast cancer samples (N = 170) Successful tests (N = 148)

N = 118 80%

Figure 1. RECAP test results: 16% of primary breast cancers B N = 148 showed HRD. A, Schematic representation of 100 RAD51 positive (HRP) the primary breast cancers obtained for RECAP testing. B, Percentage of RAD51 foci RAD51 intermediate (HRi) 90 positive tumor cells among geminin-expressing 80 RAD51 negative (HRD) nuclei in the 148 successful tests. A total of 118 breast cancer samples were HRP (>50% Geminin 70 positive cells with RAD51 foci), 24 were HRD < 60 ( 20% Geminin positive cells with RAD51 foci), and six samples were HRi (>20%/<50% 50 Geminin positive cells with RAD51 foci). Black dots indicate TNBCs. 40

10 % Geminin+ cells with >5 RAD51 foci 0 1 12 23 34 45 56 67 78 89 100 111 122 133 144

BRCA2 on the 16 HRD and HRi samples that remained unex- a subclone of the tumor) and a BRCA2 duplication (Table 1). In plained to identify possible LGRs. Large BRCA1 deletions were total, BRCA defects have thus been identiﬁed in 16/23 HRD and found in four samples and BRCA2 deletions in two samples 2/6 HRi samples. (Table 1), of which one tumor harbored both a BRCA1 and Silencing of BRCA1 was validated in tumors with BRCA1 BRCA2 deletion. Two tumors [M248 (HRD) and M112 (HRi)] promoter hypermethylation (n ¼ 7) and BRCA1 LGRs (n ¼ 4) showed extensive chromosomal instability as they contained by RNAscope in situ RNA hybridization (Supplementary a mosaic BRCA1 deletion (meaning the deletion was present in Fig. S2; Table 1). All tumors displaying BRCA1 promoter M106 M211 M093 M077 M260 M271 P001 M270 M253 M188 M096 M182 M119 M277 M028 M094 M248 M232 M156 M055 M275 M231 M057 P002 M131 M141 M209 M278 M112 RECAP test BRCA2 mutation BRCA1 promotor methylation BRCA1/2 LGRs Mutational signatures MSI >10% TILS ER+

RECAP negative

RECAP intermediate

Yes

BRCA mutational signature

MSI mutational signature

Not performed

Figure 2. Characteristics of HRD and HRi tumors. Overview of molecular alterations and TIL counts in each HRD or HRi tumor (NB, data for one HRD tumor is absent, because DNA was unavailable). ER, estrogen receptor. High TILs were deﬁned as >10% TILs.

OF4 Clin Cancer Res; 2018 Clinical Cancer Research

Functional Assay Reveals HRD in Breast Cancer Beyond BRCA

Table 1. Overview of samples with BRCA defects BRCA1 promotor BRCA2 mutation hypermethylation BRCA1 mRNA Sample Mutation Pathogenicity (score) RNAscope BRCA LGRs Tumor % RECAP status M057 c.517G>C Pathogenic $ >50% Negative (HRD) p.G173R P002 c.7617þ1G>T Pathogenic 60% Negative (HRD) M096 c.7285G>T Pathogenic $ 50–70% Negative (HRD) p.E2429X M188 c.3846_3847del Pathogenic Macrodissected Negative (HRD) p.T1282fs (FFPE) M231 c.755_758del Pathogenic 65% Negative (HRD) p.D252fs M275 c.3269delT Pathogenic 33% Negative (HRD) p.M1090fs M213 c.1708A>C Benign >50% Positive (HRP) p.N570H M114 c.6829T>C Benign 40% Positive (HRP) p.L2277L P9 c.3445A>G Benign 50% Positive (HRP) p.M1149V M211 c.5054C>T Benign (0.02) Positive >50 Negative (HRD) p.S1685L M106 c.6347A>G Benign (0.01) Positive 50–70 Negative (HRD) p.H2116R P001 þ (NA) Negative — Negative (HRD) M028 þ (NA) Negative — Negative (HRD) M119 þ (0.56) Negative 50–70 Negative (HRD) M131 þ (0.82) Heterogeneous >70 Negative (HRD) M182 þ (0.45) Negative >50 Negative (HRD) M277 þ (0.24) Heterogeneous 20 Negative (HRD) M141 þ (0.29) Negative 50 Intermediate (HRi) M094 (0.01) Negative BRCA1 deletion 50–70 Negative (HRD) M232 (0.02) Positive BRCA 1 þ 2 deletion >70 Negative (HRD) M248 (0.01) Positive Mosaic BRCA1 deletion, 50–70 Negative (HRD) BRCA2 duplication M112 c.6935A>T Benign (0.01) Positive Mosaic BRCA1 deletion, 50 Intermediate (HRi) p.D2312V BRCA2 duplication M156 (0.01) Positive BRCA2 deletion >70 Negative (HRD) M093 (0.01) Positive WT >70 Negative (HRD) M260 (0.01) Positive WT 30 Negative (HRD) M270 (0.01) Positive WT 50–70 Negative (HRD) M271 (0.01) Positive WT 30 Negative (HRD) M077 (0.01) Positive WT >50 Negative (HRD) M253 (0.01) Positive WT 50 Intermediate (HRi) M209 (0.01) Positive WT 50 Intermediate (HRi) M055 (0.01) Positive WT 50–70 Intermediate (HRi) M278 (0.01) 10 Intermediate (HRi) NOTE: Methylation score between 0.0 and 1.0, cut-off for presence of promotor hypermethylation is >0.2. $ Mutations were somatic. hypermethylation showed absence of BRCA1 RNA, except for tumors did not show any BRCA defects and therefore remained two heterogeneous samples that contained BRCA1 positive and unexplained (Fig. 2). To further characterize these HRD tumors, negative areas (Supplementary Fig. S2). As expected, the BRCA1 functional features correlating with the HRD phenotype were deletion in tumor M094 led to a total absence of BRCA1 mRNA determined. (Supplementary Fig. S2). The mosaic BRCA1 deletions did not result in complete BRCA1 silencing (Supplementary Fig. S2). HRD tumors show more tumor infiltrating lymphocytes than Neither did the BRCA1 deletion in M232, however this HRP tumors tumor also harbored a BRCA2 deletion that can explain the Recently, a subgroup of patients with TNBC was identified who HRD phenotype. showed good response to immune checkpoint inhibition through After thorough analysis of the BRCA genes using several programmed death-ligand 1 (PD-L1) blockade (27, 28). This techniques, 13/23 HRD and 1/6 HRi samples could be subgroup was characterized by having >10% TILs and high CD8 explained by deficiencies in BRCA. Moreover, 3/23 HRD lymphocyte counts in the tumor centers (28). Because 11/17 (M131, M277, and M248) and 1/6 HRi (M112) tumors also TNBCs in our study showed HRD, we hypothesized that the harbored BRCA deficiencies (BRCA1 promoter methylation or RECAP test might select for a specific subgroup of patients with mosaic BRCA1 deletions), which explain the HRi and partially TNBC who might benefit from PD-L1 therapy. We found that the HRD phenotype, as these tumors showed heterogeneous significantly more HRD tumors (6/22) had >10% TILs than HRP BRCA1 mRNA expression. Finally, 7/23 HRD and 4/6 HRi tumors (0/28) (P ¼ 0.004; Fig. 3). Also, tumors with BRCA defects

www.aacrjournals.org Clin Cancer Res; 2018 OF5

Meijer et al.

A <1% TILs 15% TILs 70% TILs

100 µm B P = 0.004 P = 0.026 P = 0.001 80 80 80

60 60 60

40 40 40 % TILs % TILs % TILs 20 20 20

0 0 0

D P C A R R H HRi H NB RC T B

Non-TNBC Non-BRCA

Figure 3. HRD tumors more frequently showed >10% tumor inﬁltrating lymphocytes than HRP tumors. A, HE sections were scored for stromal TILs. Examples of samples with <1%, 15%, and 70% TILs are shown respectively. B, More HRD tumors (6/22) had >10% TILs than HRP tumors (0/28; P ¼ 0.004). Also, tumors with BRCA defects showed more frequently >10% TILs compared with non-BRCA tumors (P ¼ 0.001), which was also true for TNBCs compared to ER/PRþ tumors (P ¼ 0.026). Signiﬁcance was calculated by Fisher exact test.

showed more frequently >10% TILs compared with non-BRCA similar distributions of the mutational signatures (Supplemen- tumors (P ¼ 0.001), which was also true for TNBCs compared tary Fig. S3). with ER/PRþ tumors (P ¼ 0.026). Mutational signature 3 is related to failure of DSB repair by HR and associated with germline and somatic BRCA1/2 defects in HRD tumors show mutational signatures related to BRCA breast, pancreatic, and ovarian cancers (31). Signature 3 was deficiencies and microsatellite instability present in 6 of 10 analyzed samples (M94, M95, M119, M131, Next, WES was performed to determine the molecular land- M141, and M221; Fig. 4). APOBEC-related mutagenesis (pre- scape of HRD tumors. A selection of HRD (n ¼ 8) and HRi (n ¼ 3) dominantly C>GorC>T substitutions in TCA or TCT motifs) is tumors with BRCA1/2 mutations/deletions, BRCA1 promoter captured in signatures 2 and 13, which arise through activity of the hypermethylation, and BRCA WT tumors and one HRP tumor AID/APOBEC family. BRCA-related signatures could also be were subjected to WES. First, mutational load was determined in identified to a lesser extent in samples (M211 and M094) having these tumors and high mutational load did not correlate with a high mutational burden of signatures 2 and 13. high numbers of TILs. Second, we did not identify commonly mutated genes other than BRCA1/2, which might explain the Microsatellite instability in HRD breast cancers functional HR defect. Third, WES data were used to identify One tumor (M077) showed a high mutational load and mutational signatures which are specific combinations of muta- high contributions of signatures 15, 20, and 26, which are tions that arise due to a certain underlying mutational or DNA related to microsatellite instability (MSI) and mismatch repair processes (29). repair (MMR) deficiency (http://cancer.sanger.ac.uk/cosmic/ Mutational signatures were derived from the WES data from signatures). MSI is a common feature in endometrial and HRD/HRi tumors of which matching normal DNA was gastro-intestinal cancers, caused by either germline (Lynch available to filter out germline variants (n ¼ 10) to explore novel syndrome) or somatic mutations in one of the MMR genes mechanisms related to or underlying the HRD phenotype. and/or promoter hypermethylation (32, 33). Presence of MSI Because discussion in the field exists that mutational signatures in tumor M077 was confi rmed using pentaplex PCR and IHC can only be faithfully obtained from WGS instead of WES data, for the MMR proteins, which showed absence of MLH1 and we first carried out a pilot experiment comparing mutational PMS2, caused by MLH1 promoter methylation (Supplemen- signature analysis using all somatic mutations in five BrC WGS tary Fig. S4). datasets (30) and filtered these WGS datasets to only contain A set of 44 tumors (20 HRD, 6 HRi, 18 HRP) was subjected to somatic mutations on exonic regions. Both methods resulted in MSI analysis by pentaplex PCR and IHC for the MMR proteins.

OF6 Clin Cancer Res; 2018 Clinical Cancer Research

Functional Assay Reveals HRD in Breast Cancer Beyond BRCA

100

Relative contribution 25

Figure 4. HRD and HRi tumors showed mutational signatures related to BRCA and MSI. Relative and absolute contribution of the mutational signatures in 0 WES datasets is depicted, and total numbers of single-nucleotide mutations and percentage of 0 TILs in the sample.

1,000

1 6 12 22 Filtered (<3%) 2 7 13 23 Signatures 3 8 15 26 2,000 4 16 29

Absolute contribution 10 5 11 20 30

Sample M211 M094 M077 M131 M093 M141 M106 M112 M096 M119

Mutational = 611) = 123) load (n = 2850) (n = 1280) (n = 765) (n (n = 267) (n = 260) (n = 237) (n = 227) (n (n = 101) TILs 10% <1% <1% <1% 1% 30% 1% 60% 1% <1%

One HRD tumor (M188) of which mutational signatures were cohort, as compared with our previous report. Therefore, both not available, also showed MSI. MSI was never detected in cohorts were combined to achieve more power to thoroughly any HRP tumors (Fig. 5). Tumor M188 showed absence of investigate the molecular mechanism underlying the HRD MSH2 and MSH6, caused by a homozygous deletion of MSH2 phenotype. Sixteen HRD samples showed deficiencies in (Supplementary Fig. S4). The two MSI BrCs (M077 and M188) BRCA1/2 (BRCA mutations, deletions, or promoter hyper- were TN and BRCA WT and ER positive and BRCA2 mutated, methylation), whereas seven HRD tumors were non-BRCA respectively (Fig. 2). The incidence of MSI within the HRD related, demonstrating that HRD tumors encompass more group (2/20, 10%) is significantly higher than the incidence of than only BRCA-deficient tumors. MSI in the unselected breast cancer population (1%; P ¼ 0.017; Several different HRD tests have been designed to identify refs. 34, 35). HRD tumors in addition to the BRCA mutated or promoter methylated tumors to enlarge the population of patients with breast cancer that could benefit from treatments targeting the Discussion HR pathway. For example, the BRCAness classifier, which is Here, a unique series of fresh primary breast cancer tissues based on specific genomic patterns derived from copy number (n ¼ 148) has been analyzed for HRD using the functional data of BRCA1/2 mutated breast cancers that also occur in RECAP test. This first large validation study describes that sporadic cancers (10, 11) and the Myriad MyChoice HRD test, functional assessment of HR in a pseudo-diagnostic setting is which is a combined score of three different structural chro- achievable and produces robust and interpretable results for mosomal aberrations [telomeric allelic imbalance (TAI), large- most patients (87%). We found that the percentage of HRD scale transition (LST), and loss of heterozygosity (LOH); tumors detected by the RECAP test is similar in this larger ref. 36]. Both the BRCAness classifier and MyChoice are robust,

www.aacrjournals.org Clin Cancer Res; 2018 OF7

Meijer et al.

MSI A n = 2

HRP HRD n = 18 n = 20

HRi n = 6

B MSH2 MSH6 MLH1 PMS2

M077

M188

100 µm

Figure 5. Two HRD breast tumors showed MSI. A, MSI was found in two HRD tumors, but not in any HRP or HRi tumors. B, IHC staining of MLH1, PMS2, MSH2, and MSH6.

easily applicable in the clinic, and have also been validated to material. The percentage of non-BRCA HRD tumors (approx- predict in vivo response to high dose chemotherapy and neo- imately 33%) detected by the HRDetect and the RECAP test are adjuvant platinum-based therapy, respectively, in patients with quite comparable, although it remains elusive whether these TNBC (37, 38). As opposed to the neo-adjuvant setting, the tests identify the same tumors, therefore comparison of several MyChoice HRD test did not predict response to PARPi therapy HRD tests within the same patient cohort is required. in platinum-sensitive recurrent ovarian cancer (39). These The major strength of this study is that functional diagnos- genomic HRD tests have the drawback that they do not deter- tics have been applied to a unprecedented large collection of mine the real-time HR status, also it remains unclear whether tumors. The advantage of the RECAP test over genetic tests, is all HRD cases are identified. In theory, the functional RECAP its functional character for exploring the HR phenotype rather test can also detect reversion of the HRD phenotype in BRCA- than the static nature of genomic tests. Also, the RECAP test is deficient tumors, that have been treated with various DNA feasible in samples with low tumor percentage, because the damaging chemotherapies that may have induced resistance. microscopic read-out allows differentiation between tumor Moreover, the BRCAness classifier focuses on TNBC only, and stromal cells. The RECAP test has a high success rate whereas the RECAP test identifies HRD independent of hor- (87%) and results are available within one week after the monal status. More recently, a HRD test based on several biopsy procedure. In this study, reproducibility and robust- genomic signatures has been published, HRDetect (40). This ness of the RECAP test is validated in an independent set of test has recently been shown to predict in vivo response to 129 tumors. For the molecular analyses to unravel the platinum-based therapies in advanced breast cancer (41). mechanisms underlying the HRD phenotype, the 41 samples HRDetect relies on whole genome sequencing and is therefore included in our previous publication were also included, to more expensive and has a longer turnaround time for biopsy achieve more power. The main limitation of this study is that results, hampering its clinical implementation. Furthermore, although prospective trials evaluating the predictive value of the tumor cell percentage needs to be above 50% for reliable RECAP for in vivo patient response to PARPi have been initi- results, which is not a prerequisite for the RECAP test. How- ated, results are not yet available. However, previously a ever, HRDetect has the advantage that it can be performed functional RAD51 test performed on biopsies obtained from on frozen material, whereas the RECAP test requires fresh patients 24 hours after start of therapy correlated with

OF8 Clin Cancer Res; 2018 Clinical Cancer Research

Functional Assay Reveals HRD in Breast Cancer Beyond BRCA

response to anthracycline-based therapies, indicating that continuing DNA replication. As MSI tumors have many neo- functional assessment of HR can have predictive value for antigens, PD-L1 blockade therapy showed antitumor activity in therapy response (13). phase I trials (45). Recently, a firstreportofapatientwithMSI Among the spectrum of BRCA defects, we have not identified BrC showing a profound response to PD-L1 blockade was any BRCA1 mutations. This is somewhat remarkable but could be published (46). Moreover, tumors with high numbers of TILs due to a selection bias, as the Erasmus Medical Center is special- are generally more sensitive to immunotherapy (47). Interest- ized in hereditary BrC and all patients with TNBC are tested, ingly, in our cohort, the two MSI tumors comprise a different therefore most families with hereditary BRCA1 mutations have subset of HRD tumors than the ones with high TILs. Thus, the been identified and carriers are offered strict screening programs RECAP test identifies not only tumors with BRCA defects (n ¼ or undergo prophylactic surgery in The Netherlands (42). Also, in 16), but also a subgroup of breast cancers that might respond this study there is a selection bias for tumor size, as the tumor well to immunotherapy due to either MSI (n ¼ 2) or high TIL should be large enough to provide residual material without counts (n ¼ 6). compromising standard diagnostic procedures. Since BRCA The RECAP test is a robust functional HR assay, detecting both mutation carriers are offered strict screening programs, tumors BRCA1/2-deficient and BRCA1/2-proficient HRD tumors. Func- are often identified at an early stage and residual material is not tional assessment of HR in a pseudo-diagnostic setting is achiev- available for the RECAP test. Also, many BRCA1 mutation carriers able and produces robust and interpretable results. Clinical trials with TNBC are treated with neoadjuvant chemotherapy, which evaluating the predictive value of the RECAP test for in vivo was an exclusion criterion for this study. response to PARPi have been initiated. Clinical consequences of BRCA1 promoter methylation are unclear (43). In the current study, BRCA1 promoter methyl- Disclosure of Potential Conflicts of Interest ation resulted in absence of BRCA1 RNA in four samples, but H.J. Dubbink reports receiving commercial research grants from AstraZeneca. in two tumors there was still heterogeneous expression of W.N.M. Dinjens reports receiving speakers bureau honoraria from Roche, and is fl BRCA1 RNA (Supplementary Fig. S2). In these tumors, the a consultant/advisory board member for Amgen. No potential con icts of interest were disclosed by the other authors. percentage of cells with RAD51 foci was 1% and 2%, respectively. The discrepancy between the very low HRD score and Authors' Contributions heterogeneous BRCA1 RNA could be explained by sampling Conception and design: T.G. Meijer, N.S. Verkaik, K.A.T. Naipal, A. Jager, from different areas of the tumor, since the tumor sample for D.C. van Gent the RECAP test was irradiated and an unirradiated tumor Development of methodology: W.N.M. Dinjens, P.M. Nederlof, sample was used for molecular analyses. This sampling error H.J G. van de Werken, J.W.M. Martens, A. Jager, D.C. van Gent is not limited to this study, but also occurs in regular diag- Acquisition of data (provided animals, acquired and managed patients, nostics when biopsies are obtained from a certain region of a provided facilities, etc.): T.G. Meijer, N.S. Verkaik, A.M. Sieuwerts, K.A.T. Naipal, C.H.M. van Deurzen, M.A. den Bakker, H.-J. Dubbink, heterogeneous tumor. Tumor heterogeneity in BRCA1 promot- T.D. den Toom, W.N.M. Dinjens, E. Lips, P.M. Nederlof, A. Jager er methylated tumors is very important for clinical decisions Analysis and interpretation of data (e.g., statistical analysis, biostatistics, on PARPi use. If subsequent studies reveal that this computational analysis): T.G. Meijer, J. van Riet, H.-J. Dubbink, phenomenon is observed in a large fraction of these tumors, T.D. den Toom, W.N.M. Dinjens, E. Lips, P.M. Nederlof, M. Smid, PARPi may not be very effective in tumors with BRCA1 pro- H.J G. van de Werken, J.W.M. Martens, A. Jager, D.C. van Gent moter methylation. Writing, review, and/or revision of the manuscript: T.G. Meijer, N.S. Verkaik, A.M. Sieuwerts, K.A.T. Naipal, C.H.M. van Deurzen, H.-J. Dubbink, We identified 6 HRi BrCs in the current cohort. Only in one of W.N.M. Dinjens, E. Lips, P.M. Nederlof, M. Smid, H.J G. van de Werken, the HRi tumors, distinct areas of RAD51 negative and RAD51 R. Kanaar, J.W.M. Martens, A. Jager, D.C. van Gent positive tumor cells were observed, suggesting clonal heteroge- Administrative, technical, or material support (i.e., reporting or organizing neity. The other HRi tumors all showed interspersed RAD51 data, constructing databases): A.M. Sieuwerts, J. van Riet, H.F.B.M. Sleddens, negative as well as RAD51 positive tumor cells. As a BRCA defect A. Jager was found in 2/6 HRi tumors, they biologically resemble HRD Study supervision: R. Kanaar, A. Jager, D.C. van Gent tumors. However, whether HRi tumors benefit from PARPi treatment remains to be elucidated. Acknowledgments Mutational signature analyses were performed to explore The authors thank many colleagues from the Departments of Molecular Genetics, Medical Oncology, and Pathology at Erasmus MC as well as from the novel mechanisms related to or underlying the HRD pheno- Maasstad Hospital, who contributed to the collection of patient material. We type. One HRD tumor that showed a large contribution of three thank Lindsey Oudijk for her assistance with the TILs scoring. We thank Ronald signatures related to MSI and MMR deficiency proved to be van Marion for expert technical assistance. D.C. van Gent, A. Jager, and R. Kanaar truly MSI. Using pentaplex PCR and IHC, MSI was discovered in have received funding from the Dutch Cancer Society (Alpe d'Huzes grant two HRD (2/20, 10%) but not in HRP or HRi tumors. This number EMCR 2014-7048 and grant number EMCR 2008-4045). This work is fi incidence is much higher than in the unselected breast cancer part of the Oncode Institute, which is partly nanced by the Dutch Cancer Society and was funded by the gravitation program CancerGenomiCs.nl from population (1%; refs. 34, 35), suggesting that the RECAP test the Netherlands Organisation for Scientific Research (NWO). may also identify MSI tumors. The relation between MSI and HRD as well as the order in which tumors develop these The costs of publication of this article were defrayed in part by the deficiencies remains unclear and future research is required. payment of page charges. This article must therefore be hereby marked We hypothesize that either MSI tumors acquire HRD over time advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate due to accumulation of mutations in genes involved in HR this fact. (44), or HRD tumors acquire MSI at a later stage of tumor development as a compensatory mechanism, to lower replica- Received January 8, 2018; revised May 17, 2018; accepted August 17, 2018; tion fork instability by not repairing mismatches but rather published first August 23, 2018.

www.aacrjournals.org Clin Cancer Res; 2018 OF9

Meijer et al.

References 1. Siegel RL, Miller KD, Jemal A. Cancer Statistics, 2017. CA Cancer J Clin 22. Wang K, Li M, Hakonarson H. ANNOVAR: functional annotation of genetic 2017;67:7–30. variants from high-throughput sequencing data. Nucleic Acids Res 2. Nelson HD, Fu R, Goddard K, Mitchell JP, Okinaka-Hu L, Pappas M, et al. 2010;38:e164. Risk assessment, genetic counseling, and genetic testing for BRCA-Related 23. Blokzijl F, Janssen R, van Boxtel R, Cuppen E. MutationalPatterns: an Cancer: systematic review to update the U.S. Preventive Services Task Force integrative R package for studying patterns in base substitution catalogues Recommendation. Pacific Northwest Evidence-based Practice Center, Francis. bioRxiv Prepr 2017. doi: https://doi.org/10.1101/071761. Rockville (MD): Agency for Healthcare Research and Quality (US); 2013 24. Alexandrov LB, Nik-Zainal S, Wedge DC, Campbell PJ, Stratton MR. Dec. Report No.: 12-05164-EF-1. Deciphering signatures of mutational processes operative in human cancer. 3. Hartman AR, Kaldate RR, Sailer LM, Painter L, Grier CE, Endsley RR, et al. Cell Rep 2013;3:246–59. Prevalence of BRCA mutations in an unselected population of triple- 25. van Lier MG, Wagner A, van Leerdam ME, Biermann K, Kuipers EJ, negative breast cancer. Cancer 2012;118:2787–95. Steyerberg EW, et al. A review on the molecular diagnostics of Lynch 4. Konecny GE, Kristeleit RS. PARP inhibitors for BRCA1/2-mutated and syndrome: a central role for the pathology laboratory. J Cell Mol Med sporadic ovarian cancer: current practice and future directions. Br J Cancer 2010;14:181–97. 2016;115:1157–73. 26. Salgado R, Denkert C, Demaria S, Sirtaine N, Klauschen F, Pruneri G, et al. 5. Lord CJ, Ashworth A. The DNA damage response and cancer therapy. The evaluation of tumor-infiltrating lymphocytes (TILs) in breast cancer: Nature 2012;481:287–94. recommendations by an International TILs Working Group 2014. Ann 6. Robson M, Im SA, Senkus E, Xu B, Domchek SM, Masuda N, et al. Olaparib Oncol 2015;26:259–71. for metastatic breast cancer in patients with a germline BRCA mutation. 27. Kurozumi S, Fujii T, Matsumoto H, Inoue K, Kurosumi M, Horiguchi J, et al. N Engl J Med 2017;377:523–33. Significance of evaluating tumor-infiltrating lymphocytes (TILs) and 7. Wendie DdB, Kasmintan AS, Sophie S, Kasaian K, Leelakumari S, Jones M, programmed cell death-ligand 1 (PD-L1) expression in breast cancer. Med et al. Homologous recombination deficiency in breast cancer: a clinical Mol Morphol 2017;50:185–94. review. JCO Prec Oncol 2017:1–13. 28. Schmid P, Cruz C, Braiteh FS, Eder JP, Tolaney S, Kuter I, et al. Atezoli- 8. Abkevich V, Timms KM, Hennessy BT, Potter J, Carey MS, Meyer LA, et al. zumab in metastatic TNBC (mTNBC): Long-term clinical outcomes and Patterns of genomic loss of heterozygosity predict homologous recombi- biomarker analyses. Abstract #2986, AACR 2017. nation repair defects in epithelial ovarian cancer. Br J Cancer 2012;107: 29. Nik-Zainal S, Alexandrov LB, Wedge DC, Van Loo P, Greenman CD, Raine 1776–82. K, et al. Mutational processes molding the genomes of 21 breast cancers. 9. Birkbak NJ, Wang ZC, Kim JY, Eklund AC, Li Q, Tian R, et al. Telomeric Cell 2012;149:979–93. allelic imbalance indicates defective DNA repair and sensitivity to DNA- 30. Nik-Zainal S, Davies H, Staaf J, Ramakrishna M, Glodzik D, Zou X, et al. damaging agents. Cancer Discov 2012;2:366–75. Landscape of somatic mutations in 560 breast cancer whole-genome 10. Joosse SA, Brandwijk KI, Devilee P, Wesseling J, Hogervorst FB, Verhoef S, sequences. Nature 2016;534:47–54. et al. Prediction of BRCA2-association in hereditary breast carcinomas 31. Alexandrov LB, Nik-Zainal S, Wedge DC, Aparicio SAJR, Behjati S, Biankin using array-CGH. Breast Cancer Res Treat 2012;132:379–89. AV, et al. Signatures of mutational processes in human cancer. Nature 11. Joosse SA, van Beers EH, Tielen IH, Horlings H, Peterse JL, Hoogerbrugge N, 2013;500:415–21. et al. Prediction of BRCA1-association in hereditary non-BRCA1/2 breast 32. Geurts-Giele WR, Leenen CH, Dubbink HJ, Meijssen IC, Post E, Sleddens carcinomas with array-CGH. Breast Cancer Res Treat 2009;116:479–89. HF, et al. Somatic aberrations of mismatch repair genes as a cause of 12. Konstantinopoulos PA, Spentzos D, Karlan BY, Taniguchi T, Fountzilas E, microsatellite-unstable cancers. J Pathol 2014;234:548–59. Francoeur N, et al. Gene expression profile of BRCAness that correlates with 33. Lynch HT, Snyder CL, Shaw TG, Heinen CD, Hitchins MP. Milestones of responsiveness to chemotherapy and with outcome in patients with Lynch syndrome: 1895–2015. Nat Rev Cancer 2015;15:181–94. epithelial ovarian cancer. J Clin Oncol 2010;28:3555–61. 34. Dudley JC, Lin MT, Le DT, Eshleman JR. Microsatellite instability as a 13. Graeser M, McCarthy A, Lord CJ, Savage K, Hills M, Salter J, et al. A marker of biomarker for PD-1 blockade. Clin Cancer Res 2016;22:813–20. homologous recombination predicts pathologic complete response to 35. Le DT, Durham JN, Smith KN, Wang H, Bartlett BR, Aulakh LK, et al. neoadjuvant chemotherapy in primary breast cancer. Clin Cancer Res Mismatch repair deficiency predicts response of solid tumors to PD-1 2010;16:6159–68. blockade. Science 2017;357:409–13. 14. Naipal KA, Verkaik NS, Ameziane N, van Deurzen CH, Ter Brugge P, Meijers 36. Timms KM, Abkevich V, Hughes E, Neff C, Reid J, Morris B, et al. Asso- M, et al. Functional ex vivo assay to select homologous recombination- ciation of BRCA1/2 defects with genomic scores predictive of DNA damage deficient breast tumors for PARP inhibitor treatment. Clin Cancer Res repair deficiency among breast cancer subtypes. Breast Cancer Res 2014; 2014;20:4816–26. 16:475. 15. Gross E, van Tinteren H, Li Z, Raab S, Meul C, Avril S, et al. Identification 37. Vollebergh MA, Lips EH, Nederlof PM, Wessels LF, Schmidt MK, van Beers of BRCA1-like triple-negative breast cancers by quantitative multiplex- EH, et al. An aCGH classifier derived from BRCA1-mutated breast cancer ligation-dependent probe amplification (MLPA) analysis of BRCA1- and benefit of high-dose platinum-based chemotherapy in HER2-negative associated chromosomal regions: a validation study. BMC Cancer 2016; breast cancer patients. Ann Oncol 2011;22:1561–70. 16:811. 38. Telli ML, Timms KM, Reid J, Hennessy B, Mills GB, Jensen KC, et al. 16. Li H, Durbin R. Fast and accurate short read alignment with Burrows- Homologous recombination deficiency (HRD) score predicts response to Wheeler transform. Bioinformatics 2009;25:1754–60. platinum-containing neoadjuvant chemotherapy in patients with triple- 17. Li H, Handsaker B, Wysoker A, Fennell T, Ruan J, Homer N, et al. The negative breast cancer. Clin Cancer Res 2016;22:3764–73. Sequence Alignment/Map format and SAMtools. Bioinformatics 2009;25: 39. Mirza MR, Monk BJ, Herrstedt J, Oza AM, Mahner S, Redondo A, et al. 2078–9. Niraparib maintenance therapy in platinum-sensitive, recurrent ovarian 18. https://broadinstitute.github.io/picard/%5Cnhttp://broadinstitute.github. cancer. N Engl J Med 2016;375:2154–64. io/picard/.BIPthbgipAa.). 40. Davies H, Glodzik D, Morganella S, Yates LR, Staaf J, Zou X, et al. HRDetect 19. Forbes SA, Beare D, Boutselakis H, Bamford S, Bindal N, Tate J, et al. is a predictor of BRCA1 and BRCA2 deficiency based on mutational COSMIC: somatic cancer genetics at high-resolution. Nucleic Acids Res signatures. Nat Med 2017;23:517–25. 2017;45:D777–83. 41. Zhao EY, Shen Y, Pleasance E, Kasaian K, Leelakumari S, Jones M, et al. 20. McKenna A, Hanna M, Banks E, Sivachenko A, Cibulskis K, Kernytsky A, Homologous recombination deficiency and platinum-based therapy out- et al. The genome analysis toolkit: a MapReduce framework for ana- comes in advanced breast cancer. Clin Cancer Res 2017;23:7521–30. lyzing next-generation DNA sequencing data. Genome Res 2010;20: 42. Drooger JC, Hooning MJ, Seynaeve CM, Baaijens MH, Obdeijn IM, Sleijfer 1297–303. S, et al. Diagnostic and therapeutic ionizing radiation and the risk of a first 21. Sherry ST, Ward MH, Kholodov M, Baker J, Phan L, Smigielski EM, et al. and second primary breast cancer, with special attention for BRCA1 and dbSNP: the NCBI database of genetic variation. Nucleic Acids Res BRCA2 mutation carriers: a critical review of the literature. Cancer Treat Rev 2001;29:308–11. 2015;41:187–96.

OF10 Clin Cancer Res; 2018 Clinical Cancer Research

Functional Assay Reveals HRD in Breast Cancer Beyond BRCA

43. Sun T, Ruscito I, Dimitrova D, Chekerov R, Kulbe H, Baron U, et al. Genetic 45. Westdorp H, Fennemann FL, Weren RD, Bisseling TM, Ligtenberg MJ, Figdor Versus Epigenetic BRCA1 silencing pathways: clinical effects in primary CG, et al. Opportunities for immunotherapy in microsatellite instable ovarian cancer patients: a study of the tumor bank ovarian cancer consor- colorectal cancer. Cancer Immunol Immunother 2016;65:1249–59. tium. Int J Gynecol Cancer 2017;27:1658–65. 46. Marleen K, Hugo MH, Petur S, Chalabi M, Schumacher TN, Blank CU, et al. 44. Zhao H, Thienpont B, Yesilyurt BT, Moisse M, Reumers J, Coenegrachts L, Profound immunotherapy response in mismatch repair-deﬁcient breast et al. Mismatch repair deﬁciency endows tumors with a unique mutation cancer. JCO Prec Oncol 2017:1–3. signature and sensitivity to DNA double-strand breaks. Elife 2014;3: 47. Gibney GT, Weiner LM, Atkins MB. Predictive biomarkers for checkpoint e02725. inhibitor-based immunotherapy. Lancet Oncol 2016;17:e542–51.

www.aacrjournals.org Clin Cancer Res; 2018 OF11

Functional Ex Vivo Assay Reveals Homologous Recombination Deficiency in Breast Cancer Beyond BRCA Gene Defects

Titia G. Meijer, Nicole S. Verkaik, Anieta M. Sieuwerts, et al.

Clin Cancer Res Published OnlineFirst August 23, 2018.

Updated version Access the most recent version of this article at: doi:10.1158/1078-0432.CCR-18-0063

Supplementary Access the most recent supplemental material at: Material http://clincancerres.aacrjournals.org/content/suppl/2019/03/28/1078-0432.CCR-18-0063.DC1

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Subscriptions Department at [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/early/2018/10/22/1078-0432.CCR-18-0063. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.