Methylation of all BRCA1 copies predicts response to the PARP inhibitor rucaparib in ovarian carcinoma

Abstract

Accurately identifying patients with high-grade serous ovarian carcinoma (HGSOC) who respond to poly(ADP-ribose) polymerase inhibitor (PARPi) therapy is of great clinical importance. Here we show that quantitative BRCA1 methylation analysis provides new insight into PARPi response in preclinical models and ovarian cancer patients. The response of 12 HGSOC patient-derived xenografts (PDX) to the PARPi rucaparib was assessed, with variable dose-dependent responses observed in chemo-naive BRCA1/2-mutated PDX, and no responses in PDX lacking DNA repair pathway defects. Among BRCA1-methylated PDX, silencing of all BRCA1 copies predicts rucaparib response, whilst heterozygous methylation is associated with resistance. Analysis of 21 BRCA1-methylated platinum-sensitive recurrent HGSOC (ARIEL2 Part 1 trial) confirmed that homozygous or hemizygous BRCA1 methylation predicts rucaparib clinical response, and that methylation loss can occur after exposure to chemotherapy. Accordingly, quantitative BRCA1 methylation analysis in a pre-treatment biopsy could allow identification of patients most likely to benefit, and facilitate tailoring of PARPi therapy.

Fig. 1

Genomic profiling of 12 HGSOC PDX. a IHC staining of PAX8, WT1, and p53 of passage one (T1) PDX tumors. Loss of p53 expression was observed for PDX #54 with a frameshift TP53 mutation (p.G199fs*8), #11 with a nonsense TP53 mutation (p.E198*), and #80 with a splice site TP53 mutation (IVS6-1G>T). b Select genomic events detected by the Foundation Medicine T5a test, BROCA assay, and BRCA1 promoter methylation testing. The Foundation Medicine T5a test was performed on PDX samples, except for case #48, where it was performed on patient HGSOC material. T5a test results and BROCA v4 assay results for PDX #11, #13, #27, #29, #56, and #62 were previously published; BROCA v6 was performed for all other PDX. c RNA-seq gene expression for genes with detected mutations or copy number changes. RNA-seq was performed on baseline patient HGSOC material samples. RNA-seq was also performed on PDX #169 and #201 samples, to verify expression levels observed in the matched HGSOC with suboptimal sample quality due to either low neoplastic cellularity or poor RNA quality (#80 inadequate quality); rep—RNA-seq library replicate

Fig. 2

Cisplatin and rucaparib responses in BRCA1/2 mutant and HR wild-type HGSOC PDX. Rucaparib and cisplatin response in a PDX #56 (BRCA1 mutant); b PDX #19 (BRCA2 mutant); c PDX #201 (HR-DNA repair gene wild-type); and d PDX #29 (HR-DNA repair gene wild-type). Recipient mice bearing PDX were randomized to treatment with vehicle or rucaparib, at the dose shown. PDX were harvested at a tumor volume of 600–700 mm³. Cisplatin response data for PDX #19, #56, and #29 were previously published. See Table 1 and Supplementary Data 2 for median TTH and p-values for survival comparison. Mean tumor volume (mm³) ± 95% CI (hashed lines are representing individual mice) and corresponding Kaplan–Meier survival analysis. Censored events are represented by crosses on Kaplan–Meier plot; n = individual mice

Fig. 3

BRCA1 promoter methylation in HGSOC PDX and rucaparib response. a A diagram of two modes of BRCA1 promoter methylation observed in four PDX #11, #62, #48, and #169. Homozygous methylation status was assigned when % of methylation was close to 100%, therefore all observed copies were methylated. Heterozygous methylation status was assigned when both, methylated and unmethylated, copies were observed. b BRCA1 methylation in four HGSOC PDX (#62, #48, #169, #11) assessed by MS-ddPCR (mean ± 95% CI); n = 2–3 mice for each treatment and PDX model. c RAD51 foci formation 4 h after 10 Gy irradiation was observed in PDX #169 with heterozygous BRCA1 methylation and not in PDX #11 and PDX #62 with homozygous BRCA1 methylation. d Quantification of ex vivo γH2AX and RAD51 foci formation in geminin-positive cells 4 h after 10 Gy irradiation (mean ± 95% CI). γH2AX foci are observed at the sites of DNA damage, and RAD51 foci are observed at the sites of HR pathway repair; n = 12 (four fields of view from three independent experiments) for each treatment and PDX model. Untreated and irradiated cells were compared by multiple t-tests for γH2AX and RAD51 foci formation. ***p < 0.001; ns not significant. e Responses to cisplatin and rucaparib in vivo treatment observed in chemo-naive PDX #62 with homozygous BRCA1 methylation. f RECIST 1.1 measurements of three monitored tumor lesions in patient #11, with homozygous methylation of BRCA1, treated with rucaparib. g CT scans of the two largest monitored lesions prior to and during rucaparib treatment of the patient #11. h, i Responses to cisplatin and rucaparib in vivo treatment observed in PDX #48 and #169 with heterozygous BRCA1 methylation. Recipient mice bearing PDX were randomized to treatment with vehicle or rucaparib, at the dose shown. PDX were harvested at a tumor volume of 600–700 mm³ (see Table 1 and Supplementary Data 2 for median TTH and p-values for survival comparison). Mean tumor volume (mm³) ± 95% CI (hashed lines are representing individual mice) and corresponding Kaplan–Meier survival analysis. Censored events are represented by crosses on Kaplan–Meier plot; n = individual mice. HOM homozygous, HET heterozygous

Fig. 4

Assessment of HR deficiency and rucaparib sensitivity in BRCA1-methylated cell lines. a RAD51 foci formation assessed 6 h post exposure to 10 Gy irradiation in HR-competent OC cell line (OV90), immortalized fallopian tube cell line (FT282), OC cell line with heterozygous BRCA1 methylation (OVCAR8), OVCAR8 derivative with RAD51C KO, and HGSOC cell line with homozygous BRCA1 methylation (WEHICS62). b Quantification of RAD51 foci formation in EdU-positive cells for OV90, FT282, OVCAR8, OVCAR8 derivative with RAD51C KO, and WEHICS62. RAD51 foci formation ability was compared to the untreated controls. At least 170 EdU-positive cells were counted for each cell type and treatment (multiple fields of view from three independent experiments). Mean ± SEM. c Colony formation assay assessing rucaparib response at 14 days in HR-competent OC cell line (PEO4), at 10 days in OC cell line with heterozygous BRCA1 methylation (OVCAR8) and HR-deficient OC cell line (PEO1), and at 21 days in HGSOC cell line with homozygous BRCA1 methylation (WEHICS62); n = 3 independent experiments. Mean ± SEM. d In vitro rucaparib response assessed by cell count proliferation time course assay using IncuCyte ZOOM of OC cell lines OVCAR8, OVCAR8 derivative with RAD51C KO and WEHICS62. One of three similar independent experiments shown. Mean ± SEM; * denotes p < 0.05 for post-IR WEHICS62 % comparison with irradiated OVCAR8 and OV90 counterparts. IR irradiated, Cntrl untreated control, ND not detected

Fig. 5

Homozygous BRCA1 methylation and rucaparib response in the ARIEL2 Part 1 trial. a Kaplan–Meier progression-free survival analysis of patients with HGSOC with homozygous BRCA1 methylation in the pre-treatment tumor biopsy, which was of high confidence based on adequate neoplastic cellularity (homozygous BRCA1 methylation (high confidence)), compared with patients with HGSOC in which there had ever been any other evidence of BRCA1 methylation (ever any BRCA1 methylation), compared with all other patients in the ARIEL2 Part 1 trial without any BRCA1 methylation (BRCA1/2 mutant vs. BRCA1/2 wild-type non-BRCA1-methylated subgroups). Shaded areas represent 95% CI for homozygous BRCA1 methylation (high confidence) and ever any BRCA1 methylation, other groups. b Genome-wide LOH % assessed in the pre-treatment biopsies compared across subgroups: homozygous BRCA1 methylation (high-confidence), (n = 6); ever any BRCA1 methylation, (n = 6); BRCA1/2 mutant, (n = 27); and BRCA1/2 wild-type non-BRCA1-methylated, (n = 96). Boxplot—median, whiskers—95% CI, dots represent individual samples. c Best percentage change from baseline in sum of longest diameter of target lesions according to RECIST 1.1 compared across subgroups: homozygous BRCA1 methylation (high confidence), (n = 6); ever any BRCA1 methylation, (n = 15); BRCA1/2 mutant, (n = 40); and BRCA1/2 wild-type non-BRCA1-methylated, (n = 143). Boxplot—median, whiskers—95% CI, dots represent individual samples. d Best percentage change from baseline in sum of longest diameter of target lesions according to RECIST 1.1 in the BRCA wild-type LOH-high subgroup of patients by BRCA1 methylation status. Each bar represents percentage change from baseline in sum of the longest diameter of target lesions for an individual patient according to RECIST 1.1. In some patients, although best percentage change of >30% was observed, the response was not investigator confirmed and thus classified as stable disease (SD) or progressive disease (PD). PR partial response, PD progressive disease, SD stable disease, CR complete response