Olaparib (O) in advanced triple negative breast cancer (aTNBC) patients (pts) with BRCA1/2 promoter methylation: GEICAM/2015-06 study (COMETA-Breast)

1093 Background: Epigenetic silencing by aberrant BRCA1/2 promoters’ methylation ( BRCA-meth) can be responsible for a dysfunctional BRCA protein. BRCA-meth occurs in 15-57% of TNBC pts. BRCA-meth breast cancer (BC) display pathologic features and genetic profiles like germline BRCA1/2-mutated (g BRCA1/2m) carriers. O is a PARP inhibitor approved for treating g BRCAm HER2-negative advanced BC (aBC) pts. These are the results of a phase II study assessing the O efficacy in aTNBC pts with BRCA1/2-meth (NCT03205761). Methods: We included aTNBC pretreated pts (≥1 line) with centrally confirmed somatic BRCA1/2-meth (in most recent lesion(s) available) and no g BRCA1/2m. O 300mg b.i.d. was administered. BRCA1/2 CpG island was considered methylated if value was ≥25%. Overall response rate (ORR) (complete response [CR] + partial response [PR]) according to RECIST 1.1) was the primary objective. Thirty-one evaluable pts were required based on an optimal 2-stage Simon’s design (α error=0.05, 1-β error=80%), estimating an ORR increase of 24%, with a null hypothesis of 30%. Whole exome/transcriptome and 105-genes NGS assays (Tempus xE & xF) were performed on pretreatment tumor and sequential ctDNA of a long-term responder. Results: Eleven pts received ≥1 cycle of O (ITT population). Median age was 51 years (37-64), and 8 pts were postmenopausal. All M0 pts at diagnosis had neo-/adjuvant chemotherapy (CT). Median O exposure duration was 8 (1-88) weeks and relative dose-intensity was 97% (76-100), with any dose modifications in 6 pts. ORR was 9% (1 PR/11; 95% CI, 0.2-41). Clinical benefit rate (CR + PR + stable disease of any duration) was 36% (4/11; 95% CI, 11-69). Most pts (n=9) discontinued O due to BC progression. Median PFS was 2 months (95% CI, 1-4), and median OS was 9 months (95% CI, 1-14). One pt with confirmed PR was on O for >20 months. AEs were reported in 3 pts in line with O safety profile. No AE led to O discontinuation and no SAEs were reported. The statistical assumptions for the 1 st Simon’s model stage were not met (≥ 4/12 pts with CR or PR), so the recruitment did not proceed onto the 2 nd stage. Preliminary NGS analysis on the long-term responder identified alterations on tumor (TP53, BCL11A, MYC) and on-treatment ctDNA (PMS2). Conclusions: In this proof-of-concept study, O did not show clinically nor statistically significant antitumor activity in pretreated aTNBC pts with BRCA1/2-meth . Clinical trial information: NCT03205761 . [Table: see text]