435P - Phase II single arm study of durvalumab and olaparib (D+O) in heavily pretreated triple-negative breast cancer (TNBC) with or without germline BRCA mutation (gBRCAm)

Background

PARP inhibitors (PARPis) induce STING-cGAS activation and increase tumor mutational load in TNBC preclinical models, making TNBC sensitive to immune checkpoint blockade. The MEDIOLA phase II study demonstrated a 63.3% overall response rate (ORR) defined as CR+PR with D+O in PARPi naïve gBRCAm BC (1 median number of prior therapies [range 0-4]). We evaluated D+O activity in heavily pretreated TNBC.

Methods

Key eligibility criteria include metastatic TNBC, adequate performance status and organ function. Patients (pts) with prior use of D or PARPi, and active CNS or leptomeningeal disease were not eligible. No limit on the number of prior systemic therapies. For germline BRCA wild type (gBRCAwt), Simon’s 2-stage design with a target of 30% ORR with α=0.1 and β=0.1 was used. If ≥ 2 responses in the first stage (16 pts), 9 additional pts were planned to enroll. 5 ≥ of 25 responses warrant a further later stage study. For gBRCAm, a pilot single stage design was used with a stopping rule. If 1≥ response in the first 5 pts, another 5 pts were planned to enroll. RP2D determined in our earlier phase I study; D 1500 mg iv q4w and O 300 mg BID po (28-day cycle) was administered until PD or intolerable toxicity. The primary endpoint was ORR by RECIST v1.1 (CT q2mo). Secondary objectives were safety by CTCAE v4.03, clinical benefit rate (CBR) defined by CR+PR+SD≥4mo, and PFS.

Results

14 of 15 pts were RECIST evaluable (11 gBRCAwt and 3 gBRCAm). The median number of prior therapies are 3 (range 0-9). A majority of pts (78.6%; 11/14) had ≥ 3 prior systemic therapies and 64.3% (9/14) had previous platinum. Of note, 4 pts had either HR+/HER2-BC (n=3) or HR+/HER2+BC (n=1) at diagnosis which converted to TNBC at recurrence. ORR was 28.5% (3/3 gBRCAm [3 PR] and 1/11 gBRCAwt [PR for 10.3 mo]). One pt with gBRCAm remains on the study with PR >4 yr (96% reduction at best). CBR was 42.9% (3/3 gBRCAm and 3/11 gBRCAwt). Median PFS was 3.7 mo (IQR 1.9-10.3). There were no new safety signals. Accrual is ongoing.

Conclusions

D+O exhibited modest clinical activity in heavily pretreated TNBC pts with or without gBRCAm. Comprehensive biomarker and pharmacodynamic studies are ongoing for the exceptional responder with gBRCAm.

Clinical trial identification

NCT02484404 Release date: June 29, 2015.

Editorial acknowledgement

No editorial assistance

Legal entity responsible for the study

J-M. Lee.

Funding

Intramural Research Program of the NCI, CCR (grant number: #ZIA BC011525; to J-M. Lee) Durvalumab and olaparib were supplied to the CCR, NCI under a Cooperative Research and Development Agreement between the CCR/NCI and AstraZeneca. Research funding was also provided by AstraZeneca to the institution.

Disclosure

All authors have declared no conflicts of interest.