POLO: Radiologic assessment of the impact of maintenance olaparib in patients (pts) with metastatic pancreatic cancer (mPaC).

Abstract

412 Background: The phase III POLO study (NCT02184195) demonstrated a benefit of maintenance olaparib over placebo in the radiologically assessed primary endpoint of progression-free survival (PFS) in pts with mPaC (median 7.4 vs 3.8 months [mo]; 12-mo rate 34% vs 15%). The impact of radiologic assessment of pancreatic lesions, which is considered challenging, was explored. Methods: Tumors were assessed using Response Evaluation Criteria in Solid Tumors version 1.1 by blinded independent central review (BICR) in pts with mPaC treated with maintenance olaparib or placebo. PFS was analyzed in subsets of pts based on various event criteria. Results: All 154 randomized pts had mPaC prior to chemotherapy, of whom 122 had disease in the pancreas at POLO baseline (BL); 34% (53/154) had pancreas-only target lesions (TL), 26% (40/154) also had ≥1 TL outside of the pancreas, and in 19% (29/154) pancreatic disease was recorded as non-TL. Sensitivity analyses were consistent with the primary PFS analysis (Table), including when all pancreas lesion assessments were discounted (median PFS 7.4 vs 4.7 mo; 12-mo rate 38% vs 22%). Of 53 pts with pancreas-only TLs at BL, 34 had disease progression (PD); in 20 pts this was not solely based on TL measurements (16 had new lesions; 4 had multiple-cause PD). Confirmed objective responses occurred during study maintenance treatment in 20% of olaparib pts (18/92) and 10% of placebo pts (6/62). In pts with pancreas-only TLs at BL there were 7 responses in the olaparib arm (1 complete response [CR], 6 partial responses [PR]) and 2 (2 PR) in the placebo arm. In pts who had ≥1 TL outside of the pancreas at BL there were 11 (1 CR, 10 PR) and 4 (4 PR) responses, respectively. Responses were generally durable irrespective of TL location. Conclusions: The significant PFS benefit with maintenance olaparib over placebo shown in the primary analysis was consistent across all sensitivity analyses and was not impacted by radiologic assessment of pancreatic TLs. Taken together, these findings suggest that contrary to historically held belief, primary pancreas TLs may be appropriate for inclusion as sites of RECIST-evaluable disease and for assessment of treatment outcome. Clinical trial information: NCT02184195. [Table: see text]