Time without symptoms or toxicity (TWiST) in patients with newly diagnosed advanced ovarian cancer receiving maintenance olaparib or placebo plus bevacizumab: Analysis of PAOLA-1/ENGOT-ov25 phase III trial.

5562 Background: In the Phase III PAOLA-1/ENGOT-ov25 trial (NCT02477644), maintenance olaparib plus bevacizumab (bev) provided a significant progression-free survival (PFS) benefit, compared with placebo plus bev, in patients (pts) with newly diagnosed advanced ovarian cancer in response after platinum-based chemotherapy (Ray-Coquard et al. NEJM 2019). A subgroup analysis revealed a substantial PFS benefit in HRD-positive (including BRCA1/BRCA2 mutation) pts (median PFS 37.2 vs 17.7 months) leading to US and EU labels for this combination. We analyzed TWiST in PAOLA-1, using several definitions of toxicity (TOX) and by molecular subgroups. Methods: Pts were randomized 2:1 to maintenance olaparib (300 mg bid) plus bev (15 mg/kg, Day 1, q3w) or placebo plus bev, stratified by first-line treatment outcome and tumor BRCAm status. TWiST is defined as PFS minus time with TOX after randomization and before disease progression or censoring for progression, and was a prespecified exploratory endpoint. Definitions of significant symptoms or TOX were explored using the following approaches: 1) all grade ≥2 adverse events (AEs); and 2) all grade ≥2 AEs selected to be correlated to olaparib (fatigue, nausea, vomiting, and anemia). Area under PFS curve was split into TOX:TWIST ratio and compared between the two arms. Results: Between May 7, 2015 and August 31, 2017, 806 eligible pts were randomly assigned to olaparib plus bev (n = 537) or placebo plus bev (n = 269), with a median duration of treatment with olaparib of 17.3 months (range 0.03–33.0) for olaparib plus bev arm and 15.6 months (range 0.07–26.2) for placebo plus bev at data cutoff for primary endpoint analysis. In the Intent-to-treat population, median duration of TWiST for all grade ≥2 AEs for olaparib plus bev vs placebo plus bev arms was 14.1 months (95% confidence interval [CI] 12.5–16.1) vs 7.7 months (95% CI 5.9–9.1), respectively; and 21.9 months (95% CI 20.2–22.5) vs 16.6 months (95% CI 14.6–18.0) considering only grade ≥2 AEs linked to olaparib. The difference was particularly significant within the HRD-positive subgroup where median duration of TWiST was 24.4 months (95% CI 19.3–not evaluable [NE]) vs 7.4 months (5.8–11.2) for TOX linked to all grade ≥2 AEs, and 36.6 months (95% CI 31.9–NE) vs 17.4 months (15.1–19.4) linked to olaparib AEs only for the olaparib plus bev and placebo plus bev arms, respectively. Conclusions: The substantial PFS benefit provided by maintenance olaparib plus bev vs placebo plus bev in pts with newly diagnosed advanced ovarian cancer was supported by significant TWiST benefit. TWiST analyses, including all grade ≥2 symptoms or toxicity related to treatment, confirmed the clinically meaningful benefit from the combination, notably in the HRD-positive subgroup. Clinical trial information: NCT02477644.