Article Text
Abstract
Introduction Improved overall survival (OS) with PARP inhibitors maintenance therapy in platinum-sensitive relapsed ovarian cancer (PSROC) was not determined. We performed treatment-free survival (TFS) with quality-adjusted OS analyses to measure their effects.
Methods We pooled available data for TFS analysis from three trials (Study 19, SOLO2, and ARIEL3). OS was divided into time on protocol treatment exposure (T), time to subsequent treatment initiation or death (TFS), and time after the first subsequent therapy or death (REL). TFS with quality-adjusted OS analyses were calculated by multiplying mean time in each health state by its assigned utility (quality-adjusted OS = ut × T + TFS + urel × REL) for two cohorts: all (BRCAm, BRCAwt, or unknown) or BRCAm patients, the area under each KM curve estimated using restricted mean time with threshold utility analyses.
Results Restricted mean treatment duration was longer with PARP inhibitors than with placebo (all patients: 16.3 vs. 7.4 months, difference, 8.9 months, 95%CI 8.1–9.7; BRCAm: 17.1 vs. 8.1 months, 9.0, 8.1–9.9). Mean TFS was longer with PARP inhibitors (7.9 months) than with placebo (4.3 months; difference, 3.6, 2.0–5.1) among BRCAm patients. REL was longer with placebo (all patients: 11.7 vs. 20.5 months, -8.8, -10.1 to -7.6; BRCAm: 8.9 vs. 19.7 months, -10.8, -12.5 to -9.0). PARP inhibitors provided more quality-adjusted OS than placebo with a wider range of utility-weight values for BRCAm patients (-6.0 to +7.0 months) rather than all patients (-7.0 to +5.0 months).
Conclusion/Implications BRCAm patients rather than all-patients population benefited from PARP inhibitors maintenance therapy in PSROC.