Quality of life (QoL) in patients with recurrent ovarian clear cell carcinoma (rOCCC) receiving durvalumab (D) versus physician's choice chemotherapy (PCC) in the randomized phase II MOCCA/APGOT-OV2/GCGS-OV3 trial.

e17564 Background: In the phase II MOCCA/ APGOT-OV2/ GCGS-OV3 trial, D was not superior to PCC in prolonging progression-free survival in patients with rOCCC. Fewer all-grade and high-grade adverse events were observed for patients treated with D compared with PCC. We analysed health-related QoL by treatment arm. Methods: In this multicentre, open-label, randomised phase II trial, eligible patients were randomly assigned (2:1), using dynamic block randomization with block size of 6, and stratification by Eastern Cooperative Oncology Group performance status, to receive D (1500mg on day 1, in 28-day cycles) or PCC until disease progression, intolerable toxicity or withdrawal of consent. Patients treated with PCC were able to receive any agent deemed appropriate by the Investigator, however the addition of anti-angiogenic drugs was prohibited. Patients with disease progression on PCC were allowed to crossover to D. Patients completed European Organisation for Research and Treatment of Cancer (EORTC) QLQ-C30 and QLQ-OV28 QoL questionnaires at baseline and every 4 weeks until end of study treatment. The minimal important difference for clinically relevant change was fixed at 10 points. Linear mixed-effects models were used to estimate the treatment effect on QoL outcomes, accounting for the intra-subject correlation in QoL scores at repeated visits from treatment cycles 2 to 6, while adjusting for baseline QoL scores (cycle 1 day 1) and time of visit. We included all patients whose QoL scores were available at both baseline and at least one visit from cycles 2 to 6. Results: 47 patients were randomly assigned to treatment with D (31 patients) or PCC (16 patients). Compliance to QoL questionnaires was 81.6 % (D) and 87.5% (PCC) at baseline. QoL data was available for 22.6-81.6% (D) and 31.3-87.5% (PCC) of patients over time. Longitudinal analysis of QoL data by domain revealed no significant differences in global health status, functional scales or most of the symptom scales based on EORTC QLQ-C30 or QLQ-OV28 between patients treated with D compared with PCC, except for the Constipation symptom scale on QLQ-C30 which showed an estimated mean difference of -20.8 points (95% confidence interval (CI), -31.5 to -10.2; p = 0.001) and the Hair loss symptom scale on QLQ-OV28 which showed an estimated mean difference of -19.6 points (95% CI -36.6 to -2.6; p = 0.025). Conclusions: D and PCC did not lead to significant differences in patient-reported global health status or functional scales. However, PCC was associated with increased patient-reported constipation and hair loss compared with D in rOCCC. Clinical trial information: NCT03405454 .