Atezolizumab Improves Overall Survival in Late-Relapsing Recurrent Ovarian Cancer
By: Lisa Astor
Posted: Friday, October 9, 2026
Adding atezolizumab to bevacizumab and platinum-based chemotherapy lengthened median overall survival, a secondary endpoint, by about 5 months in patients with late-relapsing recurrent ovarian cancer, according to the final analysis of the phase III ATALANTE/ENGOT-ov29 trial published in JCO Oncology Advances.
Checkpoint inhibition has reshaped care in endometrial and cervical cancers, but PD-L1 inhibitors added to standard therapy have frequently failed to show benefit in recurrent ovarian cancer studies. In ATALANTE/ENGOT-ov29, the dual primary endpoints of progression-free survival in the intention-to-treat (ITT) and PD-L1–positive populations missed the prespecified significance threshold (ITT hazard ratio [HR] = 0.83; 95% confidence interval [CI] = 0.69–0.99; two-sided P = .041). Follow-up continued, however, for the secondary endpoint of overall survival.
The double-blind trial enrolled 614 patients with nonmucinous epithelial ovarian cancer in first or second relapse more than 6 months after their last platinum therapy. Patients were randomly assigned 2:1 to receive bevacizumab plus six cycles of an investigator-chosen platinum doublet, along with either atezolizumab (n = 410) or placebo (n = 204) for up to 24 months. They were stratified by platinum-free interval, PD-L1 status, and chemotherapy regimen.
At the October 2023 cutoff, median follow-up reached 60.1 months and 447 patients (73%) had died. Median overall survival was 35.8 months (95% CI = 32.9–41.0) with atezolizumab vs 30.6 months (95% CI = 27.8–33.2) with standard therapy (HR = 0.83; 95% CI = 0.69–0.98), and the survival curves grew farther apart over time. Estimated 5-year overall survival rates were 27% with atezolizumab and 17 without. PD-L1–positive tumors showed no amplified benefit (HR = 0.84; 95% CI = 0.60–1.17), although median overall survival in that group was 43.0 with the combination regimen vs 33.7 months without atezolizumab. Treatment effect varied by prior chemotherapy lines and platinum-free interval (interaction P < .05).
Time to first subsequent therapy or death also favored atezolizumab (median, 14.4 vs 12.4 months; HR = 0.83), as did time to second subsequent therapy or death (HR = 0.83). Overall, 82% of patients went on to further treatment, and close to two-thirds were rechallenged with platinum.
The authors, led by Jean-Emmanuel Kurtz, MD, PhD, of Hôpitaux Universitaires de Strasbourg and GINECO, described the survival gain as clinically meaningful, while noting the exploratory nature of the analysis.
Disclosure: The study was supported by F. Hoffmann-La Roche, Ltd.
