Abstract
The AGO-OVAR16 study was designed to test the efficacy, safety, and tolerability of pazopanib maintenance after first-line chemotherapy in patients with newly diagnosed advanced ovarian cancer (AOC).
Nine hundred and forty patients with histologically confirmed AOC, International Federation of Gynecology and Obstetrics (FIGO) stage II–IV, were randomized in a 1:1 ratio to receive either 800 mg pazopanib once daily or placebo for up to 24 months, unless there was disease progression, toxicity, withdrawal of consent, or death. The primary endpoint (investigator-assessed progression-free survival [PFS]) was met and previously reported. The results of final analyses of overall survival (OS) are reported here.
A third OS interim analysis showed futility and led to study closure and a final OS analysis after last patient last visit. At the time of the final OS analysis, 494 (89.7% of the planned 551) events had occurred. No difference was observed in OS between pazopanib and placebo. The hazard ratio (HR) was 0.960 (95% confidence interval [CI]: 0.805–1.145), and the median OS from randomization was 59.1 months in pazopanib and 64.0 months in placebo arms. For the East Asian patients, similar to the first three interim OS analyses, a numerical negative trend was observed favoring placebo (HR, 1.332; 95% CI: 0.863–2.054). Exploratory analyses showed a trend for a longer time to first subsequent anti-cancer therapy or death with pazopanib over placebo (HR, 0.829; 95% CI: 0.713–0.965), with a median estimate of 19.0 and 14.5 months, respectively. No new safety signals were observed.
Although pazopanib prolonged PFS, this was not associated with improvement in median OS.
Clinical trial information.ClinicalTrials.gov: NCT00866697.
•In primary analysis, pazopanib given as maintenance therapy for 2 years prolonged PFS significantly, in women with AOC.•The current final OS analysis did not show significant survival benefit.•Although OS represents the most practical endpoint for evaluating superiority, PFS is an acceptable endpoint.•The OS benefit could be masked by several confounding factors.•Lack of a significant OS benefit in this study may be attributed to the long post-progression survival in this setting.