Background
Phase III RCT (GOG 213). 673 patients with platinum-sensitive recurrent ovarian, fallopian tube, or primary peritoneal cancer (first recurrence, platinum-free interval ≥6 months). 2×2 factorial design: (1) secondary cytoreductive surgery vs no surgery; and (2) bevacizumab 15 mg/kg q3wk + carboplatin AUC 5 + paclitaxel 175 mg/m² (6 cycles) then bevacizumab maintenance until progression vs the same chemotherapy without bevacizumab. Primary OS analysis reported here (Coleman NEJM 2019) supersedes earlier PFS publications.
Interventions and follow up
Arm A: Carboplatin + paclitaxel (6 cycles), then observatio
Arm B: Bevacizumab 15 mg/kg + carboplatin + paclitaxel (6 cycles), then bevacizumab maintenance
Primary endpoint: Overall survival
mFollow up: 49.9 month
Arm B: Bevacizumab 15 mg/kg + carboplatin + paclitaxel (6 cycles), then bevacizumab maintenance
Primary endpoint: Overall survival
mFollow up: 49.9 month
Results
Median OS: 42.2 months (bev) vs 37.3 months (no bev), HR 0.82 (95% CI 0.68–0.996), P=.045
Median PFS: 13.8 months (bev) vs 10.4 months (no bev), HR 0.61, P<.001
ORR: 78.5% (bev) vs 59.1% (no bev)
Median PFS: 13.8 months (bev) vs 10.4 months (no bev), HR 0.61, P<.001
ORR: 78.5% (bev) vs 59.1% (no bev)
Adverse events
Main adverse events: Grade ≥3 hypertension: 24.9% (bev) vs 0.8% (no bev). Grade ≥3 proteinuria: 8.1% vs 0%. Thromboembolic events: 7.7% (bev) vs 2.9% (no bev). GI perforation: 1.5% (bev) vs 0.9% (no bev). Fistula: 1.5% vs 0.6%. Treatment discontinuation due to AEs: 22% (bev) vs 4% (no bev).
Conclusions
Adding bevacizumab to carboplatin + paclitaxel followed by bevacizumab maintenance significantly improved OS (42.2 vs 37.3 months, HR 0.82) and PFS in platinum-sensitive recurrent ovarian cancer — the first phase III trial to demonstrate an OS benefit for bevacizumab in ovarian cancer.
Key Limitations
Key Limitations: The OS benefit, while statistically significant, was modest (4.9 months median difference; HR 0.82). Toxicity (hypertension, thromboembolic events) is significant and requires monitoring. The bevacizumab dose (15 mg/kg q3wk) is higher than used in some other trials. Not all patients can tolerate bevacizumab long-term. HRD/BRCA status was not prospectively stratified — PARP inhibitor combinations are increasingly preferred for BRCA-mutated patients.
Clinical Context
GOG 213 provided the first OS evidence for bevacizumab in recurrent ovarian cancer and established carboplatin + paclitaxel + bevacizumab as a standard option for platinum-sensitive recurrence. In BRCA-mutated patients, PARP inhibitor combinations (e.g., cediranib + olaparib, SOLO3) may be preferred. In non-BRCA/non-HRd patients, bevacizumab-containing regimens remain an important option. The surgery component of GOG 213 showed secondary cytoreduction did not improve OS (published separately).
References
References: Coleman RL et al, N Engl J Med 2019 (GOG 213)