GOG-240
Cisplatin-paclitaxel or topotecan-paclitaxel, each with or without bevacizumab, in stage IVB, recurrent or persistent cervical carcinoma
Phase 3N = 452NCT00803062
- Population
- Metastatic, persistent or recurrent cervical carcinoma; GOG performance status 0 or 1; measurable disease; no previous chemotherapy for recurrence
- Design
- Randomised, open-label phase 3 trial with a 2-by-2 factorial design: chemotherapy (cisplatin 50 mg/m2 plus paclitaxel 135 or 175 mg/m2, or topotecan 0.75 mg/m2 on days 1 to 3 plus paclitaxel 175 mg/m2) with or without bevacizumab 15 mg/kg, in 21-day cycles until progression, unacceptable toxic effects or complete response. Stratified by GOG performance status, previous radiosensitising platinum-based chemotherapy and disease status. 452 patients enrolled at centres in the USA, Canada and Spain (Lancet 2017 abstract).
- Primary endpoint
- OS
- Arms
- Chemotherapy alone (both regimens combined) (control, n = 225): Cisplatin plus paclitaxel, or topotecan plus paclitaxel, every 21 days
- Chemotherapy plus bevacizumab (both regimens combined) (experimental, n = 227): The same two chemotherapy regimens plus bevacizumab 15 mg/kg every 21 days
- Cisplatin-paclitaxel backbone (with or without bevacizumab) (control, n = 229): Cisplatin 50 mg/m2 plus paclitaxel 135 or 175 mg/m2
- Topotecan-paclitaxel backbone (with or without bevacizumab) (experimental, n = 223): Topotecan 0.75 mg/m2 days 1 to 3 plus paclitaxel 175 mg/m2
- People
- Tewari KS, senior author Monk BJ, Gynecologic Oncology Group
OSHR 0.71Δ median OS+3.7 monthsΔ ORR+12 pointsPFSHR 0.67 (95% CI 0.54–0.82)Δ median PFS+2.3 monthsOSHR 0.77 (95% CI 0.62–0.95)
Show the numbers
| Endpoint | Analysis | Chemotherapy plus bevacizumab (both regimens combined) / Topotecan-paclitaxel backbone (with or without bevacizumab) | Chemotherapy alone (both regimens combined) / Cisplatin-paclitaxel backbone (with or without bevacizumab) | Effect |
|---|---|---|---|---|
| OSRandomised patients (N = 452)Overall survival, primary endpoint; a 30% reduction in the hazard ratio for death was prespecified as clinically important | interimTewari 2014 | median 17 months | median 13.3 months | HR 0.71P=0.004 (one-sided)The abstract gives a 98% confidence interval of 0.54 to 0.95 for the hazard ratio, so no ci95 is entered. The PMC full text (PMC4010094) states that the bevacizumab analyses come from the second data freeze (12 December 2012; 271 deaths, median follow-up 20.8 months); the 2017 Lancet abstract calls this the second interim analysis. |
| ORRRandomised patients (N = 452) | interimTewari 2014 | 48% | 36% | P=0.008Secondary endpoint; P value as given in the paper (two-sided test, per the full text). |
| PFSRandomised patients (N = 452) | interimTewari 2014 | median 8.2 months | median 5.9 months | HR 0.67 (95% CI 0.54–0.82)Secondary endpoint; second data freeze (median follow-up 20.8 months, 271 deaths). Read in the open-access full text (PMC4010094). |
| OS | finalTewari 2017 | median 16.8 months | median 13.3 months | HR 0.77 (95% CI 0.62–0.95)P=0.007Prespecified final analysis (cutoff 450 patients with 346 deaths; 348 deaths by 7 March 2014). Overall survival was analysed in the intention-to-treat population. |
| OSRandomised patients (N = 452) | finalTewari 2023 | median 13.8 months | median 16.3 months | HR 1.12 (95% CI 0.91–1.38)P=0.28Backbone comparison with and without bevacizumab pooled within each backbone, as the paper presents it. The abstract does not state which arm is the reference for the hazard ratio. It also reports the pairs without bevacizumab (15 vs 12 months; HR 1.10, 95% CI 0.82-1.48) and with bevacizumab (17.5 vs 16.2 months; HR 1.16, 95% CI 0.86-1.56); those arms are not entered separately. |
| Endpoint | Analysis | Chemotherapy plus bevacizumab (both regimens combined) / Topotecan-paclitaxel backbone (with or without bevacizumab) | Chemotherapy alone (both regimens combined) / Cisplatin-paclitaxel backbone (with or without bevacizumab) | Effect |
|---|---|---|---|---|
| OSRandomised patients (N = 452)Overall survival, primary endpoint; a 30% reduction in the hazard ratio for death was prespecified as clinically important | interimTewari 2014 | median 17 months | median 13.3 months | HR 0.71P=0.004 (one-sided)The abstract gives a 98% confidence interval of 0.54 to 0.95 for the hazard ratio, so no ci95 is entered. The PMC full text (PMC4010094) states that the bevacizumab analyses come from the second data freeze (12 December 2012; 271 deaths, median follow-up 20.8 months); the 2017 Lancet abstract calls this the second interim analysis. |
| ORRRandomised patients (N = 452) | interimTewari 2014 | 48% | 36% | P=0.008Secondary endpoint; P value as given in the paper (two-sided test, per the full text). |
| PFSRandomised patients (N = 452) | interimTewari 2014 | median 8.2 months | median 5.9 months | HR 0.67 (95% CI 0.54–0.82)Secondary endpoint; second data freeze (median follow-up 20.8 months, 271 deaths). Read in the open-access full text (PMC4010094). |
| OS | finalTewari 2017 | median 16.8 months | median 13.3 months | HR 0.77 (95% CI 0.62–0.95)P=0.007Prespecified final analysis (cutoff 450 patients with 346 deaths; 348 deaths by 7 March 2014). Overall survival was analysed in the intention-to-treat population. |
| OSRandomised patients (N = 452) | finalTewari 2023 | median 13.8 months | median 16.3 months | HR 1.12 (95% CI 0.91–1.38)P=0.28Backbone comparison with and without bevacizumab pooled within each backbone, as the paper presents it. The abstract does not state which arm is the reference for the hazard ratio. It also reports the pairs without bevacizumab (15 vs 12 months; HR 1.10, 95% CI 0.82-1.48) and with bevacizumab (17.5 vs 16.2 months; HR 1.16, 95% CI 0.86-1.56); those arms are not entered separately. |
Data table
| Endpoint | HR | 95% CI |
|---|---|---|
| PFS | 0.67 | 0.54–0.82 |
| OS | 0.77 | 0.62–0.95 |
| OS | 1.12 | 0.91–1.38 |
- Chemotherapy alone (both regimens combined) (control)
- Chemotherapy plus bevacizumab (both regimens combined) (experimental)
Median-only curves assume a constant hazard (exponential) and fit plateaus, as seen with some immunotherapies, poorly.
Data table (modelled values)
| Months | Chemotherapy alone (both regimens combined) | Chemotherapy plus bevacizumab (both regimens combined) |
|---|---|---|
| 0 | 100.0% | 100.0% |
| 11 | 56.4% | 63.9% |
| 21 | 33.5% | 42.5% |
| 32 | 18.9% | 27.1% |
| 43 | 10.6% | 17.3% |
- Chemotherapy alone (both regimens combined) (control)
- Chemotherapy plus bevacizumab (both regimens combined) (experimental)
Median-only curves assume a constant hazard (exponential) and fit plateaus, as seen with some immunotherapies, poorly.
Data table (modelled values)
| Months | Chemotherapy alone (both regimens combined) | Chemotherapy plus bevacizumab (both regimens combined) |
|---|---|---|
| 0 | 100.0% | 100.0% |
| 5 | 55.6% | 65.5% |
| 10 | 30.9% | 42.9% |
| 15 | 17.2% | 28.1% |
| 21 | 8.5% | 16.9% |
Appears in
Sources
Tewari KS, Sill MW, Long HJ 3rd, Penson RT, Huang H, Ramondetta LM, et al. Improved survival with bevacizumab in advanced cervical cancer. N Engl J Med. 2014;370(8):734-43.
Tewari KS, Sill MW, Penson RT, Huang H, Ramondetta LM, Landrum LM, et al. Bevacizumab for advanced cervical cancer: final overall survival and adverse event analysis of a randomised, controlled, open-label, phase 3 trial (Gynecologic Oncology Group 240). Lancet. 2017;390(10103):1654-1663.
Tewari KS, Sill MW, Birrer MJ, Penson RT, Huang H, Moore DH, et al. Final survival analysis of topotecan and paclitaxel for first-line treatment of advanced cervical cancer: An NRG oncology randomized study. Gynecol Oncol. 2023;171:141-150.