AGO-OVAR-3
A randomized clinical trial of cisplatin/paclitaxel versus carboplatin/paclitaxel as first-line treatment of ovarian cancer
Phase 3N = 798
- Population
- Patients with advanced ovarian cancer, FIGO stage IIB-IV
- Design
- Randomised phase III non-inferiority trial. 798 patients with FIGO stage IIB-IV ovarian cancer were randomly assigned to six courses at 3-week intervals of paclitaxel plus cisplatin (PT) or paclitaxel plus carboplatin (TC). Primary endpoint the proportion of patients without progression at 2 years; secondary endpoints toxicity, response, quality of life, overall and progression-free survival.
- Primary endpoint
- PFS
- Arms
- Paclitaxel plus cisplatin (PT) (control): Paclitaxel plus cisplatin, six courses at 3-week intervals (doses not given in the abstract)
- Paclitaxel plus carboplatin (TC) (experimental): Paclitaxel plus carboplatin, six courses at 3-week intervals (doses not given in the abstract)
- People
- du Bois A, Arbeitsgemeinschaft Gynäkologische Onkologie Ovarian Cancer Study Group
Δ 24-month PFS−2.5 pointsΔ median PFS−1.9 monthsΔ median OS−0.8 months
Show the numbers
| Endpoint | Analysis | Paclitaxel plus carboplatin (TC) | Paclitaxel plus cisplatin (PT) | Effect |
|---|---|---|---|---|
| PFSRandomised patients (N=798); analysis population not stated in the abstractProportion of patients without progression at 2 years | primarydu 2003 | 37.5% at 2 y | 40% at 2 y | Primary endpoint, a proportion at 24 months rather than a time-to-event median. Difference 2.5%, one-sided 95% CI minus infinity to 8.2%. |
| PFSRandomised patients (N=798); analysis population not stated in the abstract | primarydu 2003 | median 17.2 months (15.2–19.3) | median 19.1 months (16.7–21.5) | Secondary endpoint. The abstract states that the difference was not statistically significant and gives no p-value. |
| OSRandomised patients (N=798); analysis population not stated in the abstract | primarydu 2003 | median 43.3 months (37.2–47.8) | median 44.1 months (40.2–49.4) | Secondary endpoint. The abstract states that the difference was not statistically significant and gives no p-value. |
| Endpoint | Analysis | Paclitaxel plus carboplatin (TC) | Paclitaxel plus cisplatin (PT) | Effect |
|---|---|---|---|---|
| PFSRandomised patients (N=798); analysis population not stated in the abstractProportion of patients without progression at 2 years | primarydu 2003 | 37.5% at 2 y | 40% at 2 y | Primary endpoint, a proportion at 24 months rather than a time-to-event median. Difference 2.5%, one-sided 95% CI minus infinity to 8.2%. |
| PFSRandomised patients (N=798); analysis population not stated in the abstract | primarydu 2003 | median 17.2 months (15.2–19.3) | median 19.1 months (16.7–21.5) | Secondary endpoint. The abstract states that the difference was not statistically significant and gives no p-value. |
| OSRandomised patients (N=798); analysis population not stated in the abstract | primarydu 2003 | median 43.3 months (37.2–47.8) | median 44.1 months (40.2–49.4) | Secondary endpoint. The abstract states that the difference was not statistically significant and gives no p-value. |
- Paclitaxel plus carboplatin (TC) (experimental)
- Paclitaxel plus cisplatin (PT) (control)
Median-only curves assume a constant hazard (exponential) and fit plateaus, as seen with some immunotherapies, poorly.
Data table (modelled values)
| Months | Paclitaxel plus carboplatin (TC) | Paclitaxel plus cisplatin (PT) |
|---|---|---|
| 0 | 100.0% | 100.0% |
| 12 | 61.7% | 64.7% |
| 24 | 38.0% | 41.9% |
| 36 | 23.4% | 27.1% |
| 48 | 14.5% | 17.5% |
- Paclitaxel plus carboplatin (TC) (experimental)
- Paclitaxel plus cisplatin (PT) (control)
Median-only curves assume a constant hazard (exponential) and fit plateaus, as seen with some immunotherapies, poorly.
Data table (modelled values)
| Months | Paclitaxel plus carboplatin (TC) | Paclitaxel plus cisplatin (PT) |
|---|---|---|
| 0 | 100.0% | 100.0% |
| 28 | 63.9% | 64.4% |
| 55 | 41.5% | 42.1% |
| 83 | 26.5% | 27.1% |
| 110 | 17.2% | 17.7% |