JCOG1113/FUGA-BT
Combination gemcitabine plus S-1 versus gemcitabine plus cisplatin for advanced/recurrent biliary tract cancer: the FUGA-BT (JCOG1113) randomized phase III clinical trial
Phase 3N = 354
- Population
- Chemotherapy-naive patients with recurrent or unresectable biliary tract cancer
- Design
- Randomised phase 3 non-inferiority trial in 33 institutions in Japan of gemcitabine plus S-1 (GS) versus gemcitabine plus cisplatin (GC) in chemotherapy-naive patients with recurrent or unresectable biliary tract cancer and ECOG performance status 0 or 1. The primary end point was overall survival; the calculated sample size was 350 (one-sided alpha 5%, power 80%, non-inferiority margin for the hazard ratio 1.155). 354 patients were enrolled between May 2013 and March 2016.
- Primary endpoint
- OS
- Arms
- Gemcitabine plus S-1 (experimental): Gemcitabine plus S-1 (doses not stated in the abstract)
- Gemcitabine plus cisplatin (control): Gemcitabine plus cisplatin (doses not stated in the abstract)
- People
- Morizane C
OSHR 0.945Δ median OS+1.7 monthsPFSHR 0.86 (95% CI 0.70–1.07)Δ median PFS+1 months
Show the numbers
| Endpoint | Analysis | Gemcitabine plus S-1 | Gemcitabine plus cisplatin | Effect |
|---|---|---|---|---|
| OSEnrolled patients (the abstract does not name the analysis population) | primaryMorizane 2019 | median 15.1 months | median 13.4 months | HR 0.945P=0.046 (for non-inferiority)The confidence interval of the hazard ratio is a 90% interval, 0.78 to 1.15, not a 95% interval. The P value is for non-inferiority of gemcitabine plus S-1; this was a non-inferiority design, not a superiority test. The abstract gives no confidence intervals for the medians. |
| PFSEnrolled patients (the abstract does not name the analysis population) | primaryMorizane 2019 | median 6.8 months | median 5.8 months | HR 0.86 (95% CI 0.70–1.07)Secondary end point; the abstract gives no P value. |
| Endpoint | Analysis | Gemcitabine plus S-1 | Gemcitabine plus cisplatin | Effect |
|---|---|---|---|---|
| OSEnrolled patients (the abstract does not name the analysis population) | primaryMorizane 2019 | median 15.1 months | median 13.4 months | HR 0.945P=0.046 (for non-inferiority)The confidence interval of the hazard ratio is a 90% interval, 0.78 to 1.15, not a 95% interval. The P value is for non-inferiority of gemcitabine plus S-1; this was a non-inferiority design, not a superiority test. The abstract gives no confidence intervals for the medians. |
| PFSEnrolled patients (the abstract does not name the analysis population) | primaryMorizane 2019 | median 6.8 months | median 5.8 months | HR 0.86 (95% CI 0.70–1.07)Secondary end point; the abstract gives no P value. |
Data table
| Endpoint | HR | 95% CI |
|---|---|---|
| PFS | 0.86 | 0.7–1.07 |
- Gemcitabine plus S-1 (experimental)
- Gemcitabine plus cisplatin (control)
Median-only curves assume a constant hazard (exponential) and fit plateaus, as seen with some immunotherapies, poorly.
Data table (modelled values)
| Months | Gemcitabine plus S-1 | Gemcitabine plus cisplatin |
|---|---|---|
| 0 | 100.0% | 100.0% |
| 9 | 66.2% | 62.8% |
| 19 | 41.8% | 37.4% |
| 28 | 27.7% | 23.5% |
| 38 | 17.5% | 14.0% |
- Gemcitabine plus S-1 (experimental)
- Gemcitabine plus cisplatin (control)
Median-only curves assume a constant hazard (exponential) and fit plateaus, as seen with some immunotherapies, poorly.
Data table (modelled values)
| Months | Gemcitabine plus S-1 | Gemcitabine plus cisplatin |
|---|---|---|
| 0 | 100.0% | 100.0% |
| 4 | 66.5% | 62.0% |
| 9 | 40.0% | 34.1% |
| 13 | 26.6% | 21.1% |
| 17 | 17.7% | 13.1% |