CheckMate 648 (Japanese subgroup): reconstructed Kaplan–Meier curves
First-line nivolumab plus ipilimumab or chemotherapy versus chemotherapy alone in advanced esophageal squamous cell carcinoma: a Japanese subgroup analysis of open-label, phase 3 trial (CheckMate 648/ONO-4538-50). Curves rebuilt as patient-level data (Guyot 2012) and checked against the statistics printed on the figure.
| Publication | Esophagus. 2022 Nov 19;20(2):291–301. doi: 10.1007/s10388-022-00970-1 |
|---|---|
| DOI | 10.1007/s10388-022-00970-1 |
| Registry | NCT03143153 |
| Phase | Phase 3 |
| Conditions | Various Advanced Cancer |
| Interventions | Nivolumab, Ipilimumab, Cisplatin, Fluorouracil |
| Sponsor | Bristol-Myers Squibb |
| Enrolment | 970 |
| Source licence | Figure published under CC BY 4.0 |
Overall survival — Patients with tumor-cell PD-L1 >=1% (Japanese subpopulation)
From figure panel a · validated against the values printed on the figure · curve source: vision model
Reconstructed from the published curve. Median: NIVO+IPI 20.3 months; Chemo 8.9 months. Survival at 12 months: NIVO+IPI 70%; Chemo 37%. Hazard ratio NIVO+IPI vs Chemo 0.46 (95% CI 0.29–0.71).
| Arm | N | Events | Median (Months) |
|---|---|---|---|
| NIVO+IPI | 66 | 37 | 20.3 |
| Chemo | 65 | 50 | 8.9 |
- NIVO+IPI vs Chemo: reconstructed HR 0.46 (95% CI 0.29–0.71)
Landmark survival (reconstructed)
| Time | NIVO+IPI | Chemo |
|---|---|---|
| 6 mo | 86.0% | 75.0% |
| 12 mo | 70.3% | 36.7% |
| 18 mo | 53.3% | 18.5% |
| 24 mo | 45.0% | 18.5% |
| 36 mo | 34.7% | — |
Checks against the published figure
- NIVO+IPI median: printed 20.2, reconstructed 20.3 pass
- Chemo median: printed 9, reconstructed 8.9 pass
- hazard ratio: printed 0.46, reconstructed 0.456 pass
Overall survival — All randomized patients (Japanese subpopulation)
From figure panel b · validated against the values printed on the figure · curve source: vision model
Reconstructed from the published curve. Median: NIVO+IPI 17.1 months; Chemo 10.8 months. Survival at 12 months: NIVO+IPI 62%; Chemo 47%. Hazard ratio NIVO+IPI vs Chemo 0.69 (95% CI 0.51–0.93).
| Arm | N | Events | Median (Months) |
|---|---|---|---|
| NIVO+IPI | 131 | 81 | 17.1 |
| Chemo | 137 | 96 | 10.8 |
- NIVO+IPI vs Chemo: reconstructed HR 0.69 (95% CI 0.51–0.93)
Landmark survival (reconstructed)
| Time | NIVO+IPI | Chemo |
|---|---|---|
| 6 mo | 86.8% | 78.1% |
| 12 mo | 62.3% | 46.9% |
| 18 mo | 48.2% | 29.9% |
| 24 mo | 36.3% | 24.7% |
| 36 mo | 29.0% | — |
Checks against the published figure
- NIVO+IPI median: printed 17.6, reconstructed 17.1 pass
- Chemo median: printed 11, reconstructed 10.8 pass
- hazard ratio: printed 0.68, reconstructed 0.687 pass
Progression-free survival per BICR — Patients with tumor-cell PD-L1 >=1% (Japanese subpopulation)
From figure panel c · validated against the values printed on the figure · curve source: vision model
Reconstructed from the published curve. Median: NIVO+IPI 5.4 months; Chemo 4.1 months. Survival at 12 months: NIVO+IPI 29%; Chemo 11%. Hazard ratio NIVO+IPI vs Chemo 0.81 (95% CI 0.52–1.28).
| Arm | N | Events | Median (Months) |
|---|---|---|---|
| NIVO+IPI | 66 | 45 | 5.4 |
| Chemo | 65 | 37 | 4.1 |
- NIVO+IPI vs Chemo: reconstructed HR 0.81 (95% CI 0.52–1.28)
Landmark survival (reconstructed)
| Time | NIVO+IPI | Chemo |
|---|---|---|
| 6 mo | 37.1% | 30.0% |
| 12 mo | 28.7% | 11.4% |
| 18 mo | 23.9% | — |
| 24 mo | 23.9% | — |
Checks against the published figure
- NIVO+IPI median: printed 5.4, reconstructed 5.38 pass
- Chemo median: printed 4.2, reconstructed 4.12 pass
- hazard ratio: printed 0.84, reconstructed 0.811 pass
Progression-free survival per BICR — All randomized patients (Japanese subpopulation)
From figure panel d · validated against the values printed on the figure · curve source: vision model
Reconstructed from the published curve. Median: NIVO+IPI 4.1 months; Chemo 4.1 months. Survival at 12 months: NIVO+IPI 24%; Chemo 17%. Hazard ratio NIVO+IPI vs Chemo 1.08 (95% CI 0.80–1.46).
| Arm | N | Events | Median (Months) |
|---|---|---|---|
| NIVO+IPI | 131 | 96 | 4.1 |
| Chemo | 137 | 80 | 4.1 |
- NIVO+IPI vs Chemo: reconstructed HR 1.08 (95% CI 0.80–1.46)
Landmark survival (reconstructed)
| Time | NIVO+IPI | Chemo |
|---|---|---|
| 6 mo | 33.8% | 35.1% |
| 12 mo | 24.0% | 17.2% |
| 18 mo | 17.7% | 14.3% |
| 24 mo | 17.7% | 14.3% |
Checks against the published figure
- NIVO+IPI median: printed 4.2, reconstructed 4.1 pass
- Chemo median: printed 4.3, reconstructed 4.14 pass
- hazard ratio: printed 1.16, reconstructed 1.08 pass
Duration of response per BICR — Responders among all randomized patients (Japanese subpopulation)
From figure panel f · validated against the values printed on the figure · curve source: vision model
Reconstructed from the published curve. Median: NIVO+IPI 10.0 months; Chemo 6.9 months. Survival at 12 months: NIVO+IPI 50%; Chemo 29%. Hazard ratio NIVO+IPI vs Chemo 0.73 (95% CI 0.39–1.37).
| Arm | N | Events | Median (Months) |
|---|---|---|---|
| NIVO+IPI | 47 | 23 | 10.0 |
| Chemo | 33 | 18 | 6.9 |
- NIVO+IPI vs Chemo: reconstructed HR 0.73 (95% CI 0.39–1.37)
Landmark survival (reconstructed)
| Time | NIVO+IPI | Chemo |
|---|---|---|
| 6 mo | 69.0% | 50.1% |
| 12 mo | 49.7% | 29.5% |
| 18 mo | 40.4% | 29.5% |
| 24 mo | 40.4% | 29.5% |
Checks against the published figure
- NIVO+IPI median: printed 9.9, reconstructed 9.98 pass
- Chemo median: printed 6.9, reconstructed 6.86 pass
Overall survival — Patients with tumor-cell PD-L1 >=1% (Japanese subpopulation)
From figure panel a · validated against the values printed on the figure · curve source: vision model
Reconstructed from the published curve. Median: NIVO+Chemo 17.4 months; Chemo 9.2 months. Survival at 12 months: NIVO+Chemo 64%; Chemo 36%. Hazard ratio NIVO+Chemo vs Chemo 0.54 (95% CI 0.35–0.82).
| Arm | N | Events | Median (Months) |
|---|---|---|---|
| NIVO+Chemo | 62 | 39 | 17.4 |
| Chemo | 65 | 50 | 9.2 |
- NIVO+Chemo vs Chemo: reconstructed HR 0.54 (95% CI 0.35–0.82)
Landmark survival (reconstructed)
| Time | NIVO+Chemo | Chemo |
|---|---|---|
| 6 mo | 87.1% | 77.6% |
| 12 mo | 64.4% | 35.6% |
| 18 mo | 48.2% | 17.6% |
| 24 mo | 46.0% | 17.6% |
Checks against the published figure
- NIVO+Chemo median: printed 17.3, reconstructed 17.4 pass
- Chemo median: printed 9, reconstructed 9.25 pass
- hazard ratio: printed 0.53, reconstructed 0.54 pass
Overall survival — All randomized patients (Japanese subpopulation)
From figure panel b · validated against the values printed on the figure · curve source: vision model
Reconstructed from the published curve. Median: NIVO+Chemo 15.5 months; Chemo 10.9 months. Survival at 12 months: NIVO+Chemo 61%; Chemo 46%. Hazard ratio NIVO+Chemo vs Chemo 0.71 (95% CI 0.53–0.97).
| Arm | N | Events | Median (Months) |
|---|---|---|---|
| NIVO+Chemo | 126 | 77 | 15.5 |
| Chemo | 137 | 95 | 10.9 |
- NIVO+Chemo vs Chemo: reconstructed HR 0.71 (95% CI 0.53–0.97)
Landmark survival (reconstructed)
| Time | NIVO+Chemo | Chemo |
|---|---|---|
| 6 mo | 85.4% | 77.9% |
| 12 mo | 60.9% | 46.3% |
| 18 mo | 43.7% | 30.4% |
| 24 mo | 38.9% | 25.2% |
| 36 mo | 22.0% | — |
Checks against the published figure
- NIVO+Chemo median: printed 15.5, reconstructed 15.5 pass
- Chemo median: printed 11, reconstructed 10.9 pass
- hazard ratio: printed 0.73, reconstructed 0.714 pass
Progression-free survival per BICR — Patients with tumor-cell PD-L1 >=1% (Japanese subpopulation)
From figure panel c · validated against the values printed on the figure · curve source: vision model
Reconstructed from the published curve. Median: NIVO+Chemo 7.0 months; Chemo 3.9 months. Survival at 12 months: NIVO+Chemo 29%; Chemo 12%. Hazard ratio NIVO+Chemo vs Chemo 0.55 (95% CI 0.35–0.88).
| Arm | N | Events | Median (Months) |
|---|---|---|---|
| NIVO+Chemo | 62 | 46 | 7.0 |
| Chemo | 65 | 37 | 3.9 |
- NIVO+Chemo vs Chemo: reconstructed HR 0.55 (95% CI 0.35–0.88)
Landmark survival (reconstructed)
| Time | NIVO+Chemo | Chemo |
|---|---|---|
| 6 mo | 55.1% | 30.5% |
| 12 mo | 29.0% | 12.4% |
| 18 mo | 12.4% | — |
| 24 mo | 12.4% | — |
Checks against the published figure
- NIVO+Chemo median: printed 7, reconstructed 7.05 pass
- Chemo median: printed 4.2, reconstructed 3.88 pass
- hazard ratio: printed 0.56, reconstructed 0.554 pass
Progression-free survival per BICR — All randomized patients (Japanese subpopulation)
From figure panel d · validated against the values printed on the figure · curve source: vision model
Reconstructed from the published curve. Median: NIVO+Chemo 7.0 months; Chemo 4.1 months. Survival at 12 months: NIVO+Chemo 24%; Chemo 18%. Hazard ratio NIVO+Chemo vs Chemo 0.76 (95% CI 0.56–1.04).
| Arm | N | Events | Median (Months) |
|---|---|---|---|
| NIVO+Chemo | 126 | 95 | 7.0 |
| Chemo | 137 | 77 | 4.1 |
- NIVO+Chemo vs Chemo: reconstructed HR 0.76 (95% CI 0.56–1.04)
Landmark survival (reconstructed)
| Time | NIVO+Chemo | Chemo |
|---|---|---|
| 6 mo | 51.6% | 35.4% |
| 12 mo | 24.4% | 18.3% |
| 18 mo | 14.0% | 15.3% |
| 24 mo | 12.2% | 15.3% |
| 36 mo | 12.2% | — |
Checks against the published figure
- NIVO+Chemo median: printed 6.8, reconstructed 7.05 pass
- Chemo median: printed 4.3, reconstructed 4.09 pass
- hazard ratio: printed 0.76, reconstructed 0.762 pass
About these data
Pseudo-IPD reproduces the published curves and numbers at risk; the rows are not real patients. Cite the publication (DOI above) and TrialCurve. The source figure is © the authors (CC BY 4.0); the charts here are redrawn from the reconstruction.
Load these data in R or Python
# R
library(survival)
d <- read.csv("https://trialcurve.com/data/trials/checkmate-648-japanese-subgroup-1.csv")
survfit(Surv(time, event) ~ arm, data = d)
coxph(Surv(time, event) ~ arm, data = d)
# Python
import pandas as pd
from lifelines import KaplanMeierFitter
d = pd.read_csv("https://trialcurve.com/data/trials/checkmate-648-japanese-subgroup-1.csv")
for arm, g in d.groupby("arm"):
KaplanMeierFitter().fit(g.time, g.event, label=arm).plot_survival_function()
Suggested citation: Esophagus. 2022 Nov 19;20(2):291–301. doi: 10.1007/s10388-022-00970-1 doi:10.1007/s10388-022-00970-1. Reconstructed data: TrialCurve (2026), https://trialcurve.com/trials/checkmate-648-japanese-subgroup
Report a problem with these data
Related trials
- CheckMate 648: Duration of response (per BICR); Overall survival; Progression-free survival (per BICR)
- RATIONALE-306 (Japanese subgroup, 3-year): Overall survival; Progression-free survival (investigator assessment)
- KEYNOTE-590 (Japanese subgroup, extended follow-up): Overall survival; Progression-free survival
- NEOCRTEC5010: Disease-free survival; Overall survival
- NEOCRTEC5010 (long-term follow-up): Disease-free survival; Overall survival
- ATTRACTION-2 (Japanese subgroup): Overall survival; Progression-free survival
Need reconstructed data for other trials? Upload any Kaplan–Meier figure, or ask about the TrialCurve oncology library.
Reconstruct a figure Browse all trials