KEYNOTE-590
Pembrolizumab plus chemotherapy versus chemotherapy alone for first-line treatment of advanced oesophageal cancer: a randomised, placebo-controlled, phase 3 study
Phase 3N = 749NCT03189719
- Population
- Previously untreated, locally advanced unresectable or metastatic oesophageal cancer or Siewert type 1 gastro-oesophageal junction cancer, regardless of PD-L1 status; ECOG performance status 0 to 1. Histology per the 5-year report: squamous-cell carcinoma 274 vs 274, adenocarcinoma 99 vs 102 (pembrolizumab vs placebo groups)
- Design
- Randomised (1:1), placebo-controlled, double-blind phase 3 study across 168 centres in 26 countries. Pembrolizumab 200 mg or placebo, plus 5-fluorouracil and cisplatin, every 3 weeks for up to 35 cycles; randomisation stratified by region, histology and performance status. 749 patients enrolled between July 2017 and June 2019.
- Primary endpoint
- OS, PFS
- Arms
- Pembrolizumab plus chemotherapy (experimental, n = 373): Pembrolizumab 200 mg plus 5-fluorouracil and cisplatin, every 3 weeks for up to 35 cycles
- Placebo plus chemotherapy (control, n = 376): Placebo plus 5-fluorouracil and cisplatin, every 3 weeks for up to 35 cycles
- People
- Sun JM
OSHR 0.57 (95% CI 0.43–0.75)Δ median OS+5.1 monthsOSHR 0.72 (95% CI 0.60–0.88)Δ median OS+2.8 monthsOSHR 0.62 (95% CI 0.49–0.78)Δ median OS+4.1 months
Show the numbers
| Endpoint | Analysis | Pembrolizumab plus chemotherapy | Placebo plus chemotherapy | Effect |
|---|---|---|---|---|
| OSSquamous-cell carcinoma with PD-L1 CPS of 10 or more | interimmedian follow-up 22.6 monthsSun 2021 | median 13.9 months | median 8.8 months | HR 0.57 (95% CI 0.43–0.75)P<0.0001First interim analysis; primary endpoint population. |
| OSSquamous-cell carcinoma | interimmedian follow-up 22.6 monthsSun 2021 | median 12.6 months | median 9.8 months | HR 0.72 (95% CI 0.60–0.88)P=0.0006First interim analysis. |
| OSPD-L1 CPS of 10 or more (both histologies) | interimmedian follow-up 22.6 monthsSun 2021 | median 13.5 months | median 9.4 months | HR 0.62 (95% CI 0.49–0.78)P<0.0001First interim analysis. |
| OSAll randomised patients | interimmedian follow-up 22.6 monthsSun 2021 | median 12.4 months | median 9.8 months | HR 0.73 (95% CI 0.62–0.86)P<0.0001First interim analysis. Grade 3 or higher treatment-related adverse events: 72% (266 patients) with pembrolizumab plus chemotherapy vs 68% (250 patients) with placebo plus chemotherapy. |
| PFSSquamous-cell carcinoma | interimmedian follow-up 22.6 monthsSun 2021 | median 6.3 months | median 5.8 months | HR 0.65 (95% CI 0.54–0.78)P<0.0001First interim analysis. |
| PFSPD-L1 CPS of 10 or more (both histologies) | interimmedian follow-up 22.6 monthsSun 2021 | median 7.5 months | median 5.5 months | HR 0.51 (95% CI 0.41–0.65)P<0.0001First interim analysis. |
| PFSAll randomised patients | interimmedian follow-up 22.6 monthsSun 2021 | median 6.3 months | median 5.8 months | HR 0.65 (95% CI 0.55–0.76)P<0.0001First interim analysis. |
| OSAll randomised patients (749) | long term-updatemedian follow-up 58.8 monthsMetges 2025 | median 12.3 months; 10.6% at 5 y | median 9.8 months; 3% at 5 y | HR 0.72 (95% CI 0.62–0.84)Extended follow-up. Median PFS 6.3 vs 5.8 months (HR 0.64, 95% CI 0.54 to 0.75); the 5-year PFS rate was 5.5% with pembrolizumab plus chemotherapy and not reached with placebo plus chemotherapy. Grade 3 or worse treatment-related adverse events: 71.9% vs 67.6%. |
| Endpoint | Analysis | Pembrolizumab plus chemotherapy | Placebo plus chemotherapy | Effect |
|---|---|---|---|---|
| OSSquamous-cell carcinoma with PD-L1 CPS of 10 or more | interimmedian follow-up 22.6 monthsSun 2021 | median 13.9 months | median 8.8 months | HR 0.57 (95% CI 0.43–0.75)P<0.0001First interim analysis; primary endpoint population. |
| OSSquamous-cell carcinoma | interimmedian follow-up 22.6 monthsSun 2021 | median 12.6 months | median 9.8 months | HR 0.72 (95% CI 0.60–0.88)P=0.0006First interim analysis. |
| OSPD-L1 CPS of 10 or more (both histologies) | interimmedian follow-up 22.6 monthsSun 2021 | median 13.5 months | median 9.4 months | HR 0.62 (95% CI 0.49–0.78)P<0.0001First interim analysis. |
| OSAll randomised patients | interimmedian follow-up 22.6 monthsSun 2021 | median 12.4 months | median 9.8 months | HR 0.73 (95% CI 0.62–0.86)P<0.0001First interim analysis. Grade 3 or higher treatment-related adverse events: 72% (266 patients) with pembrolizumab plus chemotherapy vs 68% (250 patients) with placebo plus chemotherapy. |
| PFSSquamous-cell carcinoma | interimmedian follow-up 22.6 monthsSun 2021 | median 6.3 months | median 5.8 months | HR 0.65 (95% CI 0.54–0.78)P<0.0001First interim analysis. |
| PFSPD-L1 CPS of 10 or more (both histologies) | interimmedian follow-up 22.6 monthsSun 2021 | median 7.5 months | median 5.5 months | HR 0.51 (95% CI 0.41–0.65)P<0.0001First interim analysis. |
| PFSAll randomised patients | interimmedian follow-up 22.6 monthsSun 2021 | median 6.3 months | median 5.8 months | HR 0.65 (95% CI 0.55–0.76)P<0.0001First interim analysis. |
| OSAll randomised patients (749) | long term-updatemedian follow-up 58.8 monthsMetges 2025 | median 12.3 months; 10.6% at 5 y | median 9.8 months; 3% at 5 y | HR 0.72 (95% CI 0.62–0.84)Extended follow-up. Median PFS 6.3 vs 5.8 months (HR 0.64, 95% CI 0.54 to 0.75); the 5-year PFS rate was 5.5% with pembrolizumab plus chemotherapy and not reached with placebo plus chemotherapy. Grade 3 or worse treatment-related adverse events: 71.9% vs 67.6%. |
Data table
| Endpoint | HR | 95% CI |
|---|---|---|
| OS | 0.57 | 0.43–0.75 |
| OS | 0.72 | 0.6–0.88 |
| OS | 0.62 | 0.49–0.78 |
| OS | 0.73 | 0.62–0.86 |
| PFS | 0.65 | 0.54–0.78 |
| PFS | 0.51 | 0.41–0.65 |
| PFS | 0.65 | 0.55–0.76 |
| OS | 0.72 | 0.62–0.84 |
- Pembrolizumab plus chemotherapy (experimental)
- Placebo plus chemotherapy (control)
Median-only curves assume a constant hazard (exponential) and fit plateaus, as seen with some immunotherapies, poorly.
Data table (modelled values)
| Months | Pembrolizumab plus chemotherapy | Placebo plus chemotherapy |
|---|---|---|
| 0 | 100.0% | 100.0% |
| 9 | 63.8% | 49.2% |
| 17 | 42.8% | 26.2% |
| 26 | 27.3% | 12.9% |
| 35 | 17.5% | 6.3% |
- Pembrolizumab plus chemotherapy (experimental)
- Placebo plus chemotherapy (control)
Median-only curves assume a constant hazard (exponential) and fit plateaus, as seen with some immunotherapies, poorly.
Data table (modelled values)
| Months | Pembrolizumab plus chemotherapy | Placebo plus chemotherapy |
|---|---|---|
| 0 | 100.0% | 100.0% |
| 8 | 64.4% | 56.8% |
| 16 | 41.5% | 32.2% |
| 24 | 26.7% | 18.3% |
| 32 | 17.2% | 10.4% |
Appears in
Sources
Sun JM, Shen L, Shah MA, Enzinger P, Adenis A, Doi T, et al. Pembrolizumab plus chemotherapy versus chemotherapy alone for first-line treatment of advanced oesophageal cancer (KEYNOTE-590): a randomised, placebo-controlled, phase 3 study. Lancet. 2021;398(10302):759-771.
Metges JP, Kato K, Sun JM, Shen L, Enzinger PC, Adenis A, et al. Pembrolizumab plus chemotherapy versus chemotherapy for advanced esophageal cancer: 5-year extended follow-up for the randomized phase III KEYNOTE-590 study. ESMO Open. 2025;10(12):105854.