KEYNOTE-177
A Phase III Study of Pembrolizumab (MK-3475) vs. Chemotherapy in Microsatellite Instability-High (MSI-H) or Mismatch Repair Deficient (dMMR) Stage IV Colorectal Carcinoma (KEYNOTE-177)
Phase 3N = 307NCT02563002
- Population
- Adults aged 18 years or older with previously untreated MSI-H or dMMR metastatic colorectal cancer and ECOG performance status 0 or 1
- Design
- Open-label, randomised phase 3 trial at 193 academic medical centres and hospitals in 23 countries. Patients were randomised 1:1 (blocks of four) to pembrolizumab 200 mg every 3 weeks or investigator's choice of mFOLFOX6 or FOLFIRI every 2 weeks, with or without bevacizumab or weekly cetuximab. After progression, patients on chemotherapy could cross over to pembrolizumab for up to 35 cycles. Co-primary endpoints were progression-free survival and overall survival.
- Primary endpoint
- PFS, OS
- Arms
- Pembrolizumab (experimental, n = 153): Pembrolizumab 200 mg intravenously every 3 weeks
- Chemotherapy (control, n = 154): Investigator's choice of mFOLFOX6 or FOLFIRI every 2 weeks, with or without bevacizumab or cetuximab
- People
- André T, KEYNOTE-177 Investigators
PFSHR 0.60 (95% CI 0.45–0.80)Δ median PFS+8.3 monthsΔ ORR+10.7 pointsOSHR 0.74 (95% CI 0.53–1.03)PFSHR 0.59 (95% CI 0.45–0.79)Δ median PFS+8.3 months
Show the numbers
| Endpoint | Analysis | Pembrolizumab | Chemotherapy | Effect |
|---|---|---|---|---|
| PFSAll randomised patients (analysis population not stated in the abstract) | interimmedian follow-up 32.4 monthsAndré 2020 | median 16.5 months | median 8.2 months | HR 0.60 (95% CI 0.45–0.80)P=0.0002Second interim analysis. Median follow-up is from randomisation to data cutoff (range 24.0 to 48.3 months). Overall survival was still blinded at this analysis (66% of required events). |
| ORRAll randomised patients (analysis population not stated in the abstract)Overall response (complete or partial) by RECIST version 1.1 | interimmedian follow-up 32.4 monthsAndré 2020 | 43.8% | 33.1% | Among patients with an overall response, 83% in the pembrolizumab group and 35% in the chemotherapy group had ongoing responses at 24 months. |
| OS | finalmedian follow-up 44.5 monthsDiaz 2022 | — | median 36.7 months | HR 0.74 (95% CI 0.53–1.03)P=0.036Median overall survival was not reached (95% CI 49.2 to not reached) with pembrolizumab and 36.7 months (95% CI 27.6 to not reached) with chemotherapy; the upper bounds are not representable here. The prespecified alpha of 0.025 for superiority was not met. 93 patients (60%) crossed over from chemotherapy to anti-PD-1 or anti-PD-L1 therapy (56 to on-study pembrolizumab, 37 to off-study therapy). |
| PFS | finalmedian follow-up 44.5 monthsDiaz 2022 | median 16.5 months (5.4–38.1) | median 8.2 months (6.1–10.2) | HR 0.59 (95% CI 0.45–0.79)Updated progression-free survival analysis reported with the final overall survival analysis. |
| OSAll randomised patients (analysis population not stated in the abstract) | long term-updatemedian follow-up 73.3 monthsAndré 2025 | median 77.5 months; 54.8% at 5 y | median 36.7 months; 44.2% at 5 y | HR 0.73 (95% CI 0.53–0.99)Follow-up of more than 5 years (range 64.9 to 89.2 months). 57 patients (37.0%) assigned to chemotherapy crossed over to pembrolizumab per protocol and 39 (25.3%) received a PD-(L)1 inhibitor off protocol, an effective crossover rate of 62%. The abstract gives no p value for the overall survival comparison. |
| PFSAll randomised patients (analysis population not stated in the abstract) | long term-updatemedian follow-up 73.3 monthsAndré 2025 | median 16.5 months | median 8.2 months | HR 0.60 (95% CI 0.45–0.79) |
| DORPatients with an objective response | long term-updatemedian follow-up 73.3 monthsAndré 2025 | median 75.4 months | median 10.6 months | Ranges were 2.3+ to 80.1+ months with pembrolizumab and 2.8 to 71.5+ months with chemotherapy (+ denotes ongoing response); these are ranges, not confidence intervals. |
| Endpoint | Analysis | Pembrolizumab | Chemotherapy | Effect |
|---|---|---|---|---|
| PFSAll randomised patients (analysis population not stated in the abstract) | interimmedian follow-up 32.4 monthsAndré 2020 | median 16.5 months | median 8.2 months | HR 0.60 (95% CI 0.45–0.80)P=0.0002Second interim analysis. Median follow-up is from randomisation to data cutoff (range 24.0 to 48.3 months). Overall survival was still blinded at this analysis (66% of required events). |
| ORRAll randomised patients (analysis population not stated in the abstract)Overall response (complete or partial) by RECIST version 1.1 | interimmedian follow-up 32.4 monthsAndré 2020 | 43.8% | 33.1% | Among patients with an overall response, 83% in the pembrolizumab group and 35% in the chemotherapy group had ongoing responses at 24 months. |
| OS | finalmedian follow-up 44.5 monthsDiaz 2022 | — | median 36.7 months | HR 0.74 (95% CI 0.53–1.03)P=0.036Median overall survival was not reached (95% CI 49.2 to not reached) with pembrolizumab and 36.7 months (95% CI 27.6 to not reached) with chemotherapy; the upper bounds are not representable here. The prespecified alpha of 0.025 for superiority was not met. 93 patients (60%) crossed over from chemotherapy to anti-PD-1 or anti-PD-L1 therapy (56 to on-study pembrolizumab, 37 to off-study therapy). |
| PFS | finalmedian follow-up 44.5 monthsDiaz 2022 | median 16.5 months (5.4–38.1) | median 8.2 months (6.1–10.2) | HR 0.59 (95% CI 0.45–0.79)Updated progression-free survival analysis reported with the final overall survival analysis. |
| OSAll randomised patients (analysis population not stated in the abstract) | long term-updatemedian follow-up 73.3 monthsAndré 2025 | median 77.5 months; 54.8% at 5 y | median 36.7 months; 44.2% at 5 y | HR 0.73 (95% CI 0.53–0.99)Follow-up of more than 5 years (range 64.9 to 89.2 months). 57 patients (37.0%) assigned to chemotherapy crossed over to pembrolizumab per protocol and 39 (25.3%) received a PD-(L)1 inhibitor off protocol, an effective crossover rate of 62%. The abstract gives no p value for the overall survival comparison. |
| PFSAll randomised patients (analysis population not stated in the abstract) | long term-updatemedian follow-up 73.3 monthsAndré 2025 | median 16.5 months | median 8.2 months | HR 0.60 (95% CI 0.45–0.79) |
| DORPatients with an objective response | long term-updatemedian follow-up 73.3 monthsAndré 2025 | median 75.4 months | median 10.6 months | Ranges were 2.3+ to 80.1+ months with pembrolizumab and 2.8 to 71.5+ months with chemotherapy (+ denotes ongoing response); these are ranges, not confidence intervals. |
Data table
| Endpoint | HR | 95% CI |
|---|---|---|
| PFS | 0.6 | 0.45–0.8 |
| OS | 0.74 | 0.53–1.03 |
| PFS | 0.59 | 0.45–0.79 |
| OS | 0.73 | 0.53–0.99 |
| PFS | 0.6 | 0.45–0.79 |
- Pembrolizumab (experimental)
- 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 | Chemotherapy |
|---|---|---|
| 0 | 100.0% | 100.0% |
| 10 | 65.7% | 42.9% |
| 21 | 41.4% | 16.9% |
| 31 | 27.2% | 7.3% |
| 41 | 17.9% | 3.1% |
- Chemotherapy (control)
Median-only curves assume a constant hazard (exponential) and fit plateaus, as seen with some immunotherapies, poorly.
Data table (modelled values)
| Months | Chemotherapy |
|---|---|
| 0 | 100.0% |
| 23 | 64.8% |
| 46 | 41.9% |
| 69 | 27.2% |
| 92 | 17.6% |
Appears in
Sources
André T, Shiu KK, Kim TW, Jensen BV, Jensen LH, Punt C, et al. Pembrolizumab in Microsatellite-Instability-High Advanced Colorectal Cancer. N Engl J Med. 2020;383(23):2207-2218.
Diaz LA Jr, Shiu KK, Kim TW, Jensen BV, Jensen LH, Punt C, et al. Pembrolizumab versus chemotherapy for microsatellite instability-high or mismatch repair-deficient metastatic colorectal cancer (KEYNOTE-177): final analysis of a randomised, open-label, phase 3 study. Lancet Oncol. 2022;23(5):659-670.
André T, Shiu KK, Kim TW, Jensen BV, Jensen LH, Punt CJA, et al. Pembrolizumab versus chemotherapy in microsatellite instability-high or mismatch repair-deficient metastatic colorectal cancer: 5-year follow-up from the randomized phase III KEYNOTE-177 study. Ann Oncol. 2025;36(3):277-284.