Background
Phase III randomized open-label trial. 303 patients with previously untreated (first-line) dMMR/MSI-H unresectable or metastatic colorectal cancer, ECOG PS 0–1. Three-arm design comparing nivolumab + ipilimumab, nivolumab monotherapy, and investigator-choice chemotherapy. Primary analysis reports the nivolumab + ipilimumab vs chemotherapy comparison.
Interventions and follow up
Arm A: Nivolumab 240 mg IV q2wk + ipilimumab 1 mg/kg IV q6wk until progression or unacceptable toxicity (ipilimumab up to 2 years)
Arm B: Investigator's choice chemotherapy: FOLFOX4 or FOLFIRI ± bevacizumab or cetuximab until progression or unacceptable toxicity
Primary endpoint: Progression-free survival (PFS) — nivolumab + ipilimumab vs chemotherapy
Median follow-up: 24.3 months
Arm B: Investigator's choice chemotherapy: FOLFOX4 or FOLFIRI ± bevacizumab or cetuximab until progression or unacceptable toxicity
Primary endpoint: Progression-free survival (PFS) — nivolumab + ipilimumab vs chemotherapy
Median follow-up: 24.3 months
Results
mPFS: not reached vs 5.9 months, HR 0.21 (95% CI 0.13–0.35), P<.001
ORR: 71% vs 37%
24-month PFS rate: 72% vs 14%
OS: immature at primary analysis; trend favoring nivolumab + ipilimumab
ORR: 71% vs 37%
24-month PFS rate: 72% vs 14%
OS: immature at primary analysis; trend favoring nivolumab + ipilimumab
Adverse events
Grade ≥3 treatment-related AEs: 23% (nivolumab + ipilimumab) vs 48% (chemotherapy)
Most common grade ≥3 (nivo+ipi): diarrhea/colitis 4%, hepatitis 4%, lipase increased 3%; immune-mediated AEs any grade 47%
Discontinuation: 11% (nivo+ipi) vs 19% (chemotherapy)
Most common grade ≥3 (nivo+ipi): diarrhea/colitis 4%, hepatitis 4%, lipase increased 3%; immune-mediated AEs any grade 47%
Discontinuation: 11% (nivo+ipi) vs 19% (chemotherapy)
Conclusions
Nivolumab + ipilimumab produced a dramatic improvement in first-line PFS vs chemotherapy in dMMR/MSI-H mCRC (HR 0.21, P<.001), with 72% progression-free at 24 months vs 14%. With ORR 71% and substantially lower grade ≥3 toxicity than chemotherapy (23% vs 48%), the trial establishes nivo+ipi as a preferred first-line therapy for dMMR/MSI-H mCRC.
Key Limitations
OS data immature at primary analysis — long-term survival benefit not yet confirmed. Three-arm design with sequential reporting means the nivolumab-alone comparison is reported separately. Exclusively dMMR/MSI-H population (∼5% of mCRC) limits population reach. Chemotherapy backbone and targeted agent choice varied across the control arm, introducing heterogeneity. Patients with BRAF V600E mutations — with poorer prognosis — may have driven some benefit.
Clinical Context
CheckMate 8HW is the first phase III trial to demonstrate superiority of dual checkpoint blockade over chemotherapy in first-line dMMR/MSI-H mCRC, building on KEYNOTE-177 (pembrolizumab vs chemo, HR 0.60) with a markedly better HR of 0.21. FDA approval of nivolumab + ipilimumab for first-line dMMR/MSI-H mCRC followed. ESMO increasingly favors dual checkpoint blockade over single-agent anti-PD-1 as a preferred first-line choice, pending OS maturation.