COMMIT / NRG-GI004 / SWOG S1610 Trial

Design - Randomized Phase III NCI-sponsored cooperative-group trial (NRG-GI004/SWOG S1610): atezolizumab + mFOLFOX6 + bevacizumab vs single-agent atezolizumab in first-line dMMR/MSI-H metastatic colorectal cancer. Study closed early after meeting its primary endpoint (ASCO GI 2026).

Progression-free survival (primary) - Median PFS 24.5 months with the triplet vs 5.3 months with atezolizumab monotherapy (HR 0.439, 95% CI 0.23-0.84, P=0.0103). ORR 80.6% vs 46%; early progression 2.8% vs 32.4%.

Overall survival - OS data were immature at the interim analysis and OS was not a primary endpoint; follow-up is ongoing after early closure.

Safety - Higher Grade >=3 AEs with the triplet, driven by the chemotherapy backbone (neutropenia, peripheral neuropathy). Atezolizumab monotherapy had lower toxicity but a much higher early-progression rate (32.4% vs 2.8%).

Regulatory / sponsor - Investigational - not FDA approved in this indication; pembrolizumab (KEYNOTE-177) remains the label-approved 1L standard for dMMR/MSI-H mCRC. Lead sponsor: National Cancer Institute, conducted through NRG Oncology and SWOG.

Compiled and reviewed by the KOL Pulse research team, led by Brian Shields, Founder, KOL Pulse. Last updated August 19, 2026.

Progression-Free Survival (PFS) — Primary Endpoint

Median PFS was 24.5 months with atezolizumab + mFOLFOX6 + bevacizumab vs. 5.3 months with atezolizumab monotherapy (HR 0.439, 95% CI 0.23-0.84, P=0.0103). ORR: 80.6% vs. 46%. Early progression: 2.8% vs. 32.4% — triplet eliminates the ~30% primary resistance seen with atezolizumab alone. Study closed early after meeting primary endpoint.

Overall Survival (OS)

Overall survival data immature at interim analysis; not a primary endpoint. Trial closed early after primary PFS endpoint met; OS follow-up ongoing.

Safety & Tolerability

Higher Grade ≥3 AEs with the triplet driven by chemotherapy backbone (neutropenia, peripheral neuropathy). Atezolizumab monotherapy had lower toxicity but a much higher early progression rate (32.4% vs. 2.8%).

Clinical Implications

⚠️ Does not supplant KEYNOTE-177 but addresses primary resistance. Addresses the ~40% primary resistance to PD-1 monotherapy seen in KEYNOTE-177. Triplet improves PFS and ORR significantly vs. atezolizumab monotherapy — but without head-to-head data vs. pembrolizumab monotherapy. Potential option for fit dMMR/MSI-H patients prioritizing deeper responses and reduced early-progression risk. Does not supplant KEYNOTE-177 as the label-approved standard.

What is the COMMIT trial?

COMMIT (NRG-GI004/SWOG S1610; NCT02997228) is an NCI-sponsored randomized Phase III trial in first-line dMMR/MSI-H metastatic colorectal cancer. It compared atezolizumab combined with mFOLFOX6 chemotherapy and bevacizumab against single-agent atezolizumab. The study closed early after meeting its primary endpoint, with results presented at ASCO GI 2026.

What did COMMIT show?

Median PFS was 24.5 months with atezolizumab + mFOLFOX6 + bevacizumab vs 5.3 months with atezolizumab monotherapy (HR 0.439, 95% CI 0.23-0.84, P=0.0103). ORR was 80.6% vs 46%, and early progression fell from 32.4% with monotherapy to 2.8% with the triplet - largely eliminating the primary resistance seen with checkpoint-inhibitor monotherapy.

Is the COMMIT regimen FDA approved for MSI-H colorectal cancer?

No. The atezolizumab + mFOLFOX6 + bevacizumab triplet is investigational in this indication, and atezolizumab is not FDA approved for MSI-H colorectal cancer. Pembrolizumab monotherapy, based on KEYNOTE-177, remains the label-approved first-line standard for dMMR/MSI-H metastatic colorectal cancer.

What is the safety profile of the COMMIT triplet?

Grade >=3 adverse events were higher with the triplet, driven by the chemotherapy backbone - mainly neutropenia and peripheral neuropathy. Atezolizumab monotherapy was less toxic but carried a much higher early-progression rate (32.4% vs 2.8%), so the added chemotherapy toxicity was offset by a dramatic drop in early progression.

Why is COMMIT important?

COMMIT addresses the roughly 40% primary-resistance rate seen with PD-1 monotherapy in KEYNOTE-177 by pairing a checkpoint inhibitor with chemotherapy and bevacizumab up front. It significantly improves PFS and ORR versus atezolizumab alone, though without head-to-head data against pembrolizumab monotherapy, so it does not supplant KEYNOTE-177 as the label-approved standard - it is a potential option for fit patients prioritizing deeper responses and lower early-progression risk.