In the phase III COMMIT trial, a regimen combining atezolizumab plus bevacizumab and standard chemotherapy significantly improved progression-free survival and response rates over atezolizumab monotherapy in patients with previously untreated mismatch repair deficient (dMMR) or microsatellite instability–high (MSI-H) metastatic colorectal cancer.1
“The COMMIT trial supports FOLFOX [leucovorin, fluorouracil, and oxaliplatin] and bevacizumab plus atezolizumab as a potential treatment option for selected patients with dMMR/MSI-H metastatic colorectal cancer,” said Caio Max Sao Pedro Rocha Lima, MD, of Atrium Health Wake Forest Baptist Cancer Center in North Carolina, who presented the findings at the 2026 ASCO Gastrointestinal Cancers Symposium. “The combination reduced early progression by about 10-fold and almost doubled both partial and complete response [rates compared with atezolizumab monotherapy].”
Median progression-free survival with the combination was 24.5 months compared with 5.3 months with atezolizumab alone (hazard ratio [HR] = 0.439, 95% confidence interval [CI] = 0.23–0.84; P = .0103). “This hazard ratio represents a substantial reduction in the risk of progression,” Dr. Rocha Lima commented.
As background, immunotherapy with single-agent pembrolizumab, a PD-1 inhibitor, is a standard of care in this population; however, in the previously reported KEYNOTE-177 trial nearly half the patients treated with pembrolizumab monotherapy progressed within 12 months.2 “These findings highlight the need for strategies aimed at improving early disease control beyond PD-1/PD-L1 monotherapy,” he said.
COMMIT: Study Rationale and Design
There is a strong rationale for the combination tested in this study, according to Dr. Rocha Lima. Preclinical colorectal cancer models showed synergistic tumor growth inhibition and prolonged survival with PD-L1 inhibition plus oxaliplatin. In the adjuvant setting, in stage III dMMR/MSI-H colorectal cancer, treatment with FOLFOX and atezolizumab significantly improved 3-year disease-free survival over chemotherapy alone in the ATOMIC trial; 3-year disease-free survival rates were 86.4% vs 76.6%, respectively (HR = 0.50; P < .0001).3 The combination of anti-VEGF agents, checkpoint inhibitors, and chemotherapy—the treatment tested in the COMMIT trial—has demonstrated synergistic activity in preclinical models by promoting vascular normalization and enhancing immune cell infiltration.
The COMMIT trial was initially designed as a three-arm study comparing FOLFOX and bevacizumab, FOLFOX and bevacizumab plus atezolizumab, and atezolizumab alone. After the KEYNOTE-177 results were published in 2020, the study was amended and the FOLFOX and bevacizumab arm, after enrollment of 20 patients, was closed. Additionally, due to slow accrual, the target sample size was reduced from 211 patients to 100 patients, providing 89% power to detect a true HR of 0.50, randomized between the remaining atezolizumab monotherapy vs FOLFOX and bevacizumab plus atezolizumab arms, corresponding to 120 patients across all three arms.
On March 31, 2025, enrollment to the COMMIT trial was suspended, after 82 patients were randomly assigned to the two comparative arms, because of the results of the CheckMate 8HW trial.4 Based on a preplanned interim analysis result, reviewed by the data monitoring committee in July 2025, the COMMIT study was subsequently permanently closed to further accrual.
The results of COMMIT presented at the ASCO Gastrointestinal Cancers Symposium included 41 patients in each of the remaining two arms. Patients had dMMR/MSI-H metastatic colorectal cancer without prior systemic treatment for metastatic disease. One cycle of FOLFOX or CAPOX (capecitabine and oxaliplatin) with or without bevacizumab was allowed prior to enrollment.
Other Key Findings
The progression-free survival rate at 12 months was 66.7% with the combination vs 35.1% with single-agent atezolizumab; at 24 months these rates were 53.7% and 31.6%, respectively.
“I want to highlight interesting subgroup observations,” Dr. Rocha Lima said. “Across higher-risk features—including older age, BRAF V600E mutation, prior adjuvant therapy, right-sided primary tumors, and higher baseline disease burden—the hazard ratios were numerically more favorable for the combination. These findings suggest that the regimen may be particularly beneficial for patients with more aggressive disease biology or higher tumor burden, although these analyses are exploratory.”
In addition to meeting its primary endpoint of progression-free survival, the overall response rate also favored the regimen of atezolizumab, bevacizumab, and FOLFOX at 86.1% compared with 46.0% with atezolizumab monotherapy. This included complete responses in 36.1% vs 18.9% in the monotherapy arm, and a disease control rate at 12 months of 64.7% vs 32.4%, respectively.
At the time of this analysis, there was no difference regarding overall survival (HR = 1.04, 95% CI = 0.47–2.28; P = .90). The 24-month overall survival rate was the same in both groups, 67%; the number of events remains limited, he noted.
Safety and Next Steps
Grade 3 to 5 adverse events occurred in 82.9% of the combination group and in 43.9% of the atezolizumab monotherapy group. The most common grade 3 or 4 adverse events in the experimental arm included neutropenia (26.8%), infection (26.8%), and hypertension (19.5%). Immune-related adverse events were similar in both arms.
There were five grade 5 adverse events in the study. One death occurred in the atezolizumab monotherapy arm from disease progression; in the combination arm, one was attributed to disease progression, two were sudden deaths after prolonged treatment exposure occurring after cycles 16 and 18, and one death was due to a fatal hepatic hemorrhage following liver surgery performed after cycle 12. In that case, bevacizumab had been discontinued early and oxaliplatin had been stopped several cycles prior to surgery.
The investigators plan correlative biomarker analyses to better identify patient subgroups most likely to benefit from the FOLFOX, bevacizumab, and atezolizumab combination. “These studies will evaluate tumor and immune microenvironment features, circulating biomarkers, and molecular correlates of response and resistance, to refine patient selection and inform future prospective trials,” he said.
Expert Point of View
COMMIT trial discussant Eric Scott Christenson, MD, of the Sidney Kimmel Comprehensive Cancer Center at Johns Hopkins Medicine, noted that the regimen of atezolizumab and FOLFOX, with or without bevacizumab, led to an improvement in progression-free survival, from 5.2 months to 24.5 months; however, an overall survival benefit was not achieved and toxicity was concerning.
“Is atezolizumab plus FOLFOX/bevacizumab a better option than atezolizumab alone? I would say, only in very select patients. But was atezolizumab the ideal front-line treatment in this space to begin with? Again, I would say in relatively select patients,” he maintained. “The real question for the majority of patients with metastatic dMMR [mismatch repair deficient] colorectal cancer is whether atezolizumab plus FOLFOX/bevacizumab is a viable alternative to ipilimumab/nivolumab.”
CheckMate 8HW evaluated the addition of the anti–CTLA-4 agent ipilimumab to the anti–PD-1 agent nivolumab, demonstrating a median progression-free survival that was not reached at 47 months, and a 14% absolute benefit at 1 year with the two immunotherapies over nivolumab alone and maintained at 2 and 3 years. Similarly, in COMMIT, the addition of chemotherapy led to a significant improvement in progression-free survival over atezolizumab, an anti–PD-L1 monotherapy.
“The plateau of the monotherapy arms was dramatically different between these two trials, with a relatively swift drop-off with atezolizumab alone and a less dramatic one with nivolumab alone,” he noted. “Whether this is related to patient selection or other factors is hard to know.” In short, he said it is difficult to compare the absolute progression-free survival numbers for the two studies as the control groups performed so differently.
But to determine for sure which regimen is preferred, overall survival data will be critical. This endpoint is immature for CheckMate 8HW (partly because nivolumab performed so well). For atezolizumab plus FOLFOX, no overall survival difference has emerged at 24 months despite an initial separation of the curves, “raising the question as to whether there were some ambiguous dMMR findings…and misclassifications…. Some patients may be falling off before they have the chance to get chemotherapy,” he posited. “I think it’s important to discern this with further analysis…. It’s pretty striking that we saw this really impressive difference in progression-free survival when we added chemotherapy to atezolizumab, but [it] did not ultimately [result in] an overall survival benefit,” he said.
In addition, he observed that ipilimumab/nivolumab has a slightly more favorable side-effect profile. In COMMIT, the rate of grade 3 to 5 adverse events almost doubled, and there were four deaths (9.8%) with the addition of chemotherapy. In contrast, he said, “It is striking that the toxicity of adding on ipilimumab to nivolumab was quite modest in CheckMate 8HW,” adding that while treatment with atezolizumab plus FOLFOX/bevacizumab can be prolonged, this “comes at the cost of toxicity.”
“For the time being, ipilimumab/nivolumab remains my first-line choice for fit patients with dMMR metastatic colorectal cancer,” Dr. Christenson concluded. “The lack of overall survival benefit with atezolizumab and FOLFOX at this time does limit my enthusiasm for its widespread adoption.”
DISCLOSURE: Dr. Rocha Lima reported no relevant personal financial disclosures. Dr. Christenson reported relationships with Boston Scientific, Parabilis, Roche, Seres Therapeutics, Sirtex, TATUM Bioscience, UroGen, and NextCure.
REFERENCES
1. Rocha Lima CMS, Yothers G, George TJ, et al. Colorectal Cancer Metastatic dMMR Immunotherapy (COMMIT) study: A randomized phase III study of atezolizumab (atezo) monotherapy versus mFOLFOX6/bevacizumab/atezo (FFX/bev) in the first-line treatment of patients (pts) with deficient DNA mismatch repair (dMMR) or microsatellite instability-high (MSI-H) metastatic colorectal cancer (mCRC)— NRG-GI004/SWOG-S1610. 2026 ASCO Gastrointestinal Cancers Symposium. Abstract 14. Presented January 10, 2026.
2. André T, Shiu KK, Kim TW, et al. Pembrolizumab in microsatellite-instability-high advanced colorectal cancer. N Engl J Med. 383(23):2207-2218, 2020.
3. Sinicrope FA, Ou F-S, Arnold D, et al. Randomized trial of standard chemotherapy alone or combined with atezolizumab as adjuvant therapy for patients with stage III deficient DNA mismatch repair colon cancer (Alliance A021502; ATOMIC). 2025 ASCO Annual Meeting. Abstract LBA1. Presented June 1, 2025.
4. André T, Elez E, Lenz HJ, et al. Nivolumab plus ipilimumab versus nivolumab in microsatellite instability-high metastatic colorectal cancer (CheckMate 8HW): a randomised, open-label, phase 3 trial. Lancet. 405:383-395, 2025.
