Dostarlimab Yields Sustained cCRs in dMMR/MSI-H Locally Advanced Rectal Cancer

At the 2026 European Society for Medical Oncology (ESMO) Congress, interim results from the phase 2 AZUR-1 trial (NCT05723562) showed that dostarlimab-gxly (Jemperli) monotherapy produced clinical complete responses (cCRs) in patients with mismatch repair–deficient (dMMR) or microsatellite instability–high (MSI-H) locally advanced rectal cancer.¹ These results were simultaneously published in Annals of Oncology.²Investigators reported that 68% of evaluable patients achieved a cCR sustained for at least 12 months without additional treatment.

How did dostarlimab perform in patients with locally advanced rectal cancer?

By investigator assessment, 110 of 154 patients (71%) achieved a cCR after dostarlimab alone, including 102 (66%) at the end-of-treatment assessment and 8 who converted from a near complete or partial response later. A cCR required a normal digital rectal examination, no visible residual tumor on endoscopy, and rectal MRI without residual diffusion restriction. Other responses were near complete response (n = 9), partial response (n = 15), stable disease (n = 10), and progressive disease (n = 8); 10 patients (7%) were not evaluable.

The rate of cCR sustained for at least 12 months (cCR12) was 68% (n = 98/144) by investigator assessment and 47% (n = 54/116) by independent central review (ICR), with respective data maturity of 94% and 75%. Adding patients who had a pathologic complete response (pCR) after dostarlimab alone brought the rate to 76% (n = 110/144). “So 76% of patients had no residual disease after just dostarlimab alone in this entire cohort,” presenting study author Andrea Cercek, MD, said during the presentation. Cercek is the section head of Colorectal Cancer and the founder and co-director of the Center for Young Onset Colorectal and Gastrointestinal Cancer at Memorial Sloan Kettering Cancer Center.

Why did central review show a lower cCR12 rate?

Only 75% of ICR cCR12 data were mature, compared with 94% of investigator data. Cercek said reviewers received only flexible sigmoidoscopy and MRI findings. “They did not have the patient present, and their assessment of clinical complete response lagged,” she said. “That’s why the data in the central review are not yet mature.” There were also patients who were in cCR but had not made it to the 12-month mark and, as she said, “as we expect this data to mature, the numbers will significantly improve.” She added that some patients classified as not in cCR by central review “are clearly in CCR and have had durable [cCR],” and that numbers are expected to rise as follow-up matures.

“This is a patient-centric endpoint, and it’s absolutely necessary to have a patient present, to have a tumor board present, and to discuss these responses, but of course, also better tools are needed to better assess some of those challenging responses,” Cercek said.

Were responses durable, and what happened to patients without a cCR?

Among 100 patients with a cCR and at least 12 months of follow-up, 98% had durable responses; the other 2 elected to undergo surgery and had a pCR. “There were absolutely no recurrences in this entire cohort,” Cercek said. Median follow-up was 26.3 months (range, 2-36), and 143 of 154 patients (93%) had no evidence of progressive disease.

Of the 40 patients (26%) without a CCR who received additional standard-of-care treatment at investigator discretion, 13 underwent surgery alone (10 with a pCR), 7 received chemotherapy or chemoradiotherapy plus surgery (5 with a pCR), and 4 received radiation plus surgery (2 with a pCR). Additional treatment consisted of non-operative management (n = 2), chemotherapy or chemoradiotherapy alone (n = 7), radiotherapy alone (n = 2), chemotherapy alone (n = 2), or refused treatment (n = 3).

What was the safety profile of dostarlimab in AZUR-1?

Patients received a median of 9 cycles (range, 2-9). Dostarlimab-related adverse effects (AEs) occurred in 98 patients (64%) at grades 1 to 2 and in 15 (10%) at grade 3 or higher; the most common grade 1 or 2 events were pruritus (14%), hypothyroidism (12%), and fatigue (10%). Immune-mediated AEs occurred in 32 patients (21%) at grades 1 to 2 and 4 (3%) at grade 3 or higher. Serious AEs of any causality occurred in 31 patients (20%), of which 8 (5%) were related to dostarlimab. Three patients (2%) discontinued treatment because of dostarlimab-related AEs, and no related fatal events occurred.

What is the AZUR-1 design and regulatory context?

AZUR-1 is a global, single-arm phase 2 trial in 154 patients with previously untreated stage II or III dMMR/MSI-H rectal cancer and no symptomatic tumor-related bowel obstruction. Patients received dostarlimab at 500 mg intravenously every 3 weeks for 9 cycles. Those with a cCR entered nonoperative management with surveillance, and those with less than a cCR proceeded to surveillance or standard-of-care chemotherapy, radiation, or surgery. The primary and secondary end points were cCR12 by ICR and by investigator, respectively.

The median age was 61.5 years (range, 23-92), 86% had stage III disease, 90% had dMMR status, and most tumors were in the middle (44%) or lower (40%) rectum. “From a standard of care perspective, many of these patients [would] have had significant toxicity from surgery, and many of them would have needed a permanent colostomy,” Cercek said. The data cutoff was Apr 20, 2026. The FDA granted accelerated approval to dostarlimab in this setting.³

The investigators concluded that AZUR-1 “demonstrates sustained clinical complete responses after dostarlimab monotherapy in dMMR/MSI-H locally advanced rectal cancer, which has the potential to eliminate chemoradiation and surgery in a large proportion of patients,” Cercek said. She described the safety profile as acceptable, with no surprises.

Disclosures: Cercek reported advisory roles with Amgen, AbbVie, Agenus, Daiichi Sankyo, Merck, GSK, Pfizer, Roche, Janssen, Summit, 3T Biosciences, Urogen, Regeneron, and Boehringer Ingelheim, and institutional research grants from GSK and Pfizer. The study was funded by GSK.

References

1. Cercek A, Bachet JB, Folprecht G, et al. Clinical complete response with dostarlimab monotherapy in dMMR/MSI-H locally advanced rectal cancer: first results from the global AZUR-1 trial. Presented at: 2026 European Society for Medical Oncology Congress; October 23-27, 2026; Madrid, Spain. Abstract LBA87.
2. Cercek A, Bachet JB, Folprecht G, et al. Clinical complete response with dostarlimab monotherapy in dMMR/MSI-H locally advanced rectal cancer: first results from the global AZUR-1 trial. Ann Oncol. Published online October 25, 2026. doi:10.1016/j.annonc.2026.10.1280
3. FDA approves immunotherapy that may allow certain patients with rectal cancer to avoid chemotherapy, radiation and surgery. News release. FDA. October 19, 2026. Accessed October 26, 2026. https://tinyurl.com/yj2k5zh4