FDA clears dostarlimab alone for dMMR rectal cancer, sparing some surgery

The Food and Drug Administration has granted accelerated approval to dostarlimab, GSK's PD-1-blocking antibody sold as Jemperli, for a small, genetically defined group of rectal cancer patients — and the approval is unusual because the drug is meant to be used alone, without the surgery, radiation and chemotherapy that normally define treatment [s1][s2]. The decision was recorded on 19 October 2026 as efficacy supplement 15 to biologics licence application 761174 [s3].

The approved use is narrow by design. It covers adults with previously untreated stage II or III locally advanced rectal cancer whose tumours are mismatch-repair deficient (dMMR) or microsatellite instability-high (MSI-H), confirmed by an FDA-authorised test [s1][s2]. That molecular signature is rare in this setting: the agency estimates only about 3% of locally advanced rectal cancers carry it [s2]. Those tumours tend to be studded with mutations, which makes them visible to the immune system and, in principle, responsive to a drug that releases the brakes on T cells.

Why "organ preservation" is the point

Standard care for locally advanced rectal cancer is demanding. It typically combines chemotherapy, radiotherapy and surgery, and the surgery itself can mean a permanent colostomy, along with lasting effects on bowel, bladder and sexual function. The premise behind this approval is that in the right tumour, immunotherapy alone might clear the cancer and let patients keep the rectum intact [s2]. Angelo de Claro, who directs the FDA's Oncology Center of Excellence, called it "a groundbreaking option that avoids surgery, radiation and chemotherapy" [s2].

What the trial showed

The approval rests on AZUR-1, a multicentre, single-arm, open-label trial in 154 treated patients with previously untreated MSI-H/dMMR stage II or III disease [s1]. Patients received 500 mg of dostarlimab intravenously every three weeks for up to nine cycles — roughly six months of treatment — unless the disease progressed or the side effects became unacceptable [s1].

The main measure was clinical complete response lasting at least 12 months, abbreviated cCR12 — meaning no detectable tumour on examination, imaging and endoscopy, sustained for a year. By blinded independent central review, 47% of 116 evaluable patients reached that bar (95% confidence interval 27% to 56%); by the treating investigators' assessment, in 144 evaluable patients, the figure was 68% (60% to 76%) [s1]. The overall response rate was 79% on central review (71% to 85%) and 82% by investigators (75% to 88%) [s1]. After a median follow-up of 26.3 months, none of the patients who had reached a complete or near-complete response had seen their cancer progress [s1]. Put plainly by the agency: after more than two years of median follow-up, almost three-quarters of patients had not needed surgery, radiotherapy or chemotherapy [s2].

How to read it

The results are striking, but the approval is conditional, and the reasons matter. This is an accelerated approval, which the FDA grants on an endpoint judged reasonably likely to predict benefit — here, tumour clearance — rather than on proven long-term survival. Continued approval can be made contingent on confirmatory evidence, and the agency is explicit that longer follow-up is needed to show how durable the responses are and how often the cancer returns [s1][s2].

Two further cautions sit in the data themselves. The trial had no comparison group, so the benefit is measured against expectation, not against standard chemoradiation given to similar patients. And the gap between the central-review and investigator cCR12 figures — 47% versus 68% — is a reminder that how "complete response" is judged changes the headline number, with the stricter, blinded read being the lower one [s1]. The eligible population is also tiny: the roughly 3% of rectal cancers that are dMMR or MSI-H [s2]. For everyone else, this approval changes nothing.

As a PD-1 blocker, dostarlimab carries the class's immune-mediated risks. Its labelling warns of immune-mediated reactions that can affect the lungs, bowel, liver, hormone glands and other organs, as well as infusion-related reactions and risks in pregnancy [s1]. The full prescribing information is being posted to Drugs@FDA [s1].

Red flags: anyone receiving immunotherapy who develops new breathlessness, persistent diarrhoea, severe abdominal pain, yellowing of the skin or eyes, or a reaction during an infusion should seek medical help promptly, as immune-mediated effects can escalate quickly. This article describes research and regulatory news and is not medical advice. Decisions about cancer treatment are for patients and their treating clinicians.