The U.S. Food and Drug Administration granted accelerated approval on October 19, 2026, to dostarlimab-gxly (Jemperli) as a standalone treatment for adults with previously untreated stage II or III locally advanced rectal cancer whose tumors have deficient mismatch repair or high microsatellite instability. The decision relied on interim findings from the single-arm, phase 2 AZUR-1 trial involving 154 patients and may allow some eligible patients to avoid chemotherapy, radiation and rectal surgery.
The approval applies only to a molecularly defined subset of rectal cancer
This is not a blanket approval for everyone diagnosed with rectal cancer. Eligible patients must be adults with previously untreated, locally advanced stage II or III disease and a tumor confirmed to be mismatch repair deficient (dMMR) or microsatellite instability-high (MSI-H) using an FDA-authorized test. Stage II and III disease may extend through the bowel wall or into nearby lymph nodes, but not into distant organs.
Mismatch repair is the machinery that corrects errors made when DNA is copied. When this machinery fails, mutations accumulate, and tumor cells may display unusual proteins that immune cells can recognize. MSI-H describes a characteristic pattern of DNA instability that often accompanies mismatch repair failure. Such tumors can be especially susceptible to therapies that release inhibitory signals on T cells.
The FDA estimated that about 3% of locally advanced rectal cancers have these markers. The exact proportion varies with the population studied and how tumors are tested. The key clinical step is molecular testing at diagnosis, not assuming eligibility from the cancer stage alone.
Dostarlimab blocks PD-1 to reactivate an immune response against the tumor
Dostarlimab is a monoclonal antibody directed against programmed cell death protein 1 (PD-1), an inhibitory receptor on immune cells. By blocking PD-1 signaling, the drug can enable T cells to mount a stronger response to cancer cells. It does not directly cut out the tumor, damage DNA as conventional chemotherapy does, or irradiate tissue.
The biological rationale is particularly strong in tumors that cannot repair DNA mismatches. Their accumulated mutations may increase the number of abnormal antigens available for immune recognition. This mechanism also helps explain why the approval depends on tumor testing rather than merely the location of the cancer.
Bites of Bio previously explained the broader immunological challenge in how T-cell exhaustion limits cancer immunotherapy. Unlike many solid tumors, mismatch repair-deficient rectal tumors appear unusually responsive to PD-1 blockade in the populations studied.
In AZUR-1, 110 of 154 patients avoided additional cancer treatment at the interim analysis
The pivotal AZUR-1 study was a multicenter, open-label, single-arm phase 2 trial. All 154 enrolled patients had previously untreated, histologically confirmed stage II or III dMMR/MSI-H locally advanced rectal cancer. They received dostarlimab 500 mg by intravenous infusion every three weeks for up to nine cycles, approximately six months of treatment unless toxicity or progression required stopping sooner.
Response evaluations combined endoscopy, rectal magnetic resonance imaging (MRI), and computed tomography (CT) of the chest, abdomen and pelvis. Patients with a clinical complete response could enter a nonoperative surveillance program, with assessments every four months for two years and every six months thereafter. Patients without an adequate response, or those with recurrence or progression, could receive standard treatment as their doctors considered appropriate.
According to the FDA, 110 of the 154 treated patients (71%) received dostarlimab alone, required no additional treatment and had no disease progression at the interim analysis. Another 26 (17%) underwent surgery, with or without other treatment; 13 (8%) received additional systemic therapy or radiation; one had progression without further treatment; and four withdrew or were lost to follow-up. These categories describe treatment pathways and disease status, not a randomized comparison against conventional care.
The trial’s primary endpoint was a clinical complete response sustained for 12 months, assessed by independent central review after the end of dostarlimab treatment. GSK reported that 54 of 116 evaluable patients (47%; 95% confidence interval 37% to 56%) had met this endpoint at the interim cutoff. Another 38 participants had reached a clinical complete response but had not yet completed the required 12 months of follow-up. Investigator assessment gave a higher 12-month sustained response figure, 98 of 144 evaluable patients (68%). Those percentages use different denominators and review methods and should not be treated as interchangeable.
A clinical complete response means no detectable cancer using the specified examinations. It does not guarantee that every cancer cell has been eradicated. Similarly, the 71% treatment-avoidance figure does not establish a 71% cure rate.
Earlier results were striking, but organ preservation requires continued monitoring
The FDA also cited an earlier Memorial Sloan Kettering study of 50 patients with previously untreated dMMR/MSI-H locally advanced rectal cancer. All 50 had no detectable evidence of cancer after dostarlimab treatment, and 93% maintained that clinical response for at least 12 months. Of 27 followed for at least three years, one experienced a local recurrence. These unusually encouraging results helped establish the feasibility of a nonoperative approach for selected patients.
Avoiding conventional treatment could be consequential. Chemotherapy, pelvic radiation and rectal surgery can cause short-term toxicity and lasting changes to bowel, bladder and sexual function, as well as fertility. Some operations require a permanent colostomy. Preserving the rectum is therefore a clinically meaningful objective, provided cancer control remains durable.
Nevertheless, the experience with blood-based detection of colorectal cancer recurrence shows why an encouraging test result must be distinguished from evidence that a management strategy improves long-term outcomes. With nonoperative management, response assessments and prompt recognition of regrowth are central parts of care, not optional extras.
Immune-related side effects can be serious despite the absence of chemotherapy
Immunotherapy is not synonymous with low-risk treatment. Because PD-1 blockade can intensify immune activity, healthy organs may also become targets. Immune-mediated complications can involve the lungs, intestines, liver, kidneys, endocrine glands and skin, and some reactions can be severe or fatal. The labeling also warns about infusion reactions, pregnancy-related harm and complications involving subsequent allogeneic stem-cell transplantation.
In the manufacturer’s AZUR-1 interim safety report, serious adverse reactions occurred in 20% of patients and adverse events led to permanent drug discontinuation in 3%. The most commonly reported reactions were fatigue (32%), rash (22%), diarrhea (18%) and itching (17%). Patients require monitoring during treatment and follow-up, with prompt investigation of symptoms that might indicate an immune-mediated reaction.
For other examples of how immune-based cancer therapies differ in mechanism and development stage, see our coverage of experimental CAR-T cell treatment for solid tumors. Dostarlimab’s new use has FDA authorization, but that does not remove the need to examine its long-term evidence.
What the evidence cannot yet answer
AZUR-1 did not randomly assign patients to dostarlimab or standard chemotherapy, radiation and surgery. Without a direct comparison group, it cannot determine whether overall survival, long-term recurrence rates or quality of life are better with dostarlimab. The interim results also include patients who have not yet accumulated enough follow-up to assess sustained response over several years.
A clinical complete response is assessed using imaging and endoscopy rather than examination of a surgically removed specimen. Microscopic residual disease may therefore go undetected. Close follow-up is essential, and some patients will still need surgery or other treatment. Eligibility is also limited to tumors with confirmed dMMR/MSI-H status; findings should not be generalized to the majority of rectal cancers that lack these biomarkers.
These limitations explain why the FDA used the accelerated approval pathway: an earlier measure of treatment effect supported authorization while additional data are gathered to confirm durable clinical benefit. The sponsor must continue following participants. Clinicians and patients still need to weigh the chance of organ preservation against treatment toxicity, surveillance demands and uncertainty about very long-term outcomes.
The trial and regulatory program were sponsored by GSK, the company that markets Jemperli. In its regulatory announcement, the company disclosed financial interests involving lead investigator Andrea Cercek. Those relationships do not invalidate the results, but are relevant when interpreting interim, sponsor-reported findings.
The key advance is that treatment for this subset of rectal cancer can now be selected according to the tumor’s DNA-repair biology before patients undergo conventional therapy. Whether that approach preserves organs without compromising survival will depend on continued follow-up and confirmatory evidence.
References
- U.S. Food and Drug Administration. FDA grants accelerated approval to dostarlimab for locally advanced rectal cancer. October 19, 2026. FDA approval summary.
- U.S. Food and Drug Administration. FDA Approves Immunotherapy That May Allow Certain Patients with Rectal Cancer to Avoid Chemotherapy, Radiation and Surgery. October 19, 2026. FDA press release.
- GSK plc. Jemperli approved in US as first and only immunotherapy for adults with MSI-H/dMMR locally advanced rectal cancer. October 19, 2026. Company regulatory filing.