FDA Approves Dostarlimab for Rectal Cancer. The Complete-Response Rate Is 47 or 68 Percent, Depending on Who Judges

On October 19, the FDA granted accelerated approval to dostarlimab (Jemperli, made by GSK) as a stand-alone treatment for adults with previously untreated stage II or III locally advanced rectal cancer whose tumors are mismatch repair deficient or microsatellite instability-high. An FDA-authorized test has to confirm the tumor type first. The agency said the review finished nearly four months ahead of schedule.

What is actually known is narrow. The drug blocks PD-1, a signal tumor cells use to hide from immune cells. The regimen is up to nine doses, one every three weeks, over about six months. The FDA's announcement says almost three-fourths of patients in the supporting trial did not need surgery, radiotherapy or chemotherapy. It does not give the exact count or the denominator behind that figure.

Two trials, two very different numbers

The approval rests on AZUR-1 (NCT05723562), a phase 2 trial with no comparison group that enrolled 154 previously untreated patients. Median follow-up was more than two years. OncLive, citing the approval materials, reports that 116 patients were evaluable by independent central review, and that 47 percent of them had a sustained clinical complete response at 12 months, the trial's primary endpoint. Investigators' own assessment put the rate at 68 percent. The overall response rate was 79 percent by central review and 82 percent by investigators.

The gap matters. A clinical complete response here means no tumor can be found on exam, endoscopy and imaging. It is a judgment call, and reviewers who did not treat the patients were stricter than those who did. Neither number is a survival result. OncLive notes that the incomplete-response and pathologic findings in patients who went on to surgery were suggested by imaging, not confirmed on tissue.

The earlier study that started this

The idea comes from a Memorial Sloan Kettering trial (NCT04165772) of 50 patients with the same tumor type. The FDA's account says no patient had detectable cancer after treatment, and 93 percent stayed cancer-free for at least 12 months. Of 27 patients followed for three years or longer, one had a recurrence, in the same area, and was treated with lymph node surgery.

Andrea Cercek, section head of colorectal cancer at Memorial Sloan Kettering, said the results show a "fundamentally different treatment approach" for this biomarker-defined group. Her institution ran the earlier trial, so she is not a neutral reader of the data. GSK's chief scientific officer, Tony Wood, called Jemperli the "first and only" approved immunotherapy for this use. That is a company description, repeated by OncLive from GSK's release.

Who this reaches, and what is not yet known

Published estimates cited by the FDA suggest about 3 percent of locally advanced rectal cancers are mismatch repair deficient or MSI-high. Most rectal cancer patients do not have this tumor type, and nothing in this approval applies to them.

Several things are not established. Longer follow-up is needed to show how long responses last and how often cancer returns, which is the condition attached to an accelerated approval. The coverage reviewed here carries no grade 3 or higher adverse event data and no comparison against standard chemoradiation plus surgery. The FDA lists fatigue, rash, diarrhea and itching as the most common side effects, each reported in at least 15 percent of patients, and warns that the drug can cause the immune system to attack healthy organs, with reactions that can be severe or fatal.

The pattern worth watching is the confirmatory evidence. Single-arm trials of 154 and 50 patients can show that tumors disappear. They cannot show that patients live longer than they would have after surgery.