Every year, more than 1.9 million people worldwide are diagnosed with colorectal cancer. For those whose tumors are caught at stage II or III, the standard playbook has long meant some combination of chemotherapy, radiation, and surgery, often stretching across months and carrying side effects that can linger for years. A small trial out of London is now challenging that entire sequence with a radically shorter approach, and the early results are hard to ignore.
In the NEOPRISM-CRC trial, led by researchers at University College London, 32 patients with stage II or III bowel cancer received pembrolizumab, a widely used immunotherapy drug, for up to nine weeks before their planned surgery. In 59 percent of those patients, surgeons found no detectable cancer remaining when they operated. And at a median follow-up of roughly 33 months, according to a University College London release published in April 2026, not a single participant across the entire cohort has relapsed. Not one.
Why these patients were selected
The trial did not enroll just anyone with bowel cancer. Every participant had tumors classified as mismatch repair-deficient (dMMR) or microsatellite instability-high (MSI-H). In plain terms, these are cancers whose internal DNA repair machinery is broken, causing them to accumulate large numbers of mutations. That heavy mutation load acts like a flashing beacon to the immune system, making these tumors unusually vulnerable to drugs like pembrolizumab, which work by releasing the brakes on immune cells so they can attack cancer.
This subgroup accounts for roughly 15 percent of all colorectal cancers, according to estimates from the National Cancer Institute. That means the vast majority of bowel cancer patients would not qualify for this approach. But for the roughly 285,000 people diagnosed globally each year whose tumors do carry these markers, the implications could be significant.
Pembrolizumab, sold by Merck under the brand name Keytruda, is a PD-1 checkpoint inhibitor already approved for use in several cancer types, including some advanced colorectal cancers. What makes the NEOPRISM-CRC trial distinctive is not the drug itself but the timing: giving it before surgery, in a compressed window, rather than reserving it for patients whose cancer has already spread or returned.
The zero-recurrence finding, in context
The most striking number from the trial is not the 59 percent complete response rate. It is the fact that zero recurrences have been recorded across all 32 patients, including those whose tumors were not fully eliminated by pembrolizumab alone. Those patients still underwent surgery and, as of the latest follow-up, remain cancer-free.
For context, recurrence rates for stage II and III colorectal cancer after standard treatment vary by substage but can range from roughly 10 to 30 percent within the first three years, depending on the study and patient population. A 33-month stretch with no relapses in a 32-person cohort is a notable departure from that baseline, even accounting for the favorable biology of dMMR tumors.
The NEOPRISM-CRC results also do not stand alone. A separate phase II trial, led by researchers at Memorial Sloan Kettering Cancer Center, tested another PD-1 inhibitor called dostarlimab in patients with dMMR rectal cancer. Early results from that study, published in The New England Journal of Medicine in 2022, showed complete clinical responses in an initial cohort of 12 patients, with follow-up data presented at subsequent oncology conferences continuing to show durable remissions. Some of those patients avoided surgery altogether.
Two different drugs. Two different research teams on different continents. Two different locations within the bowel. The same biological subtype. The same pattern of deep, sustained response. That kind of convergence is what moves a finding from intriguing anecdote toward credible biological signal.
What the trial has not yet proven
Thirty-two patients is a small number, and the trial was designed as a phase II study, meaning its purpose was to test whether the approach works well enough to justify larger, more definitive research. The formal trial registration is listed on ClinicalTrials.gov, but as of June 2026, full patient-level data on response, toxicity, and biomarker correlations have not yet appeared in a peer-reviewed journal.
That gap matters. Without detailed published data, independent researchers cannot assess how severe the side effects were, how tumor mutational burden (TMB) thresholds influenced outcomes, or whether certain patients responded far better than others. The trial used FoundationOne testing to measure TMB and stratify participants, a design feature that could eventually help clinicians predict who benefits most. But those granular findings remain unreported in the available literature.
There is also no control arm. Every participant received pembrolizumab, so there is no built-in comparison group of patients who received standard chemotherapy or chemoradiation instead. That makes it difficult to quantify exactly how much better this approach performed relative to existing treatment. The zero-recurrence result is encouraging, but without a head-to-head comparison, the precise magnitude of benefit remains an open question.
Follow-up duration is another caveat. Thirty-three months covers the window when many colorectal cancer recurrences surface, but late relapses do occur. Five-year data will be needed before clinicians can confidently say whether short-course pembrolizumab plus surgery effectively cures most treated patients or delays relapse.
And then there are the side effects. Pembrolizumab and other checkpoint inhibitors can trigger immune-related adverse events, including colitis, thyroid dysfunction, skin reactions, and, in rare cases, more serious organ inflammation. These side effects are different in character from those caused by chemotherapy and radiation, and some can become permanent. Without systematic reporting on how patients in this trial tolerated treatment and how quickly they returned to normal life, it is hard to fully weigh the promise of high response rates against the risk of new complications.
What comes next for dMMR bowel cancer treatment
The practical question facing oncologists and patients right now is straightforward: should people with dMMR stage II or III bowel cancer seek out neoadjuvant immunotherapy instead of standard treatment? The honest answer, as of mid-2026, is that the evidence is strong enough to warrant serious conversation but not yet strong enough to rewrite guidelines.
Larger trials, ideally randomized, will need to confirm whether the NEOPRISM-CRC results hold up in broader, more diverse patient populations. Researchers will also need to determine whether complete responders can safely skip surgery altogether, as the dostarlimab data have begun to suggest for rectal cancer, or whether surgery remains essential even when imaging and biopsies show no remaining disease.
For now, neoadjuvant pembrolizumab for this patient population is best pursued within clinical trials or through multidisciplinary tumor boards that can evaluate individual risk and benefit. What the NEOPRISM-CRC data have done is shift the conversation. The question is no longer whether immunotherapy works in dMMR bowel cancer. It clearly does. The question is how far that effectiveness can be pushed: whether weeks of treatment can replace months, whether some patients can keep their organs intact, and whether a 33-month streak of zero recurrences will hold at five years and beyond. Those answers are still being written, but the early chapters are remarkably promising.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
