In cancer medicine, good news tends to come in cautious increments. A drug that buys a few extra months. A treatment that shrinks tumors but doesn't eliminate them. Progress measured in percentages, survival curves, and statistical margins. Which is why the results of the NEOPRISM-CRC trial, presented at the world's largest cancer research conference in April 2026, were so striking: after nearly three years of follow-up, not one patient had experienced a cancer recurrence. Zero.
The Treatment Order Nobody Questioned
For decades, the standard approach to treating bowel cancer that can be surgically removed has followed the same basic sequence: operate first, then treat. A surgeon removes the tumor, and afterward the patient receives chemotherapy — sometimes for three to six months — to kill any cancer cells that may have escaped into the body.
This order made intuitive sense. Remove the visible cancer first, then mop up what's left. For most cancer types, this logic has remained largely unchallenged. But a team of researchers at University College London (UCL) and University College London Hospitals (UCLH) decided to question it — and what they found may permanently change how a significant subset of bowel cancer patients are treated.
Their approach was simple in concept but bold in execution: give patients immunotherapy before surgery, not after. Instead of months of chemotherapy following the operation, patients would receive a short course of a drug designed to activate the body's own immune system against the tumor. Then, and only then, would they go into the operating room.
A Genetically Different Type of Bowel Cancer
Not all bowel cancers are alike. Most colorectal cancers have what's called proficient mismatch repair — meaning the cells have functional machinery to fix DNA errors. But in about 10 to 15% of stage II and III cases, that repair machinery is broken. These tumors are described as MMR-deficient (or equivalently, MSI-high).
Think of DNA as a text document that cells copy every time they divide. Healthy cells have a spellchecker that catches errors as they're copied. In MMR-deficient tumors, the spellchecker is switched off. Errors accumulate, and over time the tumor DNA becomes riddled with mistakes — mutations stacking on mutations.
This might sound like it makes things worse. In some ways, it does: these tumors are biologically more aggressive and have historically been harder to treat with chemotherapy. But it also creates an unexpected opportunity. A tumor loaded with mutations is, from the immune system's perspective, a very conspicuous target. All those mutations produce abnormal proteins that immune cells can potentially recognize as foreign and attack.
Immunotherapy drugs called checkpoint inhibitors — including pembrolizumab, sold under the brand name Keytruda — work by removing a molecular "off switch" that cancer cells use to hide from immune surveillance. In MMR-deficient tumors, where the immune system already has a lot to work with, this approach turns out to be dramatically effective.
How the Trial Worked
The NEOPRISM-CRC trial, run across five hospitals in the United Kingdom, enrolled 32 patients with stage II or III MMR-deficient bowel cancer — specifically patients whose tumors had high-risk features that historically meant a significant chance of the cancer coming back even after successful surgery.
Instead of proceeding directly to the operating room, patients received three doses of pembrolizumab given intravenously once every three weeks — a total treatment course of nine weeks. That's it. Nine weeks of immunotherapy, then surgery.
The standard alternative would have been: surgery, then three to six months of intravenous chemotherapy with all the fatigue, nausea, and disruption to daily life that typically brings. The trial flipped that entirely.
Patients were then followed with regular scans, blood tests, and check-ups to track whether the cancer returned.
The Results: Zero Relapses
When the initial results were published, they were already remarkable: 59% of patients had a pathologic complete response — meaning when surgeons removed the tissue, no living cancer cells could be found at all. The tumor had, in the words of one patient's doctor, "melted away."
But the data presented at the American Association for Cancer Research (AACR) Annual Meeting in April 2026 added something even more striking: the long-term follow-up.
After a mean of 33 months — nearly three years — every single patient in the trial was alive and cancer-free. This included not only those who had a complete response (no cancer cells found after treatment) but also those who still had some residual disease at surgery. None of them relapsed. Zero relapses across all 32 patients.
By contrast, the expected relapse rate for this type of high-risk bowel cancer treated with the standard surgery-plus-chemotherapy approach is approximately 15 to 40% at three years. In some high-risk subgroups, it exceeds 25%.
Dr. Kai-Keen Shiu, the trial's chief investigator at UCL Cancer Institute, described the findings as "extremely encouraging" and said they "strengthen our confidence that pembrolizumab is a safe and highly effective treatment" for this group. The side effect profile was also favorable: serious immune-related adverse effects occurred in fewer than 7% of patients, and no one had to stop treatment because of them.
A Blood Test That Predicts Who Responds
One of the most clinically significant findings from the trial goes beyond the survival data. Researchers developed a personalised blood test — based on detecting fragments of tumor DNA circulating in the bloodstream — that could show, as early as after the very first dose of pembrolizumab, whether the treatment was working.
The concept, called circulating tumor DNA (ctDNA) monitoring, works like this: tumors constantly shed tiny fragments of their DNA into the blood. By measuring whether these fragments decrease — or disappear entirely — doctors can get a real-time window into what's happening inside the tumor, without needing a scan or a biopsy.
In the trial, the results were striking. After just one cycle of pembrolizumab (three weeks), 24% of patients had no detectable tumor DNA in their blood at all. After three cycles, 58% had cleared it completely. After surgery, all 25 patients tested showed no detectable tumor DNA.
The researchers identified three distinct groups based on how quickly tumor DNA disappeared: "super molecular responders" who cleared it after one dose (and had a 100% complete response at surgery), "dynamic responders" who cleared it gradually, and "poor molecular responders" who took longer. This early blood signal matched surgical outcomes with remarkable accuracy.
The practical implication: a blood test taken after the first dose of immunotherapy may one day tell doctors which patients are responding so completely that surgery itself could potentially be avoided — or scaled back. That possibility is still being studied, but the groundwork is now in place.
A Real Patient's Story
Christopher Burston, 73, from Dorset in the UK, was diagnosed with bowel cancer in February 2023 after a routine postal screening kit returned a positive result. A subsequent colonoscopy revealed a stage III tumor — "quite a substantial lump," as he later described it.
A few weeks after his diagnosis, his oncologist suggested he might be eligible for the NEOPRISM trial. Despite needing to travel to London for treatment, he agreed. He received three doses of pembrolizumab over nine weeks, then underwent surgery in May 2023.
"The outcome of the surgery was essentially that the cancer had melted away," he recalled. "These were the doctor's words. The immunotherapy had had an almost immediate effect."
He spent a week in hospital recovering from surgery, experienced minimal side effects, and has since returned to normal life. More than three years after his diagnosis, he remains cancer-free.
"I feel very lucky," he said, "that I've reached the stage where my main problem is age rather than cancer or any illness."
What This Means for Patients Today
The NEOPRISM-CRC trial is a Phase II study — meaning it was designed primarily to test safety and gather early evidence of effectiveness, with a relatively small number of patients. The results are not yet sufficient to change treatment guidelines on their own, but they are compelling enough to have prompted significant attention at AACR 2026, one of the world's most important oncology conferences.
For patients and families navigating a bowel cancer diagnosis right now, several practical points are worth knowing:
- Ask about your tumor's genetic profile. The patients in this trial had MMR-deficient (also called MSI-high) tumors. If you or a loved one has been diagnosed with stage II or III colorectal cancer, it is now standard practice at most major centers to test for this genetic marker. If your tumor is MMR-deficient, you may be eligible for trials or emerging treatments based on this approach. Ask your oncologist whether this test has been done.
- Pembrolizumab is already approved for some colorectal cancers. The drug used in this trial (Keytruda/pembrolizumab) is FDA-approved for metastatic (stage IV) MSI-high colorectal cancer. Its use before surgery in earlier-stage disease is still investigational, but the approval pathway exists and the evidence base is growing rapidly.
- Ask about clinical trial eligibility. Larger trials building on NEOPRISM-CRC's results are being planned. A patient navigator or your oncology team can help identify whether you qualify for an active study. CancerInsight's clinical trial search tool can also help you find relevant open studies near you.
- Rising cases in younger adults. Colorectal cancer rates among people under 50 have been increasing for two decades. If you have a family history of bowel cancer, inflammatory bowel disease, or unexplained changes in bowel habits, discuss screening options with your doctor regardless of your age.
What Comes Next
The UCL team is now working to expand the NEOPRISM-CRC trial and gather the larger dataset needed for regulatory submissions. The goal is to establish pre-operative pembrolizumab as a recognized treatment option for MMR-deficient bowel cancer — potentially replacing post-operative chemotherapy for patients who respond to it.
The researchers are also pursuing the blood test angle aggressively. If ctDNA monitoring can reliably identify patients whose cancer has been entirely eradicated by immunotherapy before surgery, it opens a genuinely transformative possibility: that some patients may eventually avoid surgery altogether, treated with immunotherapy alone and monitored closely. That outcome is not yet proven or recommended, but it is now being actively investigated.
What is already clear from NEOPRISM-CRC is this: for patients with MMR-deficient bowel cancer, the immune system — properly activated — can eliminate a stage III tumor more effectively than months of chemotherapy. After three years and zero relapses, that is no longer a hypothesis. It is a clinical fact, and the field of oncology is beginning to reorganize around it.