Every patient with stage II or III mismatch-repair-deficient bowel cancer who received nine weeks of pembrolizumab before surgery remained cancer-free at roughly 33 months of follow-up, according to updated results from the NEOPRISM-CRC trial led by University College London researchers. None of the patients relapsed. The finding raises a pointed question for oncologists: can a short course of a single immunotherapy drug replace months of post-operative chemotherapy for a specific genetic subset of colorectal cancer?
Zero relapses at 33 months and why oncologists are paying attention
The standard path for patients diagnosed with localized bowel cancer typically involves surgery followed by chemotherapy, a regimen that carries well-documented side effects including neuropathy, fatigue, and immune suppression. NEOPRISM-CRC flipped that sequence. Patients with confirmed dMMR or MSI-high tumors received up to nine weeks of pembrolizumab, a PD-1 checkpoint inhibitor, before their operations. The trial, registered as this clinical study, was designed to test the safety and efficacy of this pre-surgical approach in localized colorectal cancer rather than the advanced disease settings where immunotherapy had already shown benefit.
The biological logic is straightforward. Tumors with mismatch-repair deficiency accumulate far more mutations than their repair-proficient counterparts. Those mutations generate abnormal proteins that the immune system can recognize as foreign, but only if the brakes on T-cell activity are released. PD-1 blockade does exactly that. A foundational study published in the New England Journal of Medicine established that mismatch-repair deficiency predicts response to PD-1 blockade across tumor types, including a colorectal cancer cohort. That work gave investigators the rationale to test pembrolizumab earlier in the disease course, before cancer cells had a chance to spread.
What makes the 33-month follow-up striking is not just the absence of relapse but the durability of that absence. Stage II and III colorectal cancers carry meaningful recurrence risks even after successful surgery and chemotherapy. A zero-relapse rate across the treated group at nearly three years suggests that pre-surgical immunotherapy may be doing something chemotherapy does not: training the immune system to hunt down residual cancer cells that survive the operation.
For patients, the implications could be profound. If a short course of pembrolizumab can eliminate the need for adjuvant chemotherapy in this molecularly defined group, they may avoid long-term toxicities without sacrificing cure rates. For health systems, a shift from months of combination chemotherapy to a limited period of immunotherapy would require careful cost and resource analysis, but it could also streamline care pathways and reduce hospital visits associated with cytotoxic infusions.
How prior pembrolizumab evidence built the case for NEOPRISM-CRC
NEOPRISM-CRC did not emerge from thin air. Phase 3 data published in the New England Journal of Medicine had already demonstrated that pembrolizumab showed clear activity in MSI-high advanced colorectal cancer. That trial focused on patients with metastatic disease, a later and more dangerous stage. The question NEOPRISM-CRC posed was whether the same drug could work even better when given earlier, while the tumor is still contained and the immune system has not been worn down by widespread disease.
The hypothesis driving the trial goes beyond simple tumor shrinkage. When pembrolizumab is given before surgery, it encounters the primary tumor while it is still intact and shedding antigens. That exposure may prime a broader and more durable T-cell response than what occurs after the tumor has already been removed. In theory, those primed T cells then patrol the body for micrometastases, the tiny clusters of cancer cells too small to detect on scans but large enough to seed a recurrence months or years later. The 33-month data from the UCL team is consistent with this idea, though proving the mechanism will require additional laboratory work.
Circulating tumor DNA, or ctDNA, offers one potential way to track that immune surveillance in real time. Researchers in other neoadjuvant immunotherapy trials have used serial ctDNA measurements to identify patients whose blood clears of tumor fragments within months of treatment, a signal that correlates with long-term disease control. Whether the NEOPRISM-CRC investigators collected and analyzed ctDNA data has not been detailed in the available institutional reporting, leaving an important mechanistic question unanswered.
Unanswered questions about sample size, peer review, and broader applicability
The results reported by UCL are encouraging, but several gaps prevent them from changing clinical practice on their own. The most significant is the absence of a full peer-reviewed publication of the three-year outcomes. The data so far comes from an institutional release from University College London, which confirms the zero-relapse finding at approximately 33 months but does not disclose the exact number of patients enrolled, the breakdown by cancer stage, or the rate of pathological complete response at the time of surgery.
Sample size matters enormously here. A zero-relapse rate in a small cohort tells a different story than the same rate in a much larger population. Early-phase trials can produce dramatic results that fade when tested in broader, more diverse groups of patients. The registry entry for NEOPRISM-CRC outlines a phase 2 design with neoadjuvant pembrolizumab followed by surgery, but without the full dataset it is impossible to know how representative the participants were in terms of age, comorbidities, and tumor location.
Another open question is how these findings might apply beyond clearly defined dMMR or MSI-high cancers. In metastatic disease, PD-1 blockade has shown limited benefit in mismatch-repair-proficient colorectal tumors. It remains uncertain whether moving immunotherapy earlier in the disease course could overcome that resistance, or whether the dramatic responses in NEOPRISM-CRC are tightly linked to the hypermutated biology of dMMR disease. Until more detailed subgroup analyses are available, oncologists will have to assume the latter.
Safety is also a key consideration. The UCL report describes the neoadjuvant pembrolizumab regimen as well tolerated, but does not provide granular data on immune-related adverse events or surgical complications. In other cancers, neoadjuvant immunotherapy has occasionally led to inflammatory reactions that complicate operations. Surgeons and anesthetists will want reassurance that short-course PD-1 blockade does not increase perioperative risk, particularly in older patients or those with multiple medical conditions.
What comes next for neoadjuvant immunotherapy in colorectal cancer
For now, the NEOPRISM-CRC experience is best viewed as a proof-of-concept that demands confirmation. Larger, possibly randomized trials comparing neoadjuvant pembrolizumab plus surgery against standard surgery followed by chemotherapy will be needed to determine whether this strategy can become a new standard of care. Those studies will also need to standardize key endpoints, including pathological response rates, ctDNA clearance, and quality-of-life measures.
Translational research will be equally important. Detailed tissue and blood analyses could clarify why some dMMR tumors respond so profoundly to neoadjuvant PD-1 blockade and whether specific immune signatures predict long-term remission. Such work may help refine patient selection, ensuring that those most likely to benefit receive immunotherapy while others are spared ineffective treatment.
In the meantime, clinicians treating patients with stage II or III dMMR colorectal cancer face a familiar tension between exciting early data and the need for robust evidence. Off-trial use of neoadjuvant pembrolizumab may be tempting, especially for patients keen to avoid chemotherapy, but it risks outpacing the data and making it harder to enroll future studies. Many experts are likely to recommend that patients who are interested in this approach seek participation in clinical trials where outcomes can be systematically captured.
The headline result from NEOPRISM-CRC – zero relapses at nearly three years after a brief pre-surgical course of pembrolizumab – is undeniably striking. It reinforces the idea that timing and tumor biology matter as much as drug choice in cancer therapy. Whether this early success will translate into a durable change in practice will depend on how quickly and convincingly the research community can fill in the missing details.
More from Morning Overview
*This article was researched with the help of AI, with human editors creating the final content.
