What makes a cancer cell itself: its genes, its hiding place, or the enemies it manages to outlive?
That sounds like a philosophy seminar with worse catering, but it is the right question for mismatch repair-deficient colorectal cancer. These tumors have broken DNA spellcheck. Every time the cell copies its genome, typos pile up like unread emails. Many of those typos create strange new proteins, called neoantigens, that can wave at the immune system like tiny molecular name tags saying, "Hello, I am suspicious."
By all rights, that should make these cancers easy targets. And here is where it gets interesting: they are both highly visible and annoyingly capable of escape. Statistically speaking, that is rude.
The Immune System Has a Red Pen
In this new Gut study, Xavier-Ferreira and colleagues asked a deceptively sharp question: what does immunity do to the evolution of mismatch repair-deficient colorectal tumors?
They used a mouse model of MMR-deficient colorectal cancer and compared tumors growing in immunocompetent hosts with tumors growing in immunodeficient hosts. In plain English: some mice had a working immune security team, and some had the biological equivalent of leaving the back door open.
The difference was not subtle. In immunocompetent animals, the immune system reduced tumor incidence by wiping out early clones that carried immunogenic mutations. That is immunoediting: the immune system does not just fight cancer, it edits the cast list. The clones that shout "please notice me" get removed. The quieter, sneakier clones keep auditioning.
In immunodeficient mice, tumors carried higher mutational burden, more genetic diversity, and mutations that were missing from tumors in immune-intact animals. Translation: without immune pressure, the tumor genome gets to be messier, louder, and more chaotic. A sample size may be small, but the biological vibe is not.
The Plot Twist Is Suppression
The second mechanism is less about hiding the tumor's identity and more about ruining the neighborhood.
Even in immunocompetent hosts, tumors that survived had a more suppressive tumor microenvironment: fewer cytotoxic lymphocytes, more checkpoint proteins, and a local immune climate that said, "No fighting in the lobby." If T cells are tiny bodyguards, the tumor microenvironment is a sketchy building manager who keeps deactivating their key cards.
That explains the paradox. MMR-deficient tumors can be packed with mutations and still progress. They are visible enough to attract immune attention, but the tumors that survive are the ones that either lost the most obvious targets or learned to make the immune response tired, blocked, or weirdly polite.
Why This Matters Beyond Mouse Drama
This matters because MMR-deficient colorectal cancer is already one of immunotherapy's biggest success stories. In metastatic MSI-H/dMMR colorectal cancer, pembrolizumab improved progression-free survival versus chemotherapy in KEYNOTE-177, and long-term follow-up supports durable benefit for many patients. Nivolumab plus ipilimumab also beat chemotherapy in the phase 3 CheckMate 8HW trial, with a progression-free survival hazard ratio reported around 0.21, which in statistics language is less a nudge and more a chair being thrown across the room.
In earlier-stage disease, the numbers have been eyebrow-raising too. The NICHE-2 study of neoadjuvant nivolumab plus ipilimumab in locally advanced dMMR colon cancer reported pathologic response in 98% of patients and pathologic complete response in 68%. When a trial hands you a 98%, even the confidence interval starts sitting up straighter.
But not everyone responds. Some tumors resist. Some relapse. Some look immunogenic on paper but behave like they read the footnotes and found a loophole.
This new study helps explain why: the immune system may remove the most targetable clones early, leaving behind tumors that are less obviously immunogenic and more locally suppressive. That has real implications for neoantigen vaccines, combination checkpoint blockade, and therapies aimed at reactivating exhausted immune cells or reshaping the tumor microenvironment.
The Takeaway
MMR-deficient colorectal cancer is not simply "hot" or "cold." It is edited. It is negotiated. It is the product of a long argument between genetic chaos and immune pressure.
If these findings hold up in broader human studies, the next generation of treatment may need to do two jobs at once: reveal the tumor's best neoantigen targets and fix the immune neighborhood so T cells can actually do their job. Because apparently cancer therapy is not just about finding the villain. It is also about making sure security can get into the building.
References
-
Xavier-Ferreira H, Vilarinho A, Cavadas B, et al. Dual mechanism of immune escape shapes the genetic and immunogenic landscape of mismatch repair-deficient colorectal tumours. Gut. 2026. https://doi.org/10.1136/gutjnl-2025-336834
-
Williams CJM, Peddle AM, Kasi PM, et al. Neoadjuvant immunotherapy for dMMR and pMMR colorectal cancers: therapeutic strategies and putative biomarkers of response. Nature Reviews Clinical Oncology. 2024. https://doi.org/10.1038/s41571-024-00943-6
-
Chalabi M, et al. Neoadjuvant immunotherapy in locally advanced mismatch repair-deficient colon cancer. New England Journal of Medicine. 2024;390:1949-1958. https://doi.org/10.1056/NEJMoa2400634
-
André T, et al. Nivolumab plus ipilimumab in microsatellite-instability-high metastatic colorectal cancer. New England Journal of Medicine. 2024. https://doi.org/10.1056/NEJMoa2402141
-
Feng Y, Ma W, Zang Y, et al. Spatially organized tumor-stroma boundary determines the efficacy of immunotherapy in colorectal cancer patients. Nature Communications. 2024;15:10259. https://doi.org/10.1038/s41467-024-54710-3
Disclaimer: The image accompanying this article is for illustrative purposes only and does not depict actual experimental results, data, or biological mechanisms.