Prostate cancer has a well-documented survival trick: many tumors quietly switch off the molecular signal that lets the immune system recognize them as diseased tissue. A study published in Nature Biomedical Engineering, led by researchers at Duke University School of Medicine and the University of Rochester Medical Center, describes a CRISPR-based tool that reverses this specific evasion mechanism in prostate cancer cells — restoring the immune “visibility” that lets T cells find and attack them.
How prostate cancer hides in plain sight
Healthy cells constantly display fragments of their internal proteins on their surface using a molecular display case called MHC class I (MHC-I). T cells patrol the body reading these displays; if a fragment looks abnormal, the T cell kills the cell presenting it. Many prostate tumors evade this surveillance by shutting down MHC-I display, becoming what researchers call “immune cold” — effectively invisible to the very immune cells that immunotherapy drugs are designed to activate. That invisibility is a major reason checkpoint-inhibitor immunotherapies, which have transformed treatment for some cancers, have had limited success in prostate cancer.
The research team traced part of that shutdown to a gene called SPSB1. In prostate cancer cells, the messenger RNA (mRNA) that encodes SPSB1 gets shortened at its 3′ untranslated region (3′ UTR) — a regulatory tail attached to the end of the transcript. That shortening removes elements that normally restrain how much SPSB1 protein gets made, so the cell produces more of it. SPSB1 protein, in turn, targets components of the MHC-I complex for degradation, stripping the antigen display off the cell surface.
Editing the message without cutting it
Rather than cutting DNA, the team used a CRISPR-Cas13 system engineered to bind RNA directly. Standard gene-editing CRISPR tools (like Cas9) cut DNA to disable or alter a gene permanently. This approach instead uses a catalytically inactive (“dead”) Cas13 protein that binds to a specific site on the SPSB1 mRNA transcript without cutting it, physically blocking the shortening event at that site. The result is that the mRNA stays in its full-length form, its regulatory elements stay intact, SPSB1 protein production drops back toward normal, and MHC-I complexes are no longer degraded as aggressively — restoring antigen display on the cell surface.
In mouse models of prostate cancer, restoring MHC-I this way made tumors substantially more recognizable to the immune system and more responsive to immunotherapy. “If we can hit it with immunotherapy and another synergistic drug that pumps up the immune response, we could potentially cure it,” said Eric J. Wagner, PhD, of the University of Rochester Medical Center, one of the study’s senior researchers.
A genuine two-campus collaboration
The project brought together Duke’s Department of Pathology and Duke Cancer Institute with the University of Rochester’s Department of Biochemistry and Biophysics and Center for RNA Biology, with additional cancer-biology support from Rochester’s Wilmot Cancer Institute and Roswell Park Comprehensive Cancer Center in Buffalo. That combination reflects the two disciplines the work sits between: Duke’s cancer-pathology expertise in how prostate tumors actually behave, and Rochester’s RNA-biology expertise in building programmable tools that manipulate mRNA processing with precision. Large multi-author studies like this one — the published paper lists more than twenty co-authors across the two institutions — are increasingly how translational cancer research gets done, and they raise their own research-administration questions around multi-site data governance, authorship attribution, and cross-institutional funding agreements that CASRAI’s own standards work (CRediT taxonomies, data management planning) is built to support.
What this result does — and doesn’t — show
This is preclinical research: the findings come from engineered cell lines and mouse models, not from a human clinical trial. Turning an RNA-binding CRISPR tool into a deliverable therapy involves substantial additional work — safe and efficient delivery into human tumors, confirmation that the effect holds in more genetically diverse human cancers, and formal safety and efficacy testing before any trial in patients could begin. The researchers themselves frame the result as a proof of concept for combining MHC-I restoration with immunotherapy, not as a treatment ready for the clinic.
Still, the approach is notable for its precision: rather than editing the genome permanently, it corrects a specific RNA-processing error that tumors exploit, in principle limiting off-target effects compared to permanent DNA edits. That distinction is likely to matter as RNA-targeting CRISPR tools move into wider use across oncology research.
Source: Huang, F., Yuan, F., Li, K., Cui, Y., Li, L., Ye, W., et al. “Programmable mRNA 3’UTR engineering restores MHC-I and overcomes immune evasion in prostate cancer.” Nature Biomedical Engineering (2026). https://doi.org/10.1038/s41551-026-01720-9. Coverage: ScienceDaily, July 26, 2026.
