Highlights
Low PTEN expression in microsatellite stable colorectal cancer was associated with reduced intratumoral CD8+ T-cell infiltration and poorer outcomes in patients treated with immune checkpoint blockade.
The study identifies a non-canonical role for PTEN: direct stabilization of KEAP1 through its C2 domain, thereby restraining NRF2 activation.
PTEN deficiency promoted NRF2-dependent selective autophagy, leading to lysosomal degradation of MHC-I and impaired tumor antigen presentation.
Pharmacologic NRF2 inhibition with ML385 restored surface MHC-I expression and improved anti-PD-1 responsiveness in PTEN-deficient models, supporting a tractable therapeutic strategy.
Background and Clinical Context
Microsatellite stable colorectal cancer accounts for the large majority of colorectal cancer cases and remains notably resistant to immune checkpoint blockade. In contrast to mismatch repair-deficient or microsatellite instability-high tumors, MSS colorectal cancers usually have a lower neoantigen burden, less inflamed tumor microenvironment, and limited baseline T-cell infiltration. As a result, anti-programmed cell death protein 1 and related immunotherapies have not shown broad activity in this population, creating a major unmet need.
PTEN is a well-established tumor suppressor with canonical functions in antagonizing phosphatidylinositol 3-kinase signaling. PTEN loss is recurrent across many solid tumors and has been linked to aggressive biology, altered metabolism, and treatment resistance. In MSS colorectal cancer, PTEN loss occurs in roughly 19% to 36% of cases, but its direct role in immune exclusion has been incompletely defined. The present study by Cai and colleagues, published in Gut, addresses this gap by connecting PTEN deficiency to a specific mechanism of antigen-presentation failure: autophagic degradation of MHC class I molecules through an NRF2-dependent pathway.
This is clinically important because intact MHC-I surface expression is essential for CD8+ T-cell recognition of tumor cells. Tumors that lose MHC-I can become effectively invisible to cytotoxic T cells, even when immune checkpoint inhibitors release inhibitory brakes. A mechanism that simultaneously reduces MHC-I, excludes T cells, and drives primary resistance to anti-PD-1 therapy would be highly relevant for biomarker development and rational combination treatment.
Study Design and Methods
The investigators used an integrative translational approach combining patient-level correlative analyses, public genomic data, single-cell immune profiling, and mechanistic laboratory experiments. Clinical cohorts and The Cancer Genome Atlas database were analyzed to assess the relationship between PTEN expression, immune infiltration, and outcomes with immune checkpoint blockade. Although the abstract does not provide cohort size or hazard estimates, the clinical analyses were positioned to test whether PTEN-low tumors exhibit a less inflamed phenotype and worse therapeutic response.
To define the immune microenvironment, the authors employed single-cell RNA sequencing and flow cytometry. These approaches enabled detailed characterization of immune-cell subsets, especially CD8+ T-cell abundance and functional states. To dissect the underlying biology, the study incorporated immunohistochemistry, mass spectrometry, co-immunoprecipitation, and confocal microscopy. This multimodal design is a strength because it links a clinically observed phenotype to molecular interactions and subcellular trafficking events.
The mechanistic focus centered on the PTEN-KEAP1-NRF2 axis. KEAP1 normally restrains NRF2 by promoting its degradation, thereby limiting antioxidant and stress-response transcriptional programs. The investigators asked whether PTEN influences this pathway independently of its classic phosphatase signaling role, and whether such regulation could affect selective autophagy and ultimately MHC-I stability.
Key Findings
PTEN-low MSS colorectal cancer is immunologically cold and clinically resistant
Across clinical and TCGA-based analyses, low PTEN expression correlated strongly with reduced CD8+ T-cell infiltration. In patients receiving immune checkpoint blockade, PTEN-low status was also associated with poorer survival. Even without detailed numerical effect sizes in the abstract, the directionality is biologically coherent and clinically significant. It places PTEN among candidate biomarkers for identifying immune-excluded MSS colorectal cancers.
Single-cell RNA sequencing and flow cytometry further supported an immune-depleted phenotype in PTEN-deficient tumors. The key translational message is that PTEN loss is not simply a genomic alteration associated with tumor growth; it appears to shape the tumor-immune interface in a way that directly undermines the activity of T-cell-dependent immunotherapies.
PTEN has a non-canonical immune-regulatory function through KEAP1 stabilization
The most novel aspect of the study is the identification of a non-canonical role for PTEN. The authors found that PTEN physically interacts with KEAP1 via the PTEN C2 domain and stabilizes KEAP1 by shielding it from p62-mediated degradation. This finding moves beyond the conventional view of PTEN as mainly a lipid phosphatase acting on PI3K-AKT signaling.
In PTEN-deficient cells, KEAP1 becomes destabilized and degraded, releasing repression of NRF2. This leads to NRF2 hyperactivation and an enhanced transcriptional program favoring selective autophagy. The work therefore places PTEN upstream of a KEAP1-NRF2 stress-response circuit that can reprogram antigen presentation.
NRF2-dependent selective autophagy drives MHC-I loss
Mechanistically, NRF2 activation transcriptionally upregulated components of the selective autophagy machinery. This, in turn, promoted selective lysosomal degradation of MHC-I. The resulting reduction in surface MHC-I effectively blocked recognition by CD8+ T cells and impaired their effector function.
This is an important conceptual advance. Many prior studies of immune resistance have focused on defective antigen processing, interferon pathway disruption, beta-2 microglobulin loss, or transcriptional downregulation of MHC-I. Here, the defect appears to be post-translational and trafficking-based: MHC-I is produced but is selectively removed through autophagic-lysosomal degradation. That provides a distinct and potentially druggable route to immune escape.
The use of confocal microscopy, co-immunoprecipitation, and mass spectrometry strengthens this mechanistic chain. Rather than a simple association between PTEN loss and reduced MHC-I, the authors describe a plausible causal cascade linking PTEN deficiency to KEAP1 destabilization, NRF2 activation, autophagy induction, and loss of antigen-presenting machinery.
NRF2 inhibition restores immune visibility and sensitizes tumors to anti-PD-1 therapy
A particularly relevant translational result is that ML385, a pharmacologic NRF2 inhibitor, restored cell-surface MHC-I levels in PTEN-deficient tumors and sensitized these tumors to anti-programmed cell death protein 1 therapy. This rescue experiment is critical because it suggests the pathway is not merely descriptive but therapeutically actionable.
For MSS colorectal cancer, where checkpoint blockade alone is largely ineffective, the prospect of converting an immune-resistant tumor into one that can present antigen and engage CD8+ T cells is compelling. If validated in additional preclinical systems and ultimately in biomarker-selected clinical trials, NRF2 blockade could emerge as a rational combination partner for PD-1 inhibition in PTEN-deficient disease.
Mechanistic and Clinical Interpretation
This study adds to a growing body of literature showing that oncogenic and tumor-suppressor pathways can directly remodel anti-tumor immunity. PTEN loss has previously been implicated in immune resistance in other cancers, including melanoma, where impaired T-cell infiltration and altered cytokine networks have been described. The present report extends that paradigm to MSS colorectal cancer and provides unusually detailed mechanistic resolution.
The biological model is attractive for several reasons. First, it explains immune exclusion through a cell-intrinsic mechanism centered on antigen presentation. Second, it offers a biomarker framework: PTEN-low tumors may be especially likely to harbor defective MHC-I display despite the absence of classical antigen-processing gene mutations. Third, it identifies NRF2 as a tractable signaling node. While NRF2 has complex context-dependent roles, including cytoprotection in normal tissues, its hyperactivation in tumors is increasingly recognized as a driver of treatment resistance and metabolic adaptation.
Clinically, the work raises the possibility that PTEN expression, PTEN genomic status, or downstream KEAP1-NRF2-autophagy signatures could help stratify patients with MSS colorectal cancer for combination immunotherapy studies. It also suggests that apparent “cold” tumors may not always be devoid of relevant antigens; rather, they may fail to display them effectively on the cell surface.
Strengths and Limitations
The study’s major strengths are its integrative design and mechanistic depth. The investigators moved from patient datasets to single-cell immune phenotyping and then to molecular dissection of protein interactions and intracellular trafficking. This kind of bench-to-bedside-to-bench framework is particularly well suited to translational oncology.
Several limitations should be kept in mind. The abstract does not report sample sizes, effect sizes, hazard ratios, or confidence intervals, so the magnitude and precision of the clinical associations cannot be judged from the available information alone. The extent to which PTEN loss was defined by protein expression, mutation, deletion, or functional inactivation is also important, because these measures may not be interchangeable in practice.
In addition, ML385 is a useful research tool, but the clinical readiness of NRF2-targeted therapy remains uncertain. Systemic NRF2 inhibition could have toxicity implications, given NRF2’s role in oxidative stress responses in normal tissues. Another key question is whether this mechanism is uniformly operative across all PTEN-deficient MSS colorectal cancers or only within specific molecular contexts, such as distinct consensus molecular subtypes, stromal states, or co-occurring genomic alterations.
Finally, as with many translational studies, successful rescue in preclinical models does not guarantee clinical benefit. Biomarker-driven early-phase trials will be required to determine whether restoring MHC-I is sufficient to meaningfully improve checkpoint inhibitor activity in patients.
Implications for Practice and Research
This study is unlikely to change routine management immediately, but it meaningfully sharpens the translational agenda in MSS colorectal cancer. For clinicians and trialists, PTEN may warrant consideration as a negative predictive biomarker for immunotherapy responsiveness, pending validation in larger contemporary cohorts. For laboratory investigators, the report highlights autophagic regulation of MHC-I as a fertile area for further study, alongside more established pathways of antigen-presentation loss.
Future work should address several priorities: validation of PTEN-low status as a biomarker in independent immunotherapy-treated MSS colorectal cancer cohorts; development of clinically feasible assays for PTEN and pathway activation; exploration of combination strategies involving NRF2 inhibition, autophagy modulation, or both; and careful assessment of on-target toxicities. It will also be important to determine whether similar PTEN-KEAP1-NRF2-MHC-I interactions operate in other PTEN-deficient solid tumors.
Funding and Trial Registration
The abstract provided does not specify funding sources. No ClinicalTrials.gov registration number is listed, which is consistent with a translational mechanistic study rather than a registered interventional clinical trial.
Conclusion
Cai and colleagues provide a compelling explanation for why a subset of MSS colorectal cancers may remain refractory to checkpoint blockade. PTEN deficiency appears to promote immune evasion through KEAP1 destabilization, NRF2 hyperactivation, selective autophagy, and lysosomal degradation of MHC-I, culminating in loss of CD8+ T-cell recognition. The rescue of this phenotype with ML385 makes the pathway therapeutically relevant, not merely descriptive.
For a disease setting in which effective immunotherapy remains elusive, this study offers both mechanistic clarity and a credible translational hypothesis: PTEN-deficient MSS colorectal cancer may be rendered more visible to the immune system by targeting NRF2-dependent autophagy. The next challenge is to convert this elegant biology into clinically effective, biomarker-guided treatment strategies.
References
1. Cai R, Zhan W, Lyu X, Yang X, Wu Z, Cheng Y, Guo C, Feng J, Fu Y, Xie Y, Qin G, Deng Y, Zhang J. PTEN deficiency impairs MHC-I-mediated tumour immunity via NRF2-dependent autophagy in microsatellite stable colorectal cancer. Gut. 2026-05-22. PMID: 42173678.
2. Le DT, Durham JN, Smith KN, Wang H, Bartlett BR, Aulakh LK, Lu S, Kemberling H, Wilt C, Luber BS, et al. Mismatch repair deficiency predicts response of solid tumors to PD-1 blockade. Science. 2017;357(6349):409-413. PMID: 28596308.
3. Ganesh K, Stadler ZK, Cercek A, Mendelsohn RB, Shia J, Segal NH, Diaz LA Jr. Immunotherapy in colorectal cancer: rationale, challenges and potential. Nat Rev Gastroenterol Hepatol. 2019;16(6):361-375. PMID: 30886395.
4. Peng W, Chen JQ, Liu C, Malu S, Creasy C, Tetzlaff MT, Xu C, McKenzie JA, Zhang C, Liang X, et al. Loss of PTEN promotes resistance to T cell-mediated immunotherapy. Cancer Discov. 2016;6(2):202-216. PMID: 26645196.
