Human leukocyte antigen (HLA) class I molecules enable cytotoxic T cells to recognize and eliminate cancer cells. Downregulation or loss of HLA expression represents a well-established mechanism of immune evasion. Across many tumor types, alterations in HLA expression have shown prognostic and predictive implications.3,4 However, the role of HLA status in prostate cancer remains poorly understood. Prior studies have been limited by small sample sizes or publicly available datasets. For example, an early study of 42 primary prostate cancer samples linked loss of HLA class I immunostaining to tumor relapse,5 and a genomic analysis of 628 primary tumors from the TCGA cohort reported a suppressive chromatin state with a loss of HLA class I expression.6 These findings hint that HLA dysregulation may contribute to prostate cancer’s immune-cold phenotype, but comprehensive data across disease stages have been lacking.
Our current study leverages the Caris Precision Oncology Alliance (POA) dataset, representing one of the largest real-world, multi-omic cohorts to date, encompassing 8,040 prostate cancers, including both primary and metastatic tumor biopsies. Beyond its scale, this dataset uniquely integrates genomic, transcriptomic, and immunophenotypic data from routine clinical testing, allowing robust characterization of HLA expression in the context of real-world tumor biology. This design provides an opportunity to examine the continuum of HLA regulation and its association with other concurrent molecular alterations, immune signaling, and potential therapeutic biomarkers.
Our analysis7 revealed that prostate cancer exhibits among the lowest HLA expression levels across all tumor types. Moreover, metastatic lesions demonstrated significantly lower HLA expression than primary tumors, suggesting that disease progression may involve further immune evasion through suppression of HLA-mediated antigen presentation. Next, we found that tumors with high HLA expression were enriched for TP53 and PTEN mutations and displayed elevated PD-L1 and CTLA-4 mRNA expression, findings consistent with those from the TCGA and SU2C datasets. We surmise that associations support a model in which TP53 and PTEN mutations can result in neoantigens that may stimulate tumor-infiltrating lymphocyte activity in prostate cancer.8
We also observed that HLA-high tumors were more likely to harbor dMMR, MSI-H, and TMB-H status. Given prior evidence that low HLA expression can drive ICI resistance even in TMB-high tumors,9 our findings suggest that HLA expression may represent a valuable biomarker for identifying patients more likely to benefit from ICI therapy. Unfortunately, the limited number of ICI-treated cases in our dataset precluded formal evaluation of ICI treatment outcomes by HLA status, underscoring the need for prospective clinical validation.
Perhaps the most intriguing and unexpected finding was that higher HLA-A and HLA-B expression in primary tumors was associated with worse prostate cancer prognosis, contrary to observations in some other malignancies.10 This paradox may reflect the high prevalence of tumor suppressor gene alterations within HLA-high tumors or the presence of dysfunctional immune modulation. Further mechanistic studies, particularly conducted with spatial resolution of immune cells, may clarify how HLA expression impacts the T-cells in the tumor microenvironment.
In clinical practice, HLA expression is already reported in some comprehensive genomic profiling tests, but this currently does not impact clinical decision-making for prostate cancer patients. While we have found that HLA expression status exhibits significant associations with tumorigenic pathways and clinical outcomes, and we posit that assessment of HLA expression status in future studies should become a priority, as this may enhance our understanding of patient outcomes and immune cell activity in the prostate tumor microenvironment.
Written by: Pornlada Likasitwatanakul, MD,1,2 and Emmanuel S. Antonarakis, MD,3
- Department of Medicine, University of Minnesota-Twin Cities, Minneapolis, MN, USA.
- Department of Medicine, Siriraj Hospital, Mahidol University, Bangkok, Thailand.
- Clark Endowed Professor of Medicine, Director of GU Oncology, Associate Director of Translation, Division of Hematology, Oncology and Transplantation, University of Minnesota, Masonic Cancer Center, Minneapolis, MN.
- Chesner, L. N. et al. Androgen Receptor Inhibition Increases MHC Class I Expression and Improves Immune Response in Prostate Cancer. Cancer Discov. 15, 481–494 (2025).
- Markowski, M. C. et al. Bipolar androgen therapy plus nivolumab for patients with metastatic castration-resistant prostate cancer: the COMBAT phase II trial. Nat. Commun. 15, 14 (2024).
- Chowell, D. et al. Patient HLA class I genotype influences cancer response to checkpoint blockade immunotherapy. Science 359, 582–587 (2018).
- Montesion, M. et al. Somatic HLA Class I Loss Is a Widespread Mechanism of Immune Evasion Which Refines the Use of Tumor Mutational Burden as a Biomarker of Checkpoint Inhibitor Response. Cancer Discov. 11, 282–292 (2021).
- Carretero, F. J. et al. Frequent HLA class I alterations in human prostate cancer: molecular mechanisms and clinical relevance. Cancer Immunol. Immunother. CII 65, 47 (2015).
- Rodems, T. S. et al. Reversible epigenetic alterations regulate class I HLA loss in prostate cancer. Commun. Biol. 5, 1–16 (2022).
- Likasitwatanakul, P. et al. HLA class I expression shapes the tumor immune microenvironment and influences prognosis in prostate cancer. Prostate Cancer Prostatic Dis. 1–10 (2025) doi:10.1038/s41391-025-01045-9.
- Malekzadeh, P. et al. Neoantigen screening identifies broad TP53 mutant immunogenicity in patients with epithelial cancers. J. Clin. Invest. 129, e123791.
- Goodman, A. M. et al. MHC-I genotype and tumor mutational burden predict response to immunotherapy. Genome Med. 12, 45 (2020).
- Paulson, K. G. et al. Acquired cancer resistance to combination immunotherapy from transcriptional loss of class I HLA. Nat. Commun. 9, 3868 (2018).