A m6Avalue predictive of prostate cancer stemness, tumor immune landscape and immunotherapy response

Abstract

The molecular mechanisms underpinning prostate cancer (PCa) progression are incompletely understood, and precise stratification of aggressive primary PCa (pri-PCa) from indolent ones poses a major clinical challenge. Here, we comprehensively dissect, genomically and transcriptomically, the m6A (N6-methyladenosine) pathway as a whole in PCa. Expression, but not the genomic alteration, repertoire of the full set of 24 m6A regulators at the population level successfully stratifies pri-PCa into three m6A clusters with distinct molecular and clinical features. These three m6A modification patterns closely correlate with androgen receptor signaling, stemness, proliferation and tumor immunogenicity of cancer cells, and stroma activity and immune landscape of tumor microenvironment (TME). We observe a discrepancy between a potentially higher neoantigen production and a deficiency in antigen presentation processes in aggressive PCa, offering insights into the failure of immunotherapy. Identification of PCa-specific m6A phenotype-associated genes provides a basis for construction of m6Avalue to measure m6A methylation patterns in individual patients. Tumors with lower m6Avalue are relatively indolent with abundant immune cell infiltration and stroma activity. Interestingly, m6Avalue separates PCa TME into fibrotic and nonfibrotic phenotypes (instead of previously reported immune-proficient or -desert phenotypes in other cancer types). Significantly, m6Avalue can be used to predict drug response and clinical immunotherapy efficacy in both castration-resistant PCa and other cancer types. Therefore, our study establishes m6A methylation modification pattern as a determinant in PCa progression via impacting cancer cell aggressiveness and TME remodeling.

The m6A methylation landscape, molecular characterization and clinical relevance in prostate adenocarcinoma

“…Increasing studies have revealed the m 6 A regulatory patterns of PCa and correlated these modification patterns with the tumor immune cell infiltration microenvironment characteristics ( 49 – 51 ). In addition, a recent paper found that m 6 A reader HNRNPC can regulate Treg cell abundance as a possible mechanism for m 6 A methylation-mediated response against CTLA-4, indicating that activation of the immune microenvironment by targeting m 6 A regulators may serve as a potential therapeutic approach for advanced PCa( 52 ).Our study synthetically analyzed the relationship between the expression of m 6 A regulators and immune characteristics and drug sensitivity of PCa patients.…”

The established chemokine-related prognostic gene signature in prostate cancer: Implications for anti-androgen and immunotherapies

“…Zou et al. proposed that PCa patients with low m6Avalues respond better to immunotherapy and had a longer survival time ( 29 ). Consistently, our results indicated that the CRPGS was positively associated with m6Avalue, and low-risk PCa patients had lower m6Avalues and m6Alevels, which implied that low-risk patients exhibited lower levels of m6A status and had a better response to immunotherapy.…”

“…The established CRPGS was compared with our two previously published PCa signatures ( 16 , 24 ) and another five PCa molecular classifiers to evaluate the wide application of CRPGS ( 25 – 29 ), which was described in the Supplementary Materials and Methods.…”

“…For RFS analysis, PCa patients with high riskscore + C2/C3 had poorer survival (P = 0.004, Figures 7I, J). Zou et al (29) generate the m6Avalue and m6Alevel to assess the immune landscape, stemness, and drug response of PCa. High-risk patients exhibited a high m6Alevel (P = 0.036, Figures 8A, B), and riskscore was positively associative with m6Avalue, and patients with a high riskscore had a higher m6Avalue (all P < 0.001, Figures 8C, D).…”

The functional roles of m6A modification in prostate cancer

“…Recently Zou et al. [84]. divided PCa into three groups with distinct molecular and clinical characteristics according to the expression profile of m6A regulators, and established an m6A value based on the scoring characteristics of m6A phenotype‐related genes to quantify m6A modifications in individual PCa patient pattern.…”