BackgroundCovalent RNA modifications, such as N-6-methyladenosine (m6A), have been associated with various biological processes, but their role in cancer remains largely unexplored. m6A dynamics depends on specific enzymes whose deregulation may also impact in tumorigenesis. Herein, we assessed the differential abundance of m6A, its writer VIRMA and its reader YTHDF3, in testicular germ cell tumors (TGCTs), looking for clinicopathological correlates.MethodsIn silico analysis of TCGA data disclosed altered expression of VIRMA (52%) and YTHDF3 (48%), prompting subsequent validation. Formalin-fixed paraffin-embedded tissues from 122 TGCTs (2005–2016) were selected. RNA extraction, cDNA synthesis and real-time qPCR (Taqman assays) for VIRMA and YTHDF3 were performed, as well as immunohistochemistry for VIRMA, YTHDF3 and m6A, for staining intensity assessment. Associations between categorical variables were assessed using Chi square and Fisher’s exact test. Distribution of continuous variables between groups was compared using the nonparametric Mann–Whitney and Kruskal–Wallis tests. Biomarker performance was assessed through receiver operating characteristics (ROC) curve construction and a cut-off was established by Youden’s index method. Statistical significance was set at p < 0.05.ResultsIn our cohort, VIRMA and YTHDF3 mRNA expression levels differed among TGCT subtypes, with Seminomas (SEs) depicting higher levels than Non-Seminomatous tumors (NSTs) (p < 0.01 for both). A positive correlation was found between VIRMA and YTHDF3 expression levels. VIRMA discriminated SEs from NSTs with AUC = 0.85 (Sensitivity 77.3%, Specificity 81.1%, PPV 71.6%, NPV 85.3%, Accuracy 79.7%). Immunohistochemistry paralleled transcript findings, as patients with strong m6A immunostaining intensity depicted significantly higher VIRMA mRNA expression levels and stronger VIRMA immunoexpression intensity (p < 0.001 and p < 0.01, respectively).ConclusionAbundance of m6A and expression of VIRMA/YTHDF3 were different among TGCT subtypes, with higher levels in SEs, suggesting a contribution to SE phenotype maintenance. VIRMA and YTHDF3 might cooperate in m6A establishment in TGCTs, and their transcript levels accurately discriminate between SEs and NSTs, constituting novel candidate biomarkers for patient management.Electronic supplementary materialThe online version of this article (10.1186/s12967-019-1837-z) contains supplementary material, which is available to authorized users.
