PO-324 Detection of high-risk prostate cancer biomarkers by RNA sequencing and qPCR method

IntroductionNew prognostic biomarkers for prostate cancer have the potential to overcome the clinical challenge of therapy decision and overtreatment. Present diagnostic and prognostic tests are still limited in specificity resulting in a large number of false positives and unnecessary biopsies. Furthermore, they do not enable a proper stratification between men with a high risk for an aggressive disease course requiring comprehensive therapy scheme after surgery and men with a low risk of disease recurrence cured after prostatectomy or eligible for active surveillance. In particular, patients with Gleason score 6 and 7 tumours (low and mid stage) are difficult to stratify for the appropriate therapy or for active surveillance as conservative management approach.Material and methodsTherefore, we aimed to evaluate the potential of novel or known prognostic biomarkers in high-risk and low-risk prostate cancer tissues, and adjacent normal tissues by RNA sequencing and qPCR techniques. We also investigated tumour heterogeneity by including different foci from primary prostate cancer. A set of prognostic biomarker candidates was identified upon RNA sequencing of radical prostatectomy tumours of patients (n=25) with or without biochemical relapse (>5 year follow-up), as well as adjacent benign tissues. The candidate genes were retrospectively validated by qPCR method in the same and in an independent patient cohort (n=59). Expression variance of genes was investigated in different tumour foci of four primary tumours.Results and discussionsOverall, 16 prognostic biomarker candidates were selected from RNA sequencing data analysis according to differential expression between high-risk cancer, low-risk cancer and benign tissues. In total 10/16 candidates were technically sound upon qPCR of in the same cohort. We could clearly show that qPCR is a very robust and sensitive method to verify RNA sequencing data. Additionally, known tumour markers like AMACR and ERG showed expected signatures associated with the clinical phenotype and FISH-based gene fusion status. Data from different tumour loci indicated high expression variances across tumour sections. Independent validation of candidate genes could not confirm significant differential expression between the patient risk groups.ConclusionTumour heterogeneity might impede the detection and validation of diagnostic and prognostic biomarkers in primary prostate cancer tissues.