The clinical course of prostate cancer (PCa) is highly variable, demanding an individualized approach to therapy and robust prognostic markers for treatment decisions. We here present a random forest-based classification model to predict aggressive behaviour of PCa. DNA methylation changes between PCa cases with good or poor prognosis (discovery cohort with n=78) were used as input. The model was validated with data from two independent PCa cohorts from ICGC and TCGA. Ranking of cancer progression-related DNA methylation changes allowed selection of candidate genes for additional validation by immunohistochemistry. We identified loss of ZIC2 protein expression as a promising novel prognostic biomarker for PCa in >12,000 tissue micro-array tumors. The prognostic value of ZIC2 proved to be independent from established clinico-pathological variables including Gleason, stage, nodal stage and PSA. In summary, we have developed a PCa classification model which either directly or via expression analyses of the identified top ranked candidate genes might help in decision making related to the treatment of prostate cancer patients.
Overall design
The study included 39 patients with good and 39 patients with bad prognosis prostate cancer patients. Sample selection was based on the following criteria: good prognosis is indicated by presence of organ confined disease (pT2) and lack of recurrence for at least 5 years. In contrast, bad prognosis is defined as systemic presence of metastatic disease, indicated by biochemical (PSA based) recurrence within 3 years and no response to local radiation therapy.
Less...
More...