Dataset Information

Random forest-based modelling to detect biomarkers for prostate cancer progression.

ABSTRACT: BACKGROUND:The clinical course of prostate cancer (PCa) is highly variable, demanding an individualized approach to therapy. Overtreatment of indolent PCa cases, which likely do not progress to aggressive stages, may be associated with severe side effects and considerable costs. These could be avoided by utilizing robust prognostic markers to guide treatment decisions. RESULTS:We present a random forest-based classification model to predict aggressive behaviour of prostate cancer. DNA methylation changes between PCa cases with good or poor prognosis (discovery cohort with n?=?70) were used as input. DNA was extracted from formalin-fixed tumour tissue, and genome-wide DNA methylation differences between both groups were assessed using Illumina HumanMethylation450 arrays. For the random forest-based modelling, the discovery cohort was randomly split into a training (80%) and a test set (20%). Our methylation-based classifier demonstrated excellent performance in discriminating prognosis subgroups in the test set (Kaplan-Meier survival analyses with log-rank p value ?12,000 tissue micro-arrayed PCa cases. Loss of ZIC2 protein expression was associated with poor prognosis and correlated with significantly shorter time to biochemical recurrence. The prognostic value of ZIC2 proved to be independent from established clinicopathological variables including Gleason grade, tumour stage, nodal stage and prostate-specific-antigen. CONCLUSIONS:Our results highlight the prognostic relevance of methylation loss in PMD regions, as well as of several candidate genes not previously associated with PCa progression. Our robust and externally validated PCa classification model either directly or via protein expression analyses of the identified top-ranked candidate genes will support the clinical management of prostate cancer.

SUBMITTER: Toth R

PROVIDER: S-EPMC6805338 | biostudies-literature | 2019 Oct

REPOSITORIES: biostudies-literature

ACCESS DATA

Publications

Random forest-based modelling to detect biomarkers for prostate cancer progression.

Toth Reka R Schiffmann Heiko H Hube-Magg Claudia C Büscheck Franziska F Höflmayer Doris D Weidemann Sören S Lebok Patrick P Fraune Christoph C Minner Sarah S Schlomm Thorsten T Sauter Guido G Plass Christoph C Assenov Yassen Y Simon Ronald R Meiners Jan J Gerhäuser Clarissa C

Clinical epigenetics 20191022 1

<h4>Background</h4>The clinical course of prostate cancer (PCa) is highly variable, demanding an individualized approach to therapy. Overtreatment of indolent PCa cases, which likely do not progress to aggressive stages, may be associated with severe side effects and considerable costs. These could be avoided by utilizing robust prognostic markers to guide treatment decisions.<h4>Results</h4>We present a random forest-based classification model to predict aggressive behaviour of prostate cancer. ...[more]

PMID: 31640781

Similar Datasets

Project description:PURPOSE: Prostate cancer is a bimodal disease with aggressive and indolent forms. Current prostate-specific-antigen testing and digital rectal examination screening provide ambiguous results leading to both under-and over-treatment. Accurate, consistent diagnosis is crucial to risk-stratify patients and facilitate clinical decision making as to treatment versus active surveillance. Diagnosis is currently achieved by needle biopsy, a painful procedure. Thus, there is a clinical need for a minimally-invasive test to determine prostate cancer aggressiveness. A blood sample to predict Gleason score, which is known to reflect aggressiveness of the cancer, could serve as such a test. MATERIALS AND METHODS: Blood mRNA was isolated from North American and Malaysian prostate cancer patients/controls. Microarray analysis was conducted utilizing the Affymetrix U133 plus 2·0 platform. Expression profiles from 255 patients/controls generated 85 candidate biomarkers. Following quantitative real-time PCR (qRT-PCR) analysis, ten disease-associated biomarkers remained for paired statistical analysis and normalization. RESULTS: Microarray analysis was conducted to identify 85 genes differentially expressed between aggressive prostate cancer (Gleason score ?8) and controls. Expression of these genes was qRT-PCR verified. Statistical analysis yielded a final seven-gene panel evaluated as six gene-ratio duplexes. This molecular signature predicted as aggressive (ie, Gleason score ?8) 55% of G6 samples, 49% of G7(3+4), 79% of G7(4+3) and 83% of G8-10, while rejecting 98% of controls. CONCLUSION: In this study, we have developed a novel, blood-based biomarker panel which can be used as the basis of a simple blood test to identify men with aggressive prostate cancer and thereby reduce the overdiagnosis and overtreatment that currently results from diagnosis using PSA alone. We discuss possible clinical uses of the panel to identify men more likely to benefit from biopsy and immediate therapy versus those more suited to an "active surveillance" strategy.

Dataset Information

Random forest-based modelling to detect biomarkers for prostate cancer progression.

Publications

Random forest-based modelling to detect biomarkers for prostate cancer progression.

Similar Datasets

OmicsDI is part of the ELIXIR infrastructure