Abstract 5019: Impact of the ERG status on the prognostic relevance of prostate cancer biomarkers: a survey of 13,000 patients

Abstract

Abstract The TMPPRS:ERG fusion is the most frequent genomic alteration in prostate cancer leading to androgen dependent up-regulation of the ERG transcription factor. ERG activation deregulates more than 1,600 ERG-responsive genes, and some molecular alterations have been linked to either the subset of ERG negative or ERG positive cancers. While studies in large cohorts of prostate cancers treated by radical prostatectomy have demonstrated that ERG fusion alone has no impact on patient outcome, there is growing evidence that the prognostic relevance of many other biomarkers may be different in cancers with and without ERG fusion. In order to better understand the impact of the ERG status on the prognostic relevance of prostate cancer biomarkers, we took advantage of our molecular database obtained from immunohistochemistry and fluorescence in situ hybridization (FISH) analyses of more than 50 markers with respect to expression levels and gene copy number changes in 13,665 prostate cancers that had been assembled in a tissue microarray (TMA) format. Markers with strongest associations to presence of ERG fusion included CRISP3, HOOK3, RBM3, LPCAT1, BAZ2A, and deletions of 3p and PTEN (p<0.0001 each). Markers with strongest associations to absence of ERG fusion included SPINK1 and deletions of 6q and 5q (p<0.0001 each). Alterations of 47 markers were linked to reduced time to biochemical recurrence if all cancers were analyzed regardless if the ERG status. However, subset analyses revealed, that the prognostic relevance of 18 (38%) of these 47 markers was driven either by the subset of ERG positive (n = 7) or ERG negative cancers only (n = 11). Combined analysis of these markers identified ERG status dependent molecular classifiers that provided prognostic information exclusively in ERG negative or in ERG positive prostate cancers. Multivariate modeling indicated that the classifiers were independent of established prognostic factors, including pre-surgical (PSA, Gleason grade on biopsy, clinical stage) and post-surgical (pT stage, Gleason of the radical prostatectomy, nodal stage, resection margin stage) parameters. In conclusion, the results of our study demonstrate that the impact of many candidate prognostic markers in prostate cancer strongly depends on the ERG status. Our growing molecular database attached to more than 13,000 prostate cancers allows for the continuous validation of candidate prognostic markers and their impact in molecularly defined subsets of the disease. Citation Format: Ronald Simon, Martina Kluth, Claudia Hube-Magg, Maria Christina Tsourlakis, Sarah Minner, Christoph Burdelski, Katharina Grupp, Nathaniel Melling, Asmus Heumann, Christina Koop, Thomas Steuber, Markus Graefen, Hartwig Huland, Guido Sauter, Thorsten Schlomm. Impact of the ERG status on the prognostic relevance of prostate cancer biomarkers: a survey of 13,000 patients. [abstract]. In: Proceedings of the 107th Annual Meeting of the American Association for Cancer Research; 2016 Apr 16-20; New Orleans, LA. Philadelphia (PA): AACR; Cancer Res 2016;76(14 Suppl):Abstract nr 5019.