Abstract LB-477: Micro-RNA expression patterns to classify prostate cancer risk

Abstract

Abstract Progression risk from early stage localized prostate cancer (CaP) to more aggressive and lethal phenotypes is not precisely predicted. This leads to the over-treatment of men whose disease might otherwise remain indolent. Tumor and serum microRNA (miR) expression show promise as functional biomarkers of CaP progression risk but considerable ambiguity exists over which miR are altered, in which direction and at what point in disease. To address these ambiguities we undertook a genome-wide miR expression comparison using a micro-array approach in key and isogenic CaP cell lines and non-malignant controls. Specifically, cell pairs that represent; Initiation (non-malignant RWPE-1 and RAS-transformed RWPE-2 cells); Progression (non-malignant HPr-1AR and LNCaP cells); Recurrence (LNCaP and LNCaP-C4-2 cells). Comparison between these cell line pairs identified 56 altered miR (≥1.5 fold, p<0.05). Specifically, 11 were altered in Initiation (2 down and 9 up-regulated), 43 in Progression (33 down and 10 up-regulated) and 19 in Recurrence (3 down and 16 up-regulated). Technical validation by Q-PCR supported these findings, as did biological validation in a test-set of 36 locally confined CaP tumors and matched normal pairs. Q-PCR analyses in a longitudinal tumor series from the murine CaP TRAMP model revealed that the expression of murine homologs of several miR (e.g. miR-138 and 205) was significantly altered prior to animals displaying palpable tumors. These findings supported further their utility as early stage disease markers. Importantly, biological validation included analyses in a publically available micro-array data-base of miR from 78 human locally tumors (cBio Cancer Genomics Portal). Approximately 800 miR were in common between the two micro-array platforms and 46, of the 56 miR identified in our cell line studies, were very significantly altered in these primary tumors. Enrichment analysis identified the TGF-β pathway as a key miR-targeted network and supported enhanced TGF-β signaling in the CaP cell lines. To test the ability of miR to discern disease risk at early stage we investigated the expression and association of the 56 miR identified above, with the 24 month risk of biochemical progression following radical prostatectomy, using a cohort of men who had surgery at RPCI. We measured expression of the 56 miR by Q-PCR in 33 men who displayed biochemically progression and 67 non-progressed CaP control patients. Samples were matched on age, PSA levels and Gleason score. We are now establishing the association of individual and multiple serum miR expression at baseline with stage, grade, pre-surgery PSA levels and risk of progression. In parallel we measured the expression of the 56 miR in the tumors of 32 men within this cohort, to measure tumor serum correlations. These analyses will determine the serum miR expression that best differentiate i) men by CaP prognostic variables and ii) who experience biochemical progression. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 103rd Annual Meeting of the American Association for Cancer Research; 2012 Mar 31-Apr 4; Chicago, IL. Philadelphia (PA): AACR; Cancer Res 2012;72(8 Suppl):Abstract nr LB-477. doi:1538-7445.AM2012-LB-477