Abstract Clinical incidence and mortality rates of prostate cancer vary dramatically among different countries. In a recent study, we associated the difference in clinical incidence with certain somatic genomic alterations, including PTEN deletion and TMPRSS2:ERG fusion, which are found more commonly in Western but rarely in Chinese cancer samples (Mao et al, Cancer Res, 2010; 70:5207-12). TMPRSS2:ERG fusion, which is transcriptionally controlled by androgen receptor (AR), has been reported in more than 50% of prostate cancers. We investigated whether androgen level and AR activity, which varies between Chinese and Western men, could induce this fusion gene and consequently prostate cancer development. Since AR activity is inversely associated with its exon1 CAG repeats length, we determined the differences in repeat length between TMPRSS2:ERG fusion positive and negative UK samples and also between UK and China prostate cancer patients. We found significantly shorter CAG repeat lengths in the UK compared to Chinese cases (P<0.05). We also observed short CAG repeat lengths in the fusion positive (average n=20.3) compared to negative (average n=21.3) cases, although it is not statistically significant (p=0.14) and potentially due to the limited number of samples. We then investigated whether AR stimulation by androgen can induce TMPRSS2:ERG fusion and spatial genome reorganization. We induced the fusion not only in prostate cancer cells, but also in non-malignant prostate epithelial cells following long-term exposure to dihydrotestosterone (DHT). The induced fusion is associated with androgen-stimulated TMPRSS2 and ERG gene proximity. During our study, androgen-induced TMPRSS2 and ERG proximity and fusion were reported, but only in cancer cells (Mani et al, Science, 2009; 326:1230; Lin et al, Cell, 2009; 139:1069-83; and Haffner et al, Nat Genet, 2010; 42:668-75). We also observed that androgen-stimulated gene proximity is associated with AR transcriptional activity, which can be blocked by either RNA polymerase or AR inhibitors. Finally, we investigated the role of double strand breaks in the induction of TMPRSS2:ERG fusion. We associated high frequency of fusion induction with low expression of PIWIL1, which acts as a double strand breaks protector. Overexpression of PIWIL1 prevents DHT induced TMPRSS2:ERG fusion. Interestingly, long term DHT exposure tends to reduce expression of PIWIL1. This study demonstrates that enhanced AR activity can induce genetic alterations such as TMPRSS2:ERG fusion and consequently result in prostate carcinogenesis. The difference in physiological androgen level and AR activity may account for the variation in prostate cancer risk between Western and Chinese men and suggests that control of androgen level or AR activity can potentially prevent prostate carcinogenesis. Citation Format: {Authors}. {Abstract title} [abstract]. In: Proceedings of the 102nd Annual Meeting of the American Association for Cancer Research; 2011 Apr 2-6; Orlando, FL. Philadelphia (PA): AACR; Cancer Res 2011;71(8 Suppl):Abstract nr 3048. doi:10.1158/1538-7445.AM2011-3048