Abstract LB-371: ERG protein expression in diagnostic prostate cancer tissues among West African men

Background: Fusion of androgen regulated TMPRSS2 gene and the ERG oncogene is a common recurrent chromosomal aberration in prostate cancer, which leads to elevated expression of ERG oncoprotein. ERG expression detected by immunohistochemistry (IHC) has been validated as a proxy of TMPRSS2-ERG gene fusion. The prevalence of TMPRSS2-ERG gene fusion has been shown to vary by race being highest in Caucasian prostate cancer patients, followed by African American and then Asian. However, the prevalence of TMPRSS2-ERG gene fusion in African prostate cancer populations remains to be determined. Methods and Materials: There were 428 prostate cancer patients diagnosed during 2004-2012 in the Ghana Prostate Study who had formalin-fixed paraffin-embedded needle biopsy tissues available for analysis. Pathologists centrally scored 262 tissue samples for ERG expression and Gleason patterns, after excluding 166 without sufficient tissues. ERG-positive nuclear immunostaining of normal endothelial cells was used as an internal control and only nuclear immunostaining of prostate cancer cells was deemed ERG-positive. We evaluated ERG expression status in relation to clinicopathological characteristics using logistic regression models. To assess potential selection bias, we performed sensitivity analyses comparing clinicopathological characteristics extracted from medical records by inclusion/exclusion status. Results: We found 47 (18%) tissue samples were ERG-positive. Since recently preserved tissues appeared more likely to be ERG-negative (P = 0.044), all logistic regression models were adjusted for calendar year of diagnosis. Negative ERG staining was significantly associated with higher Gleason score (P = 0.035), in particular with higher primary Gleason grade (P = 0.028). Age at diagnosis, PSA concentration at diagnosis, and tribal group were not significantly associated with ERG expression status. Sensitivity analyses revealed that excluded tissues were more likely to be preserved in earlier years (P = 0.033) and to have lower Gleason scores (P = 0.003). Conclusions: This first study of ERG expression in an African prostate cancer population indicates that the prevalence of TMPRSS2-ERG gene fusion is similar to that of African American patients but lower than those with European ancestry. Whether this lower proportion of TMPRSS2-ERG gene fusion leads to a greater proportion of aggressive prostate cancer in African men or is the result of diagnostic selection of a higher proportion of symptomatic, high Gleason score prostate cancers in such populations remains unknown. Citation Format: Cindy Ke Zhou, Denise Young, Edward D. Yeboah, Richard B. Biritwum, Andrew A. Adjei, Yao Tettey, Evelyn Tay, Shelley Niwa, Ann L. Truelove, Isabell A. Sesterhenn, Shiv Srivastava, Robert N. Hoover, Ann W. Hsing, Michael B. Cook. ERG protein expression in diagnostic prostate cancer tissues among West African men. [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 LB-371.