Abstract 1568: Clinical implications of tmprss2-erg fusion and tumor location in prostate cancer

Abstract

Abstract Introduction: There is limited data on the impact of race on tumor location and TMPRSS2-ERG (ERG) gene fusion status in Prostate Cancer (PCa). African American men (AAM) often lack the ERG fusion and are likely to experience poor outcomes. Nonetheless, the extent to which ERG status, race, and tumor location impacts PCa outcomes remains unclear. Methods: This multi-institutional study included a retrospective cohort of 495 AAM and 422 European American men (EAM) who received prostatectomy at Moffitt Cancer Center (MCC) and University of Pennsylvania Health system (UPHS). To assess the race-dependent variation in ERG protein expression and its association with tumor location and PCa pathology, ERG antibody staining on 4 micron sections of FFPE blocks were carried out on all samples. ERG expression was categorized as ERGpos and ERGneg by two pathologists. Additionally, the staining pattern of ERG was analyzed within ERGpos tumors. A review of anatomic location of the dominant tumor lesions was performed and analyzed using methods of categorical analysis and logistic regression. Alpha value of < 0.05 was considered as statistically significant. Results: In all samples, analysis of ERG immunohistochemistry (IHC) yielded 39.7% ERGpos and 60.6% ERGneg. Variability in histopathologic expression of ERG was significant between races with AAM being more likely to lack TMPRSS2-ERG fusion (ERGneg) compared to EAM, (62% vs 38%, p < 0.001). Lack of ERG fusion was also associated with anatomically distinct tumor location with 72.1% of anteriorly located tumors having ERGneg staining, while only 27.9% were ERGpos, (p < 0.001). Among ERGpos patients, AAM (65.2%) showed significant heterogeneous expression of ERG compared to EAM (31.8%). Furthermore, ERGneg status was associated with higher pretreatment PSA. In a multivariate logistic regression model adjusted for tumor location and race, ERGneg tumors were more likely to have Gleason Score 4+3=7 (Grade Group 3) or higher compared with ERGpos tumors, Odds ratio 2.07, 95% CI 1.20 - 3.56, p = 0.008. Conclusions: The absence of TMPRSS2-ERG fusion is associated with higher grade disease and is a strong predictor for anatomic tumor location within the prostate gland. The clinical implications of ERGneg tumors among AAM and its association with anteriorly located tumors can be vital in addressing disparities in PCa. Citation Format: Kosj Yamoah, Shivanshu Awasthi, Priti Lal, Arash O. Naghavi, Shelly Mahajan, Jong Y. Park, Rob Rounbehler, Travis Gerke, Anders Berglund, Julio PowSang, Jasreman Dhillon, Timothy R. Rebbeck. Clinical implications of tmprss2-erg fusion and tumor location in prostate cancer [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2019; 2019 Mar 29-Apr 3; Atlanta, GA. Philadelphia (PA): AACR; Cancer Res 2019;79(13 Suppl):Abstract nr 1568.