[854] TMPRSS2-ERG Gene Fusion Defines a Metastatic Phenotype of Prostate Cancer

S Perner, M Svensson, R Hossain, J Day, J Groskopf, MD Hofer, R Kuefer, DS Rickman, MA Rubin. Weill Cornell Medical Center, New York, NY; Gen-Probe, Inc., San Diego, CA; University Hospital of Ulm, Ulm, Germany; Brigham and Women's Hospital, Boston, MA

Background: Histologic and molecular heterogeneity is a hallmark of prostate cancer (PCa), suggesting that distinct cancer foci may arise independently. Previously work demonstrated that clonal intrafocal homogeneity and interfocal heterogeneity of TMPRSS2-ERG fusion status exists in multifocal PCa. This poses a significant limitation to the development of prognostic and predictive biomarkers as diagnostic needle biopsies may inaccurately represent the driving tumor(s). PCa progression is closely tied to the development of metastatic disease and previous work suggests that TMPRSS2-ERG fused PCa have a more aggressive natural history. The aim of this study was to interrogate multifocal PCa to determine if the TMPRSS2-ERG tumor foci were more apt to metastasize than fusion negative foci.
Design: We studied samples from 33 PCa patients who underwent radical prostatectomy and lymphadenectomy and that had at least two distinct cancer foci in the prostate and at least one lymph node (LN) metastasis. Each focus (PCa and LN) was assigned a primary and secondary Gleason grade and assessed for fusion status by fluorescence in-situ hybridization (FISH). Fusion transcript was assessed by quantitative RT-PCR in a subset of cases.
Results: 15 out of the 26 cases (58%) exhibited interfocal homogeneity with regard to fusion status (9 cases being fusion negative in all foci and 6 cases being fusion positive in all foci). Two out of the 26 cases revealed interfocal heterogeneity with regard to fusion mechanism. 9 out of the 26 cases (42%) revealed interfocal heterogeneity with regard to fusion status. In all PCa cases that were fusion positive regardless of diameter we found that the corresponding LN metastasis were also fusion positive. The fusion status in the LN metastasis did not necessarily correspond to the fusion status of the largest PCa focus or to the fusion status of the focus with the highest Gleason score.
Conclusions: In this unique cohort, we found that tumor size and Gleason score was not the best predictor of pelvic LN metastasis. In this cohort, TMPRSS2-ERG fusion status was the strongest determinant of metastatic dissemination. These findings may have important clinical implications for the screening and diagnosis of PCa.
Category: Genitourinary (including renal tumors)

Monday, March 9, 2009 1:00 PM

Poster Session II # 110, Monday Afternoon