Clinical implications of TMPRSS2-ERG gene fusion expression in patients with prostate cancer treated with radical prostatectomy.

PURPOSE: Molecular prognostic factors may be useful tools for prostate cancer that complement classic clinicopathological factors. Genetic rearrangements between TMPRSS2 and ETS have been described for prostate cancer but their clinical significance is still unclear. We analyzed the association of the TMPRSS2-ERG fusion gene with prostate cancer outcome in patients treated with radical prostatectomy. MATERIAL AND METHODS: We analyzed prostate cancer samples from 226 patients treated with radical prostatectomy from 1996 to 2002 with a median followup of 84 months (range 9 to 153). TMPRSS2-ERG fusion gene expression was determined by reverse transcriptase-polymerase chain reaction. Clinicopathological and molecular variables were related to biochemical and clinical progression-free survival by the Kaplan-Meier proportional risk log rank test. A Cox proportional hazards model using stepwise selection was used to identify independent predictors of poor outcome. RESULTS: TMPRSS2-ERG fusion was detected in 114 cases (50.4%). We noted no association between fusion gene status and prostate cancer clinicopathological characteristics. However, when patients were grouped by TMPRSS2-ERG fusion gene status, different clinicopathological prognostic factors defined each group for biochemical and clinical progression-free survival. Prostate specific antigen, specimen Gleason score and margin status were independent prognostic factors in patients with prostate cancer expressing the fusion gene. In the nonexpressing TMPRSS2-ERG group the prognostic factors were cT, Gleason score and margins. CONCLUSIONS: TMPRSS2-ERG fusion gene status classifies patients with prostate cancer treated with radical prostatectomy into groups defined by different prognostic factors. This could be the basis for designing more refined treatment strategies.