A Prospective Study Assessing the Predictive Value of TMPRSS2-ERG Gene Fusion and PTEN Deletion in High Risk Prostate Cancer Patients

One of the biggest problems facing prostate cancer patients and their treating physicians is who needs to be treated and when. Common clinical and pathological parameters are useful (PSA, Gleason score, etc.) but do not clearly predict who will benefit from treatment and who will fail. Genetic markers for tumor aggressivity would be of greater value. The finding that the TMPRSS2-ERG gene fusion is associated with an increase risk of cancer progression is important. TMPRSS2 is controlled by androgen (testosterone) and ERG is part of a family of proteins which have a role in controlling cell growth, cell specialization and producing tumors. As a consequence of this gene fusion, production of the ERG protein increases in the presence of testosterone and could be key to the development of prostate cancer, resistance to treatment and poor outcome. The PTEN gene is known to have a role as a tumor suppressor. Its deletion is a contributing factor in the development of prostate cancers and poor outcome. The coexistence of the two markers could be associated with a higher risk of recurrence.

To date there have been no studies regarding the presence of either of these two markers or their coexistence in high risk prostate cancer patients who, despite radiation therapy and androgen suppression, develop biochemical failure (their PSA levels rise once again). Patients participating in the PCS IV study (high risk prostate cancer treated with radiation therapy plus either 18 or 36 months of hormonal suppression) who have had biochemical failure or 3 years of follow-up post hormonal therapy will be approached.

Tumor blocks from consenting patients will be collected and analyzed for the presence of the TMPRSS2-ERG gene fusion and the PTEN deletion at the Pathology Department of the Jewish General Hospital. Statistical analysis will be carried out to see whether either or both markers are present, whether they are associated with certain clinical and pathological high risk factors, and whether they can be used to predict which patients will fail treatment.