Changing the patterns of failure for high-risk prostate cancer patients by optimizing local control

Standard therapies for high-risk prostate cancer have resulted in suboptimal outcomes with both local and distant failures. Prostate-specific antigen (PSA) and distant metastases rates as well as biopsy outcomes are reported after a regimen of trimodality therapy with hormonal, radioactive seed, and external beam radiation therapy to demonstrate how patterns of failure are changed when local control is optimized.

From 1994 to 2003, a total of 360 patients with high-risk prostate cancer were treated with trimodality therapy. Patients were defined as being at high risk if they possessed at least one of the following high-risk features: Gleason score 8 to 10, PSA>20, clinical stage t2c to t3, or two or more intermediate risk features: Gleason score 7, PSA>10 to 20, or stage t2b. Patients were followed for a median of 4.25 years (range, 2 to 10 years).

The actuarial 7-year freedom from PSA failure and freedom from distant metastases (FFDM) rates were 83% and 89% respectively. Patients (n=51) developing PSA failure exhibited aggressive disease behavior with short PSA doubling times (median, 5 months) and a 7-year freedom from distant metastases rate of 48%. Local control was high. The last posttreatment biopsy results were negative in 97% of cases (68 of 70 patients). In multivariate analysis, only PSA>20 predicted biochemical failure (p=0.04), and only seminal vesicle status predicted developing distant failure (p=0.01).

Trimodality therapy results in excellent local control that alters patterns of failure, resulting in similar actuarial biochemical and distant failure rates. Most failures appear to be distant and exhibit biologically aggressive behavior.