Defining a dose-response relationship with radiotherapy for prostate cancer: Is more really better?

Purpose: Data were reviewed addressing the association between radiation th erapy (RT) dose and treatment outcome for localized prostate cancer to help clarify the existence of a potential dose-response relationship. Methods and Materials: Articles were identified through the MEDLINE databas e, CancerLit database, and reference lists of relevant articles. Studies we re categorized into four groups based upon the endpoint analyzed, including biochemical control (BC), local control (LC), pathologic control (PC), and cause-specific survival (CSS). The impact of increasing RT dose with each endpoint was recorded. Results: Twenty-two trials involving a total of 11,297 patients were identi fied. Of the 11 trials addressing the association of RT dose with LC, 9 sho wed statistically significant improvements'. Of the 12 trials that reported BC with RT dose, all showed statistically significant improvements. Two ou t of 4 studies analyzing PC with increasing dose showed a positive correlat ion. Finally, 3 out of 9 studies addressing RT dose with CSS showed statist ically significant improvements. Despite inconclusive results, patients wit h poor risk features (e.g., prostate-specific antigen [PSA] greater than or equal to 10, Gleason score [GS] greater than or equal to 7, or tumor stage greater than or equal to T2b) were most likely to benefit from increasing dose with respect to each endpoint. However, the optimal RT dose and the ma gnitude of benefit of dose escalation could not be identified. Conclusions: Although RT dose appears to correlate with various measures of treatment outcome, objective, high-quality data addressing this critical i ssue are still lacking. At the present time, the absolute improvement in ou tcome due to dose escalation, the subset of patients benefitting most, and the optimal dose remain to be defined. (C) 2001 Elsevier Science Inc.