Improvement in relapse-free survival throughout the PSA era in patients with localized prostate cancer treated with definitive radiotherapy: Year of treatment an independent predictor of outcome

Purpose: In patients treated with radical prostatectomy in the prostate-specific antigen (PSA) era, it has been demonstrated that the year of treatment in the PSA era is associated with better pathologic parameters and outcomes, independently of other well-recognized parameters such as clinical stage, pretreatment PSA level, or Gleason score. The purpose of the present study was to study a similar phenomenon with definitive radiotherapy (RT). Methods and Materials: The inclusion criteria were as follows: clinical Stage T1-T2, available pretreatment PSA level and biopsy Gleason score, treatment delivered before January 2000 with standard fractionation external beam radiotherapy to at least 70 Gy, no adjuvant androgen deprivation (AD), all neoadjuvant AD limited to ≤6 months, and a minimum of 3 years of PSA follow-up. A total of 467 cases treated between January 1986 and December 1999 were included. Short-course AD in the adjuvant or neoadjuvant setting for ≤6 months was given in 124 cases (27%). The median radiation dose was 74 Gy (range 70.0-78.0). A conformal technique was used in 293 cases (63%). The median follow-up was 62 months (range 37-189). A total of 4931 follow-up PSA levels were available for analysis (average 11 per patient). A multivariate analysis for factors affecting biochemical relapse-free survival rates using the proportional hazards model was performed for all cases using the following variables: age (continuous variable), race (black vs. white), clinical T stage (T1-T2a vs. T2b-T2c), pretreatment PSA (continuous variable), biopsy Gleason score (continuous variable), use of AD (yes vs. no), radiation dose (continuous variable), and year of treatment (continuous variable: 1986-1999). Results: The projected 8-year biochemical relapse-free survival rate was 74%. The projected 5-year biochemical relapse-free survival rate for the 143 patients treated in the 1986-1995 period was 58% vs. 82% for the 324 patients treated in the 1996-1999 period (p <0.001). The difference was attributable to a multitude of factors (earlier stage cancer, higher radiation doses, shorter follow-up). To study the confounding effects of these factors on the year of therapy, a multivariate analysis was performed. The multivariate analysis revealed the initial PSA level (p <0.001), Gleason score (p <0.001), RT dose (p = 0.045), and year of treatment (p <0.001) to be independent predictors of outcome. Age (p = 0.41), race (p = 0.14), T stage (p = 0.10), and use of AD (p = 0.58) were not. Conclusion: When controlling for tumor, treatment, and follow-up parameters, the year in which RT was performed was still an independent predictor of outcome, consistent with observations made for radical prostatectomy patients. This indicates a more favorable presentation of localized prostate in current years probably related to a combination of factors such as screening and increased patient awareness leading to earlier diagnosis. Outcome predictions should be based on contemporaneous series.