Factors affecting recurrence rates after prostatectomy or radiotherapy in localized prostate carcinoma patients with biopsy Gleason score 8 or above

BACKGROUND: To review the biochemical recurrence-free survival (bRFS) rates of treatment with either external beam radiotherapy or radical prostatectomy in patients with biopsy Gleason score 8 or above in the prostate specific antigen (PSA) era. METHODS: A total of 297 localized prostate carcinoma patients diagnosed with biopsy Gleason score 8 or above were treated between 1987 and 2000 at the Cleveland Clinic with either prostatectomy or radiotherapy (RT). All patients had pretreatment PSA (iPSA) levels. Androgen deprivation was given as part of the initial treatment in 154 patients (52%). Radical prostatectomy (RP) was the primary treatment in 115 patients (39%) and 182 patients (61%) received RT.The median radiation dose was 70.2 Gy (range, 60.0-78.0 Gy). The median follow-up time was 42 months (range, 1-153 months). RESULTS: For the 297 patients, the 5 and 8-year bRFS rates were 45% (95% confidence interval [CI] 38-52%) and 31% (95% CI 20-42%), respectively. The 5-year bRFS rates for iPSA 10 or less versus. iPSA above10 were 66% (95% CI 54-78%) versus 31% (95% CI 22-40%), respectively (P < 0.001). The 5-year bRFS rates for use of androgen deprivation versus no androgen deprivation were 62% (95% CI 52-72%) versus 35% (95% CI 27-42%), respectively (P < 0.001). The 5-year bRFS rates for RT patients versus RP patients were 47% (95% CI 38-57%) versus 42% (95% CI 31-53%), respectively (P = 0.051). For RT patients, the 5-year bRFS rates for use of androgen deprivation versus no androgen deprivation were 71% versus 29%, respectively (P < 0.001). For RP patients, the 5-year bRFS rates for use of androgen deprivation versus no androgen deprivation were 39% versus 43%, respectively (P = 0.90). Multivariate time-to-failure analysis using the proportional hazards model showed iPSA level (P < 0.001) and the use of androgen deprivation (P = 0.001) to be the only independent predictors of biochemical recurrence. Age (P = 0.44), race (P = 0.80), clinical T stage (P = 0.10), biopsy Gleason scores (P = 0.39), and treatment modality (P = 0.13) were not independent predictors of biochemical failure. A total of 104 patients had Stage T1 or T2 disease, low iPSA levels (/= 8). For all 104 patients, the 5-year bRFS rate was 64%. For 52 of the 104 patients who received 6 months of androgen deprivation or less, the 5-year bRFS rate was 78%. CONCLUSION: Patients with localized prostate carcinoma with a biopsy Gleason score 8 or less have lower recurrence rates if iPSA levels are 10 or less. Biochemical control rates were encouraging for patients with biopsy Gleason score 8 or above, clinical Stage T1-T2, and iPSA levels less than or equal to 10 ng/mL treated with adjuvant androgen deprivation given only for 6 months or less. If longer follow-up confirms these findings, these patients might not need prolonged androgen deprivation for periods exceeding 6 months following local therapy.