Report of a multicenter Canadian phase III randomized trial of 3 months vs. 8 months neoadjuvant androgen deprivation before standard-dose radiotherapy for clinically localized prostate cancer.

Purpose: To evaluate the effect of 3 months vs. 8 months of neoadjuvant hormonal therapy before conventional dose radiotherapy (RT) on disease-free survival using prostate-specific antigen PSA and biopsies as end points for clinically localized prostate cancer.

Methods And Materials: Between February 1995 and June 2001, 378 men were randomized to either 3 or 8 months of flutamide and goserelin before conventional-dose RT (66 Gy) at four participating centers. The median patient age was 72 years (range, 50-84 years). The stage distribution was 17% T1c, 35% T2a, 34% T2b-T2c, 13% T3-T4. The Gleason score (GS) was < or =6 in 51%, 7 in 38%, and 8-10 in 11%. The median baseline PSA level was 9.7 ng/mL (range, 1.3-189 ng/mL). Of the 378 men, 26% were low risk (Stage T1c-T2a, GS < or =6, PSA <10 ng/mL), 43% were intermediate risk (Stage T2b or GS 7 or PSA 10-20 ng/mL), and 31% were high risk (Stage T3 or GS 8-10 or PSA >20 ng/mL). The two arms were balanced in terms of age, GS, T stage, risk group, and presenting PSA level. The median follow-up was 44 months (range, 10-84 months), and 361 patients were available for evaluation.

Results: The 8-month arm achieved a lower PSA level before starting RT (0.37 vs. 0.74 ng/mL, p < or =0.001) and had a greater downsizing of the prostate (mean volume 26.6 cm(3) vs. 30.5 cm(3), p < or =0.001). However, the actuarial freedom from failure rate (biochemical by American Society for Therapeutic Radiology and Oncology definition, local or distant) for the 3-month vs. 8-month arms at 3 years was 66% vs. 68% and by 5 years was 61% vs. 62%, respectively (p = 0.36). No statistically significant difference was noted in the types of failure between the two arms (crude final status): biochemical, 22.2% vs. 22.3%; local, 10.2% vs. 6.5%; and distant, 3.4% vs. 4.4% (p = 0.61). Two-year post-RT biopsies were done in 57% (n = 205). Negative biopsies were obtained in 68% of the 3-month and 77% of the 8-month patients; 18% and 14% had indeterminate biopsies and 14% and 9% were positive for residual cancer (p = 0.34) in the two arms, respectively. The median PSA level for nonfailing patients was 0.50 ng/mL in both the 3-months and 8-month arms. A suggestion of improvement was found in the 8-month arm for disease-free survival at 5 years for high-risk patients (39% vs. 52%) but did not achieve statistical significance.

Conclusion: A longer period of neoadjuvant hormonal therapy before standard-dose RT does not appear to confer a benefit in terms of disease-free survival or to alter failure patterns. Failure was delayed in the 8-month arm, but this advantage was lost by 5 years of follow-up. A suggestion of benefit was noted with a longer period of hormonal therapy for high-risk patients.