Purpose: A prospective phase 3 trial was initiated to determine whether 8 compared with 3-month neoadjuvant hormonal therapy reduces prostate specific antigen (PSA) recurrence rates after radical prostatectomy. Our interim analysis includes secondary end points of differences in biochemistry, pathology and adverse events between the 2 groups.
Materials and methods: Men with clinically confined prostate cancer were randomized to receive 7.5 mg. leuprolide intramuscularly monthly and 250 mg. flutamide orally 3 times daily for 3 or 8 months before radical prostatectomy. Our study was powered to detect a 35% decrease in PSA recurrence, assuming a 30% recurrence rate in the 3-month arm after 3 years.
Results: A total of 547 men were randomized between August 1995 and April 1998. Men in the 8 and 3-month groups were equally stratified for T stage (29% T1c, 70% T2), Gleason grade (68% less than 4, 32% 4 or greater) and pretreatment PSA (63% less than 10, 27% 10 to 20 and 10% greater than 20 microg./l.). Mean pretreatment PSA was slightly higher in the 8-month compared with the 3-month group (11.64 versus 9.95 microg./l., respectively, p = 0.0539). A total of 44 men withdrew from study before surgery and, therefore, were nonevaluable. Preoperative PSA nadir was less than 0.1 microg./l. in 43.3% versus 75.1% (p <0.0001), and 0.3 microg./l. or greater in 21% versus 9.2% after 3 versus 8 months, respectively (p <0.0006). Mean serum PSA decreased 98% to 0.12 microg./l. after 3 months, with a further 57% to 0.052 microg./l. from 3 to 8 months. Transrectal ultrasound determined that prostatic volume decreased 37% from a mean of 40.6 to 25.4 cc after 3-month neoadjuvant hormonal therapy (p = 0.0001) and a further 13% to 22.2 cc after 8 months (p = 0.03). Mean hemoglobin decreased 15% (148.2 to 125.4 gm./dl.) after 3-month neoadjuvant hormonal therapy but stabilized thereafter. Radical prostatectomy was completed in 500 men, while surgery was aborted intraoperatively in 3. Positive margin rates were significantly lower in the 8 than 3-month group (12% versus 23%, respectively, p = 0.0106). There were no fatal adverse events and no differences between the 2 groups in the severity or causality (p = 0.287, 0.0564) of adverse events, or incidence of increased liver enzymes or diarrhea (p = 0.691, 0.288, respectively). However, men in the 8-month group noticed a higher number of newly reported adverse events (4.5 versus 2.9, p <0.0001) and higher incidence of hot flushes than the 3-month group (87% versus 72%, respectively, p <0.0001).
Conclusions: Ongoing biochemical and pathological regression of prostate tumors occurs between 3 and 8 months of neoadjuvant hormonal therapy, suggesting that the optimal duration of neoadjuvant hormonal therapy is longer than 3 months. Longer followup is needed to determine whether longer therapy alters PSA recurrence rates.