Short-term androgen suppression and radiotherapy versus intermediate-term androgen suppression and radiotherapy, with or without zoledronic acid, in men with locally advanced prostate cancer (TROG 03.04 RADAR): an open-label, randomised, phase 3 factorial trial

Background We investigated whether 18 months of androgen suppression plus radiotherapy, with or without 18 months of zoledronic acid, is more effective than 6 months of neoadjuvant androgen suppression plus radiotherapy with or without zoledronic acid. Methods We did an open-label, randomised, 2 × 2 factorial trial in men with locally advanced prostate cancer (either T2a N0 M0 prostatic adenocarcinomas with prostate-specific antigen [PSA] ≥10 μg/L and a Gleason score of ≥7, or T2b–4 N0 M0 tumours regardless of PSA and Gleason score). We randomly allocated patients by computer-generated minimisation—stratified by centre, baseline PSA, tumour stage, Gleason score, and use of a brachytherapy boost—to one of four groups in a 1:1:1:1 ratio. Patients in the control group were treated with neoadjuvant androgen suppression with leuprorelin (22·5 mg every 3 months, intramuscularly) for 6 months (short-term) and radiotherapy alone (designated STAS); this procedure was either followed by another 12 months of androgen suppression with leuprorelin (intermediate-term; ITAS) or accompanied by 18 months of zoledronic acid (4 mg every 3 months for 18 months, intravenously; STAS plus zoledronic acid) or by both (ITAS plus zoledronic acid). The primary endpoint was prostate cancer-specific mortality. This analysis represents the first, preplanned assessment of oncological endpoints, 5 years after treatment. Analysis was by intention-to-treat. This trial is registered with ClinicalTrials.gov, number NCT00193856. Findings Between Oct 20, 2003, and Aug 15, 2007, 1071 men were randomly assigned to STAS (n=268), STAS plus zoledronic acid (n=268), ITAS (n=268), and ITAS plus zoledronic acid (n=267). Median follow-up was 7·4 years (IQR 6·5–8·4). Cumulative incidences of prostate cancer-specific mortality were 4·1% (95% CI 2·2–7·0) in the STAS group, 7·8% (4·9–11·5) in the STAS plus zoledronic acid group, 7·4% (4·6–11·0) in the ITAS group, and 4·3% (2·3–7·3) in the ITAS plus zoledronic acid group. Cumulative incidence of all-cause mortality was 17·0% (13·0–22·1), 18·9% (14·6–24·2), 19·4% (15·0–24·7), and 13·9% (10·3–18·8), respectively. Neither prostate cancer-specific mortality nor all-cause mortality differed between control and experimental groups. Cumulative incidence of PSA progression was 34·2% (28·6–39·9) in the STAS group, 39·6% (33·6–45·5) in the STAS plus zoledronic acid group, 29·2% (23·8–34·8) in the ITAS group, and 26·0% (20·8–31·4) in the ITAS plus zoledronic acid group. Compared with STAS, no difference was noted in PSA progression with ITAS or STAS plus zoledronic acid; however, ITAS plus zoledronic acid reduced PSA progression (sub-hazard ratio [SHR] 0·71, 95% CI 0·53–0·95; p=0·021). Cumulative incidence of local progression was 4·1% (2·2–7·0) in the STAS group, 6·1% (3·7–9·5) in the STAS plus zoledronic acid group, 1·5% (0·5–3·7) in the ITAS group, and 3·4% (1·7–6·1) in the ITAS plus zoledronic acid group; no differences were noted between groups. Cumulative incidences of bone progression were 7·5% (4·8–11·1), 14·6% (10·6–19·2), 8·4% (5·5–12·2), and 7·6% (4·8–11·2), respectively. Compared with STAS, STAS plus zoledronic acid increased the risk of bone progression (SHR 1·90, 95% CI 1·14–3·17; p=0·012), but no differences were noted with the other two groups. Cumulative incidence of distant progression was 14·7% (10·7–19·2) in the STAS group, 17·3% (13·0–22·1) in the STAS plus zoledronic acid group, 14·2% (10·3–18·7) in the ITAS group, and 11·1% (7·6–15·2) in the ITAS plus zoledronic acid group; no differences were recorded between groups. Cumulative incidence of secondary therapeutic intervention was 25·6% (20·5–30·9), 28·9% (23·5–34·5), 20·7% (16·1–25·9), and 15·3% (11·3–20·0), respectively. Compared with STAS, ITAS plus zoledronic acid reduced the need for secondary therapeutic intervention (SHR 0·67, 95% CI 0·48–0·95; p=0·024); no differences were noted with the other two groups. An interaction between trial factors was recorded for Gleason score; therefore, we did pairwise comparisons between all groups. Post-hoc analyses suggested that the reductions in PSA progression and decreased need for secondary therapeutic intervention with ITAS plus zoledronic acid were restricted to tumours with a Gleason score of 8–10, and that ITAS was better than STAS in tumours with a Gleason score of 7 or lower. Long-term morbidity and quality-of-life scores were not affected adversely by 18 months of androgen suppression or zoledronic acid. Interpretation Compared with STAS, ITAS plus zoledronic acid was more effective for treatment of prostate cancers with a Gleason score of 8–10, and ITAS alone was effective for tumours with a Gleason score of 7 or lower. Nevertheless, these findings are based on secondary endpoint data and post-hoc analyses and must be regarded cautiously. Long- term follow-up is necessary, as is external validation of the interaction between zoledronic acid and Gleason score. STAS plus zoledronic acid can be ruled out as a potential therapeutic option. Funding National Health and Medical Research Council of Australia, Novartis Pharmaceuticals Australia, Abbott Pharmaceuticals Australia, New Zealand Health Research Council, New Zealand Cancer Society, University of Newcastle (Australia), Calvary Health Care (Calvary Mater Newcastle Radiation Oncology Fund), Hunter Medical Research Institute, Maitland Cancer Appeal, Cancer Standards Institute New Zealand.