Subgroup analysis of patients with localized prostate cancer treated within the Dutch-randomized dose escalation trial

Purpose: To investigate the effect of dose escalation within prognostic risk groups in prostate cancer. Patients and methods: Between 1997 and 2003, 664 patients with localized prostate cancer were randomly assigned to receive 68- or 78-Gy of radiotherapy. Two prognostic models were examined: a risk group model (low-, intermediate-, and high-risk) and PSA-level groupings. High-risk patients with hormonal therapy (HT) were analyzed separately. Outcome variable was freedom from failure (FFF) (clinical failure or PSA nadir + 2 mu g/L). Results: In relation to the advantage of high-dose radiotherapy, intermediate-risk patients benefited most from dose escalation. However no significant heterogeneity could be demonstrated between the risk groups. For two types of PSA-level groupings: PSA < 10 and >= 10 mu g/L, and <8, 8-18 and >8 mu g/L, the test for heterogeneity was significant (p = 0.03 and 0.05, respectively). Patients with PSA 8-18 mu g/L (n = 297, HR = 0.59) derived the greatest benefit from dose escalation. No heterogeneity could be demonstrated for high-risk patients with and without HT. Conclusion: Intermediate-risk group derived the greatest benefit for dose escalation. However, from this trial no indication was found to exclude low-risk or high-risk patients from high-dose radiotherapy. Patients could be selected for high-dose radiotherapy based on PSA-level groupings: for patients with a PSA <8 mu g/L high-dose radiotherapy is probably not indicated, but should be confirmed in other randomized studies. (C) 2010 Elsevier Ireland Ltd. All rights reserved. Radiotherapy and Oncology 96 (2010) 13-18