Adverse pathologic features - including positive surgical margins, extracapsular extension, and seminal vesicle invasion - are common findings at radical prostatectomy for clinically localised disease. In the absence of further treatment, patients with such features are at high risk of biochemical and local recurrence and the development of metastatic disease. Radiotherapy is the only known curative treatment modality in patients with locally recurrent disease. Its optimal use following surgery, however, is controversial. In this review, the evidence for immediate (adjuvant) and selective early salvage approaches to postoperative radiotherapy are presented. The results of three randomized controlled trials comparing adjuvant radiotherapy to observation are discussed. The trial with the longest follow-up now demonstrates that, compared to a policy of observation, adjuvant radiotherapy prolongs overall survival and prevents distant metastases. It is not clear, however, whether this survival advantage would remain if adjuvant radiotherapy were compared to a strategy of close monitoring with salvage radiotherapy instituted at the time of biochemical failure. A selective salvage strategy has the advantage of avoiding ‘unnecessary’ radiotherapy, and its toxicity, for those patients destined not to fail after surgery. Large-scale randomized controlled trials comparing these two approaches are underway.