Objective To systematically review the effectiveness and safety of radiotherapy combined with short-term or long-term hormonal therapy for prostate cancer. Methods Databases including EMbase, PubMed, Web of Science, CENTRAL and CBM were searched from inception to August 2012 to collect the randomized controlled trials (RCTs) on radiotherapy combined with short-term or long-term hormonal therapy for prostate cancer. According to the inclusion and exclusion criteria, data of the included studies were extracted, and the methodological quality was evaluated. Then meta-analysis was performed using RevMan 5.1, and the evidence qualities and recommendation levels were determined according to the GRADE System. Results A total of 6 RCTs involving 3157 patients were included. The results of meta-analysis showed that there were no significant differences in the overall survival rate (RR=0.95, 95%CI 0.91 to 1.00) and the disease-free survival rate (RR=0.73, 95%CI 0.46 to 1.13) between the radiotherapy plus short-term hormonal therapy group (the short-term group) and the radiotherapy plus long-term hormonal therapy group (the long-term group). The long-term group was superior to the short-term group in biochemical failure-free survival rate (RR=0.81, 95%CI 0.68 to 0.97), clinical progression rate (RR=1.61, 95%CI 1.44 to 1.80), and prostate cancer-specific mortality (RR=1.44, 95%CI 1.16 to 1.80). Based on the GRADE system, the evidence level of biochemical failure-free survival was moderate with a weak recommendation; the evidence level of disease-free survival was low with a weak recommendation; the evidence level of overall survival was high with a weak recommendation; and the evidence levels of clinical progression rate and prostate cancer-specific mortality were high with a b recommendation. Conclusion Currently, the limited evidence shows extending the length of hormone therapy is beneficial for patients with localized prostate cancer and locally advanced prostate cancer, especially for patients with high Gleason score, but it cannot raise overall survival rate and disease-free survival rate. This conclusion still needs to be further proved by more high-quality and large-scale RCTs.