Purpose: To determine the optimal dose of single fraction conventional palliative radiation therapy for the relief of pain caused by bone metastases.
Material and methods: Ovid version of EMBASE and EMBASE Classic, MEDLINE, the Cochrane Database of Systematic Reviews and the Cochrane Central Register of Controlled Trials were searched for relevant randomised controlled trials. Pain response data were standardised according to the clinical trial endpoints recommended by the International Bone Metastases Consensus Working Party.
Results: From 2696 references we selected 26 articles for review. These described 24 trials that cumulatively randomised 3233 patients to 28 single fraction arms: two arms received 4 Gy, one 5 Gy, one 6 Gy, twenty-two 8 Gy, one 10 Gy and one 8-15 Gy. Eighty-four percent of all patients received 8 Gy and this imbalance precluded formal modelling analyses for different doses. Efficacy endpoints and pain assessment times varied. In general, higher doses produced better pain response rates. The overall (OR) and complete (CR) pain response rates for different doses according to available intention-to-treat (ITT) and assessable patient (A) figures were: 4 Gy [OR(ITT)=23-47%, OR(A)=44-47%, CR(ITT)=15-18%, CR(A)=15-26%], 5 Gy [OR(A)=72%, CR(A)=55%], 6 Gy [OR(ITT&A)=65%, CR(ITT&A)=21%], 8 Gy [OR(ITT)=21-81%, OR(A)=31-93%, CR(ITT)=9-52%, CR(A)=14-57%], 10 Gy [OR(A)=84%, CR(A)=39%]. In trials that directly compared different single fraction doses, 8 Gy was statistically superior to 4 Gy.
Conclusions: 8 Gy was by far the most commonly administered single fraction dose within 24 randomised trials of conventional radiation therapy for the palliation of bone metastases. 8 Gy should be the standard dose against which future treatments are compared due to its reproducible pain response rate and its established safe profile. The optimal dose for the relief of pain remains an open question, however, 8 Gy produced statistically superior pain response rates compared to 4 Gy in trials directly comparing those doses, and in general across all trials doses of 8 Gy or more produced numerically superior pain response rates compared to doses less than 8 Gy.
Copyright © 2012 Elsevier Ireland Ltd. All rights reserved.