Accuracy of pelvic radiotherapy: prospective analysis of 90 patients in a randomised trial of blocked versus standard radiotherapy

摘要: The aim of this study was to assess the accuracy pelvic radiotherapy during a trial blocked at Royal Marsden Hospital, UK. Prospective evaluation performed on 90 patients receiving CT planned using weekly anterior-posterior and lateral portal films. Field placement errors (FPEs) were calculated by comparing field centres each film with designated point interest. Data evaluated calculate overall treatment simulator differences, number error free treatments, mean treatment-simulator position evaluate role systematic versus random errors. Age, weight, disease site, treatment, fractionation, conventional techniques assessed for their effect accuracy. absolute between films anterior right-left (ARL) 0.25 cm, superior-inferior (ASI) 0.32 (LAP) 0.42 (LSI) 0.28 cm. On average centre displaced 0.66 cm (standard deviation, S.D. = 0.34) from that intended. day 29% 45% had least one 0.5 error. Overall 59% treatments 9% 1.0 more than intended in 66% over 1 14% treatments. Analysis variance showed both occurred all directions. Though similar magnitude direction (variance sigma 2 0.06-0.09 cm2); 4-fold variation being greatest LAP (sigma 0.19 cm2) ARL 0.048 cm2). No factor consistently predicted worse outcome Hypofractionated less accurate LSI (P > 0.05). Systematic associated hypofractionation < 0.01) and, weight 0.03) age We conclude significant can occur especially direction. These results suggest absence customised immobilisation device, cover 95% errors, margins 0.6 RL SI directions 0.9 AP should be allowed planning clinical target volumes. However, ideally, determine own margin requirements according local practice.