Background There is limited information available on the optimal Planning Target Volume (PTV) expansions and image guidance for post-prostatectomy intensity modulated radiotherapy (PP-IMRT). As the prostate bed does not move in a uniform manner, there is a rationale for anisotropic PTV margins with matching to soft tissue. The aim of this study is to find the combination of PTV expansion and image guidance policy for PP-IMRT that provides the best balance of target coverage whilst minimising dose to the organs at risk. Methods The Cone Beam CT (CBCT) images (n = 377) of 40 patients who received PP-IMRT with daily online alignment to bony anatomy (BA) were reviewed. Six different PTV expansions were assessed: 3 published PTV expansions (0.5 cm uniform, 1 cm uniform, and 1 + 0.5 cm posterior) and 3 further anisotropic PTV expansions (Northern Sydney Cancer Centre (NSCC), van Herk, and smaller anisotropic). Each was assessed for size, bladder and rectum coverage and geographic miss. Each CBCT was rematched using a superior soft tissue (SST) and averaged soft tissue (AST) match. Potential geographic miss was assessed using all PTV expansions except the van Herk margin. Results The 0.5 cm uniform expansion yielded the smallest PTV (median volume = 222.3 cc) and the 1 cm uniform expansion yielded the largest (361.7 cc). The Van Herk expansion includes the largest amount of bladder (28.0 %) and rectum (36.0 %) and the 0.5 cm uniform expansion the smallest (17.1 % bladder; 10.2 % rectum). The van Herk PTV expansion had the least geographic miss with BA matching (4.2 %) and the 0.5 cm uniform margin (28.4 %) the greatest. BA matching resulted in the highest geographic miss rate for all PTVs, followed by SST matching and AST matching. Changing from BA to an AST match decreases potential geographic miss by half to two thirds, depending on the PTV expansion, to
