Dependence of intrafraction prostate motion within the pelvis on fraction duration during whole pelvic intensity modulated arctherapy (IMAT) versus dynamic IMRT.

Abstract

190 Background: To compare the 3-dimensional intra-fraction variations of prostate position within the pelvis with whole-pelvic fixed-field intensity-modulated radiation therapy (IMRT) vs. intensity-modulated arc therapy (IMAT) in high-risk prostate cancer. Methods: Fifteen PCa patients underwent whole pelvic radiotherapy using either dynamic IMRT with a sliding window technique (n= 8) or IMAT (n= 7). All the patients had a kV cone-beam computed tomography (CBCT) before and immediately after each fraction of IMRT or IMAT. Intra-fraction motions of the prostate were determined using a 2-step procedure performed on each pre- and post-treatment imaging: 1) planning CT and CBCT were matched on bony structures after automatic semi-rigid fusion alongside the 3 axis (x, y, z), 2) planning CT and CBCT were matched on the prostate with respect to intra-prostatic markers: xsoft, ysoft, zsoft. The position of the prostate within the pelvis for each pre- and post-treatment study points was defined as xpros= (xbone – xsoft), ypros= (ybone – ysoft) and zpros= (zbone – zsoft). Rectum and bladder were outlined on each CBCT with the aim to assess changes in rectal or vesical repletion during each fraction. Organ distension was assessed by measuring the average rectal cross-sectional area (rCSA; defined as the rectal volume divided by length), and the area of the bladder when evaluated 2.5cm above the prostate base (A-blad) on pre- and post-treatment CBCT. Results: Two hundred and ninety four CBCT were reviewed for this analysis. The average fraction duration was shorter with IMAT than with IMRT (4’49’’, vs. 11’00’’, p< 0.001). During fractions of IMRT the prostate showed statistically significant shifts in the longitudinal (p= 0,049) and lateral (p=0,013) axis while it was not statistically significant during fractions of IMAT. Intra-fraction rCSA increased neither during IMAT nor IMRT whereas A-blad increased only during fractions of IMRT but with no correlation with prostate displacements. Conclusions: The prostate moves within the pelvis during an IMRT course which could lead to a greater daily geographic miss when compared to the IMAT technique.