Accuracy of image-guided radiotherapy of prostate cancer based on the BeamCath urethral catheter technique.

Abstract

To examine the accuracy of the BeamCath urethral catheter technique for prostate localization during radiotherapy. Sixty-four patients were CT scanned twice with the BeamCath catheter, and once without the catheter. The catheter contains radiopaque fiducial markers for prostate visualization on setup images. It is held in place by a balloon inflated with air in the bladder. The repeated CT scans were co-registered and the relative shifts of the BeamCath isocenter fiducial, the prostate center-of-mass, and external skin markers were evaluated. The displacement of the BeamCath isocenter fiducial relative to its position at the planning CT scan was also determined on setup films for 53 consecutive patients (222 setup films). The standard deviation (SD) of the prostate movement relative to the BeamCath isocenter fiducial was 0.9 mm in the left-right (LR), 2.8 mm in the cranial-caudal (CC), and 1.6 mm in the anteroposterior (AP) directions, respectively. When the balloon radius differed more than 1mm between the CT scans (14 of 64 cases) the SD in the CC direction increased to 4.5 mm. The SD of the prostate movement relative to the pelvic bone was 0.6 mm (LR), 2.4 mm (CC), and 2.8 mm (AP), while the prostate movement relative to external skin markers was considerably larger. Removal of the catheter resulted in a mean cranial prostate movement of 1.5mm relative to the pelvic bone. Caudal catheter displacements of 7-30 mm were observed in 5% of the setup films. In these cases, recatherization was necessary to obtain reliable prostate localization. The BeamCath catheter technique markedly improved prostate localization in all directions when compared with skin markers. In the AP direction, the BeamCath technique was also superior to the use of bony structures. However, in the CC direction the catheter position was very vulnerable to changes in the balloon volume.