Comparison of prostate positioning guided by three-dimensional transperineal ultrasound and cone beam CT

The accuracy of a transperineal three-dimensional ultrasound system (3DUS) was assessed for prostate positioning and compared to fiducial- and bone-based positioning in kV cone beam computed tomography (CBCT) during definitive radiotherapy of prostate cancer. Each of the 7 patients had three fiducial markers implanted into the prostate before treatment. Prostate positioning was simultaneously measured by 3DUS and CBCT before each fraction. In total, 177 pairs of 3DUS and CBCT scans were collected. Bone-match and seed-match were performed for each CBCT. Using seed-match as a reference, the accuracy of 3DUS and bone-match was evaluated. Systematic and random errors as well as optimal setup margins were calculated for 3DUS and bone-match. The discrepancy between 3DUS and seed-match in CBCT (average ± standard deviation) was 0.0 ± 1.7 mm laterally, 0.2 ± 2.0 mm longitudinally, and 0.3 ± 1.7 mm vertically. Using seed-match as a reference, systematic errors for 3DUS were 1.2 mm, 1.1 mm, and 0.9 mm; and random errors were 1.4 mm, 1.8 mm, and 1.6 mm, on lateral, longitudinal, and vertical axes, respectively. By analogy, the difference of bone-match to seed-match was 0.1 ± 1.1 mm laterally, 1.3 ± 3.8 mm longitudinally, and 1.3 ± 4.5 mm vertically. Systematic errors were 0.5 mm, 2.2 mm, and 2.6 mm; and random errors were 1.0 mm, 3.1 mm, and 3.9 mm on lateral, longitudinal, and vertical axes, respectively. The accuracy of 3DUS was significantly higher than that of bone-match on longitudinal and vertical axes, but not on the lateral axis. Image-guided radiotherapy of prostate cancer based on transperineal 3DUS was feasible, with overall small discrepancy to seed-match in CBCT in this retrospective study. Compared to bone-match, transperineal 3DUS achieved higher accuracy on longitudinal and vertical axes. (orig.)