Cone-Beam CT Images as an Indicator of QACT During Adaptive Proton Therapy of Extremity Sarcomas

Periodic quality assurance CTs (QACTs) are routine in proton beam therapy. In this study, we wanted to test if the necessity of a QACT can be determined by evaluating the change in beam path length (BPL) on daily cone-beam (CBCT). In this Institutional Review Board–approved study, we retrospectively analyzed 959 CBCT images from 78 patients with sarcomas treated with proton pencil beam scanning (PBS). Plan on 17 QACTs out of a total of 243 were clinically determined to be replanned for various reasons. Daily CBCTs were retrospectively analyzed by automatic ray-tracing each beam from the isocenter to the skin surface along the central axis. A script was developed for this purpose. Patterns of change in BPL on CBCT images were compared to those from adaptive planning using weekly QACTs. Sixteen of the 17 adaptive replans showed BPL changes ≥4 mm for at least one of the beams on 3 consecutive CBCT sessions. Similarly, 43 of 63 non-adaptively planned patients had BPL changes <4 mm for all of the beams. New QACT criterium of a BPL change of any beam ≥4 mm on 3 consecutive CBCT sessions resulting in a sensitivity of 94.1% and a specificity of 68.3%. Had it been the BPL as the QACT predictor, a total of 37 QACTs would have been collected than 243 QACTs practiced in the clinic. The use of BPL changes on CBCT images represented a significant reduction (85%) in the total QACT burden while maintaining treatment quality and accuracy. QACT can be performed only when it is needed, but not in a periodic manner. The benefits of reducing QACT frequency include reducing imaging dose and optimizing the use of patient time and staff resources.