A Comprehensive Beam-Level Assessment of PBS Migration and Utilization at a Multi-System Proton Therapy Center

Purpose To provide a beam-level comparison of pencil-beam scanning (PBS) proton therapy at a high-volume center operating both the ProteusPLUS (P-PLUS) and ProteusONE (P-ONE) platforms, and to evaluate how system capabilities influenced room utilization during a multi-year transition to PBS. Methods and Materials Log files were analyzed for three rooms: P-PLUS Gantry Room 2 (GTR2; 2018–2024), P-PLUS Gantry Room 1 (GTR1; mid-2023–2024), and P-ONE Compact Gantry Room (CGTR; 2020–2024). A script was written to extract beam parameters while clinical and demographic information were obtained from the MOSAIQ database. Data streams were merged, then grouped and analyzed by site, age, and temporal factors. Results The final dataset included 120,367 beams for the P-PLUS and 95,587 for the P-ONE. During an 18 month comparative period, mean throughput was 27.4±5.4 patients/day (GTR2), 22.9±5.1 (GTR1), and 27.0±4.6 (CGTR), with mean session times of 28.2±2.5, 25.6±1.7, and 26.9±2.4 minutes, respectively. Case allocation reflected platform/nozzle differences: GTR2 primarily treated prostate, breast, and thorax/abdomen targets, while GTR1 and CGTR focused on prostate, head-and-neck, brain, and CNS cases. CGTR treated more pediatric patients (25.2% vs 7.1% in GTR2 and 5.8% in GTR1) and cases involving anesthesia. Maximum cyclotron energy was used more on GTR2 compared to CGTR (7.5% vs 0.4%), while range-shifter use was similar (71.1% vs 67.0%); couch flips occurred in 77% of P-ONE treatments. Conclusions Expanding PBS capacity substantially shifted room utilization and case mix, driven mainly by differences in spot and field size, and secondarily by imaging, layer-switching, and pediatric-oriented room features. The P-ONE’s smaller field size did not materially constrain prostate treatment with involved lymph nodes, frequent couch flips did not inhibit high operational capacity, and the P-ONE’s lower minimum energy reduced range-shifter reliance and the need for maximum cyclotron energy. This real-world analysis provides guidance for case selection, workflow optimization, and future proton system design.