Fluence-corrected pencil beam algorithm for carbon ion therapy patient quality assurance in solid phantoms

Solid phantoms streamline carbon ion therapy quality assurance (QA), but introduce dose deviations as treatment planning system's (TPS) pencil beam algorithms (PBA) calculate dose in water. This study develops a fluence-corrected PBA, enabling accurate patient-specific QA directly in solid phantoms at Heavy Ion Medical Machine (HIMM) centers. Beam modeling was done in TOPAS Monte Carlo simluation platform to match the integrated depth doses (IDD) and phase space parameters to the clinical commissioning data. Depth-dose data in PTW RW3 slab phantom and in water were simulated, from which energy- and depth-dependent fluence correction (FC) factors (FCFs) were derived. The FCFs were validated experimentally and incorporated into the PBA. 4 types of QA workflows were compared for 2 patient plans at varying equivalent depths: 1) phantom CT without FC-PBA 2) phantom CT with FC 3) virtual CT with stopping power ratio overwritten to 1 with FC, and 4) virtual CT overwritten with phantom SPR with FC. A simplified depth-dependent correction table was fitted from measurement. The simulated FCFs increased with energy and depth from about 0.4% at 2 centimeters to 5% at 25 centimeters, which agreed with the measurement within ±0.5%. At depths < 100 mm, the gamma pass rates were similar for all 4 workflows. For depths > 100 mm, the gamma pass rates increased from (81, 77, 69)% for workflow 1 to (99, 99, 97)% on average for workflows 2-4, under 3 mm/3%, 2 mm/3% and 2 mm/2% criteria, respectively. Workflow 2 and 4 gave similar results with mean difference 0.1(±0.6)%. The simplified table gave 0.0231% FC per millimeter of RW3 thickness, which agreed to the simulation within 1%. Carbon ion therapy QA in the RW3 requires FCFs particularly for > 100 mm depths. Under ≥ 2% pass rate criteria, either the fluence-corrected PBA or the simplified table could be adopted on a virtual phantom to support easier QA workflow.