RapidArc Dynamic (RAD) multi-mechanical axis optimization achieves enhanced OAR sparing in cervical cancer: A...

To overcome limitations of conventional VMAT for cervical cancer, specifically restricted OAR sparing due to static collimators and delivery constraints inherent to dynamic modulation, we systematically evaluate RapidArc Dynamic (RAD) technology. This first comprehensive assessment focuses on RAD's multi-axis coordination for optimized pelvic radiotherapy planning.

Twenty cervical cancer patients treated with TrueBeam whose energy was 6MV X-ray were retrospectively analyzed. The prescription was 45 Gy/25Fractions. Seven RAD-based planning strategies with varying collimator rotation modes (Optimize [OPT] vs. Optimize Between Static Angles [OBSA] vs. Static [STAT]) and aperture sizes (15 vs. 30 cm) and different static angle ports were developed using identical optimization objectives. Re-optimized clinical plan using Eclipse v18.1. All RAD and clinical plans were normalized so that the target received a certain dose with 95% target coverage for final dosimetric comparison. Dosimetric comparisons employed ANOVA (with LSD post hoc) for parametric data or Kruskal-Wallis's test (with Mann-Whitney U post hoc) for non-parametric data, with significance set at p < 0.05.

RAD plans showed slightly inferior conformity and homogeneity compared to clinical plans, they achieved superior dose fall-off gradients. RAD's advantage was particularly evident in sparing organs at risk (OARs). Specifically, for the bladder, significant reductions were noted in $D_{mean}$ (8.29%-13.05%), $V_{20\mathrm{Gy}}$ (4.04%-26.71%), and $V_{30\mathrm{Gy}}$ (26.98%-32.46%) compared to the clinical plans. Similarly, for the rectum, reductions in $D_{mean}$ (9.21%-15.58%), $V_{20\mathrm{Gy}}$ (5.03%-14.77%), $V_{30\mathrm{Gy}}$ (18.24%-28.11%), and $V_{40\mathrm{Gy}}$ (23.73%-31.22%) were observed. Other OARs also benefited from improved dosimetric parameters. While both the 2ARC+OBSA+15 and 2ARC+STAT+15 plans outperformed other RAD configurations, the 2ARC+OBSA+15 plan generally provided superior results. Notably, although RAD plans overall showed significantly higher small intestine D_2cc, the 2ARC+OBSA+15 configuration achieved comparable doses to clinical plans. In terms of monitor units (MU) and plan complexity, RAD plans were more complex but maintained a high verification pass rate of over 97%.

The preliminary exploration of RAD plans reveals superior capability in sparing the doses of OARs. Moreover, in the RAD plans, the optimization of collimator angles and aperture at 15 cm yields better dosimetric outcomes. Therefore, further exploration of different mechanical axis combinations is of great significance in future studies.