ESTRO 2026 - Abstract Book PART II

S1890

Physics - Dose prediction/calculation, optimisation and applications for photon and electron planning

ESTRO 2026

and fixed integrated gantry positions to the standard of care across five treatment sites. Phys Med Biol. 2024;69(5):055007. Keywords: RapidArc Dynamic, Dynamic Collimator, Scalp Digital Poster 2634 Dosimetric and Treatment Efficiency Comparison of IRIS versus MLC Collimators in CyberKnife Radiosurgery for Brain AVMs Raghavendra Holla, Bhooshan Zade Radiation Oncology, RubyHall Clinic, Pune, India Purpose/Objective: Precise dose delivery and optimal treatment efficiency are essential in stereotactic radiosurgery (SRS) for brain arteriovenous malformations (AVMs), owing to their complex shape and proximity to critical neural structures. The introduction of the Multileaf Collimator (MLC) in the CyberKnife S7 platform offers an alternative to the traditional IRIS variable aperture collimator, potentially enhancing conformity and reducing treatment time. This study aimed to perform a comparative evaluation of dosimetric outcomes and treatment efficiency between IRIS and MLC-based CyberKnife plans for AVM targets using the VOLO optimizer. Material/Methods: A retrospective dataset of 18 consecutive AVM patients previously treated with the CyberKnife S7 system using MLC collimators was analyzed. For each patient, a corresponding IRIS plan was retrospectively generated using identical planning parameters, beam orientations, and coverage objectives to enable direct comparison. The target volumes ranged from 2.01 to 10.86 cc. Dosimetric metrics including the normal brain volume receiving 12 Gy (V12Gy), minimum, maximum, and mean target doses, and conformity index (nCI) were compared using paired t-tests. Treatment efficiency was assessed through monitor units (MUs), total beam-on time, and overall treatment delivery duration. Statistical significance was set at p < 0.05 Results: The normal brain volume receiving 12 Gy ranged from 3.84–14.60 cc for IRIS and 3.83–13.26 cc for MLC, with significantly lower V12Gy observed in MLC plans across all target volume groups (p < 0.05). The MLC- based plans consistently achieved comparable conformity and target coverage to IRIS without compromising dosimetric quality. Treatment efficiency analysis revealed substantial reductions in both MUs and delivery time for MLC plans (mean MUs: 3152; mean time: 15 minutes) compared to IRIS (mean MUs: 6242; mean time: 31 minutes). The improvement in

exhibited minor dose spills. Based on blinded evaluations, P1 was identified as the most balanced and clinically preferred plan.

Conclusion: RAD demonstrated excellent dosimetric and clinical feasibility for scalp irradiation, achieving substantial brain dose reduction and improved delivery efficiency compared with conventional multi-arc VMAT. Among all plans, P1 (single-arc RAD) achieved the best balance of conformity, skin coverage, and brain sparing. These findings highlight the potential of RAD as an efficient and clinically favorable planning technique for complex irregular targets, with applicability extending beyond scalp cases. References: 1. Dunn L, Kron T, Franich R, et al. Rotational arc treatments with static ports and dynamic collimation reduce cardiopulmonary dose for esophageal cancer patients and diminish model-based predicted benefit of proton therapy. Radiother Oncol. 2024;192:110– 118.2. Feain I, Franich R, Dunn L, et al. Comparison of advanced dynamic arc therapy with collimator rotation

Made with FlippingBook - Share PDF online