Decision-making between DCA and VMAT in Linac-based SRS for brain lesions: A dosimetric analysis based on tumor size and geometry and introduction of Qasym

Abstract

Introduction

Linac-based SRS provides a highly precise noninvasive treatment option for intracranial lesions. DCA and VMAT are commonly used Linac-based techniques. There are no standardized guidelines for technique selection, particularly considering the geometric properties of the lesions. We thus sought to compare DCA and VMAT to define a workflow for clinical decision-making. This study introduces Q_asym, a novel parameter quantifying lesion asymmetry, and compares the dosimetric performance of DCA and VMAT.

Materials and methods

Between 2018 and 2021, a total of 89 brain metastases from 24 patients with lesion volumes up to 4.3 cm³ were included. VMAT and DCA plans were created for each lesion with identical field parameters, resulting in a total of 178 evaluable plans. The parameters included the CI, DSI, and MUs. Various approaches for quantifying asymmetry were explored and their impact on CI and DSI was investigated. Additionally, we introduced the novel Q_asym index, designed to quantify lesion asymmetry.

Results

VMAT resulted in lower CI values, especially for volumes exceeding 2 cm³. With a prescription dose of 20 Gy, DSI_90% values were comparable between VMAT and DCA for < 2 cm³, while VMAT achieved lower values for larger volumes. VMAT showed higher DSI_50% and DSI_25% values for < 2 cm³, which reversed for larger volumes. V_2Gy values were similar for both techniques for < 2 cm³ but were lower for VMAT in larger volumes. Q_asym significantly correlated with CI (p < 0.01). Analysis of ΔCI% and ΔDSI% revealed that VMAT outperformed DCA for lesions > 2 cm³ and smaller volumes with a Q_asym > 1.2. DCA required significantly fewer MUs (p < 0.01).

Conclusion

This study provides detailed dosimetric information on two commonly applied planning techniques for Linac-based SRS treatment for brain metastases. These findings might support decision making for optimal technique selection upon volumetric features. Q_asym provides a practical tool for these purposes.