Tumor-driven SRS VMAT planning: Regression models for intermediate and low dose spillage

Area of Science:

• Radiation Oncology
• Medical Physics
• Neurosurgery

Background:

• Stereotactic radiosurgery (SRS) using volumetric modulated arc therapy (VMAT) is increasingly used for brain metastases.
• Existing guidelines for high-dose conformity in SRS are insufficient for intermediate and low-dose regions.
• Standardization of intracranial SRS planning requires better understanding of dose distribution.

Purpose of the Study:

• To develop regression models for standardizing intracranial SRS planning.
• To explore tumor-specific characteristics and novel dosimetric parameters for predicting dose distribution.
• To provide patient-specific guidance for optimizing SRS plan quality.

Main Methods:

• Retrospective analysis of 290 VMAT SRS plans from 151 patients.
• Introduction of new dosimetric quantities: R6Gy (6 Gy cloud volume ratio) and %D1cm (max dose at 1 cm).
• Evaluation of R50%, V12Gy, number of metastases (n), and total planning target volume (PTVTotal) using correlation and regression analyses.

Main Results:

• Strong correlations between PTVTotal and dosimetric metrics in single-fraction SRS; weaker correlations in three-fraction SRS.
• Power-law regression models best described R50% and R6Gy.
• Linear regression models best described %D1cm and V12Gy; moderate correlations found between n and dosimetric metrics.

Conclusions:

• Proposed regression-based models predict radiation dose spill based on tumor burden and volume.
• These models offer a framework for model-based SRS planning.
• The findings aim to improve consistency, optimize plan quality, and support standardization in SRS treatment planning.