Patient-specific factors associated with tumour motion in lung stereotactic body radiation therapy from real-time tumour tracking traces

TL;DR

This study identifies patient-specific factors that influence lung tumor motion during radiation therapy, helping improve treatment precision.

Contribution

The study quantifies how factors like BMI, lung function, and tumor location affect tumor motion in lung SBRT.

Findings

Lower lung lobes showed up to 8.2 mm more motion than upper lobes in the superior-inferior direction.

Better pulmonary function and higher BMI were associated with increased tumor motion.

Prior lung surgery or radiotherapy reduced left-right tumor motion by 0.5 mm.

Abstract

Body mass index increased motion by 0.2 mm per unit.

•Tumour motion analysed with respect to patient factors using multivariate analysis.•Lower lung lobes showed up to 8.2 mm more motion than upper lung lobes.•Each 10 % decrease in forced expiratory volume in 1 s reduced motion by 0.4 mm.•Prior lung surgery or radiotherapy reduced motion by 0.5 mm.•Tumour diameter increased motion by 0.02 mm per unit.

Tumour motion analysed with respect to patient factors using multivariate analysis.

Lower lung lobes showed up to 8.2 mm more motion than upper lung lobes.

Each 10 % decrease in forced expiratory volume in 1 s reduced motion by 0.4 mm.

Prior lung surgery or radiotherapy reduced motion by 0.5 mm.

Tumour diameter increased motion by 0.02 mm per unit.

Respiratory motion is a major source of geometric uncertainty in lung stereotactic body radiation therapy (SBRT), necessitating individualized motion management. The study aimed to examine the lung tumour motion against patient- and tumour-specific factors.

Motion traces were obtained from 109 retrospective CyberKnife fiducial tracking lung treatments, recorded at 25 Hz through correlation of fiducial and external marker signals. Log files were collected and motion was quantified in the superior-inferior (SI), left–right (LR) and anterior-posterior (AP) directions. Each treatment delivery node (∼2700 per patient) was individually analysed. Clinical data included demographics, comorbidities, prior lung treatments, pulmonary function, and tumour location, size, and histology. Statistical analyses used multivariate and univariate approaches.

Tumour location strongly predicted SI motion, with lower lobes showing up to an 8.2 mm greater motion than upper lobes. Previous surgery or radiotherapy moderately reduced LR motion (–0.5 mm), while tumour diameter showed a weak positive association with LR motion (+0.02 mm per mm). Percentage predicted forced expiratory volume in one second (FEV1) showed moderate positive correlations with AP (+0.01 mm per %) and SI (+0.04 mm per %) motion. Body-mass index (BMI) weakly increased SI motion (+0.2 mm per kg/m2).

Tumour location primarily determined SI motion, with additional increases linked to better pulmonary function and higher BMI. LR motion was greater in patients without prior lung treatment and with larger tumours, while greater AP motion occurred with better pulmonary function. Lung motion variation was quantified to support sub-millimetre SBRT precision.

Figures