Respiratory motion effects and plan robustness for lattice radiation therapy

TL;DR

This study shows how breathing motion affects the dose distribution in lattice radiation therapy, suggesting strategies to manage motion for better treatment outcomes.

Contribution

The study quantifies the impact of respiratory motion on lattice radiation therapy dose ratios and proposes motion management cutoffs.

Findings

Hot spheres lose coverage with increasing motion, while cold spheres show mixed responses.

1.5 cm sphere plans are less sensitive to motion than 1 cm sphere plans.

Tumor deformation may alter dose distributions unpredictably.

Abstract

Lattice radiation therapy (LRT), a form of spatially fractionated radiation therapy (SFRT), has shown promise in treating bulky tumors. It consists of hot and cold spots within the tumor, and the ratio between these doses is thought to be important in clinical outcomes. Respiratory motion is expected to degrade these peak-to-valley dose ratios (PVDRs) but has largely been ignored in LRT literature and clinical practice. This work aims to quantify the response of LRT dose distributions to motion to better inform motion management decisions and improve clinical outcomes.

Respiratory motion of peak-to-peak amplitudes from 0.3 to 2 cm was simulated in 1 and 1.5 cm sphere lattice plans in a uniform phantom using dose perturbations. A similar analysis was repeated with retrospective patient plans to understand the effect of anatomical variation in motion response. Finally, deformed and rigid dose transformations were compared to understand the effect of tumor deformation on motion response.

Hot spheres (VTVH) lost coverage with increasing motion, while cold spheres (VTVL) experienced either an increase (anterior–posterior motion) or a decrease (superior–inferior motion). The ratio of the two (VTVH/VTVL D95%) decreased with increasing motion. Patient data results generally agreed with phantom results; 1.5 cm sphere plans were less sensitive to motion than 1 cm sphere plans, with statistical differences between the two responses. Mean (standard deviation) percent changes from no motion to 1 cm superior–inferior motion for patient plans are as follows with phantom values listed after: VTVH D80% change was −13.8 (1.2), −12.7 for 1.5 cm spheres and −15.8 (1.8), −13.8 for 1 cm spheres; and VTVH/VTVL D95% change was −7.1 (4.0), −15.0 and −13.0 (3.5), −12.0, respectively. Generally, no significant difference was observed between deformed and perturbed dose distributions.

Large tumor motion can degrade LRT dose distributions, which could lead to less effective treatments. Suggested cutoffs of 1 cm motion for 1.5 cm sphere plans and 0.5 cm motion for 1 cm sphere plans are proposed above which motion limiting strategies should be applied. Caution should be applied for tumors with large deformations, as these may change the dose distributions in unexpected ways.

Figures