Results of a multicentre dosimetry audit using a respiratory phantom within the EORTC LungTech trial

Introduction: The EORTC 22113-08113 LungTech trial assesses the safety and efficacy of SBRT for centrally located NSCLC. To insure protocol compliance an extensive RTQA procedure was implemented. Methods: Twelve centres were audited using a CIRS008A phantom. The phantom was scanned using target inserts of 7.5 mm and 12.5 mm radius in static condition. For the 7.5 mm insert a 4DCT was acquired while moving according to a cos6 function. Treatment plans were measured using film and an ionization chamber. Wilcoxon's signed-rank tests were performed to compare the three plans across institutions. A Spearman correlation was calculated to evaluate the influence of factors such as PTV, slice thickness and total number of monitor units on the dosimetric results. Results: The reference output dose median [min, max] variation was 0.5% [−1.1, +1.5]. The median deviations between chamber doses and point-planned doses were 1.8% [−0.1; 6.7] for the 7.5 mm and 1.1% [−2.8; 5.0] for the 12.5 mm sphere in static situation and 3.2% [−3.2; 15.7] for the dynamic situation. Film gamma median pass rates were 92.0% [68.0, 99.0] for 7.5 mm static, 96.2% [73.0, 99.0] for 12.5 mm static and 71.0% [40.0, 99.0] for 7.5 mm dynamic. Wilcoxon's signed-rank tests showed that the dynamic irradiations resulted in significantly lower gamma pass rates compared to the 12.5 mm static plan (p = 0.001). The total number of MUs per plan was correlated to both film and IC results. Conclusion: An end-to-end audit was successfully performed, revealing important variations between institutions especially in dynamic irradiations. This shows the importance of dosimetry audits and the potentials for further technique and methodology improvements.