ESTRO 2025 - Abstract Book

S3593

Physics - Quality assurance and auditing

ESTRO 2025

919

Poster Discussion Comprehensive analysis of radiotherapy linac quality control (QC) processes via a UK survey with multivariate analysis Elisha L Tassano-Smith 1,2 , Antony L Palmer 1,2 , Wojciech Polak 1 , Andrew Nisbet 2 1 Medical Physics, Portsmouth Hospitals University NHS Trust, Portsmouth, United Kingdom. 2 Medical Physics and Biomedical Engineering, University College London, London, United Kingdom Purpose/Objective: Comprehensive analysis of current linac QC practice from a UK survey, conducted on behalf of the Institute of Physics in Engineering and Medicine (IPEM), was performed to establish consensus practice and understand differences in approach between centres. Material/Methods: A survey comprising 330 questions explored linac QC practice at UK radiotherapy centres. Multivariate analysis was performed, assessing correlation between tests, measurement frequency, tolerances, number of linacs, staff per linac, and time spent performing QC. A ‘frequency score’ (FS), indicating propensity to perform QC regularly, was calculated for each centre by assigning tests a score correlated to their measurement frequency (annual=1, monthly=12) and averaging over all tests performed. Scores were compared to previous survey results [1] and IPEM Report 81 [2]. Results: 70% of UK centres provided comprehensive information. Data tables of consensus and range of QC practice were determined. High consensus was found in the frequency of standard output measurements with 60% of centres performing this monthly, equal to previous survey results. Change in frequency was found in tests that have evolved methodology: 20% increase in monthly MLC picket fence QC as measurements moved from film to EPID; large variability in image quality QC methods, frequency, and tolerances, a potential result of limited IPEM 81 guidance. Fig.1 presents an example of the advanced data analysis undertaken. FS, normalised to FS calculated from IPEM Report 81 (nFS), plotted as a function of the number of linacs, with multivariate analysis of staff per linac and time spent performing QC indicated within the data point format (point size and colour). The interquartile range is also indicated. Data correlation (Fig.1) suggests as the number of linacs per centre decreases below 5 the lower limit of nFS increases. The relative staff numbers and time spent performing QC for these centres is variable suggesting these factors alone do not cause this trend. Centre A likely demonstrates correlation between greater time spent performing QC and greatest nFS. Centre B has a lower nFS but greater time spent performing QC compared to Centre C with the same number of linacs and staff, indicating potential scope for increased efficiency.