ESTRO 2024 - Abstract Book

S4834

Physics - Quality assurance and auditing

ESTRO 2024

Freeman Hospital, Northern Centre for Cancer Care, Newcastle upon Tyne, United Kingdom

Purpose/Objective:

To audit compliance with published Lung SABR guidelines and investigate best practice for Breast radiotherapy across England. To demonstrate the ease of quality assurance of radiotherapy treatment plans from different hospitals using cloud-based dicom analysis tools.

Material/Methods:

In 2019, NHS England made big-data tool ProKnow (Elekta, Stockholm, Sweden) available to the 49 radiotherapy (RT) providers to enable sharing and learning from radiotherapy plans. RT providers were invited to submit patient plans into 2 sets of collections for Lung SABR and Breast. Each site-specific set has multiple collections to cover the range of dose prescriptions, laterality, PTV size, age and nodal involvement. ProKnow uses scorecards to extract dose data from plans and display as histograms or box-plots for the collection of patients. The Lung SABR scorecards compare doses to the UK Lung SABR consortium guidelines [1] and the Breast collections benchmark the current range of doses against published expert consensus recommendations, and may inform future UK guidelines.

Results:

NHS England ProKnow is now populated with more than 26000 patients from 47 providers. Some specific dose regimes are used widely nationally with a large potential pool of patients to analyse, and others have a relatively small number of suitable patients in each provider. Therefore, there are both heavily and sparsely populated collections within the 2 collection sets. The most common UK breast protocol is 26Gy/5# and both left and right breast collections were analysed to compare the planned mean heart dose (MHD) across 18 RT providers. The mis-filed rate for both these collections (wrong laterality, wrong dose/fr) was below 1.4%. Uploaded patients from the range of providers show that 99.9% of patient plans have MHD below 2Gy. Median MHD For 1161 left-sided breast was 0.65Gy (range 0.07 to 2.15Gy). Figure 1 shows the distribution of MHD in the left and right breast 26Gy/5# collections, broken down by RT provider. The colour bands represent 0.5Gy increments in MHD from 0Gy (bright green) to 2Gy (pink). This shows that there are small differences in planning practice or in patient population across the country, but quality assurance of all plans exhibits low doses to the heart. Analysis of other metrics is also available to the providers directly. This audit has highlighted that the lack of clear ontology for the breast PTV structure (49% of plans are missing a structure with this exact name) makes multi-provider analysis of target coverage challenging. Figure 2 shows Lung SABR data and is the aggregate scorecard data across a range of metrics for PTV 20-40cc 55Gy/5# with colour coding to visualise the UK guidelines’ mandatory and optimal values. Auditing data across all dose regimens and PTV sizes shows more than 80% of patient across all collections meet the recommended PTV D99% constraint, and the dose to 0.5cc of the Heart and dose to 0.1cc of the Spinal Canal meets the mandatory dose limits for all patients. A small number of Lung SABR patients have been entered into the wrong collection (PTV size incorrectly sorted) but this is less than 3% of patients. Feedback has been provided to individual RT providers to improve patient sorting in the future.