Conibear, John Richard; (2018) Assessment of Target Volume and Organ at Risk Contouring Variability within the Context of UK Head and Neck and Lung Cancer Radiotherapy Clinical Trials. Doctoral thesis (M.D(Res)), UCL (University College London).

Preview

Text Assessment of Target Volume and Organ at Risk Contouring Variability within the Context of UK Head and Neck and Lung Cancer Radiotherapy Clinical Trials.pdf - Accepted Version Download (2MB) | Preview

Abstract

AIMS: Radiotherapy quality assurance (RTQA) is now a requirement of radiotherapy trials since poor target volume and organ at risk (OAR) contouring has been shown to impact on patient outcomes within the context of clinical trials. The first hypothesis for this research is that statistically significant inter-observer variation exists amongst clinical oncologists’ target volume and OAR contours within the context of the pre-trial quality assurance (QA) benchmark cases for four different UK radiotherapy trials. The second hypothesis is directed towards confirming that RTQA feedback during the pre-trial benchmark period does influence contouring for head and neck cancers. MATERIALS/METHODS: Four radiotherapy trials (ART-DECO, COSTAR, IDEAL and i-START trials) that require all prospective investigators to submit pre-accrual benchmark cases were selected. All benchmark cases until November 2012 were collected in DICOM format. The investigator contours were grouped into either target contours (TARGET) i.e. CTV1, parallel OARs (OAR-P) i.e. parotid glands, lungs and heart or serial OARs (OAR-S) i.e. brainstem, spinal cord and oesophagus. These were then analysed using a tumour management group (TMG) consensus contour to determine whether statistically significant differences existed between them. The local conformity index (L-CI) for each structure was also calculated for analysis. RESULTS: Analysis of the pre-trial benchmark cases revealed statistically significant differences (p=<0.05) between clinical oncologists’ target volume, serial and parallel OAR contours. Analysis of the resubmitted head and neck pre-trial benchmark cases also revealed statistically significant differences between first and subsequent submission contours. CONCLUSIONS: This research revealed that a statistically significant difference does exist in clinical oncologists’ target volume and OAR contours within the pre-trial QA benchmark cases for both lung and head and neck cancers. It was also revealed that RTQA feedback during the pre-trial benchmark period had a positive and statistically significant impact on head and neck clinician contouring.

Download activity - last month

Export as CSVXMLJSON

Download activity - last 12 months

Export as XMLCSVJSON

Downloads by country - last 12 months

Export as JSONXMLCSV

Archive Staff Only

View Item