Assessment of contour variability in target volumes and organs at risk in lung cancer radiotherapy

Aims: This study aimed to examine whether any significant differences existed in trial protocol compliance in target volumes (TV) and organs at risk (OARs) contouring amongst clinical oncologists specialised in lung cancer radiotherapy. Materials/methods: Two lung radiotherapy trials that require all prospective investigators to submit pre-trial outlining quality assurance (QA) benchmark cases were selected. The contours from the benchmark cases were compared against a set of reference contours which were defined by the trial management group (TMG). In order to quantify the degree of variation in TV and OARs contouring, the matching index (MI), Dice coefficient (DICE), Jaccard index (JI), Van‘t Riet Index and geographical miss index (GMI) were calculated. Results: A total of 198 structures contoured by 21 clinicians were collected from the outlining benchmark cases. There were 40 clinical target volumes (CTV), 32 spinal cord, 36 oesophagus, 36 heart and 54 lungs volumes included in the study. Analysis of the pre-trial benchmark cases revealed statistically significant differences (p ≤ 0.05) in trial protocol compliances between clinical oncologists’ target volume and organs at risk contours. Our results demonstrated that the lung contours had the highest level of conformity, followed by heart, CTV, spinal cord and oesophagus respectively. Conclusions: This study showed that there was a statistically significant difference in trial protocol compliance for lung clinical oncologists’ TV and OARs contouring within the pre-trial QA benchmark cases. Trial protocol compliances of TV and OARs delineation can be identified through assessing outlining QA benchmark cases. Keywords: Contouring variability, Lung radiotherapy, Trials quality assurance

Abstract licence: CC BY-NC-ND

A person counts once a day per paper. We keep no cookie and no address: a reader is a hash that changes every day. Any month under five reads shows no number.