Coverage with dosimetric concordance index (CDCI): a tool for evaluating dosimetric impact of inter-observer target variability in brachytherapy

Purpose: The aim of this study was to propose an index for evaluating dosimetric impact of inter-observer target delineation variability in brachytherapy. Material and methods: The coverage with dosimetric concordance index (CDCI) is expressed as CDCIcommon and CDCIpair. The CDCIcommon is the mean coverage of target volume with common volume irradiated by prescription dose among all observers and represents the condition of worst target coverage. CDCIpair is the generalized form of CDCI, which is mean target coverage with common prescription volume obtained between all possible pairs of observers and represents more realistic coverage of target with dosimetric concordance. The index was used to evaluate the dosimetric impact of target delineation variability in optimized conformal plans on target volumes of five radiation oncologists for twenty patients of multi-catheter interstitial partial breast brachytherapy. Results: The mean decline of 5.6 ±3.2% and 11.3 ±5.7% in CDCIpair and CDCIcommon, respectively, was observed comparing to coverage index (CI) of target volume in all patients due to inter-observer target variability. CDCIcommon and CDCIpair were found to have significant linear correlation (r = 0.964, p < 0.000). The difference between CDC and CI increased with the mean relative target volume among observers. Significant correlation (r = 0.962, p < 0.000) was also noted for the difference (Δ) in CDCIcommon and CDCIpair with CI of target volume. Conclusions: The recommended indices and difference between the dosimetric coverage of target volume (CI) with CDCI (ΔCDCI) can be used for evaluating dosimetric impact of the inter-observer target delineation variability.