TU‐C‐17A‐06: Evaluating IMRT Plan Deliverability Via PTV Shape and MLC Motion

Abstract

Purpose:For step‐and‐shoot intensity‐modulated radiation therapy (IMRT) plans, the dosimetry and deliverability can be affected by the number and shape of the segments used. Thus, plan deliverability is likely related to target volume and shape. We investigated whether the sphericity of target volumes and the previously proposed Modulation Complexity Score (MCS) could be used together to improve the detection of IMRT fields that failed quality assurance (QA).Methods:526 and 353 IMRT fields from 32 prostate and 28 head‐and‐neck (H' N) patients, respectively, were analyzed. MCS was used to quantify the complexity of multi‐leaf collimator shapes and motion patterns for each field. Sphericity was calculated using the surface area and volume of each patient’s planning target volume (PTV). Logistic regression models with MCS‐alone or MCS and sphericity terms were fit to PlanUNC IMRT pass/fail results (5% dose difference, 4mm distance‐to‐agreement criteria) using SAS 9.3 (Cary, NC). Model concordance, discordance and area under the curve (AUC) were used to quantify model accuracy.Results:Mean (±1 standard deviation) MCS for prostate and H' N were 0.58(±0.15) and 0.40 (±0.14), respectively. Mean sphericity scores were 0.75(±0.05) for prostate and 0.63 (±0.12) for H' N. Both metrics were significantly different between treatment locations (p<0.01, Wilcoxon Rank Sum Test) indicating greater complexity in shape and variations for H' N PTVs. For prostate, concordance, discordance and AUC using MCS alone were 80.8%, 18.7% and 0.811. Including sphericity in the model improved these to 81.7%, 17.7% and 0.820. For H' N, the original concordance, discordance and AUC were of 72.9%, 26.9% and 0.729. Including sphericity into the model improved these metrics to 76.5%, 23.2% and 0.729.Conclusion:Sphericity provides a quantitative measure of PTV shape. While improvement in IMRT QA failure detection was modest for both prostate and H' N plans, including sphericity in the model does improve IMRT field failure prediction.