Do a priori expectations of plan quality offset planning variability in head and neck IMRT?. (March 2020)

• Record Type: Journal Article
• Authors: Alves, N.; Ventura, T.; Mateus, J.; Capela, M.; Lopes, M.C.
• Is Part Of: Radiation physics and chemistry. Volume 168(2020:Mar.)
• Publication Date: 2020-03
• Subjects: Radiotherapy -- Head & neck -- IMRT -- Dosimetry -- PlanIQ; Radiation chemistry -- Periodicals; Radiometry -- Periodicals; Radiation -- Periodicals; Chimie sous rayonnement -- Périodiques; 539.2
• DOI: 10.1016/j.radphyschem.2019.108580
• Languages: English
• ISSNs: 0969-806X

Abstract

Abstract: Introduction: Radiotherapy planning is still a rather biased and planner dependent process, presenting large output variability. This work aims to evaluate whether the feasibility estimation tool, from PlanIQ software version 2.2 by Sun Nuclear Corporation, contributes to a significant improvement in intensity modulated radiation therapy plans' quality for head and neck patients in Tomotherapy, as well as whether it leads to greater homogeneity across different planners. Methods: 28 head and neck clinical cases distributed by three different planners (A, B, C) were planned: 1) without the use of PlanIQ's feasibility tool and 2) using as guidelines to the organs-at-risk (OAR) constraints the feasibility dose-volume histogram (DVH) with feasibility level f =0.1. In both scenarios the resulting DVHs were compared to the closest feasibility DVH curve, and the mean or maximum dose to the OARs as well as the target coverage were assessed. Finally, a validation of the results by SPIDERplan was carried out. Results: For planner A, 3 out of 17 OARs showed statistically significant lowering of the mean f value with PlanIQ. As for planner B, only 1 OAR achieved this result, and most of the OAR presented a higher mean f value in scenario 2. Planner C had the greatest improvement with 12/17 OARs showing significant differences. It was verified that the dose limits clinically demanded for the OAR were generally met in both scenarios by all planners, and that the improvements in the sparing of OAR in scenario 2 for planners A and C were not at the expense of the coverage of the target volumes. These results were corroborated by the SPIDERplan scores, which also showed that the standard deviation of the global plan scores achieved by the different planners was lower with the use of the tool. Conclusion: For planners A and B, the usage of PlanIQ did not show a great improvement in the sparing of organs-at-risk whereas for planner C it made a major difference. In any case, PlanIQ contributed to a lower plan quality output variability among the different planners. Highlights: PlanIQ's feasibility tool contributes to greater sparing in OAR for 2 of the 3 planners. The lower the sparing prior to its use the greater the impact of the tool. The most benefited OAR were the spinal cord and brainstem. The tool contributed to lower output variability across the different planners.