A 7-Year Institutional Experience in Dosimetric Quality Improvement for Oropharyngeal Cancer: Progress to Date

Abstract

We have previously reported a low incidence of treatment-related toxic effects after treatment for oropharyngeal carcinoma compared to historical data. Multiple quality improvement steps have been taken at our institution over the past 7 years: the adoption of volumetric modulated arc therapy (VMAT) over step-and-shoot IMRT (SS-IMRT), daily image guidance, improved planning algorithms and the implementation of a specialized dosimetry team. Here we present temporal trends in improvement and factors associated with benefit. 222 patients treated definitively for oropharyngeal carcinoma from December 2009 through August 2016 were identified from an IRB-approved registry. Each patient was planned with a two or three-dose level prescription delivered to planned target volumes (PTVs) of gross tumor, microscopic high-risk, or low-risk nodes. Specialized planners were designated for treatment planning after July 2013, while VMAT was gradually adopted after June 2014. Treatment plan quality was reviewed based on tumor coverage and dosimetric endpoints to organs-at-risk (OARs). The evolutions of the endpoints over 7 years were also evaluated. Wilcoxon signed-rank test was used to compare the differences in non-parametric continuous variables and Spearman’s rho test for correlations. 134 patients were treated with SS-IMRT and 88 with VMAT. The PTVs were well covered by the prescription doses (average conformity index: 1.02±0.02). Significant dose reductions in multiple OARs were observed in the patients treated with VMAT (see Table). No statistically-significant differences in the dosimetric endpoints were found between the patients planned by specialized planners (average p>0.05). Furthermore, negative correlations between all the endpoints and treatment date were identified, indicating a temporal trend towards better OAR sparing. The results were consistent with our previous clinical observation of the reduced occurrences of treatment related toxicity. Our results demonstrate that the adoption of VMAT significantly improved OAR sparing for oropharyngeal cancer patients, which is a contributing factor for the declining incidences of treatment related toxicity. Other factors such as planning algorithms and the adoption of specialized planners likely contributed but to a lesser degree.Abstract 2787; TableMedian (Gy)VMATMedian (Gy)SS-IMRTWilcoxon test p-valueSpearman correlation coefficientSpearman test p-valueBrainstem D0.03cc20.026.4<0.0001-0.38<0.0001Spinal cord D0.03cc33.341.5<0.0001-0.68<0.0001Larynx Dmean26.636.0<0.0001-0.61<0.0001Supraglottis Dmean39.246.4<0.0001-0.27<0.0001Oral cavity Dmean32.836.60.04-0.170.02OARpharynx Dmean45.752.6<0.0001-0.46<0.0001Contralateral submandibular Dmean37.238.20.72-0.070.57Contralateral parotid Dmean24.225.40.09-0.46<0.0001 Open table in a new tab