Associations between dosimetry and patient-reported outcomes in oropharyngeal cancer: post hoc analysis of elective neck de-escalation trials.

Abstract

When using IMRT for oropharyngeal squamous cell carcinoma (OPSCC), the extent to which patient-reported outcomes (PRO) improve at progressively lower dose distributions is unclear. In this study, we evaluated the relationship between PROs and dosimetry in OPSCC patients on two elective neck irradiation (ENI) de-escalation trials. Eligible patients in this analysis had a diagnosis of OPSCC and were treated on two ENI de-escalation trials. Contoured OARs included oral cavity (OC), floor of mouth (FoM), superior/middle constrictors (SMC), parotid glands (PG), submandibular glands (SMG), and tubarial glands (TG). OAR metrics were dichotomized at the median and in tertiles. The primary endpoints were 3- and 12-month PROs: the overall MD Anderson Dysphagia Inventory (MDADI) and Dry Mouth (DM) and Sticky Saliva (SS) scores from the EORTC HN35. The analysis includes 86 patients. The median mean doses were contralateral (CL, 15.7 Gy) and ipsilateral PG (26.3 Gy), contralateral SMG (35.3 Gy), OC (29.3 Gy), and SMC (47 Gy). OC, SMC and TG were significant predictors of 12-month MDADI, with a dose-response below the lowest tertile. Three-month DM scores were associated with OC doses, CL PG V15 and CL SMG V30, but only FoM and CL SMG V30 were predictors of 12 month DM. Oral cavity, FOM, and SMG doses were associated with 12 month SS outcomes. In a cohort of patients treated with de-escalated IMRT, the critical impact of OC and SMG doses became evident. More work is needed to understand both dosimetric and non-dosimetric predictors of long-term PROs.