Validation of the total dysphagia risk score (TDRS) in head and neck cancer patients in a conventional and a partially accelerated radiotherapy scheme.

Abstract

A risk model, the total dysphagia risk score (TDRS), was developed to predict which patients are most at risk to develop grade ⩾2 dysphagia at 6 months following radiotherapy (RT) for head and neck cancer. The purpose of this study was to validate this model at 6 months and to investigate the power at earlier and later time-points. A second aim was to see if this model can be used in a partially accelerated RT regimen. 164 patients from 3 different centres treated with RT between 2008 and 2014 were included in the current study. Both physician-scored dysphagia and QoL data were prospectively obtained. The TDRS of all patients was correlated with the physician-scored dysphagia and the QoL data. To validate this prediction model, we tested the validity in terms of calibration and discrimination. Partial acceleration had no influence on the TDRS. Regarding physician-scored dysphagia, there was a significant correlation with dysphagia grade ⩾2 at 1, 3, 6 and 9 months. The area-under-the-curve at 1 month was 0.85; at 3 months 0.80; at 6 months 0.85; at 9 months 0.86 and 0.79 at 12 months. Regarding QoL, TDRS correlates with PEG-tube usage at 6 and 12 months. We found significant correlations between TDRS and dysphagia grade ⩾2 and PEG-tube usage.