Purpose: To determine the parameter values of the Lyman-Kutcher-Burman (LKB) normal tissue complication probability (NTCP) model for the pharyngeal constrictors (PC) and superior pharyngeal constrictors (SPC) regarding the endpoint of dysphagia, 12 months after radiotherapy for oropharynx cancer.
Methods: 204 patients with HPV-associated oropharyngeal squamous cell carcinoma were treated with 60Gy IMRT. Dysphagia was assessed based on the PRO-QoL, CTCAE and EAT10 scoring systems. In the first two scoring systems, patients with a change in symptom severity (from baseline) of ≥2 were considered responders, whereas for EAT10 the threshold was 8 in a scale of 40. The clinical data was fitted by the Lyman-Kutcher-Burman (LKB) NTCP model.
Results: Dysphagia was observed in 13.2%, 4.0% and 5.3% of the patients based on the PRO-QoL, CTCAE and EAT10 scoring systems, respectively. The values of the D50, m and n parameters of the LKB model for PRO-QoL were 65.1Gy, 0.21 and 0.1 for PC and 81.6Gy, 0.30 and 0.29 for SPC. Statistically significant Odds Ratios (ORs) of 4.6 were found for PC for a gEUD threshold of 52Gy and 2.6 for SPC for a gEUD threshold of 55Gy. For CTCAE, no statistically significant ORs were found. The values of the D50, m and n parameters for EAT10 were 70.3Gy, 0.10 and 0.06 for PC and 70.5Gy, 0.13 and 1.0 for SPC. The AUC values were 0.72. Statistically significant ORs of 7.9 with gEUD threshold of 58 were found for PC and 44 with gEUD threshold of 61 for SPC.
Conclusion: The LKB NTCP model could fit the clinical data and determine the dose-response curves of the pharyngeal constrictors and superior pharyngeal constrictors regarding dysphagia for the PRO-QoL and EAT10 scores. gEUD<50Gy to the PC or SPC was found to significantly reduce by more than 5 times the risk for dysphagia.
NTCP, Dose Response, Radiation Effects
TH- Response Assessment: Modeling: other than machine learning