ESTRO 2024 - Abstract Book

S1273

Clinical - Head & neck

ESTRO 2024

in HNC patients treated in an US-centre: a cohort with very distinct demographics and toxicity assessment compared to the development cohort.

Material/Methods:

Patients with oropharyngeal cancer treated with definitive radiotherapy with or without chemotherapy at the US centre were consecutively included between 2015 and 2020, resulting in 435 patients that adhered to the inclusion criteria of Van den Bosch et al. (2021). NTCP models predicting dysphagia, aspiration, xerostomia, sticky saliva, taste, and fatigue at 6 months after radiotherapy were evaluated. All endpoints were patient-rated (Dutch centres: EORTC QLQ-C30 & QLQ-HN35, US-centre: MDASI-HN), except for dysphagia, which was reported by physicians (Dutch centres: CTCAE, US-centre: PSS-HN diet normalcy). Moderate-to-severe patient-rated toxicities in the Dutch centre were translated to a MDASI- HN score of ≥4 based on clinical experience since intervention is needed for patients with these scores. Grade 2 dysphagia was translated to a PSS- HN diet normalcy score of ≤60. Models were evaluated with direct external validation (i.e. the CITOR model coefficients were used) and with a refit in the US cohort (i.e. same variables in the model, but a re-estimation of the model coefficients). The organs-at-risk included as predictors in the NTCP models are depicted per domain in Figure 1. Model performance was assessed by discrimination (Area Under the Curve (AUC)) and calibration (intercept and slope).