Prediction model for tube feeding dependency during chemoradiotherapy for at least four weeks in head and neck cancer patients: A tool for prophylactic gastrostomy decision making

Background & aims: Chemoradiation and bioradiation (CRT/BRT) for locally advanced head and neck squamous cell carcinoma (LAHNSCC) often comes with high toxicity rates, interfering with oral intake and leading to temporary tube feeding (TF) dependency. High-quality scientific evidence for indicators of prophylactic gastrostomy insertion is not available. The aim of this retrospective cohort study was to develop a prediction model to identify patients who need prophylactic gastrostomy insertion, defined as the expected use of TF for at least four weeks. Methods: Four-hundred-fifty LAHNSCC patients receiving CRT/BRT with curative intent between 2013 and 2016 were included in the study. Primary outcome was TF-dependency for four weeks or longer. Patient, tumor, and treatment characteristics were extracted from the medical records and their effects on the use of TF were analyzed using univariable and multivariable analysis. The prediction model was internally validated using bootstrapping techniques. Results: Sixty-five percent (294/450 patients) required TF for four weeks or longer. Variables included in the model were: body mass index and adjusted diet at start of CRT/BRT, percentage weight change at baseline, World Health Organization performance status, tumor subsite, TNM-classification, CRT/BRT, mean radiation dose on the contralateral submandibular and parotid gland. The corrected Area Under the Curve after internal validation was 72.3%, indicating good discriminative properties of the prediction model. Conclusions: We developed and internally validated a prediction model that is intended to estimate TF-dependency for at least four weeks in LAHNSCC patients treated with CRT/BRT. This model can be used as a tool to support personalized decision making on prophylactic gastrostomy insertion.