Factors associated with non-completion of radiation therapy (RT) or radiochemotherapy (RTCT) in head and neck squamous cell carcinoma (HNSCC): A prospective study.

Abstract

e17568 Background: Although numerous factors have been reported to increase RT/RTCT toxicity, identification of patients with high risk of toxicities is lacking. Methods: The monocentric NutriNeck study (NCT02900963) recruited all HNSCC patients between 2014 and 2018 treated by RT or RTCT. Patients fed by enteral nutrition before RT initiation were excluded. All clinical and biological, planned RT or RTCT parameters were collected at baseline. A daily monitoring of the weight and a weekly evaluation of clinical, biological and food intake during RT/RTCT were performed. A nutritional intervention by enteral feeding was systematically proposed in case of 2 kg weight loss under treatment. Complete treatment was defined as full planned RT doses without more than 3 days interruption, and full planned doses of cisplatin and cetuximab if indicated. Primary endpoint assessed factors associated with non-completion of the planned treatment. Secondary endpoint assessed factors associated with use of enteral nutrition during treatment. A p value < 0.01 was considered significant. Results: Among the 249 HNSCC patients included, 118 (47%) were treated by RT and 131 (53%) by RTCT. 63 patients (25%) did not complete treatment: 3/118 (2.5%) in the RT group compared to 60/131 (45.8%) the RTCT group, p < 0.0001. Median weight lost during treatment (7 vs 3.8 kg, p < 0.0001) and grade 3 toxicities (53 vs 33%, p = 0.006) were more frequent in patients with incomplete RT/RTCT. Nutritional intervention was performed in 111 patients: 28 (23.7%) with RT and 83 (62.6%) with RTCT (p < 0.0001). Week 4 was associated with the higher rate of enteral nutrition initiation in both RT and RTCT groups. Among RTCT group, weight loss during treatment (OR 1.2 for each kg, p < 0.0001) and use of enteral feeding (OR 2.7, p = 0.005) were associated with incomplete treatment in multivariate analysis. Of note, 48 patients refused enteral nutrition while indicated, they had comparable grade 3 toxicities and weight loss during treatment compared to the 111 patients who accepted enteral nutrition. Conclusions: Incomplete treatment is overwhelmingly associated with RTCT rather than RT. Baseline clinical, biological features or treatment volume/dose are not associated with incomplete treatment. Enteral nutrition initiated during treatment is poorly efficient.