Malnutrition is a recognized risk factor for reduced survival in patients with head and neck cancer (HNC). However, evidence remains limited regarding whether, and to what extent, pre-treatment nutritional parameters contribute to local recurrence and metastasis. So, the aim was to estimate the association between pre-treatment nutritional parameters and these outcomes such as mortality, local recurrence, and metastasis over a five-year follow-up in patients with HNC, and to assess the prognostic validity of the nutritional parameters.

A prospective cohort study was conducted with 84 HNC patients treated at a reference outpatient clinic, followed for up to five years after treatment initiation. Nutritional status was assessed using the Patient-Generated Subjective Global Assessment (PG-SGA), Global Leadership Initiative on Malnutrition (GLIM), and the ESPEN criteria, as well as the Weight Loss Grading System (WLGS), and the presence of cachexia. Kaplan-Meier curves and Cox regression analyses were performed to evaluate the associations between pre-treatment nutritional parameters and clinical outcomes. The sensitivity and specificity of these parameters were also estimated to determine their predictive potential.

High rates of mortality (76.2%), local recurrence (31.1%), and metastasis (38.1%) were observed. At the beginning of the study, most patients were at nutritional risk (34.5% PG-SGA score ≥ 10 to < 16, and 36.9% PG-SGA score ≥ 16), 27.4 to 67.9% were malnourished depending on the diagnostic criteria, and 47.6% presented with cachexia. All nutritional parameters, except WLGS and ESPEN criteria, were associated with higher mortality and shorter survival. The PG-SGA score was the only parameter associated with local recurrence, while cachexia, nutritional risk, and malnutrition according to PG-SGA were associated with metastasis.

Pre-treatment nutritional risk, malnutrition (by GLIM and PG-SGA), and cachexia were associated with mortality, which was also associated with metastasis. Notably, nutritional risk (PG-SGA score > 16) was the only parameter associated with all three outcomes. From a clinical perspective, we suggest that combining PG-SGA and cachexia assessment may enhance prediction of mortality and metastasis in HNC patients, respectively.