Higher Risk of Recurrence in Patients Treated for Head and Neck Cancer with Low BMI and Elevated Levels of C-Reactive Protein

Simple Summary Head and neck cancer (HNC) treatment poses several challenges in clinical practice, and treatment side effects can be debilitating due to the close proximity of important anatomical structures. Cancer recurrence post-treatment presents some of the most challenging HNC management issues. This prospective study identifies high-risk groups for recurrence of head and neck cancer, based on commonly accessible clinical parameters. In this study with 272 HNC patients, elevated pre- and post-treatment CRP levels, low BMI and advanced stage at admission indicate higher risk for recurrence of disease. Using these parameters, a risk model is proposed which may be useful for estimating the probability of cancer recurrence and allow the identification of high and low-risk patients. This prospective study identifies high-risk groups for recurrence of head and neck cancer by BMI and circulating inflammatory response markers. Head and neck cancer patients from three Swedish hospitals were included (n = 272). Leukocyte and thrombocyte counts, CRP levels, and BMI were measured pre-treatment and post-treatment. Associations between the four factors and treatment failure (residual tumor, loco-regional failure, general failure/distant metastasis) were assessed using a Cox proportional hazards model adjusted for sex, age at the initial visit, smoking status, cancer stage, and hemoglobin count. CRP level was the only significant single variable, with an average increase in risk of recurrence of 74% (p = 0.018) for every doubling. The predictive power of a combined model using all variables was highest during the initial months after treatment, with AUC under the ROC curve 0.75 at the 0-3 month timepoints. Patients with elevated pre- and post-treatment CRP levels are at higher risk for recurrence of disease. Male patients with low post-treatment BMI, advanced stage, and high CRP at any time post treatment are at high risk for recurrence. The combined model may be useful for