Abstract
BACKGROUND: Performing a Geriatric Assessment (GA) is recommended to help guiding treatment decisions in vulnerable older patients. As a full GA is time consuming and expensive, frailty screening methods are worth investigating. Sarcopenia and the neutrophil-to-lymphocyte ratio (NLR) could be easily available biomarkers related to frailty. OBJECTIVES: We investigated the relationships between NLR, sarcopenia, and frailty, and assessed the potential of NLR as a frailty screening method in older patients with head and neck cancer (HNC). In addition, the relationship between the NLR and each GA item was investigated. DESIGN: Retrospective study. PARTICIPANTS AND MEASUREMENTS: 148 older (≥ 60 years) patients with HNC who had undergone pretreatment a GA, routine blood sample, handgrip strength (HGS) and head and neck CT or MRI for skeletal muscle mass measurement (SMM). The GA to determine frailty was assessed as outcome. Sarcopenia was defined as the combination of low SMM and low HGS. RESULTS: The mean age was 70 (6.08 SD) years. A total of 95 (64 %) patients had an elevated NLR and 21 (14 %) had sarcopenia. Based on the GA, 56 (38 %) patients were determined as frail. Patients with an elevated NLR were more likely to be frail and sarcopenic compared to patients with a normal NLR. NLR score showed a significant though weak correlation with frailty (r= - 0.287, p < 0.05). Using GA as reference standard, the sensitivity, specificity, positive predictive value, negative predictive value and accuracy to predict frailty for NLR were 83 %, 47 %, 49 %, 83 % and 61 %, respectively. In multivariate regression analysis, the significant predictors for frailty were comorbidity, skeletal muscle index (SMI), and NLR. For elevated NLR, SMI and frailty were predictors. Elevated NLR was associated with nutritional status (OR 3.56, P = 0.02) and comorbidity (OR 3.81, P = 0.02) as independent GA items. CONCLUSION: Increased NLR is frequently observed in older HNC patients, often in combination with low SMI and frailty. There is a significant correlation between NLR and frailty. However, the accuracy of NLR to predict frailty based on GA is limited.