A prediction tool for malnutrition and sarcopenia in patients with gastroenteropancreatic neuroendocrine neoplasms: results from NUTRIGETNE (GETNE-S2109) study

Abstract

INTRODUCTION: Malnutrition and sarcopenia are prevalent among patients with gastroenteropancreatic neuroendocrine neoplasms (GEP-NENs), necessitating simple assessment tools to identify potentially malnourished patients. After describing the nutritional status of those receiving anticancer treatment in Spain, the development of a nutrition screening tool for GEP-NENs was proposed as an exploratory objective of the NUTRIGETNE study. METHODS: Malnutrition was defined by Global Leadership Initiative on Malnutrition (GLIM) criteria. Sarcopenia by European Working Group on Sarcopenia in Older People 2 (EWGSOP2) criteria. Patients were randomized 4:1 into training and validation sets to build a least absolute shrinkage and selection operator (LASSO) model. The model was corrected by sex, age, and body mass index. Predictors selected according to the area under the curve criterion were then used to fit the optimal model. RESULTS: This study included 399 patients. Most had grade 1 or 2 (86.2%) tumors, originating mostly in the small intestine (43.9%) and pancreas (41.6%). Prevalence of malnutrition and sarcopenia was 61.9 and 15%, respectively. The predictive model had an external accuracy of 65.3 and 71.2% for malnutrition and sarcopenia, respectively. The stronger predictive factors for malnutrition were poor tumor differentiation (odds ratio [OR]: 4.4; 95% confidence interval [CI]: 1.2-22), nausea/vomiting (OR: 3.4; 95% CI: 1.4-9.3), having difficulties for a long walk (OR: 3; 95% CI: 1.8-5) and diabetes (OR: 2.2; 95% CI: 1.3-3.9). Sarcopenia was correlated with having troubles in activities of daily living (OR: 19.7; 95% CI: 3.1-390.9), or difficulties for a long walk (OR: 6.1; 95% CI: 2.6-15.5). CONCLUSION: A nomogram based on these models was developed. We generated a diagnostic tool for malnutrition and sarcopenia in patients with GEP-NENs to identify high-risk cases.