LAUSR

BACKGROUND We aimed to compare the performance of established inflammation and nutrition-based prognostic indices with a relatively novel index 'mid-upper arm circumference (MUAC)' in outpatients with advanced cancer. METHODS This study was a secondary analysis of a prospective cohort study that enrolled 200 outpatients with advanced cancer visiting a medical oncology clinic at a tertiary hospital. All patients were followed until death, and the Glasgow Prognostic Score (GPS), modified GPS (mGPS), Prognostic Nutritional Index (PNI), neutrophil/lymphocyte ratio (NLR), C-reactive protein/albumin ratio (CAR), and MUAC were compared by calculating the area under the receiver operating characteristic curves (AUROCs). RESULTS The mean age of the patients was 64.4 years, 64.0% were male, and the median overall survival was 32.4 weeks [95% confidence interval (CI): 5.6-142.7]. Overall, all indices showed similarly high AUROCs for estimating 12-week (0.68 to 0.75) and 24-week survival (0.67 to 0.74). When confined to the GPS, mGPS, and MUAC, the AUROCs for 12-week survival were 0.75 (95% CI: 0.66-0.82), 0.74 (95% CI: 0.65-0.82), and 0.72 (95% CI: 0.64-0.79), respectively. For 24-week survival, the AUROCs were 0.70 (95% CI: 0.62-0.76), 0.67 (95% CI: 0.60-0.74), and 0.72 (95% CI: 0.64-0.79), respectively. MUAC had the highest specificity for estimating 12-week survival (86.0%), while GPS showed the highest sensitivity for estimating 12-week survival (81.1%). CONCLUSIONS Inflammation and nutrition-based prognostic indices showed similar acceptable accuracies in estimating the 12- and 24-week survival of oncology outpatients. Notably, a simple and non-invasive index MUAC, showed comparable performance with established indices including GPS and mGPS.