Admission-Based Tumor and Immunonutritional Profiling for Six-Month Mortality after Colorectal Cancer Hospitalization

Abstract

Hospitalized patients with colorectal cancer (CRC) may deteriorate because tumor burden, systemic inflammation, and nutritional decline interact during and after admission. We evaluated whether the albumin-derived neutrophil-to-lymphocyte ratio (Alb-dNLR), calculated from admission laboratory data, was associated with predefined six-month all-cause mortality and compared it with carcinoembryonic antigen (CEA), serum albumin, dNLR, and recorded nutrient-intake change. Medical records of 569 adults with CRC admitted to Dr. Wahidin Sudirohusodo Hospital, Makassar, Indonesia, were screened; 327 were excluded because laboratory components required for biomarker calculation were incomplete and no imputation was performed. The final cohort included 242 patients. Vital status was determined for six months from the index admission. Discrimination was assessed using receiver operating characteristic curves, and independent associations were examined with a prespecified logistic regression model. During follow-up, 59 patients died (24.4%) and 183 remained alive (75.6%). Alb-dNLR showed modest discrimination (AUC, 0.618; 95% CI, 0.533-0.703; p = 0.007). The ROC-derived cut-off of 1.5 corresponds to an Alb-dNLR score of 2 on the 0-2 ordinal scale, yielding low sensitivity (20.3%) but high specificity (93.4%). After adjustment, Alb-dNLR was the only variable independently associated with six-month mortality (OR, 2.405; 95% CI, 1.124-5.149; p = 0.024). These findings support Alb-dNLR as a highly specific adjunct for recognizing a high-risk inflammatory-nutritional phenotype, not as a stand-alone screening test.