BackgroundThe albumin-to-creatinine ratio (ACR) is known to predict prognosis in liposarcoma patients, but its role in other tumors remains unclear. This study aimed to evaluate the prognostic relationship between ACR and common solid tumors.MethodsData from the Investigation on Nutrition Status and Clinical Outcome of Common Cancers (INSCOC) between 2013 and 2022 were used to analyze patients under 65 years old with solid tumors. Patients were divided into a training cohort (n = 12,027) and a validation cohort (n = 7,985) using simple random sampling. Correlation analysis, Kaplan–Meier method, and restricted cubic spline analysis were conducted to explore ACR's relationship with overall survival (OS). Multivariable logistic regression assessed associations between ACR and Patient—Generated Subjective Global Assessment (PG-SGA), Length of Stay (LOS), and Karnofsky Performance Status (KPS).ResultsIn Cox regression, higher ACR levels were associated with better OS in solid tumor patients. Specifically, when using the cutoff value with low ACR as the reference, higher ACR levels were significantly associated with improved OS. For nasopharyngeal carcinoma (HR = 0.49, 95% CI: 0.35–0.67, P < 0.001), gastrointestinal tract tumors (HR = 0.84, 95% CI: 0.74–0.95, P = 0.007), and urogenital neoplasms (HR = 0.55, 95% CI: 0.43–0.71, P < 0.001), higher ACR levels were linked to better OS. When ACR was categorized into tertiles, the results were consistent with those observed using the cutoff value. In gastrointestinal tract tumor patients, higher ACR levels were linked to lower PG-SGA scores and improved KPS scores (P < 0.05). In urogenital neoplasm patients, higher ACR levels were associated with improved KPS scores (P < 0.05).ConclusionElevated ACR levels were significantly associated with improved OS in cancer patients, particularly in nasopharyngeal carcinoma, gastrointestinal tract tumors, and urogenital neoplasms. ACR was also linked to better nutritional and functional status, suggesting its potential as a prognostic biomarker.
Read full abstract