Albumin and neutrophil combined prognostic grade for predicting overall survival in colorectal cancer: a retrospective cohort study.

Background: Despite significant advances in the management of colorectal cancer (CRC), accurate prognostic stratification remains a clinical challenge. Although several inflammation- and nutrition-based prognostic scores have been evaluated in CRC, their clinical utility is often limited by inconsistent cutoff values and overlapping predictive information. While the Glasgow Prognostic Score (GPS) has demonstrated robust prognostic performance in CRC patients, its dependence on C-reactive protein (CRP) measurement may limit its practicality in clinical settings. To address this limitation, this study aimed to investigate the prognostic value of a novel albumin and neutrophil combined prognostic grade (ANPG) system and to develop a clinically applicable nomogram for predicting overall survival (OS) in CRC patients following curative resection. Methods: A retrospective analysis was conducted on 660 consecutive patients with primary CRC who underwent R0 resection between December 2017 and December 2018. The ANPG was constructed based on preoperative serum albumin levels and neutrophil counts, with optimal cutoff values determined by receiver operating characteristic (ROC) curve analysis. Prognostic factors were identified using univariate and multivariate Cox proportional hazards regression models. A predictive nomogram was developed and internally validated via bootstrap resampling (800 iterations) and time-dependent ROC analysis. Decision curve analysis (DCA) was performed to assess clinical utility. Results: The median follow-up duration was 2442 days with 108 cancer-specific deaths recorded. The ANPG demonstrated superior discriminative ability (AUC = 0.637, 95% CI: 0.588-0.687, P < 0.001) compared to established inflammatory and nutritional markers, including the neutrophil-to-lymphocyte ratio (NLR), platelet-to-lymphocyte ratio (PLR), systemic immune-inflammation index (SII), fibrinogen-to-albumin ratio (FAR), and fibrinogen-neutrophil-lymphocyte ratio (F-NLR). Kaplan-Meier survival analysis revealed significant OS differences across ANPG grades (Log-rank χ² = 24.423, P < 0.001), with 5-year OS rates of 93.7%, 83.2%, and 74.4% for grades 0, 1, and 2, respectively. Multivariate Cox regression analysis identified ANPG (grade 1 vs. 0: hazard ratio [HR] = 2.190, P = 0.020; grade 2 vs. 0: HR = 3.256, P = 0.001), age (HR = 1.032, P = 0.001), carbohydrate antigen 19-9 (CA19-9) (HR = 1.002, P = 0.003), histological type (HR = 1.954, P = 0.005), and TNM stage as independent prognostic factors. The nomogram incorporating these variables(retaining carcinoembryonic antigen for clinical relevance and model performance) achieved a concordance index of 0.806 (95% CI: 0.788-0.824) with excellent calibration, significantly outperforming TNM staging alone in predictive accuracy. DCA showed greater net benefit than TNM for 5-year OS. Conclusions: The ANPG score, derived from routinely available laboratory parameters, provides a practical and accessible tool for prognostic stratification in CRC patients. Additionally, the developed nomogram provides a clinically valuable tool for individualized survival prediction and risk stratification in this patient population.