Research on the application value of peripheral blood parameters in the diagnosis of early-onset colorectal cancer
10.3760/cma.j.cn311367-20250106-00002
- VernacularTitle:外周血检测在早发性结直肠癌诊断中的应用价值研究
- Author:
Wenxuan YAN
1
;
Junhai ZHEN
;
Wenhao SU
;
Jixiang ZHANG
;
Fei LIAO
;
Weiguo DONG
Author Information
1. 武汉大学人民医院全科医学科，武汉 430060
- Publication Type:Journal Article
- Keywords:
Early-onset colorectal cancer;
Systemic immune-inflammation index;
Peripheral blood parameters;
Neutrophil-to-lymphocyte ratio;
Lymphocyte-to-monocyte rat
- From:
Chinese Journal of Digestion
2025;45(4):256-265
- CountryChina
- Language:Chinese
-
Abstract:
Objective:To evaluate the value of peripheral blood systemic immune-inflammation index (SII), neutrophil-to-lymphocyte ratio (NLR), lymphocyte-to-monocyte ratio (LMR), carcinoembryonic antigen (CEA), D-dimer, and albumin (ALB) alone or their combination in the diagnosis of early-onset colorectal cancer (EOCRC).Methods:From January 1, 2023 to November 30, 2024, 104 patients with EOCRC (EOCRC group) hospitalized at Renmin Hospital of Wuhan University were enrolled. During the same period, by simple random sampling method, 104 patients with benign colorectal polyps (benign polyp group) and 104 healthy individuals for health examinations (healthy control group) from outpatient department were enrolled. The peripheral blood parameters (including neutrophil count, lymphocyte count, CEA, and others) and pathological characteristics of EOCRC (including TNM stage, tumor differentiation grade, and depth of invasion) were collected. The relationship between peripheral blood parameters and EOCRC pathological features were analyzed. Receiver operating characteristic curves (ROC) were plotted, and the area under the curve (AUC) was calculated to evaluate the diagnostic value. Multivariate logistic regression analysis was performed to analyze the peripheral blood parameters which independently correlated with EOCRC and a combined diagnostic model was established. Simple random sampling method was used to divide the subjects in the negative control group (healthy control group + benign polyp group) and positive group (EOCRC group) into a training set (218 cases) and a validation set (94 cases) at a ratio of 7∶3, and the diagnostic performance of the combined diagnostic model in the training and validation sets was assessed. Hosmer-Lemeshow test and calibration curve were used to evaluate the fit and consistency of the model. Independent sample t-test, one-way ANOVA, Mann-Whitney U test and Kruskal-Wallis H test were used for statistical analysis. Results:EOCRC group had the highest levels of SII(744.03 (473.01, 1 246.28), 437.77 (342.28, 607.47), 497.31 (385.76, 721.63)×10 9/L), NLR(2.42 (1.76, 3.94), 1.96 (1.54, 2.52), 1.91 (1.55, 2.75)), CEA (3.58 (1.96, 20.85), 1.31 (0.95, 1.93), 1.21 (0.76, 2.11) μg/L) and D-dimer (0.36 (0.20, 0.90), 0.19 (0.12, 0.28), 0.18 (0.12, 0.30) mg/L), and the lowest levels of LMR(3.51±1.56, 4.38±1.37, 4.72±1.84) and ALB(42.40 (39.90, 44.70), 44.57 (42.83, 46.25), 44.95 (43.10, 46.58) g/L) than benign polyp group and healthy control group, and the differences were statistically significant ( H=31.18, 16.21, 76.72 and 47.72, F=15.40, H=34.19; all P<0.001). In EOCRC patients, there were statistically significant differences in SII and LMR between patients with different tumor invasion depth ( Z=-2.48, t=2.31; both P<0.05), in CEA between patients with different TNM stage, with or without lymph node metastasis and distant metastasis( Z=-2.68, -2.50 and -2.65; all P<0.05), in D-dimer between patients with different TNM stage, differentiation grade, invasion depth, and with or without lymph node metastasis and distant metastasis ( Z=-2.50, -2.60, -2.06, -2.14 and -3.33; all P<0.05), and in ALB between patients with or without distant metastasis ( Z=-2.52, P=0.012).The AUC of combination of SII, NLR, LMR, CEA, D-dimer, and ALB in differential diagnosis of the healthy control group and the EOCRC group was 0.914 (95% confidence interval (95% CI): 0.870 to 0.958, P<0.001), and the AUC of the combination in differential diagnosis of the benign polyp group and the EOCRC group was 0.904 (95% CI: 0.857 to 0.950, P<0.001). The results of multivariate logistic regression analysis revealed that SII, NLR, LMR, CEA, and ALB were all independently correlated with EOCRC (all P<0.05). The diagnostic model for EOCRC was established by the combination of SII, NLR, LMR, CEA, and ALB, and the AUC of the model in the training set and validation set was 0.911 and 0.883, respectively. The Hosmer-Lemeshow goodness-of-fit test indicated good model fit ( P=0.437). Calibration curve analysis showed strong consistency between predicted probabilities and actual probabilities, and the mean absolute error was 0.015. Conclusions:SII, NLR, LMR, CEA, D-dimer, and ALB all demonstrate diagnostic value in the diagnosis of EOCRC. The combined diagnostic model based on SII, NLR, LMR, CEA, and ALB demonstrates excellent diagnostic performance, which may serve as an adjunctive diagnostic approach for EOCRC.
