Diagnostic predictive value of peripheral blood count and inflammatory markers in gastric cancer

Abstract

Abstract ObjectiveTo investigate the correlation between the systemic immuno-inflammatory index (SII), platelet-to-lymphocyte ratio (PLR),monocyte/lymphocyte ratio (MLR) and neutrophil-to-lymphocyte ratio (NLR) and gastric cancer and to evaluate the predictive value of SII, NLR, MLR and PLR for gastric cancer diagnosis. Methods Clinical data (age, sex, BMI, blood indexes, carcinoembryonic antigen, and alpha-fetoprotein) were retrospectively collected from 318 patients with a primary diagnosis of gastric cancer, gastric polyps or chronic gastritis treated in Chaohu Hospital of Anhui Medical University. The patients were divided into training and validation groups at a ratio of 7:3 by applying the Rand random function. After identifying the indicators with significant intergroup differences, the receiver operator characteristic curves (ROC) were plotted, and the optimal cut-off value, area under the curve (AUC), OR and 95% confidence interval (95% CI) of the independent risk factors for the diagnostic prediction of gastric cancer were analyzed to establish a diagnostic prediction model. This model was applied to the validation group to assess the discriminatory degree of the model. Results ROC analysis showed that the optimal thresholds for monocyte count, SII, NLR, MLR, PLR, methemoglobin, carcinoembryonic antigen, leukocytes, and lymphocytes were 0.41×109/L, 554.44, 3.14, 0.24, 151.54, 1.43 µg/L, 5.01 µg/L, 4.34×109/L, and 1.68× 109/L, respectively. Univariate analysis showed that sex, age, BMI, monocyte count, SII, NLR, MLR, PLR, alpha-fetoprotein, carcinoembryonic antigen, leukocyte count, and lymphocyte count were associated with gastric cancer. Multivariate analysis showed that the AUC of the diagnostic prediction model was 0.846 (95% CI 0.809-0.884, P < 0.001) in the training group, and when the diagnostic optimal critical value of the training group was applied to the validation group, the model had an AUC of 0.887 (95% CI 0.835-0.939, P < 0.001). These results suggest that the model has good differentiation. The SII, PLR and carcinoembryonic antigen values were lower in gastric cancer patients after treatment than before treatment. Conclusion SII, NIR, MLR, and PLR are independent risk factors for gastric cancer, and MIR, PLR, alpha-fetoprotein, and carcinoembryonic antigen could be jointly used for the diagnosis and screening of gastric cancer.