The Value of the Preoperative Hematological Inflammatory Markers for Predicting Lymphovascular Invasion in Gastric
Cong Li1, Yanyan Wang1, Shumei Liang2
1Department of Clinical Laboratory, Jiangsu Cancer Hospital & Nanjing Medical University Affiliated Cancer Hospital & Jiangsu Institute of Cancer Research, Nanjing, People's Republic of China.
Background:
Gastric cancer (GC) is a prevalent malignant tumor of the digestive system, with high morbidity and mortality. Lymphovascular invasion (LVI) is a risk factor for recurrence and metastasis of GC and an independent predictor of poor postoperative prognosis in GC patients. Accumulating evidence suggests that blood-based inflammatory indicators are linked with the pathogenesis of GC. The study delves into hematological inflammatory markers to explore their potential applications in preoperative LVI in GC.
Methods:
We retrospectively collected and analyzed data from inpatients diagnosed GC at our hospital from December 2020 to October 2024, with LVI confirmed by pathological examination after surgery. Preoperative hematological inflammatory indicators were calculated from peripheral blood samples. Lasso logistic regression and multivariate logistic regression analyses were utilized to verify independent risk factors for LVI-positive GC patients and constructed the nomogram model. The correlation between hematological inflammatory indicators and clinical TNM staging was assessed through Spearman's analysis.
Results:
In the study, 624 patients were pathological diagnosed with LVI-positive. The variables screened by Lasso logistic regression including CA199, PLR, SII, PNI, cT staging, and cN staging. Multivariate logistic regression analysis confirmed that CA199 (OR = 1.349, 95% CI: 1.016-1.790), PLR (OR = 1.396, 95% CI: 1.005-1.939), PNI (OR = 0.706, 95% CI: 0.515-0.968), cT staging (OR = 1.737, 95% CI: 1.511-1.996) and cN staging (OR = 2.272, 95% CI: 1.975-2.613) were independent risk factors of LVI-positive. PLR was positively correlated with cT and cN staging, while PNI was negatively correlated with both. The C-statistic of the nomogram model was 0.845, with a sensitivity of 86.86% and a specificity of 69.90%.
Conclusion:
We provided a new perspective on the clinical application of preoperative hematological inflammatory markers in the auxiliary diagnosis of LVI-positive in GC patients.
