In this study, we analyzed the clinical information of patients undergoing radical gastrectomy, including gender, age, family history of tumors, clinical and blood biochemical indicators such as body mass index (BMI), carcinoembryonic antigen (CEA), carbohydrate antigen 199 (CA199), and peripheral blood platelet-lymphocyte ratio (PLR) in the two weeks before surgery. Univariate analysis showed that TNM stage, CEA, CA199, and PLR were associated with postoperative recurrence and metastasis in patients undergoing radical gastrectomy. Multivariate analysis revealed that TNM stage, CEA, and PLR were independent risk factors for postoperative recurrence and metastasis in patients undergoing radical gastrectomy, with TNM stage carrying the highest weight (HR = 3.69, P < 0.001), while the systemic inflammation marker PLR (HR = 1.43, P < 0.010) and tumor marker CEA (HR = 1.44, P < 0.011) had comparable weights. The 1-, 3-, and 5-year ROC curve areas (AUCs) indicated that the model could effectively predict postoperative recurrence and metastasis in patients. Subsequently, we validated the model in a validation group and found that the model could effectively divide patients into high and low-risk groups based on the model score, and there was a significant difference in survival between the two groups. The 1-, 3-, and 5-year AUCs for predicting DFS indicated that the model had good sensitivity and specificity in the validation group as well.
The platelet-to-lymphocyte ratio (PLR) is a representative indicator of systemic inflammation and immune status, calculated by dividing the peripheral blood platelet count by the peripheral blood lymphocyte count. However, the mechanism by which PLR affects the survival outcomes of cancer patients remains unclear. Lymphocytes play a critical role in cancer immune surveillance and the prevention of malignant tumor development[11]. The body's inflammatory state can lead to poor immune response, thereby mediating tumor immune escape[12]. Platelet count is another indicator of systemic inflammatory response and potential microvascular thrombosis. Aggregated platelets can promote tumor growth by releasing pro-angiogenic factors within the tumor microvasculature. Additionally, platelets inhibit tumor cell extravasation by enhancing endothelial cell contraction induced by tumor cells, as well as promoting tumor cell proliferation and metastasis by enhancing cell adhesion and cross-extracellular matrix diffusion. Therefore, lymphocyte reduction and platelet increase are considered adverse prognostic markers for various cancers[13]. Moreover, the easy availability of PLR in clinical practice has greatly enhanced its potential application.
Although PLR is a promising and useful indicator for predicting postoperative DFS in gastric cancer patients, a single PLR alone may not provide sufficient accuracy. Therefore, we combined TNM staging and CEA to construct a DFS risk prediction model. These three factors have the advantage of being easily obtained clinically, and our training and validation groups showed good sensitivity and specificity. We hope this can provide some reference for clinical decision-making by medical professionals.
However, there are several limitations to our study. First, although we designed a large cohort retrospective study with 3,392 patients from two centers, it does not meet the multicenter standard. Second, the retrospective design of this study may have introduced bias. Third, we did not evaluate postoperative radiotherapy and immunotherapy, as well as other factors that may affect patient survival, mainly because the number of patients receiving such treatments was small.