Development and validation of a nomogram model for predicting overall survival in patients with gastric carcinoma

The prevalence and mortality rates of gastric carcinoma are disproportionately elevated in China, with the disease's intricate and varied characteristics further amplifying its health impact. Precise forecasting of overall survival (OS) is of paramount importance for the clinical management of individuals afflicted with this malignancy.

To develop and validate a nomogram model that provides precise gastric cancer prevention and treatment guidance and more accurate survival outcome prediction for patients with gastric carcinoma.

Data analysis was conducted on samples collected from hospitalized gastric cancer patients between 2018 and 2020. Least absolute shrinkage and selection operator, univariate, and multivariate Cox regression analyses were employed to identify independent prognostic factors. A nomogram model was developed to predict gastric cancer patient outcomes. The model's predictability and discriminative ability were evaluated via receiver operating characteristic curves. To evaluate the clinical utility of the model, Kaplan-Meier and decision curve analyses were performed.

A total of ten independent prognostic factors were identified, including body mass index, tumor-node-metastasis (TNM) stage, radiation, chemotherapy, surgery, albumin, globulin, neutrophil count, lactate dehydrogenase, and platelet-to-lymphocyte ratio. The area under the curve (AUC) values for the 1-, 3-, and 5-year survival prediction in the training set were 0.843, 0.850, and 0.821, respectively. The AUC values were 0.864, 0.820, and 0.786 for the 1-, 3-, and 5-year survival prediction in the validation set, respectively. The model exhibited strong discriminative ability, with both the time AUC and time C-index exceeding 0.75. Compared with TNM staging, the model demonstrated superior clinical utility. Ultimately, a nomogram was developed via a web-based interface.

This study established and validated a novel nomogram model for predicting the OS of gastric cancer patients, which demonstrated strong predictive ability. Based on these findings, this model can aid clinicians in implementing personalized interventions for patients with gastric cancer.