Survival analysis of gastric malignancy patients: identifying key prognostic factors

Abstract

Background: Gastric malignancies are common worldwide, with a high incidence rate and mortality. Relevant studies are needed to further demonstrate the harmfulness of gastric malignancies to promote awareness of its prevention. The aim of the study was to analyze the clinicopathological features and prognosis of gastric malignancies.

Methods: The Surveillance, Epidemiology, and End Results (SEER) database was used to obtain the clinical data of 122,793 patients diagnosed with gastric malignancies from 2000 to 2019. Along with performing univariate and multivariate analyses, the associated survival rates of each variable were analyzed using SPSS software. Columnar line graph prediction models were developed and validated using R software.

Results: In total, 122,793 gastric malignancy patients were included in this study, including 74,303 males (60.5%) and 48,490 females (39.5%). The predominant age group among patients was 60-74 years, comprising a total of 45,603 individuals (37.1%). The follow-up time was 0-239 months, the median follow-up time was 11.7 months, and 91,869 patients (74.8%) died. Gastric adenocarcinoma was the main pathological type, accounting for 96,259 patients (82.7%). The main disease site was the cardia of the stomach, accounting for 34,019 patients (34.0%); most (109,706; 89.3%) patients lived in cities. Univariate and multivariate analyses showed that gender, age, tumor size, tumor location, American Joint Committee on Cancer (AJCC) stage, pathological type, rural/urban, and treatment were independent risk factors for prognosis (P<0.001). The Concordance index (C-index) of the nomogram prognostic model was 0.763±0.002, and the areas under the receiver operating characteristic (ROC) curve (AUC) of the 1-, 3-, and 5-year survival rates were 0.76, 0.79, and 0.79, respectively. The calibration curve showed that the predicted survival rate of the nomogram was in good agreement with the observed survival rate.

Conclusions: The prognosis for tumors located in the greater curvature of the stomach is relatively favorable. The level of care available in a patient's city is directly correlated with their prognosis. Notably, the outcomes for gastric stromal tumors (GSTs) and gastric neuroendocrine neoplasms (G-NENs) are significantly more favorable compared to other pathological types. Patients who meet surgical criteria should undergo surgery as early as possible to enhance survival duration.

Keywords: Gastric malignancies; gastric neuroendocrine neoplasms (G-NENs); greater curvature; prognosis; survival.

Figure 1

Kaplan-Meier univariate analysis. Rural/urban: 1, counties in metropolitan areas with a population of 1 million or more; 2, counties in metropolitan areas with a population of 250,000 to 1 million; 3, counties in metropolitan areas with a population of less than 250,000; 4, nonmetropolitan counties adjacent to a metropolitan area; 5, nonmetropolitan counties not adjacent to a metropolitan area. AJCC, American Joint Committee on Cancer; OS, overall survival.

Figure 2

Multivariate Cox regression analysis of gastric malignancies. Rural/urban: 1, counties in metropolitan areas with a population of 1 million or more; 2, counties in metropolitan areas with a population of 250,000 to 1 million; 3, counties in metropolitan areas with a population of less than 250,000; 4, nonmetropolitan counties adjacent to a metropolitan area; 5, nonmetropolitan counties not adjacent to a metropolitan area. AJCC, American Joint Committee on Cancer; CI, confidence interval; HR, hazard ratio; S + C + R, surgery + chemotherapy + radiotherapy.

Figure 3

Nomogram for predicting 1-, 3-, and 5-year OS gastric malignancies. Gender: 1, male; 2, female. Age (years): 1, 0–44; 2, 45–59; 3, 60–74; 4, ≥75. Tumor size (cm): 1, 0–5; 2, >5. Primary site: 1, gastric greater curvature; 2, gastric lesser curvature; 3, fundus; 4, gastric body; 5, gastric antrum; 6, pylorus; 7, cardia; 8, anterior/posterior wall. AJCC stage: 1, IA; 2, IB; 3, IIA; 4, IIB; 5, IIIA; 6, IIIB; 7, IIIC; 8, IV. Pathology: 1, mesenchymal tumor; 2, neuroendocrine tumors; 3, other; 4, adenosquamous carcinoma; 5, adenocarcinoma; 6, squamous carcinoma. Rural/urban: 1, counties in metropolitan areas with a population of 1 million or more; 2, counties in metropolitan areas with a population of 250,000 to 1 million; 3, counties in metropolitan areas with a population of less than 250,000; 4, nonmetropolitan counties adjacent to a metropolitan area; 5, nonmetropolitan counties not adjacent to a metropolitan area. Treatment: 1, surgery; 2, surgery + chemotherapy; 3, surgery + chemotherapy + radiotherapy; 4, surgery + radiotherapy; 5, chemotherapy + radiotherapy; 6, chemotherapy; 7, radiotherapy. AJCC, American Joint Committee on Cancer; OS, overall survival.

Figure 4

The closer the ROC curve is to the top left corner, the better the performance of the model. The closer the calibration curve is to the ideal line (45 degree line), the better the calibration of the model. (A-C) ROC curves of the column line graph model predicting the OS of gastric malignancies at 1, 3, and 5 years. (D-F) Calibration curves of the column line graph model predicting the OS of gastric malignancies at 1, 3, and 5 years. AUC, area under the curve; ROC, receiver operating characteristic; OS, overall survival.