Analysis of risk and prognostic factors for pulmonary metastasis in gastric cancer: a study based on the Surveillance, Epidemiology, and End Results database

Abstract

Background: Pulmonary metastasis in patients with gastric cancer (GC) is closely associated with adverse clinical outcomes and reduced survival rates. This study aimed to investigate the incidence, risk factors, and prognostic factors of pulmonary metastasis in GC patients. Methods: A retrospective cohort study was conducted using data from the Surveillance, Epidemiology, and End Results (SEER) database (2010–2021), involving 48,474 GC patients, of whom 2,694 (5.56%) had pulmonary metastasis. Descriptive statistics, multivariable logistic regression, and Cox regression analyses were performed using R software, complemented by Kaplan-Meier survival curves and receiver operating characteristic curve construction. Results: Logistic regression revealed that the risk of pulmonary metastasis was significantly higher in patients with squamous cell carcinoma than adenocarcinoma [adjusted odds ratio (aOR) 1.575, 95% confidence interval (CI): 1.152–2.120], while other pathological types showed a lower risk (aOR 0.269, 95% CI: 0.214–0.333). Stage T4 patients had a significantly higher risk than T1 (aOR 1.487, 95% CI: 1.130–1.954). Surgical intervention (aOR 0.198, 95% CI: 0.145–0.265) and clearance of four or more lymph nodes (aOR 0.489, 95% CI: 0.330–0.725) were associated with reduced pulmonary metastasis risks. Conversely, patients with liver, brain, and bone metastases exhibited significantly increased risks of pulmonary metastasis (aOR 3.888, 95% CI: 3.568–4.238; aOR 4.434, 95% CI: 3.480–5.631; and aOR 2.883, 95% CI: 2.568–3.234, respectively). Multivariate Cox regression analysis of overall survival (OS) and cancer-specific survival (CSS) demonstrated that patients with other epithelial tumors had significantly higher mortality risks [hazard ratio (HR) 1.194, 95% CI: 1.019–1.399; HR 1.191, 95% CI: 1.006–1.409]. Conversely, surgical treatment significantly reduced mortality risks (HR 0.632, 95% CI: 0.473–0.843; HR 0.659, 95% CI: 0.486–0.894), as did chemotherapy (HR 0.322, 95% CI: 0.295–0.351; HR 0.336, 95% CI: 0.307–0.369). Single patients (never married) exhibited higher mortality risks (HR 1.142, 95% CI: 1.020–1.278; HR 1.159, 95% CI: 1.030–1.305), as did patients with liver metastasis (HR 1.240, 95% CI: 1.144–1.344; HR 1.275, 95% CI: 1.171–1.388). Patients with primary lesions located in the lower stomach showed increased mortality risk (HR 1.289, 95% CI: 1.110–1.496; HR 1.203, 95% CI: 1.026–1.410), and those with bone metastases also increased OS mortality risk (HR 1.183, 95% CI: 1.071–1.307). The median OS for patients with pulmonary metastasis was 2 months, compared to 14 months for those without (P<0.001). Conclusions: Surgical treatment and chemotherapy significantly prolonged OS and CSS. Pulmonary metastasis in GC is associated with extremely poor survival rates. Comprehensive screening for highrisk patients, combined with detailed clinical and pathological evaluations, is essential to improve survival outcomes.