Treatment outcomes of surgery and chemotherapy for pulmonary metastases from non-lung cancers: a propensity score-matched analysis

BACKGROUND: The optimal treatment for pulmonary metastases has not been determined, and the survival benefit of surgical resection in selected patients remains controversial. The purpose of this retrospective study was to explore whether surgery can prolong survival in patients with pulmonary metastases compared with chemotherapy, and to analyze the factors that may affect the long-term survival of patients with pulmonary metastases. METHODS: We retrospectively analyzed the medical records of patients with pulmonary metastases from June 2012 to June 2019. Propensity score matching (PSM) was used to balance factors that might affect survival between the two groups. The data were analyzed by Kaplan-Meier survival analysis and Cox proportional hazards models to compare the survival of the surgery group and the chemotherapy group. RESULTS: A total of 202 patients with pulmonary metastases were enrolled in the study, with 43 patients in the surgery group and 43 in the chemotherapy group after screening and PSM. After PSM, patients in the surgery group had better survival than those in the chemotherapy group, with 5-year overall survival (OS) rates of 75.1% and 48.0%, respectively (P=0.017). Univariate analysis of the two groups of patients found that the treatment method, the number of metastases, and the total diameter of metastases were prognostic factors, but multivariate analysis did not find independent prognostic factors. In the surgical group, univariate analysis found that disease-free interval (DFI), the number of metastases, surgical methods, resection scope and surgical route were prognostic factors, and multivariate analysis showed that only DFI was an independent prognostic factor. In the chemotherapy group, DFI and the response of metastases to chemotherapy were found to be prognostic factors in univariate analysis, but no independent prognostic factors were found in multivariate analysis. CONCLUSIONS: Surgery does not provide a significant survival advantage. For patients undergoing surgery, longer DFI predicts better survival.