[Comparison of the efficacy of different surgical strategies in the treatment of patients with initially resectable gastric cancer liver metastases]
L Li,Y H Gao,L Zang,K Xue,B Ke,L Shang,Z Q Tang,J Yu,Y R Liang,Z R He,H L Zheng,H Huang,J P Xiong,Z Y He,J Y Li,T T Lu,Q Y Song,S H Liu,Y W Chen,Y Tang,H Liang,Z Qiao,L Chen
DOI: https://doi.org/10.3760/cma.j.cn112139-20240126-00053
2024-05-01
Abstract:Objective: To examine the impact of varied surgical treatment strategies on the prognosis of patients with initial resectable gastric cancer liver metastases (IR-GCLM). Methods: This is a retrospective cohort study. Employing a retrospective cohort design, the study selected clinicopathological data from the national multi-center retrospective cohort study database, focusing on 282 patients with IR-GCLM who underwent surgical intervention between January 2010 and December 2019. There were 231 males and 51 males, aging (M(IQR)) 61 (14) years (range: 27 to 80 years). These patients were stratified into radical and palliative treatment groups based on treatment decisions. Survival curves were generated using the Kaplan-Meier method and distinctions in survival rates were assessed using the Log-rank test. The Cox risk regression model evaluated HR for various factors, controlling for confounders through multivariate analysis to comprehensively evaluate the influence of surgery on the prognosis of IR-GCLM patients. A restricted cubic spline Cox proportional hazard model assessed and delineated intricate associations between measured variables and prognosis. At the same time, the X-tile served as an auxiliary tool to identify critical thresholds in the survival analysis for IR-GCLM patients. Subgroup analysis was then conducted to identify potential beneficiary populations in different surgical treatments. Results: (1) The radical group comprised 118 patients, all undergoing R0 resection or local physical therapy of primary and metastatic lesions. The palliative group comprised 164 patients, with 52 cases undergoing palliative resections for gastric primary tumors and liver metastases, 56 cases undergoing radical resections for gastric primary tumors only, 45 cases undergoing palliative resections for gastric primary tumors, and 11 cases receiving palliative treatments for liver metastases. A statistically significant distinction was observed between the groups regarding the site and the number of liver metastases (both P<0.05). (2) The median overall survival (OS) of the 282 patients was 22.7 months (95%CI: 17.8 to 27.6 months), with 1-year and 3-year OS rates were 65.4% and 35.6%, respectively. The 1-year OS rates for patients in the radical surgical group and palliative surgical group were 68.3% and 63.1%, while the corresponding 3-year OS rates were 42.2% and 29.9%, respectively. A comparison of OS between the two groups showed no statistically significant difference (P=0.254). Further analysis indicated that patients undergoing palliative gastric cancer resection alone had a significantly worse prognosis compared to other surgical options (HR=1.98, 95%CI: 1.21 to 3.24, P=0.006). (3) The size of the primary gastric tumor significantly influenced the patients' prognosis (HR=2.01, 95%CI: 1.45 to 2.79, P<0.01), with HR showing a progressively increasing trend as tumor size increased. (4) Subgroup analysis indicates that radical treatment may be more effective compared to palliative treatment in the following specific cases: well/moderately differentiated tumors (HR=2.84, 95%CI 1.49 to 5.41, P=0.001), and patients with liver metastases located in the left lobe of the liver (HR=2.06, 95%CI 1.19 to 3.57, P=0.010). Conclusions: In patients with IR-GCLM, radical surgery did not produce a significant improvement in the overall prognosis compared to palliative surgery. However, within specific patient subgroups (well/moderately differentiated tumors, and patients with liver metastases located in the left lobe of the liver), radical treatment can significantly improve prognosis compared to palliative approaches.