A predictive scoring model to select suitable patients for surgery on primary tumor in metastatic esophageal cancer

Abstract

Abstract Background Surgery on primary tumor (SPT) has been a common treatment strategy for many types of cancer. Aims This study aimed to investigate whether SPT could be considered a treatment option for metastatic esophageal cancer and to identify the patient population that would benefit the most from SPT. Methods Data from 18 registration sites in the Surveillance, Epidemiology, and End Results Program database (SEER database) were analyzed to select patients with metastatic esophageal cancer. Multivariate Cox regression analysis was used to identify potential risk factors for pre‐treatment survival. Variables with a p‐value of less than 0.05 were used to construct a pre‐treatment nomogram. A pre‐surgery predictive model was then developed using the pre‐surgery factors to score patients, called the “pre‐surgery score”. The optimal cut‐off value for the “pre‐surgery score” was determined using X‐tile analysis, and patients were divided into high‐risk and low‐risk subsets. It was hypothesized that patients with a low “pre‐surgery score” risk would benefit the most from SPT. Results A total of 3793 patients were included in the analysis. SPT was found to be an independent risk factor for the survival of metastatic esophageal cancer patients. Subgroup analyses showed that patients with liver or lung metastases derived more benefit from SPT compared to those with bone or brain metastases. A pre‐treatment predictive model was constructed to estimate the survival rates at one, two, and three years, which showed good accuracy (C‐index: 0.705 for the training set and 0.701 for the validation set). Patients with a “pre‐surgery score” below 4.9 were considered to have a low mortality risk and benefitted from SPT (SPT vs. non‐surgery: median overall survival (OS): 24 months vs. 4 months, HR = 0.386, 95% CI: 0.303–0.491, p < 0.001). Conclusion This study demonstrated that SPT could improve the OS of patients with metastatic esophageal cancer. The pre‐treatment scoring model developed in this study might be useful in identifying suitable candidates for SPT. The strengths of this study include the large patient sample size and rigorous statistical analyses. However, limitations should be noted due to the retrospective study design, and prospective studies are needed to validate the findings in the future.

Keywords

• advanced esophagus cancer
• nomogram
• predictive scoring model
• surgery on primary tumor
• X‐tile