The Prognosis Effects Of Primary Tumor Resection For Patients With Metastatic Gastrointestinal Stromal Tumors:A Preliminary Population-Based Analysis

Background:Gastrointestinal stromal tumors（GIST）are the most frequent mesenchymal neoplasms of the gastrointestinal（GI）tract,which can occur anywhere along the GI tract,approximately accounting for 0.1%-3.0%of gastrointestinal tumors.At present,GIST is mainly treated with surgery and molecular targeted drugs.Surgery is a standard radical treatment for patients with resectable GIST.However,it is still controversial on the efficacy of surgery in patients with metastatic GIST.Moreover,it is still unclear whether a specific population will benefit from surgical treatment.Objective:The purpose of this study is to assess whether and for whom primary tumor resection has a survival benefit on metastatic GIST patients undergoing chemotherapy,and to further analyze the risk factors affecting their prognosis,so as to provide evidence for clinical treatment of advanced GIST.Methods:We selected patients with metastatic GIST who met the inclusion criteria between 2010 and 2016 from the Surveillance,Epidemiology,and End Results（SEER）database,and retrospectively collected and analyzed clinical data.According to research purpose,we divided these patients into the resection group and non-resection group to compare the overall survival（OS）and cancer-specifific survival（CSS）of the two groups.The Kaplan-Meier method and Log-rank test were used for survival analysis.The LASSO regression model and COX regression model were performed to evaluate independent risk factors affecting prognosis.In addition,we performed stratification analysis with age,tumor site,and tumor size to reveal the effects of these variables on surgery.Results:1.In total,455 patients with metastatic GIST were considered qualified,of whom 235 underwent primary tumor resection and 220 did not.The Kaplan-Meier survival curve showed that patients in the resection group were associated with improved prognosis than those without surgery.The median survival time of the resection group and the nonresection group were 72 months and 40 months,respectively.The5-year OS for the resection group（62.2%）is superior to the nonresection group（34.3%）.The 5-year CSS for the resection group（63.5%）is superior to the nonresection group（36.8%）,among groups P<0.001.2.The COX model included age,gender,marital status,tumor location,tumor size, lymph node status,and surgery for OS,indicating that age（HR=1.591,P=0.008）is an independent risk factor for poor survival.Primary tumor size between 5 and 10 cm（HR=0.349,P=0.034）and surgical treatment（HR=0.581,P=0.009）are protective factors for the prognosis of patients.3.For CSS,the COX model included age,gender,marital status,insurance status,tumor location,tumor size,lymph node status,and surgery,which also shows that tumor size between 5 and 10 cm（HR=0.309,P=0.022）and surgery（HR=0.663,P=0.042）were predictors of decreased hazard of death,whereas increasing age（HR=1.686,P=0.006）was predictive of poorer survival.4.We conducted subgroup analysis based on age,tumor location and tumor size.For OS,regardless of age<65 years or≥65 years,the 5-year OS for the resection group is superior to the nonresection group（age<65 OS:67.9%vs 40.5%;age≥65OS:49.3%vs 26.3%）;regardless of whether the tumor is located in the stomach or the small intestine,the 5-year OS for the resection group is superior to the nonresection group（stomach OS:53.4%vs 34.1%;small intestine OS:65.7%vs47.1%）;regardless of the tumor size is 5-10cm or>10cm,the 5-year OS for the resection group is superior to the nonresection group（tumor size between 5 and 10cm OS:68.6%vs 41.8%;tumor size>10cm OS:59.4%vs 44.7%）.For CSS,the5-year CSS for the resection group is superior to the nonresection group when the age was less than 65 years or more than 65 years or the tumor was located in the stomach（P<0.05）.In addition,the log-rank test revealed that there were no survival differences between resection and non-resection groups in the following subtypes:tumor originated from colorectum,tumor originated from other sites,tumor size<2cm,and tumor size between 2 and 5 cm（P>0.05）.5.Our interaction tests showed that patients who presented with tumor sizes between 5 and 10 cm had a stronger association of surgery with the reduction of overall death or cancer-specific death(p_int=0.001 for OS;p_int=0.002 for CSS).6.The results of binary logistic regression analysis showed that patients older than65 years with tumors located in the small intestine were much less likely to perform surgery than other patients,especially compared to younger counterparts with tumors in the stomach.Conclusions:Compared with chemotherapy alone,surgical resection combined with chemotherapy could improve the prognosis of patients.In addition,our study preliminarily demonstrates that resection of primary tumors might prolong survival period in cases of well-selected patients with metastatic GIST,especially those with a primary tumor between 5 and 10 cm and a tumor located in the stomach.We recommend that comprehensive treatment options including surgical resection should be actively considered in the case of fully individualized evaluation.It should to be emphasized that this study is a retrospective analysis which still needs to be validated by prospective trials with large samples.