Adjuvant Chemotherapy Regimens And Chemosensitivity Genes Screening For Resectable Gastric Cancer

Background and objective Gastric cancer is a serious threat to human life,which is the third most common cancer and the second leading cause of cancer death in China,has led serious medical and health berden to China.Ajduvant chemotherapy has been demonstracted to prolong the overall survival(OS)and disease-free survival(DFS)in patients with resectable gastraic cancer.The present study,firstly,comprehensively collected the clinical practice guidelines published by international and national organizations to assess and survey the guideline recommendations of adjuvant chemotherapy for resectable gastric cancer.Secondly,to resolve the preblem on which adjuvant chemotherapy was the best one,we used Bayesian network meta-analysis approach and GRADE system to assess and screen the available adjuvant chemotherapy regimens.Thirdly,we assessed the top two regimens from Bayesian network meta-analysis using retrospective cohort study design.What's more,we used system bioinformatics technology to screen the sensitivity genes to the adjuvant chemotherapy for resectable gastric cancer in the gene chip databases,and demonstrated the clinical application value of the candidate gene by immunohistochemistry.Finnaly,we provided a comprehensive evidence to support the precision adjuvant chemotherapy in patients with resectable gastric cancer from the two aspects of screening optimal adjuvant chemotherapy regimens to accurately predict the effect of adjuvant chemotherapy.Methods 1)Cross-sectional survey:Searching the main Chinese and English databases,main guideline websites and Google to comprehensively identify clinical practice guidelines for gastric cancer published in English and Chinese,and included guidelines that contained the recommendations of postoperative adjuvant chemotherapy.We assessed the methodological and reporting quality of guidelines with AGREE II tool and RIGHT statement,and investigated the recommendations of postoperative adjuvant chemotherapy in these guidelines and their evidence sources and recommended strength.2)Bayesian network meta-analysis:Randomized controlled trials(RCTs)that assessed adjuvant chemotherapy regimens for resectable gastric cancer were collected by main Chinese and English databases.The risk of bias of included RCTs was assessed using the modified Cochrane risk of bias assessment tool.A Bayesian network meta-analysis and a single-arm meta-analysis were conducted to comprehensively analysis the effectiveness and safety of all currently available adjuvant chemotherapy regimens.Based on the methods developed by the GRADE working group,the certainty of evidence in the network meta-analysis was also evaluated.3)Retrospective cohort study:According to the eligibility criteria,patients who underwent D2 radical resection and treated by XELOX or S-1monotherapy were collected,patients who received any neoadjuvant therapy were excluded.The Kaplan-Meier curve was used to analyze the influence factors of survival outcomes.Log Rank test was used to compare the differences between XELOX and S-1 monotherapy for OS and DFS,and the Cox multivariate regression analysis was also performed.4)Screening adjuvant chemosensitive genes:Systematic bioinformatics technology was used to collect the data from the GEO database that received postoperative adjuvant chemotherapy for resectable gastric cancer,and based on the median of each patient's prognostic index(PI),we divided patients to chemosensitive and chemoresistence groups.The relationship between genes expression and survival in the two groups was analyzed and compared,and the differentially expressed genes between the two groups were screened.Then,TCGA database was used for external validation,and the adjuvant chemosensitivity gene for gastric cancer was finally screened.5)Clinical validation assessment of candidate genes:we collected tissue blocks of patients with gastric cancer who only received postoperative adjuvant chemotherapy.The expression levels of candidate genes proteins in gastric cancer tissues were detected by immunohistochemistry.The chi-square test and Log Rank test were used to analyze the relationship between the expression levels of candidate genes and clinical characteristics and survival.Results 1)A total of 14 guidelines were included,which were from 9 countries and areas,published between 2006 and 2018.Only five guidelines conducted a comprehensive search and reported the database name retrieved and search strategies.Eight guidelines reported the criteria used for evidence rating and recommendation strength,five of which were self-defined.Nine guidelines were based on evidence and expert consensus to inform the recommentations.Based on the AGREE II tool,only one guideline from the Scottish Intercollegiate Mentoring Network was strongly recommended(average score of 71.47%in 6 domains);eight guidelines were generally recommended(40.00%?average score<60.00%).The main problematic domains included the application,rigor and participation.Based on the RIGHT reporting statement,the overall reporting rates of the 14 guidelines ranged from22.86%to 68.57%,and the reporting rates of the 6 guidelines were between 40.00%and 70.00%.A total of 18 recommendations for postoperative adjuvant chemotherapy were included in the 14 guidelines:4 recommendations did not report the evidence sources,7 were based on published single RCTs,and 7 were based on published systematic reviews or meta-analyses.All guidelines recommend postoperative adjuvant chemotherapy for resectable gastric cancer,but the recommended chemotherapy regimens varied in different guidelines,including 5-Fu-based chemotherapy regimens,FMA regimens,S-1 monotherapy,SOX regimens,XELOX regimens,LF regimens,and FLOFOX program,etc.2)A total of 45 RCTs were included for Bayesian network meta-analysis,enrolling 12221 patients with 58.60 of average age,and involving 13 adjuvant chemotherapy regimens.Most of RCTs were from China,Japan,the United States and Italy.Twenty-six RCTs were assigned to be high risk of bias and 19 were low risk of bias.The results of network meta-analysis showed that compared with surgery alone,all postoperative adjuvant chemotherapy regimens significantly improved the OS and DFS of patients with resectable gastric cancer.Among them,high to moderate certainty evidence showed S-1 monotherapy(OS:HR=0.67,95%CrI:0.54-0.83,high certainty;DFS:HR=0.64,95%CrI:0.54-0.77,moderate certainty)and XELOX regimen(OS:HR=0.66,95%CrI:0.51-0.85,high certainty;DFS:HR=0.58,95%CrI:0.47-0.72,high certainty)were more effective.The side effects of S-1 monotherapy were lighter than the XELOX regimen.The most common side effect after S-1 monotherapy was nausea(3.70%),while the most common side effect of XELOX regimen was neutral mitochondrial reduction(35.70%).3)A total of 243 patients with resectable gastric cancer were enrolled,of them 117 in the XELOX group and 126 in the S-1 monotherapy group.The mean follow-up was 28.86 months.There was no significant difference in baseline characteristics between the two groups(P>0.05).Survival analysis showed that the1-,2-,and 3-year survival rates of XELOX were 97.43%,92.31%,and 81.20%,respectively,and the 1-,2-,and 3-year DFS rates were 91.45%,82.91%,and 74.36%,respectively.The 1-,2-,and 3-year survival rates of S-1 monotherapy were 94.40%,87.30%,and 80.95%,respectively.The the 1-,2-,and 3-year DFS rates were 84.92%,75.40%,and 68.25%,respectively.The?~2 test showed there were no significant differences in the 1-,2-,and 3-year survival rates and DFS rates between two groups(P>0.05).Cox regression analysis showed that there were no significant differences between XELOX and S-1 monotherapy in improving OS(HR=0.96,95%CI:0.54-1.72;P=0.894)and DFS(HR=1.26,95%CI:0.78-2.02;P=0.336).4)GSE84437chip data included 431 samples involving 29,362 genes was analyzed.17chemosensitivity genes were screened initially by Cox and LASSO regression analysis.Then we matched 17 chemosensitivity genes with TCGA samples,calculated PI values and performed Cox survival analysis.Finally,three genes were indentified:ITGA2,GAD1 and SLC25A15.We further introduced these three genes into the GSE84437 and TCGA datasets.Only GAD1 gene in GSE84437 was differentially expressed between the chemosensitivity group and the chemo-insensitivity group,and it was related to patient survival(HR=0.64,95%CI:0.48-0.84;P=0.002).In the TCGA dataset,only GAD1 and SLC25A15 genes were associated with survival(HR=0.31,95%CI:0.10-0.92;P=0.035;HR=0.23,95%CI:0.07-0.76;P=0.009;respectively).5)A total of 30 pathological specimens were used to verify the relationship between the expression of SLC25A15 and GAD1 gene proteins and survival.The results showed that the 1-year survival rate of patients with high expression of GAD1 protein was higher than low expression group(100%vs.70.60%;Log Rank test P=0.050).A higher 1-year survival rate was found in high expression of SLC25A15 protein,but the difference of survival between high and low expression groups was not statistically significant(91.70%vs.77.80%;Log Rank test P=0.646).Conclusions The adjuvant chemotherapy regimens for resectable gastric cancer recommended by the clinical practice guidelines are inconsistent,and the quality of the guidelines needs to be improved.Among the different adjuvant chemotherapy regimens,moderate to high certainty evidence shows that S-1 monotherapy and XELOX regimen significantly improve patients'survival time compared with other regimens;however,the effect between S-1 monotherapy and XELOX regimen are similar,and S-1 monotherapy presents lower side effects.Gastric cancer patients with relatively high expression of GAD1 gene may have better survival after adjuvant chemotherapy than thoese with low expression of GAD1,and the predictive ability of SLC25A15 for adjuvant chemotherapy still needs further study.