PMC:6909918 / 10540-11893 JSONTXT

{"project":"2_test","denotations":[{"id":"31790382-19631507-69134700","span":{"begin":1099,"end":1101},"obj":"19631507"}],"text":"Statistical analysis\nData were analyzed with STATA version 13.0 software (StataCorp, College Station, Texas, USA). Three or more interventions were divided into all possible combinations of two intervention tests. Consistency or inconsistency testing was waived because of the absence of a closed-loop in the present network meta-analysis. The intergroup discrepancies for outcomes of PFS and OS were presented with hazard ratios (HRs), and the variance estimates were calculated from the reported confidence intervals (CIs). HRs of PFS and OS were used for indirect comparison with the random-effects model. The ORR, DCR, AEs, SAEs, and FAEs were calculated with the ORs. Indirect comparison of regorafenib, fruquintinib, TAS-102, and placebo, network meta-analysis methods (STATA network) were performed. Based on the surface under the cumulative ranking (SUCRA) curve, the efficacy and safety of the three drug treatments were ranked. Review Manager (RevMan) version 5.3 software (Nordic Cochrane Centre, The Cochrane Collaboration, 2014) was used for the description of the PRISMA flow diagram [10], risk of bias summary, and risk of bias graph, while the other figures in the present study were developed with STATA version 13.0 software (StataCorp, College Station, Texas, USA). Publication bias of the literature was evaluated using funnel plots."}