| Objective: In patients with advanced non-small cell lung cancer(NSCLC),the use of immune checkpoint inhibitors(ICIs)can bring significant survival benefits if the disease progresses after the first-line chemotherapy,but the relative efficacy of various regimens is unknown.Based on this situation,network meta-analysis was conducted to compare the efficacy and safety of second-line ICIs after the first-line platinum-based double-drug chemotherapy in advanced NSCLC patients without sensitive gene mutation,so as to provide evidence-based basis for clinical decision-making.Methods: Through searching Chinese databases(Wanfang database and CNKI),English language databases(Pub Med,EMBASE and the Cochrane Library),and international conferences(ASCO,AACR,ESMO and WCLC),randomized controlled trials(RCTs)comparing the efficacy of ICI monotherapy and docetaxel(DOC)from January 1,2000to December 31,2021 were included.Outcome measures contained overall survival(OS),progression-free survival(PFS),objective response rate(ORR),grade 3-4 treatmentrelated adverse events(TRAEs)and grade 5 TRAEs.Pairwise meta-analysis was conducted using Review Manager 5.3.Network meta-analysis was performed using R language 4.0.4 and JAGS 4.3.0 under the framework of Bayesian theory.In addition,subgroup analysis was performed according to tumor histological types and Programmed Cell Death-Ligand 1(PD-L1)expression levels,and the surface under the cumulative ranking(SUCRA)was used to compare the efficacy of ICIs,also to make a relative ranking.Results: Finally,8 RCTs were identified and included,containing 4,999 patients and 6remedies: DOC,nivolumab(NIVO),atezolizumab(ATEZO),pembrolizumab(PEMBRO),tislelizumab(TISLE)and sintilimab(SINTI).In the direct comparison,compared with DOC,ICI monotherapy significantly prolonged OS(HR=0.72,95%CI:0.66-0.78,P<0.00001)and PFS(HR=0.80,95%CI: 0.68-0.94,P=0.005),improved ORR(OR=2.04,95%CI: 1.26-3.31,P=0.004)and reduced the incidence of grade 3-4 TRAEs(OR=0.13,95%CI: 0.08-0.21,P<0.00001),but there was no statistical difference in reducing the incidence of grade 5 TRAEs(P>0.05).In the indirect comparison,NIVO,ATEZO and TISLE had no significant difference in improving OS,PFS and ORR(P>0.05).According to SUCRA values,TISLE was the optimal drug to improve OS,PFS and ORR(0.851,0.905 and 0.884,respectively).These three ICIs also showed no significant difference in safety outcomes(P>0.05),after comparing SUCRA values,it was found that TISLE was the optimal drug to reduce the incidence of grade 3-4 TRAEs(0.883)and ATEZO was the best for reducing the incidence of grade 5 TRAEs(0.797).Through analyzing OS and grade 3-5 TRAEs,the priority ranking was TISLE,NIVO and ATEZO.In the subgroup analysis that compared with DOC,ICIs significantly improved OS both in histological and Programmed Cell Death-ligand 1(PD-L1)expression level subgroups(P<0.0001),however,there was no statistical difference in prolonging OS among several ICIs that were included in corresponding subgroups(P>0.05).According to the SUCRA value of each drug,TISLE was the optimal one to prolong OS in patients with non-squamous cell carcinoma(0.731),squamous cell carcinoma(0.829),PD-L1≥1%(0.851)and PD-L1 <1%(0.685).Conclusion: TISLE,NIVO and ATEZO had high effectiveness and safety in the second-line immunotherapy of advanced NSCLC patients who failed in the first-line platinum-based double-drug chemotherapy,among them,TISLE was the best choice.In subgroup analysis,regardless of tumor histological type and PD-L1 expression level,TISLE had the best efficacy,followed by NIVO. |