Abstract: OBJECTIVE To analyze the adverse reactions of immunological check-point inhibitors causing hypothyroidism by network meta-analysis. METHODS Search for randomized controlled trials published in Embase, PubMed, Cochrane Library, CNKI, Wanfang, and Weipu. Data analysis was performed using WinBUGS 1.4 in conjunction with Stata 14.0 software. RESULTS A total of 2 435 articles were searched and 19 articles were included in the inclusion exclusion criteria, involving 12 253 patients and seven kinds of immunological checkpoint inhibitors used in the market. During the data extraction process, three immunological checkpoint inhibitors of atezolizumab, avelumab, and tremllimumab were found to have no hypothyroidism report, and the remaining four immunological checkpoint inhibitors were administered in 19 different regimens. The top 3 rankings for the probability of hypothyroidism in 19 different dosing regimens were as followed:nivolumab 3 mg·kg^-1, once every 2 weeks+ipilimumab 1 mg·kg^-1, once every 6 weeks, four times of combined administration(PrBest 90.1%, SUCRA 99.1%), nivolumab 240 mg, once every 2 weeks, four times in total(PrBest 4.6%, SUCRA 92.8%), durvalumab 10 mg·kg^-1, once every 2 weeks(PrBest 1.9%, SUCRA 58.8%). The lowest dose schedule for probability ordering were ipilimumab 3 mg·kg^-1, once every three weeks, four times in total, ipilimumab 3 mg·(2 kg)^-1, once every three weeks, four times in total, ipilimumab 10 mg·kg^-1, once every three weeks, four times in total(all PrBest was 0, SUCRA were 11.4%, 11.5% and 13.6%). CONCLUSION The combined administeration of nivolumab 3 mg·kg^-1, once every 2 weeks+ipilimumab 1 mg·kg^-1, once every 6 weeks, four times of combined administration result in the highest probability of hypothyroidism, the second risk is nivolumab 240 mg, once every 2 weeks, four times in total. The three dosing regimens of ipilimumab are ranked lower and they are safer.