Background and Aims <p>We aimed to explore whether saxagliptin, a dipeptidyl peptidase-IV inhibitor, could ameliorate glucose fluctuations and maintain β-cell function in T1DM.</p> Methods and Results <p>A multicentre, open-label, randomised trial was performed, including 184 T1DM patients from six medical centres. These patients received insulin with or without saxagliptin at 5 mg per day for 24 weeks. The primary endpoint was the change from the baseline value of the MAGE, as measured by a CGMS after 24 weeks. The secondary endpoints included the change from baseline value of islet function during the 3-hour BMTT, HbA1c, and insulin dosage. The exploratory analysis was the influence of SNPs in the incretin-related genes on saxagliptin treatment outcomes. No differences were observed between the two groups in MAGE after treatment for 24 weeks. The change of C-peptide_max levels from baseline to 24 weeks in SAXA group (insulin plus saxagliptin) was higher than in CONT group (insulin only) [<i>p</i> = 0.040]. No difference were observed between the groups in HbA1c, insulin dosage after 24 weeks. In SAXA group, rs10305439, rs10305441 of GLP1R and rs6233 of PCSK1/3 were associated with HbA1c response (<i>p</i> = 0.026, 0.019, and 0.048 respectively); the G allele of rs2143734 of GLP1R were associated with lower change of fasting C-peptide from baseline (<i>p</i> = 0.029)</p> Conclusions <p>The saxagliptin did not ameliorate glucose fluctuations; however, it appeared to maintain β-cell function to some extent, and SNPs in the incretin-related gene may indicate responsiveness to DPP-IV inhibitors in T1DM.</p> <p><b>ClinicalTrials.</b> Gov number, NCT 02307695</p>