Objectives: Predicting the risk of de novo donor specific antibody (dnDSA) is important for long term allograft outcomes after kidney transplantation. We investigated whether HLA class II eplet mismatch was related to dnDSA development and analyzed its combined impact with tacrolimus levels kidney transplantation outcomes. Methods: A total of 347 kidney transplants were included. We used the HLA matchmaker for the single molecular eplet, total eplet, and antibody (Ab) verified eplet mismatch analyses. ROC analysis was used to identify HLA-DR or DQ molecular specific thresholds for the risk of dnDSA development. Next, patients were divided into four groups according to the combination of eplet and tacrolimus (TAC) levels. Time-weighted tacrolimus trough level (TAC-C0) of 5 ng/ml within the 1st post-transplant year or TAC-C0 time-weighted coefficient variability (TWCV) of 20% was applied to find the synergic effects on dnDSA development. Results: De novo HLA class II DSA was detected in 25 (7.2%) patients. High level of mismatch for single molecular eplet (DQ ≥ 10), total eplet (DQ ≥ 12) and Ab verified eplet (DQ ≥ 4) significantly correlated with HLA class II dnDSA development. In the combined analysis of eplet results and TAC-C0, class II dnDSA developed mostly in patients with low TAC-C0 and high eplet mismatch. In the multivariate analyses, low TAC-C0 and high eplet mismatch showed highest hazard ratio for the development of dnDSA. However, no significant synergic effect was observed in dnDSA development according to TWCV.  Conclusions: In conclusion, the determination of HLA class II eplet mismatch may improve the risk stratification for dnDSA development, especially in conjunction with tacrolimus trough levels. These findings could help personalize post-transplant follow-up and tailor immunosuppressive therapy.