Comparison of Extended-Release and Immediate-Release Tacrolimus for Preventing De Novo Human Leukocyte Antigen (HLA) Sensitisation in Patients with Failed Allografts

Background: No consensus exists for maintenance immunosuppression in patients returning to dialysis following allograft failure. De novo HLA is undesirable for those suitable for retransplantation. Continuation of low dose tacrolimus may provide protection against broadening sensitisation, but it has a narrow therapeutic index. The use of extended release (ER-FK) preparations may enable better adherence less complications. Methods: This is an interim analysis of a prospective study, randomising patients to either receive ER-FK or immediate release IR-FK to achieve a trough level of 3-5ng/ml. The primary end point was dnHLA development at 2 yr post-failure. Additional clinical variables,included screening for medication adherence (BAASIS©) and quality of life measures (QoL) (ED-5D-5L) at baseline, 3,6,12,18,24 months. Results: There was no difference in baseline characteristics of 35 patients with at least 1yr follow-up. At 6 months, 42% of patients had developed dnHLA which increased to 54% by 1yr, with the corresponding number of DSA being 23% and 34%; no differences were seen between the ER-FK and IR-FK groups. Adherence scores were low overall, and significantly lower in IR-FK group at 3 months (p=0.02). 25% of the IR-FK group had a trough level <3 at any time point compared with 8% of the ER-FK (p=0.26). Despite randomisation, baseline QoL measures were lower in the IR-FK, which were reflected in the self-reported health visual analogue scale Figure 1. IR-FK group had significantly worse diabetic control (p=0.009) and also higher CRP (p=0.026). Infection-related admissions were significantly higher with IR-FK(p=0.05). 4 patients died. 5 have been retransplanted. Conclusion: We found no difference in dnHLA by FK preparation, but 'sub-therapeutic' levels were more frequent with IR-FK. There was a significant burden of poor QoL measures, which correlated with co-morbidity and may confound both HLA responses and adherence. Multiple considerations are likely required to reduce dnHLA in patients with failed allografts.