LONG-TERM SAFETY OF SIROLIMUS IN LIVER TRANSPLANT RECIPIENTS
Ashraf T, Siddiqui MB, Khorfan K, and Moonka D. LONG-TERM SAFETY OF SIROLIMUS IN LIVER TRANSPLANT RECIPIENTS. Hepatology 2020; 72:823A-824A.
Background: Sirolimus (SRL) has been used as an alternative to calcineurin inhibitors (CI) to help spare renal function in liver transplant (LT) recipients. SRL received a black-box warning in LT because of concerns about hepatic artery thrombosis (HAT). Prior to this, our center initiated a large number of LT patients on SRL. This study assesses the long-term safety of SRL in LT recipients. Methods: This is a retrospective, single-center study of all LT recipients who initiated SRL from 2001-2006. A group that remained on SRL was compared to a group that discontinued SRL (CNI group) for adverse events. Groups were compared for LT outcomes. Kaplan-Meier analyses with Cox regression models were done to determine risk factors (RF), including SRL use, for death or end-stage renal disease (ESRD). Results: 159 patients were included. 109 (68.5%) remained on SRL and 50 (31.4%) discontinued (CNI). There were no differences between groups in age at transplant, gender, race, etiology of liver disease or time after LT to SRL initiation. The mean and median follow up time for the SRL group was 703.87 and 739.71 weeks respectively and was 703.58 and 757.15 weeks for the CNI group. In comparing SRL to CNI groups, there were no differences in HAT (2.8 vs. 2.0%), CAD (6.4 vs. 10%) or CVA (5.7% vs. 11.9%). Fewer patients on SRL died (47.7% vs. 58.0%) or developed ESRD (19.3 vs. 32.0%) though neither difference was significant (P=0.228: P=0.078). No significant differences were noted in GFR between the two groups up to 10 years after LT. In the evaluation of risk factors to time to ESRD, CNI (SRL discontinuation) had a small trend towards earlier onset ESRD on univariate analysis (HR:1.56, CI:0.78, 3.23, p=0.207 but even less so on multivariate analysis (HR:1.366, CI:0.675-2.763) (Figure 1). In the evaluation of RF for time to death, CNI was associated with greater mortality from time of transplant (HR:1.46, CI: 1.06-2.03, p=0.023) and from time of SRL initiation (HR:1.64, CI:1.06- 2.55, p=0.026) (Figure 1). However, when including age, diabetes and GFR at LT in multivariate analyses, the effect of CNI was no longer significant. Conclusion: The results describe one of the longest follow-up periods for SRL use in LT patients, with mean follow up extending over a decade. The results demonstrate that patients receiving SRL after LT did not have an increased risk of developing diabetes, CAD, CVA or HAT. Furthermore, SRL was at least equivalent to CNI in terms of ESRD and mortality. This suggests that SRL, in appropriate LT patients, is safe and that LT patients currently receiving SRL can continue to do so.