Recurrence of autoimmune hepatitis cholestatic variant syndromes after liver transplantation affects graft and patient survival

Abstract

Background & aims: A significant proportion of patients with variant syndromes (VSs), namely autoimmune hepatitis/primary biliary cholangitis or autoimmune hepatitis/primary sclerosing cholangitis, require liver transplantation (LT) despite treatment. The frequency of disease recurrence and the effect on graft survival are yet to be clarified. The aim of this international, multicentric, retrospective study is to evaluate the risk factors associated with recurrence and the impact of the disease recurrence after LT on graft and patient survival.

Methods: We evaluated 166 patients undergoing LT for VS in 33 centers in North America, South America, Europe, and Asia. Clinical data before and after LT, biochemical data within the first 12 months after LT, and immunosuppression after LT were analyzed to identify patients with a higher risk of recurrence of autoimmune disease based on a histological and radiological diagnosis. Cumulative probabilities of graft and overall survival after LT were calculated using a semi-Markov model.

Results: The autoimmune pattern of recurrence resembled the original VS in 19 cases (61%). Recurrence of autoimmune liver disease (rALD) after LT was observed in 23% and 33% of patients after 5 and 10 years, respectively. Increased alkaline phosphatase (hazard ratio [HR] 1.60, 95% confidence interval [CI] 1.13-2.25, p <0.01) and alanine aminotransferase (HR 1.25, 95% CI 1.01-1.53, p = 0.03) at 12 months after LT and acute rejection (HR 3.58, 95% CI 1.60-7.73, p <0.01) were associated with a higher risk of VS recurrence, whereas the use of predniso(lo)ne was associated with a reduced risk (HR 0.30, 95% CI 0.14-0.64, p <0.01). After adjusting for alanine aminotransferase and alkaline phosphatase at 12 months, the use of predniso(lo)ne was found to be independently and negatively associated with recurrent disease. The rALD was found to be significantly associated with graft loss and patient survival in the multivariate Cox regression analysis with a time-dependent covariate. The 5- and 10-year probabilities of graft survival were 68% and 41% in patients with recurrent VS compared with 83% and 60% in patients without recurrent disease, respectively (p = 0.01). The overall survival was significantly reduced in patients with recurrent disease (p = 0.01), with event probability at 5 and 10 years of 75% and 49% vs. 84% and 60% in patients without recurrence, respectively.

Conclusions: rALD after LT is frequent and is associated with elevation in liver enzymes within the first year after LT and rejection episodes. According to our data, VS recurrence appears to be associated with poorer graft and patient survival. Further studies are needed to explore strategies that can prevent VS recurrence or mitigate its potential impact.

Impact and implications: This study investigated the recurrence of autoimmune liver diseases (rALD) in patients transplanted for variant syndromes (VSs) and its effect on graft and patient survival. The findings reveal a significant association between rALD and poorer graft and overall survival, highlighting the need for preventive strategies. This research is crucial for transplant physicians and healthcare providers, as it underscores the impact of early liver enzyme monitoring and tailored immunosuppressive therapy on long-term outcomes. These insights can inform more effective post-LT management protocols, potentially improving patient prognosis.

Keywords: Autoimmunity; Liver transplantation; Recurrence.

Graphical abstract

Fig. 1

Cumulative probability of rALD. The probabilities of rALD were 23%, 33%, 55%, and 77% at 5, 10, 15, and 20 years, respectively. The cumulative probabilities were calculated using the Kaplan–Meier method. rALD, recurrent autoimmune liver disease.

Fig. 2

Overall and graft survival. (A) Overall survival of VS patients after LT. The probabilities of survival at 5, 10, 15, and 20 years were 84%, 60%, 50%, and 38%, respectively. Cumulative probabilities were calculated using the Kaplan–Meier method. (B) Graft survival of VS patients after LT. The graft survival probabilities at 5, 10, 15, and 20 years were 83%, 60%, 41%, and 36%, respectively. Cumulative probabilities were calculated using the Kaplan–Meier method. (C) Graft survival in patients with and without recurrent VS after LT using the semi-Markov models (‘clock reset’ model) (p = 0.01). Cumulative probabilities were calculated using the Kaplan–Meier method. Patients who had no rALD during their follow-up are represented by the pink line. Patients who developed rALD are represented by the pink line only until they developed rALD. These patients are then censored and switched to a new survival curve (green line) once they have rALD. The time is then reset as time 0 for their further follow-up. (D) Overall survival in patients with and without recurrence of autoimmune liver disease after LT using the semi-Markov models (‘clock reset’ model) (p = 0.001). Cumulative probabilities were calculated using the Kaplan–Meier method. Patients who had no rALD during their follow-up are represented by the pink line. Patients who developed rALD are represented by the pink line only until they developed rALD. These patients are then censored and switched to a new survival curve (green line) once they have rALD. The time is then reset as time 0 for their further follow-up. LT, liver transplantation; rALD, recurrent autoimmune liver disease; VS, variant syndrome.