T-cell specific antibody induction versus corticosteroid induction immunosuppression for liver transplant recipients: a meta-analysis – Scientific Reports

In this meta-analysis, we sought to compare the efficacy and safety of T-cell specific antibody induction and corticosteroid induction in liver transplant recipients. Our pooled analysis showed that T-cell specific antibody induction did not decrease the incidence of acute rejection compared to corticosteroid induction. However, antibody induction significantly reduced CMV infection, HCV recurrence, DM, and hypertension while the incidence of graft loss and mortality was similar. There was no significant difference in the incidence of renal failure requiring dialysis, but antibody induction significantly decreased serum cholesterol levels. Although the required information size was not reached for all outcomes as revealed by TSA, cumulative evidence crossed the trial sequential boundary for DM and HCV recurrence. However, our results should be interpreted cautiously given the low quality of evidence for most outcomes, high risk of bias and significant heterogeneity of study protocols of the included RCTs, and insufficient information size.

No significant difference in the biopsy-proven acute rejection was found between the antibody induction and corticosteroid induction. We obtained the same results when comparing the subgroups of antibodies with corticosteroids; interleukin-2 receptor antibodies, polyclonal antibodies, basiliximab, and daclizumab These results were consistent in a recent meta-analysis¹⁰. Therefore, we could conclude that the use of antibody induction was not associated with an increased risk of acute rejection, which may allow the substitution of corticosteroid induction with antibody induction.

Our enrolled population had heterogeneity regarding the underlying liver disease. It is well-known that the risk of acute rejection varies depending on the indication of liver transplantation, especially autoimmune liver disease²⁷. In patients with autoimmune hepatitis, acute cellular rejection (ACR) or steroid-resistant ACR may occur more frequently than those with other liver diseases^28,29. Therefore, it is possible that the effect of T-cell specific antibody induction with steroid avoidance or minimization on the incidence of acute rejection may differ in the subgroups with autoimmune hepatitis. It is important to consider the underlying liver disease when interpreting the results of our meta-analysis, and further studies are needed to evaluate the effect of T-cell specific antibody induction with steroid avoidance or minimization on the incidence of acute rejection across different subgroups of liver transplantation.

Patients undergoing liver transplantation with autoimmune diseases such as autoimmune hepatitis, biliary cirrhosis, and primary sclerosing cholangitis, there is debate regarding the benefit or concerns of steroid avoidance or minimization in the immunosuppression regimen^30,31. Patients with autoimmune hepatitis receive corticosteroids prior to transplant. The issue is whether weaning steroids after transplantation contributes to the risk of recurrent autoimmune hepatitis. A previous study of 74 patients with autoimmune liver disease reported acceptable rates of survival and acute cellular rejection without corticosteroid maintenance dose³². Another study of 66 patients showed that the association between steroid withdrawal and recurrence of autoimmune hepatitis was not significant³³. However, the first reported cases of autoimmune hepatitis occurred during weaning or withdrawal of steroid or on a background of long-term calcineurin inhibitor monotherapy³¹. Furthermore, previous studies of steroid withdrawal did not include a separate analysis for patients with autoimmune diseases or excluded such patients because of concerns regarding acute rejection. The use of steroid avoidance or minimization in patients with autoimmune diseases should be evaluated on a case-by-case basis, given the risk of acute rejection and complications associated with steroid-free regimens. We could not provide subgroup analysis for these patients because our meta-analysis also included only a few trials with a small proportion of patients with autoimmune liver disease^21,25.

Meanwhile, we found some significant benefits of antibody induction. The risk of DM was significantly decreased in the T-cell antibody induction group, which was consistent in previous meta-analyses^8,10,11 and our subgroup analyses of different immunosuppressive regimens with corticosteroids. Although the required information size was not reached, the cumulative Z-curve crossed the O’Brien-Fleming boundary for benefit.

The risk of CMV infection was also significantly decreased in antibody induction which was consistent in a recent meta-analysis¹⁰, while not in another previous meta-analysis⁸. This result was consistent in most of our subgroup analysis but there was no difference in the risk of CMV infection when basiliximab was compared to corticosteroids. For this subgroup of basiliximab, the pooled estimate was imprecise and included only two studies.

Furthermore, the risk of HCV recurrence was significantly decreased in the T-cell antibody induction group. Although the required information size was not reached in TSA, the cumulative estimate crossed the trial sequential boundary for benefit. However, none of our subgroup analyses showed significant results for this outcome, which means further studies are required to interpret significant results for HCV recurrence at the subgroup levels. It should be noted that the current use of antivirals both pre and post-transplant may limit the relevance of our findings in the context of HCV recurrence^34,35,36. The American Society of Transplantation Consensus Conference on the use of hepatitis C viremic donors in solid organ transplantation has provided guidelines on the use of HCV-positive organs for transplantation³⁷. Although 16.9% of HCV-infected liver recipients receive liver graft from HCV-positive donors, the possible positive impact of steroid avoidance in hepatitis C viremic donors has not been studied. Furthermore, a previous meta-analysis reported different results regarding HCV recurrence¹⁰, which may be because our updated meta-analysis added a recent trial. The risk of hypertension was significantly decreased in the T-cell antibody induction group. However, some of our subgroup analyses did not show consistent results when comparing daclizumab to corticosteroids and when comparing antibody induction with an intraoperative steroid bolus to corticosteroids.

In patients undergoing liver transplantation with autoimmune diseases such as autoimmune hepatitis, biliary cirrhosis, and primary sclerosing cholangitis, there is debate regarding the benefit or concerns of steroid avoidance or minimization in the immunosuppression regimen^30,31. Patients with autoimmune hepatitis receive corticosteroids prior to transplant. The issue is whether weaning steroids after transplantation contributes to the risk of recurrent autoimmune hepatitis. A previous study of 74 patients with autoimmune liver disease reported acceptable rates of survival and acute cellular rejection without corticosteroid maintenance dose³². Another study of 66 patients showed that the association between steroid withdrawal and recurrence of autoimmune hepatitis was not significant³³. However, the first reported cases of autoimmune hepatitis occurred during weaning or withdrawal of steroid or on a background of long-term calcineurin inhibitor monotherapy³¹. Furthermore, previous studies of steroid withdrawal did not include a separate analysis for patients with autoimmune diseases or excluded such patients because of concerns regarding acute rejection. The use of steroid avoidance or minimization in patients with autoimmune diseases should be evaluated on a case-by-case basis, given the risk of acute rejection and complications associated with steroid-free regimens. We could not provide subgroup analysis for these patients because our meta-analysis also included only a few trials with a small proportion of patients with autoimmune liver disease^21,25.

Previous meta-analyses reported no significant differences in renal function between antibody induction and corticosteroids^8,10,11. This was consistent in our meta-analysis regarding renal failure requiring dialysis. However, in the T-cell specific antibody group, GFR and serum creatinine levels were significantly increased. These results are conflicting with each other because antibody induction was favorable regarding an increase in GFR and steroid induction was favorable regarding the outcome of serum creatinine. Postoperative use of calcineurin inhibitor and its different trough level may act as a confounding factor and we cannot conclusively say the pure effect of immunosuppression regimen on renal function. Also, given the effect size for these outcomes does not correspond to the levels of clinical significance, the high risk of publication bias due to the small number of studies—three for each outcome—included in the analysis, and the insignificance of the most direct outcome of renal failure requiring dialysis, we believe that renal function may not be significantly different between groups.

Our meta-analysis results suggest that T-cell specific antibody induction with steroid avoidance or minimization could lead to a significant reduction in metabolic complications, including diabetes mellitus and hypertension when compared to corticosteroid induction. NASH patients had a higher risk of developing metabolic syndrome after liver transplantation, which can lead to long-term complications such as cardiovascular disease³⁸. Therefore, there is a potential relevance of the metabolic advantage of steroid avoidance to individuals with NASH. However, we could not find any previous study that compared the metabolic outcomes between steroid-free immunosuppression and steroid-based immunosuppression in NASH patients undergoing liver transplantation. Among our included trials, only one trial enrolled a small number of patients with NASH⁴, we could not perform subgroup analysis due to mixed baseline liver diseases of that study. Further studies are required to evaluate whether the benefit of metabolic outcomes by steroid-free immunosuppression is greater in patients with NASH compared to those without.

There were no significant differences in other secondary outcomes including mortality, graft loss, acute rejection requiring treatment, corticosteroid-resistant rejection, and infection, which was consistent in the subgroups comparing different antibodies and immunosuppressive regimens with corticosteroids. Other outcomes such as adverse events, malignancy, the total length of hospital stay, and hyperlipidemia were also not significant. Two previous meta-analyses reported the same results^8,10. While serum cholesterol levels were significantly higher in the corticosteroid group, we deemed this difference insignificant due to the mean cholesterol levels in the corticosteroid group (about 180 mg/dL) being low enough to ignore the difference.

Three previous meta-analyses reported the use of T-cell specific antibody induction for corticosteroid avoidance^8,10,11. However, only one meta-analysis included RCTs comparing T-cell specific antibody induction with corticosteroids with identical concomitant immunosuppressive regimens between groups¹⁰. Wang et al.⁸ and Goralczyk et al.¹¹ analyzed only studies on interleukin-2 receptor antibodies. Also, Wang et al. compared studies with different immunosuppressive regimens other than antibodies and corticosteroids, making it impossible to compare the pure impact of antibody induction versus corticosteroid induction. Goralczyk et al. included non-randomized studies, decreasing the quality of evidence.

Compared to previous meta-analyses^8,10,11, our study analyzed additional outcomes such as acute rejection requiring treatment, length of hospital stay, and serum cholesterol levels. We also performed an additional subgroup analysis to address the effect of omitting the intraoperative bolus dose of corticosteroids. Meta-regression analyses of our primary outcome using recipient age, sex ratio, MELD score, and cold ischemia time were also performed. Regarding secondary outcomes, we observed antibody induction reduced the risk of hypertension and serum cholesterol levels. Compared to a previous meta-analysis¹⁰, the quality of evidence for CMV infection, HCV recurrence, and DM has been increased to ‘moderate’ due to our updated analysis.

Our results should be interpreted cautiously due to the following important limitations.

Firstly, daclizumab has been withdrawn from the market after reports of autoimmune encephalitis in Europe³⁹. Analysis including trials with daclizumab now has a limited value.

Secondly, due to the high risk of bias and imprecision, the majority of our outcomes were at a ‘low’ or ‘very low’ quality of evidence. The majority of the included trials had a high risk of bias due to not blinding participants, personnel, and outcome assessors^{4,5,18,19,20,21,22,23,24}. Thirdly, significant heterogeneity of study protocols should also be considered. Most RCTs differed in inclusion criteria, the type and dose of antibody or corticosteroid, and the type and dose of concomitant other immunosuppressive drugs. Fourthly, the required information size was not reached for all outcomes in TSA and the O’Brien-Fleming boundary for benefit was not crossed for CMV infection and hypertension. Fifthly, we could not find means and standard deviations for recipient age, cold ischemia time, and total hospital stay in one RCT²³. We used medians as means and assumed standard deviations by dividing ranges by four. This method could be used only when the distribution of our variable follows the normal distribution.

In conclusion, according to our updated meta-analysis, T-cell specific antibody induction appears to significantly reduce cytomegalovirus infection and HCV recurrence and metabolic complications including DM and hypertension when compared to corticosteroid induction. However, most included RCTs were assigned a high risk of bias, and the required sample size was not reached for all outcomes. Furthermore, included RCTs were heterogeneous in the immunosuppressive regimens and study protocols. Therefore, additional RCTs with high quality and adequate study power are still needed to assess the efficacy and safety of T-cell specific antibody induction for a higher quality of evidence.