Conversion From Calcineurin Inhibitors to Mammalian Target of Rapamycin Inhibitors in Kidney Transplant Recipients: A Systematic Review and Meta-Analysis of Randomized Controlled Trials

Background A systematic review and meta-analysis were performed to investigate the efficacy and safety of conversion from calcineurin inhibitors (CNIs) to mammalian target of rapamycin inhibitors (mTORi) in kidney transplant recipients (KTRs). Methods MEDLINE, EMBASE, PubMed, and Cochrane Library were searched to identify randomized controlled trials (RCTs) that compared the continuation of CNI with conversion to mTORi therapy. Results Twenty-nine RCTs (5,747 KTRs) were included in our analysis. Meta-analysis of the glomerular filtration rate (SMD 0.20; 95%CI 0.10–0.31; P<0.01) and malignancy (RR 0.74; 95%CI 0.55–0.99; P=0.04) demonstrated a significant advantage of mTORi conversion over CNI continuation. However, the risk of acute rejection (RR 1.58; 95%CI 1.22–2.04; P<0.01), infection (RR 1.55; 95%CI 1.01–1.31; P=0.04), proteinuria (RR 1.87; 95%CI 1.34–2.59; P<0.01), leukopenia (RR 1.56; 95%CI 1.27–1.91; P<0.01), acne (RR 6.43; 95%CI 3.43–12.04; P<0.01), and mouth ulcer (RR 11.70; 95%CI 6.18–22.17; P<0.01) were higher in the mTORi group. More patients in the conversion group had to discontinue study medication (RR 2.52; 95%CI 1.75–3.63; P<0.01). There was no significant difference between the two groups with regard to death, graft loss, diabetes, chronic allograft nephropathy, and interstitial fibrosis/tubular atrophy. Conclusions Posttransplant patients have a better graft function and lower incidence of malignancy after conversion from CNI to mTORi therapy. However, this conversion strategy may be prevented by the higher drug discontinuation rate due to mTORi-associated adverse events, such as more acute rejection, infection, proteinuria, leukopenia, acne, and mouth ulcer, indicating that conversion therapy may only be a treatment option in selected patients.


INTRODUCTION
Most kidney transplant recipients (KTRs) currently receive calcineurin inhibitor (CNI) therapy, which has remarkably reduced acute rejection (AR) episodes and improved early graft survival (1). However, long-term CNI exposure may induce irreversible nephrotoxicity, resulting in progressive graft dysfunction (2,3). The CNI can also promote cardiovascular events and malignancies, which are the leading causes of premature death with a functioning graft (4,5). This discrepancy has prompted investigations into CNI retention strategies, which maintain adequate immunosuppressive effects without compromising safety (6).
Mammalian target of rapamycin inhibitors (mTORi) are new immunosuppressants that exhibit little or minimal nephrotoxicity (7)(8)(9). The dual immunosuppressive and antineoplastic properties of mTORi offer a distinct advantage for this class of drugs in the treatment of patients who receive kidney transplant (10,11). Nevertheless, major concerns associated with the de novo introduction of mTORi include the risk of AR, impaired wound healing, and prolonged delayed graft function (DGF), limiting the adoption of mTORi as a first-line immunosuppressant (9,12,13).
Thus, increasing interest has been directed toward exploiting the synergistic actions of CNIs and mTORi by administering the CNI at the initial stage of high-risk AR after transplantation and then converting it into mTORi before the onset of CNI-induced irreversible nephrotoxicity (14). Evidence regarding the clinical benefit of this conversion strategy is conflicting. Budde et al. (15) demonstrated that mTORi combined with CNI withdrawal improved kidney function while maintaining efficacy and safety. In contrast, several studies reported that mTORi therapy not only did fail to show any overall clinical benefit but also mTORi conversion was associated with more adverse events (16) and discontinuations (17). Thus, further evidence for the validity of mTORi conversion strategy is urgently needed (18). Our study aimed to systematically identify and evaluate the currently available evidence from randomized controlled trials (RCTs) on the outcomes of conversion from the CNI to mTORi for KTRs.

MATERIALS AND METHODS
Our systematic review was conducted and reported in accordance with the Preferred Reporting Items for Systematic Reviews and Meta-Analyses (PRISMA) guidelines (19).

Eligibility Criteria
Studies were eligible if they met the following criteria: (a) RCTs of conversion from CNIs to mTORi maintenance immunosuppressive regimen; (b) participants were the recipients of a single organ kidney transplant (first or repeat) from a living or deceased donor; (c) initial immunosuppression consisted of a CNI but not an mTORi; and (d) KTRs were randomly assigned to either continue with current CNI therapy or be converted from the CNI to mTORi. Dose modifications of CNIs, mTORi, and other concomitant immunosuppressants were permissible after randomization. Randomized trials with less than 20 cases in each group or abstracts of major transplant conferences were excluded. KTRs with a history of malignancy (other than adequately treated non-melanoma skin carcinoma) before randomization were excluded. To minimize participant overlap, when the same study was reported many times, the study with long follow-up period and complete case report was identified as the primary data source.

Search Strategy
MEDLINE, EMBASE, PubMed, and Cochrane Library were searched from the start date of each resource up to November 12, 2020 (the last literature search) using logical combinations of relevant keywords and medical subject headings that contained all spellings of the following terms: kidney transplantation, calcineurin inhibitor, CNIs, tacrolimus, cyclosporine, conversion, mTORi, sirolimus, and everolimus. No restrictions regarding the language of the publication, age of KTRs, or concomitant immunosuppressants were applied. The results were supplemented by manually searching reference lists of relevant reviews and included articles. All citations identified by this search strategy were evaluated by two independent reviewers (JZ and QZ) using titles, abstracts, and, where required, the full text to determine eligibility.

Outcome Measures
The primary endpoints were graft function, AR, graft loss, and death. The secondary outcomes were adverse events (AEs), infections, malignancies, and discontinuation of medication.

Data Extraction
Eligible RCTs were referred to throughout this paper by the first author (JZ and QZ) and the year of the earliest peer-reviewed publication. Data extraction was independently performed by two reviewers (JZ and QZ) using a predesigned form on which study design, participant characteristics, interventions, and outcome data were recorded. Missing information was requested from the trial authors or sponsors. They met to combine their findings, and the information was subsequently entered into Review Manager (RevMan) (Version 5.4, Copenhagen: The Nordic Cochrane Centre, The Cochrane Collaboration).

Quality Assessment and Statistical Analyses
Cochrane Collaboration's tool was used to assess the risk of bias of randomized controlled trials (20). RevMan 5.4 was utilized for the execution of the meta-analysis. Dichotomous outcomes were expressed as risk ratios (RRs) with 95% confidence intervals (CIs), and continuous variables were expressed as mean differences (MDs) with 95% CI. Data were pooled using the standardized mean difference (SMD) as the summary effect size metric. P<0.05 was considered statistically significant. Metaanalysis was conducted using the fixed effect model in the absence of heterogeneity; otherwise, the random effect was applied. Heterogeneity was quantified via the Cochrane Q (P<0.1) and I 2 statistics (I 2 >50%). If heterogeneity existed, a sensitivity analysis was performed to explore possible sources of heterogeneity.

Search and Selection
The search and selection flow chart is illustrated in Figure 1. Initial literature searches identified 1,488 citations across all databases. The titles and abstracts were reviewed, followed by full-text review of potentially eligible articles and abstracts, of which 29 RCTs met the inclusion criteria (15,17,. The baseline overview of included studies is reported in Table 1.

Description of Eligible Trials
There were 29 eligible RCTs, in which a total number of 5,747 KTRs were included, whose characteristics are summarized in Table S1. Among these 29 trials, 22 trials reported the time of conversion to mTORi within 6 months after transplantation; the rest were more than 6 months. A total of 16 RCTs reported the impressive therapy of conversion from the CNI to everolimus. The patient sample size ranged from 49 to 830, and follow-up duration was from 6 to 24 months.

Quality of Included Trials
The quality of the 29 clinical trials was evaluated by Cochrane Collaboration's tool ( Figure S1). A total of 14 studies exhibited an adequate random sequence generation process, 15 trials described the methods used for allocation concealment, 2 RCTs illustrated performance bias, and 4 studies had unreported data. In short, the studies included had low to moderate risk of bias.

Primary Endpoints
The meta-analysis results of the primary endpoints are reported in Figures 2 and 3. A total of 22 trials reported renal function; the meta-analysis of the estimated glomerular filtration rate (eGFR) demonstrated a statistically significant advantage for mTORi conversion over CNI continuation (SMD 0.20; 95%CI 0.10-0.31; P<0.01) in Figure 2A. A total of 28 trials reported AR; we found that patients converted to mTORi were at a higher risk of developing AR (RR 1.58; 95%CI 1.22-2.04; P<0.01) in Figure 2B. A total of 22 trials reported graft loss; there was no difference between patients converted to mTORi and those remaining on the CNI (RR 1.14; 95%CI 0.75-1.74; P=0.53) in Figure 3A. A total of 22 trials reported mortality; we also found that there was no statistical difference in mortality between the mTORi conversion group and the CNI continuation arm (RR 0.99, 95%CI 0.64-1.55; P=0.97) in Figure 3B.

DISCUSSION
The meta-analysis of the 29 eligible RCTs (involving 5,747 KTRs) demonstrated that the conversion from the CNI to  mTORi after kidney transplantation was associated with an improvement in graft function and a reduced incidence of malignancy. There was no significant difference between the groups with respect to IF/TA or chronic allograft nephropathy, indicating that the improvement in graft function after conversion may primarily be due to effects other than structural improvement (48). Ten years ago, mTOR inhibitors were more commonly used, but only 1.9% of recipients were prescribed with them at transplant in 2016, increasing to 4.3% at 1-year posttransplant (49). Current studies reported that the combination of low-dose CNI (tacrolimus) and everolimus can make the difference in order to obtain the best clinic benefit for KTRs (50). Nevertheless, conversion to mTORi therapy may be prevented by the high discontinuation rate (51) due to the intolerable AEs, such as the development of serious infections, AR, mouth ulceration, proteinuria, anemia, leucopenia, thrombocytopenia, dyslipidemia, acne, diarrhea, and edema. All of these changes were statistically significant between the two groups. Thus, the application of mTORi conversion must be weighed during clinical decision-making, with the potential benefits and risks carefully assessed for each KTR. Previous systematic review had suggested that conversion to sirolimus was associated with an improvement in short-term renal function, while none of the identified studies investigated everolimus (18,52). Similarly, Lim et al. (53) reviewed RCTs comparing delayed conversion from CNIs to mTORi indicating that patients converted to mTORi up to 1-year posttransplant had higher GFR compared with those remaining on the CNI. Those findings were highly consistent with our meta-analysis, but we also found that there was no difference in CMV and BKV infection, and no difference in new-onset diabetes. Wolf et al. (54) concluded that the incidence of infections is lower when mTORi is combined with a CNI compared to a standard CNI therapy following renal transplantation, while we found that mTORi conversion therapy had more risk of infections.
In this systematic review, we identified all mTORi conversion RCTs involving both sirolimus and everolimus with heterogeneous treatment and varying follow-up period. We included the RCTs with more than 20 cases in each group, which made the evidence in this review stronger and more convincing. We also evaluated some outcomes (graft function, acute rejection, graft loss, death) and some secondary outcomes that are not commonly reported in other systematic reviews such as drug discontinuation and adverse events. Consequently, the current study provides critical information to guide clinical decision-making on the mTORi conversion strategy after kidney transplantation.
Nevertheless, there were potential caveats limiting the generalizability of our findings. First, besides mTORi-induced AEs, the open-label nature of the 29 RCTs may have Fifth, due to the difference in the number of included cases, we did not distinguish the difference between CsA and TAC conversion. Determining the optimal timing or indication for mTORi conversion requires a balance between avoiding the high rejection risk in the initial posttransplant period and minimizing the progressive development of CNI-related nephrotoxicity. Conversion ≥1-year posttransplant may be insufficient to prevent progression to graft dysfunction, except in KTRs with normal graft function and without proteinuria (14). Conversion in the first 6 months after transplantation seems appropriate for maintenance therapy in KTRs after careful screening (14). Refining selection criteria may help both in identifying patients who will profit most from switching and in alleviating the need to reintroduce CNI therapy. Primarily based on the current consensus, the target population for mTORi A B FIGURE 2 | The forest plot of studies that compared conversion from CNIs to mTOR inhibitors versus maintenance of CNI therapy for the outcomes of (A) renal function (eGFR) and (B) acute rejection (AR).
conversion should be KTRs with a baseline GFR of more than 40 ml/min and normal urinary protein excretion, an absence of previous AR and subclinical rejection, and appearance of donorspecific antibody (14,25). Periodic surveillance biopsies may also help to identify a subset of KTRs who do not develop subclinical rejection or irreversible pathological changes in the renal interstitial to guide mTORi conversion therapy (56).
The high rate of drug discontinuation in the conversion group illustrates the challenge of handling a new spectrum of AEs that differs from more familiar CNI-related AEs, including dyslipidemia, acne, mouth ulceration, proteinuria, and AR (57,58). Most AEs occurred early after conversion and in some cases resolved without intervention within a few weeks. Some of the discontinuations in the conversion group could have been avoided with better patient selection and more refined management of mTORi-related AEs. Specific patient monitoring and the concomitant use of lipid-lowering agents, angiotensin-converting-enzyme inhibitors, angiotensin receptor blockers, and induction therapies should be considered for combating dyslipidemia, proteinuria, and AR (16,59). The identification of the optimal timing of conversion, appropriate candidates, dosing regimens for conversion, and safe management of mTORi-related AEs will be key in minimizing the need for discontinuation, which currently limits the broad applicability of this strategy.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/Supplementary Material. Further inquiries can be directed to the corresponding authors.

AUTHOR CONTRIBUTIONS
JZ designed the study and wrote the manuscript. XF, QZ, and LL performed the meta-analysis and did the data extraction. SF, YF, TS, and ZH checked the results. XW and TL designed the