Improved Medication Adherence with the Use of Extended-Release Tacrolimus in Liver Transplant Recipients: A Pilot Randomized Controlled Trial

Background Nonadherence to immunosuppression in liver transplant recipients (LTRs) leads to deterioration in health outcomes. Once-dailyextended-release tacrolimus (TAC-ER) may improve adherence when compared to twice-dailyimmediate-release tacrolimus (TAC-IR). Methods We conducted a randomized controlled study to evaluate medication adherence, clinical efficacy, and safety of TAC-ER in stable LTR. All patients >18 years who underwent liver transplantation before 6 months were eligible. Patients were randomized 1 : 1 to continued TAC-IR or conversion to TAC-ER. The primary outcome was change in medication adherence from baseline to 9 months, assessed using BAASIS. Secondary outcomes were tacrolimus trough levels, safety, and quality of life. Results Thirty-one patients were consented and randomized to either of the two groups: conversion to TAC-ER (n = 15) or continued TAC-IR (n = 16). Six patients in the TAC-ER group withdrew after randomization due to apprehension about switching medication (n = 2), unwillingness to travel (n = 2), and increased liver tests after conversion (n = 2, both were acute rejections despite therapeutic tacrolimus levels and were considered unrelated to TAC-ER). We compared the results of nine patients in the TAC-ER group that completed the study with those of sixteen in the TAC-IR group. At baseline, there was no difference in tacrolimus trough levels between groups. Improved adherence was observed in the TAC-ER group as 100% of patients reported at least one period of full adherence during the study period (100% vs. 62.6%, p = 0.035). Tacrolimus trough levels and liver tests were comparable between groups throughout the study. There were no differences in eGFR, HbA1c, or QoL between the groups. Conclusion TAC-ER improved medication adherence while maintaining comparable trough levels, liver function, and QoL as TAC-IR in LTR.


Introduction
Liver transplant recipients (LTRs) are required to take immunosuppressive medications throughout their life. Nonadherence is common and is variable over time [1]. It is estimated that roughly 20-62% of adults and about 50% of pediatric LTR are nonadherent to their immunosuppressive medication regimen [1,2]. Te most serious outcomes of immunosuppression nonadherence (IMNA) include acute/ chronic graft rejection and death. Much of the data pertaining to IMNA derives from the kidney transplant population, but studies have demonstrated worsened graft survival in LTR with IMNA [3]. Meta-analytic data have shown that, on average, 6.7 per 100 LTR per year are nonadherent with immunosuppressant medications [4]. However, this is likely underestimated, with other data reporting IMNA rates as high as 15-40% among LTR [5]. A Scottish article estimated a 30% chronic rejection rate and 10% mortality due to IMNA [6]. In addition to unfavorable patient outcomes, complications related to IMNA contribute to massive costs to the healthcare systems-an estimated $15 to 100 million annually for solid-organ transplants [2].
Nonadherence to medication is not unique to liver transplant patients; it has been seen in patients with kidney, pancreas, heart, and lung transplant as well [7,8]. Currently, the most common immunosuppressive maintenance protocol prescribed for solid-organ transplants is a triple therapy comprising a calcineurin inhibitor (tacrolimus/cyclosporine), mycophenolate, and a steroid agent [9]. Tese agents are favored among physicians for their well-understood mechanisms of action and among patients for their relative costs compared to the newer biologic agents. However, when prescribed triple therapy in addition to antimicrobial prophylactic medications, transplant patients face a high pill burden and are required to take multiple tablets daily for the rest of their life [10]. Tis is likely a signifcant factor contributing to IMNA, despite patients understanding that these medications are required for their survival [11].
One of the medications in question, tacrolimus, is formulated as either a twice-dailyimmediate-release capsule (TAC-IR, Prograf ® ) or as a once-dailyextended-release formulation labeled as Envarsus XR ® (TAC-ER). Notably, another extended-release formulation (Advagraf/Astagraf ) has been commonly used in medication adherence research. Reports in the literature have documented the benefts of the extended-release formulation of tacrolimus, with evidence of increased medication adherence in kidney and kidneypancreas transplant recipients [12].
We conducted a pilot randomized controlled study to compare the potential benefts and safety of once-daily TAC-ER/Envarsus XR ® versus twice-daily TAC-IR in LTR. Te primary outcome is a change in self-reported medication adherence from baseline to 3 months. Te secondary outcomes include change in the quality of life (QoL) and safety.

Study Design.
Tis was a single-center, open-label, twoarm, parallel-group, randomized controlled trial (NCT03386305), enrolling patients from Jan 2018 through June 2021. Te study was approved by the local Institutional Review Board (IRB#5024), and written informed consent was obtained from all patients in accordance with the Declaration of Helsinki. Eligible LTR Patients were as follows: (1) ≥ 18 years; (2) at least > 6 months and <5 years post-LT; (3) had stable kidney and liver tests (defned as serum creatinine ≤ 2.5 mg/dL and liver function tests (AST/ALT/alkaline phosphatase (ALP) ≤ 2 times upper limit of normal) for at least 4 weeks before enrollment; (5) on a stable dose of TAC-IR confrmed with maintaining tacrolimus trough levels between 3-12 ng/mL for at least 4 weeks before enrollment. Exclusion criteria were as follows: combined liver-kidney transplantation; concomitant use of a medication with a known cytochrome (CYP) 450 interaction with tacrolimus; received treatment for rejection within 30 days of enrollment; recurrent or active hepatitis C infection or receiving a hepatitis C antiviral agent; and documented gastrointestinal malabsorption. Randomization was conducted using a pregenerated computerized list. Patients were randomized 1 :1 to continue TAC-IR at their current dose or be converted to TAC-ER (Envarsus XR ® ; Veloxis Pharmaceuticals, Inc.) at a dose conversion ratio of 0.8 :1. Subsequent dose adjustments were permitted to maintain tacrolimus trough levels at 4-8 ng/mL, as assessed 7 and 14 days after conversion. Patients converted to TAC-ER were provided the study drug and followed for 9 months, at which time they could decide to continue TAC-ER or return to TAC-IR. Routine adjustment of background immunosuppression (e.g., steroid tapering) was permitted per clinician judgment of transplant surgeons and hepatologists.

Study Outcomes.
Te primary outcome was change in patient-reported immunosuppression adherence from baseline to 9 months, assessed using the Basel Assessment of Adherence with Immunosuppressive Medications Scale (BAASIS). BAASIS measures taking, skipping, and dose reduction of drugs, with a recall period of 4 weeks. It consists of 4 questions with a 6-point response scale (ranging from never to every day). An additional overall adherence is ranked on a scale of 0 to 100 using a visual analog scale. It can be completed by patients themselves or by an interviewer.
Te change in QoL was assessed as a secondary endpoint using PROMIS-29 (Patient-Reported Outcomes Measurement Information System). PROMIS is a result of the NIH's support to develop a psychometrically validated, dynamic system to measure QOL. PROMIS-29 V2.0 comprises a set of 29 questions evaluating the following seven QOL domains: physical function, anxiety, depression, fatigue, sleep disturbance, social function, and pain [13]. Te scores are reported as a T score for all domains except pain (mean 50, SD � 10) centered on the sample representative of the 2000 US general census considering demographic variables. Pain intensity is assessed using a single item, on a 0-10 scale. It is available in the public domain for research use.
BAASIS and PROMIS-29 were assessed at baseline and at 4 weeks, 3 months, 6 months, and 9 months after the enrollment. Additional secondary efcacy and safety endpoints included tacrolimus trough levels, liver tests (AST, ALT, ALP, GGT), kidney function (serum creatinine, BUN, and eGFR), and hemoglobin A1c collected at baseline and at 3, 6, and 9 months from baseline. Te incidence of allograft rejection was collected throughout the study.

Statistical Methods.
Given the pilot nature of the study and the complexity of enrolling this population, we aimed to enroll 30 patients in this study. Demographics, safety profle, IMNA, and QoL were compared between the two groups at the baseline and study completion. Adherence to the TAC-IR or TAC-ER is an evolving variable. Te overall adherence status was considered by reporting at least onetime full adherence between week 4 and month 9. Continuous variables were examined utilizing the t-test or Mann-Whitney U test. Chi-squared testing was used for categorical variables. P values were 2-sided, and alpha was set at 0.05. All analyses and graphs were performed using the

Results
31 patients were consented and randomized. Six patients in the TAC-ER group withdrew after randomization due to apprehension about switching medication while stable (n � 2), unwillingness to travel to receive study medications (n � 2), and increased liver tests after conversion (n � 2, both were deemed acute rejections, despite therapeutic tacrolimus levels but were considered unrelated to TAC-ER conversion). Of these 6, fve were male, 4 were African American, and 2 were Caucasian. 9 patients in the TAC-ER group that completed the study, were compared with 16 subjects in the TAC-IR.

Medication Adherence.
Using the BAASIS instrument, we considered three levels of adherence: never taken as prescribed, sometimes taken as prescribed, and always taken as prescribed.
At baseline, fewer patients in the TAC-ER group noted IMNA within the four weeks prior to enrollment; however, this was not statistically signifcant (33% vs. 62%, p � 0.16). During the study period, the pattern of adherence changed at 4 weeks, 3 months, 6 months, and 9 months. Full adherence (always taken as prescribed) throughout the entire study was only reported in three patients, two in the TAC-ER group and one in the TAC-IR group; however, this was not statistically signifcant. Te proportion of patients who reported at least one-time full adherence over the study period was greater in the TAC-ER vs. the TAC-IR group (100% vs. 62%, p � 0.035). Notably, the percentage of patients reporting better adherence remained higher in the study group than in the control over time. Figure 2). At the baseline, physical function and social function in our study population were below the mean (SD) 50 (10) of the general population. No component of PROMIS-29 difered between groups. However, the mean score for sleep disturbances trended towards better sleep in the TAC-ER group (54.3 (±4.2) vs. 51.2 (±3.6), p � 0.069). No signifcant change in other PROMIS domain scores was found over time.

Discussion
Nonadherence has been identifed as a major modifable risk factor for poor outcomes in liver transplantation by the Consensus on Managing Modifable Risk in Transplantation (COMMIT) group [14]. Studies have shown that IMNA is often a persistent problem and one which may actually worsen with time [15,16]. Minimizing pill burden and using once-daily dosing is a potential strategy to address nonadherence [17,18]. We report better medication adherence with the use of once-daily TAC-ER when compared to twicedaily TAC-IR. Tere were no diferences in clinical or safety outcomes between the two study groups.
Clinical data using TAC-ER in the setting of liver transplantation are limited. A phase 2 study published by Alloway et al. demonstrated safe conversion to TAC-ER in 57 stable LTR who were a median of 32.2 months after transplant [19]. Of these, 43 subjects completed 52 weeks of treatment in an extension phase of the study. Te mean therapeutic dose was 30% lower with TAC-ER after conversion (6.1 mg and 4.8 mg). Tere were three discontinuations due to adverse events in the TAC-ER group, and one possibly related rejection in the TAC-ER group which was resolved. In a pharmacokinetic phase 2 study, 58 LTR were randomized to de novo TAC-ER or TAC-IR, with 35 subjects completing a 52-week extension period [20]. Adverse event rates were similar between groups, and there were 6 and 4 rejections observed in the TAC-ER and TAC-IR groups, respectively. Our study did not show signifcant diferences in adverse events or rejections in a 9-month study period after conversion.
Adherence is unstable and varies over time. Longerterm studies with TAC-ER would be of great interest, especially considering a recent European Liver Transplant Registry study of over 13,000 patients, which demonstrated superior long-term patient and graft survival (up to 8 years) in LTR either initiated and maintained on, or converted to, TAC-ER [21]. Our study has notable limitations, particularly high drop-out rates, low sample size, and imbalance in a number of patients in the two comparative arms. Tere were patient-related factors such as unwillingness to take the risk of converting to a new medication when they were already on a stable dose and difculty in coming to the clinic for research visits and picking up study medications (while their current medications could be shipped directly to their home). Moreover, the lack of blinding may have added bias to our  Journal of Transplantation patient's responses, though admittedly it is difcult to blind patients when the number of pills taken during the day is by nature impossible to hide from the patient.
In conclusion, this study suggests improved adherence and acceptable clinical outcomes after conversion to TAC-ER in a stable liver transplant population. A larger and longer-term study is needed to assess the impact of improved adherence on clinical outcomes after conversion to TAC-ER.

Data Availability
Data are available from the corresponding author upon request.

Conflicts of Interest
Te authors declare that they have no conficts of interest.