Personalized Immunosuppression in Transplantation: Role of Biomarker Monitoring and Therapeutic Drug Monitoring

By Amitava Dasgupta

Personalized Immunosuppression in Transplantation: Role of Biomarker Monitoring and Therapeutic Drug Monitoring provides coverage of the various approaches to monitoring immunosuppressants in transplant patients, including the most recently developed biomarker monitoring methods, pharmacogenomics approaches, and traditional therapeutic drug monitoring.

The book is written for pathologists, toxicologists, and transplant surgeons who are involved in the management of transplant patients, offering them in-depth coverage of the management of immunosuppressant therapy in transplant patients with the goal of maximum benefit from drug therapy and minimal risk of drug toxicity.

This book also provides practical guidelines for managing immunosuppressant therapy, including the therapeutic ranges of various immunosuppressants, the pitfalls of methodologies used for determination of these immunosuppressants in whole blood or plasma, appropriate pharmacogenomics testing for organ transplant recipients, and when biomarker monitoring could be helpful.

• Focuses on the personalized management of immunosuppression therapy in individual transplant patients
• Presents information that applies to many areas, including gmass spectrometry, assay design, assay validation, clinical chemistry, and clinical pathology
• Provides practical guidelines for the initial selection and subsequent modifications of immunosuppression therapy in individual transplant patients
• Reviews the latest research in biomarker monitoring in personalizing immunosuppressant therapy, including potential new markers not currently used, but with great potential for future use
• Explains how monitoring graft-derived, circulating, cell free DNA has shown promise in the early detection of transplant injury in liquid biopsy

• Language: English
• Publisher: Elsevier Science
• Release date: Aug 8, 2015
• ISBN: 9780128011331

Book preview

Germany

Chapter 1

Overview of the pharmacology and toxicology of immunosuppressant agents that require therapeutic drug monitoring

Michael C. Milone,    Hospital of the University of Pennsylvania, Philadelphia, PA, USA

This chapter provides an overview of the mechanisms of action, pharmacological parameters, reference range, and toxicological aspects of various immunosuppressants that require therapeutic drug monitoring. These immunosuppressants include cyclosporine, tacrolimus, sirolimus, everolimus, and mycophenolic acid. Chemical structures of these immunosuppressants are also provided.

Keywords

Cyclosporine; tacrolimus; sirolimus; everolimus

1.1 Introduction

More than 100,000 solid organ and 50,000 allogeneic bone marrow transplants are currently performed worldwide each year. Outcomes vary widely depending on the transplant type and underlying disease; however, solid organ allograft survival has improved significantly during the past quarter century coinciding with the introduction of new immunosuppressive drugs (ISDs). ISDs are critical to transplantation success due to the potent cellular and humoral immune mechanisms that restrict allogeneic transplantation. Whereas early ISDs consisted primarily of glucocorticoids and antimetabolite drugs to block lymphocyte proliferation, several ISDs with differing mechanisms of action have been introduced, including the recent introduction of the first biologic agent, belatacept (Nulojix, a CTLA4–Ig fusion protein), which interferes with a critical step in the initiation of T cell-mediated immunity. Although these agents have significantly improved outcomes, their benefits often come at a cost of increased risk of infection as well as toxicity. The use of ISDs to control allograft rejection and graft-versus-host disease is very difficult because errors can lead to serious and sometimes fatal consequences for the transplant recipient.

One of the greatest challenges to effectively using ISDs is their widely variable pharmacokinetic behavior across individuals. This pharmacokinetic variability makes it difficult to predict a priori an individual’s response to a drug following administration of a particular dose. Applying knowledge of a drug’s concentration and pharmacokinetic behavior within an individual to the clinical use of a drug, often termed therapeutic drug monitoring (TDM), has therefore become a standard approach to ISD therapy aimed at mitigating the risks associated with the use of these drugs. An important prerequisite to successful TDM is the ability to measure a drug of interest. Using modern technologies that are available within most analytical chemistry laboratories, the measurement of drugs, including ISDs and their metabolites, is readily achieved as described in the following chapters of this book.

Unfortunately, having the plasma or whole blood concentration of a drug is not enough for proper patient management. Effective use of drug concentration data also requires a thorough understanding of the pharmacodynamics relationship between drug exposure and important clinical outcomes of toxicity or efficacy. Like pharmacokinetics, the pharmacodynamics of ISDs also vary greatly across individuals [1], but a measured drug concentration does not provide insight into this variability. Biomarkers of organ function, tissue injury, and immune function provide some insight into the pharmacodynamics of ISDs. In the broadest sense, TDM may be considered to encompass an array of testing modalities beyond traditional concentration monitoring, such as the use of serum creatinine to monitor the nephrotoxic effects of drugs such as the calcineurin inhibitors, tacrolimus, or cyclosporine. Several of the subsequent chapters are devoted to exploring these biomarker-based testing approaches. In this chapter, the ISDs currently approved for use in solid organ and bone marrow transplantation are discussed with a focus on their pharmacology and clinical use.

1.2 Calcineurin Inhibitors

Currently, two calcineurin inhibitors (CNIs), cyclosporine and tacrolimus, are commonly used clinically as immunosuppressants. This section provides an overview of these two drugs.

1.2.1 Cyclosporine A

Introduced in the 1980s, cyclosporine A (CsA) revolutionized the care of transplant patients through its potent inhibition of acute cellular transplant rejection. Although its use has gradually been replaced by tacrolimus, it is currently used in approximately 10% of transplants. It is typically used in combination with other immunosuppressive drugs such as mycophenolic acid, azathioprine, and glucocorticoids. Originally isolated in 1969 from the soil fungus Tolypocladium inflatum by Hans Peter Frey, a biologist working at Sandoz Pharmaceuticals, CsA is a lipophilic, cyclic endecapeptide composed of N-methylated amino acids, making it resistant to intestinal digestion as shown in Figure 1.1A. It is highly lipophilic and only slightly water-soluble. It derives its primary immunosuppressive activity by selectively binding to cyclophilin A, a peptidylprolyl isomerase present within the cytoplasm of cells. Once bound, the CsA/cyclophilin complex inhibits the enzymatic activity of the calcineurin (CN), a heterodimeric, calcium-dependent serine/threonine phosphatase composed of CNA and CNB subunits that is activated by the rapid rise in intracellular calcium following T cell receptor engagement. CN removes a critical regulatory phosphorylation on nuclear factor of activated T cells (NFATc) triggering its translocation to the nucleus of T cells, where it synergizes with other factors to mediate the transcription of a large number of genes, including interleukin-2 (IL-2), an important cytokine for T cell proliferation, and CD40 ligand (CD40L), an important costimulatory ligand for B cells, as schematically diagrammed in Figure 1.1B. Although the calcineurin–NFATc pathway is critical to T cell activation, this pathway plays a role in diverse cell types, including neurons [2,3], skeletal and cardiac myocytes [4,5], and endothelium [6]. These non-immune roles of calcineurin–NFATc signaling may contribute to the toxicity observed with the clinical use of cyclosporine, which includes nephrotoxicity, neurologic toxicity (e.g., tremors and headaches), and diabetes.

Figure 1.1 Calcineurin inhibitors and their mechanism of action.

(A) Structure of cyclosporine A; (B) structure of tacrolimus;(C) schematic of calcineurin–NFATc signaling pathway in T cells that is inhibited by CsA and TRL. CRAC channel, calcium release activated channel (Orai1); PLCγ, phospholipase Cγ; IP3, inositol triphosphate.

Due to the highly lipophilic nature of CsA, the original therapeutic formulation of CsA (Sandimmune) was an oral solution of the drug dissolved in oil. This solution was then mixed with a liquid such as juice prior to consumption. Early pharmacokinetic studies revealed that CsA absorption with this formulation was slow and erratic with poor bioavailability, leading to significant intra- and interindividual variability in CsA exposure. Studies of CsA given to healthy volunteers by intravenous (IV) and oral routes demonstrated a median oral bioavailability of 21.2% [7]. CsA is highly protein bound and exhibits a large volume of distribution at steady state that ranges from 3 to 5 L/kg due to the high affinity for cyclophilins within tissues including red blood cells (RBCs) [8]. As a result of the extensive binding to RBCs, whole blood concentrations of CsA are commonly used for most pharmacokinetic (PK) studies. In addition to highly variable bioavailability, CsA also displays significant variability in clearance that spans greater than an order of magnitude (0.63–23.9 ml/min/kg) in healthy individuals [7].

Due to the poor oral bioavailability observed with these early preparations of CsA, formulations based on an oil-based microemulsion (Neoral or Gengraf) were developed in an effort to improve absorption [9]. The oral bioavailability of the microemulsion formulations was significantly improved. Bioavailability is still lower in liver transplant recipients compared to kidney transplant recipients [8]. Biliary flow and the presence of bile is a major factor affecting intestinal absorption of CsA, as illustrated by the greater than fourfold increase in bioavailability observed in liver transplant patients following T-tube clamping [10]. The improved bioavailability of CsA microemulsion is paralleled by improvements in absorption kinetics leading to a more consistent time to peak concentration and superior dose linearity with exposure. Despite these improvements in formulation, significant pharmacokinetic variability remains, with the dose-adjusted area under the concentration curve (AUC) of microemulsion-formulated cyclosporine demonstrating a greater than 20% coefficient of variation (CV) across individuals [11].

In addition to the wide variability in absorption, the variability in CsA metabolism and elimination is also clinically important. CsA is extensively metabolized to more than 25 different metabolites primarily via the cytochrome P450 3A (CYP3A) system [8,12–14]. Excretion is mostly biliary, with greater than 90% of the parent drug eliminated by this route. Renal excretion in urine accounts for only approximately 6% of drug elimination, with the vast majority excreted as CsA metabolites. As a result, renal failure has minimal effect on the clearance of CsA compared with the dramatic alterations in CsA absorption and clearance in patients with liver failure. CsA is also highly bound to cyclophilins within tissues, including RBCs. Due to the high protein binding, little CsA is also removed by hemodialysis [15]. CsA is subject to numerous drug and food interactions. Grapefruit and red wine, as well as herbal medicines such as St. John’s wort, exhibit significant interactions with CsA through their common metabolism by the CYP3A enzymatic system and membrane transport by P-glycoprotein (also known as MDR1). Commonly co-administered immunosuppressive drugs such as corticosteroids and sirolimus also show clinically relevant effects on CsA pharmacokinetics.

The relationship between CsA exposure and clinically relevant endpoints such as nephrotoxicity and organ rejection was investigated early during the use of CsA due to the highly variable pharmacokinetic behavior of the drug. In one of the earliest studies of CsA pharmacokinetics, Lindholm et al. reported on a population of 160 consecutive kidney transplant patients treated with once-daily IV or oral CsA [16]. Although transplant rejection (40%) and graft loss (23%) were significantly higher in this study compared with the incidence observed with current induction and maintenance immunosuppressive regimens, patients with higher CsA concentrations had significantly lower rates of graft rejection and higher rates of graft survival at 1 year. Subsequent studies have supported the pharmacodynamic relationship between CsA exposure and clinically relevant endpoints [17–22]. These studies have also confirmed the wide variability in pharmacokinetic behavior of this drug, particularly during the early part of the dose interval. Thus, concentration monitoring of CsA is generally considered a standard of care across transplant centers.

Although AUC0–12 h provides the best measure of drug exposure, the impracticality of making these AUC measurements, particularly in the outpatient setting, has led to the use of other surrogate measures of exposure. Because most of the variability in CsA pharmacokinetics occurs during the initial 4 h following dosing, AUC over this early post-dose period (e.g., AUC0–4 h) has been explored as a surrogate for the full-dose interval AUC0–12 h; however, even these abbreviated sampling approaches pose real challenges to collection in the clinic [22]. Pre-dose concentration (C0) represents the simplest measure of CsA exposure. Unfortunately, the correlation between C0 and AUC0–12 for CsA is relatively poor. Reported r² (r = correlation coefficient) values for the relationship between C0 and AUC0–12 h or AUC0–4 h generally fall within the 0.4–0.6 range. C0 also appears to be a poor predictor of CsA efficacy or toxicity [17–22]. In a prospective study by Grant et al. that compared the pharmacokinetics of Neoral and Sandimmune formulations of CsA, AUC0–6 h demonstrated a significant correlation with graft rejection, with patients in the lowest quartile of AUC exposure showing a more than twofold increased incidence of rejection compared to those in the highest exposure quartile. No significant relationship between C0 and efficacy or toxicity endpoints in either formulation group could be demonstrated [19].

Despite the limitations, C0 (trough or pre-dose concentration) remains a commonly used single, timed concentration for CsA monitoring. The typical target pre-dose concentrations vary significantly across transplant type, time post-transplant, and transplant center [23]. Selection of a target concentration should also take the method used for measurement into consideration. Early PK studies used immunoassays with limited specificity for CsA relative to metabolites. Differences between some analytical platforms were reported to be as large as 100% [24]. More recent versions of automated CsA immunoassays, such as the chemiluminescent microparticle immunoassay marketed by Abbott Laboratories or the Roche automated electrochemiluminescence immunoassay, appear to show improved agreement with liquid chromatography–mass spectrometry-based assays, with a mean positive bias of less than 10% for both methods [25,26].

Although no single, timed concentration is likely to provide as much information regarding CsA exposure as an AUC, 1- or 2-h post-dose concentration, close to the Cmax for microemulsion-formulated CsA, has been proposed as a significantly better surrogate for the 12-h dose interval AUC compared with C0. Correlations between these early post-dose time points and AUC0–12 are reported with r² values generally greater than 0.8 [17–21,27–29]. Based on the improved correlation with AUC, the 2-h post-dose concentration (C2) has been advocated as a single concentration monitoring alternative to C0 in several transplant settings. Knight and Morris systematically reviewed the literature for studies directly comparing C2 and C0 monitoring in both de novo and stable kidney, liver, heart, and lung transplant recipients [30]. Although most retrospective studies demonstrate a relationship between C2 and clinically relevant endpoints such as rejection and nephrotoxicity, prospective studies of the benefits of C2 monitoring on clinically relevant endpoints are limited. Of the 10 randomized, controlled studies comparing C2 to C0 monitoring, only a single study demonstrated a significant improvement in rejection and nephrotoxicity with C2 monitoring; however, this study lacks many important details, such as the C0 target range used or the fraction of patients achieving the target concentration [31]. Thus, currently, there appears to be insufficient evidence to support C2 monitoring as superior to C0 despite the improved correlation with AUC.

1.2.2 Tacrolimus

Tacrolimus, also known as FK506, was introduced into clinical practice in 1989 as an alternative to cyclosporine, and it achieved US Food and Drug Administration (FDA) approval for use in patients following liver transplantation in 1994. Tacrolimus has assumed a central role in the primary prophylaxis against organ rejection, with approval in most transplant settings. In 2012, approximately 90% of kidney and liver transplant recipients were treated with a tacrolimus-based immunosuppressive regimen [32]. Tacrolimus, like CsA, is typically used in combination with other immunosuppressive drugs.

Tacrolimus is a macrocyclic lactone (macrolide) compound that was originally isolated from Streptomyces tsukubaensis in 1984. It is very poorly soluble in water but highly soluble in alcohol. Due to the poor aqueous solubility, IV formulations such as Prograf contain tacrolimus solubilized in polyoxyl 60 hydrogenated castor oil (HCO-60) mixed with alcohol.

The mechanism of action for tacrolimus-induced immunosuppression is very similar to that for CsA. Tacrolimus demonstrates high-affinity binding to a distinct family of ubiquitously expressed peptidyl-prolyl isomerases termed FK506-binding proteins (FKBPs) or immunophilins. Although immunophilins share functional activity with the cyclophilins that bind CsA, they do not share amino acid similarity. The tacrolimus–FKBP complex that forms within the cytoplasm of T lymphocytes binds to the calcineurin complex blocking its phosphatase activity with similar inhibitory effects on NFATc translocation: T cell activation and T cell proliferation to those observed with cyclosporine (see Figure 1.1). Interestingly, tacrolimus shares structural similarity and binding to immunophilins with sirolimus and everolimus; however, these ISDs mediate their immunosuppressive effects by a very distinct mechanism from tacrolimus and CsA. As a result of the overlapping mechanism of action for tacrolimus and CsA, they share many of the same adverse effects, such as renal and neurologic toxicity.

The oral bioavailability of tacrolimus, like that of CsA, varies widely from as low as 5% to as high as 95% in some individuals [33,34]. In studies of liver transplant patients, tacrolimus exhibited a mean oral bioavailability of approximately 25%. Unlike CsA, little change in absorption was observed with T-tube closure, suggesting that bile does not play an important role in absorption [35]. Tacrolimus displays an apparent volume of distribution of 1.94 ± 0.053 L/kg in healthy individuals with extensive binding to FKBP in most tissues including RBCs. Like CsA, whole blood concentrations are the primary measurements for most PK studies.

In healthy individuals, clearance following IV administration is estimated at 0.040 ± 0.008 ml/min/kg; however, clearance varies substantially, with some studies showing as much as a 50% CV. Tacrolimus undergoes extensive metabolism by the liver and gastrointestinal cytochrome P450 (CYP) enzyme system, with less than 0.5% of the parent drug excreted unchanged in the feces and urine. Much of the variability in elimination as well as absorption may be explained by genetic differences among individuals in the cytochrome P450 enzymes and the P-glycoprotein drug transporter (see Chapter 5). At least 15 different metabolites of tacrolimus have also been described. Some of the metabolites of tacrolimus exhibit immunosuppressive activity; however, metabolites of tacrolimus with immunosuppressive activity generally represent only a small fraction of the total immunosuppressive activity of tacrolimus [36–38]. Tacrolimus pharmacokinetics is also affected by age, gender, several drug interactions, and especially liver function [39]. Significant diurnal variation has also been observed with lower clearance after the morning dose compared with the evening dose, resulting in an average 20% difference in exposure [40].

Although pharmacokinetics is variable, tacrolimus exposure appears to correlate with the important clinical endpoints of rejection and toxicity. Laskow et al. were the first to document a clear pharmacokinetic–pharmacodynamic relationship for tacrolimus in a prospective, concentration-controlled exposure escalation trial [41]. In this study, 120 kidney transplant recipients were prospectively randomized during the first 2-week period following transplant to three different levels of tacrolimus exposure with low (C0, 5–15 ng/mL), intermediate (C0, 16–25 ng/mL), or high (C0, 26–40 ng/mL) target blood trough concentrations adjusted over 42 days. Logistic regression analysis of trough concentrations in relation to both rejection and toxicity endpoints demonstrated a significant relationship, with more rejection events occurring in the low exposure group and more toxicity, including severe, life-threatening toxicity, observed in the high target exposure group. Serum creatinine and estimated glomerular filtration rate were not significantly different among the three target groups; however, the follow-up period in this study was short. Venkataramanan et al. conducted a prospective study of liver transplant recipients receiving tacrolimus with a longer 4-month follow-up period and demonstrated a strong concentration–response relationship between rejection as well as nephrotoxicity. Toxicity was observed even at the lowest exposures, with the lowest probability of toxicity (<10%) at tacrolimus trough concentrations less than 5 ng/mL rising to greater than 80% probability of nephrotoxicity at concentrations greater than 20 ng/mL [42]. These pharmacodynamics relationships were further supported by several subsequent studies that were thoroughly reviewed by Staatz and Tett [39].

Tacrolimus dosing is presumed to benefit from concentration-monitored therapy due to the narrow therapeutic index of this drug and the overall variable pharmacokinetic behavior. The package insert currently recommends concentration-monitored therapy, which is considered the standard of care in all centers using the drug. Although widely used, there are limited data from concentration-controlled studies to support its use. The study by Laskow et al. represents the only prospective trial that evaluated the ability of concentration-controlled therapy to achieve desired exposure levels [41]. Unfortunately, due to ethical concerns with study design, a control arm managed without TDM was not included for comparison to demonstrate a benefit to monitored therapy.

Monitoring of trough (pre-dose) tacrolimus concentration (C0) is by far the most commonly applied TDM approach supported by the published, pharmacodynamic relationships observed with trough measurements [41,42]. This is undoubtedly due to the ease of collecting this specimen over other timed specimens or the multiple samples required over a dose interval for AUC estimation. The strength of the correlation between C0 and exposure as assessed in a full-dose interval AUC has been the subject of some debate. The correlation between trough tacrolimus and AUC is variable, with r² values ranging from 0.97 to 0.34 in published studies [43–48]. The wide differences in the correlation between C0 and AUC across studies is unclear, but they may be related to differences in the study populations or analytical methods used for measurement. The introduction of new, extended-release formulations of tacrolimus with altered PK may necessitate a re-evaluation of monitoring methods; however, early PK data from studies of these formulations suggest that C0 and AUC correlations may be similar despite the altered PK [49,50].

Alternative single time point concentrations and limited sampling strategies (LSS) have been explored to enhance the accuracy of estimating the tacrolimus AUC while retaining a sample collection scheme that is reasonable to apply to the outpatient clinic setting. Ting et al. provides the most recent critical analysis of LSS for tacrolimus AUC estimation [51]. Of the seven published approaches available in 2006, all utilized multiple linear regression to derive relationships between either single or multiple timed tacrolimus concentrations within the initial 6-h period following tacrolimus dosing and tacrolimus AUC. Only a single published study by Dansirikul et al. used prospectively collected data and provided validation of the derived LSS equation in a separate group of patients [52]. A few additional studies have been published beyond those reviewed by Ting et al. with similar results and limitations as those of the previously published studies [53–58]. These approaches may provide some utility in the evaluation of challenging patients; however, the improvement over traditional C0 monitoring remains unknown. The transferability of LSS derived in one patient population to another, given the variability in transplant type, concomitant drug therapy, and genetic factors (e.g., CYP3A5 genetic differences among racial groups) that influence tacrolimus pharmacokinetics presents an additional major challenge to using any LSS. Caution must therefore be exercised with their use.

Notwithstanding the lack of randomized controlled trial evidence to support the use of TDM to optimize tacrolimus therapy, the substantial evidence showing a correlation between whole blood tacrolimus concentrations and toxicity has led to monitoring as standard of care. C0 monitoring remains the primary approach even in light of the known limitations; however, AUC measurements should be considered for some patients, particularly those with clinical findings of rejection or toxicity that are inconsistent with the apparent level of immunosuppression.

1.3 Antimetabolite Drug

Mycophenolic acid (MPA; formulated as the 1,4-morpholinoethyl ester of MPA prodrug (mycophenolic acid mofetil, CellCept)), an antimetabolite drug, was first FDA approved for prevention of kidney transplant rejection in 1995. It has since become the predominant antimetabolite ISD used in the transplant setting. MPA is a fungal metabolite originally described by Bartolomeo Gosio in 1893 as an antibiotic with activity toward Bacillus anthracis, making it the first antibiotic purified from a mold [59]. However, immunosuppressive activity of MPA was recognized by Planterose almost 80 years later [60]. Its first clinical use was in the treatment of psoriasis [61–63].

As an ISD, the primary mode of action of MPA is noncompetitive inhibition of the enzyme inosine 5′-monophosphate dehydrogenase (IMPDH; EC 1.1.1.205). Two isoforms of IMPDH have been identified, and both are sensitive to MPA, with the IMPDH-II isoform showing slightly more sensitivity to inhibition by MPA [64]. IMPDH plays an important role in the de novo synthesis of guanine nucleotides by catalyzing the conversion of IMP to the critical precursor xanthine 5′-monophosphate (XMP), as shown in Figure 1.2. Because lymphocytes rely heavily on de novo nucleotide synthesis for their proliferation and function [65], inhibition of IMPDH by MPA results in a significant block in cell-mediated adaptive immunity [66–68].

Figure 1.2 Schematic of guanine nucleotide synthesis showing mechanism of action for MPA.

HGPRT, hypoxanthine–guanine phosphoribosyl transferase; XMP, xanthosine-5′-monophosphate; IMP, inosine-5′-monophosphate; GMP, guanosine-5′-monophosphate; GDP, guanosine-5′-diphosphate; GTP, guanosine-5′- triphosphate; dGDP, deoxyguanosine-5′-diphosphate; dGTP, deoxyguanosine-5′-triphosphate; PRPP, phosphoribosyl pyrophosphate; AMP, adenosine-5′-monophosphate; ADP, adenosine-5′-diphosphate; ATP, adenosine-5′- triphosphate; dADP, deoxyadenosine-5′-diphosphate; dATP, deoxyadenosine-5′-triphosphate; PNP, purine nucleotide phosphorylase.

MPA is currently used clinically both as prodrug, mycophenolic acid mofetil (MMF), and an enteric-coated sodium salt (EC-mycophenolate sodium [EC-MPA], Myfortic). MMF is rapidly hydrolyzed, mostly in the upper gastrointestinal tract, to produce MPA and hydroxyethyl morpholine, an inactive metabolite that is rapidly metabolized and excreted in urine [69]. Peak concentrations are typically achieved within 1 or 2 h following oral dosing of MMF and 1.5–2.75 h for the sodium salt with a lag in absorption of 0.25 to 1.25 h, most likely due to the enteric coating. The bioavailability of MPA is relatively high, with MMF exhibiting oral bioavailability of 80–90% and comparable bioavailability for EC-MPA [70,71]. Once absorbed, MPA exhibits complex pharmacokinetics. Distribution within blood is primarily within the plasma compartment, with 97–99% bound to albumin [72,73]. The 12-h dose-interval MPA plasma concentration versus time profile is characterized by rapid absorption, reaching maximal concentration within 1 h followed by rapid distribution to tissues and falling plasma concentration, reaching a plateau within 3 or 4 h. A second peak concentration is often present within the latter half of the dosing interval. This latter peak is variable in both time and intensity, with complete absence in some patients. MPA is metabolized mostly via the UDP glucuronyl transferase (UGT) system within the liver [74]. The primary inactive metabolite is the phenolic glucuronide MPAG, which is transported from liver cells into bile most likely via the ATP binding cassette transporter, MDR1-related protein 2 (Mrp2) [75–78]. Following biliary excretion, MPAG can be converted back into MPA by glucuronidase produced presumably by intestinal bacteria, resulting in reabsorption of MPA. This enterohepatic cycle (EHC) of excretion followed by reabsorption is responsible for the secondary peak in concentration often observed with MPA, and EHC is estimated to contribute 10–60% of the overall AUC for MPA [79].

Despite the complex PK of MPA, exposure as assessed by AUC displays a relatively linear relationship with dose [71,80]. Clearance of MPA, however, varies considerably across individuals as well as within an individual [79,81]. Studies in healthy individuals show interindividual variability of 25–30% for AUC following a single 1-g dose of MMF. It is also clear from studies in transplant recipients that MPA PK changes over time following transplantation, with lower maximal concentration (Cmax) and lower AUC during the early post-transplant period followed by a gradual rise during the first 3 or 4 months [79]. A lag in absorption is also noted in the early post-transplant where Cmax is often not reached until 4 or 5 h compared with the typical achievement of Cmax at approximately 1 h in the stable post-transplant period. Although drug absorption may contribute to some of the changes in PK, these changes are also observed with MPA administered intravenously. It is therefore likely that the change in PK is related, in part, to restrictive clearance mechanisms that likely result from the high protein binding of MPA [82]. The free (non-protein-bound) fraction of MPA, which is the fraction metabolized, is likely higher in the immediate post-transplant period due to lower albumin concentration or altered protein binding associated with displacement by drugs or endogenous compounds present in disease states such as renal and hepatic failure [72]. As albumin concentration normalizes in the post-transplant period, MPA clearance declines, presumably leading to the observed increase in AUC. These changes in MPA PK over time have been observed in most transplant settings, with the most pronounced change observed in the liver transplant