Therapeutic Drug Monitoring of Mycophenolic Acid

Mycophenolic acid (MPA)-based therapies are widely used in combination with calcineurin inhibitors as maintenance immunosuppression for kidney transplant recipients (1). The two MPA therapies used in clinical transplantation are mycophenolate mofetil (MMF [brand name CellCept, Roche Pharmaceuticals, Nutley, NJ]) and mycophenolate sodium (MPS [brand name Myfortic, Norvartis Pharmaceuticals, Nutley, NJ]). MMF has been used for more than a decade and is a prodrug of MPA. The standard dosage of MMF in combination with cyclosporine (CsA) is 1 g given twice daily, although the dosage may be somewhat lower when co-administered with tacrolimus. MPS is an enteric coated form of MPA that was more recently introduced into the clinical arena. A dosage of 720 mg of MPS provides bioequivalence to a dosage of 1000 mg of MMF in kidney transplant patients (2). Immunosuppression afforded by MPA is achieved via reversible and uncompetitive inhibition of inosine monophosphate dehydrogenase (IMPDH), resulting in inhibition of guanine nucleotide biosynthesis (3,4). This consequently leads to suppression of both new DNA synthesis and other pathways that depend on a continuous supply of guanine nucleotide pool, such as T cell surface antigens and other glycosylated membrane proteins (4). Although there has been increased interest to incorporate MPA therapeutic drug monitoring into routine clinical practice (5–9), this has not yet become widespread in the United States for several possible reasons, including (1) lack of availability of US Food and Drug Administration–approved automated simple assays, (2) attainment of low rejection rates using empiric dosing of MMF in many maintenance immunosuppression regimens, (3) the complex pharmacokinetics (PK) of MPA, and (4) absence of overt organ toxicity. Here we discuss and review the pertinent information and study data regarding (1) our current understanding of MPA PK and major factors that can influence MPA clearance, (2) the performance characteristics of MPA analysis methods and prospects for more widespread availability of simple automated methods, (3) results and limitations of clinical outcome studies in renal transplant patients, (4) the status of ongoing prospective trials to evaluate various concentration-control approaches for dosing MMF in renal transplant patients under a variety of contemporary maintenance immunosuppression protocols, and (5) suggested rationale and guidelines for monitoring MPA. MPA PK MMF, the morpholinoethylester of MPA 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 (10). Within blood, MPA is largely distributed in plasma, where it is avidly bound to human serum albumin (10). The 12-h dose-interval MPA plasma concentration versus time profile, AUC0–12 h, is characterized by rapid absorption that reaches maximal concentration within 1 h followed by rapid distribution to tissues and falling plasma concentration, reaching a plateau within 3 to 4 h (Figures 1 and 2). A small fraction of MPA circulates in a free, unbound form, but most is bound to serum albumin. MPA is metabolized mostly via the uridine diphosphate glucuronyltransferase (UGT) system. The primary inactive metabolite is the phenolic glucuronide mycophenolic acid glucuronide (MPAG), which is transported from liver cells into bile most likely via the ATP-binding cassette transporter MDR1-related protein 2 (Mrp2) (11). Biliary MPAG then enters the gastrointestinal tract, where, under the catalytic action of glucuronidase that is shed from the intestinal flora, it is hydrolyzed back to MPA, which is then recycled into the bloodstream, the so-called enterohepatic circulation (EHC) pathway (Figure 1). The EHC pathway results in secondary peaks that typically occur between 4 and 8 h after ingestion of a dose of MMF. Besides MPAG, MPA may also be metabolized to the minor acylglucuronide, AcMPAG. Of note, the enteric coating of MPS results in maximal MPA concentrations that are achieved later in comparison with MMF. Several factors can significantly alter the elimination of MPA and can therefore influence MPA plasma concentrations. Such factors include one or more of the following situations: Significant renal or liver disease, drug–drug interactions, genetics, diarrhea, and patient noncompliance. Renal and Liver Dysfunction Kidney or liver disease can influence clearance by increasing the free fraction of MPA. Thus in kidney recipients who receive concomitant CsA and have poor renal function in the first days after transplant surgery, there can be a two- to three-fold increase in the free fraction (12). It is important to appreciate that in this situation, free MPA concentrations are usually stable, not increased, and comparable to the average values obtained in patients with good early renal function, at least in patients who are on CsA co-therapy and on a 1-g twice daily dose of MMF (12). The increased MPA free fraction is believed to be caused by increased concentrations of MPAG that competitively displace MPA from albumin binding sites, alterations of albumin produced by uremia, that reduces the binding avidity for MPA and by low albumin concentration that reduces binding capacity for MPA (10). As predicted by the equation clearance = free fraction × intrinsic clearance, for restrictively cleared drugs, the increased MPA free fraction results in enhanced MPA clearance, leading to reduced total MPA AUC values but stable and not increased free MPA concentrations. As renal function improves to baseline in these patients, the average free fraction approaches values obtained during stable graft function, a pattern that is paralleled by clearance and total MPA AUC values normalizing as well but free MPA concentrations remaining stable. This situation is comparable to that obtained for restrictively cleared drugs, such as phenytoin, in which this drug's free fraction can increase by two-fold or more in patients with uremia but free phenytoin concentrations remain stable as long as the dosage of the drug remains the same. Patients with good early kidney transplant function have near-normal MPA free fraction values and MPA clearance from the early posttransplantation period and consequently MPA AUC levels that typically resemble those obtained when graft stabilization has occurred (12). Similar perturbations can be observed in the early posttransplantation period in liver transplant patients as a result of one or more disturbances of MPA binding caused by renal dysfunction and elevated MPAG concentrations, hyperbilirubinemia that competitively displaces MPA from albumin binding sites, and hypoalbuminemia. A more extreme picture emerges in transplant patients with chronic kidney disease, in whom greater increases in free fraction and free MPA concentrations can occur as a result of even greater increased MPA free fraction (13,14) and abrogation of the restrictive clearance mechanism. Possible direct influences of renal and hepatic disease on UGT activity or on the EHC pathway have not been established, although there are renal failure animal model data showing inhibition of UGT activity and consequent reduced rate of MPA glucuronidation by the isolated perfused liver prepared from the diseased animal (15). Thus, in the special circumstance of chronic renal failure, the possible reduction of intrinsic clearance combined with the observed elevation of MPA free fraction could explain the observed higher free MPA but normal total MPA concentrations (13,14). In patients with chronic renal failure, measurement of both total and free MPA is warranted for accurate assessment of the possibility of overexposure to very high concentrations of free MPA that can predispose the patient to increased risk for low white blood cell counts and infection (13,14). Drug–Drug Interactions Drugs may interfere with MPA clearance by different mechanisms. Agents such as corticosteroids and rifampin (16), when introduced into the regimen, stimulate the biosynthesis of UGT, thereby increasing MPA clearance with consequent decreased MPA concentration. Discontinuation of such inducing drugs reduces MPA clearance and consequently increases MPA concentrations. Such effects result from changes in the intrinsic UGT enzymatic clearance of MPA, and the degree of effect may vary depending on genetic factors. Another mechanism by which drug–drug interactions may affect MPA exposure is by suppression of the EHC pathway by certain antibiotics secondary to loss of glucuronidase activity normally shed by gastrointestinal tract bacteria (17,18). The prime example of a drug that suppresses the EHC pathway is CsA, which, when co-administered with MPA therapy, results in a significant reduction of dosage-adjusted MPA concentrations. This can be seen clearly in Figure 2, which shows the much lower concentrations from 3 through 12 h for renal transplant patients who receive CsA and MMF. For renal transplant patients who receive tacrolimus and MMF, the presence of a greater contribution of the EHC to the dose interval MPA AUC also gives rise to greater variability in the postabsorption MPA concentrations (see the sample times from 3 h on in Figure 2, showing the consistently higher and more variable MPA concentrations in the tacrolimus patients). This explains the observation in many studies of the approximately 50% lower MPA trough concentrations, adjusted for MMF dosage, in patients on concomitant CsA compared with patients on concomitant tacrolimus. The proposed mechanism for this interaction supported by animal model data is the inhibition of transport of MPAG by CsA at the Mrp2 transporter site in the hepatocyte canalicular membrane (19–21). This is also the likely explanation for the differential MPA drug dosage requirement that is typically observed in patients who are on tacrolimus versus CsA. Genetics The interindividual PK variability observed with MPA use is likely explained in part by genetic variation. MPA is metabolized both in the liver and in the intestine by at least three enzymes of the UGT system: UGT1A8, UGT1A9, and UGT1A10. In addition, there is evidence that a fourth UGT, UGT2B7, is responsible for production of the acylglucuronide metabolite that might contribute to some of the gastrointestinal toxicity that is associated with MPA (22). Genetic variability within these important biotransformation pathways has been described, and this is predicted to have effects on the PK of MPA. This prediction is supported by a study in renal transplant patients that showed a 50% decrease in MPA trough concentrations in patients with promoter variants within the UGT1A9 gene (23). Another study conducted in healthy volunteers further supports the impact of the UGT1A9 −275/−2152 promoter polymorphisms on MPA exposure and suggests a further role for UGT1A8 and UGT2B7 polymorphisms (24). In addition to variation in metabolism, transport of MPA and its metabolites is potentially affected by genetic variation. The main MPA glucuronide metabolite, MPAG, is believed to be transported into bile via Mrp2, the same transporter responsible for bilirubin export from hepatocytes into bile. Mrp2 is the transporter site at which the inhibition of MPAG export into bile by CsA but not tacrolimus is believed to occur (20). Polymorphisms in this and other that might affect and EHC of MPA have been (20). in renal transplant patients have effects of Mrp2 polymorphisms on MPA AUC values PK genetic factors may also contribute to variability of MPA. Polymorphisms within the and that decrease the of the increase or alter these such that become or more to inhibition by MPA, may and affect both the and the toxicity of this the in MPA PK and in combination with serum concentration monitoring may dosing of MMF. The information to is to on the of genetic information for this studies at understanding the impact of genetic variation on MPA PK and are clearly under and or other disease that the gastrointestinal tract has the to result in absorption of MPA and consequent MPA plasma concentrations. is that can an in MPA exposure in a patient in routine clinical Of note, two to compared MPA PK in stable of and white study could a in PK between the two This as a significant to MPA clearance and also as significant to MPA PK variability in two PK Several have PK for MPA in renal transplant patients In of these a PK model that the of concentration versus time for renal transplant patients was first this and using the PK model and the widely used effects the MPA concentration data to the model and the most likely factors and variability in MPA concentrations The model that of this to rapid increase in MPA concentration during the absorption that is by a followed by a distribution and an elimination Thus, for studies in which MPA is and CsA is the calcineurin a model with absorption provides an PK In these one is the blood and and the other is the such as and with which MPA in the rapidly The gastrointestinal tract is the site from which drug enters the (Figure 1). In one approximately of the patients CsA and the other tacrolimus as concomitant calcineurin MPAG and MPA concentration in the and this addition of that MPAG, produced in liver important transport into the intestinal tract and to MPA followed by into the (Figure 1). This model the MPA concentration data for the tacrolimus patients well but not of the data for the patients who concomitant CsA. This MPAG back to MPA is with to be in tacrolimus patients but in CsA the use of PK and including the early posttransplantation period 3 to major factors that significantly with MPA clearance include clearance, plasma albumin concentration, and CsA daily dosage In a of data in which MPA PK values from 1 after transplantation to it was that kidney function, plasma albumin concentration, concentration, and CsA concentration could explain of the and of the variability in MPA exposure In a to this the further showed that for the first a decreased MPA clearance from to with kidney function, for this of patients, from to albumin concentration from to concentration from to 12 and CsA trough concentration from to In a study in patients with stable renal function after the was with clearance and of this in the model in a reduction of the variability of clearance from to a very such as kidney function, and liver function not with MPA clearance The of effect of renal or liver function on MPA clearance in this study is an result renal function and liver function stable in this patient Possible genetic factors that may significantly influence MPA clearance, not in of these PK studies that incorporate genetic factors be to more influence on MPA clearance in various transplant of the studies to has free MPA concentration and this is one of the PK that to be in studies that model the increase in MPA concentration with time after It is not that this be a significant variable during stable graft function when the free fraction of MPA is stable. for of MPA AUC The dose interval MPA AUC0–12 h is as the most PK of risk for rejection but is to in routine clinical such as the trough concentration or not well with the MPA in the early posttransplantation period (10). Several have that a good of MPA AUC0–12 h can be obtained by during 2 to 3 h, and with performance have been for and renal transplant patients and for that include MMF CsA, or sample for and and for that include CsA or tacrolimus are in use in prospective studies of MPA therapeutic drug monitoring A recently study of the performance of a for patients showing and values comparable to the values obtained using the AUC0–12 h Although the of the dose interval MPA AUC 2 to 3 h has the important of a it is important to that to a patient and drug (10). for MPA The methods used to total and free MPA concentration include with as well as that can be automated has been and is used in a of for total MPA concentration. The performance characteristics of these methods have been compared and in in (10). with or with is used as a in clinical trials in routine clinical practice in some clinical and have been for the and A new automated has been with as the The data for this new in comparison with and methods have for total and free MPA concentration values using from renal and transplant patients This new has recently been by the US Food and Drug for use in patient the lack of availability of a simple MPA that can be automated is one of the limitations in the of monitoring in many this may an for more widespread use of MPA a the measurement of the effect of a drug or its effect may more with clinical outcome than a PK This is the for the and of that activity in In this of the by factors including genetics, is for use in therapeutic monitoring of MPA PK has a more in therapeutic trials as well as in clinical practice as a result of the in the A in blood cells has been the by studies using this are the (1) the significant in baseline activity for the outcome observed in (2) low activity a higher risk for (3) high activity a higher risk for rejection studies on patient be to evaluate these between MPA and The addition of MMF to immunosuppression has been associated with decreased rates of rejection and decreased graft the of therapies into most contemporary maintenance in the decade has to an of of the other to calcineurin In the of immunosuppression the of MPA monitoring is As is for other maintenance MPA PK are widely with dose-interval AUC adjusted for MMF dosage, to in organ transplant patients concentration-control studies renal transplant patients who concomitant CsA, showed that when concentration of the MPA dose interval AUC0–12 h is in a and and AUC values of at least to are achieved in the early posttransplantation the risk for rejection can be major one the versus study and the other the are ongoing to the clinical of of MPA AUC and of trough concentrations compared with empiric The of both of these more the routine in that guidelines for dosage but to to The on the results of these two trials are The addition of MMF to immunosuppression has been associated with decreased rates of rejection and decreased graft the time of of MMF into clinical empiric dose the practice at most and the has been in a review of contemporary therapy, that is in combination with many other simple to use and free from organ toxicity and the and of the variable MPA PK and of the between MPA concentration and risk for rejection and effects and the for more therapy, in patients, has to further of these The current practice of dosing MMF to toxicity of the gastrointestinal tract or with dosage when toxicity is has been characterized by one as to of but to for or dosage The routine availability of MPA measurement in the most can an measurement of MPA concentration. The is of the MPA concentration values of trough In various significant have been observed between MPA concentrations and (1) the inhibition of activity in obtained from transplant patients during the 12-h dose (2) inhibition of new DNA synthesis in (3) rejection in a and (4) risk for rejection in renal transplant patients (10). The study of the between MPA AUC as well as trough concentration values and the risk for early rejection was first for a of renal transplant patients This was followed by a in which MMF dosing was on a dosing and the dose interval MPA AUC to one of three or This study the that MPA AUC predicted the risk for rejection in the early posttransplantation period for the widely used CsA MMF in use at that time in renal transplant patients important results from this that (1) reaching or a MPA AUC achieved a significantly lower rate of rejection as compared with the rate at the lower MPA AUC and (2) the risk for rejection decreased as the MPA AUC increased In a of a prospective concentration-control versus study that was conducted in and used a AUC of with a on three in the concentration-control this was A of the study be for a more analysis of this In the MPA trough values also with rejection risk but significantly as a result of the greater variability of trough concentration values In three review studies of MPA PK data in or renal transplant patients who MMF CsA the early MPA AUC predicted the risk for rejection 2). In two studies that renal transplant patients who MMF CsA MPA AUC not with the rate of but this may have been to the small of rejection in these of patients and of In addition to the possibility that the low of rejection with MPA concentrations, a possible for this is that one AUC on posttransplantation was used for prediction of early rejection This time in to the time of rejection may not have been and this of when to sample and which of sample or further studies have been conducted on the between MPA concentration and risk for early rejection for the combination of MMF tacrolimus studies the between rejection risk and MPA AUC or trough concentration significant in the MPA AUC or trough concentration observed between patients with rejection and those who not rejection when concentrations of these patients into to AUC0–12 h values values of for MPA and for the of rejection higher = in the with the low MPA and low tacrolimus AUC values compared with the with the high MPA and high tacrolimus AUC values The risk for rejection was for the patients who low values for one of the two drugs A significant was for the for MPA trough concentration and risk for rejection within the first after transplantation The and prospective therapeutic drug monitoring trials more data on which to the of and sample for renal transplant patients who receive concomitant tacrolimus The data of the upper of the therapeutic are for several reasons, including the variable of plasma in to time of effects and the variable of patient factors from study to study variable risk for effects For renal transplant patients who are on CsA there is further reduction in rejection at AUC values of higher exposure on the of this is also the possibility that other PK such as free MPA concentration and the concentration of the acylglucuronide, may be of effects trials to evaluate various concentration-control approaches for dosing MMF in renal transplant patients under a variety of immunosuppression and including measurement of free MPA and the acylglucuronide more data pertinent MPA PK to risk for rejection and for Drug have suggested therapeutic for MPA AUC and trough concentrations when using MMF in combination with CsA or tacrolimus. combined with CsA, the are 1 to and to for trough concentrations and For the combination with the of to and to for trough and AUC have been suggested ongoing concentration-control versus trials in renal transplant patients the for assessment of these in the of contemporary A proposed for assessment of MPA exposure is in that the ongoing trials more data on which to the of sample and the analysis of MPA therapeutic has and from Roche and from has and from Roche and for metabolism, enterohepatic and elimination of mycophenolic acid The gastrointestinal tract the and upper small is the site where mycophenolate mofetil is hydrolyzed to produce MPA, the which is into the circulation with the circulation the the MPA the into and including and in the The glucuronide metabolites including the primary inactive mycophenolic acid glucuronide and the are produced by uridine diphosphate (UGT) in tract and The glucuronide metabolites produced by the liver are likely via the MDR1-related protein 2 (Mrp2) into bile. into MPAG is hydrolyzed back to MPA via shed by tract MPA in the tract the thereby the EHC MPAG also in the circulation after its production from MPA in various of MPAG from the circulation via renal The effect of cyclosporine (CsA) on MPAG transport into bile is by the MPA plasma concentration values versus time during a 12-h dose The are the values for average MPA concentration and the values are the of variation from the × of the and the average MPA concentrations at time and MPA concentration values for or renal transplant patients who receive as concomitant immunosuppression CsA or tacrolimus The are in the or for the CsA or the patients, For both the CsA and patient the values during the first 2 h are comparable but are consistently greater in the versus the CsA PK studies in renal transplant between MMF PK and clinical in renal factors that can alter MPA for MPA monitoring in renal transplant patients