September 2008: Volume 34, Number 9
Monitoring MPA in Solid-Organ Transplant Recipients
Does TDM Reduce Organ Rejection and Adverse Events?
By John R. Bell
When a patient receives a solid-organ transplant, the clinician can choose from two immunosuppressive calcineurin inhibitors to prevent organ rejection. Since the 1990s, most have chosen to add mycophenolate mofetil (MMF) or mycophenolate sodium to the therapeutic regimen, both of which are converted to mycophenolic acid (MPA). Although the drug has been associated with lower rejection rates in several trials, it also is associated with an increased incidence of nausea, vomiting, diarrhea, and other symptoms. For patients who experience GI side effects, many clinicians reduce the dosage until the effects subside. But some laboratorians believe adjusting the dosage based on the patient’s level of monitored serum MPA is a better way to avoid toxicity while maintaining an immunosuppressive dose.
“There is a growing group of laboratorians, as well as clinicians, who are in favor of monitoring MPA,” said Susan Maynard, PhD, director of chemistry and toxicology at Carolinas Medical Center in Charlotte, N.C. “If it requires a certain amount of MPA to provide the immunosuppressant effect, and if different people have different pharmacokinetic parameters, it makes sense to know what their levels are,” said Maynard.
Now, a new report, titled Utility of Monitoring Mycophenolic Acid in Solid Organ Transplant Patients, conducted by researchers from McMaster University, Hamilton, Ont., for the Agency for Healthcare Research and Quality (AHRQ), has systematically examined the body of evidence from published studies in hopes of clarifying the benefit to patients. The report called the evidence so far inconclusive and stated that research into monitoring of MPA is still “in its infancy.” But despite the report’s conclusions, some TDM experts say it’s worthwhile.
The Challenge of Trough Levels
One difficulty in monitoring MPA lies in acquiring a value that accurately reflects true serum levels over time, said William M. Bennett, MD, medical director of renal transplantation at Legacy Good Samaritan Hospital, Portland, Ore. “Technically doing a blood test for MPA is certainly possible,” he said. “The issue is, what does it mean? The drug has very complex pharmacokinetics, and it’s difficult to foresee a practical way to do monitoring if you just do a trough blood level, the way most other immunosuppressants are monitored.”
Alexander Vinks, PharmD, PhD, director of the division of clinical pharmacology Cincinnati Children’s Hospital Medical Center and professor of pediatrics and pharmacology at the University of Cincinnati School of Medicine, believes that monitoring makes sense—but he agrees trough levels are often of little value. “The goal is to optimize exposure response,” he said. “To predict exposure for a drug like MPA, you cannot rely on trough monitoring only—because the drug has enterohepatic recycling. And that means that the trough cannot predict the hills that were there before. It’s like trying to stand in a valley and predict how high the mountain was before you came to the valley, but there are two mountains, and you may see only one.”
Others, however, say trough levels have proved to be useful for the majority of patients after graft stabilization, when receiving tacrolimus. Leslie M. Shaw, PhD, noted that trough levels of MPA do correspond with AUC levels in patients treated with MMF and tacrolimus, as shown by recent findings from the OptiCept trial, which investigated dosing of MMF in renal patients. The final results of the trial were presented at the International Congress of the Transplant Society in Sydney in August. Two of its major conclusions are that greater MPA exposure is highly correlated with reduced risk of acute rejection and monitored patients with reduced tacrolimus dosing had reduced treatment failure rates, fewer rejections, and significantly lower withdrawal rates. This is important, because tacrolimus is now used in 80% of transplant patients, noted Shaw, a co-investigator on the OptiCept trial and a professor of pathology and laboratory medicine at the University of Pennsylvania.
The OptiCept trial’s earlier results showed no statistically significant lower rate of rejection for patients whose MPA levels were monitored. This was perhaps one reason for the inconclusive nature of the AHRQ report and Bennett’s reservations about monitoring. “Adjustment based on trough blood level probably leads to over- and underdosing,” Bennett said. “It just hasn’t seemed to correlate with drug exposure or with clinical events. There is tremendous discrepancy in individual absorption.”
Yet the OptiCept trial was designed simply to show that monitoring was not inferior to symptom-based dose adjustment, noted Shaw. “You can safely reduce the calcineurin inhibitor, using MPA monitoring, and you don’t see increases in bad outcomes.” He also pointed to a 1998 prospective double-blind, concentration-control study in renal transplant patients from Michael D. Hale, MD, and colleagues, known as RCCT, which Shaw said “showed unequivocally that maintenance of patients in a target AUC range of 30–60 mg/hr/L achieved a substantial statistically significant reduction in early acute rejection rates,” compared with renal transplants maintained at lower targeted AUC values (Clin Pharmacol Ther 1998;64:672-83; Transplantation 1999;68:261-66). “Trough concentrations also were significantly correlated with acute rejection but were noisier,” he said.
The European Practice
Maynard, who served as a peer reviewer for the AHRQ report, also believes monitoring makes sense. Yet her institution, like most in the United States, does not monitor MPA—despite having the capability to do so via HPLC/mass spectrometry. One reason MPA monitoring is not widespread, she said, is the difficulty in measuring MPA without such highly specialized lab instruments. “Tacrolimus, sirolimus, and cyclosporine may be measured by HPLC/ mass spectrometry, and immunoassay kits are also marketed by various diagnostic manufacturers. So even any small hospital in the U.S. has the capability to monitor these other immunosuppressants via immunoassay,” she explained. “However, mycophenolic acid testing is available only on the Roche chemistry analyzer, and unless the lab has that system, the only other option is HPLC/ mass spectrometry, which is not widely used.”
That may be one reason MPA monitoring is more widespread in Europe, Maynard said. “The last time I looked, there were only 37 institutions in the United States reporting mycophenolic acid proficiency testing through the College of American Pathology.” This contrasts with the far greater prevalence in Europe. “That might be due to more centralized TDM analysis in Europe, the fact that MPA has been available in Europe longer, or greater availability of pharmacokinetically assisted dosing in Europe.”
Teun van Gelder, MD, PhD, an internist and clinical pharmacologist at Erasmus University, Rotterdam, the Netherlands, has been an author or coauthor of numerous studies of MPA monitoring, including the RCCT study. He also said regulatory differences in Europe versus the US may contribute to the disparity. “It may have to do with the fact that the enzyme multiplied [EMIT] immunoassay has been on the market for many years in Europe, and not in the US,” he said. “Apart from availability of an easy assay, the effort DadeBehring [now Siemens] has done to teach doctors on the added value of TDM for MPA may have contributed to the implementation in Europe.”
Another native of the Netherlands, Vinks, suggested reimbursement and infrastructure as factors in the discrepancy. In the United States, “pathology is doing the tests, but they’re not incentivized to do interpretation, because they’re being reimbursed very little. Plus pathologists are not trained in doing this.” Vinks formerly ran a centralized therapeutic drug monitoring toxicology service for the city of The Hague, comprising five hospitals, 18 nursing homes, one children’s hospital, and two psychiatric clinics. In Europe “the clinical pharmacy people have their own labs, so they run the TDM. And they see it as part of their job not only to provide a number but also provide a decent interpretation.” That interpretation is important, he emphasized. “A physician cannot do anything with [just] a number. The interpretive step and bringing it back to the physician in a format that he or she can really understand and use is a whole different type of great opportunity—with some challenges.”
Others acknowledged that reimbursement is lacking in the United States but noted that the overall cost for TDM is low for this drug. “Medicare only covers the immunosuppressant drug costs for disabled patients—usually kidney patients, due to disability qualification—or those qualified for Medicare by age. And then, only for the first 3 years post transplant,” Maynard said. However, “the drug is not as expensive [to monitor] as the other drugs.” Shaw shares this sentiment. “The cost to monitor this medicine is relatively small when you compare it to the cost of rehospitalization or extra visits to the hospital,” he said. “It’s dollars of cost; it’s not hundreds or thousands of dollars of cost.”
The real question, however, is whether monitoring has an impact on organ rejection rates. This is a difficult question, because the current cocktail of antirejection medications is so effective, noted Shaw, who also served as a technical expert for the AHRQ report. “We now have the lowest rates of acute rejection in the history of the field. The challenge for immunosuppressant drug monitoring is all the greater, since there still is no reliable test of immune system suppression, a pharmacodynamic type of test,” he said. “The need for periodically checking MPA concentration exposure, in my view, is to know if your patient is getting an adequate exposure to this immunosuppressive drug,” he explained. “This is relevant because this drug, like the others, has wide pharmacokinetic variability. You don’t know how much it will vary from month to month within a given patient until you track it.”
Clinicians at Shaw’s center order MPA concentration monitoring in renal and heart transplant patients and some other transplant patients, and his laboratory provides the service for several other centers. “Our MPA therapeutic drug monitoring database has accumulated MPA concentration data over time for each of those patients, such that we have a good handle on the within-patient variability, and it ranges anywhere from a coefficient of variation of about 10% up to 75%. One important goal of TDM is to reduce that variability. In addition to tracking MPA concentrations for each patient, our database has at least 20 demographic and clinical characteristics for each monitored patient. This permits ready review of the individual patient’s history and permits reporting of individualized interpretation of the MPA concentration data.”
He noted that monitoring MPA is different in several ways from the monitoring of the calcineurin inhibitors or rapamycin. “This drug does not exert organ toxicity as [strongly as] the calcineurin inhibitors can. So the urgency for monitoring is less for MPA than for those drugs,” he explained. “I think the issue is that in order for anyone who is not familiar with it to understand and use MPA pharmacokinetic data in transplant patients, they’re going to have to study it in some detail, because this drug has lots of sources of variability,” Shaw added. “And they include individual variability of clearance. That depends on disease state, patient compliance, patient age—pediatric versus adult—and genetic factors. And drug-drug interactions are a very important consideration, too. I think the variability in the mind of the physician is that since the drug has a lower toxicity profile compared with the calcineurin inhibitors, there’s less worry about intoxicating the patient. The urgency is not the same.” Another source of variability is changing pharmacokinetics relating to time post transplant. “For some individuals, the MPA concentration exposure per unit of MMF dose increases with time—but not for all patients,” Shaw said.
A recent review, published after the AHRQ report was released, found that the AUC level of total MPA was associated with the risk of acute rejection but not with toxicity; it concluded that because most studies have been retrospective and/or based on monitoring trough levels in patients on cyclosporine, the evidence in favor of monitoring is “weak.” However, the authors did note that monitoring multiple measures immediately post transplant for an AUC may have benefit (Transplantation 2008;85:1675-85).
Finding the Right Sampling Model
What options exist beyond trough levels and AUC curves to monitor MPA? The answer may be Bayesian type estimation. Vinks explained: “You have a computer program, and you enter all the patient information. Then you have a pharmacokinetic model, which is basically the summary of everything that you have learned from all the patients that you have treated to date.” The transplant physician can then enter the current patient’s clinical parameters and receive an evidence-based estimate of what the MPA levels will be at different time points, he said. One publicly available example is based in France, at the National Institute of Health and Medical Research (INSERM) (See Sidebar).
A Web Resource for Monitoring Immunosuppressants
A useful resource for labs considering an MPA monitoring program has been developed by a group at the University Hospital Center Limoges, France.
For MPA, the group uses a three-sample schedule, a graph of which is available on the website, as well as algorithms for all other immuno-suppressants.
“The algorithm can come up with a very good estimate of what the profile looks like for a given patient at that point,” said Alexander Vinks, PharmD, PhD. He said that this group has tested the algorithm versus the current standard of care. “They found less rejection in a period of 1 year.”
The information can be found online.
“If you do it intelligently, monitoring is always better than not monitoring,” said Vinks. However, “most studies haven’t done it right, so they could not find any difference between an adequately monitored group and an unmonitored group.” With AUC testing, two to four samples are usually collected, at between 0 and 4 or 0 and 6 hours, he said, because most of the action happens early. “Most people don’t use any prior information. They just take the numbers as they come and use an algorithm that they developed from their own data or from the data of others.”
Shaw noted that the three randomized, controlled trials of TDM for MPA have enrolled approximately 1,800 patients and that the aggregate of all three study populations could yield an analysis that is more conclusive as to the benefit of monitoring. There have been preliminary discussions among the researchers about this possibility, he said.
Eternal Vigilance—the Price of Lower Toxicity?
Despite support for monitoring among laboratorians, the AHRQ report didn’t go that far.
“It’s very difficult for the scientific community to do these prospective trials, because they’re costly, and nobody is investing in this,” Vinks explained. “So what you see is a lot of underpowered studies that then report it doesn’t work. And so you get this continuous controversy.”
Shaw noted that the randomized controlled trials of TDM for MPA—which total four if the 1998 study from Hale, van Gelder, and colleagues is included—make up a larger evidence base for monitoring of MPA than for any other drug. He believes the preponderance of the evidence is in favor of monitoring. “I think that when you consider whether to monitor, you can simply say, ‘Do I want to have an objective assessment of how much drug is in my patient’s body, versus not knowing that?’ So when you ask it that way, I think it’s pretty logical to conclude that it makes perfect sense that you would want to know and make dosing decisions from that information, rather than simply from an empirical dosing schedule.”
Disclosures: Teun van Gelder, MD, PhD, has received consulting fees and grant support from Hoffman-La Roche, Wyeth, and Siemens. Leslie M. Shaw, PhD, has received funding from Hoffman-LaRoche and Novartis.
An Expert Access program on this topic with Susan Maynard, PhD, is available on the AACC website. Search “Events” to locate it.