We appreciate the comments by Kuypers et al. regarding our study on the association of maintenance immunosuppressive agents and BK nephropathy (BKN) (1). We are in an agreement with the authors that mycophenolate mofetil (MMF) dose is not a reliable surrogate for drug exposure, and multiple factors, such as renal dysfunction, serum albumin level, and concomitant medications, could significantly affect the concentration of its metabolites (2–6). However, the clinical value of mycophenolic acid (MPA) therapeutic drug monitoring remains a controversial and supportive evidence regarding its added benefit is conflicting (7). Although, many retrospective studies found an association between the risk of acute rejection and total MPA area under the curve (AUC0–12 h) (7–9), the relationship between adverse effects of MMF, such as anemia, leukopenia, gastrointestinal symptoms, and infection, and the AUC0–12 h is less well established (7, 9, 10). Some studies have found better correlation between the incidence of infection or leucopenia with free rather than total MPA level (2–4). Others have suggested that adverse events are better predicted by MMF dose, rather than blood concentration (8, 11). We agree with the authors that the wide interindividual and intraindividual variability in MMF pharmacokinetics makes its dose a less than optimal indicator of its effect at the cellular level. For the same reason, a single MPA trough level suffers from similar shortcomings and is the reason that there is no consensus on its use for drug monitoring or its optimal target level. Therefore, acknowledging the pitfall of using the drug dose, we are not sure that MPA trough level is a better surrogate for drug exposure and effect. Moreover, the use of complete or abbreviated AUC for comparing the drug exposure between the cases and controls in this setting is not practically feasible, because a single testing after the diagnosis may not reflect the exposure over time, and retrospective AUC measurements at different time points would not be possible. To better study this matter, a prospective cohort study would be a much better design, but the relatively low incidence of the disease would require enrolment of a large cohort, which would make it an expensive study and impair its feasibility. Published guidelines advocate reduction in immunosuppression as the first step after a diagnosis of BKN (12). However, no specific approach has been established regarding whether calcineurin inhibitors (CNI) or antiproliferative agents should be reduced first. Clinical trials to validate different strategies are lacking and published reports on different immunosuppression reduction protocols contain mixed results (13). Therefore, the question on how to manage immunosuppressive agents in this patient population remains uncertain. In our study, we suggested reduction in tacrolimus level as the first step, while maintaining the MMF dose, based on our observation of a significant association between BKN and tacrolimus level (1). Even though, similar to other strategies, there is no clinical trial to support this recommendation, evidence continue to accumulate supporting the benefit of significant reduction in CNI. Recently, Egli et al. (14) explored BK virus-specific T-cell response as a marker of antiviral immune protection. The study found that BK virus-specific T-cell response in vitro inversely correlated with tacrolimus concentration, although there was no relationship with MMF concentration. This further supports the idea that reduction in CNI level would be a better first step for the management of BKN. The risk of rejection after reduction in immunosuppression remains a valid concern, as acute rejection rates as high as 75% have been reported after reduction in immunosuppression with various protocols (13). However, we believe that with our suggested strategy, this risk will not be higher than the traditional reduction in MPA and maintenance of CNI as the first step. In this setting, we agree that the therapeutic drug monitoring for MPA level might have a role in the prevention of acute rejection (7–9). In conclusion, we admit that well-designed clinical trials would be the best way of studying the role of differential reduction in immunosuppressive agents in patients with BKN and use of standard pharmacokinetic methods to examine the drug exposure provides more accurate data; however, in the absence of such studies, the use of observational clinical data such as ours and in vitro studies would provide a better rationale for a therapeutic approach than following practices that are not based on tangible evidence. Wana Manitpisitkul1 Abdolreza Haririan2 1Department of Pharmacy, University of Maryland Medical Center Baltimore, Maryland 2 Division of Nephrology University of Maryland Medical Center Baltimore, Maryland
Read full abstract