Pharmacokinetic/Pharmacodynamic Drug Interaction Between Rosiglitazone and Mycophenolate Mofetil in Kidney Transplantation

Anemia continues to be a major problem in kidney transplant recipients, being mainly related to iron deficiency, gastrointestinal blood loss, and toxicity of immunosuppressants or other pharmacologic agents, such as inhibitors of the renin-angiotensin system (1–3). Here, we report a serious drug-to-drug interaction between mycophenolate mofetil (MMF) and rosiglitazone (4). A 47-year-old male patient received his first cadaveric renal allograft in January 1998. He rapidly recovered renal function (serum creatinine 1.30 mg/dL on day 15 after transplant), and no episodes of acute graft rejection occurred in the early posttransplant period. The patient was on maintenance immunosuppression with cyclosporine A (given at 275 mg daily and kept constant during all the observational period) and MMF (5). Nifedipine and furosemide were started early after transplantation to control high blood pressure. He also received sodium bicarbonate and allopurinol to control hyperuricemia and calcitriol to treat elevated Ca/P. On March 2006, a progressive decline of red blood cell (RBC) count was documented. Erythropoietin therapy was started and anemia partially recovered. On January 2007 the patient was enrolled in a pilot study aimed at evaluating the metabolic and renal effects of rosiglitazone in proteinuric renal transplant recipients. At baseline evaluation serum creatinine concentration and urinary protein excretion were 1.81 mg/dL and 2.06 g per 24 hr, respectively. At this time point mycophenolic acid (MPA) pharmacokinetic measurement was also performed after the morning oral dose of MMF (1,000 mg every 12 hr). The dose-normalized MPA area under the curve (AUC) was 92.2 mg·hr/L/g MMF. Previous reports have shown that MMF exerts a negative effect on erythropoiesis, and that hemoglobin concentrations correlated negatively with MPA plasma levels (6). According to these observations, the daily MMF dose was halved (from 1,000 to 500 mg twice per day). Thereafter, the patient started rosiglitazone therapy (4 mg/day). A week later, ramipril was also introduced to optimize blood pressure control, as requested by the study protocol. After beginning rosiglitazone, a progressive decline in RBC count was shown reaching a nadir value of 3.3 × 106 cells/μL 3 months later, despite the appropriate erythropoietin treatment. Hemoglobin (10.5 g/dL) and hematocrit (31%) were also reduced as compared with baseline. Anemia was not caused by gastrointestinal blood loss, iron deficiency, or hemolysis (lactate dehydrogenase 307 mg/dL and total bilirubin 0.52 mg/dL), and was characterized by erythrocytes with normal mean corpuscular volume (92.8 fL) and mean corpuscular hemoglobin concentrations (34.1 g/dL). At this time, the MPA AUC was twofold higher (194 mg·hr/L/g MMF) than the value measured at baseline visit. Rosiglitazone was discontinued, and progressive increase in RBC count was observed thereafter (Fig. 1). In parallel, hemoglobin (12.2 g/dL) and hematocrit (37%) increased. A third MPA pharmacokinetic evaluation, performed 2 months after rosiglitazone withdrawal, documented a marked reduction in the MPA AUC (131 mg·hr/L/g MMF).FIGURE 1.: Profiles of mycophenolic acid area under the curve and red blood cell count before, during, and after rosiglitazone administration. Time course and dosing of immunosuppressants and other drugs are also reported.Anemia has been reported as a rare event in patients treated with rosiglitazone (7, 8), and thought to be related to the drug-induced plasma volume expansion without effect on erythropoiesis (9). However, in our patient serum sodium concentration did not change significantly during rosiglitazone treatment (serum Na: before, 135.6 mEq/L; during 136.2 mEq/L; recovery 137.4 mEq/L) and body weight had not increased, which argues against a major direct effect of the drug on the RBC count. Instead, we thought that rosiglitazone- associated anemia was caused by interaction of rosiglitazone with MMF metabolism, which eventually translated to suppression of erythropoiesis. This possibility is supported by the significant reduction in the MPA AUC observed after rosiglitazone discontinuation, which paralleled a partial resolution of the anemia, suggesting a potential novel drug-to-drug interaction between MMF and rosiglitazone. The nature of this interaction is unknown. Because both drugs share, at least in part, the same metabolic fate (7, 10), preferential metabolism of rosiglitazone may have taken place, ultimately resulting in less MPA metabolism and eventually high exposure of the patient to the drug. In conclusion, we have first reported a clinically relevant drug interaction between rosiglitazone and MMF, which results in very high MPA plasma concentrations and severe anemia. It is, therefore, advisable to apply strict MPA monitoring with rapid drug dose adjustments when rosiglitazone is given to patients receiving MMF as part of their immunosuppressive therapy. Dario Cattaneo Alessandra Bitto Sara Baldelli Monica Cortinovis Department of Medicine and Transplantation Azienda Ospedaliera Ospedali Riuniti Bergamo and Mario Negri Institute for Pharmacological Research Bergamo, Italy Center for Research on Organ Transplantation "Chiara Cucchi De Alessandri & Gilberto Crespi" Azienda Ospedaliera Ospedali Riuniti Bergamo and Mario Negri Institute for Pharmacological Research Bergamo, Italy Eliana Gotti Department of Medicine and Transplantation Azienda Ospedaliera Ospedali Riuniti Bergamo and Mario Negri Institute for Pharmacological Research Bergamo, Italy Norberto Perico Giuseppe Remuzzi Department of Medicine and Transplantation Azienda Ospedaliera Ospedali Riuniti Bergamo and Mario Negri Institute for Pharmacological Research Bergamo, Italy Center for Research on Organ Transplantation "Chiara Cucchi De Alessandri & Gilberto Crespi" Azienda Ospedaliera Ospedali Riuniti Bergamo and Mario Negri Institute for Pharmacological Research Bergamo, Italy