Impact of carbapenem antibiotics on mycophenolate mofetil in hematopoietic stem cell transplant patients with interruption of the enterohepatic recycling: a retrospective study

BACKGROUND: Mycophenolate mofetil (MMF), a prodrug of mycophenolic acid (MPA), is widely used in the prophylaxis for graft-versus-host disease (GVHD) following hematopoietic stem cell transplantation (HSCT). MPA undergoes enterohepatic recycling (EHR). Oral antibiotics can affect MPA concentration by reducing intestinal flora-mediated EHR. However, the effect of intravenous antibiotics on MPA concentration is not clear, especially in patients whose EHR is already interrupted. This study was conducted to determine whether intravenous carbapenem antibiotics (CBP) influence the pre-dose plasma concentration (C(0)) of MPA in HSCT patients when the EHR of MPA is interrupted by cyclosporine and gut decontamination. METHODS: The HSCT patients who received immunosuppressive therapy with MMF and cyclosporine, as well as treatment with CBP were screened as potential candidates. Patients who lacked MPA C(0) measurements before or during CBP use, had combination therapy of rifampin with MMF, or switched from IV to oral MMF were excluded. The liver/renal function, demographic information, albumin/cyclosporine concentration, MPA C(0) and medication information were collected. The changes in the MPA C(0) before and during CBP use were evaluated, and the influence of related clinical factors was also estimated. RESULTS: CBP resulted in a significant reduction in the MPA C(0) from 0.65±0.33 to 0.43±0.30 µg/mL. Linear regression analysis indicated a weak correlation between the dose-normalized C(0) of MPA and the dosage of CBP during CBP use (r(2)=0.129, P=0.009). Univariate and multivariate analysis confirmed that the MPA C(0) had no relevance to rifaximin administration (P=0.249–0.700), demographics (P=0.118–0.599), fluctuation of plasma albumin (ALB, P=0.943 and 0.609) and cyclosporine concentrations (P=0.647 and 0.112), or liver and renal functions (P=0.078–0.887) no matter whether the CBP were used. However, compared with the non-gut decontamination group, larger interindividual variabilities and smaller decreases in MPA C(0) (6.60% vs. 41.73%) during CBP therapy were seen in the gut decontamination group, although it was a nonsignificant trend. CONCLUSIONS: CBP decreased the MPA C(0) in Chinese HSCT patients even when MMF is used in combination with cyclosporine and rifaximin. If antibiotics must be used, and CBP in particular, therapeutic drug monitoring should be performed to ensure adequate exposure.