Higher Mycophenolic Acid (MPA) Trough Levels Result In Lower Day 100 Severe Acute GVHD Without Increased Toxicity In Double-Unit Cord Blood Transplantation (CBT) Recipients

Mycophenolate mofetil (MMF) is frequently combined with cyclosporine-A (CSA) as immunosuppression in unrelated donor CBT. Recent evidence suggests that therapeutic drug monitoring of mycophenolic acid (MPA), the active metabolite of MMF, is advisable based on intra- and inter-patient variability of MMF pharmacokinetics. Moreover, an increased incidence of acute graft-versus-host disease (aGVHD) has been associated with low unbound MPA AUCs in hematopoietic stem cell transplantation. This is highly relevant in cord blood transplantation (CBT) as aGVHD is a leading cause of transplant-related morbidity and mortality. However, as AUCs are cumbersome, a limited pharmacokinetic parameter such as MPA troughs would be ideal. Additionally, the toxicity associated with MPA trough levels is not established and is of concern in CBT especially due to the theoretical risk of myelosuppression from high MPA levels.

We evaluated the association between serial (weeks 1-6) total MPA serum trough levels and transplant outcomes in pediatric and adult recipients of double-unit CB grafts. Patients were transplanted between 8/2009 and 11/2012 for hematologic malignancies with 4-6/6 HLA-A,-B antigen, -DRB1 allele donor-recipient matched units and intravenous CSA/MMF was commenced on day -3. To evaluate the association between trough levels and outcomes, the trough levels were dichotomized into < 2 mcg/mL and ≥ 2 mcg/mL for toxicity (neutrophil and platelet engraftment, gastro-intestinal toxicity as measured by TPN duration, and CMV infection) at each time point. For the efficacy (aGVHD prevention) analysis, mean week 1 and 2 trough levels were split at < 0.5 versus ≥ 0.5 mcg/mL.

Eighty-three patients (median age 44 years, range 1-71) had MPA serum trough levels drawn weekly for the first 6 weeks after CBT. Sixty-nine (83%) received myeloablative (MA) conditioning and 45 (54%) were CMV seropositive. Diagnoses included acute leukemia (n = 51), myeloproliferative disorders/MDS (n = 7), or lymphoma (n = 25). Median trough levels by week were 0.9, 0.9, 0.6, 0.6, 0.9 and 1.3 mcg/mL. The increase in trough levels over time reached significance (p = 0.03). Recipients of MA conditioning had lower MPA troughs than those who received non-myeloablative conditioning (p = 0.02). Younger age (0-15 years old) was associated with lower trough levels (p = 0.002). By time dependent Cox regression analysis, there was no association between trough levels and toxicity as measured by speed and success of neutrophil engraftment, duration of TPN in myeloablative CBT recipients, or time to CMV viremia in seropositive patients. However, MA CBT recipients with a MPA trough ≥ 2 mcg/mL had enhanced platelet recovery (p = 0.005). In a competing risk 2-week landmark efficacy analysis, there were no differences in rates of grade II-IV aGVHD (61% vs 57%, p = 0.52) according to trough level. However, patients with a low MPA trough early post-CBT had nearly triple the incidence of severe (grade III-IV) aGVHD (27.8% vs 9.5%, p = 0.06, Figure 1).

This analysis suggests that higher total MPA serum trough levels are safe from the standpoint of toxicity and may protect against severe aGVHD. The beneficial effect on platelet engraftment was unexpected but may be explained by a lower incidence of severe aGVHD. Prospective investigation of the utility of MPA trough measurements, and ultimately intervention based on drug monitoring in a larger CBT series is warranted.

No relevant conflicts of interest to declare.