The utility of trough mycophenolic acid levels for the management of lupus nephritis

Abstract

Background: Monitoring of mycophenolic acid (MPA) levels may be useful for effective mycophenolate mofetil (MMF) dosing. However, whether commonly obtained trough levels are an acceptable method of surveillance remains debatable. We hypothesized that trough levels of MPA would be a poor predictor of area under the curve (AUC) for MPA.

Methods: A total of 51 patients with lupus nephritis who were on MMF 1500 mg twice a day and had a 4-h AUC done were included in this study. MPA levels were measured prior to (C0) and at 1 (C1), 2 (C2) and 4 (C4) h, followed by 1500 mg of MMF. The MPA AUC values were calculated using the linear trapezoidal rule. Regression analysis was used to examine the relationship between the MPA trough and AUC. Differences in the MPA trough and AUC between different clinical and demographic categories were compared using t-tests.

Results: When grouped by tertiles there was significant overlap in MPA, AUC 0-4 and MPA trough in all tertiles. Although there was a statistically significant correlation between MPA trough levels and AUC, this association was weak and accounted for only 30% of the variability in MPA trough levels. This relationship might be even more unreliable in men than women. The use of angiotensin-converting enzyme inhibitors or angiotensin receptor blockers was associated with increased MPA trough levels and AUC at 0-4 h (AUC0-4).

Conclusion: Trough levels of MPA do not show a strong correlation with AUC. In clinical situations where MPA levels are essential to guide therapy, an AUC0-4 would be a better indicator of the adequacy of treatment.

FIGURE 1

Relationship between MPA trough plasma concentration and MPA AUC_0–4. All subjects on oral MMF 1500 mg twice a day. (A) Plot of trough plasma concentrations verse MPA AUC_0–4 with a regression line. (B) Box plot of tertiles of trough plasma concentrations and MPA AUC_0–4. (C) Plot of trough MPA levels and the difference between measured MPA AUC_0–4 and the MPA AUC_0–4 predicted from the regression equation. The shaded area depicts 95% CI for the differences between predicted and measured AUC_0–4.

FIGURE 2

Relationship of (A) MPA trough levels and (B) MPA AUC_0–4 levels and clinical variables. 1 = male, 2 = female; 1 = White, 2 = African American; 1 = smoker, 2 = nonsmoker; 1 = DM, 2 = no DM; 1 = HTN, 2 = no HTN; 1 = hyperlipidemia, 2 = no hyperlipidemia. Error bars depict the standard deviation.

FIGURE 3

The relationship between (A) MPA AUC_0–4 levels and (B) MPA trough levels with concurrent medication use. 1 = steroid use, 2 = no steroid use; 1 = Plaquenil use, 2 = no Plaquenil use; 1 = ACE inhibitor use, 2 = no ACE inhibitor use; 1 = diuretic use, 2 = no diuretic use. Error bars depict the standard deviation.