Original ArticleMycophenolic Acid Estimation by Pooled Sampling: A Novel StrategyMathew, Binu S MD*; Fleming, Denise H PhD*; Prasanna, Samuel MSc†; Basu, Gopal DM‡; Chandy, Sujith J MD*; John, George T DM‡Author Information From the *Clinical Pharmacology Unit; †Biostatistics Unit; and ‡Department of Nephrology, Christian Medical College, Vellore, Tamil Nadu, India. Received for publication September 10, 2009; accepted January 12, 2010. Correspondence: Denise H. Fleming, PhD, Clinical Pharmacology Unit, Christian Medical College Hospital, Vellore, Tamil Nadu, India, 632004 (e-mail: [email protected]). Therapeutic Drug Monitoring: April 2010 - Volume 32 - Issue 2 - p 141-144 doi: 10.1097/FTD.0b013e3181d3d051 Buy Metrics Abstract The aim of the study was to determine the reliability of estimating area under the curve from 0 to 6 hours (AUC0-6) of mycophenolic acid (MPA) by pooling the blood samples from different sampling time points. Eighty 6-hour concentration-time profiles were obtained from 68 patients on mycophenolate mofetil and the MPA AUC0-6 was calculated. In the pooled strategy, each of the equally spaced time point samples was pooled into two samples. Two rectangles were created instead of multiple trapezoids and the sum of their areas equal to the MPA AUC0-6. The linear correlation (r2), intraclass correlation, bias, and precision were calculated between the pooled MPA AUC0-6 and the MPA AUC0-6 derived from measurements at different time points. Pharmacokinetic profiles of an additional 20 patients were obtained to study the possibility of using fewer time points to create a single pooled sample to obtain MPA AUC0-6. The linear correlation (r2) and intraclass correlation between pooled and measured MPA AUC0-6 was 0.982 and 0.979, respectively. There was a highly significant correlation (r2) of 0.978 between the pooled versus measured for both MPA AUC0-3 and MPA AUC3-6. The mean bias and precision (95% confidence interval) for pooled with total measured MPA AUC0-6 was -6.4% (-7.8% to -4.94%) and 7.37% (6.21%-8.54%), respectively. The pooled sample approach using only five time points to estimate MPA AUC0-6 had an unacceptable bias and precision. Pooling 10 samples to a set of two samples gave a highly accurate measure of MPA AUC0-6. The advantages for a central laboratory are the high throughput of samples and the transportation of only two specimens from other centers, all of which leads to a reduction in cost. This approach is extremely useful for studies aimed at examining the bioavailability of mycophenolate mofetil in different ethnic populations within India. © 2010 Lippincott Williams & Wilkins, Inc.