Published: June 21, 2011 r2011 American Chemical Society 6237 dx.doi.org/10.1021/ac200820q | Anal. Chem. 2011, 83, 6237–6244 ARTICLE pubs.acs.org/ac A Convenient Strategy for Quantitative Determination of Drug Concentrations in Tissue Homogenates Using a Liquid Chromatography/Tandem Mass Spectrometry Assay for Plasma Samples Hao Jiang, † Jianing Zeng,* ,† Naiyu Zheng, † Hamza Kandoussi, † Qianping Peng, ‡ Jack L. Valentine, ‡ Robert W. Lange, ‡ and Mark E. Arnold † † Analytical & Bioanalytical Development, and ‡ Drug Safety Evaluation, Research and Development, BristolÀMyers Squibb Company, Princeton, New Jersey 08540, United States ’ INTRODUCTION In tissue sample analyses performed via liquid chromatogra- phyÀtandem mass spectrometry (LC-MS/MS), standards and analytical quality controls (QCs) are ideally prepared in drug- free tissue homogenates to maintain the matrix consistency between the standards, analytical QCs, and tissue homogenate samples. In these cases, 1À9 analyte working solutions were either spiked onto drug-free solid tissue before homogenization or directly spiked into tissue homogenates to prepare standards and analytical QCs, with tissue samples homogenized and extracted in the same way as the standards and analytical QCs. However, using a substitute matrix such as 4% bovine albumin, 10 water, 11 or blood 12,13 to prepare standards and analytical QCs for tissue sample analysis was also reported when a drug-free tissue was not available. It was observed that analyte recoveries were different between the standards and tissue homogenate samples, and they caused significant data biases. In addition, the impact on data accuracy was even more significant when an analogue internal standard was used. 12 Therefore, a correction factor, the recovery difference between the standards and tissue samples, was used to correct reported study sample drug concentrations. In this study, the data bias resulting from the recovery difference between standards and tissue homogenate samples was defined as the tissue matrix effect. In occasional cases, the tissue matrix effect was minimized when standards were prepared with the homo- genization solution, even when a stable isotopically labeled internal standard was not used. 13 In this manuscript, we successfully used a validated LC-MS/ MS method for the quantitative determination of BMS-650032 (an inhibitor of hepatitis C virus NS3 protein; see Figure 1) in dog plasma to determine BMS-650032 in dog liver homogenate samples. The standards and analytical QCs were prepared in plasma, whereas the tissue samples were homogenized in acet- onitrile/Hank’s buffered salt solution (20:80, v/v). The matrix differences were minimized by diluting the tissue homogenate samples with dog plasma before sample extraction. The tissue matrix effect was evaluated by calculating the percentage devia- tion (%Dev) of the measured concentrations of homogenate QCs (after dilution with plasma at different dilution factors) from their nominal concentrations in order to determine a minimal required dilution factor (MRDF), at which the tissue matrix effect was not significant (i.e., the %Dev value of the measured from the nominal concentration was minimal). The tissue samples from several drug safety evaluation studies have Received: March 31, 2011 Accepted: June 21, 2011 ABSTRACT: Quantitative determination of drug concentrations in tissue homogenates via liquid chromatographyÀtandem mass spectrometry (LC-MS/MS) is commonly conducted using the standards and analytical quality controls (QCs) prepared in the same matrix (tissue homo- genates), to keep the matrix and its effects consistent on the analytes during sample extraction and analysis. In this manuscript, we proposed to analyze tissue homogenate samples using an LC-MS/ MS assay with the standards and analytical QCs prepared in plasma after tissue homogenate samples were appropriately diluted with plasma. BMS-650032 was used as a model compound, and its validated dog plasma assay was used for dog liver sample analyses. The tissue matrix effect was evaluated by diluting liver homogenate QCs with drug-free plasma at different dilution factors to determine the minimum required dilution factor (MRDF) at which tissue matrix has insignificant impact to the plasma assay. The percentage deviation of the measured concentration from the nominal concentration was used as an indicator of the tissue matrix effect. The results suggested that the tissue matrix effect was decreased as the plasma dilution factor increased. Based on the results of the tissue matrix effect evaluation, liver homogenate samples were analyzed after appropriate dilutions with plasma at the MRDF or greater dilution factors. The results confirmed that this approach generates accurate data, and the process is very convenient and economic. This approach has been used on the analyses of different tissues (liver and brain) and biofluid (bile) to support several drug development programs.