734   An International Interlaboratory Quality Control (QC) Program for Therapeutic Drug Monitoring (TDM) in HIV Infection.

R. Aarnoutse*, D. Burger, C. Verweij-Van Wissen, E. Van Ewijk-Beneken Kolmer, E. Wuis, and Y. Hekster.
Univ. Med. Ctr. Nijmegen.

Introduction:Analytical methods for antiretroviral drugs are being applied in TDM for antiretroviral drugs and in pharmacokinetic (PK) studies. Intra- and interlaboratory QC procedures are required to ensure that these methods have sufficient accuracy, precision and specificity. We started an international interlaboratory QC program for measurement of protease inhibitors (PIs) and non-nucleoside reverse transcriptase inhibitors (NNRTIs) in plasma. Results of the second round are presented.

Methods:Drug-free plasma from HIV-negative volunteers was spiked with indinavir, nelfinavir, ritonavir, saquinavir, efavirenz and nevirapine to obtain 6 QC samples, each containing the 4 PIs or the 2 NNRTIs in varying concentrations (low, median and high). Thirteen laboratories in the US, Canada, Europe and Australia participated in this round of the program. They were requested to analyse the samples within 4 weeks after dispatch and were asked to provide details about their analytical procedures. Twenty percent limits around the weighed-in concentrations of the drugs were considered to be appropriate thresholds for a satisfactory measurement.

Results:Laboratories used HPLC-UV or LC-MS to measure the drugs. Measurements of indinavir, nelfinavir, ritonavir and saquinavir yielded satisfactory results in 72%, 77%, 83% and 92% of the analyses, respectively. For efavirenz and nevirapine these percentages were 74% and 62%. Only one laboratory performed all measurements within 20% limits. Eight laboratories reported <100% but more than 75% of the concentrations within 20% limits. The remaining four laboratories had less satisfactory results. Three laboratories appeared to use methods with systematic errors in one direction. One laboratory made reporting errors and another one made a factor 10 error. Low drug concentration levels were as accurately measured as median and high concentrations. All laboratories were informed about their own performance to enable them to optimize their methods.

Conclusions:The observed variability in the performance of laboratories may have important implications for TDM or PK studies. A quality control program for measurement of antiretroviral drugs is a useful instrument to monitor accuracy of analytical methods. The program will be open for other laboratories to participate.