Implications of methodologic bias for therapeutic drug monitoring.

Abstract

T clinical laboratory has limited ability to determine the therapeutic interval for therapeutic drug monitoring. Laboratory personnel are skilled at determining the reference interval for endogenous analytes; however, these procedures do not apply to exogenous analytes, such as therapeutic drugs. In most cases, the therapeutic interval and the toxic concentration for a drug are determined during preclinical trials in conjunction with the pharmaceutical manufacturer or studies reported in the literature and most often are based on nonimmunoassay techniques. The book Clinical Guide to Laboratory Tests, edited by Norbert W. Tietz, PhD, provides a therapeutic interval and toxic concentration for most drugs monitored by the clinical laboratory.1 This book is a valuable reference to a laboratory in establishing a therapeutic interval and a toxic concentration for drugs. In fact, several laboratories use this reference or other books on therapeutic drug monitoring as a basis for their therapeutic intervals and toxic concentrations. Such references are needed because the laboratory cannot give patients a therapeutic drug to determine the concentration for clinical efficacy and a toxic concentration. The underlying assumption is that the structure and molecular weight of a therapeutic drug are known, and for a given methodology the quantification of the drug would be consistent. What happens when instruments using similar methodology give significantly different results for a therapeutic drug? An example is therapeutic drug monitoring of tacrolimus (FK506). Recent reports in the literature, including the proceedings of a consensus meeting, recommend time-dependent therapeutic intervals of tacrolimus of 10 to 15 ng/ mL during the first 3 months and 5 to 10 ng/mL for maintenance therapy. These recommendations are based on data from multiple centers using the microparticle enzyme immunoassay (MEIA) method.2 The earliest clinical trials directed toward establishing a therapeutic range for tacrolimus employed methylene chloride extraction of whole blood and quantitation by dual antibody enzyme-linked immunosorbent assay (ELISA). Subsequently, this ELISA format was compared with the Abbott MEIA, and al-