Urine drug screening results from volunteers in phase I clinical pharmacology studies: are we being misled?

Abstract

The general intent of phase I clinical pharmacology studies is to demonstrate the safety and tolerability of investigational new drugs in healthy human volunteers. There is emerging evidence that people who volunteer for these studies are not always truthful with investigators during the screening process. All healthy volunteers who participate in studies at the SmithKline Beecham Clinical Pharmacology Unit in Philadelphia, Pennsylvania, are required to submit to urine drug testing. During 11 months of 1996, a total of 1,469 urine samples were collected and tested for eight different drugs or classes of drugs of abuse. The urine samples collected during the first five months of 1996 were all analyzed using EMIT (Syva Corporation) and interpreted according to the guidelines established by the National Institutes of Drug Abuse (NIDA). Of 534 samples, 12 (2.2%) were reported as positive. During the last 6 months of 1996, a new methodology using a fluorescence polarization immunoassay (FPIA) was used. This assay had lower limits of quantification than EMIT, and more stringent interpretation guidelines than those of the NIDA were used. Of 935 samples analyzed by FPIA, 89 (9.5%) were positive. Of the 89 positive test results, 59 were below the cut-offs specified by the NIDA guidelines and would have been reported as negative. Interpretation of urine drug screen results according to the NIDA guidelines is not acceptable for clinical pharmacology investigations.