Laboratory data in healthy volunteers: reference values, reference changes, screening and laboratory adverse event limits in Phase I clinical trials

Objective: Laboratory data are key evaluation procedures for Phase I clinical pharmacology for two reasons. Firstly, laboratory data are used within the screening process to exclude subjects with asymptomatic diseases, which could result in increased danger to themselves or confuse interpretation of the study results. Secondly, during study implementation, safety evaluation and in particular maximum tolerated dose determination have to be done by a case-by-case analysis, sometimes using laboratory adverse events (LAEs). Thus, relevant limits are needed to discriminate between a usual common variation and a significant abnormality, which is considered to be a LAE. This report presents laboratory data distribution, reference values and reference changes and, based on previously published new methods, suggests inclusion limits at screening and laboratory adverse event limits for analysis during study implementation. Subjects and methods: Nine hundred and twenty-seven young healthy male volunteers were recruited in one centre (Association de Recherche Thérapeutique). A standard screening process was carried out. Protocols were approved by the local ethics committee. Blood sampling was performed in the same conditions. Reference values (at screening and at baseline) were determined by a non-parametric procedure selecting 2.5% and 97.5% of the distribution of data. Reference changes were also defined as the 2.5–97.5% interval of distribution of the variations between the end of treatment and baseline. Inclusion limit and LAE limit methods of determination used had been specified in previous articles. Results: Detailed results of laboratory data distribution, reference values at screening and at baseline, reference changes, inclusion limits and LAE limits are presented in tables with number of subjects, mean, median, standard deviation, minimal and maximal values and the 2.5–97.5% interval for each laboratory parameter. Conclusion: The key aims of this paper are to provide clinical pharmacologists with data, reference values or changes obtained in the real conditions of Phase I study implementation, and to propose relevant limits, either for screening as inclusion limits, or during studies as LAE limits. Thus, these data, reference values and specific limits improve the capacity to screen healthy volunteers and to analyse LAEs during Phase I studies. Received: 30 July 1998 / Accepted in revised form: 25 November 1998