Studies on the variation in clinical laboratory data and safety evaluation of pharmaceuticals

The safety of pharmaceuticals has become increasingly important not only in daily medical treatment but also in clinical trials. Although clinical laboratory data are more objective than clinical symptoms, the determination as to whether they indicate abnormal variations depends largely upon the clinical judgment of physicians. The process of determination has not been sufficiently objectified. The present study investigated the indices of criteria for variations in clinical laboratory data obtained in clinical trials. Then, detection rates of abnormal variations were compared between our determination method that employs the reference change value (RCV) expressing the width of biological variation for each test component and conventional determination methods. The study also demonstrated that by combining standard values and the RCV for determination, abnormal variations were found at a rate greater than 50%. The method we propose was applied to the safety evaluation of pharmaceuticals. In clinical trials on the antiviral drug ribavirin administered alone, components of laboratory tests were selected that should be noted in studies on its effects. Expect for decreases in red blood cell counts and hemoglobin values, which are closely associated with anemic symptoms and well known to hepatologists, the increasing trend in platelet counts and decreasing trend in albumin were found to be laboratory test components that should be paid attention to, even though they may not be obvious.