1
INTRODUCTION: HISTORY OF METABOLITE SAFETY IN DRUG DEVELOPMENT

Dennis A. Smith1 and Suzanne L. Iverson2,3

1 The Maltings, Kent, UK

2 Drug Metabolism Discussion Group (DMDG), Leicester, UK

3 Department of Medical Physics and Biomedical Engineering, Sahlgrenska University Hospital, Gothenburg, Sweden

1.1 PEOPLE, EVENTS, AND REACTION

Change, whether social or technological, often is catalyzed by a convergent set of events and influences. Drug testing using animals became important in the twentieth century. In 1937 a preparation of sulfanilamide, using diethylene glycol (DEG) as a solvent and called the preparation "elixir sulfanilamide," was marketed. DEG is highly toxic and the preparation led to the deaths of more than a hundred people. No animal testing had been conducted, a step that would have highlighted the risks. The public outcry caused by this incident led to the passing of the 1938 Federal Food, Drug, and Cosmetic Act requiring safety testing of drugs on animals before they could be marketed. This event would have remained perhaps as a footnote, with drug metabolism remaining an academic pastime of minor note, but further events and people would catalyze dramatic change.

Richard Tecwyn Williams was one of the founding scientists in the systematic study of the metabolism of chemicals including drugs. He detailed this in a major book Detoxication Mechanisms which was published in 1947. His work led to the widely adopted phase 1 (oxidation, hydrolysis, etc.) and phase 2 (conjugation) divisions of drug metabolism. Following his appointment in 1949 to Chair of Biochemistry at St Mary's Hospital Medical School in London, a growing group of researchers studied in the field. The group focused on many aspects of drug metabolism, but species differences in metabolism became a key topic, building on observations described in the aforementioned book. For instance, in 1956 together with Parke, he published a paper describing "species differences in the ortho-hydroxylation and para-hydroxylation of aniline." As part of his later research activities (Schumacher et al. 1965), he examined the metabolism of thalidomide (a key component in the changes; see as follows) showing hydrolysis products and hydroxylated metabolites in various species, but a constant theme was species differences. Typical findings of Williams' group are exemplified by studies with sulphamethomidine (Bridges et al. 1969). In the rat, rabbit, and monkey, the main metabolite was the N4-acetyl derivative. In man, the major metabolite was the sulphamethomidine N1-glucuronide, which was also excreted by the monkey but not by the rat or rabbit. Many other drugs were examined, all leading to descriptions of species differences of a major or minor form such as amphetamine, methamphetamine, etc. (Caldwell et al. 1972). In many of these publications, the use of 14C labeled drugs was used to obtain detailed results. The studies created an awareness that the products of metabolism excreted by animals could differ markedly from those excreted by man. Again this course of research could have remained an academic pursuit, but events dictated otherwise.

The drug thalidomide was first marketed in 1957 in West Germany under the trade name Contergan as a sedative. Generally very well tolerated it was prescribed for a number of CNS indications, but crucially it became commonly administered to counteract nausea and alleviate morning sickness in pregnant women. The drug was licensed out to other distributors and was widely used. A terrible consequence of its wide acceptance was the realization that thalidomide was a human teratogen. Throughout the world, about 10?000 cases were reported of infants with phocomelia (a malformation in which the limbs were absent or present as stumps), with a high morbidity (50% survival). Deformities were also observed in the eyes, hearts, and alimentary and urinary tracts. At this time there was no legal requirement for animal studies to evaluate risk in pregnancy, although they were already established, albeit without rigorous guidelines or protocols on species selection, duration, and size of dosing.

This drug tragedy led to a complete change in the way drugs could be marketed. For instance, in the United Kingdom in 1963, Sir Derrick Dunlop set up a committee to investigate the control and introduction of new medicines which resulted in the Committee on Safety of Drugs being established. This evolved into the Committee on Safety of Medicines. These moves paralleled the Kefauver Harris Amendment in the United States and Directive 65/65/EEC1 in the EU. The principal change was the enforcement that applicants needed to prove efficacy and to disclose all side effects encountered in testing prior to marketing approval.

1.2 THE RISE OF INDUSTRIAL DRUG METABOLISM

The metabolic fate of drugs in animals and man, in these times, was not a major focus or priority research topic for the innovator company. Illustrative of this period is propranolol, the first full antagonist of ß-adrenoceptors which was discovered in 1962, first marketed in 1964. Analytical methods for propranolol were first published in 1965 with techniques relying on the optical (fluorescence) properties of the drug, while it was not until 1967 (Bond 1967) that preliminary reports on its metabolic fate first appeared.

Williams' book title implied that metabolism was primarily a detoxification step, a concept that is still mainly correct. However growing evidence indicated that metabolism could also lead to toxicity. These observations concerned metabolites which were intrinsically reactive. Pioneering work on the carcinogenicity of polycyclic aromatic hydrocarbons and other planar heterocyclic aromatic compounds (Boyland 1969; Ames et al. 1972) had shown that the reactive metabolites were the ultimate toxin. This finding was broadened by early studies of liver necrosis in rodents. Many studies demonstrated that enhanced toxicity was associated with induction of liver enzymes by agents such as phenobarbital and attenuated toxicity with the inhibition of drug-metabolizing enzymes by agents such as SKF525A. Radiolabeled studies showed that this toxicity was accompanied by the irreversible covalent binding of drug-related material. By the 1970s Gillette et al. (1974) was able to establish that cellular necrosis, hypersensitivity, and blood dyscrasias could result from the formation of reactive metabolites.

The increased focus on drug safety, the realization that species differences could occur, and the concept that drugs could be metabolized to reactive metabolites and bind to organs, combined to lead to the decision that part of drug safety should have a drug metabolism component. Thus toxicology species and eventually human would be examined for overall excretion of the drug and its products, evidence that the drug had not been sequestered in the body (of animals) and a view as to what metabolites were produced. Metabolite profiling, where urine and feces extracts were compared quantitatively between species and eventually human, by techniques such as thin-layer chromatography became common practice. These experiments were conducted after the synthesis of a 14C- or 3H-labeled version of the candidate drug of interest. These early studies lacked sophistication as methodology was not advanced. Identifying metabolites required considerable workup of fairly large quantities. In vitro reagents such as human microsomes and hepatocytes were not available to allow early species comparisons. The principal focus was on excreted metabolites as evidenced with work on propranolol. The determination that 4-hydroxypropranolol, a major circulating metabolite of propranolol, had pharmacological activity equivalent to the parent was not made until 1971 (Fitzgerald and O'Donnell, 1971), 7?years after its launch.

Studies tended to report the identity of excreted metabolites. Circulating drug-related material was often reported as the proportion of parent present in the total radioactivity measured in plasma. Regardless of dose, results were quoted as percentage and not absolute amounts. Typical of this is isoxepac (6,11-dihydro-11-oxodibenz[be]oxepin-2-acetic acid) which Illing and Fromson reported in 1978. The disposition was studied in rat, rabbit, dog, rhesus monkey, and human. Fecal excretion of radioactivity occurred in the rat (26-37%) and dog (33-49%), whereas in the other species elimination was mainly urinary (<83%). Biliary excretion accounted for 18-52% of the dose in the rat and dog. Enterohepatic circulation was demonstrated in both species. Plasma of all species was found to contain mainly unchanged isoxepac. The compound was rapidly eliminated from plasma of dog, rhesus monkey, and man but was more slowly eliminated in rat and rabbit. In the rabbit and dog the principal metabolites were the glycine and taurine conjugates of isoxepac, respectively, whereas in the rhesus monkey and man, isoxepac was excreted unchanged or as the glucuronide. This species difference was highlighted even though the metabolites were conjugates. It is arguable that emphasis was put on species differences to justify publication. The emphasis from excreted metabolites to both circulating metabolites and excreted metabolites was a gradual process over the next decade, and authors began to state more implications for their findings. For instance, the disposition of amlodipine,...