control over its use and a major part of the responsibility for the judicious application of the compound now rests with the medical profession and the scientific organizations and societies interested in drug therapy. It is therefore hoped that the many local, national, and international organizations which are concerned with effects and side effects of drugs will join with the pharmaceutical manufacturer and the Government agencies charged with surveillance of drugs in an effort to assure the most beneficial and safe application of our modern chemical weapons against disease. It is proposed to review the various aspects of drug safety evaluation under the following subheadings: 1. Preclinical toxicity testing in animals. 2. Safety evaluation in human subjects. 3. Classification of adverse drug effects in man. 4. Predictability of drug toxicity in humans from preclinical animal experiments; critique of the presently employed methods, proposals for improvements, and major areas of future research. 5. Conclusions. 1. Preclinical toxicity testing of new drugs in animals Before a newly-discovered drug may be administered to human subjects, preliminary toxicity tests in animals are performed. The results of these tests must be filed with the FDA before the compound is shipped to the clinical investigators. There is a considerable divergence of opinion among the experts in the field as to the extent and methodological details of such studies. Although our presently employed experiments are often subject to criticism, it must be acknowledged that in the past 15 years the investigations of a few toxicologists and among them most prominently Dr. A. J. Lehman, Director, Division of Pharmacology, and his coworkers in the FDA, have provided workable guidelines which assure excellent standards for the preclinical toxicity testing of drugs in animals. These methods and rules are now essentially accepted all over the world. The principle of the toxicity tests involves single or continuous administration of the compounds to groups of animals at various dose levels and the observation of the drug effects on general health and development, blood formation, and function and structure of the vital organs. The extent and duration of these studies depend on the scope of the clinical trials planned for the drug; but in the majority of cases the investigations are performed in two stages; namely, (1) an acute and subacute toxicity test which precedes initial tolerance and efficacy studies in man and (2) a chronic toxicity test which is conducted with compounds in extended clinical trials and those scheduled for marketing. The procedures presently used in our laboratories are summarized in table 1. All drugs which will be given to women in childbearing age must also be administered to pregnant animals. TABLE 1.-Toxicological tests in animals I. ACUTE 1. LD50 determination in 4 species: mice, rats, guinea pigs, and dogs. 2. Pyramiding single-dose studies in dogs. 3. Local effects (topical or parenteral agents). II. SUBACUTE 1. Thirteen weeks' administration to 4 groups of 16 rats (3 dose levels). 2. Thirteen weeks' administration to 4 groups of dogs (3 dose levels). III. CHRONIC 1. One to 11⁄2 years' administration to 4 groups of 50 rats (3 dose levels). 2. One year's administration to 4 groups of 8 dogs (3 dose levels). 3. Six months to 11⁄2 years' administration to a third species if indicated. 4. Reproduction experiments in rats (litter test). IV. FUNCTION TESTS USED IN EXPERIMENTAL TOXICOLOGY 1. Hematology hematocrit total erythrocyte counts total leukocyte counts hemoglobin erythrocyte sedimentation differential rates leukocyte counts reticulocyte counts 2. Liver function tests Bromsulphalein retention Direct and indirect serum bilirubin Thymol turbidity Serum alkaline phosphatase Serum glutamic oxaloacetic transaminase (SGO-T) Serum glutamic pyruvic transaminase (SGP-T) Serum isocitric dehydrogenase 3. Renal function tests Blood urea nitrogen Urinalysis color, transparency, pH, specific gravity, albumin, flucose, bilirubin, acetone microscopic examination (RBC, WBC, triple phosphate crystals, amorphous phosphate crystals, calcium oxalate crystals, epithelial cells, bacteria). V. AUTOPSY AND HISTOLOGICAL EVALUATION Complete autopsy of all animals of the subacute and chronic tests. The evaluation and interpretation of these animal tests require much experience. It is understandable that during the prolonged administration of a drug many functional and structural changes of the organs may develop. The FDA requires that in all subacute and chronic toxicity tests the drug is administered at a minimum of three dose levels of which one dose level must be increased to produce clearcut toxicity. This means that in all preclinical and subsequent animal studies toxic effects will be noted, and it is often most difficult to determine whether or not these changes which must be considered as a consequence of intentional overdosage are of any practical significance for the use of a drug at therapeutic dose level in humans. 2. Safety evaluation in human subjects After satisfactory completion of initial animal toxicity studies the drug is administered to a small number of human subjects. The initial 37-272-64-pt. 2- -10 dose, determined in collaboration with the toxicologist, pharmacologist, and clinical pharmacologist, is low and is subsequently increased slowly. The initial clinical trial is conducted on inpatients or volunteers in a specially equipped hospital section and the subjects are under constant supervision by the clinical pharmacologist and trained nurses. Laboratory examinations of the blood, kidney, and liver function and controls of blood pressure, ECG and the neurological status are conducted at regular intervals. The dose is increased until subjective intolerance symptoms (sleepiness, perspiration, nausea, etc.) or objective signs of a pharmacological drug effect (e.g., hypotension, oversedation, diuresis) are noted or until the maximal dose previously agreed upon, based on pharmacological and toxicological data, is reached. This transition from the animal experiment to the first few human patients is certainly one of the most important steps in the development of a new drug, but it is not a hazardous and dangerous phase as it is sometimes believed to be by people not involved in drug research. As a matter of fact, it is our experience that if the proper animal studies have been done and if the above described precautions are rigidly adhered to, particularly the monitoring and supervision by the clinical pharmacologist, there is hardly a chance that the volunteers involved in phase I trials will be seriously injured by an unexpected toxic drug effect. The safety evaluation continues in subsequent phases of the clinical investigation. In phase II, during which the therapeutic effectiveness is established, precautions, and supervision of the patients are just as rigorous as during phase I, but as more experience is gained laboratory controls and examination by the physician may be spaced further apart. During phase III, the extended clinical trial, the drug is turned over to specially qualified clinical investigators in medical schools, hospitals, and sometimes in private practice. This permits administration to a larger number of patients under clinical and experimental conditions. In this phase an attempt is made to obtain information about safety and efficacy of the drug for mass use. In special cases, particularly for drugs which are administered over many years, a prolonged tolerance trial in humans is initiated. This trial should, if possible, be done in a patient population requiring continuous administration of the drug. Extensive clinical and laboratory controls are performed weekly or biweekly in the first 2 months and at monthly intervals thereafter. A control group treated with a placebo or an established drug must be included in such studies. 3. Classification of adverse drug effects in man For the evaluation of the efficiency of the animal and human safety testing procedure, it is necessary to classify the various adverse drug reactions observed in man. A classification which is based on the mechanism of toxic drug effects has been proposed by us and is summarized in table 2. In our discussion only those undesirable and harmful effects will be included which may be encountered during proper administration of the correct dose. It therefore does not contain damages which may occur after accidental or voluntary overdosage or after improper use of the drug, such as intravenous injection of a solution suitable only TABLE 2.-Drug-related toxic manifestations in humans Designation Related to desired pharmacological, biochemical, or endocrine effects. Exaggerated effect at recommended dose. Related to desired pharmacological, biochemical, or endocrine effects. Drug acting on wrong target organ. Related to desired pharmacological, biochemical, or endocrine effects, requiring preexisting pathology which is not drug related. Related to desired pharmacological, biochemical, or endocrine effects, requiring contributing iatrogenic and other exogenous factors. Related to undesired pharmacological, biochemical, or endocrine effects. Related to undesired pharmacological, biochemical, or endocrine effects, requiring preexisting pathology which is not drug related. Related to undesired pharmacological, biochemical, or endocrine effects, requiring contributing iatrogenic and other exogenous factors. Related to sensitization and allergic reactions. Related to sensitization and allergic reactions requiring contributing exogenous factors. Related to idiosyncrasy and other unknown mechanisms. Related to interference with absorption of nutrients. Related to interference with natural defense mechanisms. Related to tissue storage or precipitation of drugs or metabolities. Toxic effects on the fetus. Related to tissue irritation and damage on direct contact. Organ damage due to overdosage. for intramuscular administration. These cases belong in the area of forensic toxicology which is outside the present considerations. We have this table of toxic changes and I will refer to it later on, but I think I should come to the main point now, which is: 4. Predictability of drug toxicity in humans from preclinical animal experience; critique of the presently employed methods, proposals for improvement and major areas of future research In this part of our discussion it will be shown that only a small percentage of the drug-induced toxic changes in man may be recognized or predicted by the routine animal toxicity experiment. Various attempts have therefore been made to make these preclinical toxicity studies more meaningful and some of these newer possibilities which we include in our "extended toxicity program" will be mentioned. It will also be analyzed at what stage of the clinical drug evaluation the various forms of adverse reactions may become evident. Since the situation is quite different from one form of drug reaction to the other, the various classes of adverse drug effects will be treated individually. A schematic summary of these considerations is presented in table 3 which I would like to draw to your attention. I propose that we go through the various forms of toxicities and analyze how they could be predicted or avoided by animal and human toxicity studies. Class A. Toxic manifestations related to desired drug effect without preexisting pathology Now, let us take subclass A-1, exaggerated effect at recommended dose. Hypersensitivity of certain patients to a drug may lead to marked exaggeration of the therapeutic effect and thus cause serious complications. The animal toxicity experiment is not able to predict such an event and since hyperreactive patients may be rare, the danger is usually only recognized in phase III clinical studies or after years of general use of the drug. Such hyperreactions are sometimes due to genetically determined abnormalities of drug metabolism and it is. therefore useful to study dose-response curves and the rates of metabolic inactivation of each drug in a substantial number of human subjects. You can see if there are patients who are hyperreactive to a particular drug, they may not be encountered until very late in our clinical trials. Next, subclass A-2, drug action on the wrong target organ. Most drugs act on various organs and it is not unusual that their effect on one target organ is beneficial whereas their action on other systems causes disturbing side effects (that is, dryness of mouth, hypertension, palpitations). The classical toxicity experiment is rarely, if ever, capable of detecting these reactions which, however, are often recognized or predicted by a comprehensive pharmacological analysis. We have therefore included in our preclinical toxicity studies a substantial number of pharmacological tests which are designed to recognize functional side effects, particularly those related to the autonomic nervous system. Class B. Toxic manifestations which are related to drug effects which require preexisting pathology or disease In patients with preexisting congenital or acquired abnormalities and diseases therapeutic drug effects may be potentiated or reversed and therefore cause toxic symptoms. The toxicity experiment in animals is inadequate to predict such events which may not become rec |