Testing for dependence liability in animals and man (Revised 1972)


I. Introduction
II. Method of submission of compounds for review by the Committee
III. Procedures employed in dependence studies
Primary physical dependence
The grading of drug effects
Criteria for identification of compounds with specific morphine-like properties in single-dose suppression tests
Criteria for the identification of compounds with morphine-antagonist properties in morphine-dependent monkeys
Single dose administration
Substitution for morphine
Primary physical dependence
IV. ADDENDUM 6- Adjuvant procedures in animals
Findley method 8
Wakeley method 8


Pages: 25 to 39
Creation Date: 1973/01/01

Testing for dependence liability in animals and man (Revised 1972)

By the Committee on Problems of Drug Dependence,

Division of Medical Sciences, National Academy of Sciences-National Research Council, Washington, D.C.,

21 November 1972

I. Introduction

In 1966, the Committee on Problems of Drug Dependence of the National Academy of Sciences-National Research Council prepared a document entitled: " Testing for Dependence Liability in Animals and Man " . [ 1] Since that time, the Committee's interest in dependence liability has continued to broaden, and the likelihood of abuse of a wide variety of substances which act on the central nervous system has similarly increased. Additionally, there have been significant developments in procedures for estimating dependence liability, and some changes have occurred with respect to the Committee's role in this problem. Thus, it is appropriate to revise the document that was published in 1966.

It is not the policy generally of the National Academy of Sciences-National Research Council to involve itself in testing programmes; however, a unique situation was encountered in drug dependence and drug abuse. The first Committee on Drug Addiction of the National Research Council, which existed under that name from 1929 to 1939, conducted its own chemical, pharmacological and clinical research programmes, and it was closely concerned with the development of research at the United States Public Health Hospital for the treatment of addicts at Lexington, Kentucky. This concern led to the establishment of the Addiction Research Center by the National Institute of Mental Health. It also led to the development of established procedures for the evaluation of morphine-like effects and physical dependence properties in man.

When the National Research Council's Committee was reorganized in 1947 as the " Committee on Drug Addiction and Narcotics ", it was given authority to accept funds and to support research on various aspects of drug addiction. One of the first projects to receive support was a study of physical dependence in the monkey at the University of Michigan. This, in turn, led to development of the basic evaluation programme for physical dependence capacity of potential analgesic agents.

It was recognized very early in connexion with the projects in man and the monkey that an impartial body was needed which would accept compounds to be tested and would also serve in the role of referee as to results of the tests. Thus, investigators would be relieved of any hint of bias or pressure from interested parties. With the concurrence of, and indeed at the request of, investigators at the Addiction Research Center and at the University of Michigan, this Committee of the National Research Council - now known as the Committee on Problems of Drug Dependence - accepted this role. This paper discusses how the Committee interprets and carries out its role. It also discusses the specific techniques employed by the co-operating laboratories.

The Committee on Problems of Drug Dependence provides assistance in the determination of drug dependence properties of all types. It also aids in the determination of the abuse liability of substances that act on the central nervous system, whether or not these substances are potentially medicinal agents. The Committee supports programmes for studies in animals, reviews the data on new compounds, makes recommendations with respect to tests in man, and advises on initiation of control procedures when, in its opinion, these are required. The Committee welcomes submission of compounds for evaluation and advises the producer on the results of this evaluation. No fees are involved. The Committee receives financial support from the pharmaceutical industry in the United States and other countries in basic support of its activities. Contributions from the pharmaceutical industry, whether of a new or a continuing nature, are both welcome and essential in the Committee's work. All communications with respect to the testing programme should be directed to the Committee's Executive Secretary. [ 2]

II. Method of submission of compounds for review by the Committee


  1. Time. A compound may be submitted at any time in the course of its development; however, if the results of evaluation are to be most helpful, submission should be at an early stage.

  2. Information required. The name of the compound and its structural formula should be submitted. A summary of physiochemical data, including the base or salt form, solubility, stability, and any limiting properties, should be provided. A summary of pharmacological data and an indication of the producer's specific interest and purpose with respect to the compound should also be included.

  3. Quantity. Generally, not less than one gramme ought to be provided as the initial sample; however, potency will obviously have major bearing on the amount required. If the substance is reputed to have analgesic activity, a portion will be used to test for this property by the mouse hot-plate method [ 3] in the Laboratory of Chemistry at the National Institutes of Health. This test insures that data are available by a uniform method with respect to this parameter for all compounds reviewed by the Committee.

    The quantity is 1-2 grammes for a compound with an analgesic potency approximately equal to that of morphine; however, the amount is subject to adjustment upward or downward according to the ratio of activity with that of morphine. If comparison only with some other standard than morphine is available, an initial sample of not less than 5 grammes should be submitted if possible.

  4. Coding and identification. Upon receipt of the compound in the office of the Committee on Problems of Drug Dependence, the compound is initially forwarded to the Laboratory of Chemistry of the National Institutes of Health where an NIH code number is assigned. The compound is thence forwarded solely under that number to the testing laboratory of the Department of Pharmacology, University of Michigan Medical Center. Reference to the compound is by code number only until all studies are completed. On the other hand, correspondence with the producer is by means of the producer's code number.

  5. Reporting. On occasion, interim reports are submitted to indicate the direction of the results. Interim correspondence may also be employed in consultation or with respect to extension of the tests to be performed. When all tests are completed, a report is sent to the producer. At the time this report is submitted, or shortly thereafter, it is expected that the producer will authorize revelation of the name and structure of the compound concerned. Prior to issuance of the final report, this information is held in confidence by the Committee. Subsequent to the producer's authorization, the nature of the compound and the results of the evaluation are included in the Committee's annual report. These data are also published in the Proceedings of the Committee's Annual Scientific Meeting and may be quoted following this publication.


  1. and 2. (Time and information required). These requirements are the same as those noted above in connexion with section A (Dependence of morphine type and related properties in the Rhesus monkey).

  2. Quantity. If comparison with a barbiturate has been made by another laboratory technique, this comparison may be used as a basis. For a compound with a sedative potency approximately equal to that of sodium barbital, not less than 5 grammes will be required.

  3. Coding and identification. These requirements are the same as those noted in connexion with section A (above) except that the reference to testing activity at the University of Michigan is not involved.

  4. Reporting. The statements with respect to reporting noted in connexion with section A (above) are applicable.


Techniques of self-administration have been developed in the Department of Pharmacology of the University of Michigan Medical Center as well as in other laboratories. These techniques have been extensively described elsewhere and are being studied intensively for attendant conditions and interpretations. Several approaches have been tried: ( a) making the drug available to the naive animal for spontaneous initiation and continuation of self-administration, ( b) response in terms of self-administration after a period of programmed administration; ( c) response of the animal to an alternative presentation (substitution) after sustained self-administration of a known dependence-producing drug has been established. A uniform procedure and method of evaluation is being worked out. When this is accomplished, the same conditions noted above with respect to sections A and B will apply.


  1. Time. Testing of compounds in man can be recommended only after the name and structure of the compound have been revealed, preliminary study for dependence properties in animals has been completed, and pharmacological data have been made available in accordance with pertinent regulations of the Food and Drug Administration.

  2. Information required. The producer must submit 15-20 copies of a report setting forth the nature and properties of the compound. The results of all pharmacological and clinical tests should be given, to include acute and chronic (not less than 3 months) toxicity data in several species. Toxicity information should include indication of the cause of death whenever possible. Information on respiratory, circulatory, behavioural and incidental effects should be provided.

    There must also be evidence that the compound has been administered to man, together with an indication of the compound's relative potency and side effects. Note should be included as to whether the actions of the drug are antagonized by specific opiate antagonists, e.g., nalorphine, levallorphan or naloxone. It is also desirable to include data on the rate and extent of dosage increase in chronic administration (tolerance development), cross-tolerance, and whether acute tolerance and physical dependence have been demonstrated in animals.

    The producer's report is distributed to the Chief of the Section of Clinical Pharmacology of the Addiction Research Center, Lexington, Kentucky and to members of the Committee on Problems of Drug Dependence. On this basis, a judgement is reached with respect to adequacy of information on the safety and effectiveness of the compound as well as on the desirability of tests in man at the Addiction Research Center. The Center is not bound by the Committee's judgement but prefers to be guided by it.

  3. Quantity. Generally, an amount of the order of 10-25 grammes will be required. The quantity and form in which the material is submitted are determined by direct correspondence with the Chief of the Addiction Research Center, and shipment of material is directly to his office.

  4. Sponsorship. Since these are clinical trials, the Food and Drug Administration requires submission of an Investigational New Drug Application (IND). The producer (the American counterpart if a foreign producer is involved) must assume responsibility for submitting the IND, viz., for applying to the Food and Drug Administration for approval of clinical trial. The producer assumes concomitant responsibility for providing necessary reports of the clinical trial to the Food and Drug Administration.

  5. Reporting. The Chief of the Section of Clinical Pharmacology, Addiction Research Center, will report on an interim basis. On completion of the clinical study, the results will be communicated to the Committee on Problems of Drug Dependence through the Executive Secretary. Notification to the producer will be made immediately by the Committee's Executive Secretary. Protocols on studies accomplished will be made available by the Chief of the Section of Clinical Pharmacology when required for compliance with regulations of the Food and Drug Administration.


Evaluation programmes in this area are urgently needed. The same principles will apply as noted above in connexion with section D.

III. Procedures employed in dependence studies


Initial evaluation is based upon the fact, well established in animals and man, that any chemical substance capable of completely suppressing all the specific signs of morphine abstinence is equally capable of creating physical dependence during chronic administration. In so far as the monkey and man respond similarly to these agents, the test has predictive value in estimating dependence liability for the human for most chemical classes and for narcotic analgesics with a high level of intrinsic activity.

The primary and principal objective is to ascertain the physical dependence capacity of a drug, i.e. its ability to suppress abstinence signs in the morphine-dependent monkey or the degree to which the total spectrum of signs is affected. Single dose suppression thus provide a measure of the physical dependence capacity of the test drug. The test initially is qualitative in nature; subsequently, quantitative studies may be undertaken.

Physical dependence may be characterized by degrees, viz.:

  1. High. The drug produces complete suppression of all abstinence signs with doses that reveal no other overt pharmacological effect.

  2. Intermediate. Complete suppression of all morphine abstinence signs is obtainable, but only with doses that elicit other pharmacological actions which are manifested by such signs as ataxia, stupor, muscarinic effects, etc.

  3. Low. Although some suppression of all signs may occur, attempts to achieve complete suppression with larger doses result in the intervention of toxic reactions, convulsions, coma, etc. before such complete suppression is achieved.

  4. None. The drug fails to produce specific suppression of morphine abstinence signs. Non-specific depressants may obscure individual signs.

Single dose suppression involves use of a colony of 50 to 100 morphine-dependent Rhesus monkeys. The animals are maintained in a state of physical dependence by the hypodermic administration of 3 mg/kg of morphine sulphate every 6 hours without interruption. After a stabilization period of 60 days, these monkeys can be used for testing at weekly intervals.

For testing, regular morphine injections are withheld until abstinence signs of intermediate intensity are present (14 hours). If left untreated, the intensity of abstinence will increase progressively over the ensuing several hours, but the administration of morphine or of any drug with morphine-like physical dependence capacity results in a partial or complete suppression of the signs of abstinence. If the dose is in excess of that required for complete suppression, typical signs of narcotic depression result.

A pre-selected quantity of the test drug is administered subcutaneously by someone other than the observer to 2 withdrawn monkeys. The intensity of abstinence signs, or of drug depression, is graded just prior to, and at intervals of ½, 1, 2, 3, et seq. hours after administration until the monkeys have returned to the pre-injection level of excitability. If the dose of the test drug produces more effect than the maintenance dose of morphine (3 mg/kg), the dose is halved; if less effect is produced, the dose is doubled for a subsequent test. The procedure is repeated until a dose is determined that is approximately equivalent to 3 mg/kg of morphine, or until side effects appear in such intensity as to preclude further testing. Thus, determination is reached as to the high, intermediate, low or no physical dependence capacity of the test drug.

Physical dependence capacity and single-dose suppression potency are not synonymous and may not even parallel each other in relation to morphine. Only the former is of major predictive significance in relation to dependence in man. A compound of relatively low potency, 20 mg equivalent to 3 mg/kg of morphine (ratio 0.15) may have a high physical dependence capacity, and in this respect be equal to morphine in its potential as a drug of abuse, only requiring large quantities to be effective.

Primary physical dependence

When it is deemed advisable (most often in cases of questionable significance of physical dependence capacity in single-dose tests), the ability of a drug to produce physical dependence is determined directly by chronic administration to monkeys which have not previously received morphine or a related agent. The purpose here is to determine the highest level of physical dependence that can be obtained with maximum tolerable doses of the test drug. Increasing doses are administered for at least 31 days at intervals (usually of 2 to 6 hours) that correspond to the drug's duration of action. A group of at least 3 monkeys is employed. The development of physical dependence is monitored at the 14th and 28th days of treatment by administering 2 mg/kg of nalorphine HCl subcutaneously. The administration of nalorphine precipitates an abstinence syndrome of one-half hour to four hours in duration. The severity of this syndrome corresponds to the degree of physical dependence developed at the time. After drug administration is terminated, the monkeys are observed for at least 10 days for signs of abstinence. From this experiment, physical dependence is graded. A low grade is inferred if only mild abstinence signs develop, an intermediate grade if the abstinence signs are of intermediate severity, and a high grade if severe or very severe abstinence signs result during withdrawal or after administration of nalorphine.

The grading of drug effects

In the single-dose suppression technique, effective doses of test drugs reduce the intensity of abstinence signs partially or completely. Alternatively, having completely suppressed the abstinence signs, the effective doses may then produce typical signs of narcotic depression. The abstinence signs and the depression are graded. The classification of abstinence signs being:

  1. Mild. The signs are apprehension, continual yawning, rhinorrhea, lacrimation, hiccup, shivering, perspiration on face, chattering, quarrelling and fighting. Signs in this category are liable to be overlooked or considered of no significance by the untrained observer.

  2. Intermediate. The signs are tremor, anorexia, pilomotor activity, muscle twitching and rigidity, and holding the abdomen (cramps).

  3. Severe. The signs are extreme restlessness, assumption of peculiar attitudes, vomiting, severe diarrhoea, erection and continued masturbation, inflammation of the eyelids and conjunctivae (insomnia), continual calling and crying, lying on side with eyes closed, and marked spasticity.

  4. Very severe. The signs are docility in the normally excitable animal, dyspnea, pallor, strabismus, dehydration, weight loss, prostration, circulatory collapse and, occasionally, death.

Criteria for identification of compounds with specific morphine-like properties in single-dose suppression tests

Whereas morphine will suppress all the manifestations of the morphine abstinence syndrome, other narcotic analgesics suppress some abstinence signs more readily than others. A group of signs, however, is always suppressed by dependence-producing analgesics. These include: behavioural hyper-irritability and hyperalgesia. Further, dependence-producing analgesics not only suppress behaviours that increase during morphine abstinence, but they also restore behaviours that decrease in abstinence, e.g., eating, climbing, non-aversive social interactions, etc. The actions of drugs that suppress only 1 or 2 signs, for example restlessness or emesis, are not interpreted as indications of specific morphine-like activity but as non-specific signs of central depressant effects or as anti-emetic properties. Barbiturates, benzodiazepines and phenothiazines do not restore the behaviour of morphine-abstinent monkeys to that shown by non-withdrawn animals. Further, these substances do not suppress the signs of behavioural hyperirritability and hyperalgesia when given in sub-anesthetic or sub-paralytic doses.


Morphine-like drugs with high dependence capacity have the following pharmacological properties: ( a) morphine-like CNS depression is produced by single doses; ( b) the CNS depressant effects can be antagonized by low doses of nalorphine and naloxone; ( c) tolerance and cross-tolerance can be demonstrated on chronic administration; ( d)the morphine abstinence syndrome is suppressed; ( e) physical dependence of the morphine type is produced on chronic administration; ( f) cross-dependence between all of these drugs can be demonstrated regardless of their chemical class; ( g) self administration behaviour is reinforced. All compounds with high physical dependence capacity have all of these properties. Not every known narcotic analgesic has been tested for all of these properties, but there appear to be no exceptions to the rule. One important implication of this pharmacologic generalization is that, when a compound is found to have high physical dependence potential, it is practically certain that the compound will have all of the other properties listed above. This includes the capacity to reinforce self-administration as a primary effect before physical dependence is produced.

A great deal of evidence indicates that the primary factor in compulsive self-administration is the behavioural reinforcing property of the drugs of dependence (psychologic dependence). Physical dependence plays a secondary, although a very strong, reinforcing role in the processes that maintain compulsive drug-seeking behaviour. On the other hand, the identification of strong morphine-like physical dependence capacity in a compound implies with a high degree of probability that the compound will also have primary reinforcing effects on self-administration behaviour. The great strength of the current programme of the Committee in evaluating the physical dependence capacity of morphine-like analgesics lies in the fact that the single-dose suppression technique is the most specific and most reliable procedure for identification of drugs with high levels of morphine-like intrinsic activity, viz., drugs that are certain to produce primary physical dependence and also to initiate compulsive self-administration behaviour. Because of this, there has never been a serious need to test each narcotic analgesic for "psychologic" dependence capacity.

Narcotic antagonists

The considerations noted above apply to narcotic analgesics but not to analgesics that have morphine-antagonist actions. There are at least 3 types of morphine antagonists: ( a) pure antagonists; ( b) antagonists with morphine-like agonist actions, viz., partial agonists; and ( c) antagonists with cyclazocine-like agonist actions. The antagonist analgesics constitute a heterogenerous group of substances, the full characterization of which demands a great deal of study. Unequivocal evaluation of the dependence potential of many of these compounds requires direct assessment of their capacity to produce physical dependence and to generate self-administration behaviour.

Morphine-dependent animals are the subjects that are most sensitive to the actions of morphine antagonists. Tests in morphine-dependent monkeys have made possible the discovery of antagonistic properties in compounds not otherwise suspected of having such activity. They have also made possible the determination of relative potencies with reference to standard antagonists, estimates of the duration of action of such antagonists, and the presence of other types of pharmacological activity.

Tests with morphine antagonists include: ( a) studies of the capacity of test drugs to produce a characteristic exacerbation of abstinence during single-dose suppression tests in morphine-dependent monkeys; ( b) dose-response studies of the abstinence-precipitating properties of these compounds in non-withdrawn morphine-dependent monkeys; ( c) studies of the ability of the test compound to reverse the depressant effects of single large doses of morphine or meperidine in non-dependent monkeys; ( d) studies of the possible presence of partial morphine-like agonist properties in the antagonists. Such studies determine whether or not these substances produce characteristic morphine-like effects, acute and chronic, in non-dependent animals. Further, pharmacologic analysis is undertaken to characterize the relative ceilings of morphine-like action of these substances, ( e) studies of the possible presence of cyclazocine-like agonist properties in the antagonists by determining whether or not they produce a characteristic pattern of CNS depression in non-dependent animals and whether or not they produce physical dependence of the cyclazocine type.

Criteria for the identification of compounds with morphine-antagonist properties in morphine-dependent monkeys

Just as narcotic analgesics other than morphine suppress abstinence in a manner different from that of morphine itself, so narcotic antagonists other than nalorphine precipitate acute abstinence syndromes that are somewhat different from those obtained with nalorphine. For example, many antagonists, particularly those with N-cyclopropylmethyl groups, are strong CNS depressants, and their administration to morphine-dependent monkeys produces a complex syndrome. In that syndrome, signs of precipitated abstinence occur in combination with such signs of central depression as decreased locomotor activity, decreased muscle tone, ataxia, and ptosis. Fortunately, however, a number of compounds the morphine-antagonist properties of which have been well established precipitate acute abstinence syndromes in the monkey. The syndrome is marked by: ( a) signs of behavioural hyper-irritability as noted above in the section on methods; ( b) signs of hyperalgesia, especially contact-induced aggression, persistent masturbation, bodily postures indicative of general malaise; and ( c) some autonomic signs such as salivation, retching and vomiting.

Only those compounds that are capable of reproducing at least all of the above signs of precipitated abstinence are judged to behave as morphine antagonists.

It is further required: ( a) that the severity of the precipitated syndrome be doserelated over at least a four-fold range of doses; and ( b) that the various signs corresponding to the different degrees of abstinence severity be present in roughly the same proportions as they occur after administration of nalorphine or levallorphan, or during morphine deprivation. Amphetamine-like CNS stimulants, psychotomimetics, and drugs with convulsant properties do not reproduce the picture of abstinence that is precipitated by antagonists in morphine-dependent monkeys.


As in the determination of morphine-like physical dependence capacity, it is assumed that any substance that will substitute for sodium barbital in a barbital-dependent dog, suppressing completely a barbiturate withdrawal syndrome, may be expected to produce dependence of barbiturate type during prolonged administration.

Dogs are made physically dependent on sodium barbital by the daily oral administration of a single dose of 100 mg/kg. After 6 weeks of this treatment, the drug is withdrawn for a period of 5 days, and the ensuing abstinence signs are observed. The abstinence syndrome is somewhat different in detail in each dog. After this 5-day period, barbital administration is resumed for at least one month. A test drug is then substituted in appropriate dose, and at appropriate intervals, for a period of 5 days. All treatment is then discontinued for 5 days, and the abstinence syndrome is observed and compared with the control withdrawal. Barbital treatment may be re-established and the dog used for another substitution in about another month.

The sedative potency of a test drug and its duration of action must be determined in normal dogs before substitution in dependent animals. The dose that is estimated to be equivalent in sedative potency to 30 mg/kg of sodium barbital is the dose used in substitution. Administration is at sufficiently frequent intervals so that no period intervenes between doses when the dog is free of the depressant effects of the test drug.

Dogs that are physically dependent on sodium barbital (100 mg/kg/day orally) begin to show mild abstinence signs about 20 hours after the last dose. The signs become progressively more intense, reaching a peak during the 72-96 hour period of withdrawal. The signs then diminish gradually in severity and disappear completely in about 10 days. The signs may be classified as:

  1. Mild. The signs are: Tachycardia (10-20 per cent), hyperpnea, weight loss (10-15 per cent), and tremors.

  2. Intermediate. The signs are: Anorexia, nervousness, restlessness, insomnia, and fighting.

  3. Severe. The signs are hyperthermia (1.5-2.0°C.), fasciculations, convulsions, and delirium.

Convulsions and delirium are the most dramatic and specific characteristics of the barbiturate abstinence syndrome. The other signs may be said to be less specific, since they may result from a variety of causes. On the other hand, when the lesser signs are associated with the withdrawal of a drug and are accompanied by convulsions and delirium, they may be considered to be specific and sensitive signs. The lesser signs are also much more uniform in their occurrence. Convulsions and delirium are more variable. Some dogs show convulsions as early as 12 hours after the last dose, but convulsions do not usually appear until after 48 hours. The convulsions may be localized, or they may spread to involve the entire body. Delirium appears in a variety of forms, but it is constant in appearance for a particular dog. Some dogs show catatonic behaviour, remaining in one position for many hours. Others snap and snarl at, or retreat from, non-existent objects.

Sedative action as determined in normal dogs, and substitution potency in the dependent dog, may not be related dosewise or in any other way. Sedative action is used as the starting point for substitution, and such substitution is begun 24 hours after the last dose of barbital. Thus, if abstinence signs appear, the dose is increased until these signs are completely suppressed or until toxic signs appear.

If complete substitution, i.e., complete suppression of the barbiturate abstinence syndrome, is attained at or near the comparable sedative dosage, it is probable that physical dependence will develop in man with the new agent under the same conditions as with a barbiturate. If a much higher dose is required for substitution, or if no suppression of abstinence occurs, the agent presumptively is one which produces less or no dependence of barbiturate type. Suppression of convulsions alone (of the abstinence syndrome) cannot be taken as evidence of adequate substitution. Sodium barbital at a dose of 50 mg/kg orally will prevent the convulsions without affecting other elements of the abstinence syndrome.


The subjects in the studies at the Addiction Research Center are former opiate dependent persons (addicts), male, in good physical condition, who are serving a sentence for violation of narcotic or other law and who have volunteered for the experiment. The objectives are: ( a) to determine the drug effects and the reaction of the subject to them in comparison with morphine or another drug in his experience; ( b) to determine the ability of the drug to substitute for morphine in a morphine-dependent person, thus suppressing the morphine abstinence syndrome, and ( c) to determine the ability of the drug to produce primarily tolerance and dependence on prolonged administration.

Single dose administration

Starting with small doses that are based on previous experience, a new drug is administered by various routes depending upon its solubility and other properties. The dose is increased, 2 or 3 men being given the drug at each dose level, until a definite effect is obtained or until signs of impending toxicity are seen.

To this point, administration is single blind, viz., the nature of the drug is unknown to the subject. The subject is, however, allowed to express his feelings about the drug in comparison with his previous drug experience. The new agent is compared with an appropriate standard, usually morphine, using a 4, 5 or 6 point assay and a placebo. Most commonly, a latin-square block design is used in which dose order is randomized. The block or experiment may contain 6 to 12 individuals.

Pupillary diameter is measured photographically. Respiratory rate, blood pressure, and body temperature are measured before and after each dose. Both patients and observers are required to fill out a single-dose questionnaire, and patients are required to complete a subjective drug-effect questionnaire. By analysis of signs and symptoms, these questionnaires make possible the calculation of the relative potency of the new agent in comparison with a standard. They also allow a qualitative comparison and the classification of the agent as an agonist of the morphine type or, alternatively, of the nalorphine type. A precise estimate of the relative potency of the drug with regard to a variety of parameters is made from the single-dose test. This will serve as a guide in the determination of the optimal dose level to be administered in direct addiction studies.

Substitution for morphine

A group of 5 to 12 individuals is given morphine sulphate subcutaneously and regularly, the dose being increased rapidly to 240 mg per day and stabilized at that level for not less than 30 days. Beginning with the evening dose (10 p.m.), and continuing through 24 hours, a coded substance (experimental drug, placebo, or the stabilization morphine dose are given as unknowns) is administered in place of the regular morphine doses.

Beginning the next morning, 14 hours after the last regular dose of morphine, observations are made at hourly intervals. The results are recorded and evaluated according to the Himmelsbach hourly score for abstinence phenomena.

Regular morphine administration is resumed at the end of the 24 hours, and the substitution procedure is repeated at weekly intervals until each individual has received in random order at least one dose of the experimental drug, a saline injection as a placebo, the regular stabilization dose of morphine, and frequently one or more doses of one or more other agents.

The mean hourly Himmelsbach point score for all patients is plotted for each coded medication, and comparison is made between the abstinence score for the experimental drug and morphine on the one hand (complete suppression of abstinence) and the placebo on the other hand (no suppression of abstinence). The purpose is to determine the abstinence suppression potency of the new drug relative to morphine.

Alternatively, especially if the new drug has not substituted satisfactorily for morphine, its substitution may be continued for 10 days with some dose adjustment aimed at attaining complete substitution. An impression is gained in this way of the individual's satisfaction with the substitute. At the end of the 10 day period, the new drug is replaced by placebo without the patient's knowledge. Through the 10 days and the following week, observations are made for morphine withdrawal phenomena. Ultimately, comparison is made qualitatively and quantitatively with an abrupt morphine withdrawal in the same individual.

Subjects dependent on 240 mg of morphine per day have served in a standard dependence regimen, but more recently it has been found that subjects who are dependent on lesser doses of morphine may serve a useful purpose in the classification of drugs. Thus, some of the partial agonists that will not substitute for morphine but rather will precipitate abstinence in subjects dependent on 240 mg of morphine per day, will substitute for morphine in subjects dependent upon 60 mg per day.

Subjects dependent on both 60 and 240 mg per day can be used to determine the relative potency of new agents in substituting for morphine and also for precipitating abstinence.

Primary physical dependence

If a compound has been shown to be quantitatively or qualitatively different from morphine, if it is a new chemical type, and - particularly - if the introduction of the compound into clinical practice is contemplated, chronic administration may be undertaken to determine, directly, tolerance and dependence development. Individuals previously dependent on morphine or a related drug and who are free of drugs for some months are given the new agent regularly. Intervals of administration, doses, and routes are those deemed appropriate in light of previous experience. On the basis of the potency determinations made in single-dose and substitution studies, attempts are made to accustom patients to a dose level of the new agent that is comparable to 240 mg of morphine.

The dose is increased slowly, giving the patient ample time to become tolerant or to adjust to undesirable side effects of the drug. When a maximal tolerated dose level is reached, or, alternatively, when a dose equivalent to 240 mg of morphine is reached, the patient is stabilized. Attempts to precipitate abstinence are made, using both nalorphine and naloxone.

Nalorphine will precipitate abstinence in patients dependent on drugs of the morphine type, but will substitute in patients who are dependent upon drugs of the nalorphine type. Naloxone, on the other hand, will precipitate abstinence in patients who are dependent on morphine-like drugs as well as in patients who are dependent on drugs that have characteristics of nalorphine. Following precipitation studies with nalorphine and/or naloxone, the patient is abruptly withdrawn. Signs and symptoms of abstinence are carefully analysed to determine whether the withdrawal syndrome most resembles dependence of the morphine type or dependence of the specific opiate antagonist type.

Two other procedures may be employed: A short-term cross-over repeated administration test or a patient's choice experiment. In the first of these, a series of medications, including morphine and a placebo, are administered to 8 drug-free, formerly morphine-dependent individuals. Administration is double blind, in random order, and in ascending dosage, each for 18 or 19 days. On the 19th or 20th day, unknown to the patient or observer, a placebo is substituted for the previous medication. Substitution is then continued for 10 days. At that point, the individual is started on another of the medications, and the whole procedure is repeated until each person has received each of the medications. Since no difficulty is experienced in increasing morphine dosage to 240 mg a day in 18 days, such a regimen of administration of morphine is employed. A parallel schedule of increasing dosage is employed for all other drugs, modified as necessary by slower development of tolerance, excessive sedation, or signs of toxic effects. The initial dose in each instance (placebo excepted) is that judged to be equivalent in effectiveness to morphine, based on the 24-hour substitution experiment.

In the patient's choice experiment, applicable to substances which may be given intravenously, each of 6 subjects is given at 3-day intervals a sample high dose of each medication to be compared, including morphine, on a single-blind basis. The patient is asked to rate the effects of the drugs in the order of his preference and to elect to take none, one, or any in increasing doses for 7 days. The patient has the further option that he can request discontinuance of a drug and/or further participation in the experiment at any time. At the end, the subject again rates the drugs in order of preference. The advantages of the procedure are that: ( a) it brings out rapidly among drugs characteristics that determine suitability for intravenous administration; ( b) it permits the subject to make selections among drugs as to whether and how long he would like to take one before trying another, ( c) it brings out differences among subjective effects and the subject's immediate like or dislike of a drug by his usually preferred route of administration. Thus, determination can be made whether or not the patient would be inclined to use the drug in this way if it were available to him.


A review of the data accumulated at the Addiction Research Center permits a judgement of the likelihood of tolerance and physical dependence development. Judgement can also be made on the probably subjective reaction to, and abuse liability of, compounds tested in the Center. In actual practice, these judgements have been confirmed in the past and are the basis for sound recommendations on the need for, and degree of, narcotics control.

IV. ADDENDUM [ 6] - Adjuvant procedures in animals

Efforts to develop dependence-free alternates for strong analgesics, sedatives (barbiturate-like), stimulants (amphetamine-like), etc., have led to the synthesis of compounds that have patterns of pharmacologic activity that differ, sometimes markedly, from the patterns of action of the respective prototypes. This has been especially striking with analgesics, and on many occasions the standard procedures to evaluate dependence potential have proven to be insufficient for clear characterization. This has been true both with regard to the evaluation of physical dependence capacity and estimation of psychologic dependence. Consequently, work continues on the development of new procedures that complement the standard studies in various ways, although the predictive value of these newer procedures is not yet clear.

Testing for dependence liability in animals and man 37


Mice. Repeated injections of morphine may be used to produce physical dependence, but a more rapid and convenient procedure is to implant a pellet of morphine base subcutaneously. After a few days, abstinence signs are observable upon removal of the pellet, but much more dramatic effects are elicitable using naloxone-precipitated withdrawal. The abstinence syndrome in the mouse includes defecation, urination, sniffing, increased motor activity, tremors, sometimes convulsions, and, most characteristically, stereotyped jumping. The jumping response can be used as an index for estimating the degree of dependence. This degree can be assessed without removing the pellet (the standard is a pellet containing 75 mg of morphine) by determining the amount of naloxone that is required to precipitate jumping from a circular platform. The median effective dose (ED 50) of naloxone is easily determined. The greater the degree of dependence, the lower is the ED 50 of naloxone. Alternatively, dependence intensity can be quantified by a graded response, viz., the total of jump attempts (average of a number of animals) after a fixed dose of naloxone.

Rats. After pellet implantation or repeated administration of morphine, naloxone promptly precipitates an abstinence syndrome. The signs of this syndrome, in order of appearance, include rearing on the hind legs, sniffing and exploratory behaviour, hypothermia, ear blanching, teeth chattering and "wet dog" shakes, attempts to escape from a jar, or stereotyped jumping from a platform, rhinorrhea, abnormal posturing, salivation, swallowing, vocalization on contact, ptosis, and diarrhea. A point-scoring system can be used to quantify the degree of dependence. Also, an all-or-none basis estimation can be made of the naloxone ED 50, preferably for more than one withdrawal sign. The more reliable indices are diarrhoea, abnormal posturing, ear blanching, swallowing, ptosis, teeth chattering, escape attempts, and "wet dog" shakes. The last two are highly characteristic, but they require higher doses of naloxone.

The signs after abrupt withdrawal in both mouse and rat are similar to those in precipitated withdrawal, but they are much slower in onset, less intense, and persist longer. A convenient procedure for assessment of the degree of dependence through abrupt withdrawal is to follow the body weight. Dependent animals lose considerable weight over a 24-hour period.


Procedures have been developed at the University of Michigan to complement the standard tests in those cases where there is a need for fine grain resolution of pharmacological activity or where drug side effects make interpretation of the results of standard tests questionable. These special studies have made use of objective behavioural measures of various sorts, including operant behaviour. Physiological measurements have also been used. In these, advantage has been taken of the marked changes in body temperature produced by narcotic withdrawal in the monkey as well as of the ability of narcotics to reverse the effects of morphine withdrawal. [ 7]

In recent years, other investigators have greatly expanded self-administration techniques in animals. These investigators have studied the capacity of a growing number of drugs to generate and maintain self-administration; and such research is particularly important in that it offers promise for pre-clinical identification of compounds that might produce psychological dependence in humans. This promise may exist, even though the conditions for extrapolation of results to prediction are not satisfactorily understood at present. This expanded effort proceeds along many lines. As examples: ( a) studies to determine whether or not the drug will generate steady self-administration and also to determine the strength of the drug's reinforcing activity; ( b) studies of the maximum work output that animals will exert to receive a drug injection; ( c) studies in which, following substitution of one drug for another, the preference of the animal is determined with respect to the two drugs or with respect to one drug as against a non-drug behavioural reinforcer, e.g., food, termination, or avoidance of electrical shocks, etc.

Findley method [ 8]

In this method, baboons are first prepared for self-administration by means of an in-dwelling venous catheter. They are chronically restrained within individual chambers and are subject to a continuous experimental programme that provides for: ( a) adaptation and training of subjects within their automated environment; ( b) establishment of initial drug dependency, or drug substitutions; ( c) experiments for determination of response costs per drug infusion which will reduce stable-state dependence to 25 per cent or less of the original intake levels.

Animals are allowed 24-hour access to food by means of a simple timing schedule that is programmed concurrently with the drug programme. The drug programme provides each subject with the opportunity to initiate a "trial" every 3 hours by means of an automated switch and stimulus light. Completion of a required number of responses (response cost) begins a "trial ". Each "trial" affords the animals a choice between two alternative stimulus conditions with equal work requirements but with different chemical consequences. Following a modest avoidance requirement in one stimulus, the drug is infused. Following parallel requirements in the other stimulus, saline is infused.

After an animal has been trained and stabilized with a given compound, the response cost to initiate a trial is increased each 7 days until the drug intake has been reduced to criterion levels. Dependencies established with different drugs necessitate differing response requirements for their abatement. Preliminary data suggest that these effects are independent of dosage variations provided that test conditions allow the establishment of tolerance to new doses and also provided that absolute levels of infusion are sufficient to maintain self-administration. Major dependent measures obtained on a daily basis are: ( a) food and water intake; ( b) absolute drug intake; ( c) frequency of initiated trials, and ( d) the relative frequency of choice as between the drug or the saline solution.

Wakeley method 8

In this method, monkeys are trained to press a lever to avoid electric shock and to self-administer drugs. The shock avoidance schedule is in effect for 7 hours. For the rest of the day, i.e. a period of 17 hours, lever-pressing behaviour is continuously recorded. Drugs can be characterized with respect to their ability to maintain (reinforce) lever-pressing and also by their effects on shock-avoidance behaviour. New drugs can be compared with reference drugs of dependence.

It must again be emphasized that many of the adjuvant procedures are complex, and interpretation of results demands caution. None of these procedures can be considered as routine at present, but their further use and study are worthwhile so that broader bases for prediction of dependence liability may be established.


1. Addendum 1 of Minutes, Committee on Problems of Drug Dependence, Division of Medical Sciences, National Academy of Sciences-National Research Council, Washington, D.C., U.S.A. Twenty-Eighth Meeting, New York, 9-11 February 1966. This article was reprinted in the Bulletin on Narcotics, XXII, 1, p. 11 (1970) with permission of the National Academy of Sciences-National Research Council.


2. Division of Medical Sciences, National Academy of Sciences-National Research Council, 2101 Constitution Avenue, N.W., Washington, D.C.20418. Telephone: (202) 961-1600.


3. When a compound is known, or suspected, to have properties of antagonism, it is also tested by a modification of the Nilsen procedure (cf. Perrine, T. D., Atwell, E. L., Tice, I. B., Jaconbson, A. E., and May, E. L., J. Pharm. Sci., 61, 86 (1972). This generally provides better predictiveness for man with drugs of the agonist-antagonist type.


4. No programme of this kind is presently sponsored by the Committee on Problems of Drug Dependence.


5. See footnote 4, p. 27.


6. The procedures described in the addendum are not a part of the evaluation programme presently sponsored and endorsed by the Committee on Problems of Drug Dependence. They are presented to indicate present trends in research for the estimation of drug dependence of various types. Their validity for man has not been proved.


Holtzman, S. G., and Villarreal, J. E., Morphine Dependence and Body Temperature in Rhesus Monkeys. J. Pharmacol. Exper. Ther. 166, 125-133 (1969); Holtzman, S. G. and Villarreal, J. E., Pharmacologic Analysis of the Hypothermic Responses of the Morphine-Dependent Rhesus Monkey. J. Pharmacol. Exper. Ther., 177, 317-325 (1970).


This method was developed under contract with the Bureau of Narcotics and Dangerous Drugs of the U.S. Department of Justice.