Overview of Safety Pharmacology
UNIT 10.1
Sonia Goineau,1 Martine Lemaire,1 and Guillaume Froget1 1
Porsolt S.A.S., Le Genest-Saint-Isle, France
ABSTRACT Safety pharmacology entails the assessment of the potential risks of novel pharmaceuticals for human use. As detailed in the ICH S7A guidelines, safety pharmacology for drug discovery involves a core battery of studies on three vital systems: central nervous (CNS), cardiovascular (CV), and respiratory. Primary CNS studies are aimed at defining compound effects on general behavior, locomotion, neuromuscular coordination, seizure threshold, and vigilance. The primary CV test battery includes an evaluation of proarrhythmic risk using in vitro tests (hERG channel and Purkinje fiber assays) and in vivo measurements in conscious animals via telemetry. Comprehensive cardiac risk assessment also includes full hemodynamic evaluation in a large, anesthetized animal. Basic respiratory function can be examined in conscious animals using whole-body plethysmography. This allows for an assessment of whether the sensitivity to respiratory-depressant effects can be enhanced by exposure to increased CO2 . Other safety pharmacology topics detailed in this unit are the timing of such studies, ethical and animal welfare issues, and statistical evaluation. C 2013 by John Wiley & Sons, Inc. Curr. Protoc. Pharmacol. 63:10.1.1-10.1.8. Keywords: safety pharmacology r risk assessment r ICH S7A guidelines r central nervous system r cardiovascular system r respiratory system
INTRODUCTION Safety pharmacology is a drug discovery and development discipline that falls between the fields of classical toxicology and pharmacology (Fossa, 1994). Whereas toxicologists examine chemical-induced changes that most typically result from repeated administration of the test compound at supratherapeutic doses, safety pharmacologists catalog changes that occur after acute administration of a chemical agent at doses approximating those to be used clinically. A first attempt to define safety pharmacology is contained in the Japanese Guidelines for Nonclinical Studies of Drugs Manual (Japanese Ministry of Health and Welfare, 1995). The manual provides considerable details on the types of pharmacological studies considered essential for testing a compound in humans. The European Medicines Agency proposed a new set of principles for safety guidelines within the International Conference on Harmonization of Technical Requirements for Registration of Pharmaceuticals for Human Use (ICH) (http://www.ich. org/products/guidelines/safety/article/safetyguidelines.html). The ICH S7A guidelines deal with core battery studies that include the assessment of compound effects on the cardiovascular, central nervous, and respiratory
systems, whereas the ICH S7B guidelines focus more specifically on proarrhythmic risk assessment related to QT interval prolongation (U.S. Food and Drug Administration, 2001, 2005) (Table 10.1.1). The ICH steering committee approved the S7A and S7B guidelines in 2000 and 2001, respectively, and then incorporated them into regional regulations. More recently, specific concerns have arisen about the drug abuse and dependence liability of agents that cross the blood–brain barrier. This has been the subject of considerable discussion among national agencies, even though, to date, there have been no ICH recommendations concerning this issue. This unit describes a number of important issues relating to safety pharmacology as it is currently defined by regulatory agencies.
TERMINOLOGY Safety pharmacology is a discipline that aims to provide an integrated assessment of data that relate to the risks associated with the medicinal use of a new chemical entity (Pugsley et al., 2008). A source of confusion regarding the definition of safety pharmacology arises from the use of various terms to categorize the
Current Protocols in Pharmacology 10.1.1-10.1.8, December 2013 Published online December 2013 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/0471141755.ph1001s63 C 2013 John Wiley & Sons, Inc. Copyright
Safety Pharmacology/ Toxicology
10.1.1 Supplement 63
Table 10.1.1 Required and Additional Tests in Safety Pharmacology Studiesa
Cardiovascular system In vitro electrophysiology studies (single-cell or multicellular preparations) Telemetry (blood pressure, heart rate, and electrocardiogram) Systemic, cardiac, pulmonary, and renal hemodynamic evaluation in large animals Orthostatic hypotension (tilting test) Langendorff-perfused isolated heart Central nervous system Irwin Locomotor activity Motor coordination Barbital sleep time Convulsant/proconvulsant risk Abuse/dependence liability assessment (for drugs crossing the blood–brain barrier) Respiratory system Plethysmography Gastrointestinal system Gastrointestinal transit Gastric emptying Gastric secretion Ulcerogenic potention Emesis induction Renal/urinary system Urine volume and electrolyte excretion a Required
Overview of Safety Pharmacology
tests are in bold text. Other tests are considered as additional.
types of studies covered by the discipline. Some terms used to describe the work conducted by safety pharmacologists include general, ancillary, secondary, high-dose, and regulatory pharmacology. Moreover, experiments performed to define the mechanisms of any adverse effects noted during toxicological studies are often considered part of safety pharmacology. In its strictest sense, the word safety implies the absence of untoward effects that might endanger the patient. Thus, the term safety pharmacology could be applied to all pharmacological studies undertaken to ensure the absence of adverse effects when a drug candidate is administered in a manner, and over a dose range, that is clinically relevant. Only studies that provide some prediction of risk should be part of a safety pharmacology assessment. The primary aim of such studies is to demonstrate that, at doses thought to be appropriate for obtaining the intended therapeutic benefit, there are no other effects that could
be considered risk factors for the patient. A further aim is to determine the maximum dose that could be administered before encountering adverse events. Such studies should also be useful for establishing a bridge between therapeutic doses and those used for toxicological studies, and for determining the maximum doses that can be safely administered during Phase I human studies. While the expression high-dose pharmacology would appear relevant in this regard, it is too restrictive in that it does not include the notion that the drug candidate might have adverse effects on other systems, even at therapeutic doses. In contrast, the terms general or ancillary pharmacology encompass all studies undertaken to characterize probable therapeutic responses to the new substance. The aim of general pharmacology studies is to determine the selectivity of the drug candidate for the intended purpose. For example, a novel anticancer compound would not usually be expected to display psychotropic activity,
10.1.2 Supplement 63
Current Protocols in Pharmacology
although there might be a therapeutic advantage if the drug had antidepressant or anxiolytic effects. Studies aimed at exploring such issues would come under the category of general pharmacology, since proof of safety is not their intended purpose.
SAFETY PHARMACOLOGY VERSUS TOXICOLOGY Traditionally, safety pharmacology studies have been conducted in a toxicological context. Toxicology differs from pharmacology in that toxicologists mainly investigate structure, whereas pharmacologists mainly investigate function (Sullivan and Kinter, 1995). Although there is an entire area of study linking morphological with functional changes and defining their relationship to adverse events that can be detected during toxicological studies, several potentially adverse responses may not be detected unless specific pharmacological studies are performed. Furthermore, changes in physiological function may occur in the absence of changes in organ structure and may frequently occur at doses lower than those necessary to induce a structural change. Moreover, not all structural modifications are clearly associated with a detectable change in function. Although structural and functional changes are sometimes clearly related, it is not always possible to link them in terms of cause and effect. Thus, safety pharmacology and classical toxicology are complementary disciplines, with both providing information needed for determining the safety of a new substance. While toxicology studies aim to define maximal tolerated dose and to determine adverse effects following chronic administration, safety pharmacology studies evaluate pharmacodynamic effects after acute administration up to subtoxic doses. Today, the complementarity of these two types of risk assessment studies is best illustrated by the increase in the number of safety pharmacological endpoints in classical toxicological studies (Redfern et al., 2013).
COMPARISON OF JAPANESE AND ICH GUIDELINES The Japanese guidelines have their origins in a September 1967 notification issued by the Japanese Ministry of Health and Welfare concerning “basic policies for approval to manufacture drugs.” The 1995 version of the Japanese guidelines (Japanese Ministry of Health and Welfare, 1995) do not specifically mention safety pharmacology. Instead they de-
scribe general pharmacology as having the following aims: 1. to assess the overall profile of general pharmacological effects as compared with principal pharmacological effects, 2. to “obtain useful information on potential adverse effects,” and 3. to evaluate “effects of drugs on physiological functions not necessarily detectable in toxicological studies.” The ICH S7A guidelines specifically mention safety pharmacology and define it as “those studies that investigate the potential undesirable pharmacodynamic effects of a substance on physiological functions in relation to exposure in the therapeutic dose-range and above.” The ICH definition has the considerable advantage of restricting safety pharmacology to the assessment of risk, thereby eliminating a wide range of studies that would be needed if the definition were more general. Reducing the number of studies decreases drug development costs, which is an advantage to both the drug developer and the consumer. The Japanese guidelines differ from the ICH S7A guidelines in that they make more specific recommendations concerning the safety pharmacology tests to be employed. These are divided into Category A and Category B. Category A includes essential evaluations, whereas Category B studies are those that are to be conducted “when necessary.” For CNS evaluations, Category A includes general behavioral observations, measures of spontaneous motor activity, general anesthetic effects—including the assessment of potential synergism/antagonism with general anesthetics—proconvulsant effects, analgesia, and effects on body temperature. Category B includes effects on electroencephalographic recordings, and effects on the spinal reflex, the conditioned avoidance response, and locomotor coordination. Category A also includes requirements for studying effects on the cardiovascular and respiratory systems (e.g., respiration, blood pressure, blood flow, heart rate, and electrocardiographic recordings), the digestive system (including gastrointestinal transit and gastric emptying), water and electrolyte metabolism (e.g., urinary volume, urinary concentrations of sodium, potassium, and chloride ions) and “other important pharmacological effects.” The three vital systems that the European guidelines (that is, the ICH guidelines) include in their core battery studies are the CNS, cardiovascular (CV) system, and the respiratory system. Core battery CNS studies include
Safety Pharmacology/ Toxicology
10.1.3 Current Protocols in Pharmacology
Supplement 63
motor activity, behavioral changes, coordination, sensory/motor reflex responses, and body temperature, with the remark that “the CNS should be assessed appropriately.” A similar declaration is made with regard to CV assessment, with specific mention of blood pressure, heart rate, and electrocardiographic recordings, together with a suggestion that “in vivo, in vitro and/or ex vivo evaluations, including methods for repolarization and conductance abnormalities, should also be considered.” Comparison of the Japanese guidelines with the ICH S7A guidelines makes clear the intent of the ICH guidelines to free safety pharmacology from the constraints of a cookbook approach. On the other hand, they are vague, which causes evaluations concerning what could or should be performed to be uncertain. This is most apparent in the recommendations for follow-up studies. For the CNS, the ICH indicates that such studies should include “behavioral pharmacology, learning and memory, ligand-specific binding, neurochemistry, visual, auditory and/or electrophysiology examinations, etc.” The subject of drug dependence/abuse, although a major concern for many agents with CNS effects, receives only a one-word mention in the section titled Other Organ Systems. The ICH guidelines are much more explicit for the CV and respiratory systems. Indeed, the entire ICH S7B section is devoted to an analysis and recommendations for arrhythmogenic risk. The ICH S7A guidelines specifically mention the renal/urinary, autonomic, and gastrointestinal systems. Somewhat surprisingly, no mention is made of nausea, despite the fact that it is one of the more common adverse events associated with drug administration.
nent in the assessment of CNS safety. Several substances, including antipsychotics such as clozapine, do not induce frank convulsions at any dose, but clearly decrease the convulsive threshold. The use of dog electroencephalograms (EEGs) monitored by telemetry to evaluate proconvulsant risk is also a relatively simple, convenient, and practical alternative to the more commonly used rodent models (D¨urm¨uller et al., 2007). Even anticonvulsant activity, which in itself is not a risk factor, could be a predictor of cognition-impairing effects. Several anticonvulsants, such as benzodiazepines and NMDA receptor antagonists, are known to impair cognition. Thus, anticonvulsant activity could represent a useful initial screen for potential cognition-impairing effects. Likewise, sleep-inducing or sleepattenuating activity could be unmasked with a barbiturate interaction procedure. While some benzodiazepines—for example the anxiolytic clobazam—do not themselves induce definitive sleep, their sleep-enhancing activity is readily detected by studying their interaction with barbiturates. The same is true for psychostimulants, which may or may not induce signs of excitation in a primary observation procedure, but clearly block barbiturate-induced sleep. The ICH S7A guidelines initially placed abuse/dependence liability assessment under the heading of secondary or follow-up safety studies. Thereafter, both the EMA and FDA issued guidelines recommending that all drug candidates known to cross the blood–brain barrier should be evaluated for this property by biochemical analysis and/or behavioral responses. Nonclinical evaluation is performed using self-administration paradigms for abuse liability and withdrawal tests for dependence (Moser et al., 2011).
CNS SAFETY PHARMACOLOGY
Overview of Safety Pharmacology
The ICH S7A guidelines recommend that core battery CNS studies include measures of drug-induced signs of CNS dysfunction as well as measures of spontaneous locomotion and motor coordination. Three other measurements, originally recommended by Category A in the Japanese guidelines but dropped from the ICH S7A guidelines, are convulsive threshold, interaction with hypnotics, and effects on pain threshold (Porsolt et al., 2005). In spite of their exclusion from the ICH S7A guidelines, such measures are useful in a core battery of CNS safety pharmacology procedures (Porsolt et al., 2007). Decreases in the convulsive threshold are an important compo-
CARDIOVASCULAR SAFETY PHARMACOLOGY The ICH S7B guidelines, which deal exclusively with the evaluation of proarrhythmic risk, recommend that core battery cardiovascular studies include measures of drug candidateinduced effects on arterial blood pressure, heart rate, and electrocardiographic recordings. It is also suggested that “in vivo, in vitro and/or ex vivo evaluations, including methods for repolarization and conductance abnormalities, should also be considered.” Other surrogate biomarkers, e.g., TRIaD (triangulation, reverse-use-dependence, instability and
10.1.4 Supplement 63
Current Protocols in Pharmacology
dispersion of ventricular repolarization), are known to accurately estimate drug-candidaterelated torsadogenic risk (Picard et al., 2011). Recommended methodologies for assessing these risks include measurements of ionic currents in isolated animal or human cardiac myocytes and in cultured cell lines or heterologous expression systems with cloned human channels. Other assays include the measurement of action potential (AP) parameters in isolated cardiac preparations, analysis of specific electrophysiological parameters indicative of AP duration in anesthetized animals, and electrocardiographic measurements in conscious or anesthetized animals. The hERG (human ether-`a-go-go-related gene) channel assay is now considered to be the in vitro model of choice for cardiac proarrhythmic risk assessment. Whereas hERG channel assays, which use binding techniques and automated technology, are appropriate as a first screen for QT prolongation in the early stages of safety evaluation, the hERG channel patch-clamp technique (UNIT 10.8) is recommended for the core battery of cardiovascular studies. Although more time consuming, the patch-clamp technique is an indicator of function, as opposed to receptor affinity, and lends itself more readily to compliance with Good Laboratory Practice (GLP) (see Internet Resources, http://www.oecd.org), which is mandatory for the ICH S7A-recommended core battery studies. Nevertheless, this test may fail to identify substances with an indirect effect on potassium currents, which inhibit hERG protein trafficking from the endoplasmic reticulum to the cell surface (Kuryshev et al., 2005). Some of these indirect effects may be detected following long-term exposure to test substances in vitro. The hERG channel assay cannot be used as a stand-alone in vitro test for evaluating proarrhythmic risk. Calcium agonists, for example, lengthen the AP duration and favor the occurrence of early afterdepolarizations and/or delayed afterdepolarizations—either of which can lead to torsades de pointe. The hERG channel assay cannot detect cardiac risk related to this calcium-dependent mechanism. In addition, the hERG channel assay can frequently overestimate the cardiac risk for a new substance, since in a Purkinje fiber preparation, partial inhibition of the potassium ion channel conductance (IKr ) may not result in AP prolongation because of counteracting effects on other cardiac ion channels. For this reason, the Purkinje fiber assay (UNIT 11.3) is a necessary
adjunct for investigating the effects of a test substance on the different AP parameters. Whichever in vitro assay that is employed cannot fully mimic the in vivo situation. Because all in vitro data must be considered in the context of plasma protein binding (UNIT 7.5), pharmacokinetic parameters (UNIT 7.1), and anticipated plasma concentrations of the test substance, in vivo telemetric analyses in conscious animals remain an essential component for assessing proarrhythmic risk. Nonetheless, telemetry obtained from conscious animals does not constitute a standalone technique, as it provides little information on the mechanism responsible for any observed effect. A comprehensive systemic, cardiac, pulmonary, and renal hemodynamic evaluation in a large animal, such as the dog, is essential for an adequate evaluation of cardiovascular risk. Other risk factors, such as drug candidate– induced depression of myocardial contractility or pulmonary hypertension, are critically important, even in the absence of other cardiac electrical disorders. Improvements in telemetry techniques have increased the number of parameters that can be measured, such as left ventricular blood pressure for cardiac contractility. The number of cardiovascular parameters that can be examined in an acute hemodynamic study in the anesthetized animal yields a wealth of information regarding the mechanisms responsible for cardiovascular effects (Lacroix and Provost, 2000). Orthostatic hypotension, which is a common cardiovascular risk, is not covered by the ICH S7A guidelines. As this constitutes a major adverse effect associated with many different drug classes, it is important to determine whether it is a property of a drug candidate. The most widely used experimental model for evaluating orthostatic hypotension is the tilting test in the anesthetized rat (Hashimoto et al., 1999). With this assay, orthostatic hypotension is exacerbated by prazosin and β-adrenoceptor antagonists. While inclusion of such a test in a cardiovascular study does not constitute a major expense, it provides a more complete assessment of cardiovascular risk. Although it is obvious that proarrhythmic effects represent a major cardiovascular danger for new drug candidates, the excessive focus on this risk by the ICH S7B guidelines, together with the avalanche of reports dealing with methodological issues, have drawn attention away from the many other types of drug-induced cardiovascular dangers, such as
Safety Pharmacology/ Toxicology
10.1.5 Current Protocols in Pharmacology
Supplement 63
orthostatic hypotension, pulmonary hypertension, and valvular heart disease.
RESPIRATORY SAFETY PHARMACOLOGY
Overview of Safety Pharmacology
Drug candidates can cause changes in pulmonary function (UNIT 10.9) by direct actions on the respiratory system or as a consequence of central, metabolic (alterations in acid-base balance), or vascular (pulmonary hypertension) effects. The ICH S7A guidelines include respiration as a vital function that must be assessed during safety evaluations. The method of choice for examining whether a test substance affects airway function is whole-body plethysmography in the unrestrained animal (e.g., rat, guinea pig, mouse), using the same route of administration as in conventional toxicology studies. A number of ventilatory parameters are measured with this assay, including inspiratory and expiratory times, peak inspiratory and expiratory flows, minute volume, tidal volume, respiratory rate, relaxation time, pause, and enhanced pause. This makes it possible to differentiate between effects on respiratory control and on the mechanical properties of the lung (Murphy, 2002). As a general screen, the whole-body test is preferable to the head-out method because the animals have freedom of movement and so they can be studied over longer periods of time. A weakness of whole-body plethysmography, however, is that it is insensitive to the respiratory-depressant effects of some drugs, such as barbiturates and opioids. Attempts to increase the sensitivity of this assay by placing the animals in a CO2 enriched environment (Van den Hoogen and Colpaert, 1986; Gengo et al., 2003; Goineau et al., 2010) appear promising. Although the anesthetized dog preparation does not lend itself to the evaluation of spontaneous lung function, it is well suited for evaluating the risk of pulmonary hypertension (as mentioned above). For this reason, the anesthetized dog is an important component in a comprehensive respiratory safety evaluation. Methods currently available for measuring respiratory parameters in the conscious dog include the use of a plethysmograph chamber or a facemask equipped with a pneumotachograph attached to the snout of the animal (Murphy, 2005). The animals are restrained, however, and the measurements are performed over a short period of time. An alternative is to equip conscious dogs with respiratory belts placed around the thorax and abdomen and to use a jacketed external telemetry system (Murphy et al., 2010).
This allows for the simultaneous recording of cardiovascular and respiratory parameters in conscious, nonrestrained dogs.
TIMING OF SAFETY PHARMACOLOGY STUDIES AND GOOD LABORATORY PRACTICE An issue that is not clearly addressed in either the Japanese or ICH guidelines is the timing of safety pharmacology studies. Whereas the Japanese guidelines imply that safety data are required for marketing approval, the ICH S7A guidelines clearly state that they are needed prior to initiating Phase 1 clinical trials. Both the Japanese and ICH guidelines require that Category A or core battery studies be performed according to GLP. Many pharmaceutical companies exceed these regulatory requirements; they engage in safety pharmacology studies early in the drug discovery process, even at the very beginning of in vivo experiments (Sullivan and Kinter, 1995). A primary observation procedure, such as the Irwin test (UNIT 10.10), is frequently the first assessment in living animals for determining acute toxicity, the active dose range, and the principal effects on behavior and physiological function. In addition, very early in the discovery process, hERG procedures are often used to screen optimized lead molecules for potential proarrhythmic risk (UNIT 10.8). Such early safety screening is rarely conducted according to GLP, and therefore falls outside the requirements of regulatory agencies. Nonetheless, information gained from such tests is vital in informing the discovery program and the selection of clinical candidates. Early in the discovery process, safety studies are performed differently than they are for Phase 1 approval. An early-stage Irwin test typically employs fewer animals, the mouse rather than the rat, and examines doses selected sequentially on the basis of effects observed with previous doses to determine as quickly as possible the maximal tolerated dose. With later-stage Irwin tests, the dose range is fixed in advance, beginning with the lowest dose approximating the therapeutic dose, followed by multiples of 10 or 100, up to, but not including, a lethal dose. Likewise, early-stage hERG tests (especially non-GLP screening) employing a binding assay or a patch-clamp procedure might involve just a single high concentration of several compounds within the same chemical series rather than a range of concentrations, as would be the case in a later hERG evaluation aimed at determining an IC50 value. In
10.1.6 Supplement 63
Current Protocols in Pharmacology
addition, for GLP in vitro studies, evaluation of exposure in the test system is necessary to allow an accurate and reliable determination of safety margins. To obtain meaningful data, it is mandatory that pharmacokinetic studies be conducted on drug candidates entering safety pharmacology studies. It should also be noted that the experimental aims of the early and late tests are different. For early-stage work, usually non-GLP screening, the objective is to detect the presence of risk, which is used as a guide in the selection of lead candidates for development. Later-stage analysis, which is conducted in accordance with GLP guidelines for regulatory submission, is performed to confirm the absence of risk in the relevant dose range for the selected compound.
positives, or erroneous detections of possible risk, although bothersome, are less serious and can usually be corrected with supplementary testing. Thus, it is important to minimize the risk of false negatives, even if this increases the risk of false positives. Based on preclinical studies, a test substance that is found to not have significant safety risks, even after the use of oversensitive statistics, is more likely to be truly devoid of risk. Consequently, safety pharmacology, in contrast to efficacy pharmacology, should employ statistical procedures that possess maximal sensitivity for detecting possible effects on a dose-by-dose basis, even though this might increase the possibility of false positives.
ETHICAL AND ANIMAL WELFARE ISSUES
CONCLUSIONS
As with all procedures involving living animals, there are important ethical issues that must be taken into consideration when designing experimental protocols. The guiding principles are to use as few animals as possible and to minimize their suffering and discomfort. Since the goal of safety pharmacology is to assess the risk of side effects, the possibility of causing suffering in experimental animals is higher than in other areas of pharmacology. The investigator must remain sensitive to these issues, not only in planning and designing the experiments, but also during their performance. For example, procedures for terminating the experiment in the event of welldefined events, such as pain or death, must be established. Ethical issues must be considered within the context of the aims of the experiments, which, ultimately, are to minimize human risk. While reducing risk to humans is of paramount importance, it is still possible to devise scientifically valid experiments using a small number of laboratory animals. For example, it is now accepted that the traditional LD50 acute toxicity test, which requires the use of a large number of subjects, yields only limited information. The Irwin test obtains considerably more information from fewer animals (UNIT 10.10).
STATISTICAL EVALUATION Since identification of risk is the chief aim of a safety pharmacology test, it is essential that positive results not be overlooked, and that the generation of false negatives be kept to a minimum (Porsolt et al., 2005). False
Thanks in part to the ICH S7A guidelines, safety pharmacology can now be considered an independent discipline that falls between traditional toxicology and efficacy/discovery pharmacology. Safety pharmacology is, however, a pharmacological rather than a toxicological discipline, since it concerns the study of drug actions on physiological function rather than on physical/anatomical structure. Although it employs identical methods, safety pharmacology differs from efficacy pharmacology in that the former evaluates the potentially adverse effects of test substances on normal function, whereas the latter is aimed at establishing therapeutic potential. Both provide information critical for drug discovery and development.
LITERATURE CITED D¨urm¨uller, N., Guillaume, P., Lacroix, P., Porsolt, R.D., and Moser, P. 2007. The use of the dog electroencephalogram (EEG) in safety pharmacology to evaluate proconvulsant risk. J. Pharmacol. Toxicol. Methods 56:234-238. Fossa, A.A. 1994. Inaugural conference on general pharmacology/safety pharmacology. Drug Dev. Res. 32:205. Gengo, P.J., Pettit, H.O., O’Neill, S.J., Su, Y.F., McNutt, R., and Chang, K.J. 2003. DPI-3290 [(+)-3-((α-R)-α-((2S,5R)-4-Allyl-2, 5-dimethyl-1-piperazinyl)-3-hydroxybenzyl)-N -(3-fluorophenyl)-N-methylbenzamide]. I. A mixed opioid agonist with potent antinociceptive activity and limited effects on respiratory function. J. Pharmacol. Exp. Ther. 307:12211226. Goineau, S., Rompion, S., Guillaume, P., and Picard, S. 2010. Ventilatory function assessment in safety pharmacology: Optimization of rodent studies using normocapnic or hypercapnic conditions. Toxicol. Appl. Pharmacol. 247:191-197.
Safety Pharmacology/ Toxicology
10.1.7 Current Protocols in Pharmacology
Supplement 63
Hashimoto, Y., Ohashi, R., Minami, K., and Narita, H. 1999. Comparative study of TA-606, a novel angiotensin II receptor antagonist, with losartan in terms of species difference and orthostatic hypotension. Jpn. J. Pharmacol. 81:63-72. Japanese Ministry of Health and Welfare. 1995. Japanese Guidelines for Nonclinical Studies of Drugs Manual 1995: Guidelines for Toxicity Studies of Drugs: Guidelines for General Pharmacology Studies: Guidelines for Nonclinical Pharmacokinetic Studies. Pharmaceutical Affairs Bureau, Japanese Ministry of Health and Welfare. Yakugi Nippo, Japan. Kuryshev, Y.A., Ficker, E., Wang, L., Hawryluk, P., Dennis, A.T., Wible, B.A., Brown, A.M., Kang, J., Chen, X.L., Sawamura, K., Reynolds, W., and Rampe, D. 2005. Pentamidine-induced long QT syndrome and block of hERG trafficking. J. Pharmacol. Exp. Ther. 312:316-323. Lacroix, P. and Provost, D. 2000. Basic safety pharmacology: The cardiovascular system. Th´erapie 55:63-69. Moser, P., Wolinsky, T., Duxon, M., and Porsolt, R.D. 2011. How good are current approaches to nonclinical evaluation of abuse and dependence? J. Pharmacol. Exp. Ther. 336:588-595. Murphy, D.J. 2002. Assessment of respiratory function in safety pharmacology. Fund. Clin. Pharmacol. 16:183-196. Murphy, D.J. 2005. Comprehensive non-clinical respiratory evaluation of promising new drugs. Toxicol. Appl. Pharmacol. 207:414-424. Murphy, D.J., Renninger, J.P., and Schramek, D. 2010. Respiratory inductive plethysmography as a method for measuring ventilatory parameters in conscious, non-restrained dogs. J. Pharmacol. Toxicol. Methods 62:47-53. Picard, S., Goineau, S., Guillaume, P., Henry, J., Hanouz, J.L., and Rouet, R. 2011. Supplemental studies for cardiovascular risk assessment in safety pharmacology: A critical overview. Cardiovasc. Toxicol. 11:285-307. Porsolt, R.D., Picard, S., and Lacroix, P. 2005. International safety pharmacology guidelines (ICH S7A and S7B): Where do we go from here? Drug Dev. Res. 64:83-89. Porsolt, R.D., Castagn´e, V., D¨urm¨uller, N., and Moser, P. 2007. CNS safety pharmacology. In Nonclinical Drug Safety Assessment: Practical Considerations for Successful Registration
(W.K. Sietsema and R. Schwen, eds.) pp. 141162. FDA News, Washington, D.C. Pugsley, M.K., Authier, S., and Curtis, M.J. 2008. Principles of safety pharmacology. Br. J. Pharmacol. 154:1382-1399. Redfern, W.S., Ewart, L.C., Lain´ee, P., Pinches, M., Robinson, S., and Valentin, J.-P. 2013. Functional assessments in repeat-dose toxicity studies: The art of the possible. Toxicol. Res. 2:209234. Sullivan, A.T. and Kinter, L.B. 1995. Status of safety pharmacology in the pharmaceutical industry—1995. Drug Dev. Res. 35:166-172. U.S. Food and Drug Administration. 2001. S7A safety pharmacology studies for human pharmaceuticals (ICH Guidance for Industry). U.S. Food and Drug Administration, Rockville, Md. U.S. Food and Drug Administration. 2005. S7B nonclinical evaluation of the potential for delayed ventricular repolarization (QT interval prolongation) by human pharmaceuticals (ICH Guidance for Industry). U.S. Food and Drug Administration, Rockville, Md. Van den Hoogen, R.H. and Colpaert, F.C. 1986. Respiratory effects of morphine in awake unrestrained rats. J. Pharmacol. Exp. Ther. 237:252259.
INTERNET RESOURCES http://www.fda.gov/downloads/Drugs/Guidance ComplianceRegulatoryInformation/Guidances/ UCM074959.pdf The FDA’s S7A safety guidelines for pharmacology studies of human pharmaceuticals. http://www.fda.gov/downloads/Drugs/Guidance ComplianceRegulatoryInformation/Guidances/ UCM074963.pdf The FDA’s S7B safety guidelines for pharmacology studies of human pharmaceuticals. http://www.ich.org/products/guidelines.html Contains links to both the ICH S7A (Safety Pharmacology Studies for Human Pharmaceuticals) and S7B (The Nonclinical Evaluation of the Potential for Delayed Ventricular Repolarization (QT Interval Prolongation) By Human Pharmaceuticals) safety guidelines. http://www.oecd.org The OECD Web site contains comprehensive information about GLPs.
Overview of Safety Pharmacology
10.1.8 Supplement 63
Current Protocols in Pharmacology
UNIT 10.1
Sonia Goineau,1 Martine Lemaire,1 and Guillaume Froget1 1
Porsolt S.A.S., Le Genest-Saint-Isle, France
ABSTRACT Safety pharmacology entails the assessment of the potential risks of novel pharmaceuticals for human use. As detailed in the ICH S7A guidelines, safety pharmacology for drug discovery involves a core battery of studies on three vital systems: central nervous (CNS), cardiovascular (CV), and respiratory. Primary CNS studies are aimed at defining compound effects on general behavior, locomotion, neuromuscular coordination, seizure threshold, and vigilance. The primary CV test battery includes an evaluation of proarrhythmic risk using in vitro tests (hERG channel and Purkinje fiber assays) and in vivo measurements in conscious animals via telemetry. Comprehensive cardiac risk assessment also includes full hemodynamic evaluation in a large, anesthetized animal. Basic respiratory function can be examined in conscious animals using whole-body plethysmography. This allows for an assessment of whether the sensitivity to respiratory-depressant effects can be enhanced by exposure to increased CO2 . Other safety pharmacology topics detailed in this unit are the timing of such studies, ethical and animal welfare issues, and statistical evaluation. C 2013 by John Wiley & Sons, Inc. Curr. Protoc. Pharmacol. 63:10.1.1-10.1.8. Keywords: safety pharmacology r risk assessment r ICH S7A guidelines r central nervous system r cardiovascular system r respiratory system
INTRODUCTION Safety pharmacology is a drug discovery and development discipline that falls between the fields of classical toxicology and pharmacology (Fossa, 1994). Whereas toxicologists examine chemical-induced changes that most typically result from repeated administration of the test compound at supratherapeutic doses, safety pharmacologists catalog changes that occur after acute administration of a chemical agent at doses approximating those to be used clinically. A first attempt to define safety pharmacology is contained in the Japanese Guidelines for Nonclinical Studies of Drugs Manual (Japanese Ministry of Health and Welfare, 1995). The manual provides considerable details on the types of pharmacological studies considered essential for testing a compound in humans. The European Medicines Agency proposed a new set of principles for safety guidelines within the International Conference on Harmonization of Technical Requirements for Registration of Pharmaceuticals for Human Use (ICH) (http://www.ich. org/products/guidelines/safety/article/safetyguidelines.html). The ICH S7A guidelines deal with core battery studies that include the assessment of compound effects on the cardiovascular, central nervous, and respiratory
systems, whereas the ICH S7B guidelines focus more specifically on proarrhythmic risk assessment related to QT interval prolongation (U.S. Food and Drug Administration, 2001, 2005) (Table 10.1.1). The ICH steering committee approved the S7A and S7B guidelines in 2000 and 2001, respectively, and then incorporated them into regional regulations. More recently, specific concerns have arisen about the drug abuse and dependence liability of agents that cross the blood–brain barrier. This has been the subject of considerable discussion among national agencies, even though, to date, there have been no ICH recommendations concerning this issue. This unit describes a number of important issues relating to safety pharmacology as it is currently defined by regulatory agencies.
TERMINOLOGY Safety pharmacology is a discipline that aims to provide an integrated assessment of data that relate to the risks associated with the medicinal use of a new chemical entity (Pugsley et al., 2008). A source of confusion regarding the definition of safety pharmacology arises from the use of various terms to categorize the
Current Protocols in Pharmacology 10.1.1-10.1.8, December 2013 Published online December 2013 in Wiley Online Library (wileyonlinelibrary.com). DOI: 10.1002/0471141755.ph1001s63 C 2013 John Wiley & Sons, Inc. Copyright
Safety Pharmacology/ Toxicology
10.1.1 Supplement 63
Table 10.1.1 Required and Additional Tests in Safety Pharmacology Studiesa
Cardiovascular system In vitro electrophysiology studies (single-cell or multicellular preparations) Telemetry (blood pressure, heart rate, and electrocardiogram) Systemic, cardiac, pulmonary, and renal hemodynamic evaluation in large animals Orthostatic hypotension (tilting test) Langendorff-perfused isolated heart Central nervous system Irwin Locomotor activity Motor coordination Barbital sleep time Convulsant/proconvulsant risk Abuse/dependence liability assessment (for drugs crossing the blood–brain barrier) Respiratory system Plethysmography Gastrointestinal system Gastrointestinal transit Gastric emptying Gastric secretion Ulcerogenic potention Emesis induction Renal/urinary system Urine volume and electrolyte excretion a Required
Overview of Safety Pharmacology
tests are in bold text. Other tests are considered as additional.
types of studies covered by the discipline. Some terms used to describe the work conducted by safety pharmacologists include general, ancillary, secondary, high-dose, and regulatory pharmacology. Moreover, experiments performed to define the mechanisms of any adverse effects noted during toxicological studies are often considered part of safety pharmacology. In its strictest sense, the word safety implies the absence of untoward effects that might endanger the patient. Thus, the term safety pharmacology could be applied to all pharmacological studies undertaken to ensure the absence of adverse effects when a drug candidate is administered in a manner, and over a dose range, that is clinically relevant. Only studies that provide some prediction of risk should be part of a safety pharmacology assessment. The primary aim of such studies is to demonstrate that, at doses thought to be appropriate for obtaining the intended therapeutic benefit, there are no other effects that could
be considered risk factors for the patient. A further aim is to determine the maximum dose that could be administered before encountering adverse events. Such studies should also be useful for establishing a bridge between therapeutic doses and those used for toxicological studies, and for determining the maximum doses that can be safely administered during Phase I human studies. While the expression high-dose pharmacology would appear relevant in this regard, it is too restrictive in that it does not include the notion that the drug candidate might have adverse effects on other systems, even at therapeutic doses. In contrast, the terms general or ancillary pharmacology encompass all studies undertaken to characterize probable therapeutic responses to the new substance. The aim of general pharmacology studies is to determine the selectivity of the drug candidate for the intended purpose. For example, a novel anticancer compound would not usually be expected to display psychotropic activity,
10.1.2 Supplement 63
Current Protocols in Pharmacology
although there might be a therapeutic advantage if the drug had antidepressant or anxiolytic effects. Studies aimed at exploring such issues would come under the category of general pharmacology, since proof of safety is not their intended purpose.
SAFETY PHARMACOLOGY VERSUS TOXICOLOGY Traditionally, safety pharmacology studies have been conducted in a toxicological context. Toxicology differs from pharmacology in that toxicologists mainly investigate structure, whereas pharmacologists mainly investigate function (Sullivan and Kinter, 1995). Although there is an entire area of study linking morphological with functional changes and defining their relationship to adverse events that can be detected during toxicological studies, several potentially adverse responses may not be detected unless specific pharmacological studies are performed. Furthermore, changes in physiological function may occur in the absence of changes in organ structure and may frequently occur at doses lower than those necessary to induce a structural change. Moreover, not all structural modifications are clearly associated with a detectable change in function. Although structural and functional changes are sometimes clearly related, it is not always possible to link them in terms of cause and effect. Thus, safety pharmacology and classical toxicology are complementary disciplines, with both providing information needed for determining the safety of a new substance. While toxicology studies aim to define maximal tolerated dose and to determine adverse effects following chronic administration, safety pharmacology studies evaluate pharmacodynamic effects after acute administration up to subtoxic doses. Today, the complementarity of these two types of risk assessment studies is best illustrated by the increase in the number of safety pharmacological endpoints in classical toxicological studies (Redfern et al., 2013).
COMPARISON OF JAPANESE AND ICH GUIDELINES The Japanese guidelines have their origins in a September 1967 notification issued by the Japanese Ministry of Health and Welfare concerning “basic policies for approval to manufacture drugs.” The 1995 version of the Japanese guidelines (Japanese Ministry of Health and Welfare, 1995) do not specifically mention safety pharmacology. Instead they de-
scribe general pharmacology as having the following aims: 1. to assess the overall profile of general pharmacological effects as compared with principal pharmacological effects, 2. to “obtain useful information on potential adverse effects,” and 3. to evaluate “effects of drugs on physiological functions not necessarily detectable in toxicological studies.” The ICH S7A guidelines specifically mention safety pharmacology and define it as “those studies that investigate the potential undesirable pharmacodynamic effects of a substance on physiological functions in relation to exposure in the therapeutic dose-range and above.” The ICH definition has the considerable advantage of restricting safety pharmacology to the assessment of risk, thereby eliminating a wide range of studies that would be needed if the definition were more general. Reducing the number of studies decreases drug development costs, which is an advantage to both the drug developer and the consumer. The Japanese guidelines differ from the ICH S7A guidelines in that they make more specific recommendations concerning the safety pharmacology tests to be employed. These are divided into Category A and Category B. Category A includes essential evaluations, whereas Category B studies are those that are to be conducted “when necessary.” For CNS evaluations, Category A includes general behavioral observations, measures of spontaneous motor activity, general anesthetic effects—including the assessment of potential synergism/antagonism with general anesthetics—proconvulsant effects, analgesia, and effects on body temperature. Category B includes effects on electroencephalographic recordings, and effects on the spinal reflex, the conditioned avoidance response, and locomotor coordination. Category A also includes requirements for studying effects on the cardiovascular and respiratory systems (e.g., respiration, blood pressure, blood flow, heart rate, and electrocardiographic recordings), the digestive system (including gastrointestinal transit and gastric emptying), water and electrolyte metabolism (e.g., urinary volume, urinary concentrations of sodium, potassium, and chloride ions) and “other important pharmacological effects.” The three vital systems that the European guidelines (that is, the ICH guidelines) include in their core battery studies are the CNS, cardiovascular (CV) system, and the respiratory system. Core battery CNS studies include
Safety Pharmacology/ Toxicology
10.1.3 Current Protocols in Pharmacology
Supplement 63
motor activity, behavioral changes, coordination, sensory/motor reflex responses, and body temperature, with the remark that “the CNS should be assessed appropriately.” A similar declaration is made with regard to CV assessment, with specific mention of blood pressure, heart rate, and electrocardiographic recordings, together with a suggestion that “in vivo, in vitro and/or ex vivo evaluations, including methods for repolarization and conductance abnormalities, should also be considered.” Comparison of the Japanese guidelines with the ICH S7A guidelines makes clear the intent of the ICH guidelines to free safety pharmacology from the constraints of a cookbook approach. On the other hand, they are vague, which causes evaluations concerning what could or should be performed to be uncertain. This is most apparent in the recommendations for follow-up studies. For the CNS, the ICH indicates that such studies should include “behavioral pharmacology, learning and memory, ligand-specific binding, neurochemistry, visual, auditory and/or electrophysiology examinations, etc.” The subject of drug dependence/abuse, although a major concern for many agents with CNS effects, receives only a one-word mention in the section titled Other Organ Systems. The ICH guidelines are much more explicit for the CV and respiratory systems. Indeed, the entire ICH S7B section is devoted to an analysis and recommendations for arrhythmogenic risk. The ICH S7A guidelines specifically mention the renal/urinary, autonomic, and gastrointestinal systems. Somewhat surprisingly, no mention is made of nausea, despite the fact that it is one of the more common adverse events associated with drug administration.
nent in the assessment of CNS safety. Several substances, including antipsychotics such as clozapine, do not induce frank convulsions at any dose, but clearly decrease the convulsive threshold. The use of dog electroencephalograms (EEGs) monitored by telemetry to evaluate proconvulsant risk is also a relatively simple, convenient, and practical alternative to the more commonly used rodent models (D¨urm¨uller et al., 2007). Even anticonvulsant activity, which in itself is not a risk factor, could be a predictor of cognition-impairing effects. Several anticonvulsants, such as benzodiazepines and NMDA receptor antagonists, are known to impair cognition. Thus, anticonvulsant activity could represent a useful initial screen for potential cognition-impairing effects. Likewise, sleep-inducing or sleepattenuating activity could be unmasked with a barbiturate interaction procedure. While some benzodiazepines—for example the anxiolytic clobazam—do not themselves induce definitive sleep, their sleep-enhancing activity is readily detected by studying their interaction with barbiturates. The same is true for psychostimulants, which may or may not induce signs of excitation in a primary observation procedure, but clearly block barbiturate-induced sleep. The ICH S7A guidelines initially placed abuse/dependence liability assessment under the heading of secondary or follow-up safety studies. Thereafter, both the EMA and FDA issued guidelines recommending that all drug candidates known to cross the blood–brain barrier should be evaluated for this property by biochemical analysis and/or behavioral responses. Nonclinical evaluation is performed using self-administration paradigms for abuse liability and withdrawal tests for dependence (Moser et al., 2011).
CNS SAFETY PHARMACOLOGY
Overview of Safety Pharmacology
The ICH S7A guidelines recommend that core battery CNS studies include measures of drug-induced signs of CNS dysfunction as well as measures of spontaneous locomotion and motor coordination. Three other measurements, originally recommended by Category A in the Japanese guidelines but dropped from the ICH S7A guidelines, are convulsive threshold, interaction with hypnotics, and effects on pain threshold (Porsolt et al., 2005). In spite of their exclusion from the ICH S7A guidelines, such measures are useful in a core battery of CNS safety pharmacology procedures (Porsolt et al., 2007). Decreases in the convulsive threshold are an important compo-
CARDIOVASCULAR SAFETY PHARMACOLOGY The ICH S7B guidelines, which deal exclusively with the evaluation of proarrhythmic risk, recommend that core battery cardiovascular studies include measures of drug candidateinduced effects on arterial blood pressure, heart rate, and electrocardiographic recordings. It is also suggested that “in vivo, in vitro and/or ex vivo evaluations, including methods for repolarization and conductance abnormalities, should also be considered.” Other surrogate biomarkers, e.g., TRIaD (triangulation, reverse-use-dependence, instability and
10.1.4 Supplement 63
Current Protocols in Pharmacology
dispersion of ventricular repolarization), are known to accurately estimate drug-candidaterelated torsadogenic risk (Picard et al., 2011). Recommended methodologies for assessing these risks include measurements of ionic currents in isolated animal or human cardiac myocytes and in cultured cell lines or heterologous expression systems with cloned human channels. Other assays include the measurement of action potential (AP) parameters in isolated cardiac preparations, analysis of specific electrophysiological parameters indicative of AP duration in anesthetized animals, and electrocardiographic measurements in conscious or anesthetized animals. The hERG (human ether-`a-go-go-related gene) channel assay is now considered to be the in vitro model of choice for cardiac proarrhythmic risk assessment. Whereas hERG channel assays, which use binding techniques and automated technology, are appropriate as a first screen for QT prolongation in the early stages of safety evaluation, the hERG channel patch-clamp technique (UNIT 10.8) is recommended for the core battery of cardiovascular studies. Although more time consuming, the patch-clamp technique is an indicator of function, as opposed to receptor affinity, and lends itself more readily to compliance with Good Laboratory Practice (GLP) (see Internet Resources, http://www.oecd.org), which is mandatory for the ICH S7A-recommended core battery studies. Nevertheless, this test may fail to identify substances with an indirect effect on potassium currents, which inhibit hERG protein trafficking from the endoplasmic reticulum to the cell surface (Kuryshev et al., 2005). Some of these indirect effects may be detected following long-term exposure to test substances in vitro. The hERG channel assay cannot be used as a stand-alone in vitro test for evaluating proarrhythmic risk. Calcium agonists, for example, lengthen the AP duration and favor the occurrence of early afterdepolarizations and/or delayed afterdepolarizations—either of which can lead to torsades de pointe. The hERG channel assay cannot detect cardiac risk related to this calcium-dependent mechanism. In addition, the hERG channel assay can frequently overestimate the cardiac risk for a new substance, since in a Purkinje fiber preparation, partial inhibition of the potassium ion channel conductance (IKr ) may not result in AP prolongation because of counteracting effects on other cardiac ion channels. For this reason, the Purkinje fiber assay (UNIT 11.3) is a necessary
adjunct for investigating the effects of a test substance on the different AP parameters. Whichever in vitro assay that is employed cannot fully mimic the in vivo situation. Because all in vitro data must be considered in the context of plasma protein binding (UNIT 7.5), pharmacokinetic parameters (UNIT 7.1), and anticipated plasma concentrations of the test substance, in vivo telemetric analyses in conscious animals remain an essential component for assessing proarrhythmic risk. Nonetheless, telemetry obtained from conscious animals does not constitute a standalone technique, as it provides little information on the mechanism responsible for any observed effect. A comprehensive systemic, cardiac, pulmonary, and renal hemodynamic evaluation in a large animal, such as the dog, is essential for an adequate evaluation of cardiovascular risk. Other risk factors, such as drug candidate– induced depression of myocardial contractility or pulmonary hypertension, are critically important, even in the absence of other cardiac electrical disorders. Improvements in telemetry techniques have increased the number of parameters that can be measured, such as left ventricular blood pressure for cardiac contractility. The number of cardiovascular parameters that can be examined in an acute hemodynamic study in the anesthetized animal yields a wealth of information regarding the mechanisms responsible for cardiovascular effects (Lacroix and Provost, 2000). Orthostatic hypotension, which is a common cardiovascular risk, is not covered by the ICH S7A guidelines. As this constitutes a major adverse effect associated with many different drug classes, it is important to determine whether it is a property of a drug candidate. The most widely used experimental model for evaluating orthostatic hypotension is the tilting test in the anesthetized rat (Hashimoto et al., 1999). With this assay, orthostatic hypotension is exacerbated by prazosin and β-adrenoceptor antagonists. While inclusion of such a test in a cardiovascular study does not constitute a major expense, it provides a more complete assessment of cardiovascular risk. Although it is obvious that proarrhythmic effects represent a major cardiovascular danger for new drug candidates, the excessive focus on this risk by the ICH S7B guidelines, together with the avalanche of reports dealing with methodological issues, have drawn attention away from the many other types of drug-induced cardiovascular dangers, such as
Safety Pharmacology/ Toxicology
10.1.5 Current Protocols in Pharmacology
Supplement 63
orthostatic hypotension, pulmonary hypertension, and valvular heart disease.
RESPIRATORY SAFETY PHARMACOLOGY
Overview of Safety Pharmacology
Drug candidates can cause changes in pulmonary function (UNIT 10.9) by direct actions on the respiratory system or as a consequence of central, metabolic (alterations in acid-base balance), or vascular (pulmonary hypertension) effects. The ICH S7A guidelines include respiration as a vital function that must be assessed during safety evaluations. The method of choice for examining whether a test substance affects airway function is whole-body plethysmography in the unrestrained animal (e.g., rat, guinea pig, mouse), using the same route of administration as in conventional toxicology studies. A number of ventilatory parameters are measured with this assay, including inspiratory and expiratory times, peak inspiratory and expiratory flows, minute volume, tidal volume, respiratory rate, relaxation time, pause, and enhanced pause. This makes it possible to differentiate between effects on respiratory control and on the mechanical properties of the lung (Murphy, 2002). As a general screen, the whole-body test is preferable to the head-out method because the animals have freedom of movement and so they can be studied over longer periods of time. A weakness of whole-body plethysmography, however, is that it is insensitive to the respiratory-depressant effects of some drugs, such as barbiturates and opioids. Attempts to increase the sensitivity of this assay by placing the animals in a CO2 enriched environment (Van den Hoogen and Colpaert, 1986; Gengo et al., 2003; Goineau et al., 2010) appear promising. Although the anesthetized dog preparation does not lend itself to the evaluation of spontaneous lung function, it is well suited for evaluating the risk of pulmonary hypertension (as mentioned above). For this reason, the anesthetized dog is an important component in a comprehensive respiratory safety evaluation. Methods currently available for measuring respiratory parameters in the conscious dog include the use of a plethysmograph chamber or a facemask equipped with a pneumotachograph attached to the snout of the animal (Murphy, 2005). The animals are restrained, however, and the measurements are performed over a short period of time. An alternative is to equip conscious dogs with respiratory belts placed around the thorax and abdomen and to use a jacketed external telemetry system (Murphy et al., 2010).
This allows for the simultaneous recording of cardiovascular and respiratory parameters in conscious, nonrestrained dogs.
TIMING OF SAFETY PHARMACOLOGY STUDIES AND GOOD LABORATORY PRACTICE An issue that is not clearly addressed in either the Japanese or ICH guidelines is the timing of safety pharmacology studies. Whereas the Japanese guidelines imply that safety data are required for marketing approval, the ICH S7A guidelines clearly state that they are needed prior to initiating Phase 1 clinical trials. Both the Japanese and ICH guidelines require that Category A or core battery studies be performed according to GLP. Many pharmaceutical companies exceed these regulatory requirements; they engage in safety pharmacology studies early in the drug discovery process, even at the very beginning of in vivo experiments (Sullivan and Kinter, 1995). A primary observation procedure, such as the Irwin test (UNIT 10.10), is frequently the first assessment in living animals for determining acute toxicity, the active dose range, and the principal effects on behavior and physiological function. In addition, very early in the discovery process, hERG procedures are often used to screen optimized lead molecules for potential proarrhythmic risk (UNIT 10.8). Such early safety screening is rarely conducted according to GLP, and therefore falls outside the requirements of regulatory agencies. Nonetheless, information gained from such tests is vital in informing the discovery program and the selection of clinical candidates. Early in the discovery process, safety studies are performed differently than they are for Phase 1 approval. An early-stage Irwin test typically employs fewer animals, the mouse rather than the rat, and examines doses selected sequentially on the basis of effects observed with previous doses to determine as quickly as possible the maximal tolerated dose. With later-stage Irwin tests, the dose range is fixed in advance, beginning with the lowest dose approximating the therapeutic dose, followed by multiples of 10 or 100, up to, but not including, a lethal dose. Likewise, early-stage hERG tests (especially non-GLP screening) employing a binding assay or a patch-clamp procedure might involve just a single high concentration of several compounds within the same chemical series rather than a range of concentrations, as would be the case in a later hERG evaluation aimed at determining an IC50 value. In
10.1.6 Supplement 63
Current Protocols in Pharmacology
addition, for GLP in vitro studies, evaluation of exposure in the test system is necessary to allow an accurate and reliable determination of safety margins. To obtain meaningful data, it is mandatory that pharmacokinetic studies be conducted on drug candidates entering safety pharmacology studies. It should also be noted that the experimental aims of the early and late tests are different. For early-stage work, usually non-GLP screening, the objective is to detect the presence of risk, which is used as a guide in the selection of lead candidates for development. Later-stage analysis, which is conducted in accordance with GLP guidelines for regulatory submission, is performed to confirm the absence of risk in the relevant dose range for the selected compound.
positives, or erroneous detections of possible risk, although bothersome, are less serious and can usually be corrected with supplementary testing. Thus, it is important to minimize the risk of false negatives, even if this increases the risk of false positives. Based on preclinical studies, a test substance that is found to not have significant safety risks, even after the use of oversensitive statistics, is more likely to be truly devoid of risk. Consequently, safety pharmacology, in contrast to efficacy pharmacology, should employ statistical procedures that possess maximal sensitivity for detecting possible effects on a dose-by-dose basis, even though this might increase the possibility of false positives.
ETHICAL AND ANIMAL WELFARE ISSUES
CONCLUSIONS
As with all procedures involving living animals, there are important ethical issues that must be taken into consideration when designing experimental protocols. The guiding principles are to use as few animals as possible and to minimize their suffering and discomfort. Since the goal of safety pharmacology is to assess the risk of side effects, the possibility of causing suffering in experimental animals is higher than in other areas of pharmacology. The investigator must remain sensitive to these issues, not only in planning and designing the experiments, but also during their performance. For example, procedures for terminating the experiment in the event of welldefined events, such as pain or death, must be established. Ethical issues must be considered within the context of the aims of the experiments, which, ultimately, are to minimize human risk. While reducing risk to humans is of paramount importance, it is still possible to devise scientifically valid experiments using a small number of laboratory animals. For example, it is now accepted that the traditional LD50 acute toxicity test, which requires the use of a large number of subjects, yields only limited information. The Irwin test obtains considerably more information from fewer animals (UNIT 10.10).
STATISTICAL EVALUATION Since identification of risk is the chief aim of a safety pharmacology test, it is essential that positive results not be overlooked, and that the generation of false negatives be kept to a minimum (Porsolt et al., 2005). False
Thanks in part to the ICH S7A guidelines, safety pharmacology can now be considered an independent discipline that falls between traditional toxicology and efficacy/discovery pharmacology. Safety pharmacology is, however, a pharmacological rather than a toxicological discipline, since it concerns the study of drug actions on physiological function rather than on physical/anatomical structure. Although it employs identical methods, safety pharmacology differs from efficacy pharmacology in that the former evaluates the potentially adverse effects of test substances on normal function, whereas the latter is aimed at establishing therapeutic potential. Both provide information critical for drug discovery and development.
LITERATURE CITED D¨urm¨uller, N., Guillaume, P., Lacroix, P., Porsolt, R.D., and Moser, P. 2007. The use of the dog electroencephalogram (EEG) in safety pharmacology to evaluate proconvulsant risk. J. Pharmacol. Toxicol. Methods 56:234-238. Fossa, A.A. 1994. Inaugural conference on general pharmacology/safety pharmacology. Drug Dev. Res. 32:205. Gengo, P.J., Pettit, H.O., O’Neill, S.J., Su, Y.F., McNutt, R., and Chang, K.J. 2003. DPI-3290 [(+)-3-((α-R)-α-((2S,5R)-4-Allyl-2, 5-dimethyl-1-piperazinyl)-3-hydroxybenzyl)-N -(3-fluorophenyl)-N-methylbenzamide]. I. A mixed opioid agonist with potent antinociceptive activity and limited effects on respiratory function. J. Pharmacol. Exp. Ther. 307:12211226. Goineau, S., Rompion, S., Guillaume, P., and Picard, S. 2010. Ventilatory function assessment in safety pharmacology: Optimization of rodent studies using normocapnic or hypercapnic conditions. Toxicol. Appl. Pharmacol. 247:191-197.
Safety Pharmacology/ Toxicology
10.1.7 Current Protocols in Pharmacology
Supplement 63
Hashimoto, Y., Ohashi, R., Minami, K., and Narita, H. 1999. Comparative study of TA-606, a novel angiotensin II receptor antagonist, with losartan in terms of species difference and orthostatic hypotension. Jpn. J. Pharmacol. 81:63-72. Japanese Ministry of Health and Welfare. 1995. Japanese Guidelines for Nonclinical Studies of Drugs Manual 1995: Guidelines for Toxicity Studies of Drugs: Guidelines for General Pharmacology Studies: Guidelines for Nonclinical Pharmacokinetic Studies. Pharmaceutical Affairs Bureau, Japanese Ministry of Health and Welfare. Yakugi Nippo, Japan. Kuryshev, Y.A., Ficker, E., Wang, L., Hawryluk, P., Dennis, A.T., Wible, B.A., Brown, A.M., Kang, J., Chen, X.L., Sawamura, K., Reynolds, W., and Rampe, D. 2005. Pentamidine-induced long QT syndrome and block of hERG trafficking. J. Pharmacol. Exp. Ther. 312:316-323. Lacroix, P. and Provost, D. 2000. Basic safety pharmacology: The cardiovascular system. Th´erapie 55:63-69. Moser, P., Wolinsky, T., Duxon, M., and Porsolt, R.D. 2011. How good are current approaches to nonclinical evaluation of abuse and dependence? J. Pharmacol. Exp. Ther. 336:588-595. Murphy, D.J. 2002. Assessment of respiratory function in safety pharmacology. Fund. Clin. Pharmacol. 16:183-196. Murphy, D.J. 2005. Comprehensive non-clinical respiratory evaluation of promising new drugs. Toxicol. Appl. Pharmacol. 207:414-424. Murphy, D.J., Renninger, J.P., and Schramek, D. 2010. Respiratory inductive plethysmography as a method for measuring ventilatory parameters in conscious, non-restrained dogs. J. Pharmacol. Toxicol. Methods 62:47-53. Picard, S., Goineau, S., Guillaume, P., Henry, J., Hanouz, J.L., and Rouet, R. 2011. Supplemental studies for cardiovascular risk assessment in safety pharmacology: A critical overview. Cardiovasc. Toxicol. 11:285-307. Porsolt, R.D., Picard, S., and Lacroix, P. 2005. International safety pharmacology guidelines (ICH S7A and S7B): Where do we go from here? Drug Dev. Res. 64:83-89. Porsolt, R.D., Castagn´e, V., D¨urm¨uller, N., and Moser, P. 2007. CNS safety pharmacology. In Nonclinical Drug Safety Assessment: Practical Considerations for Successful Registration
(W.K. Sietsema and R. Schwen, eds.) pp. 141162. FDA News, Washington, D.C. Pugsley, M.K., Authier, S., and Curtis, M.J. 2008. Principles of safety pharmacology. Br. J. Pharmacol. 154:1382-1399. Redfern, W.S., Ewart, L.C., Lain´ee, P., Pinches, M., Robinson, S., and Valentin, J.-P. 2013. Functional assessments in repeat-dose toxicity studies: The art of the possible. Toxicol. Res. 2:209234. Sullivan, A.T. and Kinter, L.B. 1995. Status of safety pharmacology in the pharmaceutical industry—1995. Drug Dev. Res. 35:166-172. U.S. Food and Drug Administration. 2001. S7A safety pharmacology studies for human pharmaceuticals (ICH Guidance for Industry). U.S. Food and Drug Administration, Rockville, Md. U.S. Food and Drug Administration. 2005. S7B nonclinical evaluation of the potential for delayed ventricular repolarization (QT interval prolongation) by human pharmaceuticals (ICH Guidance for Industry). U.S. Food and Drug Administration, Rockville, Md. Van den Hoogen, R.H. and Colpaert, F.C. 1986. Respiratory effects of morphine in awake unrestrained rats. J. Pharmacol. Exp. Ther. 237:252259.
INTERNET RESOURCES http://www.fda.gov/downloads/Drugs/Guidance ComplianceRegulatoryInformation/Guidances/ UCM074959.pdf The FDA’s S7A safety guidelines for pharmacology studies of human pharmaceuticals. http://www.fda.gov/downloads/Drugs/Guidance ComplianceRegulatoryInformation/Guidances/ UCM074963.pdf The FDA’s S7B safety guidelines for pharmacology studies of human pharmaceuticals. http://www.ich.org/products/guidelines.html Contains links to both the ICH S7A (Safety Pharmacology Studies for Human Pharmaceuticals) and S7B (The Nonclinical Evaluation of the Potential for Delayed Ventricular Repolarization (QT Interval Prolongation) By Human Pharmaceuticals) safety guidelines. http://www.oecd.org The OECD Web site contains comprehensive information about GLPs.
Overview of Safety Pharmacology
10.1.8 Supplement 63
Current Protocols in Pharmacology
