EurJ Clin Pharmacol (1991) 41:89-93 0031697091001800
EuropeanJournalof C r ~ : ~ ( ~ ( ~
@beecell@9 @ Springer-Verlag 1991
Special article Racemic therapeutics - ethical and regulatory aspects E. J. Ari~ns Received: April 26,1990/Acceptedin revised form: December 8, 1990
Summary. Racemic therapeutics are fixed ratio mixtures of stereoisomers to be regarded biologically as different compounds. Usually only one of the isomers fully contributes to the therapeutics action, and the other is often classifiable as "isomeric ballast". Due to differences in turnover and pharmacokinetics, the proportion of enantiomers (1:1 in the racemate) continuously changes in plasma. The implications of the neglect of stereoselectivity for various levels in the investigation of racemic drugs are discussed and summarized in Table 2. The fact is that clinical investigators, Ethical Committees and regulatory authorities have for decades accepted invalid pharmacokinetic data on some 25% of therapeutics. That those racemates remain in use make the benefit of and necessity for kinetics generally questionable. Exposure of patients to the "isomeric ballast" present in about 50% of the most commonly used drugs will probably containe for many decades. As a result of a change in attitude of the regulatory authorities, however, for new drugs the choice in future between the racemic therapeutic or the single isomeric ballast-free drug will largely be based on a critical evaluation of the chiral characteristics with regard to their therapeutic, toxicological and pharmacokinetic aspects. Key words: Racemates therapeutics; ethical aspects, isomers, adverse effects, chirality drug stereoselectivity Since the article "Stereochemistry a basis for sophisticated nonsense in pharmacokinetics and clinical pharmacology" appeared in the European Journal of Clinical Pharmacology 1984 [1] much attention has been paid to the pharmacological aspects of drug chirality [1-13]. There has been much less interest in the ethical implications and the problems caused to regulatory authorities in particular by the many racemic therapeutics [4]. In the recent past, however, there has been some discussion of these aspects, too, which makes it worthwhile to assess the present position with some emphasis on the ethical and social aspects.
Few scientists will deny that stereoselectivity in action is a common and important phenomenon, not only in pharmacology but in particular in pharmacotherapeutics. This holds true for dynamics - desired and undesired actions - and kinetics. In the latter it is particularly important in the rate of enzymatic conversion, transport by carriers and protein binding, and thus for distribution and elimination. Stereoisomers in biology should really be considered as essentially different chemical compounds. The distinction "active" and "inactive" isomer is misleading. The isomer most potent for a particular action, e.g. the desired effect, is called the "eutomer". The other isomer, which has little often active, is the "distomer", and it often contributes to the undesired actions, and so can never be regarded as entirely harmless. Isomers that do not contribute to the desired action are to be considered as unwanted by products, "isomeric ballast", in drugs. Problems arise with the many mixtures of stereoisomers. Composite Chiral Drugs (CCDs), such as racemates, which constitute _+25% of the therapeutics on the market. What is said for racemates here is mostly applicable also to CCDs in general. Two important parameters must be considered. 1) The "eudismic ratio" (ER), the ratio of the activities (dynamics) of the eutomer and distomer [15]. It usually differs for those components in the desired and undesired actions generated at different sites of action (e. g. receptors). It is specific for a particular racemate in relation to a particular action and only rarely has the value '1'. 2) The "eudismic proportion" (EP) is the proportion of the concentrations of eutomer and distomer ([eut]/[dist]) in, for instance, plasma [14]. It is of particular significance in kinetics. For a racemate, the eudismic proportion (EP) of the enantiomers is 1 by definition. In particular after absorption, due to the commonly different rates of metabolic conversion of the enantiomers, the EP in the body fluids gradually changes. Exceptionally, e.g. in case of rapid in vivo racemisation, the racemate is found in plasma. If elimination is dominant for the eutomer, the ER which may be i for the racemate in the tablet, with grad-
90 ually decrease in the plasma, e. g. in the case of verapamil [16] to about 1/4 ( + 80% distomer). If elimination of the distomer is dominant, the EP in plasma will increase, as in the case of nivaldipine [17] to about 3 ( + 75% eutomer). In fact, the composition, ER of the "racemate" in the plasma continuously changes until a complete steady state is reached for each of the stereoisomers. The time required differs, therefore, for the two isomers. Without chiral assays the eudismic proportion (EP) in plasma remains unknown. If there is extensive presystemic or first pass loss by metabolic conversion in the liver, equal "plasma concentrations of the racemate" obtained after oral and after parenteral administration will differ in EP, and so will be bioinequivalent [16]. The plasma concentration after application of a racemate, as measured by non-chiral assays, only gives information on the sum of the eutomer and distomer, and since the eudismic proportion changes with time, it will not be directly related to the response. The various pharmacokinetic constants, such as half-life, bioavailability, persistence, etc. derived from "racemate" concentrations based on non-chiral assays, can only be notional, and are comparable to such parameters as the age or body weight of a married couple. In the exceptional case when the eudismic ratio is 1, the eudismic proportion does not matter, but complications may still arise due to stereoselectivity in the conversion to active metabolites.
Clinical pharmacology of racemic therapeutics Neglet of stereoselectivity by reliance on non-chiral assays in clinical pharmacology of racemic therapeutics is far from exceptional. A large proportion of the publications on pharmacokinetics, and the related clinical evaluation of racemic drugs, not only was but still remains heavily biased [1-6]. Often the racemic nature of the drugs is not mentioned in the publications, although many, mostly recent, reviews on chirality of drugs [1-15] clearly illustrate stereoselectivity in pharmacodynamics and kinetics, and the broad spectrum of possibilities and implications thereof. Impressive tables of pharmacokinetic constants for racemates, calculated on the basis of erroneous plasma concentrations of "racemates" by means of computer-assisted curve fitting, and based on mathematical equations derived for and valid only for single compounds, are a striking characteristic in particular of pharmacokinetic publications in clinical pharmacology. Similarly, multi-compartment systems for racemates are derived by computer assisted fitting of postulated compartmental systems to the multiphasic decline in the concentration of the "racemate", again on basis of invalid pharmacokinetic equations. The eudismic proportion (EP), the composition of the "racemate", not only changes with time but also tends to differ in the various compartments. The number of phases in the decline of the plasma "racemate" concentration increases with the number of compounds (stereoisomers) involved. Kinetics of racemates studied with non-chiral assays thus readily lead to fictional results, both in the kinetic constants and the compartmental systems derived. Data so generated are
E. J. Ari~ns: Racemictherapeutics not realistic, and, depending on the extent to which the clinician relies on them, they may even be risky [11]. Studies using chiral assays brought out remarkable complications with racemic mixtures in relation to genetic polymorphism and species-, race- and age-associated differences in pharmacokinetics. Caldwell et al. [18] analysed the metabolic chiral inversion and dispositional enantio-selectivity of the 2-arylpropionic acid, NSAID's and their biological consequences with highly informative results. Metabolic conversion of enantiomers at different rates to reactive intermediates also has the consequence that positive results obtained with the microsome-dependent Ames test of a racemates can be misleading. The distomer may be the culpit. Also, racemic prodrugs with different rates of conversion of the pro-eutomer and the pro-distomer may, if studied by non-chiral assays, lead to confusing results. The major problems arise: 1) How to reduce undesirable exposure of patients to isomeric ballast in composite chiral drugs to the scientifically feasible, technologically possible and economically acceptable limit? 2) How to avoid reports in the literature of heavily biased data on racemic therapeutics due to neglect of chirality when chiral assays are essential. An argument often heard is "these are matters for the regulatory authorities". They do have their responsibilities with regard to admission to the market of drugs, including CCD's. Their decisions must be based on scientifically valid data. A scientist never has any excuse to violate basic principles in science by referring to some administrative authority. In any case, in the study of the pharmacokinetics of racemic drugs already on the market, and in Therapeutic Drug Monitoring, regulatory authorities are no generally involved. Scientists have their own specific responsiblity as individuals and as a group. It is their duty to raise an alarm, especially, in the case of widespread fundamental errors.
Ethical aspects of drug development, evaluation and use In relation to the problems just mentioned there are two basic principles to be emphasized: 1) "Exposure of nature (including man) and its environment to xenobiotics (including drugs) is only justified if the desirable actions adequately compensate for the undesirable actions and the never fully excluded risks." This implies, amongst other things: "Avoid isomeric ballast in therapeutics." 2) "In science the notion of impossibility or incapability to do things the proper way is no excuse to do them the wrong way." Protection of humans against the risks of exposure to xenobiotics, particularly drugs, in today's chemistry-impregnated society received some attention in the World Medical Association's 'Declaration of Helsinki' [19]. Under "Recommendations guiding medical doctors in biomedical research involving human subjects" it is stated under basic principles:
E. J. Ariens: Racemic therapeutics
91
Table 1. Racemic antiarrhythmics
Tocainidec Mexiletinec Encainide d
VPC-patients efficacy (%) B or PL 40 40 80
MVA 15 15 25
adverse-effect proarrhythmia (%)* VPCs MVA 2 3 2 3 >1 10
non-cardiac (%)
potential for organ toxicity
40a 40a 10b
+++ + -
VPC ventricular premature complexes; MVA malignant ventricular arrhythmias; B or PL Benign or Potentially Lethal. a) Gastrointestinal and neurological, b) blurred vision, c) enantiomeric differences in metabolic conversion, d) formation of active metabolites. * In-
dudes an estimate of serious and potentially life-threatening proarrhythmic effects only [20]. The eudismic ratios for the therapeutic action are close to 1. How the side-effects are distributed over the stereoisomers is unknown
1.1. "Biomedical research involving human subjects must confirm to generally accepted scientific principles and should be based on adequately performed laboratory and animal experimentation and on a thorough knowledge of the scientific literature."
books pharmacology, pharmacokinetics or toxicology are the principles of stereoselectivity, stereospecificity, stereochemistry etc. properly described, or even mentioned in the index. Pharmacokinetic parameters, particularly the half-life and bioavailability of racemic therapeutics calculated from plasma "racemate" concentrations measured by non-chiral assays are amply presented [4, 5]. What do they mean? The reader may consult the library and see for himself. G o o d m a n and Gilman 7th edition (1985), for example, presents 65 pages (1668-1733) of pharmacokinetic constants for therapeutics, roughly a quarter of which are racemates. There is no indication there of the role of chirality. With few exceptions the constants are based on plasma levels determined by non-chiral assays. Students should be spared the false idea that there really is someentity like "plasma concentrations and pharmacokinetic parameters of racemates". What is real are plasma concentrations etc. of the individual stereoisomers and the continuously changing eudismic proportion. For the hundreds of therapeutics registered in the past 20 years the regulatory authorities have required pharmacokinetic data, including information on the hundreds of racemates among them. These data with few exceptions have been based on non-chiral assays. What is the sense?
1.9. "In any research on human beings each potential subject must be adequately informed about the aims, methods, anticipated benefits and potential hasards of the study etc." This implies the "informed consent". 1.8. "Reports of experimentation not in accordance with the principles laid down in this declaration should not be accepted for publication." There is something to be learn there by the Ethical Committees (the Institutional Review Boards, IRBs) supervising evaluation of drugs in humans. If racemic therapeutics or therapeutic isomer mixtures are involved, unless stereospecific assays are used invalid data will be generated. Taking into account the involvement of patients, this is unethical and thus is not acceptable. Are the patients informed about this situation? Is there honest "informed consent"? In particular, hospital pharmacists and clinical pharmacologists are in vited to review their contributions to science, in particular in the clinical evaluation the many racemic therapeutics they deal with. This will increase awareness of the pitfalls of non-chiral assays in the study of CCDs. A simple and effective way to stop the still continuing flow of deficient data generated in the study of CCDs would be extension of the insructions to authors with an indication that "The composite character of drugs which are mixtures of stereoisomers must be brought to the attention of the reader. The prefix (RS)- or rac-, e.g. rac-propranolol, in the case of racemates and (Z/E)- or cis/trans- in the case of that type of isomerism is obligatory. The implications of the composite and chiral nature of the drugs for the interpretation of the data measured and the conclusions drawn must be made explicit" [3, 5, 6].
The source of wisdom
The persistent neglect of chirality and its implications for drug action comes from deep in our education. In hardly any of the highly recommended clinically orientated text-
Racemates as drug combinations
For fixed ratio combinations, for instance of an antihypertensive [3-blocker and a saluretic, the authorities require full information on both components in the mixture, as well as on the mixture as such. It would make sense to follow a similar approach in the case of mixtures of stereoisomers. Such information would make clear when the single eutomer, will often be the case or the racemic mixture, is to be preferred. It is necessary to take into account, even in the exceptional case, that there is some form of synergism between the stereoisomers, when their 1:1 proportion only rarely will be optimal. In those cases when the eudismic ratio for the therapeutic action is approximately 1, one must still take into account that this may not be the case for adverse effects that are not inherent to the mechanism of therapeutic action. Side-effects, such as allergic reactions, mutagenicity and embryotoxicity, but also, for example, intestinal or neurological side-effects of cardiovascular therapeutics may be distributed over the enantiomers in a different way
92
Table 2. Requirements in the evaluation of racemic and Composite Chiral Drugs (CCD's) 1) PharmaceuticaI industry The composite character of chiral drugs, such as racemates, and its implications should clearly be noted. Stereoselective analytical methods necessary for proper investigation of CCDs should be made available. 2) Investigators Proposals for research projects scientifically deficient due to neglect of the chiral nature of the drug(s) to be studied demand re-education of the scientist involved. 3) Funding institutions In judging requests for grants to study CCDs the need for chiral assays should not be overlooked. 4) Ethics Committees Generation of invalid data due to disregard of stereoselectivity in evaluation of the activity CCDs in studies on patients is not ethical. In such cases "informed consent" requires special attention. 5) Investigators should avoid Neglect of stereoselectivity in pharmacodynamic and -kinetic, studies of CCDs. Calculation of pharmacokinetic constants and derivation of multicompartment systems for CCDs based on illusory racemate concentrations using pharmacokinetic equations valid only for single agents single stereoisomers, and not for CCDs. Presentation for publication of research reports and manuscripts biased due to disregard of the implications of stereoselectivity in action of CCDs. 6) Editorial boards and referees of scientific journals Proper attention should be paid to the composite character of racemic drugs and CCDs and the awareness of the authors to the implications thereof. 7) Prize awarding agencies Promotion of poor science by awarding prizes etc., for example, for astereognosis in the study of CCDs must be prevented. 8) IUPHAR (International Union of Pharmacology). FIP (F6d6ration Internationale Pharmaceutique). APha (American Pharmaceutical Association). Might take steps to stop or at least to discourage the frequently neglect of stereoselectivity in the study of CCDs, one basis for substandard science. 9) Teachers and textbook-writers In educational programmes proper attention should be paid to chirality as an important aspect of drug action, particularly of CCDs. 10) Regulatory authorities Adaptation of the rules for data to be submitted for new CCDs, namely information on individual stereoisomers as well as the mixture. Re-evaluation of those CCDs already on the market with a relatively high incidence of side-effects on a similar basis is advisable.
from the desired action. In this respect it is important, to reevaluate racemic therapeutics with a high incidence of side-effects, such as tocainide and mexiletine, antiarrhythmics with a + 40% incidence of non-cardiac side-effects (Table 1; [20]). Requirements for the evaluation of racemic therapeutics are summarized in Table 2. T h e y are applicable to the 25% CCDs in use but are usually neglected. The literature is rich in examples of such neglect, e. g. most studies on [3adrenergic agents, ~-adrenoreceptor blockers, combined [3- and c~-adrenoreceptor blockers, calcium channel block-
E. J. Ari6ns: Racemic therapeutics ers and non-steroidal antiphlogistics, all drugs in frequent use. Most clinical pharmacokinetic studies on t h e m are deficient due to the indiscriminate use of non-chiral assays. There are recent informative and illustrative (1989) articles [21-23] on racemic calcium channel blockers presenting a b r o a d spectrum of kinetic constants. They clearly demonstrate where astereognosis leads to. It is characteristic of the situation that Article 21 even was awarded the "Pharmatec prize" for its outstanding scientific qualities! This was given under auspice of the F6deration Internationale Pharmaceutique (FIP). D a t a on changes in clinical parameters, such as blood pressure, cardiac output, liver function and etc. as a function of dose and time have their own intrinsic value for CCDs and fixed ration combinations of drugs in general. In this regard References 22 and 23 are correct. Pharmacokinetic data on CCDs based on non-chiral assays should be rejected as should concealment of the composite character of the drugs studied.
Perspectives The climate is changing. In 1987 Japan, in a leading position, adopted rational requirements for racemic products. Full pharmacology and toxicology is required for each of the enantiomers and for the racemate [24]. For the latter only pharmacokinetic data based on chiral assays should and probably would be accepted. The E u r o p e a n Economic Community (EEC) in "Rules governing medicinal products in the E u r o p e a n C o m m u n i t y " Volume II, 1989 extended the section on stereoisomerism. It states: "Possible problems relating to stereoisomerism, which should be discussed in the appropriate Expert R e p o r t and cross referenced, should include: the batch to batch consistency of the ratio of stereoisomers in the various batches used - the toxicological issues - the pharmacological aspects (including evidence on which stereoisomers have the desired pharmacological properties) - p h a r m a c o k i n e t i c s including information on the relative metabolism o f the stereoisomers - extrapolation of the pre-clinical data (paying particular attention to possible problems relating to species differences in handling of the stereoisomers) - the significant clinical issues." This at least opens the possibility of requiring essential information on each of the enantiomers or stereoisomers in general, as well as on the racemate or in general composite chiral products. Further, it is stated that: "where a mixture of stereoisomers has previously been m a r k e t e d and it is now proposed to m a r k e t a product containing only one isomer, full data on this isomer should be provided." The requirements for a single isomer (the eutomer) among new drug applications should in general be analogous to those for a single compound. The US authorities ( F D A ) have come to an approach comparable to that in Japan and the E E C to be effected in 1992. This change in policy will reduce the flow of heavily biased data on CCDs in the scientific literature as far as ned drugs are concerned. Unless the editorial boards of the scientific journals join the regulatory authorities in their wisdom, for the hundreds of CCDs already on the
93
E. J. Ari~ns: Racemic therapeutics m a r k e t the contestable non-chiral kinetics will remain with us for m a n y years to come. Therapeutics containing isomeric ballast should be withdrawn from the m a r k e t as soon as ballast-free preparations are available at an acceptable price. Often only a small fraction of the price paid in the pharmacy represents the cost of the drug substance in the product. The approach indicated should hold true for generics as well as for branded products. The resulting incentive for the innovative industry and the rapid development of chiral techniques in chemistry will help to reduce chemical ballast in drug therapy. The patient, if aware of it, would definitely appreciate this result. In conclusion: (I) Acceptance by regulatory authorities of composite chirat drugs, such as racemates, should, whenever possible, be based on proper information on the individual stereoisomers as well as the mixture. It is anyway preferable to avoid chemical ballast. The question to be considered is that of weighing the advantages against the disadvantages, including the never entirely absent risks of exposure to chemical ballast. II) The study of pharmacology, dynamics and in particular of the kinetics of CCDs, such as racemates, requires chiral assays. The investigator has the moral duty to inform the reader of the composite nature of any drug studied and its implication for the interpretation of data. III) Ethical Committees amongst others have the duty to implement i n clinical research the recommendations of the World Medical Association "Declaration of Helsinki". IV) Teachers are invited to discuss with their pupils, or at least bring to their attention, the arguments advanced here. As an exercise in open minded reasoning they might be asked to write - "in the light of the forgoing a short c o m m e n t on the books on pharmacology, therapeutics, pharmacokinetics and biopharmacy r e c o m m e n d e d to them". The p r o b l e m affects about 25% of the available therapeutics, including 50% of those most commonly used. The recognition of these problems had to wait for hundred years after Pasteur generated the basic insights [25]. It should not take a further hundred years to eliminate astereognosis from general and clinical pharmacology.
References 1. Ariens EJ (1984) Stereochemistry, a basis for sophisticated nonsense in pharmacokinetics and clinical pharmacology. Eur J Clin Pharmacol 26:663-668 2. Ariens EJ, Soudijn W, Timmermans PBMWM (1982) Stereochemistry and biological activity of drugs. Blackwell Scientific Publications, Oxford, pp 1-190 3. Series on Chirality, Trends in Pharmacological Sciences (1986) 7: 20-24, 60-65,112-115,155-158, 200-205, 227-230, 281-301 4. Ariens EJ, Wuis EW (1987) Bias in pharmacokinetics and clinical pharmacology. Clin Pharmacol Ther 42:361-363
5. Ari~ns E J, Wuis EW, Veringa EJ (1988) Stereoselectivity of bioactive xenobiotics. A pre-Pasteur attitude in medicinal chemistry, pharmacokinetics and clinical pharmacology. Biochem Pharmacol 37:9-18 6. Ari~ns EJ (1990) Stereoselectivity in pharmacodynamics and pharmacokinetics. Schweiz Med Wochenschr 120:131-134 7. Drayer DE (1986) Pharmacodynamic and pharmacokinetic differences between drug enantiomers in humans: an overview. Clin Pharmacol Ther 40:125-133 8. Wainer IW, Drayer DE (1988) Drug stereochemistry. Analytical methods and pharmacology. Dekker, New York 9. Jamali F, Mehvar R, Pasutto FM (1989) Enantioselective aspects of drug action and disposition: Therapeutic pitfalls. J Pharm Sci 78:695-715 10. Smith DF (1989) The stereoselectivity of drug action. Pharmacol Toxicol 65:321-331 11. Evans AM (1988) Stereoselective drug disposition: potential for misinterpretation of drug disposition data. Br J Clin Pharmacol 26:771-780 12. Ari~ns EJ, v. Rensen JJS, Welling W (1988) Stereoselectivity of pesticides - biological and chemical problems. Elsevier, Amsterdam 13. Eichelbaum M, Gross S (1990) Stereoselectivity in drug action and disposition. N Engl J Med 14. Ari~ns EJ (1991) Chirality and isomeric ballast in bio-active products: Proceedings 3rd Netherlands Biotechnology Congress Amsterdam, in press. Elsevier, Amsterdam 15. Lehmann FPA, Rodriques de Miranda JF, Ari~ns EJ (1976) Stereoselectivity and affinity in molecular pharmacology. In: Jucker E (ed) Vo120. Birkhguser, Basel, pp 101-142 16. Vogelgesang B, Echizen H, Schmidt E, Eichelbaum M (1984) Stereoselective first-pass metabolism of highly cleared drugs: studies of the bioavailability of L- and D-verapamil examined with a stable isotope technique. Br J Clin Pharmacol 18:733-740 17. Tokuma Y, Fujiwara T, Niwa 1",.Hashimoto T, Noguchi H (1989) Stereoselective disposition of nivaldipine, a new dihydropyridine calcium antagonist, in the rat and dog. Res. Comm. Chem Pathol Pharmacol 63:24%262 18. Caldwell J, Hutt AJ, Fournel-Gigleux S (1988) The metabolic chiral inversion and dispositional enantioselectivity of the 2arylpropionic acids and their biological consequences. Biochem Pharmacol 37:105-114 19. Levine RJ (1986) Ethics and regulation of clinical research, 2nd ed. Urban and Schwarzenberg, Baltimore, pp 393-429 20. Morganroth J (1987) New antiarrhythmic agents: Mexiletine, tocainide, flecainide, encainide and amiodarone. Rational Drug Therapy. Am Soc Pharm Exp Ther 21:1-5 21. Soons PA, de Boer AG, van Brummelen R Breimer DD (1989) Oral absorption profile of nitrendipine in healthy subjects: a kinetic and dynamic study. Br J Clin Pharmaco127:17%189 22. Harten J van Burggraaf J, Lichthart GJ, Brummelen P van, Breimer DD (1989) Single- and multiple-dose nisoldipine kinetics and effects in the young, the middle-aged, and the elderly. Clin Pharmacol Ther 45:600-607 23. Dunselman PHJM, Edgar B, Scaf AHJ, Kuntze CEE, Wesseling H (1989) Pharmacokinetics of felodipine after intravenous and chronic oral administration in patients with congestive heart failure. Br J Clin Pharmaco128: 45-52 24. Smith RL, Caldwell J (1988) Racemates towards a New Year solution? TIPS 9:75-77 25. Pasteur L (1901) On the asymmetry of naturally occurring organic compounds, the foundations of stereochemistry. In: Richardson GM (ed) Memoirs by Pasteur, Van 't Hoff, Le Bel and Wislicenus. American Book Co, New York, pp 1-33 Prof. J. E. Ari~ns Groenewoudse Weg 45 NL-6524 TP Nijmegen The Netherlands
EuropeanJournalof C r ~ : ~ ( ~ ( ~
@beecell@9 @ Springer-Verlag 1991
Special article Racemic therapeutics - ethical and regulatory aspects E. J. Ari~ns Received: April 26,1990/Acceptedin revised form: December 8, 1990
Summary. Racemic therapeutics are fixed ratio mixtures of stereoisomers to be regarded biologically as different compounds. Usually only one of the isomers fully contributes to the therapeutics action, and the other is often classifiable as "isomeric ballast". Due to differences in turnover and pharmacokinetics, the proportion of enantiomers (1:1 in the racemate) continuously changes in plasma. The implications of the neglect of stereoselectivity for various levels in the investigation of racemic drugs are discussed and summarized in Table 2. The fact is that clinical investigators, Ethical Committees and regulatory authorities have for decades accepted invalid pharmacokinetic data on some 25% of therapeutics. That those racemates remain in use make the benefit of and necessity for kinetics generally questionable. Exposure of patients to the "isomeric ballast" present in about 50% of the most commonly used drugs will probably containe for many decades. As a result of a change in attitude of the regulatory authorities, however, for new drugs the choice in future between the racemic therapeutic or the single isomeric ballast-free drug will largely be based on a critical evaluation of the chiral characteristics with regard to their therapeutic, toxicological and pharmacokinetic aspects. Key words: Racemates therapeutics; ethical aspects, isomers, adverse effects, chirality drug stereoselectivity Since the article "Stereochemistry a basis for sophisticated nonsense in pharmacokinetics and clinical pharmacology" appeared in the European Journal of Clinical Pharmacology 1984 [1] much attention has been paid to the pharmacological aspects of drug chirality [1-13]. There has been much less interest in the ethical implications and the problems caused to regulatory authorities in particular by the many racemic therapeutics [4]. In the recent past, however, there has been some discussion of these aspects, too, which makes it worthwhile to assess the present position with some emphasis on the ethical and social aspects.
Few scientists will deny that stereoselectivity in action is a common and important phenomenon, not only in pharmacology but in particular in pharmacotherapeutics. This holds true for dynamics - desired and undesired actions - and kinetics. In the latter it is particularly important in the rate of enzymatic conversion, transport by carriers and protein binding, and thus for distribution and elimination. Stereoisomers in biology should really be considered as essentially different chemical compounds. The distinction "active" and "inactive" isomer is misleading. The isomer most potent for a particular action, e.g. the desired effect, is called the "eutomer". The other isomer, which has little often active, is the "distomer", and it often contributes to the undesired actions, and so can never be regarded as entirely harmless. Isomers that do not contribute to the desired action are to be considered as unwanted by products, "isomeric ballast", in drugs. Problems arise with the many mixtures of stereoisomers. Composite Chiral Drugs (CCDs), such as racemates, which constitute _+25% of the therapeutics on the market. What is said for racemates here is mostly applicable also to CCDs in general. Two important parameters must be considered. 1) The "eudismic ratio" (ER), the ratio of the activities (dynamics) of the eutomer and distomer [15]. It usually differs for those components in the desired and undesired actions generated at different sites of action (e. g. receptors). It is specific for a particular racemate in relation to a particular action and only rarely has the value '1'. 2) The "eudismic proportion" (EP) is the proportion of the concentrations of eutomer and distomer ([eut]/[dist]) in, for instance, plasma [14]. It is of particular significance in kinetics. For a racemate, the eudismic proportion (EP) of the enantiomers is 1 by definition. In particular after absorption, due to the commonly different rates of metabolic conversion of the enantiomers, the EP in the body fluids gradually changes. Exceptionally, e.g. in case of rapid in vivo racemisation, the racemate is found in plasma. If elimination is dominant for the eutomer, the ER which may be i for the racemate in the tablet, with grad-
90 ually decrease in the plasma, e. g. in the case of verapamil [16] to about 1/4 ( + 80% distomer). If elimination of the distomer is dominant, the EP in plasma will increase, as in the case of nivaldipine [17] to about 3 ( + 75% eutomer). In fact, the composition, ER of the "racemate" in the plasma continuously changes until a complete steady state is reached for each of the stereoisomers. The time required differs, therefore, for the two isomers. Without chiral assays the eudismic proportion (EP) in plasma remains unknown. If there is extensive presystemic or first pass loss by metabolic conversion in the liver, equal "plasma concentrations of the racemate" obtained after oral and after parenteral administration will differ in EP, and so will be bioinequivalent [16]. The plasma concentration after application of a racemate, as measured by non-chiral assays, only gives information on the sum of the eutomer and distomer, and since the eudismic proportion changes with time, it will not be directly related to the response. The various pharmacokinetic constants, such as half-life, bioavailability, persistence, etc. derived from "racemate" concentrations based on non-chiral assays, can only be notional, and are comparable to such parameters as the age or body weight of a married couple. In the exceptional case when the eudismic ratio is 1, the eudismic proportion does not matter, but complications may still arise due to stereoselectivity in the conversion to active metabolites.
Clinical pharmacology of racemic therapeutics Neglet of stereoselectivity by reliance on non-chiral assays in clinical pharmacology of racemic therapeutics is far from exceptional. A large proportion of the publications on pharmacokinetics, and the related clinical evaluation of racemic drugs, not only was but still remains heavily biased [1-6]. Often the racemic nature of the drugs is not mentioned in the publications, although many, mostly recent, reviews on chirality of drugs [1-15] clearly illustrate stereoselectivity in pharmacodynamics and kinetics, and the broad spectrum of possibilities and implications thereof. Impressive tables of pharmacokinetic constants for racemates, calculated on the basis of erroneous plasma concentrations of "racemates" by means of computer-assisted curve fitting, and based on mathematical equations derived for and valid only for single compounds, are a striking characteristic in particular of pharmacokinetic publications in clinical pharmacology. Similarly, multi-compartment systems for racemates are derived by computer assisted fitting of postulated compartmental systems to the multiphasic decline in the concentration of the "racemate", again on basis of invalid pharmacokinetic equations. The eudismic proportion (EP), the composition of the "racemate", not only changes with time but also tends to differ in the various compartments. The number of phases in the decline of the plasma "racemate" concentration increases with the number of compounds (stereoisomers) involved. Kinetics of racemates studied with non-chiral assays thus readily lead to fictional results, both in the kinetic constants and the compartmental systems derived. Data so generated are
E. J. Ari~ns: Racemictherapeutics not realistic, and, depending on the extent to which the clinician relies on them, they may even be risky [11]. Studies using chiral assays brought out remarkable complications with racemic mixtures in relation to genetic polymorphism and species-, race- and age-associated differences in pharmacokinetics. Caldwell et al. [18] analysed the metabolic chiral inversion and dispositional enantio-selectivity of the 2-arylpropionic acid, NSAID's and their biological consequences with highly informative results. Metabolic conversion of enantiomers at different rates to reactive intermediates also has the consequence that positive results obtained with the microsome-dependent Ames test of a racemates can be misleading. The distomer may be the culpit. Also, racemic prodrugs with different rates of conversion of the pro-eutomer and the pro-distomer may, if studied by non-chiral assays, lead to confusing results. The major problems arise: 1) How to reduce undesirable exposure of patients to isomeric ballast in composite chiral drugs to the scientifically feasible, technologically possible and economically acceptable limit? 2) How to avoid reports in the literature of heavily biased data on racemic therapeutics due to neglect of chirality when chiral assays are essential. An argument often heard is "these are matters for the regulatory authorities". They do have their responsibilities with regard to admission to the market of drugs, including CCD's. Their decisions must be based on scientifically valid data. A scientist never has any excuse to violate basic principles in science by referring to some administrative authority. In any case, in the study of the pharmacokinetics of racemic drugs already on the market, and in Therapeutic Drug Monitoring, regulatory authorities are no generally involved. Scientists have their own specific responsiblity as individuals and as a group. It is their duty to raise an alarm, especially, in the case of widespread fundamental errors.
Ethical aspects of drug development, evaluation and use In relation to the problems just mentioned there are two basic principles to be emphasized: 1) "Exposure of nature (including man) and its environment to xenobiotics (including drugs) is only justified if the desirable actions adequately compensate for the undesirable actions and the never fully excluded risks." This implies, amongst other things: "Avoid isomeric ballast in therapeutics." 2) "In science the notion of impossibility or incapability to do things the proper way is no excuse to do them the wrong way." Protection of humans against the risks of exposure to xenobiotics, particularly drugs, in today's chemistry-impregnated society received some attention in the World Medical Association's 'Declaration of Helsinki' [19]. Under "Recommendations guiding medical doctors in biomedical research involving human subjects" it is stated under basic principles:
E. J. Ariens: Racemic therapeutics
91
Table 1. Racemic antiarrhythmics
Tocainidec Mexiletinec Encainide d
VPC-patients efficacy (%) B or PL 40 40 80
MVA 15 15 25
adverse-effect proarrhythmia (%)* VPCs MVA 2 3 2 3 >1 10
non-cardiac (%)
potential for organ toxicity
40a 40a 10b
+++ + -
VPC ventricular premature complexes; MVA malignant ventricular arrhythmias; B or PL Benign or Potentially Lethal. a) Gastrointestinal and neurological, b) blurred vision, c) enantiomeric differences in metabolic conversion, d) formation of active metabolites. * In-
dudes an estimate of serious and potentially life-threatening proarrhythmic effects only [20]. The eudismic ratios for the therapeutic action are close to 1. How the side-effects are distributed over the stereoisomers is unknown
1.1. "Biomedical research involving human subjects must confirm to generally accepted scientific principles and should be based on adequately performed laboratory and animal experimentation and on a thorough knowledge of the scientific literature."
books pharmacology, pharmacokinetics or toxicology are the principles of stereoselectivity, stereospecificity, stereochemistry etc. properly described, or even mentioned in the index. Pharmacokinetic parameters, particularly the half-life and bioavailability of racemic therapeutics calculated from plasma "racemate" concentrations measured by non-chiral assays are amply presented [4, 5]. What do they mean? The reader may consult the library and see for himself. G o o d m a n and Gilman 7th edition (1985), for example, presents 65 pages (1668-1733) of pharmacokinetic constants for therapeutics, roughly a quarter of which are racemates. There is no indication there of the role of chirality. With few exceptions the constants are based on plasma levels determined by non-chiral assays. Students should be spared the false idea that there really is someentity like "plasma concentrations and pharmacokinetic parameters of racemates". What is real are plasma concentrations etc. of the individual stereoisomers and the continuously changing eudismic proportion. For the hundreds of therapeutics registered in the past 20 years the regulatory authorities have required pharmacokinetic data, including information on the hundreds of racemates among them. These data with few exceptions have been based on non-chiral assays. What is the sense?
1.9. "In any research on human beings each potential subject must be adequately informed about the aims, methods, anticipated benefits and potential hasards of the study etc." This implies the "informed consent". 1.8. "Reports of experimentation not in accordance with the principles laid down in this declaration should not be accepted for publication." There is something to be learn there by the Ethical Committees (the Institutional Review Boards, IRBs) supervising evaluation of drugs in humans. If racemic therapeutics or therapeutic isomer mixtures are involved, unless stereospecific assays are used invalid data will be generated. Taking into account the involvement of patients, this is unethical and thus is not acceptable. Are the patients informed about this situation? Is there honest "informed consent"? In particular, hospital pharmacists and clinical pharmacologists are in vited to review their contributions to science, in particular in the clinical evaluation the many racemic therapeutics they deal with. This will increase awareness of the pitfalls of non-chiral assays in the study of CCDs. A simple and effective way to stop the still continuing flow of deficient data generated in the study of CCDs would be extension of the insructions to authors with an indication that "The composite character of drugs which are mixtures of stereoisomers must be brought to the attention of the reader. The prefix (RS)- or rac-, e.g. rac-propranolol, in the case of racemates and (Z/E)- or cis/trans- in the case of that type of isomerism is obligatory. The implications of the composite and chiral nature of the drugs for the interpretation of the data measured and the conclusions drawn must be made explicit" [3, 5, 6].
The source of wisdom
The persistent neglect of chirality and its implications for drug action comes from deep in our education. In hardly any of the highly recommended clinically orientated text-
Racemates as drug combinations
For fixed ratio combinations, for instance of an antihypertensive [3-blocker and a saluretic, the authorities require full information on both components in the mixture, as well as on the mixture as such. It would make sense to follow a similar approach in the case of mixtures of stereoisomers. Such information would make clear when the single eutomer, will often be the case or the racemic mixture, is to be preferred. It is necessary to take into account, even in the exceptional case, that there is some form of synergism between the stereoisomers, when their 1:1 proportion only rarely will be optimal. In those cases when the eudismic ratio for the therapeutic action is approximately 1, one must still take into account that this may not be the case for adverse effects that are not inherent to the mechanism of therapeutic action. Side-effects, such as allergic reactions, mutagenicity and embryotoxicity, but also, for example, intestinal or neurological side-effects of cardiovascular therapeutics may be distributed over the enantiomers in a different way
92
Table 2. Requirements in the evaluation of racemic and Composite Chiral Drugs (CCD's) 1) PharmaceuticaI industry The composite character of chiral drugs, such as racemates, and its implications should clearly be noted. Stereoselective analytical methods necessary for proper investigation of CCDs should be made available. 2) Investigators Proposals for research projects scientifically deficient due to neglect of the chiral nature of the drug(s) to be studied demand re-education of the scientist involved. 3) Funding institutions In judging requests for grants to study CCDs the need for chiral assays should not be overlooked. 4) Ethics Committees Generation of invalid data due to disregard of stereoselectivity in evaluation of the activity CCDs in studies on patients is not ethical. In such cases "informed consent" requires special attention. 5) Investigators should avoid Neglect of stereoselectivity in pharmacodynamic and -kinetic, studies of CCDs. Calculation of pharmacokinetic constants and derivation of multicompartment systems for CCDs based on illusory racemate concentrations using pharmacokinetic equations valid only for single agents single stereoisomers, and not for CCDs. Presentation for publication of research reports and manuscripts biased due to disregard of the implications of stereoselectivity in action of CCDs. 6) Editorial boards and referees of scientific journals Proper attention should be paid to the composite character of racemic drugs and CCDs and the awareness of the authors to the implications thereof. 7) Prize awarding agencies Promotion of poor science by awarding prizes etc., for example, for astereognosis in the study of CCDs must be prevented. 8) IUPHAR (International Union of Pharmacology). FIP (F6d6ration Internationale Pharmaceutique). APha (American Pharmaceutical Association). Might take steps to stop or at least to discourage the frequently neglect of stereoselectivity in the study of CCDs, one basis for substandard science. 9) Teachers and textbook-writers In educational programmes proper attention should be paid to chirality as an important aspect of drug action, particularly of CCDs. 10) Regulatory authorities Adaptation of the rules for data to be submitted for new CCDs, namely information on individual stereoisomers as well as the mixture. Re-evaluation of those CCDs already on the market with a relatively high incidence of side-effects on a similar basis is advisable.
from the desired action. In this respect it is important, to reevaluate racemic therapeutics with a high incidence of side-effects, such as tocainide and mexiletine, antiarrhythmics with a + 40% incidence of non-cardiac side-effects (Table 1; [20]). Requirements for the evaluation of racemic therapeutics are summarized in Table 2. T h e y are applicable to the 25% CCDs in use but are usually neglected. The literature is rich in examples of such neglect, e. g. most studies on [3adrenergic agents, ~-adrenoreceptor blockers, combined [3- and c~-adrenoreceptor blockers, calcium channel block-
E. J. Ari6ns: Racemic therapeutics ers and non-steroidal antiphlogistics, all drugs in frequent use. Most clinical pharmacokinetic studies on t h e m are deficient due to the indiscriminate use of non-chiral assays. There are recent informative and illustrative (1989) articles [21-23] on racemic calcium channel blockers presenting a b r o a d spectrum of kinetic constants. They clearly demonstrate where astereognosis leads to. It is characteristic of the situation that Article 21 even was awarded the "Pharmatec prize" for its outstanding scientific qualities! This was given under auspice of the F6deration Internationale Pharmaceutique (FIP). D a t a on changes in clinical parameters, such as blood pressure, cardiac output, liver function and etc. as a function of dose and time have their own intrinsic value for CCDs and fixed ration combinations of drugs in general. In this regard References 22 and 23 are correct. Pharmacokinetic data on CCDs based on non-chiral assays should be rejected as should concealment of the composite character of the drugs studied.
Perspectives The climate is changing. In 1987 Japan, in a leading position, adopted rational requirements for racemic products. Full pharmacology and toxicology is required for each of the enantiomers and for the racemate [24]. For the latter only pharmacokinetic data based on chiral assays should and probably would be accepted. The E u r o p e a n Economic Community (EEC) in "Rules governing medicinal products in the E u r o p e a n C o m m u n i t y " Volume II, 1989 extended the section on stereoisomerism. It states: "Possible problems relating to stereoisomerism, which should be discussed in the appropriate Expert R e p o r t and cross referenced, should include: the batch to batch consistency of the ratio of stereoisomers in the various batches used - the toxicological issues - the pharmacological aspects (including evidence on which stereoisomers have the desired pharmacological properties) - p h a r m a c o k i n e t i c s including information on the relative metabolism o f the stereoisomers - extrapolation of the pre-clinical data (paying particular attention to possible problems relating to species differences in handling of the stereoisomers) - the significant clinical issues." This at least opens the possibility of requiring essential information on each of the enantiomers or stereoisomers in general, as well as on the racemate or in general composite chiral products. Further, it is stated that: "where a mixture of stereoisomers has previously been m a r k e t e d and it is now proposed to m a r k e t a product containing only one isomer, full data on this isomer should be provided." The requirements for a single isomer (the eutomer) among new drug applications should in general be analogous to those for a single compound. The US authorities ( F D A ) have come to an approach comparable to that in Japan and the E E C to be effected in 1992. This change in policy will reduce the flow of heavily biased data on CCDs in the scientific literature as far as ned drugs are concerned. Unless the editorial boards of the scientific journals join the regulatory authorities in their wisdom, for the hundreds of CCDs already on the
93
E. J. Ari~ns: Racemic therapeutics m a r k e t the contestable non-chiral kinetics will remain with us for m a n y years to come. Therapeutics containing isomeric ballast should be withdrawn from the m a r k e t as soon as ballast-free preparations are available at an acceptable price. Often only a small fraction of the price paid in the pharmacy represents the cost of the drug substance in the product. The approach indicated should hold true for generics as well as for branded products. The resulting incentive for the innovative industry and the rapid development of chiral techniques in chemistry will help to reduce chemical ballast in drug therapy. The patient, if aware of it, would definitely appreciate this result. In conclusion: (I) Acceptance by regulatory authorities of composite chirat drugs, such as racemates, should, whenever possible, be based on proper information on the individual stereoisomers as well as the mixture. It is anyway preferable to avoid chemical ballast. The question to be considered is that of weighing the advantages against the disadvantages, including the never entirely absent risks of exposure to chemical ballast. II) The study of pharmacology, dynamics and in particular of the kinetics of CCDs, such as racemates, requires chiral assays. The investigator has the moral duty to inform the reader of the composite nature of any drug studied and its implication for the interpretation of data. III) Ethical Committees amongst others have the duty to implement i n clinical research the recommendations of the World Medical Association "Declaration of Helsinki". IV) Teachers are invited to discuss with their pupils, or at least bring to their attention, the arguments advanced here. As an exercise in open minded reasoning they might be asked to write - "in the light of the forgoing a short c o m m e n t on the books on pharmacology, therapeutics, pharmacokinetics and biopharmacy r e c o m m e n d e d to them". The p r o b l e m affects about 25% of the available therapeutics, including 50% of those most commonly used. The recognition of these problems had to wait for hundred years after Pasteur generated the basic insights [25]. It should not take a further hundred years to eliminate astereognosis from general and clinical pharmacology.
References 1. Ariens EJ (1984) Stereochemistry, a basis for sophisticated nonsense in pharmacokinetics and clinical pharmacology. Eur J Clin Pharmacol 26:663-668 2. Ariens EJ, Soudijn W, Timmermans PBMWM (1982) Stereochemistry and biological activity of drugs. Blackwell Scientific Publications, Oxford, pp 1-190 3. Series on Chirality, Trends in Pharmacological Sciences (1986) 7: 20-24, 60-65,112-115,155-158, 200-205, 227-230, 281-301 4. Ariens EJ, Wuis EW (1987) Bias in pharmacokinetics and clinical pharmacology. Clin Pharmacol Ther 42:361-363
5. Ari~ns E J, Wuis EW, Veringa EJ (1988) Stereoselectivity of bioactive xenobiotics. A pre-Pasteur attitude in medicinal chemistry, pharmacokinetics and clinical pharmacology. Biochem Pharmacol 37:9-18 6. Ari~ns EJ (1990) Stereoselectivity in pharmacodynamics and pharmacokinetics. Schweiz Med Wochenschr 120:131-134 7. Drayer DE (1986) Pharmacodynamic and pharmacokinetic differences between drug enantiomers in humans: an overview. Clin Pharmacol Ther 40:125-133 8. Wainer IW, Drayer DE (1988) Drug stereochemistry. Analytical methods and pharmacology. Dekker, New York 9. Jamali F, Mehvar R, Pasutto FM (1989) Enantioselective aspects of drug action and disposition: Therapeutic pitfalls. J Pharm Sci 78:695-715 10. Smith DF (1989) The stereoselectivity of drug action. Pharmacol Toxicol 65:321-331 11. Evans AM (1988) Stereoselective drug disposition: potential for misinterpretation of drug disposition data. Br J Clin Pharmacol 26:771-780 12. Ari~ns EJ, v. Rensen JJS, Welling W (1988) Stereoselectivity of pesticides - biological and chemical problems. Elsevier, Amsterdam 13. Eichelbaum M, Gross S (1990) Stereoselectivity in drug action and disposition. N Engl J Med 14. Ari~ns EJ (1991) Chirality and isomeric ballast in bio-active products: Proceedings 3rd Netherlands Biotechnology Congress Amsterdam, in press. Elsevier, Amsterdam 15. Lehmann FPA, Rodriques de Miranda JF, Ari~ns EJ (1976) Stereoselectivity and affinity in molecular pharmacology. In: Jucker E (ed) Vo120. Birkhguser, Basel, pp 101-142 16. Vogelgesang B, Echizen H, Schmidt E, Eichelbaum M (1984) Stereoselective first-pass metabolism of highly cleared drugs: studies of the bioavailability of L- and D-verapamil examined with a stable isotope technique. Br J Clin Pharmacol 18:733-740 17. Tokuma Y, Fujiwara T, Niwa 1",.Hashimoto T, Noguchi H (1989) Stereoselective disposition of nivaldipine, a new dihydropyridine calcium antagonist, in the rat and dog. Res. Comm. Chem Pathol Pharmacol 63:24%262 18. Caldwell J, Hutt AJ, Fournel-Gigleux S (1988) The metabolic chiral inversion and dispositional enantioselectivity of the 2arylpropionic acids and their biological consequences. Biochem Pharmacol 37:105-114 19. Levine RJ (1986) Ethics and regulation of clinical research, 2nd ed. Urban and Schwarzenberg, Baltimore, pp 393-429 20. Morganroth J (1987) New antiarrhythmic agents: Mexiletine, tocainide, flecainide, encainide and amiodarone. Rational Drug Therapy. Am Soc Pharm Exp Ther 21:1-5 21. Soons PA, de Boer AG, van Brummelen R Breimer DD (1989) Oral absorption profile of nitrendipine in healthy subjects: a kinetic and dynamic study. Br J Clin Pharmaco127:17%189 22. Harten J van Burggraaf J, Lichthart GJ, Brummelen P van, Breimer DD (1989) Single- and multiple-dose nisoldipine kinetics and effects in the young, the middle-aged, and the elderly. Clin Pharmacol Ther 45:600-607 23. Dunselman PHJM, Edgar B, Scaf AHJ, Kuntze CEE, Wesseling H (1989) Pharmacokinetics of felodipine after intravenous and chronic oral administration in patients with congestive heart failure. Br J Clin Pharmaco128: 45-52 24. Smith RL, Caldwell J (1988) Racemates towards a New Year solution? TIPS 9:75-77 25. Pasteur L (1901) On the asymmetry of naturally occurring organic compounds, the foundations of stereochemistry. In: Richardson GM (ed) Memoirs by Pasteur, Van 't Hoff, Le Bel and Wislicenus. American Book Co, New York, pp 1-33 Prof. J. E. Ari~ns Groenewoudse Weg 45 NL-6524 TP Nijmegen The Netherlands
