Back to the 1999 MAPS MDMA Conference Page
See also the talk by Juraj and Sonja Styk
and the talk by George Greer
given at the same conference.

Researchers in several countries have proposed to recommence research into the potential beneficial uses of 3,4-methylenedioxymethamphetamine (MDMA, popularly "Ecstasy")(1). Current or proposed studies address its potential in treating post- traumatic stress disorder (PTSD) (2) and the psychological distress of terminal cancer patients (3). In light of MDMA's demonstrated abuse potential (4) and possible toxic side-effects (5) , those proposals confront regulatory authorities with some difficult choices involving the protection of research subjects and the prevention of substance abuse as well as the interests of sufferers from various disorders. This paper will consider the risks and benefits of research on potential therapeutic uses of MDMA.

It should be noted, however, that the powers claimed for MDMA - increasing insight and empathy, reducing defensiveness - are not at all specific to post-traumatic stress or to terminal cancer, or indeed to the therapeutic context at all. (6,7) Medical uses are a proper subset of possibly beneficial uses. If in fact there are drugs that, appropriately used, could enhance the well-being of individuals without diagnosed pathologies, then how best to administer them, and to regulate their administration, needs to be explored. Physicians, trained to treat the sick, may not be the best supervisors for healthy individuals who want to use drugs in a way that is neither "recreational" nor "therapeutic," and the research and regulatory paradigms developed for the study and approval of purely therapeutic agents may require adaptation to the new situation. But those questions go far beyond the scope of this essay.

Even if we restrict our attention to research into therapeutic applications of MDMA, we face a tangle of scientific and regulatory questions. Decisions about what research to pursue, and to permit, involve questions of policy toward drug-development research, policy toward the regulation of drugs in clinical use, and policy regarding control of non- medical use and abuse of psychoactive drugs.

RESEARCH POLICY

First, then, to the question of research. Here as elsewhere, costs and risks must be balanced against benefits. The costs are the economic costs of conducting the research and the forgone benefits of the other research that might be conducted with the same scarce resources of money and research talent. The risks are twofold: the risk of damage to experimental subjects, and the risk that the research will generate non-medical use outside the research context.

The primary potential benefit is, of course, the value to patients of whatever therapies might be developed. Given the famous intractability oof conditions such as PTSD and the personality disorders and the great challenge of dealing with those who are known to be dying, those benefits could be very substantial; they need to be discounted by the probability of actually developing them and putting them into practice, a probability hard to guess in advance but surely much below unity. Potential side-benefits of research include the therapeutic value the experimental subjects might derive in the course of clinical research and the basic (non-clinical) knowledge, perhaps relevant to drug abuse prevention and treatment, that might emerge in the course of clinical studies.

The risks to experimental subjects are themselves of two kinds: toxic effects (physical or behavioral) of the material itself, and adverse therapeutic outcomes (by comparison with whatever therapy would otherwise have been given). Since no therapeutic procedure - including inaction - is without risk, including the risk of failure to improve the condition being treated, it would seem that the ethically relevant question is whether there are potential subjects whose expected outcome would be better from undergoing the experimental therapy than it would be from relying on the best established therapy, or none at all. If so, and if potential subjects are given full information and the opportunity to make a reasoned choice, then the research can go forward with no injustice towards its subjects (8).

The history of medical research yields all too many examples of researchers' disregard for the well-being of their subjects, and thus fully justifies the imposition of a regulatory regime including Institutional Review Boards or the equivalent. (9) Nonetheless, it remains the case that research regulation, especially in areas where scientific controversy is linked to broader social controversy, carries with it the risk of official censorship of science for fear of reaching the "wrong" results. This problem has not been absent from the regulation of research into currently illicit psychoactives. There is irony, at least, in regulation that forbids subjects to receive known dosages of pure chemicals in settings designed for safety, even when those same subjects are voluntarily using unknown doses of the same chemicals, possibly adulterated, in much more hazardous settings.

The danger of human-subjects protection acting as scientific censorship is especially grave in the case of controlled substances, where the studies that might bring a drug's prohibited status into question can be made difficult, or even impossible, as a result of the prohibited status itself. The problem is intensified when agencies with strong organizational commitments to the maintenance of existing regulations have the power to approve or prevent research(10).

The risks to others than experimental subjects arising from non-medical use depend both on the inherent riskiness of the material in such use (11) and on the impact of research - both the fact of research approval and the reports of research results - on the extent such use. In the case of a compound such as MDMA, which is already in widespread non- medical use and widely available on the illicit market, the "supply-side" effects are likely to be negligible, but the "demand-side" issues deserve further exploration.

The benefits of research depend on both the knowledge developed and on the extent to which it changes the behavior of clinicians and regulatory authorities here and abroad. The two are quite distinct in some instances. For example, the finding that, in rats, fluoxetine blocks the changes that MDMA can otherwise induce in serotonergic neurons (12) was, according to some of those involved in that research, unsurprising, since the same effect was well-established with methamphetamine. But the result may still be important from a regulatory viewpoint. By the same token, from the viewpoint of those who believe in MDMA's potential benefits based on large numbers of clinical reports from the period preceding its scheduling (13), a published study involving random assignment of patients to experimental and control treatments would add little to what they regard as established knowledge. But the absence of such a study remains a barrier to approval for clinical use.

This problem of "regulatory science" interacts in perverse ways with the human-subjects problem. For example, an IRB might refuse permission for studies on the grounds that the results are fully predictable (as in the case of replicating the fluoxetine/MDMA studies in human subjects) and that therefore no new knowledge is to be gained to justify whatever risks are incurred by the subjects. At the same time, another IRB might refuse to permit other studies for fear of the possibly toxic effects which fluoxetine would block, and refuse to consider the co-administration of fluoxetine as an adequate safeguard on the grounds that its protective effect has been shown in rats but not in humans. There is need to guard against the risk of creating a closed system in which the research that might justify changes in current regulations is made impossible by the regulations themselves.

Those planning clinical research with MDMA and other controlled substances must accept the fact that their results, if positive, are likely to be met with intense skepticism, not only from peers but from regulators. They may need to shape their research strategies to account for the regulatory context. The less subjective the outcome measures, the greater the credibility of the results; this enhances the value of outcome measures involving physical symptoms (e.g., asthma) or directly measurable behavior (e.g., cocaine consumption) and of conditions such as depression for which there are generally recognized measurements not dependent on the therapist.

DRUG APPROVAL POLICY

As a formal matter, a drug's potential for non-medical use, other than in the context of iatrogenic addiction, plays little or no role in the drug-approval process. The question asked by drug regulators is whether the drug is safe and efficacious therapeutically, not what impact its approval might have on substance abuse by non-patients.

In fact, however, drug regulators read the same newspapers the rest of us do, and it would be unreasonable to expect them to ignore the risks in approving abusable psychoactives. Nor would it be entirely desirable; the possibility of spread from medical to non-medical use is a real one, and should not be ignored. On the other hand, since consideration of such issues is not part of the statutory responsibility of the US FDA and its counterparts in other countries, they tend to be considered informally and tacitly (14). This may give such considerations greater weight than they would have in a process that openly discussed all the potential stakes in the decision to be made; it is very difficult to respond to an objection that is never voiced.

Drug approval policy should ideally also be designed on a risk-benefit basis. Two kinds of errors can be made by regulatory authorities: they can approve a drug which is later shown to have adverse side-effects greater than its therapeutic benefits (again, in each case comparing the new drug with the best available established therapy, if any); or they can fail to approve, or delay in approving, a drug whose benefits in use exceed its risks in use. Thus the gravity of the condition being addressed, and the known or suspected risks of existing therapies, ought to be just as central to the drug-approval process as the risks of the new material.

Unfortunately, the organizational interests of the regulators do not in this instance coincide perfectly with the interests of the public. Adverse outcomes of a newly-approved drug are likely to be blamed directly on the regulatory authorities, while the lost benefits of new drugs not being given - lost therapeutic value and the adverse outcomes of existing therapies - are not obviously and directly the responsibility of the regulators. In particular, taking an existing drug off the market in the face of newly-revealed side effects can be seen as a confession that the previous approval was in error. Of course, no error need be involved; both the initial approval and the subsequent removal may have been the best course of action available given the information in hand. (Similarly, as a traveler should not be criticized either for ordering a cup of coffee at an inn he has never visited before or for not drinking it when a first sip tells him that it is weak, over- brewed, and many hours old.)

Indeed, the faster and more efficient the process of detecting side effects as they arise clinically after a drug approval and removing it from use if conditions warrant, the bolder the authorities can afford to be in approving new drugs; just as, the better the brakes on an automobile, the faster one can prudently drive it. Most drugs newly approved for clinical use have been tried in a relatively small number of subjects in the course of clinical trials. Under those circumstances, the first year of clinical use will multiply manyfold the number of persons exposed to the drug. It is not surprising that occasionally a rare but serious side-effect which had not appeared in clinical trials appears shortly after approval, as most recently demonstrated by the fenfluramine- phenteramine fiasco.

Note that this sort of risk is far smaller with MDMA than with drugs just emerging from the laboratory. Years of medical and non-medical experience, with very large numbers of users, in effect represents an informal screening for gross adverse effects (15), and in fact both the sequellae of severe hyperthermia and adverse cardiac events have been identified as real though low-probability consequences of MDMA use, or at least of MDMA use in combination with vigorous exercise and high ambient temperatures and in the absence of adequate hydration (16). The wide informal experience base reduces, though of course not to zero, the probability of finding clinically significant side-effects after approval that were missed during clinical trials.

The best system for new drug approvals would combine (1) the approval of new drugs as soon as the evidence that their benefits exceed their risks rises to the level where a prudent person would prefer that drug for his own use to the best existing alternative with (2) their speedy removal if early clinical use reveals sufficiently important and frequent adverse side-effects not revealed by pre-approval testing. Some parts of Western Europe have approached that ideal far more closely than has the United States.

Part of the explanation for the difference lies in varying assumptions about the motives and competence of clinicians, and thus in the extent to which decisions about drug therapies ought to be centralized or decentralized. Where the United States has a policy reflecting the fear that irresponsible pharmaceutical companies will deceive ignorant physicians into unwise use of new therapeutic agents, Switzerland and the United Kingdom, for example, place far more stress on physician self-regulation. Where the assumption that physicians are competent to make sound clinical judgments and will not be deceived or unduly influenced by manufacturers is justified, a decentralized system can have substantial advantages over a highly centralized one, especially where there are relatively subtle differences between groups of patients in their tendency to be helped or harmed by a given drug.

But no one would argue that drugs should be approved for routine clinical use as freely as they are for experimental use. In particular, research protocols can define patient populations, dosage levels and frequencies, and safeguards far more precisely than can be achieved in routine practice. One must also allow for the prospect that some of the practitioners who prescribe a drug will be less well-informed about the techniques of its use and its associated risks than one can expect a careful researcher to be.

In addition, when the drug involved is subject to non-medical use because it produces a subjective experience which some persons find pleasurable (or otherwise beneficial) one must be concerned not only with the results of medical use but with the impacts of medical use on non-medical use. There are two major dangers here: that drugs may be diverted from licit medical use to illicit recreational use, and that reports of therapeutic applications, either by word-of-mouth or through the media of mass communication, will stimulate illicit demand. The significance of these risks depends on the probability of diversion, the likely size of that diversion, and the damage diverted drugs might do to users and to others. The instance of diazepam (Valium) illustrates both the problem of mis- and over-prescription and of deliberate diversion for recreational use.

Even if MDMA were approved for the sort of medical use now under discussion, it is almost impossible to see how diversion could become a significant source of black-market supply. No one has proposed giving patients prescriptions for multiple doses of MDMA to take home for self-medication ad libitum. Rather, it is proposed that patients be given doses one at a time for immediate consumption in the presence of the therapist. Any leakage from such a system will necessarily be trivial, simply because the total number of legitimate doses would remain too small to allow significant diversion to go undetected.

However, the lack of a diversion problem does not guarantee the lack of substantial non- medical use. The drug can be illicitly produced from easily available materials. Given the prospect of an illicit market, the question is the extent to which medical use might contribute to demand in that market.

There are two opposite effects here. Medical use might discourage non-medical use by associating the drug with (often painful) treatment rather than with pleasure. This, and the apparent tendency of MDMA's pleasures and other benefits to exhaust themselves after a relatively small number of doses and for the drug's unpleasant side effects to grow quickly with repeated use, provides some reassurance that those who first experience the drug clinically are unlikely to form a base of steady customers for an illicit market (17). However, reports from England and the Continent of frequent and multiple-dose MDMA use among ravegoers should raise a warning flag.

On the other hand, MDMA is reported to produce empathy, insight, and feelings of wholeness. These experiences are desired by the healthy as well as the sick. To the extent that MDMA in fact produces such results, and gains a reputation for doing so, it will be in demand. It is also a stimulant, and the history of stimulant use among Europeans and North Americans - from the day the Turks, retreating from Vienna, left their coffee behind them to the current epidemic of cocaine smoking in the United States and elsewhere - suggests that stimulants find ready markets. The history of MDMA in the United States appears to be, in part, one of therapeutic use generating a market for non- therapeutic use, and recent reports from the U.S. of the use of MDMA in poly-drug combination with alcohol and other depressants are far from encouraging (18).

But the genie is already out of the bottle. On balance, it seems likely that the approval of MDMA for clinical use will tend to foster non-medical use as well, or at least speed the pace at which knowledge of the drug's effects, and interest in experiencing them, spreads. But it also seems likely that non- medical use will be present whether or not clinical use is approved, and the eventual size of the illicit market may depend very little on the decision to approve or disapprove the use of MDMA in psychotherapy. (Approval as a prescription drug for "take-home" use would be a very different matter, as the current case of Viagra demonstrates.)

MDMA AND THE LARGER PROBLEM OF DRUG ABUSE CONTROL

The size of the abuse problem - the extent, that is, of the damage done by drug use to users and by them to others - depends on the number of users, the proportion of them who consume to the point of damaging themselves or losing self-control, and the innate tendency of the drug to damage the health of its users and to cause them to behave in ways which damage others: its social dose-effect curve, if you will.

In judging a drug as a potential abuse problem, one should not stress its pharmacophysiology in preference to its behavioral pharmacology. The tendency of a drug to generate heavy chronic use or very heavy use at a single session is more important than its toxicity. Nicotine, for example, is a powerful poison: there is enough nicotine in a single cigar to kill a person if it were administered all at once. But as a behavioral fact no one takes that much nicotine at once, because the drug's unpleasant effects build up too quickly. On the other hand, nicotine in cigarette form is so powerfully reinforcing that many if not most users become compulsive users, thus exposing themselves to the long- term effects of the drug and its delivery systems.

The experience of the United States and Europe suggests that MDMA is capable of generating a substantial market as measured by the number of doses used. (Only some of those doses turn out to actually contain MDMA, and some of those are adulterated with other chemicals.) The European, but not yet the American, experience suggests the development of a significant number of truly problematic users, though again the pattern in the US may be shifting in worrisome ways.

The reason for MDMA's relatively benign profile in the United States - at least to date - seems to be that many persons use MDMA, and many of them use it repeatedly, but few use it in very large doses or very frequently. This is consistent with the studies of Beck and Rosenbaum, among others, suggesting an upward-bounded dose-effect curve and rapid and lasting tolerance formation for the drug's desired, but not its undesired, effects (19).

But the European experience indicates that this picture must have been too optimistic. Perhaps the tolerance and bounded dose-effect curve applies to the drug's more subtle emotional effects and not to its raw stimulant effect; if so, those in search of mere stimulation may find that dose-stacking and frequent use give them the effects they were seeking. Reports in the US that dance-party participants are mixing MDMA with alcohol may foreshadow the development of a similar set of problems in that country. The disapproval of alcohol use among MDMA purists may reflect alcohol's interference with its "empathogenic" and "entactogenic" effects; by contrast, the use of alcohol with stimulants - starting with Café Royale - is as common as it is dangerous.

An alternative explanation for the use of MDMA in frequencies and dosages inconsistent with the model of rapid tolerance formation would be Pavlovian conditioning or secondary reinforcement; especially in subcultures that value intoxication, early favorable experiences with MDMA may be sufficient to maintain, and even intensify, usage in the face of diminishing desired effects.

It is possible to imagine low-damage scenarios. The sort of drug use pattern reported by Beck and Rosenbaum, one without any equivalent of the small number of extremely heavy users who account for a large share both of black-market revenues and of drug-related harms in the markets for cocaine and heroin, would mean that even widespread use might coexist with relatively limited problems. A drug that does not generate either compulsive use or catastrophic bad behavior can remain a very minor source of problems even if it has a large user base. Methamphetamine, which is much cheaper and more easily available than MDMA, might tend to crowd its more expensive and difficult-to-obtain competitor from the market for sheer stimulation. If that were the case - or at least if that were the case among those likely to begin non-medical MDMA use as a consequence of their own medical use of it or reports of such use by others, or of research results involving MDMA - then the substance abuse term in the decision-making equation might be negligibly small.

But that rosy picture is not the only possible one. If therapeutic use, and even therapeutic-use research, can influence the non-medical use of drugs such as MDMA, that potential process is a legitimate issue in deciding whether to approve research efforts and New Drug Applications. But the nature and extent of the interaction are not self-evident; regulatory decisions should, to the extent possible, be based on concrete knowledge rather than speculative fears. That then leads us to inquire into the optimal design of a research program - pharmacological, behavioral, and ethnographic - into existing patterns of non-medical use.

It would be a mistake to ignore the possibility that ideas as well as chemicals might leak across the legal barrier between medical and non-medical usage. But it would equally be a mistake to assume, without evidence, that banning research or medical use is necessary, or even valuable, in preventing the spread of non-medical use of a drug with the established reputation, user base, and distribution structure of MDMA. These are researchable questions, and that research ought to go forward in parallel with clinical and pharmacological studies.

Footnotes

1. Greer G, Tolbert R. A Method of Conducting Therapeutic Sessions with MDMA. Journal of Psychoactive Drugs. 30 (1998) 4:371-380.

2. Dr Pedro Sopelana, Dr. Maria Angeles and Jose Carlos Bouso, U. of Autonoma de Madrid, Spain; Dr. Moshe Kotler and Dr. Adam Darnell, Ben-Gurion University of the Negev, Israel.

3. Dr. Charles Grob, Harbor-UCLA Medical School, United States.

4. Sannerud, CA, Brady, JV, and Griffiths, RR: Self-injection in Baboons of Amphetamines and Related Designer Drugs. In Asghar, K. and Desouza, F. (Eds.), Pharmacology and Toxicology of Amphetamines and Related Designer Drugs - NIDA Research Monograph #94, 30-42, 1989.

5. McCann UD, , Szabo Z, Scheffel U, Dannals RF, Ricaurte GA. Positron emission tomographic evidence of toxic effect of MDMA ("Ecstasy") on brain serotonin neurons in human beings. Lancet (1998) 352: 1433-37. Bolla, K.I.; McCann, U.D.; and Ricaurte, G.A. Memory impairment in abstinent MDMA ("ecstasy") users. Neurology 51:1532-1537, 1998. Hatzidimitriou, G.; McCann, U.D.; and Ricaurte, G.A. Altered serotonin innervation patterns in the forebrain of monkeys treated with MDMA seven years previously: Factors influencing abnormal recovery. Journal of Neuroscience 191(12):5096-5107, 1999. McCann, U.D.; Mertl, M.; Eligulashvili, V.; and Ricaurte, G.A. Cognitive performance in 3,4- methylenedioxymethamphetamine (MDMA, "ecstasy") users: a controlled study. Psychopharmacology 143:417-425, 1999. See also the review article Mathias, R. "Ecstasy" Damages the Brain and Impairs Memory in Humans. NIDA Notes 14 (4):10- 11,1999.

6. See Adamson S and Metzner, R. " The Nature of the MDMA Experience and its Role in Healing, Psychotherapy and Spiritual Practice." In Revision 10 (1988) 4: 59-72.

7. Watson, L and Beck, J: New Age Seekers: MDMA Use as an Adjunct to Spiritual Pursuit. J. of Psychoactive Drugs 23(3): 261-270. 1991

8. For a recent discussion of the ethical aspects of clinical research with MDMA, see three articles; Lieberman J and Aghajanian G. "Caveat Emptor: Researcher Beware." Editorial in Neuropsychopharmacology. 21 (1999) 4:471-473. Gilsman G, Vrekes R, van Gervan J, Cohen A, (Center for Human Drug Research, Leiden, The Netherlands) "MDMA Study." in Neuropsychopharmacology. 21 (1999) 4:597. Vollenweider F, Gamma A, Liechti M, Huber T. "Is A Single Dose of MDMA Harmless?" in Neuropsychopharmacology 21 (1999) 4: 598-600.

9. IRB's were formally required by the Declaration of Helsinki, adopted by the 18th World Medical Assembly, Helsinki, Finland, 1964, and as revised by the 29th World Medical Assembly, Tokyo, Japan, 1975. HHS regulations governing IRBs can be found at 45 C.F.R. ßß 46.101-46.124 (1998). FDA regulations for IRBs are at 21 C.F.R. ßß. 312, 314.

10. From 1985-1987, the US Food and Drug Administration (FDA) rejected all applications to conduct human studies with MDMA, citing the overriding need to protect human subjects from the theoretical risks of MDMA neurotoxicity. These rejections occurred while the the US Drug Enforcement Administration (DEA) was conducting administrative hearings on the proper scheduling of MDMA. A central issue in the hearings was whether MDMA had a "currently accepted medical use," an issue that might have been influenced by whether FDA-approved research in humans was taking place. A June 26, 1992 letter to Dr. Charles Grob from Dr. Lee Zwanziger, Executive Secretary, FDA Drug Abuse Advisory Committee, with background material written by Dr. Curtis Wright, Medical Review Officer of FDA's Pilot Drug Evaluation Staff, discusses four of the rejections by Dr. Paul Leber, Director, FDA Division of Neuropharmacological Drug Products. A fifth application, not mentioned in Dr. Wright's background material, was rejected in a March 5, 1987 letter from Dr. Paul Leber to Dr. Francisco Di Leo, for IND #27,281 The rejection was sustained in a March 18, 1987 letter from Dr. Robert Windom, Assistant Secretary for Health, to Rick Doblin.

11. Beck, J: The Public Health Implications of MDMA Use. In SJ Peroutka (Ed.), Ecstasy: The Clinical Pharmacological and Neurotoxicological Effects of the Drug MDMA. Norwell, MA: Kluwer Academic Publishers, 1990.

12. Schmidt CJ; Abbate GM; Black CK; Taylor VL. Selective 5-hydroxytryptamine receptor antagonists protect against the neurotoxicity of methylenedioxymethamphetamine in rats. Journal of Pharmacology and Experimental Therapeutics (1990) 255(2): 478-83. Also, Schmidt CJ; Taylor VL; Abbate GM; Nieduzak TR 5-HT2 antagonists stereoselectively prevent the neurotoxicity of 3,4-methylenedioxymethamphetamine by blocking the acute stimulation of dopamine synthesis: reversal by L-dopa. Journal of Pharmacology and Experimental Therapeutics (1990) 256(1): 230-5.

13. Adamson, S. ed. Through the Gateway of the Heart. San Francisco: Four Trees Publications. 1985. Also, Greer R and Tolbert R. Subjective Reports of the Effects of MDMA in a Clinical Setting. Journal of Psychoactive Drugs. 18 (1986) 4:319-327.

14. By contrast, in regulating the production and distribution of approved pharmaceuticals, FDA and DEA, pursuant to standards set by law, do explicitly consider the risks that manufacturers, pharmacies, physicians and patients will divert controlled substances for non-medical use or illicit sale.

15. Newmeyer, JA: Some Considerations on the Prevalence of MDMA Use. J. of Psychoactive Drugs 18(1986) 4: 361-362.

16. Dowling, GP: Human Deaths and Toxic Reactions Attributed to MDMA and MDEA. In SJ Peroutka (Ed.), Ecstasy: The Clinical Pharmacological and Neurotoxicological Effects of the Drug MDMA. Norwell, MA: Kluwer Academic Publishers, 1990

17. Dr. Franz Vollenweider reports that his administration of MDMA to MDMA- naïve subjects in the context of clinical research has not led subjects to self- administer MDMA. Follow-up studies are currently underway to investigate this issue scientifically.

18. See information about MDMA on NIDA Homepage at www.drugabuse.gov

19. Beck J and Rosenbaum M. Pursuit of Ecstasy: The MDMA Experience. Albany, State University of New York Press. 1994.