Does phentermine induce or provoke relapse of substance abuse? Statement of the Issue: FDA drug labeling and the PDR state that phentermine is contraindicated in patients with a history of drug abuse. The FDAs hypothesis is that phentermine, a category IV controlled substance, prescribed for such patients can provoke relapse of drug abuse (first hypothesis). A corollary hypothesis, which must also be true for the first hypothesis and the FDAs contraindication to be valid, is that phentermine has significant human addiction potential (second hypothesis). Origin of the Issue & Historical perspective: Phentermine was approved for marketing by the FDA in 1959. At the time phentermine was approved, far less was understood of the nature of stimulant drug abuse than is the case today. At that time, the favored hypothesis was that stimulant abuse potential was directly correlated to stimulant potency. No suggestion of abuse or dependence had appeared in any of the clinical trials conducted prior to approval. However, a series of animal studies in rats beginning in the 1950s and continued into the 1960s which examined changes in spontaneous activity showed that phentermine increased spontaneous activity more than placebo.⁽¹⁾ Even though there was complete absence of clinical evidence in humans, the FDA relied on the animal data, and assuming stimulation in rats equated to abuse potential in humans, classified phentermine as a category IV controlled substance. In reading the FDA labeling in the 2010 edition of the PDR, which differs from the original 1959 labeling in only a few details, it is clear that the FDA has always assumed that phentermine has exactly the same risks and adverse effects as amphetamine. In 1959 commonly agreed upon criteria for diagnosis of drug abuse, dependence and addiction had not been established. The first edition of the Diagnostic and Statistical Manual of Mental Disorders (DSM-I), published in 1952 listed drug addiction as a diagnosis but neither discussed nor provided criteria for the diagnosis. Several drugs of addiction were listed but amphetamine was not included; indeed the nature of amphetamine abuse, addiction, and withdrawal was poorly understood in 1959. At that time modern addiction medicine methodology and psychometric testing were in their infancy and there were no standardized examination techniques for addiction. After the 1962 passage of Kefauver-Harris amendments to the Food, Drug and Cosmetics Act, the FDA reevaluated the anorectic drugs. For ten years thereafter, as the FDA continued the re-evaluation the major perceived risk of the anorectics was addiction. Although the amphetamines clearly posed a risk of addiction, the addictive potential of the amphetamine congeners had not been thoroughly studied in humans and remained open to debate. Based on structural similarities and some anecdotal evidence some of the expert panel considering the drugs believed that phentermine also had a risk for abuse and addiction. Eventually in 1973 a compromise was reached in the debate and the anorectic drugs including phentermine were re-approved with the proviso that the labeling restrict use to a few weeks and that the drugs be prominently labeled to warn about the risk of addiction.⁽²⁾ Currently, as a result of numerous investigations into the nature of substance abuse, addiction medicine specialists have developed clearly defined criteria for substance abuse disorders. These criteria did not exist in 1959 when phentermine was approved, or in 1973 when the FDA debate was settled reaffirming phentermine was addicting and only emerged with the publication of the DSM-IV in 2000. The term addiction has been replaced with dependence and abuse. Substance dependence is defined as a maladaptive pattern of substance use, leading to clinically significant impairment or distress, as manifested by three (or more) of the following, occurring at any time in the same 12-month period: 1) Tolerance markedly increased amounts needed. 2) Withdrawal syndrome characteristic for the substance. 3) Substance taken in higher amounts or for longer duration than intended. 4) Persistent desire or unsuccessful attempts to cut down or control use. 5) A great deal of time spent in obtaining the substance (e.g. seeing multiple doctors, driving long distances), using the substance, or recovering from its effects. 6) Important activities are given up or reduced because of substance use. 7) Substance use is continued in the face of significant social, family, occupational, or legal problems caused or exacerbated by use. Substance abuse is defined as a maladaptive pattern of substance use, leading to clinically significant impairment or distress, as manifested by one (or more) of the following, occurring at any time in the same 12-month period: 1) Recurrent use resulting in failure to fulfill major role obligations at work, school or home. 2) Recurrent use in situations where it is physically hazardous. 3) Recurrent use-related legal problems. 4) Continued use despite persistent social or inter-personal problems. A diagnosis of substance dependence is of greater severity than substance abuse. Substance abuse should not be diagnosed in any patient with a prior diagnosis of dependence. Patients who have substance dependence who abruptly cease taking the substance develop a withdrawal syndrome characteristic of the substance involved. Intense cravings for the substance are a cardinal feature of substance dependence and withdrawal.⁽³⁾ The substances of abuse have been classified into categories; amphetamines are classified as stimulants with cocaine being the prototypical stimulant of abuse. Addiction medicine specialists have also developed psychometric tests for the diagnosis and severity assessment of substance abuse disorders.^(3-7) Psychometric tests have recently been used to accurately describe the features of amphetamine abuse, dependence and withdrawal.^(8-10) Currently the only investigation in the peer-reviewed medical literature (in press) which has examined the addiction potential of phentermine using modern addiction medicine methodology is the one we recently did in this practice.⁽¹¹⁾ Methods Used for Evidence Searches: The National Library of Medicine currently catalogues every paper published in the peer reviewed medical literature. PubMed, the NLM search engine, now has 19,000,000 articles catalogued and accessible online. Papers as far back as 1928 are included. Every reference in this memo is catalogued there. Any bibliographies developed using PubMed search algorithms may be stored in a personal account with automatic email notification when new papers are added to the bibliography. The result is that anyone, anywhere in the world, has instant access to the resources of the National Library of Medicine. Evidence for the two hypotheses: There is no evidence in the peer-reviewed medical literature which supports the first hypothesis that phentermine therapy can provoke relapse in patients with a history of substance abuse. Indeed, there is little in the literature to support the second hypothesis that phentermine has addiction potential in humans. On the other hand, many textbooks repeat the FDA conjectures but none of these offer any supporting evidence or cite supporting peer-reviewed literature other than animal studies, primarily in rodents, measuring stimulant activity or self-administration data. More recent animal studies in primates do not support the notion that phentermine has addictive potential.⁽¹²⁾ In fact, in studies in rhesus monkeys, phentermine is effective in decreasing cocaine self-administration suggesting it may be a useful drug in treating cocaine addiction.^(13-14) PubMed searches, limited to humans and excluding animal studies, for phentermine AND abuse, phentermine AND dependence, phentermine AND withdrawal, found only one anecdotal report of phentermine addiction, which predated the publication of diagnostic criteria for dependence, abuse, and withdrawal.⁽¹⁵⁾ These searches found a few other papers, reviews focused on drug safety, which discussed phentermine but included no data.^(16-17) The studies the FDA has relied upon in asserting both hypotheses must be considered indirect evidence since the studies were all animal studies. Thus it seems that although it is common knowledge that phentermine is addicting, direct evidence in support of this knowledge is nowhere to be found. The few papers that exist which might support the addiction potential hypothesis in human subjects are anecdotal.^(18-19) Evidence against the two hypotheses: 1. There have been numerous phentermine clinical trials reported in the peer-reviewed medical literature most which have been re-examined several times in meta-analyses.^(20-24) Adverse side effects were monitored in each trial. There were no signs of abuse, dependence or withdrawal reported in any of the clinical trials. This is direct evidence contradicting the second hypothesis. 2. A recent investigation, conducted in the authors practice, of the addiction potential of phentermine involved studying symptoms occurring in patients in a medical weight loss program who abruptly ceased taking phentermine after having been on long-term phentermine pharmacotherapy. If phentermine does indeed have addictive properties, one would expect that such patients would experience symptoms similar to those seen in amphetamine addicts when such subjects abruptly cease amphetamine use at a treatment center. Amphetamine cessation in addicts has been studied and psychometric scales devised to assess withdrawal severity.⁽²⁵⁾ The study found that long-term phentermine pharmacotherapy in the context of a weight management program did not induce amphetamine-like withdrawal symptoms. The study also indicated that long-term phentermine pharmacotherapy did not induce phentermine cravings in any degree. This was true even in patients who had been on phentermine therapy for as long as eleven years and in patients on doses higher than recommended in the package insert.^{(11, 26)} This report of symptoms following abrupt phentermine cessation is direct evidence that phentermine has little or no addiction potential; direct evidence in human subjects contradicting the second hypothesis. 3. The fact that phentermine pharmacotherapy does not induce cravings for phentermine is highly significant since intense cravings for the substance are cardinal features of substance dependence and withdrawal. The author, in another on-going study, not as yet published, is assessing cravings using the Tiffany Cocaine Craving Questionnaire⁽²⁷⁾ modified for phentermine and has found that phentermine treated subjects do not develop phentermine cravings. This too is direct evidence that the second hypothesis is not true. 4. Attention Deficit Disorder is typically treated with amphetamine drugs, category II controlled substances. There has been concern that treatment with category II drugs might induce substance abuse in treated subjects. Wilens and colleagues at Massachusetts General Hospital and Harvard Medical School has studied this question extensively and has concluded that, contrary to expectations, patients treated with amphetamine have a lower, not higher, incidence of drug abuse when compared to untreated patients.^(28-33) Given these the results of these studies in amphetamine, a drug known to much more potent than phentermine in stimulant strength and in creating adverse side effects, it seems unlikely that the first hypothesis is correct for phentermine, a category IV drug. It is true that this is indirect evidence, but it seems compelling never-the-less. If amphetamine therapy produces a decrease in incidence of substance abuse, how could the weaker drug, phentermine, produce an increase in substance abuse? These studies in ADD patients provide indirect evidence contradicting the first hypothesis. 5. Some of the acknowledged experts in obesity treatment have commented in published reviews that if phentermine indeed has addiction potential, it is vanishingly low and has not been a problem in clinical practice.^(34-36) The author of another report discussing long-term phentermine use (more than ten years) noted that he had not observed abuse or dependence even in patients who had been on phentermine as long as 40 years.⁽³⁷⁾ This too is indirect evidence against the second hypothesis. Even though this is indirect evidence I would argue clinical observations by experts with long experience with phentermine should be more convincing than studies in rodents. 6. My own experience has been that a prior history of drug abuse is not an absolute contraindication to phentermine. Patients who are currently abusing or are addicted to substances do not often present themselves to a physician for weight loss they are too preoccupied with their addiction to be concerned with being overweight. Stimulant addicts typically select much stronger stimulants than phentermine as their drug of choice. Cocaine, methamphetamine, and amphetamine are likely choices. Phentermine may be abused by stimulant addicts in desperation if their drug of choice isnt available to them or phentermine may be added to a cocktail of a variety of compounds in an attempt to achieve a stronger stimulant mix. Although one old report from a methadone treatment program indicated that 22% of urine samples from their patients contained traces of phentermine,⁽¹⁸⁾ recent informal surveys of drug treatment centers indicate that phentermine abuse or dependence is not seen or treated at such centers today. I have not encountered a patient currently addicted to stimulants in 20 years of practicing bariatric medicine in a busy weight loss clinic. Patients with a prior history of substance abuse who are currently abstinent do present themselves for weight loss. Some of these never admit their prior abuse to the physician. Some patients do reveal their past experience after varying intervals of treatment. My experience has been that when these patients are treated with phentermine they never have a return of cravings for the substance they once abused even when the substance was amphetamine, or the potently addicting amphetamine congeners such as methamphetamine, crystal methamphetamine, or 3,4-methylenedioxymethamphetamine (MDMA or ecstasy). I have treated perhaps two dozen such patients with phentermine who have at first denied prior substance abuse or dependence. As these patients have confessed to a prior substance disorder I have quizzed them for signs or re-emergence of cravings for the previous substance and for phentermine. If there are no such symptoms, I continue phentermine pharmacotherapy. These patients have not developed phentermine-induced substance abuse nor have they relapsed into substance abuse or dependence. The fact that some patients at first deny prior illicit drug abuse suggests that an unknown number of patients, unsuspected of being former addicts, are unwittingly treated with phentermine in weight loss practices across the U.S. The fact that this practice has never emerged as a problem is also compelling indirect evidence that the first hypothesis cannot be true. In my experience, phentermine does not provoke or induce cravings for either phentermine or substances of abuse. Patients who admit to prior substance abuse, dependence or withdrawal can be treated with phentermine provided the treating physician is convinced that the patient is trustworthy, but this must be done with care and caution. When phentermine treatment is undertaken the physician must continuously be watchful for the emergence of cravings for other drugs and for drug seeking behavior; should these appear, phentermine should be discontinued. The patient should be informed that phentermine pharmacotherapy for anyone with prior substance disorders is controversial and if the patient chooses to discontinue phentermine the patients wishes should be respected. My experience is anecdotal and therefore must be considered indirect evidence contrary to the second hypothesis. The fact that not a single instance has been reported in which phentermine therapy provoked relapse of a substance abuse disorder must also be considered indirect evidence contrary to the first hypothesis. 7. The clinical features of the abuse, dependence, and withdrawal syndromes of the drugs of addiction are described in the DSM-IV. Although the syndromes for stimulants of abuse are described including cocaine, amphetamine, methamphetamine, methylphenidate, and others, the DSM-IV does not mention phentermine. In fact, no syndromes of abuse, dependence or withdrawal for phentermine have ever been described even though it has been widely available for 50 years. This is another piece of indirect evidence contradicting the second hypothesis. 8. One final point is that, after 50 years of widespread use, the absence of published evidence in support of the first hypothesis is indirect evidence that the hypothesis is false. The same is true of the second; after 50 years of use there are no credible reports of phentermine addiction providing indirect evidence the second hypothesis is false. Discussion: The type and severity of adverse effects produced by any drug is influenced by the dose, route and duration of administration. Subjects who abuse amphetamine and other stimulants typically self-administer the drug by either the inhalational or intravenous route. These routes of administration typically produce very high plasma drug levels very quickly,⁽³⁸⁾ increasing the probability of producing serious adverse effects, such as addiction. Patients taking oral stimulants, as prescribed by a physician, simply do not achieve the high plasma drug levels needed to produce serious adverse effects. This fact likely explains the overall very low rate of serious adverse effects seen with the use of stimulants for the treatment of obesity and also of attention deficit disorder.⁽³⁹⁾ Patients in weight management programs are typically seen often and prescribed only enough phentermine to last until the next visit. The fact that they do not self-administer the drug by inhalation or injection is an important aspect of the question of the addiction potential of phentermine one that has been completely ignored in the controversy. The addiction potential of phentermine must be examined within the context of a weight management program whether or not substance abusers might add it to a stimulant cocktail should not be part of the argument. Summary: The existing evidence regarding the two hypotheses is summarized in Table 1 and Table 2 on the following page.

Table 1. Summary of Evidence

First Hypothesis: Phentermine given to a patient with prior substance abuse will provoke relapse of drug abuse

Evidence For/against	Evidence type	Evidence	Published Literature
FOR	Direct	None	No
	Indirect	None	No
AGAINST	Direct	None	No
	Indirect	Attention Deficit Disorder patients treated with amphetamines have lower incidence of substance abuse than do untreated ADD patients	Yes
	Indirect	No reported instances of phentermine treatment provoking relapse of substance abuse	No
	Indirect	Not seen in clinical experience	No

Table 2. Summary of Evidence

Second Hypothesis: Phentermine has significant human addiction potential.

Evidence For/Against	Evidence Type	Evidence	Published Literature
FOR	Direct	None	No
	Indirect	Older animal studies, 1950 to ~1980	Yes
AGAINST	Direct	Multiple human clinical trials, 1975 -2006	Yes
	Direct	Abrupt Phentermine Cessation Study, 2010	Yes
	Direct	Long-term Phentermine Rx does not induce phentermine cravings, no matter dose or duration, 2010	Yes
	Indirect	Modern animal studies	Yes
	Indirect	Not seen in anyones clinical experience	Yes
	Indirect	No credible reports of human addiction	No

Note that that there is absolutely no evidence, direct or indirect, to support the first hypothesis! The FDA admonition that prior substance abuse contradicts phentermine therapy is pure conjecture; no evidence supports the conjecture. Conclusions: Phentermine pharmacotherapy for obesity does not provoke substance abuse in patients with prior substance abuse. Nor does phentermine induce substance abuse in patients with no prior history of substance abuse. Phentermine pharmacotherapy for obesity does not induce phentermine substance abuse. The existing evidence does not support either hypothesis. The first hypothesis is unproven and is therefore false. The second hypothesis is unproven and is therefore false. Why hasnt the FDA changed the labeling? The FDA will never change phentermine labeling until compelled to do so. A more cogent question then is what must happen before the FDA will consider changing phentermine labeling? The answer: at least three things. First: More modern investigations into the human addiction potential of phentermine must be undertaken and published. More primate studies are needed as well. The preponderance of evidence the FDA has relied upon is evidence from older animal studies. Until the preponderance of evidence, both animal and human, clearly contradicts the second hypothesis, the FDA will not act. Second: At least one large scale, long-term, randomized, double blinded phentermine trial which satisfies current FDA Guidance must be undertaken and published. My understanding is that a group of investigators recently applied for a NIH grant to do this only to have the application rejected (personal communication, K.M Gadde, October 2009). The group is preparing to re-apply. Because phentermine is widely used as the agent of choice obesity treatment, a long-term trial investigating effectiveness and safety is in the public interest and should be undertaken. This is the major critical requirement which must be satisfied before the FDA will act. A large scale prospective trial designed to test the first hypothesis would be a difficult undertaking but is also needed. Third: The FDA must be presented with a petition to asking that phentermine labeling be changed. This should not be undertaken until the first two requirements are satisfied; otherwise the FDA will undoubtedly reject the petition. References: 1. Rossum JMv, Simon F. Locomotor activity and anorexigenic action. Psychopharmacologia 1969, 14 248-254. 2. Colman E. Anorectics on trial: A half century of federal regulation of prescription appetite suppressants. Ann Intern Med 2005, 143 (5), 380-385. 3. Heinz AJ, Epstein DH, Schroeder JR, et al. Heroin and cocaine craving and use during treatment: Measurement validation and potential relationships. J Subst Abuse Treat 2006, 31 (4), 355-364. 4. Kampman KM, Volpicelli JR, McGinnis DE, et al. Reliability and validity of the cocaine selective severity assessment. Addictive behaviors 1998, 23 (4), 449-461. 5. Ahmadi J,Kampman K, Dackis C. Outcome predictors in cocaine dependence treatment trials. Am J Addict 2006, 15 (6), 434-439. 6. Ahmadi J, Kampman K, Dackis C, et al. Cocaine withdrawal symptoms identify "Type b" Cocaine-dependent patients. Am J Addict 2008, 17 (1), 60-64. 7. Ahmadi J, Kampman KM, Oslin DM, et al. Predictors of treatment outcome in outpatient cocaine and alcohol dependence treatment. Am J Addict 2009, 18 (1), 81-86. 8. McGregor C. Amphetamine withdrawal:Nature, time course, and treatment; ph.D. Thesis. Adelaide: University of Adelaide; 2005. 9. McGregor C, Srisurapanont M, Jittiwutikarn J, et al. The nature, time course and severity of methamphetamine withdrawal. Addiction (Abingdon, England) 2005, 100 (9), 1320-1329. 10. McGregor C, Srisurapanont M, Mitchell A, et al. Psychometric evaluation of the amphetamine cessation symptom assessment. J Subst Abuse Treat 2008, 34 (4), 443-449. 11. Hendricks EJ, Greenway FL. Symptoms observed after abrupt phentermine cessation 223-p. Obesity 2009, 17 (Suppl 2), S111. 12. Alexander M, Rothman RB, Baumann MH, et al. Noradrenergic and dopaminergic effects of (+)-amphetamine-like stimulants in the baboon papio anubis. Synapse 2005, 56 (2), 94-99. 13. Wojnicki FHE, Rothman RB, Rice KC, et al. Effects of phentermine on responding maintained under multiple fixed-ratio schedules of food and cocaine presentation in the rhesus monkey. The Journal of pharmacology and experimental therapeutics 1999, 288 (2), 550-560. 14. Rothman RB,Blough BE, Baumann MH. Dual dopamine/serotonin releasers as potential medications for stimulant and alcohol addictions. AAPS Journal 2007, 9 (1), E1-10. 15. Valenta LJ. Letter: Phentermine addiction and thyroxine abuse associated with hypothyroidism. Am J Psychiatry 1975, 132 (5), 569. 16. Ioannides-Demos LL, Proietto J, Tonkin AM, et al. Safety of drug therapies used for weight loss and treatment of obesity. Drug Saf 2006, 29 (4), 277-302. 17. Greenway FL, Caruso MK. Safety of obesity drugs. Expert Opin Drug Saf 2005, 4 (6), 1083-1095. 18. Jain NC,Budd RD, Sneath TC. Frequency of use or abuse of amphetamine-related drugs. The American journal of drug and alcohol abuse 1979, 6 (1), 53-57. 19. Hoffman BF. Diet pill psychosis: Follow-up after 6 years. Canadian Medical Association journal 1983, 129 (10), 1077-1078. 20. Haddock CK, Poston WS, Dill PL, et al. Pharmacotherapy for obesity: A quantitative analysis of four decades of published randomized clinical trials. International journal of obesity and related metabolic disorders 2002, 26 (2), 262-273. 21. Ioannides-Demos LL,Proietto J, McNeil JJ. Pharmacotherapy for obesity. Drugs 2005, 65 (10), 1391-1418. 22. Li Z, Maglione M, Tu W, et al. Meta-analysis: Pharmacologic treatment of obesity. Ann Intern Med 2005, 142 (7), 532-546. 23. Weintraub M. Long-term weight control study: Conclusions. Clin Pharmacol Ther 1992, 51 (5), 642-646. 24. Kim KK, Cho HJ, Kang HC, et al. Effects on weight reduction and safety of short-term phentermine administration in korean obese people. Yonsei Med J 2006, 47 (5), 614-625. 25. McGregor C, Srisurapanont M, Mitchell A, et al. Psychometric evaluation of the amphetamine cessation symptom assessment. J Subst Abuse Treat 2007. 26. Hendricks EJ, Greenway FL. A study of abrupt phentermine cessation in patients in a weight management program. American Journal of Therapeutics 2010, In Press, Manuscript AJT-20082603. 27. Tiffany ST, Singleton E, Haertzen CA, et al. The development of a cocaine craving questionnaire. Drug and alcohol dependence 1993, 34 (1), 19-28. 28. Biederman J, Monuteaux MC, Spencer T, et al. Do stimulants protect against psychiatric disorders in youth with adhd? A 10-year follow-up study. Pediatrics 2009, 124 (1), 71-78. 29. Biederman J, Spencer TJ, Wilens TE, et al. Long-term safety and effectiveness of mixed amphetamine salts extended release in adults with adhd. CNS Spectr 2005, 10 (12 Suppl 20), 16-25. 30. Wilens TE. Attention deficit hyperactivity disorder and substance use disorders. Am J Psychiatry 2006, 163 (12), 2059-2063. 31. Wilens TE, Adamson J, Monuteaux MC, et al. Effect of prior stimulant treatment for attention-deficit/hyperactivity disorder on subsequent risk for cigarette smoking and alcohol and drug use disorders in adolescents. Arch Pediatr Adolesc Med 2008, 162 (10), 916-921. 32. Wilens TE, Adamson J, Sgambati S, et al. Do individuals with adhd self-medicate with cigarettes and substances of abuse? Results from a controlled family study of adhd. Am J Addict 2007, 16 Suppl 1 14-21; quiz 22-13. 33. Wilens TE, Faraone SV, Biederman J, et al. Does stimulant therapy of attention-deficit/hyperactivity disorder beget later substance abuse? A meta-analytic review of the literature. Pediatrics 2003, 111 (1), 179-185. 34. Pi-Sunyer FX. Use of lifestyle changes treatment plans and drug therapy in controlling cardiovascular and metabolic risk factors. Obesity 2006, 14 (suppl_3), 135S-142. 35. Bray GA, Greenway FL. Current and potential drugs for treatment of obesity. Endocr Rev 1999, 20 (6), 805-875. 36. Bray GA. Use and abuse of appetite-suppressant drugs in the treatment of obesity. Ann Intern Med 1993, 119 (7_Part_2), 707-713. 37. Frank A. The long-term management of obesity with continuing pharmacotherapy. Obes Res 2004, 12 (11), 1821-1827. 38. Evans SM,Cone EJ, Henningfield JE. Arterial and venous cocaine plasma concentrations in humans: Relationship to route of administration, cardiovascular effects and subjective effects. The Journal of pharmacology and experimental therapeutics 1996, 279 (3), 1345-1356. 39. Rothman RB, Hendricks EJ. Phentermine cardiovascular safety. American Journal of Emergency Medicine 2009, 27 1010-1013.