Abuse deterrent formulations and the Controlled Substances Act (CSA)

Drug and Alcohol Dependence 83S (2006) S23–S30

Abuse deterrent formulations and the Controlled Substances Act (CSA)夽 Frank L. Sapienza ∗ The Drug and Chemical Advisory Group, LLC, 4911 Heversham Court, Fairfax VA 22032, USA Received 11 August 2005; received in revised form 17 November 2005; accepted 21 November 2005

Abstract The Controlled Substances Act (CSA) has reduced the diversion of controlled substances at the manufacturing and distribution levels. Recent increased diversion has occurred at the retail level. Levels of diversion and abuse of controlled substances with similar abuse potential and therapeutic indications often parallel availability for medical use, while rates of diversion and abuse may be influenced by factors related to specific products, including their formulations and risk management plans. Abuse deterrent formulations may reduce abuse and attendant adverse health consequences even if the products are diverted. Their development should consider how, to what extent and by whom products containing the targeted substance are abused. It should take into consideration all potential types of abuse including “as is”, multiple doses, alternate routes of administration, physical or chemical separation of the active ingredient, compromised extended release mechanisms and abuse in combination with other substances. Industry incentives for developing abuse-resistant formulations include enhanced corporate image and potentially less restrictive scheduling or risk management plans. Scheduling is substance specific, but the CSA includes products/formulations that are differentially scheduled. Issues to be considered for differential scheduling under the CSA include: (1) whether there is legal authority to do so; (2) application of standard scheduling criteria to individual products; (3) product specific data for “eight factor analyses”; (4) development of predictive data and standards accepted by the scientific and regulatory communities; (5) use of predictive data or post marketing surveillance data; (6) international treaty obligations. These issues must be addressed before differential scheduling can be considered. © 2006 Elsevier Ireland Ltd. All rights reserved. Keywords: Abuse deterrent formulations; Controlled Substances Act; Drug scheduling; Differential scheduling

1. Introduction The Controlled Substances Act (CSA) is part of a comprehensive legislative package enacted in 1970 in response to the widespread and growing drug abuse problem in the United States. The CSA, which is Title II of the Comprehensive Drug Abuse Prevention and Control Act of 1970, consolidated several drug control laws and for the first time established a single system of controls for all abuseable substances (narcotics, stimulants, depressants and hallucinogens).1 It provides the legal framework for the federal government’s strategy against drug abuse and is primarily enforced by the Department of Justice

夽

This article is part of a supplemental issue of the journal devoted entirely to papers on how abuse liability of medications is affected by their formulation for medical use. ∗ Tel.: +1 703 239 8871 (O)/978 7921 (R); fax: +1 703 978 0893. E-mail address: [email protected]. 1 Anabolic steroids were legislatively added to the categories of controlled substances under the CSA in 1990 (Public Law 101-647, November 29, 1990). 0376-8716/$ – see front matter © 2006 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.drugalcdep.2005.11.028

with most authority delegated to the Drug Enforcement Administration (DEA). There is also a complementary role for agencies of the Department of Health and Human Services (DHHS), most notably the Food and Drug Administration (FDA). The CSA addresses problems associated with the manufacture, distribution and abuse of substances with no recognized medical use in the United States (e.g. cannabis, heroin and MDMA) as well as those with currently accepted medical uses in the U.S. (e.g. morphine, amphetamine and diazepam). Since many controlled drugs are necessary to maintain the health and general welfare of the American people, the CSA has a two-fold purpose regarding legitimate medicines: to maintain an adequate and uninterrupted supply of these controlled substances to meet legitimate needs while simultaneously reducing their diversion and abuse. The three international drug control treaties (Single Convention on Narcotic Drugs, 1961, Convention on Psychotropic Substances, 1971, and United Nations Convention Against Illicit Traffic in Narcotic Drugs and Psychotropic Substances, 1988), also stress this balance between the supply of and demand for controlled substances.

S24

F.L. Sapienza / Drug and Alcohol Dependence 83S (2006) S23–S30

2. CSA controls The CSA was enacted at a time when the diversion and abuse of pharmaceutical products containing controlled stimulants and depressants were rampant. Some estimates of the diversion of amphetamines from legitimate channels were as high as 50%; the BNDD (Bureau of Narcotics and Dangerous Drugs)2 estimated that about 20% of legitimate production was diverted (Baumgartner and Morrell, 1972). Much of this abuse and diversion was associated with insufficient controls at the manufacturer and distributor levels (GAO, 1982). The CSA addressed these problems by establishing a closed system of distribution for all who handle controlled substances. Substances were placed into one of five schedules based on their relative potential for abuse, accepted medical use, and relative dependence potential. These schedules determine the level of control, with substances in Schedules I and II being subject to the most stringent control measures. Substances in Schedules I and II have a high potential for abuse while substances in Schedules III–V have progressively less abuse potential. Substances in Schedule I have no accepted medical use and no accepted safety for use under medical supervision while substances in Schedules II–V have accepted medical uses. Abuse of substances in Schedule II leads to severe psychological or physical dependence while abuse of substances in Schedules III–V leads to moderate to low/limited physical or psychological dependence. The type of substance (narcotic or non-narcotic) and type of registrant (manufacturer, distributor, practitioner, researcher), also have an effect on the level of control. Narcotic substances and manufacturers/importers/exporters are subject to more stringent controls than the others. The key controls of the CSA are: (1) scheduling the substance, (2) registration of all handlers, (3) import/export permits or declarations, (4) quotas, (5) record keeping, reporting, monitoring and tracking, (6) security and (7) investigations and penalties. These controls are focused at the manufacturer/importer/exporter levels. Regulation of practitioners and pharmacies is left primarily to states and professional boards. As of 3 May 2005, there were approximately 1743 manufacturers, importers, exporters and distributors registered with the DEA and more than one million retail level registrants (retail pharmacy, hospital/clinic, practitioner, teaching institution and mid-level practitioner) (DEA Office of Diversion Control, 2005a,b,c). DEA is required by regulation to conduct annual inspections of all bulk manufacturers and importers of controlled substances in Schedules I and II (21 C.F.R. 1316.13, 2004). Administrative inspections of distributors, importers and exporters of controlled substances in Schedules II–V and manufacturers of controlled substances in Schedules III–V are conducted as circumstances dictate. DEA does not conduct routine inspections of practitioners; however, all DEA registrants are

2 By Executive Order No. 11727, July 10, 1973, President Nixon merged a number of drug control related agencies, including the Bureau of Narcotics and Dangerous Drugs (BNDD) into the newly created Drug Enforcement Administration (DEA) (38 FR 18357, 1973).

subject to unscheduled or complaint investigations if they are suspected to be in violation of the CSA. DEA conducts relatively few investigations at the practitioner level. For example, in FY 2003 DEA took action against less than one tenth of one percent of the registered one million doctors, with only 14 arrests (DEA Statistical Information, 2005). With the enactment of the CSA, DEA focused its activities on the upper levels of the drug distribution system where controlled substances are produced and handled in the largest quantities and are most widely distributed. During the 1960s and 1970s, diversion was widespread at the manufacturer/distributor levels (GAO, 1982). CSA controls led many of those contributing to the oversupply, diversion and abuse of stimulants and depressants to leave the market place or alter their practices to conform to the new regulations. Implementation of CSA controls, regulations and policies were quite successful in eliminating much of the opportunity for diversion of controlled substances at the manufacturer/distributor levels (GAO, 1982). Subsequently, diversion from the retail level became the major source of diversion. Based on estimates and reported drug thefts, DEA estimated that diversion of controlled substances from the retail level accounted for approximately 80–90% of the diversion (GAO, 1978). Today, diversions from the retail level and through the internet are the most significant sources of abused prescription drugs (CASA, 2005). The most common sources of diversion include: (1) indiscriminate or illegal prescribing, (2) forged or fraudulent prescriptions, (3) doctor shopping, (4) thefts, burglaries and in-transit losses, and (5) the internet. Although the DEA investigates retail level diversion, much of the oversight at this level rests with a wide array of state and local law enforcement authorities and medical and pharmacy boards. The CSA controls, because of their focus on the wholesale level, are not sufficient to address much of the retail level diversion. For example, quota requirements for bulk and dosage form manufacturers were particularly effective in reducing the overproduction and its associated diversion at the manufacturer and distributor levels, but they have not had a similar effect on diversion from the retail level. The CSA requires the DEA to establish quotas for each Schedule I and II controlled substance annually based on an estimate of legitimate medical need. Quotas set maximum limits on the total amount of each Schedule I and II controlled substance that can be manufactured in the United States (aggregate production quotas) and divides this amount among bulk manufacturers (individual manufacturing quotas) and ultimately dosage form manufacturers (procurement quotas) on a substance by substance basis. It has been suggested that quotas for the most highly abused substances or products should be reduced by an estimate of the amount diverted in order to reduce diversion at the retail level. There are several inherent difficulties with this approach. First, each product containing a Schedule II controlled substance for which a quota is granted has been approved by the FDA and marketing and promotion of these products are subject to FDA approved labeling and, in some cases, risk management plans. Within these parameters, physicians may prescribe and dispense, and companies may market and promote, these substances. Second, it is extremely difficult

F.L. Sapienza / Drug and Alcohol Dependence 83S (2006) S23–S30

to quantify the amount of diversion of a substance and even more difficult to determine the amount of a specific product that has been diverted. Third, the established quotas take into account the amount of material that is dispensed. This, in turn, determines the amount that is distributed and ultimately manufactured. There is no true estimate of legitimate medical need for individual substances; there is only an estimate of use that may include the amounts that are diverted and abused. Legitimate medical need is primarily determined by prescriptions which in essence are sales. All prescriptions are included in these estimates since it is often difficult to separate valid prescriptions from those that are illegal or issued inappropriately. Fourth, and most importantly, there is no way to ensure that a reduction in a specified quota will only apply to that which is diverted. It is entirely possible that the reduction will affect the availability of a substance/product for those who have a legitimate need for it. While quotas can set limits on the amount of a Schedule I or II substance that can be manufactured in a given year, they cannot ensure that controlled medicines get to those who need them and not to those who may divert and abuse them. 3. Abuse deterrent formulations Since the demand for products containing abuseable substances, particularly those with high abuse potential (Schedule II substances), is inherently greater than the legitimate need for them, it is unreasonable to expect that diversion of these products can be eliminated. As long as abuseable products are available, there will be attempts to divert them and some will be successful. It is reasonable to expect that diversion and abuse of substances in the same therapeutic category and schedule, and having similar pharmacology, would be similarly affected by availability. In other words, the more of such a substance that is available, the more diversion and abuse there will be. Between 1992 and 2002, prescriptions for controlled substances increased 12 times faster than the population and almost three times faster than the prescriptions for non-controlled substances. Although this increase undoubtedly improved medical care, it was also accompanied by increased diversion and abuse (CASA, 2005). Despite prevention, education and law enforcement efforts, chronic abusers continue to obtain and abuse controlled products. The ongoing search for new molecular entities with little or no abuse potential that can be used in medicines to treat conditions such as acute and chronic pain, Attention Deficit Hyperactivity Disorder (ADHD), anxiety and sleep disorders should continue to be aggressively pursued. Nevertheless, there is a need to develop products containing abuseable substances that provide greater deterrent to diversion and abuse than existing products. These products must maintain effectiveness, safety and ease of use. In addition to the goal of reduced diversion and/or abuse, successful abuse deterrent products may encourage more physicians to prescribe these medicines for more patients, thus resulting in increased treatment of pain and other conditions. They may also promote better compliance for patients and, in some cases, reduce the adverse health consequences associated with abuse. The ultimate goal is to significantly increase the difficulty for abusers to obtain the sought after effects of the

S25

controlled substance. If successful, they may reduce the illicit demand for pharmaceutical products containing controlled substances, force chronic abusers into treatment and reduce the availability of abuseable drug products for new users. In addition to addressing a public health problem and contributing to public safety, companies that develop successful abuse deterrent products may benefit from an enhanced corporate image and financial gains. There is a perception among some that the success of a product containing a controlled substance is related to its level of control under the CSA compared to similar products; thus there is considerable interest in the possibility of controlling abuse deterrent products and formulations less stringently (in a lower schedule) than the substance and current products. The following discussion will explore some of the issues relevant to the potential differential scheduling of products or formulations under the CSA. 4. Administrative scheduling The CSA provides an administrative mechanism for the Attorney General (delegated to the Deputy Administrator of the DEA) to add or transfer substances to a schedule if the substance has a potential for abuse and meets the criteria for a particular schedule and to remove a substance from the schedule if the substance does not meet the criteria for inclusion in any schedule (21 U.S.C. 811(a) 2005). As described earlier, placement in a particular schedule is based on the substance’s relative abuse potential, whether or not it has a currently accepted medical use in treatment in the U.S., and its physical and psychological dependence potentials. The CSA requires that eight specific factors be considered in determining the appropriate schedule for a substance or whether a substance should be decontrolled. They are: (1) actual or relative potential for abuse; (2) scientific evidence of the drug’s pharmacological effects; (3) the state of current scientific knowledge regarding the substance; (4) its history and current pattern of abuse; (5) the scope, duration and significance of abuse; (6) what, if any, risk there is to the public health; (7) the drug’s psychic or physiological dependence liability; (8) whether the substance is an immediate precursor of a substance already controlled (21 U.S.C. 811(C) 2005). Both the DEA and DHHS must consider these factors, including information that is predictive of abuse and dependence as well as actual abuse and dependence data. For newly marketed substances, predictive data are generally more important, while for the substances that have been available for some time, both predictive and actual abuse data are important. The legislative history of the CSA endorsed drug control (scheduling) as a preventive measure and stressed that the government “should not be required to wait until a number of lives have been destroyed or substantial problems have already arisen before designating a drug as subject to controls of the bill” (U.S. House of Representatives, 1970).3

3

The Supreme court has cautioned against using legislative history to interpret statutory text (EXXON MOBIL CORP. v. ALLAPATTAH SERVICES INC., et al., 2005), however it has also noted that it can be a useful guide to the intent of

S26

F.L. Sapienza / Drug and Alcohol Dependence 83S (2006) S23–S30

The administrative scheduling process involves the DEA, various agencies of the DHHS (primarily the FDA and National Institute on Drug Abuse (NIDA)), and interested parties (industry, consumer groups, the medical/health care communities, state/local government agencies and individual citizens). The DEA receives scheduling petitions, reviews them for legal and administrative sufficiency, collects data, considers the factors in 21 U.S.C. 811(C) and requests from the DHHS a scientific and medical evaluation and scheduling recommendation. It may also initiate investigations into the scheduling of a substance on its own. The DHHS initiates scheduling reviews for drugs going through the New Drug Application (NDA) process and is the scientific and medical authority for any scheduling action. The DHHS evaluations are binding on the DEA with respect to scientific and medical matters. DHHS’ scheduling recommendations are binding on the DEA if it recommends that a substance may not be controlled. Otherwise, DEA makes the final determination of whether a substance should be controlled and in what schedule; it also has responsibility for defending these actions in legal proceedings. Interested parties may petition the DEA to initiate a scheduling or decontrol action. Interested parties may comment on scheduling proposals published in the Federal Register and request and participate in administrative hearings on scheduling proposals. Although each administrative scheduling action is unique, all are ultimately consistent with scientifically verified and legally defensible data. Substances may also be scheduled legislatively (e.g. anabolic steroids) and to meet the requirements of international treaties (e.g. tilidine) without consideration of the eight factors and three criteria. 5. Differential scheduling The term differential scheduling is not defined or used in the CSA. However, it is being used in regulatory and scientific discussions to refer to the control of a substance in one schedule and the control of one or more formulations or products containing that substance in one or more different schedules. The CSA is generally substance specific, but the provisions of the CSA applicable to scheduling and the criteria for scheduling refer to a “drug or other substance” (21 U.S.C. 811 and 812, 2005). The definition of drug under the CSA is the same as that under the federal Food, Drug and Cosmetic Act which generally refers to a product. The CSA also requires that the Secretary of DHHS forward to the DEA information regarding the abuse potential of a drug having a stimulant, depressant or hallucinogenic effect on the central nervous system at the time that a new-drug application is submitted (21 U.S.C. 811(f)). The use of the term “drug” and not substance and reference to “new-drug applications” suggests that the possibility of differential scheduling of products was recognized at the time the CSA was enacted. Each “drug or other substance” under consideration for a differential schedul-

Congress (Wisconsin Public Intervenor v. Mortier, 501 U.S. 597, 617 (1991)). References to legislative history in this document are not meant to interpret statutory language, but to provide some background on the Congressional intent regarding the CSA.

ing action, however, must satisfy each of the three criteria for control in a specific schedule after consideration of the eight factors. This means that a product containing a controlled substance must be shown to have lower abuse potential and lead to less dependence than the substance itself if it is to be controlled in a lower schedule. 6. Differential control under the CSA The CSA as enacted in 1970 provides examples of specifically listed substances in one schedule and formulations containing them in different schedules. The CSA also provides a mechanism to exempt certain formulations and products from regulations. With one exception, specific criteria for these differential controls were either contained in the CSA or subsequently developed by expert panels. The most familiar examples of differential control involve the narcotic substances codeine, dihydrocodeinone (hydrocodone), dihydrocodeine, ethylmorphine, opium and morphine. Each of these substances is in Schedule II of the CSA. However, formulations/products containing limited amounts of these substances in combination with an isoquinoline alkaloid or one or more active non-narcotic ingredients in recognized therapeutic amounts were included in Schedule III or V of the CSA (Public Law 91-513, Section 202). Similarly, diphenoxylate was included in Schedule II while products containing limited amounts of diphenoxylate in combination with specified amounts of atropine sulfate were included in Schedule V. In these cases, the CSA specifies the criteria for Schedules II, III or V control. It is interesting to note that no criteria were established for the differential control of the Schedule II opioid oxycodone and its combination products. Thus, oxycodone and all marketed single entity and combination products are in Schedule II. The CSA included methamphetamine in Schedule III in 1970; however, it listed any injectable liquid containing any quantity of methamphetamine, including its salts and isomers, in Schedule II, indicating an understanding of the importance of the route of administration in evaluating abuse potential (Public Law 91-513, Section 202). However, it was also recognized that both bulk and dosage forms of these substances could be easily transformed into an injectable solution and that both forms of methamphetamine have the same abuse potential. The first administrative scheduling action under the CSA transferred methamphetamine and amphetamine from Schedules III to II (36 FR 12734, 1971). Of interest is the fact that two manufacturers claimed that their products (Eskatrol® , Biphetamine® and Biphetamine-T® ) should not be transferred into Schedule II, proposing, among other things, that these formulations, had less abuse potential than the amphetamines and requested a hearing on the transfer of these products. Following informal conferences between BNDD and the companies in which information on actual and relative potential for abuse of these products were exchanged, both companies dropped their requests for a hearing and the three products were also transferred to Schedule II (36 FR 13686, 1971; 36 FR 15744, 1971). The CSA allows the DEA, by regulation, to exempt a mixture or preparation containing a non-narcotic controlled substance

F.L. Sapienza / Drug and Alcohol Dependence 83S (2006) S23–S30

from the application of all or part of the regulatory provisions of the CSA if the mixture or preparation is approved for prescription use, and if it contains one or more other active ingredients which are not listed in any schedule and which are included in such combination, quantity, provision, or concentration to vitiate the potential for abuse of the product (21 U.S.C. 811(g) 2005). The exemption process is separate from and different than the administrative scheduling process. Exemptions may be granted following DEA review of product specific applications against specific criteria. Applications must include the qualitative and quantitative composition of the product, a description of the product with complete labeling, reported instances of abuse, reported and anticipated adverse effects, and the number of dosage units produced over the preceding 2 years. For products not covered by specific criteria, applications must also include, for each ingredient separately and for the combination, a summary of the pharmacology of the product, including preclinical and clinical evaluations and studies, with an emphasis on dependence potential. The criteria for these exemptions were developed by a panel of Public Health Service Physicians and FDA medical officers in 1967 and were published by the DEA in 1980 (45 FR 74091, 1980). The criteria were based upon the concept of combining a controlled substance with an amount of another active medicinal ingredient in sufficient quantities to cause early deterrent side effects. Specific ratios of the stimulant or depressant substance to the deterrent substance were developed and based upon daily threshold values for physical or psychological dependence. The most current list of “exempt prescription drugs” (1 July 2003) contains more than 300 individual products, the vast majority of which contain a barbiturate in combination with an analgesic or Bella Donna alkaloid (DEA Office of Diversion Control, 2005a,b,c). DEA may revoke any exemption following the procedures set forth in the regulations (21 C.F.R. 1308.31(c), 2004). The foregoing examples strongly suggest that the CSA was drafted in such a way to allow for the differential control of products/formulations containing controlled substances in combination with other active ingredients that are likely to reduce abuse. 7. Administrative differential scheduling actions There have been more than 150 administrative scheduling actions under the CSA since 1970, only a handful of which involved differential scheduling (DEA Office of Diversion Control, 2005a,b,c). Although the number of differential scheduling actions is small, the reasons and basis for them may provide some additional insight into future attempts at using this mechanism. Administrative differential scheduling actions involved both combination products and single entity products, and actions required to comply with international treaty obligations and legislative mandates. Differential control of some combination products followed the concept described in the previous section that the addition of certain active medicinal ingredients to a controlled substance lowers the abuse potential of the product. The differential control of difenoxin in Schedule I and of com-

S27

bination products containing limited amounts of difenoxin in combination with atropine sulfate per dosage unit in Schedules IV and V is representative of this type of differential control (43 FR 27560, 1978; 43 FR 38382, 1978) and consistent with the differential control of codeine and other opioids described above. The administrative rescheduling of three short-acting barbituric acid derivatives (amobarbital, pentobarbital and secobarbital) from Schedules III to II without the transfer of combination products containing these substances is another example (38 FR 14289, 1973; 38 FR 31310, 1973). BNDD noted that almost all of the abuse of these short-acting barbiturates involved single entity preparations or combinations of two derivatives of barbituric acid. The final rule also specified that any suppository dosage form approved by the FDA and containing one of the transferred barbituric acid derivatives would remain in Schedule III (38 FR 31310, 1973). Thus, a lack of actual abuse as determined by BNDD and Department of Health, Education and Welfare (DHEW) investigations was used as a means to differentiate certain barbiturates and single entity products containing them from certain combination products. Not all products containing potentially abuse deterrent ingredients have been differentially scheduled. Combination products containing a narcotic antagonist such as naloxone with an opioid have been considered as a potential means of deterring opioid abuse at least by those who would abuse them by the intravenous route of administration. There was significant diversion and abuse of pentazocine, a Schedule IV controlled substance, often in combination with an antihistamine, by the intravenous route of administration in the 1980s. In order to reduce this type of abuse, the existing product was reformulated to include pentazocine and naloxone (Talwin NX® ). More recently buprenorphine was formulated with naloxone (Suboxone® ) for use in treatment of opioid addiction. In both cases the single entity and combination products are controlled in the same schedule. There were no formal petitions submitted to differentially schedule either combination product, but several commenters to the recent buprenorphine rescheduling action raised the issue that single entity buprenorphine and buprenorphine with naloxone should be differentially scheduled. DEA, after its review of the comments and all relevant factors and after discussions with DHHS, concluded that when all at-risk populations and all potential patterns of abuse were considered, there was no difference between the abuse potentials of the single entity and combination products. The DEA recognized the potential value of the combination product in deterring intravenous abuse by physically dependent abusers. However, it noted that injection of the combination product in other populations will likely result in opioid agonist effects and that use of the combination by the sublingual route of administration would not be a deterrent, particularly to na¨ıve users (67 FR 62354, 2002). Although DEA noted that it encouraged the development of products that deter abuse in selective populations, it could not make the findings necessary to differentially schedule the combination product. This illustrates an important point in the consideration of any scheduling review, particularly differential scheduling; all potential populations, all routes of administration and patterns of abuse must be considered.

S28

F.L. Sapienza / Drug and Alcohol Dependence 83S (2006) S23–S30

There have been a few examples of the differential control of single entity products. The differential control of dextropropoxyphene and its preparations in 1980 was necessary for the U.S. to comply with international treaty obligations. The Commission on Narcotic Drugs (CND) voted to place dextropropoxyphene into Schedule II of the Single Convention and certain preparations for oral use into Schedule III. All forms of dextropropoxyphene were already in Schedule IV of the CSA. DEA concluded that the most appropriate method to control dextropropoxyphene and its preparations to meet international treaty obligations was to control the bulk substance dextropropoxyphene in Schedule II so that it was subject to quotas and maintain preparations containing dextropropoxyphene in Schedule IV (45 FR 48881, 1980). Since the CSA provides that control of substances required by obligations under the Single Convention be made without regard to the findings and proceedings for administrative scheduling actions, this differential control action did not involve an evaluation of the abuse or dependence potentials of the substance and its preparations. It was strictly a legal action. The differential control of gamma-hydroxy butyrate (GHB) in Schedule I and FDA approved products containing GHB in Schedule III was congressionally mandated (Public Law 106172, 2000). The law mandated DEA to place GHB into Schedule I immediately and, if and when a new drug application was approved, to place that product into Schedule III of the CSA. DEA was required to complete the scheduling action without regard to the findings and procedures of the administrative scheduling process. This action, although a form of differential scheduling, was not based on the difference between the abuse potentials of the substance and the product; instead it tried to differentiate the abuse potential of a legitimately manufactured, distributed and dispensed medicine with a very rigid risk management plan from that of an illicitly manufactured substance. Probably the most often cited example of differential scheduling of a single entity product is that of dronabinol (the (−)-delta9 trans isomer of tetrahydrocannabinol (THC)) in Schedule I and the FDA approved product containing dronabinol (Marinol® ) in Schedule III. DEA placed Marinol® into Schedule II upon FDA approval and after receipt of a scientific and medical evaluation and scheduling recommendation from the DHHS (51 FR 17476, 1986). All other stereochemical variants and formulations of THC remained in Schedule I. This action was based on findings that this FDA approved product has a currently accepted medical us in treatment in the U.S. and a high abuse potential. Subsequently, the DEA received a petition to transfer Marinol® from Schedules II to III. The petition cited as justification for rescheduling the need for better patient access to Marinol® and a lack of actual abuse since its introduction into the market. The petition further noted that separating the active ingredient from the formulation is very difficult, and therefore effectively limits abuse to the oral route of administration, indicative of a lower abuse potential. Following extensive reviews by DEA and DHHS, DEA transferred Marinol® (as described above) from Schedules II to III (63 FR 59751, 1998; 64 FR 35928, 1999). DEA concluded that, although, the pharmacological and behavioral effects of dronabinol substance are comparable to those of

delta9 -THC, suggesting a high abuse potential, there are several factors that contribute to a lower abuse potential. There was little evidence of the actual diversion or abuse of Marinol® despite increases in prescriptions since its introduction into the market in 1985 and the approval of an additional indication in 1992. Additionally, the DEA found that the product’s formulation in sesame oil, the difficulty in separating the active ingredient from the formulation which limits its abuse to the oral route of administration, and its delayed onset of behavioral effects by the oral route of administration, supported a finding of a lower abuse potential relative to substances in Schedule II. It is important to note that the transfer to Schedule III applied only to FDA approved products, which at the time of this writing is limited to Marinol® . This is the most straightforward example of the differential scheduling of a single entity formulation based on considerations of the eight factors listed in 21 U.S.C. 811(C) and the findings required by 21 U.S.C. 812. This action occurred at a time when the federal government was looking for alternative medicines in response to those who refused to accept that smoked marijuana is not medicine as determined by U.S. laws and regulations. Transferring Marinol® from Schedules II to III arguably increases access to this alternative medicine since five prescription refills and oral prescriptions are allowed under federal law for Schedule III drugs. The above scheduling actions illustrate that, although differential scheduling may be possible under the CSA, they have occurred only in unique situations. 8. Differential control issues Abuse potential is the threshold criteria for control under the CSA and includes both actual abuse and potential for abuse (21 U.S.C.811(c)). Although the CSA does not specifically define the term abuse potential, the legislative history of the CSA provides a definition taken from the Drug Abuse Control Amendments (DACA) to the Food, Drug and Cosmetic Act (section 201(v)) and its regulations as follows: (1) There is evidence that individuals are taking the drug or drugs containing such a substance in amounts sufficient to create a hazard to their health or to the safety of other individuals or of the community; or (2) There is significant diversion of the drug or drugs containing such a substance from legitimate drug channels; or (3) Individuals are taking the drug or drugs containing such as substance on their own initiative rather than on the basis of medical advice from a practitioner licensed by law to administer such drugs in the course of his professional practice; or (4) The drug or drugs containing such a substance are new drugs so related in their action to a drug or drugs already listed as having a potential for abuse to make it likely that the drug will have the same potentiality for abuse as such drugs, thus making it reasonable to assume that there may be significant diversions from legitimate channels, significant use contrary to or without medical advice, or that it has a substantial capa-

F.L. Sapienza / Drug and Alcohol Dependence 83S (2006) S23–S30

bility of creating hazards to the health of the user or to the safety of the community” (U.S. House of Representatives, 1970). The above definition shows that a determination of the abuse potential of a drug or other substance can be based upon a predictive element, evidence of actual abuse/diversion, or both. For newly marketed drugs, the predictive element has been extremely important. Years of research have resulted in the development of animal models to predict the abuse liability of individual substances (Balster and Bigelow, 2003; Griffiths et al., 2003). The results of these paradigms have been compared with and, in many cases, been successfully verified by human studies and experience. Currently there are generally accepted standards utilized by scientists and government regulators for the determination of the abuse liability of many types of substances. Usually, substances are tested against a prototypic substance in a similar pharmacologic or therapeutic class. This has been a tremendous help to regulators in determining whether a drug or substance should be controlled under the CSA, and, in some cases, the particular schedule based on relative abuse liability, before actual abuse occurs. However, similar paradigms are not generally available or accepted that differentiate the abuse liability among formulations or products containing the same controlled substance. This raises difficulties in evaluating the abuse potential of potential abuse deterrent formulations and products. It is important to understand how, by whom, by what route of administration, and under what conditions, current products containing the targeted controlled substances are abused. However, the CSA requires that all potential patterns of abuse and populations be considered when evaluating the abuse potential of a drug or other substance for administrative scheduling. Currently, many potential abuse deterrent formulations are being designed to minimize abuse by certain routes of administration, by some populations, in selected patterns of abuse or to reduce adverse effects if abused in certain ways. Although the quantity of active ingredient in a controlled product generally has little effect on its scheduling status, there has been tremendous concern over significant abuse of high dose extended release opioid products. This has led to the development of a number of formulations that attempt to make it difficult to defeat the controlled release mechanisms in these products so that high doses of the active ingredient cannot be isolated and injected or snorted. A review of such formulations under the CSA, however, also requires an evaluation of whether or not the formulation could be abused in other ways. Some products are abused in unaltered form by the labeled route of administration, either in single or multiple doses (e.g. hydrocodone and oxycodone combination products; methylphenidate and benzodiazepines). Even controlled release opioids products may be abused unaltered and orally. Some products are abused by alternate routes of administration after minimal physical manipulation (e.g. snorting immediate release methylphenidate products). In other cases, abusers attempt to remove the active ingredient from formulations for intravenous or other use. Physical separation techniques such as crushing and grinding, either alone or in combination with chemical separa-

S29

tion techniques, such as solvent extraction, acid/base extraction, or simple reactions, have been used. Studies that evaluate possible physical and chemical separation methods for individual formulations and products may be useful in determining the likelihood of defeating controlled release mechanisms in these and other formulations. The potential for abuse deterrent formulations to be abused in combination with other substances or products, to enhance the abuse potential of either substance, should also be considered. The ingenuity of the “street chemists and pharmacologists” must never be underestimated, particularly in light of the use of the Internet for the rapid and wide spread dissemination of information on abuseable substances, products and formulations. Animal models and human studies that are developed to predict the relative abuse liability of individual formulations and products should be pursued, but their verification against human experience over time is important for them to be fully accepted by regulatory authorities for potential differential scheduling purposes. The second component of the term “abuse potential” for scheduling purposes is actual abuse. Actual abuse is not a necessary condition for control under the CSA. A lack of substantial actual abuse alone is not sufficient for decontrol or decreased control of a substance with abuse potential warranting control under the CSA. Nevertheless, establishing a lack of substantial abuse of marketed products or formulations over time would be a critical, and arguably a requisite factor for the differential control of a product or formulation. There are a number of difficulties in trying to substantiate a lack of abuse of a particular product or formulation for purposes of differential scheduling. First, most current drug abuse indicator systems are generally substance specific and are not set up to track individual products with any degree of reliability. The legislative history of the CSA indicates that law enforcement data, including forensic laboratory analyses, and morbidity and mortality data should be considered in evaluating actual abuse (U.S. House of Representatives, 1970). Forensic laboratories, however, generally identify substances, not products. Toxicologists identify substances (including metabolites) and not products. Identifying specific products or formulations involved in abuse episodes often requires further investigation of each instance of abuse, diversion, trafficking, morbidity or mortality. Current indicator systems are not set up to do this. Specific programs to effectively track the diversion and abuse of individual products are needed. Second, it is necessary to establish that any lower level of actual abuse is the result of a lower abuse potential and not the result of some other phenomena, such as lower or restricted availability, current CSA controls and rigid risk management plans. The potential impact of reduced controls on the product or formulation must be balanced against any potential increased level of abuse. Third, experience has shown that substantial actual abuse of new products often is not observed for some time after marketing. Post-marketing surveillance programs should be established for a minimum of 3 years to obtain meaningful information. Fourth, standards for measuring amounts of abuse and diversion for comparison with other drugs or other substances have not been validated and accepted. Since scheduling always involves a determination of the “relative” abuse potential of drugs or other

S30

F.L. Sapienza / Drug and Alcohol Dependence 83S (2006) S23–S30

substances, comparison of the abuse and abuse potential of individual products or formulations is critical for such evaluations. Other issues that must be considered in potential differential scheduling actions include consistency with U.S. international treaty obligations, the impact on control by the States of substances and products and the impact on law enforcement and criminal justice matters. If products are scheduled differently than the substance, identification of specific products involved in a violation of the CSA becomes critical. Forensic laboratories may have difficulty in confirming the identification of specific products or formulations. Differentially scheduled products, formulations and substances could also lead to confusion regarding appropriate criminal charges, penalties and sentencing issues. Although resolution of these difficulties is not insurmountable, each must be addressed for differential scheduling to be considered. 9. Conclusion Abuse deterrent formulations hold promise as a tool to reduce the problems associated with prescription drug diversion and abuse. Products utilizing these formulations may be more difficult to abuse; others can reduce the adverse effects associated with abuse; while others can make the products less susceptible to diversion. The development of abuse deterrent formulations is consistent with the spirit of the CSA. Incentives for the development of such products include an enhanced corporate image as well as financial rewards. Although the CSA contains a number of examples of differentially scheduled formulations, there are significant hurdles to overcome if the current administrative process and criteria are to be used for new abuse deterrent products. Each of the factors to be considered and the criteria necessary for differential control of a formulation or product must be rigidly applied by regulatory authorities. Cautious introduction of potentially abuse deterrent products with properly focused risk management plans coupled with extensive post marketing surveillance for at least several years is an effective means of validating predictive data that these formulations or products have a lower abuse potential than the substance itself or other products containing it. References Balster, R.L., Bigelow, G.E., 2003. Guidelines and methodological reviews concerning drug abuse liability assessment. Drug Alc Depend. 70 (Suppl. 3), S13–S40. Baumgartner, K.C., Morrell, M.X., 1972. Pharmaceutical industry regulation by the department of justice. Syracuse Law Rev. 23 (3), 785–815. Center on Addiction and Substance Abuse at Columbia University, 2005. Under the Counter: The diversion and Abuse of Controlled Prescription Drugs in the U.S., July.

DEA Office of Diversion Control, 2005. Controlled Substance Schedules (as viewed May 27, 2005). http://www.deadiversion.usdoj.gov/ schedules/schedules.htm. DEA Office of Diversion Control, 2005. Registrant Population (as viewed May 27, 2005). http://www.deadiversion.usdoj.gov/pop/summary.htm. DEA Office of Diversion Control, Exempt RX List, 2005. Exempt Prescription Products List (as viewed May 27, 2005). http://www. deadiversion.usdoj.gov/schedules/exempt/exempt rx list 070103.pdf. DEA Statistical Information, 2005. Office of Enforcement Pharmaceutical Investigations Section. Government Accounting Office, 1978. Report to the Congress. Retail diversion of legal drugs—a major problem with no easy solution. GGD-78-22. March 10. Government Accounting Office, 1982. Report to the Congress of the United States. Comprehensive approach needed to help control prescription drug abuse. GAO/GGD-83-2. October 28. Griffiths, R.R., Bigelow, G.E., Ator, N.A., 2003. Principles of drug abuse liability assessment in laboratory animals. Drug Alc Depend. 70 (Suppl. #), S55–S72. Public Law 91-513, 1970. The Comprehensive Drug Abuse Control Prevention Act of 1970. October 27. Public Law 101-647, 1990. Title XIX—Anabolic Steroids Control Act of 1990. November 29. Public Law 106-172, 2000. Hillory J. Farias and Samantha Reid Date—Rape Drug Prohibition Act of 2000. February 18. U.S. House of Representatives, 1970. House Report 91-1444. 91st Congress 2d Session, pp. 1–70. 21 C.F.R. 1308.31(c), 2004. Title 21 Part 1300 to End Revised as of April 1, 2004. Food and Drugs. Office of the Federal Register, National Archives and Records Administration. 21 C.F.R. 1316.13, 2004. Title 21 Part 1300 to End Revised as of April 1, 2004. Food and Drugs. Office of the Federal Register, National Archives and Records Administration. 21 U.S.C. 811(a). Federal Criminal Code and Rules as amended to January 21, 2005. Title 21, Chapter 13. Drug Abuse Prevention and Control. Thomson West Publisher. 21 U.S.C. 811(C). Federal Criminal Code and Rules as amended to January 21, 2005. Title 21, Chapter 13. Drug Abuse Prevention and Control. Thomson West Publisher. 21 U.S.C. 811(g). Federal Criminal Code and Rules as amended to January 21, 2005. Title 21, Chapter 13. Drug Abuse Prevention and Control. Thomson West Publisher. 21 U.S.C. 812. Federal Criminal Code and Rules as amended to January 21, 2005. Title 21, Chapter 13. Drug Abuse Prevention and Control. Thomson West Publisher. 36 FR 12734, 1971. Federal Register, vol. 36, 12734, July 7. 36 FR 13686, 1971. Federal Register, vol. 36, 13686, July 23. 36 FR 15744, 1971. Federal Register, vol. 36, 15744, August 18. 38 FR 14289, 1973. Federal Register, vol. 38, 14289–14290, May 31. 38 FR 18357, 1973. Federal Register, vol. 38, 18357, July 10. 38 FR 31310, 1973. Federal Register, vol. 38, 31310–31311, November 13. 43 FR 27560, 1978. Federal Register, vol. 43, 27560, June 26. 43 FR 38382, 1978. Federal Register, vol. 43, 38382, August 28. 45 FR 74091, 1980. Federal Register, vol. 45, 74091–74092, November 7. 45 FR 48881, 1980. Federal Register, vol. 45, 48881, July 22. 51 FR 17476, 1986. Federal Register, vol. 51, 17476–17578, May 13. 63 FR 59751, 1998. Federal Register, vol. 63, 59751–59753, November 5. 64 FR 35928, 1999. Federal Register, vol. 64, 35928–35930, July 2. 67 FR 62354, 2002. Federal Register, vol. 67, 62354–62370, October 7.