What Will We Learn from the FDA Clinical Trials Process and What Will We Still Want to Know About Immunotherapies and Depot Medications to Treat Drug Dependence?
Thomas R. Kosten
Yale University School of Medicine
Henry R. Kranzler
University of Connecticut Health Center, Farmington
The National Academies created the Committee on Immunotherapies and Sustained-Release Formulations for Treating Drug Addition to examine issues related to the development of immunotherapies and depot medications targeted to treat drugs of abuse. This appendix was commissioned to examine the stage-wise strategy for completing clinical trials that will be part of the Food and Drug Administration (FDA) process for ensuring the safety and efficacy of these medications. The medications currently available for human use include vaccines for active immunizations against cocaine and nicotine and long- acting depot formulations of naltrexone, an opiate antagonist for alcoholism and opiate dependence. Monoclonal antibodies for passive immunotherapy are still in animal testing, but one for phencyclidine should be ready for human use within 2 years. The clinical trials to test these medications may involve individuals from three major categories: (1) addicts who overdose, (2) drug-dependent individuals who either volunteer for the medication or are mandated to use it by another agent to prevent relapse, and (3) nondependent persons who either volunteer or are inducted to receive the medication as a protection against initiating or increasing substance use (i.e., primary or second-
ary prevention, respectively). Particular attention is given to safety considerations of immunotherapies and depot medications, recognizing that some patients will continue to abuse various psychoactive substances and that these medications may be administered to pregnant women, adolescents, or children.
The FDA clinical trials process is designed to ensure safety and efficacy for specific uses or indications for new medications but not for off-label use in new diseases or in patient populations in which the medication was never studied. This appendix reviews the four phases of the FDA clinical trials process as it is likely to be implemented for the immunotherapies and depot medications currently in clinical or preclinical development. These products include depot formulations of naltrexone for alcohol and, potentially, opioid dependence; vaccination for cocaine and nicotine dependence; and monoclonal antibodies for phencyclidine (PCP), methamphetamine, and possibly cocaine. Also reviewed are the three types of treatment protocols: overdose, relapse prevention, and protection. Overall, the purpose here is to consider what might be learned during the FDA clinical trials process to inform later applications of these therapies and postmarketing experience. The surveillance that is an intrinsic part of the postmarketing experience should help to discourage premature examination of applications that may place certain populations at unacceptable risk.
TYPES OF INTERVENTION PROTOCOLS
The three types of treatment protocols—overdose, relapse prevention, and protection—are most suitably tested with different types of medication approaches: active immunization, passive immunization with monoclonal antibodies, or depot medications (Klein, 1998; Blaine et al., 1994). For example, overdose protocols most usefully employ monoclonal antibodies (passive immunization), while active immunization and depot medications have, at most, a limited role for this indication. Relapse prevention protocols can usefully employ any of these medication approaches, but different limitations exist for each approach. Protection protocols are the most speculative at present but could also use any of these three approaches. However, protection protocols must consider medical safety and frequency of administration as critical issues, since the individuals intended to benefit from these treatments do not require treatment for substance dependence. In addition, not all of the three types of intervention protocols are applicable to every abused substance. Overdose and relapse prevention protocols are very likely to be studied before these agents are approved for protection, but a protocol for protection from addiction is not likely to get controlled study because of several prac-
tical issues, such as the cost of such a long-term study in even a relatively high-risk group.
Protection protocols also are likely to be aimed at adolescents, since it is during adolescence that the majority of experimentation with substances is initiated and the potential for protection is greatest. However, protection protocols in adolescents may have three broad risks. First, they may result in medical harm to the adolescent, which will be covered in this appendix. Second, there may be psychological or social harm to the child-parent relationship resulting from parents “forcing” their adolescents to get treatments against their will or in a manner that harms parent-child trust. Third, there may be a misplaced biological focus for any antagonist protection in adolescents where much of the incentive to use illicit drugs or even tobacco is related to social, not pharmacological, effects. Adolescents want to impress peers, demonstrate rebelliousness to their parents, signal membership in a clique or subculture, and generally assert a social message. Pharmacological reinforcement of a drug may be a secondary motivation for use. Consequently, inhibiting this pharmacological reinforcement will have little effect on such motivations to use the substance and be much less cost effective than alternative protection strategies.
A typical overdose protocol might use monoclonal antibodies to reverse an acute overdose of a drug such as PCP (Owens and Mayershohn, 1986; Valentine et al., 1996). However, since monoclonal antibodies can last up to several months, it is important that safety be considered in two areas (Proksch, Gentry, and Owens, 2000). First, if an individual is dependent on the overdosed substance, withdrawal will occur after the overdose is reversed, and this withdrawal will not be suppressed by treatment with the usual modest doses of a long-acting agonist from the same pharmacological class as the targeted overdose drug. Very large doses of the agonist might be required to overcome the antagonism produced by the antibody treatment. Second, when the patient who has recovered from the overdose leaves the emergency department, he or she will continue to have a relative blockade of the abused substance. Any attempts by the patient to override this blockade could lead to the use of large amounts of an abused substance. The effects of any adulterants included in an illicit street drug would be magnified by this more intensive self-administration. Thus only a single dose of the medication would be needed to provide acute treatment, but aftercare would be critical because of the potential for the intervention to be long lasting. Using monoclonal fragments (Fab) rather than the complete humanized antibodies will be an impor-
tant consideration for overdose reversal, since these fragments have considerably shorter half-lives and should have minimal efficacy within 24 hours, rather than lasting several weeks, as is typical of the complete antibody.
Nonetheless, the economic advantages of this type of intervention could be substantial if a single monoclonal antibody injection can keep a patient from entering an intensive care unit, at an expected cost of more than $1,000 per day. Thus the cost of treatment (approximately $2,000) would have to be justified in part by the cost of continued medical care (i.e., the high cost of a day in an emergency room or several days in intensive care). The aftercare costs for substance abuse relapse prevention (discussed below) after starting monoclonal treatment or after an intensive care unit stay should be the same, although the patient will be able to enter this aftercare much more quickly following reversal of the acute overdose by the monoclonal.
Relapse Prevention Protocols
A typical relapse prevention protocol might use any of the three types of agents to enhance compliance with treatment. The psychosocial backbone of these treatments may be quite variable, however, based on comorbid psychiatric or medical disorders as well as social supports. Overall, depot medications or immunotherapies are simply components of treatment for addictions. With depot medications a monthly injection might be given, though efforts to develop formulations of naltrexone, for example, that are active for up to 6 months are under way. One important issue in the use of depot medications in general is whether a test dose of the oral medication is required to ensure that the patient can tolerate the medication well. The need for a test dose is clear in the case of naltrexone for opioid dependence, because in an individual who is currently opioid dependent, naltrexone will precipitate a severe withdrawal syndrome that is irreversible until the medication is eliminated metabolically (Kleber and Kosten, 1984). In contrast, the use of this formulation in alcoholics may not require an oral test dose.
Based on the current technology, a year of treatment with a depot medication would involve monthly injections at a potential cost of $150 each. However, a depot medication is unlikely to be effective without a substantial psychosocial intervention that is delivered relatively frequently at first, with the potential for reduced frequency over time. Based on evidence of poor compliance with oral naltrexone, which has limited value in the treatment of opioid dependence, a major focus of the psychosocial intervention would be on promoting compliance with the depot injections (Kosten and Kleber, 1984). Efforts to enhance medication com-
pliance have included contingency management and other interventions with the patient as well as the involvement of family members. For opioid addicts this intervention would also include urine toxicology monitoring as well as self-reports of drug use. The psychosocial treatment to accompany a depot medication can be expected to add $5,000 to $10,000 to the cost of the medication itself. This estimate is based on twice weekly visits for up to 6 months, with gradual reduction to monthly visits over the second half of the year, or a total of approximately 75 visits, at a cost of up to $120 per visit (Rosenheck and Kosten, 2001; French et al., 1997). This cost may be mitigated in a relatively low-intervention criminal justice setting, where depot injections or even vaccinations of an antagonist could be part of monthly visits to probation or parole officers.
The use of long-acting depot formulations of antipsychotic medications, which have been well accepted, may provide a valuable model for dissemination of depot technology for the treatment of both alcohol and drug dependence. Johnson (1984) suggests several reasons why some patients who do not respond to an oral medication may respond to the depot formulation. First, the depot formulation overcomes the problems of oral drug absorption, yielding a more predictable and constant plasma level. Second, depot medications bypass hepatic metabolism, potentially resulting in a higher brain concentration of the parent compound. Third, depots help to reduce the noncompliance associated with daily drug administration. Although the use of depot antipsychotics in the treatment of schizophrenia appears to reduce patient noncompliance, evaluation of their benefits ideally requires a three-way, double-blind comparison of patients randomly assigned to a long-acting drug, the same drug given orally, or a placebo (Kane, 1984). Similar considerations apply to FDA testing of depot medications for the treatment of alcohol and drug dependence.
In addition, safety issues must be considered. Although initial study of one depot naltrexone formulation showed it to be well tolerated by alcoholics (Kranzler, Modesto-Lowe, and Nuwayser, 1998), severe local reactions to a similar formulation have subsequently been seen (Kranzler, unpublished observations). The need for careful monitoring of depot preparations was shown in a study of a long-acting formulation of somatostatin for the treatment of acromegaly (Ayuk et al., 2002). In that study, 3 of 22 patients showed impaired glucose tolerance that was attributable to the depot medication. Use of a depot formulation of a corticosteroid to treat severe seasonal allergic rhinitis, which resulted in severe bone damage to both hips of a patient (Nasser and Ewan, 2001), also underscores the risk associated with off-label use of depot formulations.
Relapse prevention using monoclonal antibodies (passive immuniza-
tion) would be very similar to a depot medication and might require injections as infrequently as every 2 months (Proksch et al., 2000; Casadevall, 1999). However, the initial administration of these monoclonal antibodies could be uncomfortable or even unsafe if given to a drug-dependent individual. Safety is a consideration because, like a depot antagonist, monoclonal antibodies would prevent the relief of withdrawal that ordinarily results from administration of a long-acting agonist. Long-acting agonists such as methadone for heroin-dependent individuals or benzodiazepines for those dependent on sedatives or alcohol are typically used for medical safety during detoxification treatment. The complications of withdrawal from sedatives or alcohol, for example, can be severe, including seizures. Helpful medications such as nicotine replacement therapy will also be neutralized by monoclonal antibodies. Therefore, before individuals are given long-acting monoclonal antibodies, they need to be adequately detoxified, which can take 3 to 14 days, depending on the abused substance and the severity of dependence (Kosten and O’Connor, 2003). As with depot medications, a relatively intensive psychosocial intervention will also be needed during at least the first few months of treatment. The cost of such an intervention is likely to be $5,000 to $10,000, not including the cost of the monoclonal antibody itself, based on twice weekly visits initially, with gradual reduction to monthly visits over time (Rosenheck and Kosten, 2001).
Relapse prevention using active immunization has several additional complications that are not present with monoclonal antibodies or depot medications. First, four or five injections administered over 8 to 12 weeks have been required in order to elicit an antibody response sufficient to antagonize the effects of the abused substance (Kosten et al., 2002). During this induction period other interventions will be needed to maintain individuals in treatment, and these may include monoclonal antibody treatment. Abusing drugs such as cocaine or taking replacement therapy such as nicotine during the induction period will not interfere with antibody production in response to the immunizations, but because of this delay in efficacy more intensive psychosocial interventions may be required with active than with passive immunization. Second, immunization will lead to a variable antibody response among individuals (Kosten and Biegel, 2002). While monoclonal antibodies can be given at a known dosage and concentration, which will not vary widely among individuals, some patients will be unresponsive to active immunization and will produce low antibody levels that will be ineffective at antagonizing the effects of the abused drug. Even in patients who respond well to initial immunization, booster immunizations about every 4 months will be required to maintain high antibody levels, and the cost per immunization might be
about $150 for the medication alone. Similar to the use of a depot medication, the inclusion of a psychosocial component will add substantially to the cost of relapse prevention via active immunization.
Protection protocols might use any of these three types of interventions, but the psychosocial issues raised earlier in this appendix as well as medical safety are important. Determining the safety of long-term exposure to these treatment agents in a relatively large number of individuals will be difficult and expensive (Sparenborg, Vocci, and Zukin, 1997; Cohen, 1997a). Thus, if a protection protocol is to be developed, it is unlikely to occur, even for those at high risk of drug abuse, until well after overdose or relapse prevention protocols are well established. Depot medications with no effects on normal functioning might be considered for this application, but even relatively inactive antagonists such as depot naltrexone for opiates have substantial risks that would likely preclude their use for such purposes. These risks include sustained elevations of various hormones such as cortisol and the sex hormones (e.g., follicle stimulating hormone) and potential liver toxicity (Kosten et al., 1986; Morgan and Kosten, 1990). Active immunization has the potential for producing a lifetime marker of immunization due to low levels of persistent antibody to the drug that could be detected in employment screenings or other nonmedical settings (Janeway et al., 1999). Passive immunotherapies, such as the monoclonals, are less likely to have safety issues than depot medications, such as naltrexone, and do not produce any lifetime markers of their use. However, compliance with 1- to 2-hour protein infusions that are administered every other month, the potential for overriding the antibody with large doses of the abused substance, and the substantial cost of the medication (about $12,000 per year, which represents six infusions at a cost of $2,000 per infusion) are limiting factors.
The potential harm of using large doses of an abused substance to overcome the blockade is well illustrated by nicotine, where a parental desire for immunotherapy of an adolescent child is a potential issue. If an immunized adolescent smoked cigarettes to obtain several times the usual dose of nicotine, he or she would also inhale several times the usual dose of various carcinogens that are in tobacco smoke without any antibody to block the adverse effects. Finally, as indicated earlier, protection against adolescent nicotine use by inhibiting pharmacological reinforcement may be ineffective because initiation of use is more closely related to peer acceptance and social factors than to pharmacological effects of the nicotine.
Several common threads run through all of these clinical protocols. First, immunotherapies and depot medications represent only a small part
of a comprehensive clinical intervention that requires substantially greater behavioral treatment than the monthly or even twice annual contact that may be needed to administer these therapies. Second, because these therapies are long lasting, they must build on a platform of treatment that is sustained for weeks and months. For example, reversal of overdose with these agents could commit the treatment provider to a substantially longer intervention than the hours ordinarily spent in an emergency department, particularly if a complete antibody rather than a Fab fragment is given. Third, the least expensive interventions to add on to existing treatment programs may best address relatively select patient populations, such as the 30 to 40 percent of methadone-maintained patients with combined heroin and cocaine dependence in whom active cocaine immunization could complement opioid agonist therapy (Brooner et al., 1997). Its advantage is that the marginal cost of such an intervention is less than for individuals not already receiving a variety of rehabilitative services. Fourth, polydrug abuse is common, and effective immunotherapies and depot medications may require that an individual receive multiple agents to treat a range of abused substances. The technology to develop such multiple target therapies is available and feasible (Kosten and Biegel, 2002). Furthermore, the medication and administration cost for such a multivalent vaccine would probably not be substantially greater than for a monovalent vaccine targeting a single abused drug. Fifth, any of these medications may be used in ways that are unlikely to be examined during clinical testing and the FDA approval process, such as for protection in nonabusing individuals (i.e., primary prevention) or among individuals identified as experimental users (i.e., secondary prevention), creating the possibility of adverse effects in an otherwise healthy population.
THE FDA CLINICAL TRIALS PROCESS
The FDA clinical trials process, which is designed to assess safety and efficacy (but not cost efficiency) through four phases of testing, may raise specific issues in a substance-dependent population (Blaine et al., 1994). The purpose of Phase I is to establish the safety of escalating doses of medications, generally in healthy subjects. Realistically, though, some active immunotherapies can only be examined in the intended substance-abusing population, perhaps during extended abstinence, because active immunization is likely to leave a low level of antibody for a lifetime (Kosten et al., 2002). Health insurers or other agencies could use the presence of such an antibody as a marker of prior treatment for drug abuse or dependence. Active immunization requires that a series of doses be given;
no efficacy is likely from a single dose because several doses are needed for antibodies to be produced at an optimal level for efficacy. The use of multiple doses increases the risk of testing active immunization in healthy nonusers. In established drug users, testing of active immunization may need to occur in an inpatient setting for up to 3 months, to prevent the use of drugs that might interact with active immunization.
Monoclonal antibodies or depot medications can be tested with single doses in healthy nonusers or drug users who are currently abstinent, perhaps in a residential setting. The safety questions related to depot medications or monoclonals require no additional monitoring requirements beyond those involved in standard evaluations for related parenteral treatments in medicine. For monoclonal testing in abstinent drug users, the drug-free period might be up to 2 months to prevent unmonitored interactions between the antibody and the abused drug. For depot medications, safety testing is probably best done in the targeted substance abuser population but could be done in healthy nonusers. Phase I testing with drug abusers will most helpfully inform later studies in Phases II and III. Limiting testing to substance abusers also addresses ethical concerns over whether the risk of the interventions can be justified in nonusers (Cohen, 1997a).
The purpose of Phase II is to establish preliminary efficacy by optimizing the dosage of the medication and may include comparison to a placebo. At this point the proposed indication for the medication is critical for determining which outcomes and populations to target. The simplest indication to study for substance dependence is probably reversal of overdose. However, drug-dependent users who have overdosed are a problematic population because they may be unable to give informed consent. Such problems can be addressed, since they were encountered with the initial evaluation of naloxone for reversing opiate overdose and flumazenil for benzodiazepine overdose. Furthermore, ongoing problems with obtaining informed consent exist when medications are tested for acute treatment of stroke patients in the emergency department while the patient is unconscious, obtunded, or in some other way unable to communicate informed consent. Hence, the conduct of such a study is possible, although it will likely require enrolling patients who meet predetermined criteria and obtaining consent for participation after the individual has regained the capacity to give informed consent.
If abstinence initiation is the outcome, interactions between the abused substance and the immunotherapy may need to be assessed in a human laboratory setting in which the abused substance is administered.
The laboratory assessment may also help establish blocking efficacy. Human laboratory assessments have been employed in testing naltrexone for opioid antagonism by using opioid agonist administration to evaluate the magnitude and duration of blocking effects (Kleber et al., 1985). This approach was recently applied to evaluate the effects of a depot naltrexone formulation (Comer et al., 2002).
During Phase II testing an additional risk that might be considered involves radiation exposure for receptor neuroimaging. Receptor neuroimaging can be very useful when used before and after immunotherapy to assess whether the antibodies effectively reduce entry of the abused drug into the brain. When the antibody is not present, a large amount of the receptor’s radioligand (e.g., radioactive C-11 cocaine) should be displaced when the nonradioactive abused drug (e.g., cocaine), which binds to the same receptor, is taken. However, when the antibody is present, substantially less of the receptor’s radioligand should be displaced following administration of the nonradioactive drug, thereby increasing the total dose of radiation detected in the brain. Imaging technology can also be applied to examine the time course of receptor occupancy by medications administered as a depot formulation. For example, mu-opioid selective receptor agents such as carfentanil could be used to examine the time course of mu-opioid receptor blockade following depot naltrexone administration (Fowler et al., 1999; Swanson and Volkow, 2002).
Another objective during Phase II testing can be to develop immunotherapies that target multiple abused substances, since these distinct antibodies should have no significant interactions. This multitarget approach should be considered as testing of a single treatment agent to facilitate examination in a polydrug-abusing population.
Depot medications in Phase II may provide a special case in which the efficacy of the oral medication is already established, so that a small Phase II study of efficacy may be sufficient. However, oral naltrexone was not compared to placebo for its initial approval as a treatment of opioid dependence and still has no evidence of superiority to placebo to support its efficacy (Kirchmayer, Davoli, and Verster, 2002). Because of the potential FDA requirement that depot naltrexone be demonstrated to have efficacy against placebo for opioid dependence, alcohol dependence has been the initial indication for developing depot naltrexone. This approach also takes into account the higher prevalence of alcohol dependence than opioid dependence (and a concomitantly greater market potential), previous success in showing that naltrexone was superior to placebo in alcohol dependence (Volpicelli et al., 1992; O’Malley et al., 1992), and the greater difficulty of conducting relapse prevention trials in opioid addicts compared with alcoholics (Srisurapanont and Jarusuraisin, 2002). However, given the modest overall effects of naltrexone for the treatment of alcohol
dependence (Kranzler and Van Kirk, 2001; Streeton and Whelan, 2001), demonstrations of the efficacy of depot naltrexone for this indication will require large patient samples to yield adequate statistical power.
The purpose of Phase III is to establish efficacy and safety in one or more large-scale, placebo-controlled studies. The design of the studies and the specific outcomes of interest may differ with the potential indication being considered. Because the reversal of overdose has become an important consideration in the management of adverse effects of a variety of new drugs, including most recently gamma-hydroxybutyrate (Miro et al., 2002), consideration should be given to the evaluation of immunotherapies for this indication. If reversal of overdose is the outcome of interest, drug-dependent users who have overdosed will be given an immunotherapy that will probably last for several weeks beyond the time needed to reverse the overdose. This could provide the opportunity to examine relapse prevention as well. However, it involves a patient population that is likely to be very difficult to follow closely using urine toxicology testing and that may be at high risk of using large doses of the abused drug to override the blockade provided by the immunotherapy. Therefore, in these drug abusers it may not be feasible to assess the potential efficacy of the antibodies for relapse prevention; instead, a follow-up of the overdose episode would focus on safety considerations.
If relapse prevention is the outcome of interest, the focus would be on drug-dependent users who are abstinent, rather than those who have just experienced an overdose. Since the natural history of drug dependence is one of a chronic relapsing course with periods of binge use, several weeks of abstinence before randomization to the medication or placebo may be required to yield a meaningful relapse prevention study (Klein, 1998). Furthermore, because of a delay in efficacy for 6 to 8 weeks as antibody levels rise after active immunization, early treatment retention is critical and may require cointerventions such as intensive contingency management (Kosten and Biegel, 2002). To add to the complexity of subject selection and the initial procedures required for an efficacy trial, extended outcome monitoring over 6 to 12 months is needed after the initial period of abstinence to follow patients until the majority relapse back to drug use. These requirements will impose a selection bias in favor of an especially stable population of abusers of drugs such as cocaine. Finally, these efficacy studies will probably be restricted to adults and nonpregnant women using suitable methods of birth control and would not include nonabusers. Thus, these studies are likely to include a very selected population of patients and will require complex psychosocial treatment plat-
forms to assess efficacy, factors that will limit the generalization of these treatments to more usual clinical applications in actively drug-using populations. This appears to be an unavoidable limitation of a Phase III clinical trial.
Phase IV and Off-Label Uses
The purpose of Phase IV is to monitor use of the medication in clinical practice (i.e., once the medication has been approved for commercial use). This includes continued evaluation of the medication in populations not originally studied and assessment of relatively rare side effects (i.e., those occurring in less than 1 percent of patients). It is during this postmarketing phase that most of the many ethical and legal issues concerning the use of immunotherapies and, to a lesser extent, depot medications are likely to arise. Among the populations not likely to be studied in the initial three phases of FDA clinical trials process are adolescents, pregnant women, medically ill people, and prisoners. These populations may be considered for later controlled trials before off-label use becomes widespread. Off-label uses include treating patient groups with different disorders or using different types of interventions than originally approved. An example of a different disorder might be using depot naltrexone for relapse prevention in opioid dependence after it is approved for alcohol relapse prevention. An example of a different type of intervention might be using a monoclonal antibody that was approved for overdose reversal in a protection protocol. Early off-label use of immunotherapies will pose special issues for immunocompromised individuals with HIV infection or AIDS. While these patients may not be suitable for active immunization, they are potential candidates for passive immunotherapies such as monoclonal antibodies to help treat their substance dependence. These groups of medically ill patients with a relatively high prevalence among substance abusers may require the conduct of early Phase IV studies focusing on safety and not necessarily requiring placebo controls, for example.
Issues of stigmatization and coercion must be considered carefully in advance of Phase IV evaluation of immunotherapies and depot medications. Potential populations for evaluation during Phase IV may include long-term abstinent substance abusers at high risk for relapse to substance use. An example of such an application would be the prophylaxis of abstinent addicts following extended stays in prison or residential treatment settings (Cohen, 1997a). Prophylaxis or protection of high-risk substance abusers who have never been substance dependent (i.e., secondary prevention) may be considered for adolescents excluded from previous trials due to low severity of the disorder. Prophylaxis in high-risk groups with no personal history of abuse (i.e,, primary prevention), such as ado-
lescents with substantial family risk factors, may also be considered. With respect to immunotherapies, these populations should be tested using monoclonal approaches before active immunization is considered because persistent low levels of antibody to the abused drug resulting from active immunization will lead to a potential lifetime marker of substance abuse treatment.
Comparisons among treatments delivered in different settings will ultimately provide information on the most effective approach to treating a range of high-risk and affected individuals. Settings may vary in the degree to which they are suited to particular types of interventions, due to factors such as the level of medical care available on-site (Fiellin and O’Connor, 2002; Sindelar and Fiellin, 2001). For example, drug-free clinics may provide an opportunity to use depot medications because these medications have minimal need for ongoing medical evaluation after their initial administration. Although severe local reactions have been observed following some depot injections, only products with demonstrated safety in this regard are likely to receive FDA approval. Opioid agonist maintenance clinics (e.g., those dispensing methadone) have greater medical resources than do drug-free clinics, making them more feasible sites for immunotherapies that require careful medical assessments over a sustained period of time. Depot medications and immunotherapies should also be evaluated in primary care medical settings and emergency departments, since these are the settings in which drug abusers are often seen in the community.
Before additional postmarketing clinical trials are completed, some physicians may decide to use these depot medications or immunotherapies for different diagnoses or in different types of intervention protocols than their approved indications. This off-label use extends beyond simply using the medication in an additional population that may not have been included in the Phase III clinical trials. Off-label use can involve a wide variety of patient groups. Generally, new medications will be used in patients with multiple illnesses, rather than in patients meeting the precise inclusion criteria of developmental trials. Although common, off-label use has raised ethical and practical issues (Cohen, 1997b; McIntyre et al., 2000; American Academy of Pediatrics, 2002).
Interestingly, one of the changes resulting from the FDA Modernization Act of 1997 was that pharmaceutical companies were allowed to disseminate articles from peer-reviewed journals about off-label use, which had previously been forbidden. The rationale for this approach was that,
since physicians had the legal right to prescribe drugs off label, information provided to them about uses not specifically approved by the FDA would make for more informed prescribing decisions (Reh, 1998). It was also hoped that this would motivate pharmaceutical companies to do the clinical studies necessary to get these indications added to drug labeling. However, such studies are unlikely in substance-dependent patients due to concerns about adverse effects related to substance abuse during the clinical trials. In general, off-label uses of medications, even in patient groups who do not abuse substances, is associated with a higher adverse drug reaction rate than that associated with labeled indications (Choonara and Conroy, 2002). Nevertheless, pediatric off-label uses of medications are quite common, and a specific study of the treatment of poisoning in children found that 60 percent of antidotes and other useful agents were not used according to the demands of licensing systems (Lifshitz, Gavrilov, and Gorodischer, 2001). This off-label use of antidotes for poisoning may be particularly relevant to the use of immunotherapies for overdose reversal in adolescents because the overdose agent may be unknown at the time that therapy is given, potentially resulting in the administration of multiple monoclonal antibodies. Furthermore, FDA clinical trials are likely to be conducted in adults, not children, so there will be limited or no prior experience with the use of these agents in adolescents.
While the development process and the postmarketing experiences differ across various therapies, when the FDA clinical trials process is completed and there is a successful new immunotherapy or depot medication, there may still be a number of unanswered questions relevant to the common practice of off-label medication use (Cohen, 1997b; McIntyre et al., 2000; American Academy of Pediatrics, 2002). For example, depot naltrexone is being developed for alcohol dependence, but the pharmacological specificity and utility for opioids are clear. Nevertheless, it is unclear whether similar efforts will be undertaken for a new drug application (NDA) to treat opioid dependence, and off-label use of depot naltrexone for opioid dependence is likely. Thus, what studies should be done with depot naltrexone in relation to the treatment of opioid dependence before the FDA grants an indication for alcohol dependence? Overall, many unanswered questions are likely to remain concerning which subgroups of substance-abusing patients will be most effectively treated pharmacologically with immunotherapies (Kosten, 1989). Nevertheless, at this juncture it would be useful to consider the indications for which an NDA or biological license might most readily be obtained for various depot medications and immunotherapies and how these indications might then be expanded by clinicians into off-label uses.
SPECIFIC ABUSED SUBSTANCES
This section addresses four issues related to generalization of the results of FDA clinical trials and the potential for off-label use of immunotherapies or depot medications that are in clinical development for treatment of a major abused substance. The first issue to be considered is the extent to which the study participants are representative of the general treatment population. Second is the range of potential therapeutic agents to be tested. Third is the need to develop multiple therapeutic targets, including integration of these agents with existing treatments. Fourth is the testing and clinical use of these treatments in settings other than specialty clinics, including primary care settings.
The FDA testing of immunotherapies or depot medications is likely to be more informative about the treatment of nicotine dependence than the treatment of any other abused substance. The participants in FDA trials for nicotine will likely be representative of the broad clinical target populations because the therapeutic goal for an NDA would be relapse prevention rather than the reversal of overdose. Second, because these therapies can have a relatively large market and the lower cost of active immunotherapies makes them attractive, both active and passive or monoclonal immunotherapies are likely to be tested. Third, because of the range of other pharmacotherapies available for nicotine dependence, combination therapies such as with antidepressants or nicotine replacement therapy (NRT; e.g., patches) are likely to be examined (Sutherland, 2002; Sullivan and Covey, 2002; Sims and Fiore, 2002). Monoclonal antibodies could not be combined with NRT because they would bind the nicotine from the patch, making NRT ineffective. However, the combination of NRT with active immunotherapies would provide an ideal and cost-effective combination that provides detoxification using the NRT, while antibody levels rise to therapeutic levels (Woolacott et al., 2002). Fourth, testing and use of these treatments would not be limited to specialty clinics; primary care settings could readily be used, since nicotine-dependent patients are commonly treated using NRT in primary care settings. Initial human testing of a nicotine vaccine has begun with no serious adverse events and the promise of rapid development (http://www.xenova.co.uk/dc_ta_nic.html).
The process of obtaining an NDA for adult smokers can address relapse prevention well, but protection protocols for prophylactic use in adolescents will require specific testing before off-label use should be allowed. Enforcement of such a prohibition on off-label use will provide
new challenges to the FDA clinical trials process. One approach that builds on the FDA requirement for testing new pharmacotherapies in children is represented by the 1998 final rule from the FDA that pediatric studies are required, the 2001 Subpart D—Final Rule: Additional Safeguards for Children in Clinical Investigations of FDA-Regulated Products, and the Best Pharmaceuticals for Children Act (2002; Subpart D, Part 46, Title 45 CFR; http://www.fda.gov/OHRMS/DOCKETS/98fr/cd0030.pdf). Although this ruling has been overturned in the courts, it is under appeal by the FDA, and a new rule is being drafted in Congress to address the issue of mandating these studies in children.
Adolescent smoking is relatively common, and treatment interventions have been studied in this population. In addition, since a “smoking career” generally begins during adolescence, the greatest impact on reducing adult smoking rates will result from the prevention of adolescent smoking. Immunization could be used to alter the trajectory of early drug use—for example, among teenagers who smoke occasionally. This group has a greater than 70 percent chance of becoming regular smokers within a few years and so might provide a “high-incidence” target. An advantage of intervening at an early stage in the development of dependence, when the dose of the drug used is low, is that it would be more amenable to blockade by immunization. A specific consideration early in the development of dependence among adolescents, however, is that the initial incentives for drug use are more social than pharmacological. Consequently, the utility of a pharmacological intervention would probably not be substantial until the pharmacological effects of the drug become important in sustaining the dependence, such as occurs among smokers who use nicotine to reverse withdrawal. Furthermore, because development of any pharmacotherapy in adolescents raises a concern over potential interference with the normal growth process, an ideal intervention would be one that specifically targets nicotine without effects on organ or hormonal systems. Thus, immunotherapies would have to be tested first in adolescent smokers with demonstrated nicotine dependence in order to assess safety.
Passive immunization with monoclonal antibodies could be examined in relatively large groups of adolescents after initial approval of an immunotherapy for adults. This type of Phase IV study should be conducted within the context of the FDA drug development process and would provide an opportunity to consider secondary prevention in this younger patient group before any primary prevention is attempted. The use of depot formulations of nicotine, naltrexone, or bupropion in adolescents would require demonstration of the safety of the oral formulations of these medications in this age group. In addition, the safety of the depot prepa-
rations would have to be established in adults during at least 2 years of Phase IV monitoring.
Many of the same issues exist in relation to the treatment of nicotine dependence in pregnant smokers. Despite clear evidence of adverse outcomes among children born to women who smoke and the widespread acceptance of nicotine replacement therapy for smoking cessation, there is a paucity of research on pharmacological treatments for pregnant smokers (Oncken and Kranzler, 2003). Although nicotine replacement therapy is not approved for use in pregnant smokers, nearly half of obstetrics practitioners surveyed indicated that they prescribed nicotine to smokers in their practices (Oncken et al., 2000), which underscores the potential for off-label use of therapies in this population.
Cocaine and Amphetamines
Because a wide range of potential populations may not be tested in clinical trials directed toward an NDA for cocaine or amphetamines, these immunotherapies or depot medications may be poorly generalized to clinical populations for off-label use. First, in terms of populations studied, the passive immunotherapies could be most efficiently examined as overdose treatments, particularly using monoclonal antibodies that are designed to have a relatively short half-life (Carrera et al., 2001; Fox et al., 1996). Using these short-duration immunotherapies, an NDA could be obtained prior to the availability of information on the utility or safety of immunotherapies as a relapse prevention tool. Therefore, testing will need to examine the repeated administration of these monoclonal agents with no more than one or two half-lives between each administration. Such repeated dosing would be a simple extension of the labeled indication of the monoclonal for overdose reversal. Second, active immunotherapies are not useful for overdose reversal, but both active and passive or monoclonal immunotherapies are likely to be useful for relapse prevention. Because the lower cost of active immunotherapies makes them attractive in settings with limited resources, it may be critical to examine both approaches. Third, because polydrug abusers in general and stimulant abusers in particular can readily switch from cocaine to amphetamine to other “designer drug” stimulants (Petry and Bickel, 1998), multitarget immunotherapies might be encouraged to cover a range of stimulants and facilitate broader abstinence from these substances. Fourth, because of the difficulty in maintaining abstinence among stimulant abusers and the need for relatively comprehensive behavioral therapies, these immunotherapies will most likely succeed in specialty clinics. Primary care settings, which would not be likely testing sites, are also unlikely sites for off-label treatment.
Use of immunotherapies as protection protocols or for primary prevention in adolescents, prisoners, or pregnant women raises all of the issues listed above under Phase IV testing. Some applications, such as for abstinent prisoners with a prior history of cocaine dependence, are likely to be within the labeled use because the time since last stimulant use is not important to the administration of these immunotherapies. However, since no pharmacotherapy exists for these stimulants, significant social pressure may be exerted to examine the potential for immunotherapy in other off-label uses with these special groups (Kosten, 2002). Because of the potential for lifetime markers after active vaccination, this would not be a viable option for protection protocols. But even with monoclonal antibodies, the scientific information that might be obtained during the FDA clinical trials process will be inadequate to formulate guidelines for off-label uses in adolescents or pregnant women. Since animal models are the best available approximation of use in pregnant women, safety studies of immunotherapies and depot medications in pregnant animals should be a required part of any successful NDA.
No obvious candidates for depot medications to treat stimulant dependence exist. However, since disulfiram has shown some efficacy for cocaine dependence (Carroll et al., 1998), the active sulfoxide metabolite of disulfiram, which is highly potent (Hart and Faiman, 1992), is a potential agent for development as a depot formulation. Because of the potential for this medication to interact with alcohol, resulting in rare but potentially serious adverse effects, off-label uses would probably be discouraged by the liability concerns of practitioners (Wright and Moore, 1990).
Although immunotherapies for opioids were developed in primate models over 30 years ago (Bonese et al., 1974), development did not continue due to the utility of methadone and naltrexone as pharmacological treatments for opioid dependence. Nevertheless, immunotherapies might be developed in trials that inform the four issues being considered here. It is likely that the first population to be studied with immunotherapies and depot medications would be representative of the general treatment population seeking relapse prevention. Because naloxone is so cost effective for the reversal of opioid overdose, overdose reversal may not seem a profitable target for development of this type of immunotherapy (Clarke and Dargan, 2002).
The range of potential therapeutic agents for treatment of opioid dependence is likely to be wide in order to compete with the cost-effective treatments of methadone maintenance or even naltrexone maintenance.
An inexpensive active immunotherapy or a depot form of naltrexone would be more likely to succeed than a relatively expensive monoclonal antibody.
The third issue of developing multiple therapeutic targets and integrating them with existing treatments will be of particular importance in the treatment of opioid dependence. While heroin is a predominant opioid of abuse, many other synthetic opioids are abused, including the treatment agents methadone and buprenorphine (National Institute on Drug Abuse, 1998; Kintz, 2001). Use of multiple therapeutic targets would be a reasonable approach, although the advantage of having high specificity for heroin while allowing the use of other opioids for pain relief might have some value in special populations. The integration of immunotherapies with methadone treatment provides interesting possibilities, including a potential slow detoxification starting with modest doses of methadone (e.g., 30 to 40 mg daily), while the antibody levels to heroin rise over 6 to 8 weeks. Depot naltrexone, like immunotherapy, addresses compliance issues, but naltrexone appears to offer advantages of very high levels of blockade compared to the competitive antagonism of active or passive immunotherapy. Obviously, however, the use of long-acting naltrexone, as with oral naltrexone, requires that detoxification be completed prior to initiation of therapy, to avoid a severe withdrawal reaction.
Although the testing and use of these treatments in settings other than specialty clinics pose the same challenges as with treatments for stimulant dependence, the structure and medical resources of methadone maintenance clinics make them excellent sites for the transfer of this technology into the community. Due to the difficulties inherent in blinding trials involving naltrexone for opioid dependence, limited placebo-controlled data are available on the oral formulation of this medication for the treatment of opioid dependence. Consequently, the design of placebo-controlled clinical trials of depot formulations of naltrexone for opioid dependence is likely to break new ground, because unmasking the blind will be relatively easy and is likely to occur.
Other Drugs of Abuse—Phencyclidine
While immunotherapies are theoretically possible for hallucinogens, cannabis, and “club drugs,” such as MDMA (methylenedioxy-n-methylamphetamine) or ecstacy, the only monoclonal developed to date is PCP. This immunotherapy is specifically designed for the reversal of PCP overdose, but its long duration of action suggests that it also has potential for relapse prevention among PCP abusers (Owens and Mayershohn, 1986; Valentine et al., 1996; Proksch et al., 2000). Clinical trials to support an NDA might logically focus on the potential of this treatment to reverse
overdose, but its long duration of action dictates testing of its longer-term effects and its safety in outpatient substance abusers, particularly since once approved for overdose, it would likely be used off label.
The capacity to generalize the findings among study participants to general clinical use is a complex issue in relation to immunotherapies for PCP. Although as an overdose treatment the PCP monoclonal will be tested in precisely the population where it is intended for clinical use, follow-up of these patients after the overdose treatment may be very difficult. There is likely to be a low rate of contact for the weeks following overdose, since the patients are unlikely to be motivated to seek treatment or continued contact with the providers. Nevertheless, the weeks of follow-up will be most critical for assessing both the safety and potential efficacy for relapse prevention. This difficulty in follow-up suggests that a separate clinical trial may be needed to assess safety in active PCP abusers who have not had an overdose as the basis for monoclonal treatment. The availability of active vaccination as well as a passive monoclonal is important if this approach is considered for relapse prevention and not just overdose reversal. The need for multiple therapeutic targets is considerable, since the specific agent in overdoses with club drugs such as PCP, ketamine, or even MDMA is difficult to identify based on clinical presentation, and a broad-spectrum antidote would be most useful (Baskin and Morgan, 1997; Owens, 1997). The issue of using these treatments beyond the emergency department is a critical question because of the duration of their blocking effects. Unlike naloxone for opioids or flumazenil for benzodiazepines, both of which have brief durations of action, monoclonals are a sustained intervention that can be most important as an entry into treatment for substance abuse or dependence (Clarke and Dargan, 2002; Singh and Richell-Herren, 2000). This opportunity should be exploited for maximum clinical benefit and examined as part of the NDA process.
Immunotherapies are not possible for alcohol, but Phase III clinical trials of depot naltrexone are under way, and trials of depot formulations of other opioid antagonists such as nalmefene are being planned. Depot medications are not likely to have application in the treatment of either alcohol overdose (for which supportive measures and, at the extreme, hemodialysis, are the treatments of choice) or alcohol withdrawal (for which brain depressants or anticonvulsants are efficacious when administered for a relatively short period of time). Consequently, clinical trials are likely to be most informative to the degree that they extend findings of placebo-controlled trials of oral formulations of the candidate medications, which have focused on relapse prevention.
In addition to considering the development of these formulations in relation to the four issues of interest in regard to generalization of findings from the FDA clinical trials process, it is important to consider their potential application to the treatment of drug dependence. The large samples recruited for Phase III studies of these formulations should provide results that can be generalized to the population of treatment-seeking alcoholics, though as with oral naltrexone, it must be recognized that there is some selection of more motivated and compliant patients to participate in the trials. Given the difficulty of retaining opioid addicts in treatment with an opioid antagonist (Kleber and Kosten, 1984), it is likely that findings from Phase III studies of these formulations in opioid addicts will not be as readily generalized to the treatment-seeking population of opioid addicts. Consequently, once approved for the treatment of alcohol dependence, these medications are likely to be used off label for the treatment of opioid dependence. Consequently, FDA approval for alcohol dependence may necessitate evidence of the safety of these formulations for use in opioid addicts.
Since alcohol affects a variety of neurotransmitter systems, many of which have been implicated in the pathophysiology of alcohol dependence (Kranzler, 1995), the range of potential therapeutic agents to be tested in conjunction with a depot medication is great. Some of these systems (e.g., opioidergic or dopaminergic) are of obvious relevance to drug dependence, so that transfer of the technology to treat alcohol dependence will be relatively straightforward. However, depot formulations of drugs affecting neurotransmitter systems for which therapeutic effects in drug dependence are not as promising (e.g., the serotonergic system; Pettinati et al., 2000; Johnson et al., 2000) will be less readily applied across substances.
As with drug addicts, alcoholics often abuse a variety of substances (Gossop, Marsden, and Stewart, 2002), so a depot formulation that hits multiple targets could be very useful. Although the evidence supporting naltrexone treatment of nicotine dependence is not yet adequate to draw conclusions (David, Lancaster, and Stead, 2001), a depot medication that is efficacious for treatment of both alcohol and nicotine dependence would have considerable utility, given the high rate at which these disorders co-occur (Hughes, 1995). Combination therapy has not been widely used in alcoholism treatment. However, the diversity of neurotransmitter systems implicated in the disorder argues in favor of greater research attention being paid to this approach (Kranzler, 2000). The use of a targeted approach to oral therapies (Kranzler et al., 2003) raises the possibility of augmenting a depot treatment with intermittent use of an oral medication. This approach would facilitate a combination of medications to target different neurotransmitter systems—for example, depot naltrexone combined with targeted use of an alcohol-sensitizing drug to cope with high-
risk situations (Duckert and Johnsen, 1987; Annis and Peachey, 1992). As an example of an application of this strategy to the treatment of drug dependence, depot naltrexone could be combined with daily disulfiram to treat comorbid opioid and cocaine dependence (Petrakis et al., 2000).
There appears to be considerable potential for the use of depot formulations in settings other than specialty clinics, including primary care settings. In contrast to most specialty substance abuse treatment settings, the use of injectable medications is common in primary care practices. Consequently, the feasibility of their use in these settings will likely depend on demonstration in Phase III trials that low-intensity psychosocial interventions (i.e., those that can be readily applied to primary care settings) are adequate to support the efficacy of the medications.
Three broad types of intervention protocols might be tested in support of an NDA or expanded during the off-label use after approval of immunotherapies or depot medications: overdose, relapse prevention, and protection from abuse. The four-phase FDA clinical trials process to assess safety and efficacy will inform many of the important questions that must be answered prior to widespread use of these treatments. During Phases I and II, the safety of escalating doses of immunotherapies can be established in relevant patient groups and because polydrug abuse is common, multiple target approaches should be developed. The issue of multiple targets may be critical for overdose reversal with users of club drugs such as PCP because these abusers frequently do not know whether they have taken ketamine, MDMA, or other related substances and may be unaware that they are taking combinations. During Phase III, relapse prevention clinical trials would be most useful for both immunotherapies and depot medications and, where feasible, might include assessment of these treatments for abstinence initiation in active abusers. Relapse prevention studies are likely to enroll a very select population of patients, and complex (and costly) psychosocial treatment platforms may be used when assessing efficacy; both factors will limit their generalization to more usual clinical applications in actively drug-using populations. Nevertheless, these relapse prevention trials will be more relevant to the widespread application of these treatments than trials focusing on the reversal of overdose.
At best only some of the important questions will be answered during the FDA clinical trials process required for the approval of an NDA for immunotherapies and depot medications. The unanswered questions are likely to include how much behavioral treatment is needed to deliver these therapies effectively. This question includes both the frequency of treat-
ment contact (e.g., ranging from daily to monthly) and the duration over which behavioral treatment is required after the medication is administered. The setting for delivery of this care also needs to be considered to make the marginal cost of such an intervention affordable.
None of the Phase III studies are likely to address issues relevant to the prophylaxis of addiction in nonabusers (primary prevention) because of the substantial cost and long duration of this type of clinical trial to establish efficacy. Nevertheless, subjects with sustained abstinence, who are at high risk for relapse, might be approached for secondary prevention studies during Phase IV monitoring. Phase IV is where many ethical issues will arise as off-label uses of these immunotherapies or depot medications proliferate. New populations may be studied, including adolescents, prisoners, and pregnant women, and new treatment settings, such as primary care medical clinics, may be examined. The FDA testing process will provide only limited help in generalization to off-label uses, and the extent to which the process will help varies across the specific abused substances. Four issues will be important for this generalization: the nature of the NDA study population, the range of agents tested, the targeting of multiple therapeutic targets or integration with existing treatments, and the variety of settings where testing was done and treatment is provided. The issue of where treatment is provided may be a particular challenge, since many substance abuse treatment programs lack the infrastructure to deliver pharmacotherapies, particularly those that require greater medical support. Coordination between these programs and a medical setting where these immunotherapies or depot medications might be provided has typically not been successful and may lead to discontinuities in care for the patient. Off-label uses in these medical settings are likely to be provided most effectively for nicotine products. Off-label uses will probably be provided much more poorly for cocaine, amphetamines, and PCP. The difficulties with services for these drugs include limited information on their use from the FDA trials (e.g., reversal of overdose using monoclonal antibodies for PCP), the need for close coordination with substance abuse treatment settings that have limited medical backgrounds, and social pressure to make any effective treatment available. Use of depot medications such as naltrexone for alcoholism are likely to be well informed by the FDA process, but other uses of depot naltrexone such as for heroin dependence may be less thoroughly studied before the formulation is available for commercial use.
In summary, the FDA clinical trials process will provide a wealth of information about the safety and efficacy of these new medications. However, the wide range of unanswered questions posed by off-label use after approval needs ethical consideration to protect the many groups of indi-
viduals who may be offered or perhaps coerced into receiving these medical interventions.
American Academy of Pediatrics. (2002). Uses of drugs not described in the package insert (off-label uses). Pediatrics, 110(1 Pt 1), 181-183.
Annis, H.M., and Peachey, J.E. (1992). The use of calcium carbamide in relapse prevention counseling: Results of a randomized controlled trial. British Journal of Addiction, 87(1), 63-72.
Ayuk, J., Stewart, S.E., Stewart, P.M., and Sheppard, M.C. (2002). Long-term safety and efficacy of depot long-acting somatostatin analogs for the treatment of acromegaly. Journal of Clinical Endocrinology and Metabolism, 87(9), 4142-4146.
Baskin, L.B., and Morgan, D.L. (1997). Drugs detected in patients suspected of acute intoxication. Texas Medicine, 93(9), 50-58.
Blaine, J.D., Ling, W., Kosten, T.R., O’Brien, C.P., and Chiarello, R.J. (1994). Establishing the efficacy and safety of medications for the treatment of drug dependence and abuse: Methodological issues. In R. Prien and E. Robinson (Eds.), Clinical evaluation of psychotropic drugs: Principles and guidelines. New York: Raven Press.
Bonese, K.F., Wainer, B.H., Fitch, F.W., Rothberg, R.M., and Schuster, C.R. (1974). Changes in heroin self-administration by a rhesus monkey after morphine immunization. Nature, 252(5485), 708-710.
Brooner, R.K., King, V.L., Kidorf, M., Schmidt, C.W., Jr., and Bigelow, G.E. (1997). Psychiatric and substance use comorbidity among treatment-seeking opioid abusers. Archives of General Psychiatry, 54(1), 71-80.
Carrera, M.R., Ashley, J.A., Wirsching, P., Koob, G.F., and Janda, K.D. (2001). A second-generation vaccine protects against the psychoactive effects of cocaine. Proceedings of the National Academy of Sciences, USA, 98(4), 1988-1992.
Carroll, K.M., Nich, C., Ball, S.A., McCance, E., and Rounsaville, B.J. (1998). Treatment of cocaine and alcohol dependence with psychotherapy and disulfiram. Addiction, 93(5), 713-727.
Casadevall, A. (1999). Passive antibody therapies: Progress and continuing challenges. Clinical Immunology, 93(1), 5-15.
Choonara, I., and Conroy, S. (2002). Unlicensed and off-label drug use in children: Implications for safety. Drug Safety, 25(1), 1-5.
Clarke, S., and Dargan, P. (2002). Towards evidence based emergency medicine: Best BETs from the Manchester Royal Infirmary. Intravenous bolus or infusion of naloxone in opioid overdose. Emergency Medicine Journal, 19(3), 249-250.
Cohen, P.J. (1997a). Immunization for prevention and treatment of cocaine abuse: Legal and ethical implications. Drug and Alcohol Dependence, 48(3), 167-174.
Cohen, P.J. (1997b). “Off-label” use of prescription drugs: Legal, clinical and policy considerations. European Journal of Anaesthesiology, 14(3), 231-235.
Comer, S.D., Collins, E.D., Kleber, H.D., Nuwayser, E.S., Kerrigan, J.H., and Fischman, M.W. (2002). Depot naltrexone: Long-lasting antagonism of the effects of heroin in humans. Psychopharmacology, 159(4), 351-360.
David, S., Lancaster, T., and Stead, L.F. (2004). Opioid antagonists for smoking cessation. In The Cochrane Library (Issue 1). Chichester, England: John Wiley & Sons.
Duckert, F., and Johnsen, J. (1987). Behavioral use of disulfiram in the treatment of problem drinking. International Journal of the Addictions, 22(5), 445-454.
Fiellin, D.A., and O’Connor, P.G. (2002). Clinical practice: Office-based treatment of opioid-dependent patients. New England Journal of Medicine, 347(11), 817-823.
Fowler, J.S., Volkow, N.D., Wang, G.J., Ding, Y.S., and Dewey, S.L. (1999). PET and drug research and development. Journal of Nuclear Medicine, 40(7), 1154-1163.
Fox, B.S., Kantak, K.M., Edwards, M.A., Black, K.M., Bollinger, B.K., Botka, A.J., French, T.L., Thompson, T.L., Schad, V.C., Greenstein, J.L., Gefter, M.L., Exley, M.A., Swain, P.A., and Briner, T.J. (1996). Efficacy of a therapeutic cocaine vaccine in rodent models. Nature Medicine, 2(10), 1129-1132.
French, M.T., Dunlap, L.J., Zarkin, G.A., McGeary, K.A., and McLellan, A.T. (1997). A structured instrument for estimating the economic cost of drug abuse treatment. The Drug Abuse Treatment Cost Analysis Program (DATCAP). Journal of Substance Abuse Treatment, 14(5), 445-455.
Gossop, M., Marsden, J., and Stewart, D. (2002). Dual dependence: Assessment of dependence upon alcohol and illicit drugs, and the relationship of alcohol dependence among drug misusers to patterns of drinking, illicit drug use and health problems. Addiction, 97(2), 169-178.
Hart, B.W., and Faiman, M.D. (1992). In vitro and in vivo inhibition of rat liver aldehyde dehydrogenase by S-methyl N, N-diethylthiolcarbamate sulfoxide: A new metabolite of disulfiram. Biochemical Pharmacology, 43(3), 403-406.
Hughes, J.R. (1995). Clinical implications of the association between smoking and alcoholism. In J.B. Fertig and J.P. Allen (Eds.), Alcohol and tobacco: From basic science to clinical practice (National Institute on Alcohol Abuse and Alcoholism Research Monograph No 30. NIH Publ. No. 95-3931, pp. 171-186). Washington, DC: U.S. Government Printing Office.
Janeway, C.A., Travers, P., Capra, J.D., and Walport, M.J. (1999). Immunobiology. The immune system in health and disease, 4th ed. New York: Garland.
Johnson, B.A., Roache, J.D., Javors, M.A., DiClemente, C.C., Cloninger, C.R., Prihoda, T.J., Bordnick, P.S., Ait-Daoud, N., and Hensler, J. (2000). Ondansetron for reduction of drinking among biologically predisposed alcoholic patients: A randomized controlled trial. Journal of the American Medical Association, 284(8), 963-971.
Johnson, D.A.W. (1984). Observations on the use of long-acting depot neuroleptic injections in the maintenance therapy of schizophrenia. Journal of Clinical Psychiatry, 45(5 Pt 2), 13-21.
Kane, J.M. (1984). The use of depot neuroleptics: Clinical experience in the United States. Journal of Clinical Psychiatry, 45(5 Pt 2), 5-12.
Kintz, P. (2001). Deaths involving buprenorphine: A compendium of French cases. Forensic Science International , 121(1-2), 65-69.
Kirchmayer, U., Davoli, M., and Verster, A. (2004). Naltrexone maintenance treatment for opioid dependence. In The Cochrane Library (Issue 1). Chichester, England: John Wiley & Sons.
Kleber, H.D., and Kosten, T.R. (1984). Naltrexone induction: Psychological and pharmacological strategies. Journal of Clinical Psychiatry, 45(9 Pt 2), 29-38.
Kleber, H.D., Kosten, T.R., Gaspari, J., and Topazian, M. (1985). Non-tolerance to the opioid antagonism of naltrexone. Biological Psychiatry, 20(1), 66-72.
Klein, M. (1998). Research issues related to development of medications for treatment of cocaine addiction. Annals of the New York Academy of Sciences, 844, 75-91.
Kosten, T.R. (1989). Pharmacotherapeutic interventions for cocaine abuse: Matching patients to treatments. Journal of Nervous and Mental Disease, 177(7), 379-389.
Kosten, T.R. (2002). Pathophysiology and treatment of cocaine dependence. In K.L. Davis, D. Charney, J.T. Coyle, and C. Nemeroff (Eds.), Neuropsychopharmacology: The fifth generation of progress (pp. 1461-1473). Baltimore, MD: Williams and Wilkins.
Kosten, T.R., and Biegel, D. (2002). Therapeutic vaccines for substance dependence. Expert Review of Vaccines, 1(3), 363-371.
Kosten, T.R., and Kleber, H.D. (1984). Strategies to improve compliance with narcotic antagonists. America Journal of Drug and Alcohol Abuse, 10(2), 249-266.
Kosten, T.R., and O’Connor, P.G. (2003). Current concepts—management of drug withdrawal. New England Journal of Medicine, 348(18), 1786-1795.
Kosten, T.R., Kreek, M.J., Ragunath, J., and Kleber, H.D. (1986). Cortisol levels during chronic naltrexone maintenance treatment in ex-opiate addicts. Biological Psychiatry, 21(2), 217-220.
Kosten, T.R., Rosen, M., Bond, J., Settles, M., Roberts, J.S., Shields, J., Jack, L., and Fox, B. (2002). Human therapeutic cocaine vaccine: Safety and immunogenicity. Vaccine, 20(7-8), 1196-1204.
Kranzler, H.R. (1995). The pharmacology of alcohol abuse. New York: Springer-Verlag.
Kranzler, H.R. (2000). Pharmacotherapy of alcoholism: Gaps in knowledge and opportunities for research. Alcohol and Alcoholism, 35(6), 537-547.
Kranzler, H.R., and Van Kirk, J. (2001). Efficacy of naltrexone and acamprosate for alcoholism treatment: A meta-analysis. Alcoholism, Clinical and Experimental Research, 25(9), 1335-1341.
Kranzler, H.R., Modesto-Lowe, V., and Nuwayser, E.S. (1998). Sustained-release naltrexone for alcoholism treatment: A preliminary study. Alcoholism, Clinical and Experimental Research, 22(5), 1074-1079.
Kranzler, H.R., Armeli, S., Tennen, H., Blomqvist, O., Oncken, C., Petry, N., and Feinn, R. (2003). Targeted naltrexone for early problem drinkers. Journal of Clinical Psychopharmacology, 23(3), 294-304.
Lifshitz, M., Gavrilov, V., and Gorodischer, R. (2001). Off-label and unlicensed use of antidotes in paediatric patients. European Journal of Clinical Pharmacology, 56(11), 839-841.
McIntyre, J., Conroy, S., Avery, A., Corns, H., and Choonara, I. (2000). Unlicensed and off label prescribing of drugs in general practice. Archives of Disease in Childhood, 83(6), 498-501.
Miro, O., Nogue, S., Espinosa, G., To-Figueras, J., and Sanchez, M. (2002). Trends in illicit drug emergencies: The emerging role of gamma-hydroxybutyrate. Journal of Toxicology—Clinical Toxicology, 40(2), 129-135.
Morgan, C., and Kosten, T.R. (1990). Potential toxicity of high dose naltrexone in patients with appetitive disorders. In L. Ried (Ed.), Opioids, bulimia, and alcohol abuse and alcoholism (pp. 261-274). New York: Springer-Verlag.
Nasser, S.M., and Ewan, P.W. (2001). Lesson of the week: Depot corticosteroid treatment for hay fever causing avascular necrosis of both hips. British Medical Journal (Clinical Research Edition), 322(7302), 1589-1591.
National Institute on Drug Abuse (1998). National survey results on drug use from Monitoring the Future survey. Rockville, MD: National Institute on Drug Abuse.
O’Malley, S.S., Jaffe, A.J., Chang, G., Schottenfeld, R.S., Meyer, R.E., and Rounsaville, B. (1992). Naltrexone and coping skills therapy for alcohol dependence: A controlled study. Archives of General Psychiatry, 49(11), 894-898.
Oncken, C.A., and Kranzler, H.R. (2003). Pharmacotherapies to enhance smoking cessation during pregnancy. Drug and Alcohol Review, 22(2), 191-202.
Oncken, C.A., Pbert, L., Ockene, J.K., Zapka, J., and Stoddard, A. (2000). Nicotine replacement prescription practices of obstetric and pediatric clinicians. Obstetetrics and Gynecology, 96(2), 261-265.
Owens, S.M. (1997). Antibodies as pharmacokinetic and metabolic modifiers of neurotoxicity. NIDA Research Monograph, 173, 259-272.
Owens, S.M., and Mayershohn, M. (1986). Phencyclidine-specific Fab fragments alter phencyclidine disposition in dogs. Drug Metabolism and Disposition, 14(1), 52-58.
Petrakis, I.L., Carroll, K.M., Nich, C., Gordon, L.T., McCance-Katz, E.F., Frankforter, T., and Rounsaville, B.J. (2000). Disulfiram treatment for cocaine dependence in methadone-maintained opioid addicts. Addiction, 95(2), 219-228.
Petry, N.M., and Bickel, W.K. (1998). Polydrug abuse in heroin addicts: A behavioral economic analysis. Addiction, 93(3), 321-335.
Pettinati, H.M., Volpicelli, J.R., Kranzler, H.R., Luck, G., Rukstalis, M.R., and Cnaan, A. (2000). Sertraline treatment for alcohol dependence: Interactive effects of medication and alcoholic subtype. Alcoholism, Clinical and Experimental Research, 24(7), 1041-1049.
Proksch, J.W., Gentry, W.B., and Owens, S.M. (2000). Anti-phencyclidine monoclonal antibodies provide long-term reductions in brain phencyclidine concentrations during chronic phencyclidine administration in rats. Journal of Pharmacology and Experimental Therapeutics, 292(3), 831-837.
Reh, M. (1998). Changes at FDA may speed drug approval process and increase off-label use. Journal of the National Cancer Institute, 90(11), 805-807.
Rosenheck, R., and Kosten, T. (2001). Buprenorphine for opiate addiction: Potential economic impact. Drug and Alcohol Dependence, 63(3), 253-262.
Sims, T.H., and Fiore, M.C. (2002). Pharmacotherapy for treating tobacco dependence: What is the ideal duration of therapy? CNS Drugs, 16(10), 653-662.
Sindelar, J.L., and Fiellin, D.A. (2001). Innovations in treatment for drug abuse: Solutions to a public health problem. Annual Review of Public Health, 22, 249-272.
Singh, P.K., and Richell-Herren, K. (2000). Towards evidence based emergency medicine: Best BETs from the Manchester Royal Infirmary. Flumazenil and suspected benzodiazepine overdose. Journal of Accident and Emergency Medicine, 17(3), 214.
Sparenborg, S., Vocci, F., and Zukin, S. (1997). Peripherial cocaine-blocking agents: New medications for cocaine dependence. Drug and Alcohol Dependence, 48(3), 149-151.
Srisurapanont, M., and Jarusuraisin, N. (2004). Opioid antagonists for alcohol dependence. In The Cochrane Library (Issue 1). Chichester, England: John Wiley & Sons.
Streeton, C., and Whelan, G. (2001). Naltrexone, a relapse prevention maintenance treatment of alcohol dependence: A meta-analysis of randomized controlled trials. Alcohol and Alcoholism, 36(6), 544-552.
Sullivan, M.A., and Covey, L.S. (2002). Nicotine dependence: The role for antidepressants and anxiolytics. Current Opinion in Investigational Drugs, 3(2), 262-271.
Sutherland, G. (2002). Current approaches to the management of smoking cessation. Drugs, 62(Suppl 2), 53-61.
Swanson, J.M., and Volkow, N.D. (2002). Pharmacokinetic and pharmacodynamic properties of stimulants: Implications for the design of new treatments for ADHD. Behavioural Brain Research, 130(1-2), 73-78.
Valentine, J.L., Mayersohn, M., Wessinger, W.D., Arnold, L.W., and Owens, S.M. (1996). Antiphencyclidine monoclonal Fab fragments reverse phencyclidine-induced behavioral effects and ataxia in rats. Journal of Pharmacology and Experimental Therapeutics, 278(2), 709-716.
Volpicelli, J., O’Brien, C., Alterman, A., and Hayashida, M. (1992). Naltrexone in the treatment of alcohol dependence. Archives of General Psychiatry, 49(11), 867-880.
Woolacott, N.F., Jones, L., Forbes, C.A., Mather, L.C., Sowden, A.J., Song, F.J., Raftery, J.P., Aveyard, P.N., Hyde, C.J., and Barton, P.M. (2002). The clinical effectiveness and cost-effectiveness of bupropion and nicotine replacement therapy for smoking cessation: A systematic review and economic evaluation. Health Technology Assessment, 6(16), 1-245.
Wright, C., and Moore, R.D. (1990). Disulfiram treatment of alcoholism. American Journal of Medicine, 88(6), 647-655.