THE FUTURE OF PEDIATRIC PSYCHOPHARMACOLOGY

CHILD AND ADOLESCENT PSYCHOPHARMACOLOGY 0031-3955/98 $8.00 + .OO

THE FUTURE OF PEDIATRIC PSYCHOPHARMACOLOGY John T. Walkup, MD, Karen Cruz, MD, Scott Kane,

and Barbara Geller, MD

The future of pediatric psychopharmacology can be considered in many ways. Perhaps the simplest approach is to consider what new medications will become available for the treatment of childhood psychiatric disorders. Although the development of new medications is important, it is only one aspect of the future of pediatric psychopharmacology. However, before pediatric psychopharmacology can even have ,a future there is a great need to catch up on work that is long overdue. Specifically, the action, safety, and efficacy of available psychotropic medications should be studied in children and adolescents. The next step is to keep up with the rapid advances in our understanding of childhood psychiatric disorders. Without the ability to rapidly incorporate new knowledge, our field will forever be behind in providing safe and effective treatments. Child psychiatry currently extrapolates methods and strategies from adult psychiatry. In the future, for true innovation to occur, methods and approaches to childhood psychiatric disorders based on the study of children is the true future of our discipline.

Pharmacologic interventions will be an important component of the treatment of childhood psychiatric disorders in the future. This article reviews new medications that will soon be available in the United States, but it also reviews the changes that have occurred and are needed in pediatric psychiatry, including changes in the federal regulatory and funding agencies, private industry, and patient and professional groups.

THE FUTURE OF CHILD AND ADOLESCENT PSYCHIATRY

The use of medications to treat behavioral conditions in children and adolescents began in 1937, when Bradley identified the calming effect of benzedrine

From the Division of Child and Adolescent Psychiatry, Johns Hopkins Medical Institutions, Baltimore, Maryland (JTW, KC); Duke University, Durham, North Carolina (SK); and the Department of Psychiatry, Washington University, St. Louis, Missouri (BG)

PEDIATRIC CLINICS OF NORTH AMERICA

VOLUME 45 - NUMBER 5 * OCTOBER 1998 1265

1266 WALKUP et a1

in behaviorally disturbed children.6 Thirty years later, Conners and Eisenberg conducted the first controlled clinical trial of dextroamphetamine in minimal brain dysfunction.8 Much has changed in child psychiatry since that time, but despite early successes, a paucity of well-designed, controlled pharmacologic treatment trials in children and adolescents has been published. There are several reasons for the lack of controlled trials. Research in child psychiatry, in general, has lagged behind that of other medical specialties, in part because of concern about involving children in research studies but also because prevailing concep- tions about the causes of childhood psychopathology do not readily lend them- selves to scientific study. Although it is an oversimplification, the prevailing understanding of childhood psychopathology is still loosely based on psychodynamic theory. In this context, research and especially psychopharmacologic research is not yet broadly considered an essential field of study, or, worse, it is considered misguided and even unethical.

Only recently, with the advent of clear diagnostic criteria, standardized methods of assessment, and more medical conceptual models of childhood psychiatric disorders, has there been a utilization of the scientific method to understand the psychiatric problems of children and adolescents. Children with psychiatric disorders are increasingly being recruited into clinical research, but based on epidemiologic studies, the need for knowledge about childhood psychiatric disorders far outpaces the advances gained from research.

Unfortunately, there is not universal support for the study of psychiatric problems in children. Even though increasing numbers of children are treated with medication each year without the benefit of safety and efficacy data, pediatric pharmacologic research is still considered by many to be controversial. The opponents of child psychiatric research describe a number of concerns. There is fear about what might happen to children involved in research and the perception that the risk, regardless of the safety precautions, is too high. Some naively perceive the problems of children and consider research as misguided. There are also competitive conceptual models of the cause and treatment of psychiatric disorders in children, such as the often cited false dichotomy between psychodynamic and brain-based theories of origin. Some religious and political groups oppose child psychiatric research and cite the problems of children as evidence of family and social problems, which could be better solved by either religious or policy changes. Although it is clear that there is increasing momen- tum supporting the scientific study of childhood psychopathology (and that these forces will win out), there will be times in the future when such research will not be supported and progress in our understanding of childhood psychopathology will be impeded.

GOVERNMENTAL EFFORTS TO INCREASE THE AMOUNT OF INFORMATION REGARDING THE USE OF MEDICATIONS IN THE PEDIATRIC AGE GROUP

Regulatory Changes in the US Food and Drug Administration

Traditionally, healthy adults have been considered the best subjects for research, including pharmacologic research. The rationale for using only adults, especially men, may seem biased in our current view. At the time this "tradition" was established, however, preferentially enrolling men in clinical trials was a way of protecting women of childbearing age, children, and the elderly from the risks of exposure to unknown and potentially harmful compounds. As a result, a small percentage of all medications marketed in the United States,

THE FUTURE OF PEDIATRIC PSYCHOPHARMACOLOGY 1267

including psychotropic medication, have had clinical trials conducted in the pediatric age group and have adequate product labeling.

Historically, the lack of support for studies of psychotropic medications in children has come from the pharmaceutical industry, the federal government, and the public. The logic of protecting vulnerable populations from participating in clinical trials makes sense. The logic, however, seems to be changing. It is often now considered more of a health risk for medications to be used in children and other vulnerable populations off-label than it is to conduct carefully considered and well-designed clinical trials to establish efficacy, safety, and tolerability. This is especially true given the fact that once a medication receives an indication and is marketed for adults, there is much less motivation to do the research necessary to establish safety and efficacy in pediatric patients.

Despite the lack of controlled safety and efficacy studies in children, marketed pharmaceuticals are increasingly prescribed to children and other vulnerable populations without the specific knowledge of dosing, efficacy, and tolerability. Listed in Table 1 are the 10 most common medications used off-label in children, three of which are psychotropic agents. The increased use of medications off-label, especially psychotropic agents, is likely related to: (1) the rapidly increasing pharmacopoeia; (2) an impression that medications available for use in the United States are safe; (3) the fact that available medications, with a few exceptions, have a similar therapeutic action in adults and children, even if there are differences in pharmacokinetics (e.g., antibiotics or asthma medications); and (4) the fact that new m6dels for conceptualizing childhood psychiatric disorders as “illnesses” clearly facilitates the use of psychotropic medications in the pediatric population.

The lack of informative product labeling for pediatric patients poses a serious clinical dilemma to physicians. Physicians may often have to choose between using an agent with known safety data but limited usefulness (eg., imipramine for depression in children) or use a potentially more effective and safe medication that does not have adequate safety and efficacy data in children (e.g., fluoxetine for depression). In addition, even though there is an increasing awareness of similarities between adults and children, dosing and tolerability cannot be readily extrapolated from adult studies for all medication used in children. Changes in body weight, height, and surface area over the course of development and the development of metabolic and immunologic functioning

Table 1. DRUGS MOST PRESCRIBED TO CHILDREN “OFF LABEL” IN 1994

Medication Prescriptions Age Indication

Albuterol 1,626,000 <12 Asthma Promethazine (fhenergan) 663,000 < 2 Allergic reactions Ampicillin 639,000 <12 Infections Antipyrine (Auralgan otic sol.) 600,000 <16 Ear pain Cromolyn (Intal Aerosol) 399,000 < 5 Asthma Fluoxetine (Prozac) 349,000 4 6 Depression and OCD Clotrimazole (Lotrisone cream) 325,000 <12 Topical infections Sertraline (Zoloft) 248,000 4 6 Depression Methylphenidate (Ritalin) 226,000 < 6 ADHD Metaproterenol (Alupent) 184,000 < 6 Asthma

OCD, obsessive-compulsive disorder; ADHD, attention-deficit hyperactivity disorder. Boldface indicates psychotropic agents. Data from regulations requiring manufacturers to assess the safety and effectiveness of new drugs

and biological products in pediatric patients. Federal Register 62:43899, 1997.

1268 WALKUP et a1

may alter the dosing of a medication or the expected side-effect profile. For example, the use of valproate in conjunction with other anticonvulsants is not a safe combination in infants9 Also, recent information suggests that the cardiac conduction system in children is more vulnerable to the effects of desipramine than it is in teens and ad~1ts . l~ Also, because pharmaceutical companies do not consider the use of their medication in children as part of their comprehensive plan to market a medication, there is a paucity of medication formulations designed specifically for children. Without alternative formulations, parents re- sort to cutting, crushing, or mixing available pills with liquid in an effort to provide the medication for their children. For example, physicians wishing to prescribe fluoxetine for children in the late 1980s would either prescribe the adult daily dose or suggest to parents that they empty the contents of the capsule into water and create a homemade colloidal suspension. Only with the advent of liquid fluoxetine was prescribing made easier for children, even though liquid fluoxetine was created for geriatric patients, not children.

The US Food and Drug Administration (FDA), in collaboration with the National Institute of Mental Health (NIMH), has attempted to increase the available information regarding safety and efficacy of psychotropic medication in children. Before 1979, there was no information regarding pediatric safety and efficacy required in the product labeling for medications marketed in the United States. To address this problem, the FDA proposed guidelines for the clinical evaluation of psychoactive drugs in infants and children (FDA publication 79- 3055, 1979) and developed the Pediatric Use subsection to be included in the product labeling and advertisingA of prescription drugs. Although these guidelines define how medications can be evaluated in children and highlight the importance of the safety and efficacy information, the 1979 guidelines did not necessarily stimulate pediatric drug research.27 Between 1979 and 1994, when the FDA revised the Pediatric Use subsection, there were only 17 adequately sized and rigorous studies published on psychotropic medications in childreazO

The FDA revised the 1979 guidelines for the Pediatric Use subsection in 199428 to increase the amount of information available on pediatric use. The 1994 guidelines proposed that pharmaceutical companies review the existing data on the use of their medications and, based on the available information, propose revisions of the Pediatric Use subsection. The 1994 guidelines emphasized that, where the course of the disease and drug response was similar in adults and children, efficacy data from adult studies could be extrapolated to children. Additional dosing, pharmacokinetic, and safety studies would need to be done, but more expensive and difficult efficacy studies would not be needed. The FDA retained the right to review such data, and, if the data were not considered "substantial evidence" of safety and efficacy, the pediatric labeling would include the now-famous default phrase, "Safety and effectiveness in pediatric patients have not been established." In addition, if the marketed medication was widely used, potentially hazardous, or an indication would be therapeutically beneficial, then the FDA could actually require the pharmaceutical company to conduct additional studies. For a medication not yet marketed in the United States, the rules were somewhat different. Controlled trials would need to be done for a specific pediatric indication, but pharmacokinetic data could be adequate to complete the Pediatric Use subsection. Under the 1994 guidelines, the FDA would not ordinarily require the sponsor of a new drug application to conduct safety and efficacy studies in children; however, the FDA could require studies in children if there was good likelihood that the medication would ultimately be used in children. Because the studies in children often took place after the medication was on the market or after the completion of studies in adults, a special office was developed in the FDA to track and support the


voluntary efforts of pharmaceutical companies to conduct research in the pediatric population.

The response to the 1994 guidelines has been underwhelming. In 1996, only 37% (15 of 40) of “new medical entities” with potential usefulness in the pediatric population had some pediatric labeling at the time of approval. An additional 17 drugs were approved, with promises by the manufacturers to complete pediatric studies after approval. Based on previous experiences, however, the numbers of pediatric studies promised and completed after approval are few.18

Given the overall lack of research progress, President Clinton announced on August 13, 1997, proposed FDA guidelines that require drug companies to conduct safety and efficacy studies in children for new medications and also for some currently marketed medications. A medication would be required to meet the standard of “substantial evidence” of safety and efficacy if the medication was used or ”likely to be used in a substantial number of pediatric patients” or if the medication ”offered a meaningful therapeutic benefit over existing treatments for children.” In addition to mandating the pharmaceutical industry to conduct studies, the FDA also described incentives for the development of adequate ”Pediatric Use” information. The proposal included user fee reductions (i.e., fees paid to the FDA at the time of application for a new indication) or the possibility of extended exclusivity. For some agents, an additional period of exclusivity before the compound coming off patent can mean millions of dollars. Several provisos were included in the proposal to address potential concerns of the pharmaceutical industry. As in the 1994 rule, studies would not be required if the disease and response to medication were considered to be similar among adults and children. Testing for medications that were potentially hazardous to children would occur only after completion of adult studies. Similarly, pediatric studies would or could be deferred if the completion of the trials delayed the approval process of medication for life-threatening diseases. Waivers of the guidelines could be initiated by the FDA or by the pharmaceutical company. Because this was a new proposal, a period of time was provided for hearings and feedback regarding the proposed guidelines. Most of the major professional associations have provided information to the FDA regarding the medications that are in need of study. The Committee on Drugs from the American Academy of Pediatrics and the American Academy of Child and Adolescent Psychiatry has cooperated with the FDA regarding these important guidelines.

The reaction to the proposed guidelines was significant. Spokespersons for the pharmaceutical industry have been vocal in opposition even though the industry as a whole is not uniformly opposed to studying medication in children. The current concerns include (1) more clinical trials for children are being conducted in children and adolescents; (2) children in clinical trials could be exposed to compounds whose safety has not been adequately studied even in adults; (3) children who participate in trials could file lawsuits in the future despite consents given by their parents; (4) identifying and recruiting children for these studies could be very difficult, making it hard to complete the study; (5 ) waiting for pediatric studies to be completed may result in an unnecessary delay in the approval of life-saving medications for adults; (6) clinical trial methodology will have to consider differences among children and adults, including differences in assessment, measurement, and reporting of clinical information; (7) analytical tools need to be developed to measure small amount of a drug in small aliquots of serum; (8) young children have difficulty ingesting tablets and capsules; devising other forms of the drug would be costly, additional work may also be needed to stabilize a compound in liquid or make the taste more acceptable to children; and (9) the potential lack of return on investment

1270 WALKUP et a1

makes it difficult to consider funding pediatric trials, especially if the pediatric studies do not result in any additional time of exclusivity to recoup the cost of the pediatric studies.

Many of the issues raised by the pharmaceutical industry have been ad- dressed by the pediatric research community in the past, including the types of controlled trials that are child friendly'O and the ethical and legal issues in pediatric psychopharmacology research, including informed consent.z, 7, l1 Yet despite the ongoing concerns regarding research in children, concern is mount- ing about the use of medication in pediatric populations without adequate safety and efficacy data. Although it is impossible to predict the outcome of these debates, it seems clear that the scientific community and, to some extent, the public are increasingly interested in the expansion of our knowledge about psychopharmacologic treatments in children and adolescents. Given the pattern of FDA guideline development since 1979, there may come a time when studies in children will not really be an "add-on" to safety and efficacy studies in adults but will be an essential component of any new drug application. From the perspectives of professionals who work with children, it is clear that the distinc- tions that allow the exclusion of children from research studies are less tenable than ever.

In November 1997, FDA Modernization and Accountability Act was passed by Congress and signed by President Clinton. The overall goal of this legislation is to overhaul the FDA to decrease, the regulatory burden on industry, protect consumers, and make the process of approving new medications faster and the FDA more efficient. The legislation has many facets but provides new incentives for pharmaceutical developers to test their products on children. The major incentive to conduct pediatric studies is an additional 6 months of product exclusivity, which, for many medications, can mean millions of dollars for pharmaceutical companies.

Efforts by the National Institutes of Health to Facilitate Research into the Use of Medications by Children

In his confirmation hearing in 1993, the director of the National Institutes of Health (NIH), Harold Varmus, MD, pledged to facilitate the extension of biologic discoveries to clinical settings. In 1997, he outlined, in his testimony before Congress, the efforts made by the NIH to accomplish his goal-efforts to increase the number of clinical investigators; funding for research; and collaborative efforts among the federal government, academia, and private industry. Congress has also been very active not only in supporting research but also supporting clinical trials research-an area that in many respects was not as actively supported in the past. More recently, the NIH mandated that as of October 1, 1998, children be included in all research studies and, if they are not, that the rationale for their exclusion be provided. The new mandate for children's participation in research parallels the mandate to include women and minorities in all research endeavors. To support the new research initiatives, the federal government has planned a significant increase in funding over the coming year. This past year, more than $13 billion was approved for biomedical research. Under this new plan, increases in funding could be as large as 15% per year compared with the increase of 7.1% for the fiscal year 1997-1998. Because problems affecting children and mental health are priority areas, it is expected that a significant amount of new federal money will be available for treatment research, including psychopharmacologic treatment research.


Two specific efforts by the NIH are noteworthy. Within the past 5 years, the NIH, through its branch institutes, have funded the Pediatric Pharmacology Research Units (National Institute of the Child Health and Human Development). The NICHD funded seven PPRUs in 1993. The goal of the PPRUs was to establish a network of sites that can do the kinds of pediatric clinical trials necessary for non-psychiatric medications in children. An additional three PPRU sites will be added in 1998. In addition, the NIMH has actively funded research into the study of the treatment of childhood psychiatric disorder with medications, and funded seven Research Units of Pediatric Psychopharmacology (RUPPs), which were modeled after the PPRUs. The goal of the RUPPs is to conduct studies of already marketed medications for the treatment of childhood psychiatric disorders.

Efforts by the NIMH to Facilitate Research on the Use of Medications to Treat Childhood Psychiatric Disorders

In 1995, the NIMH in conjunction with the National Plan for Research on Child and Adolescent Mental Disorders, identified pediatric psychopharmacology as an important bu't neglected area of research.26 That same year, a large conference sponsored by the NIMH and the FDA gathered more than 100 individuals from academia, the pharmaceutical industry, and patient and professional groups to discuss pediatric psychopharmacologic research and to make recommendation regarding solutions to the lack of information on the safety and efficacy of psychotropic medications in children. Obstacles to studying pharmacologic treatment for childhood psychiatric disorders were discussed at this conference, including the current regulatory environment, combined pharmacotherapy, methodological and logistic issues, ethical and legal issues and issues critical to the involvement of the pharmaceutical

As a result of the conference, the NIMH commissioned reviews of the pediatric psychopharmacology literature by experts in the field to identify whether there are sufficient data already available that could add to the existing information on the Pediatric Use of psychotropic agents. These reviews would answer the challenges posed by the FDA to pharmaceutical companies in the 1994 FDA guidelines to review the existing data and to make proposals regarding changes in the product labeling of already marketed psychotropic medications. At the current time, the six commissioned reviews, covering all agents currently used in pediatric psychopharmacology, are being reviewed for publication in the Journal of the American Academy of Child and Adolescent Psychia- try.

Another result of the Conference was the establishment by the NIMH of the contractual network of Research Units of Pediatric Psychopharmacology (RUPPs). The RUPPs provide funding to the institutions to conduct definitive pharmacological treatment studies for childhood psychiatric disorders. It was the goal of the RUPPs to combine existing resources within institutions with the financial and technical resources of the pharmaceutical industry and the federal government. Initially two RUPP sites were funded, Columbia (Larry Greenhill, PI) and Johns Hopkins (Mark Riddle, PI) in 1996. A third site was added shortly thereafter (University of Pittsburgh, Boris Birmaher, PI). These 3 sites are studying selective serotonin reuptake inhibitors for serious anxiety disorders in children and adolescents. The Columbia and Hopkins sites are collaborating on a fluvoxamine trial and the Pittsburgh site is using fluoxetine. More recently a second group of RUPP sites were developed to address the pharmacological

1272 WALKUP et a1

treatment of autism and other pervasive developmental disorders. These four sites include Ohio State/Kennedy-Krieger Institute (Michael Aman, PI); Univer- sity of Indiana (Christopher McDougle, PI); University of California at Los Angeles (James McCracken, PI); and Yale University (Fred Volkmar, PI). These sites are collaborating on a trial of risperidone for autistic symptoms.

The RUPPs have been successful in collaborating with the Pharmaceutical industry. For the Columbia/Hopkins treatment study of anxiety, additional funding was obtained from the pharmaceutical industry. The very fact that the RUPP network existed and that it combined academic and federal government resources made the RUPPs credible partners for the pharmaceutical industry. Hopefully other pharmaceutical companies will use the resources of the RUPPs to test their medications in childhood psychiatric disorders. In addition to collaboration with the pharmaceutical industry, there is a plan for the PPRUs and the RUPPs to collaborate on a study of stimulants in children aged 3 to 5 years with ADHD. In this way, a larger sample of children can be recruited in a shorter period of time. The collaboration between pediatricians in the PPRUs and the psychiatrists in the RUPPs promotes interspecialty collaboration not previously done in the treatment of childhood psychiatric disorders.

The NIMH has also sponsored the largest controlled trial of ADHD. The Collaborative Multimodal Treatment Study of Children with ADHD, otherwise know as the MTA, is the first multisite cooperative agreement study of children conducted by the NIMH.' The goal of the study is to compare state-of-the-art treatment of ADHD with routine clinical care. Using a parallel-groups design, 576 children aged 7 to 9 years with ADHD were randomly assigned to one of the four treatment groups: (1) stimulant medication, (2) behavioral treatment, (3) combined medication and behavioral treatment, or (4) routine community- based clinical care of the parents' choosing. Treatment is long-term. Subjects in groups 1 to 3 were treated for 14 months, then followed-up for an additional 10 months. Children in the community control group were followed for 24 months. At the beginning of the initial treatment phase, all children in the medication groups (1 and 3) underwent a 1-month, blind titration with methylphenidate and, if necessary, open treatment with other drugs to identify the most effective stimulant and proper dose for each subject. The study began in 1994 and is finishing in 1998. In contrast to the many pharmacologic efficacy studies, this study has the methodologic capacity to change how patients with ADHD are treated. The investigators attempted to include flexibility and relevance in the design to enhance generalizability of the results. Of particular importance, the subject sample was not a purified one. The lack of extensive exclusion criteria guarantees that the subjects enrolled in the study were more similar to patients seen in clinical practice than subjects in other controlled treatment trials.

The results are not yet available. On November 1618,1998, an NIH Consen- sus Development Conference on ADHD will be held in Washington, DC. A Consensus Development Conference is held when there is reason to review what is known about a disorder, including treatment, and to write a consensus statement. A panel of esteemed scientists, not from the field being studied, reviews the testimony of experts and then writes the consensus statement. When the consensus statement on ADHD is completed, it will likely be published in the Journal of tlze American Medical Association. This will be the first Consensus Development Conference on a childhood psychiatric disorder. The MTA study is an important component of the treatment data to be presented at the conference.

Following the success of the MTA study, a request for applications was released in March 1998 for a large collaborative study of cognitive behavioral therapy and pharmacotherapy for depression in children and adolescents. The


described goal of this request for applications is to develop a very large study, similar to the MTA study for ADHD, that would collect the definitive data on the role of each form of therapy and the best approach to integrate these treatments. Such large collaborative studies seem to be the future of psychopharmacology research.

FUTURE APPROACHES TO ASSESSMENT OF CHILDHOOD PSYCHIATRIC DISORDERS

Within the past 20 years, there has been substantial progress in the assessment of childhood psychiatric disorders. Borrowing from the advances in general psychiatry, the effort to identify childhood psychiatric disorders based on valid or at least standardized, explicit, observable, diagnostic criteria free of etiologic assumptions has changed the practice of child and adolescent psychiatry. Parallel with the development of diagnostic criteria was the development of diagnostic interviews and symptom checklists. Because medical tests for psychiatric disorders are not available for children (or adults), systematically ascertaining diagnostic information became absolutely critical to assessment and conducting research. Increasingly, however, there is concern that diagnostic interviews and symptom checklists do not provide the kind of information necessary to adequately assess for diagnosis.17, 21 Two trends are occurring in the assessment of child psychiatric disorders to address these concerns. The first is a move away from strictly categorical approaches to consideration of combined categorical and diagnostic specific dimensional approaches. For example, diagnoses of learning disorders are increasingly considered more dimensional than strictly categori-

and most studies of ADHD are using both categorical and dimensional approaches. In the authors' own studies of anxiety disorders in children, categorical diagnoses are combined with dimensional measures of anxiety severity to more broadly assess the impact of medications on anxious symptoms. Because separation anxiety, social phobia, and generalized anxiety often overlap and have similar long-term outcomes, it is prudent to not only consider each disorder individually but also to consider the three disorders as potentially a single clinical entity with varying manifestation^.'^ Other efforts to add to the descrip- tive power of categorical diagnoses are the addition of impairment criteria as part of a psychiatric diagnosis. The addition of impairment criteria supports the assertion that the diagnosis in question is truly sufficiently severe to warrant intervention.

A second trend is to identify clinical markers for inclusion in specific diagnostic groups. Markers can be conceptualized as historical, such as a positive family history of an illness or behavior. Markers can also be considered behavioral, such as impairment on a continuous performance test for ADHD. Markers can also be considered biologic. Traditionally, biologic studies of psychiatric disorders have been undertaken to understand the pathophysiology of a condi- tion. More recently, investigators have begun to use diagnostic criteria in conjunction with putative biologic markers as assessment components of pharmacologic treatment studies. For example, investigators of anxiety disorders are using diagnostic challenge studies in the assessment of treatment o~tcome. '~ Clinicians studying affective disorders are using studies of sleep architecture to identify "true cases" and also to assess treatment outcome. Some clinical studies are using genotyping of study subject to predict drug response (Wilens and col- leagues, personal communication, 1998). Neuroimaging studies have been used

1274 WALKUP et a1

to demonstrate treatment change3, 5, 25 and will likely be used in the future in treatment outcome studies.

THE FUTURE OF CLINICAL TRIAL STUDY DESIGN

The FDA standard for assessing efficacy of medications in adults is two large, multisite, placebo-controlled trials. The reason double-blind, placebo-controlled trials are considered best is in part tradition but also other designs can have flaws that can impact on outcome and generalizability. If this standard were to be applied to pharmacologic treatment trials in children, it would be a very long time before adequate (by this standard) safety and efficacy data would become available. As a result of the FDA guideline changes noted earlier, the standard has been changed to one well-controlled study, even for adult disorders. Although this is a clear improvement, traditional parallel-group, placebo- controlled trials may not offer the best data on efficacy for childhood psychiatric disorders. If the chance of being on placebo is viewed negatively by the child or his or her parents, it may introduce a significant sampling bias that would distort the outcome and generalizability of the study. Investigators are increasingly considering other study designs to assess efficacy in childhood psychiatric disorders, including? (1) double-blipd, placebo substitution studies; (2) comparisons of two or more active treatments; (3) a variety of crossover trials; (4) the addition of nonpharmacologic treatments as comparison treatments; (5) the addition of long-term, open treatment at the end of a short-term, double-blind, controlled trial as an incentive for participating in the short-term trial; (6) ultra-short, controlled trials when variables of early response can be reliably identified.I3

One recently completed, industry-sponsored trial (Paroxetine for Childhood ODC) used a unique study design to establish efficacy and to address the issue of recruitment of child subjects. In this trial, subjects were treated openly for 16 weeks before either randomization to placebo or to continue on active medication. Recruiting for this study was easy. Subjects knew they were getting active drug in the first phase of the study, and they also knew that they would be randomized only if they responded to medication. If they failed to respond in the first phase, they would leave the study and could then get another form of treatment. Subjects who were randomized to placebo in the second phase would have improved already and would not necessarily lose complete symptom control before exiting the study and using the treatment information in the first phase of the study to restabilize on the active medication. This was a unique study design, but it had some limitations. Generalization of study finding may be an issue because the only patients randomized were those that had already demonstrated clinical response. In addition, it presupposes that, when a subject is transitioned to placebo in the second phase, the patient will relapse and the subject's symptoms return in the same pattern as they went away in the acute treatment phase. Although this is a reasonable assumption, it is not necessarily true. In addition, a recurrence of symptoms in the second phase of the trial needs to be differentiated from other adverse events and may make it difficult to assess outcome. When the data analysis is complete, the impact of these factors on the outcome of this study will be an important component of the discussion.

THE FUTURE OF TREATMENT

Several trends are developing in the treatment of children with psychiatric disorders. There is significant interest in developing clinical algorithms for the


treatment of routine and complex psychiatric presentations. Clinical algorithms have been developed for adult psychiatric disorders,3O and some efforts have been made to develop pharmacologic algorithms for childhood psychiatric disor- d e r ~ . ~ ~ Increasingly, managed care organizations are looking for broader treatment algorithms that include practice standards for the evaluation and best practice standards for the treatment of childhood psychiatric disorders.

There is also a trend toward the use of combined treatments, including combined medication treatments. Although combining psychosocial treatments with pharmacotherapy is often considered optimum treatment; most combined treatments have never been studied. A recent study of the combined treatment for ADHD suggests that, for many patients with ADHD, psychosocial treatment may not offer additional benefit above that which is experienced with medication alone (Abikoff, personal communication, 1998).

The use of medication in combination is controversial in child and adolescent psychiatry because of the legacy of polypharmacy; however, for patients with severe symptomatology, multiple agents are used in combination to offer benefits for a complex array of symptoms. Given the lack of efficacy studies of single agents, it is likely to be some time before studies of multiple medication combinations will be conducted. There are some notable exceptions, including the use of neuroleptic augmentation of SSRIs in patients with tics and OCD.12

THE FUTURE OF DRUG DEVELOPMENT

Estimates vary, but, on average, it takes approximately 15 years for an experimental medication to become commercially available in the United States. Although many medications are studied in the laboratory, only about 0.1% get tested in humans, and only approximately 1 in 5 (0.02% overall) medications tested in humans becomes commercially a~ailable.~

Most psychotropic medications made available for use over the past 25 years were discovered to have psychotropic effects almost by accident. The classic example is the identification of antipsychotic effects of chlorpromazine after discovering that some medications developed as antihistamines produced sedation. In many respects, this discovery revolutionized psychiatry. Many suc- cessfully treated patients were discharged from long-term treatment facilities, and, perhaps more importantly, conceptual models regarding the causes of psychoses changed, albeit slowly, from a psychologic defect to a biologic one.

Progress in drug development has occurred in four phases.” After the phase of discovery by pure serendipity, the second phase included the use of animal models to test drug effects to identify potential therapeutic candidates. Predict- able animal models of drug response were used to standardize the process. Basic scientific knowledge derived from animal models led to the third phase, in which more laboratory-based processes were used to assess the action of a potential medication. The last phase is in part a derivative of the third phase. Advances in molecular biology in combination with sophisticated computer programs allow for new molecules to be designed based on knowledge of the structure of receptors, neurotransmitters, and other neuropeptides. Medications will not only target the cell surface but may ultimately be able to be sent into cells to act directly on the pathologic mechanism. Perhaps one of the ironies of new drug development is that serendipity is much less of an issue in the identification of new and specifically targeted compounds, but serendipity may still have a role as these new medications come to the bedside. Drug develop-

1276 WALKUP et a1

ment still requires astute clinical observers to understand what disorders and what symptoms will respond to these new medi~ations.2~

The creation of medications that prevent or even correct pathologic mechanisms is the future of pediatric psychopharmacology. Given that the medications we now use work primarily to reduce symptom severity and rarely cures or consistently prevents illness, we can expect a very exciting future.

FUTURE MEDICATIONS

Antipsychotic Medications

Since the discovery of the effect of clozapine on both the positive and negative symptoms of schizophrenia, drug development has focused on the development of a medication with similar clinical effects but without the serious risk for agranulocytosis. Trends in the early 1980s focused on the development of specific and selective dopamine-blocking agents. With the success of clozapine, whose efficacy is purported to be caused by its broad-based receptor activity, the emphasis on antipsychotic drug development has shifted to the creation of agents with more broad-based receptor activity. Several new medications that also have broad-based receptor activity have subsequently been developed and marketed, including risperidone; olanzapine; quetiapine; and, most recently, ziprasidone. Many more of these medications-those that impact several receptor types-are currently under study. These new medications will likely demonstrate better tolerability and less risk for the long-term side effects (e.g., tardive dyskinesia) seen with the typical neuroleptics. One would expect that the spec- trum of their use will expand beyond the treatment of schizophrenia to the routine treatment of affective illness, such as bipolar disorder, and perhaps other childhood conditions, such as autism.

Antidepressant Medications

The number of new antidepressant medications in the pipeline is also significant. Following on the success of fluoxetine, four other SSRIs have been marketed, most recently citalopram.22 The trend to identify selective serotonergic agents has expanded with the identification of several serotonin receptor~.'~ Some of these new selective serotonin medications have antagonist activity, and others have more specific agonist activity. Another trend in the development of antidepressants is the development of medications that are selective, but selective for two receptor types; those that block the reuptake of both serotonin and norepinephrine, and serotonin and dopamine.

Mood Stabilizers

The advances in medications for epilepsy will likely be extended to the treatment of psychiatric disorders, especially mood disorders, impulsivity, and aggression. A few of the new anticonvulsants have already been piloted in adult patients with bipolar disorder.I6 Based on the success in preliminary trials, one new anticonvulsant, lamotrigine, is pursuing an indication for adult bipolar disorder.


ADHD Medications

The very long search for a highly effective, long-acting stimulant that does not suppress appetite, impact on sleep, or cause rebound hyperactivity is still underway. There is no current program of drug development specifically for ADHD, so new medications for ADHD will come from other areas of drug development. Novel antidepressants with noradrenergic or dopaminergic activity such as bupropion, venlafaxine, and mirtazapine, will likely be studied and used for ADHD. Also nonantidepressant medications with noradrenergic or dopaminergic activity will be studied. The most unique approach to the treatment of ADHD may be the use of medications developed for dementia. At the curren time there are very few of these medications available, yet many more are under study. These medications may revolutionize the treatment of dementia and perhaps even be beneficial in ADHD and the learning disorders.

SUMMARY

The future of pediatric psychopharmacology will be a collaborative effort among the public, the federal government, the pharmaceutical industry, and clinical scientists. Clinical scientists and the NIH have initiated the process. The federal regujatory guidelines are in place but may need to be amended further to truly facilitate needed research. The pharmaceutical industry is making efforts to study medications in childhood psychiatric disorders. More new drugs are being tested in humans. Many of the new agents offer the promise of more clinical benefit with milder side effects. The future is indeed promising.

References

1. Arnold LE, Hoagwood K, Jensen PS, et al: Psychopharmacology in children and adolescents: Toward clinically relevant trials. Psychopharmacol Bull 33:135, 1997

2. Arnold LE, Stoff DM, Cook E, et al: Ethical issues in biological psychiatric research with children and adolescents. J Am Acad Child Adolesc Psychiatry 34(7):929-939,1995

3. Baxter LR, Schwartz JM, Bergman KS, et al: Caudate glucose metabolic rate changes with both drug and behavior therapy for obsessive-compulsive disorder. Arch Gen Psychiatry 49:681, 1992

4. Beary JF 111: The drug development and approval process. In America PRaMo (ed): New Medicines in Development for Mental Illness, vol. 1996 Survey. Washington, DC, PhRMA, 1996, p 1

5. Berman I, Merson A, Sison C, et al: Regional cerebral blood flow changes associated with risperidone treatment in elderly schizophrenia patients: A pilot study. Psycho- pharmacol Bull 32:95, 1996

6. Bradley C: The behavior of children receiving Benzedrine. Am J Psychiatry 94577,1937 7. Coffey BJ: Ethical issues in child and adolescent psychopharmacology Child Adolesc

Psychiatr Clin North Am 4:793, 1995 8. Conners CK, Eisenberg L: Effect of dextroamphetamine-amphetamine on children.

Arch Gen Psychiatry 17:478, 1967 9. Eadie MJ, Hooper WD, Dickinson RG: Valproate-associated hepatotoxicity and its

biochemical mechanisms. Med Technology 3:8S, 1988 10. Fava M: Traditional and alternative research designs and methods in clinical pediatric

psychopharmacology. J Am Acad Child Adolesc Psychiatry 35:1292, 1996 11. Jensen PS, Hoagwood K, Fisher CB: Bridging scientific and ethical perspectives: Toward

synthesis. In Hoagwood K, Jensen PS, Fisher CB (eds): Ethical Issues in Mental Health Research with Children and Adolescents. Mahwah, NJ, Lawrence Erlbaum, 1996

1278 WALKUP et a1

12. McDougle CJ, Goodman WK, Price LH, et al: Neuroleptic addition in fluvoxamine

13. Montgomery SA: Are 2-week trials sufficient to indicate efficacy? Psychopharmacol

14. Peroutka SJ: 5-HT receptors: Past, present, and future. Trends Neurosci 18:68, 1995 15. Pine DS, Cohen P, Brook J: Emotional problems during youth as predictors of stature

during early adulthood: Results from a prospective epidemiologic study. Pediatrics 97856, 1996

refractory obsessive-compulsive disorder. Am J Psychiatry 147652, 1990

Bull 31:41, 1995

16. R & D Directions Staff Epilepsy market comes to life. R & D Directions 4:56, 1998 17. Regier DA, Kaelber CT, Rae DS, et a1 Limitations of diagnostic criteria and assessment

instruments for mental disorders. Arch Gen Psychiatry 55:109, 1998 18. Regulations requiring manufacturers to assess the safety and effectiveness of new

drugs and biological products in pediatric patients. Federal Register 62:43899, 1997 19. Riddle MA, Geller B, Ryan N: Another sudden death in a child treated with desipra-

mine. J Am Acad Child Adolesc Psychiatry 32:792, 1993 20. Riddle MA, Subramaniam G, Walkup JT Efficacy of psychiatric medications in children

and adolescents: A review of controlled studies. Philadelphia, WB Saunders, 1998, pp 269-285 Psychiatr Clin North Am

21. Rutter M: Child psychiatric disorder: Measures, causal mechanisms, and interventions. Arch Gen Psychiatry 54785,1997

22. Sambunaris A, Keppel Hesselink J, Pinder R, et al: Development of new antidepressants. J Clin Psychiatry 58:40, 1997

23. Shaywitz B, Fletcher B, Shaywitz $: Defining and classifying LD and AD/HD. J Child Neurol 10:550, 1995

24. Tallman JF, Dahl SG: New drug design in psychopharmacology. In Bloom FE, Kupfer DJ (eds): Psychopharmacology. New York, Raven Press, 1995, p 1861

25. Tamminga CA, Conley RR: The application of neuroimaging techniques to drug development. J Clin Psychiatry 58:3, 1997

26. U.S. Food and Drug Administration: FDA Bureau of Drugs Clinical Guidelines: Guide- lines for the clinical evaluation of psychoactive drugs in infants and children. Washing- ton, DC, US Government, 1979

27. U.S. Food and Drug Administration: Labelling and prescription drug advertising: Content and format of labelling for human prescription drugs. Federal Register 44:37434, 1979

28. U.S. Food and Drug Administration: Specific requirements on content and format of labelling for human prescription drugs: Revision of pediatric use subsection in the labelling. Federal Register 59:64240, 1994

29. Vitiello B, Jensen P S Medication development and testing in children and adolescents. Arch Gen Psychiatry 54:871, 1997

30. Walkup JT Clinical decision making in child and adolescent psychopharmacology. Child Adolesc Psychiatr Clin North Am 4(1):23-40, 1995

Address reprint requests to John T. Walkup, MD

Johns Hopkins Hospital CMSC 343

600 North Wolfe Street Baltimore, MD 21056

e-mail: [email protected]

THE FUTURE OF PEDIATRIC PSYCHOPHARMACOLOGY

Documents