Clinical trials in children with Down syndrome Issues from a cognitive research perspective.код для вставкиСкачать
American Journal of Medical Genetics Part C (Seminars in Medical Genetics) 142C:187 – 195 (2006) A R T I C L E Clinical Trials in Children With Down Syndrome: Issues From a Cognitive Research Perspective JAMES H. HELLER,* GAIL A. SPIRIDIGLIOZZI, BLYTHE G. CRISSMAN, JENNIFER A. SULLIVAN-SAARELA, JENNIFER S. LI, AND PRIYA S. KISHNANI Clinical and translational research play a key role in the transition of basic research discoveries to effective therapies. In Down syndrome (DS), these research approaches are not well utilized or developed to test new therapies to improve cognitive and/or adaptive function in this population. This article reviews the history of clinical trial research in children with DS from a cognitive research perspective and discusses important issues relevant to the conduct of well designed clinical trials for this population. Specific issues addressed include: funding, study design, study medication, subject recruitment and retention, safety, and efficacy challenges. The Duke Down Syndrome Research Team’s program of clinical research of cholinesterase inhibitors for individuals with DS serves as the model application for the identified research principles. It is hoped that this article will raise awareness of the unmet need for clinical research in the cognitive and adaptive function of individuals with DS, especially children with DS. ß 2006 Wiley-Liss, Inc. KEY WORDS: clinical trials; trisomy 21; cognitive deficit; language impairment; attention; memory; expressive language; rivastigmine tartrate; donepezil hydrochloride; cholinergic therapy; pediatric; Down syndrome How to cite this article: Heller JH, Spiridigliozzi GA, Crissman BG, Sullivan-Saarela JA, Li JS, Kishnani PS. 2006. Clinical trials in children with Down syndrome: Issues from a cognitive research perspective. Am J Med Genet Part C Semin Med Genet 142C:187–195. INTRODUCTION Down syndrome (DS) results from chromosomal aneuploidy (trisomy 21). The consequent gene dosage imbalance is believed to be the main cause of the phenotype. Despite the fact that the phenotype was described over 100 years ago, much about the condition remains unexplored [Jones, 1996]. Recent advances in basic, translational, and clinical research along with the development of subspecialty expertise have yielded substantial improvements in the medical care of individuals with DS. These advances in DS investigation have contributed to an increased understanding of the more complex medical and developmental issues associated with DS. Clinical and translational research play key roles in the transition of basic research discoveries to effective therapies. The randomized, double-blind, The Duke Down Syndrome Research Team (JH, GAS, BGC, JASS, PSK), established in 1997, is a collaborative group of medical sub specialists who design and conduct clinical research including the investigation of potential therapeutic effects of cholinesterase inhibitors in adults and children with Down syndrome. James H. Heller M.A., M.S., C.C.C. is a Clinical Associate, Department of Surgery, Duke University Medical Center (DUMC). He has advanced degrees in speech-language pathology (U of Minnesota) and in experimental psychology (Memorial U of Newfoundland, Canada). He is the former director of the Child Development Unit, Department of Pediatrics, DUMC and has over 25 years clinical research experience in the field of communication disorders. Gail A. Spiridigliozzi, Ph.D. is a clinical child psychologist and Assistant Clinical Professor, based in the Duke Child Development and Behavioral Health Clinic. She is also the psychologist for the Duke Down Syndrome Research Team. Her research and clinical interests include the psychoeducational profiles of children with Down syndrome, fragile X syndrome, metabolic disorders, and other genetic conditions. Blythe G. Crissman, M.S., CGC is a genetic counselor in the Department of Pediatrics at Duke University Medical Center. She serves as the primary genetic counselor and clinic coordinator of the Duke Comprehensive Down Syndrome Clinic, and is a clinical research coordinator for the Duke Down Syndrome Research Team. Jennifer A. Sullivan-Saarela, M.S., CGC is a senior genetic counselor in the Department of Pediatrics at Duke University Medical Center. She served as the research coordinator of the Duke Down Syndrome Research Team from its inception, 1997–2002. She currently serves as the coordinator and senior genetic counselor for the Duke Metabolic Clinic and Lysosomal Storage Disease Center. She is also Associate Editor of the Journal of Genetic Counseling. Jennifer S. Li, M.D., MHS is Associate Professor of Pediatrics (cardiology) and Director of Pediatric Clinical Trials at the Duke Clinical Research Institute. She has a research focus in cardiac issues in patients with Down syndrome. Priya S. Kishnani, M.D. is Associate Professor of Pediatrics and the Director of Clinical Trials in the Division of Medical Genetics at Duke University Medical Center. Her primary focus has been the translation of laboratory science into the clinic arena to advance clinical care. She has also been the coDirector of the Duke Comprehensive Down Syndrome Clinic since 1995. *Correspondence to: James H. Heller, DUMC Box 3528, 244 Bell Building, Duke University Medical Center, Durham, NC 27710. E-mail: email@example.com DOI 10.1002/ajmg.c.30103 ß 2006 Wiley-Liss, Inc. 188 AMERICAN JOURNAL OF MEDICAL GENETICS PART C (SEMINARS IN MEDICAL GENETICS): DOI 10.1002/ajmg.c placebo-controlled trial has become the gold standard for establishing efficacy and to a lesser extent safety [Alderman, 2005]. To date, clinical research in DS has focused on the utility of treatments for leukemia [Gamis, 2005]. This research has led to best practice standards that have extended the life span of individuals with DS affected by hematologic malignancies. However, as the lifespan of individuals with DS has increased significantly [Yang et al., 2002] and more individuals with DS are active members of the community, maximizing functional potential has become an immediate need. The development of educational policies, such as early intervention and inclusion in the least restricted environment, have helped focus attention on children’s abilities rather than their disabilities [Baker et al., 1994; Hollowood et al., 1995; Hassold and Patterson, 1998]. These policies bring hope that individuals with DS can lead relatively independent and productive lives. In light of the strong efforts and accomplishments in improving the lives of individuals with DS, the cognitive deficits associated with DS currently loom as a formidable barrier to even greater accomplishment. There is a paucity of published research in pharmacologic therapy for improving cognitive function in individuals with DS. Most of the available literature is based on anecdotal case There is a paucity of published research in pharmacologic therapy for improving cognitive function in individuals with DS. Most of the available literature is based on anecdotal case reports and small, single center, open-label, or placebo control studies of agents such as vitamin and mineral supplements. reports and small, single center, openlabel, or placebo control studies of agents such as vitamin and mineral supplements. A listing of clinical trials investigating the potential enhancement of cognitive function in children with DS is provided in Table I. It is the goal of researchers and practitioners to develop strategies and interventions to maximize the functional potential of all individuals with DS. Due to many circumstances, including a shortage of funding and reluctance of investigators and marketers to test unproven therapies in rigorous clinical trials, there is a limited history of shared effort towards this goal. Thus far, interventions focusing on the cognitive impairments of individuals with DS by the medical community have been notoriously slow. This may be due, in part, to their adherence to two important principles: (1) do no harm and (2) medical practice must be evidencebased. To date, there have been no carefully designed and controlled studies of a pharmacological agent that have been proven effective in improving cognitive function in children with DS. Without this kind of study, the adoption of any intervention to improve cognitive function in individuals with DS in the medical community is unlikely. In an attempt to enhance cognitive abilities, many parents choose to use complementary and alternative therapies and off-label medication for their children with DS. These parents are not willing to lose time waiting for conclusive scientific data for their children who may be at a critical stage in cognitive and/or language development. Their decision whether or not to actively seek a particular intervention is pragmatic (i.e., do I think that this intervention might help my child?). As medical researchers and practitioners, we face the dilemma of maintaining a balance between evidence from conclusive scientific data to justify practice and the need of parents to identify promising treatments for their children while their children are young enough to obtain maximum benefit. The most obvious solution for medical researchers to resolve these ARTICLE competing needs is to initiate large-scale double-blind, placebo-controlled clinical trials with potentially effective interventions. Ideally, therapies shown to be ineffective in these trials can be eliminated from consideration and therapies shown to be safe and efficacious can be adopted. Unfortunately, this solution is often not as simple as it seems. In the following sections, we will identify and discuss the clinical research issues that make this solution so challenging. FUNDING Available funding of clinical trials for enhancement of cognitive function in DS is extremely limited both federally and privately. Despite the FDA Modernization Act of 1997  and the Best Pharmaceuticals for Children Act, which include provisions for 6 months of patent protection under pediatric exclusivity, pharmaceutical companies have been reluctant to support trials in a population with special needs, particularly when the focus is on cognitive deficits. Although small grants are of merit when gathering pilot and preliminary data, more comprehensive and rigorously designed studies, with larger sample sizes or multiple site trials are needed to get a medication approved for a particular indication. These kinds of studies require substantial financial support. STUDY DESIGN While funding is a critical factor in establishing a double-blind placebocontrolled trial, it is not the only factor. Necessary preliminary studies include the establishment of an optimal dose and titration schedule through pharmacokinetic studies, potential efficacy (and how this should be measured with a robust end point), and determination of the relative toxicity of the drug. Without these, it is unlikely that sufficient funding for a definitive double-blind, placebocontrolled trial would become available. And even if sufficient funding could be obtained, without completing this preliminary work, the likelihood of a conclusive result is low. ARTICLE AMERICAN JOURNAL OF MEDICAL GENETICS PART C (SEMINARS IN MEDICAL GENETICS): DOI 10.1002/ajmg.c 189 TABLE I. Pediatric Clinical Trials for Enhancing Cognitive Function in Individuals With DS Investigator and year Drug tested Type of agent Berg et al.  Pituitary extract Heaton-Ward  Bumbalo et al.  Niacin U series vitamin Pueschel et al.  Vitamin B6, 5-hydroxytryptophan, or both Megavitamins and thyroid hormone Vitamins and Minerals Vitamins and Minerals Megavitamins Vasopressin Harrell et al.  Bennett et al.  Weathers  Smith et al.  Eisenberg et al.  Lonsdale and Kissling  Bidder et al.  Thiamine Coleman et al.  Lobaugh et al.  Multivitamins and minerals Vitamin B6 Piracetam Heller et al.  Donepezil hydrochloride Heller et al. [in press] Rivastigmine tartrate Small pilot studies provide the essential background information for the design of larger, better controlled trials. Pilot studies provide perspective on general safety and highlight specific systems and effects that may require more careful monitoring. Dose and Small pilot studies provide the essential background information for the design of large, better controlled trials. Pilot studies provide perspective on general safety and highlight specific systems and effects that may require more careful monitoring. Study design Open label Number of subjects Placebo controlled Placebo controlled 23 24 Placebo controlled 95 Vitamin compound Open label Vitamin supplement Vitamin supplement Vitamin supplement Antidiuretic hormone Vitamin supplement Placebo controlled Placebo controlled Placebo controlled Open label 20 47 56 9 5–13 years 6–17 years 7–15 years 10–42 years Placebo controlled 22 8–16 years Vitamin supplement Crossover 26 Vitamin supplement Cyclic derivative of gammaaminobutyric acid Acetylcholinesterase inhibitor Acetyl and butyl cholinesterase inhibitor Placebo Controlled Double blind crossover 19 25 6 months– 5 years 10–42 years 6–13 years Open label 7 8–13 years Open Label 11 10–18 years titration schedule effects can be investigated. These factors are particularly important in children where wide variability in dose and titration tolerance can be observed. While maximum dose and titration schedules are determined in the study planning stage, these targets may not be reached by particular subjects if the subjects demonstrate a significantly increased adverse event profile at a lower dose level or on a faster titration rate. It is important to build in flexibility in dose and/or titration rate to minimize adverse events. The loss of subjects, particularly in small sample trials, reduces the study’s internal validity [Katz, 2005]. In our studies of cholinesterase inhibitors [Heller et al., in press; Kishnani et al., 1999; Heller et al., 2003, 2004], the dose level was increased gradually and subject response to the 3 Age range Bovine pituitary extract Vitamin supplement Compound of 48 agents Vitamin 4 16 months- 4 years 6–36 years 3 months-11 years Newborn 5–9 years increased dose was carefully monitored. Dose tolerance is affected by titration rate [Kishnani et al., 2001]. If the subject could not tolerate the increased dose at the slow rate of titration, the dose was backed down to the previously tolerable level. Efficacy is a key element of clinical trial design that directly affects the external validity of the study [Pajonk, 2005]. In order for the study drug to be incorporated into standard practice, positive study results must be generalizable to real-world practice. In pilot studies, the inherent weakness of knowing that a subject is on the experimental medication and the absence of a placebo control group creates bias in data interpretation including: (a) the potential to associate all change with the study medication; and (b) the potential to 190 AMERICAN JOURNAL OF MEDICAL GENETICS PART C (SEMINARS IN MEDICAL GENETICS): DOI 10.1002/ajmg.c observe change that is not really there. Thus, despite positive pilot study outcomes, the off-label use of medications for cognitive enhancement should be discouraged [Salman, 2002]. In an attempt to maximize the external validity of a clinical trial of an experimental medication for individuals with DS, one might target general measures of adaptive function or quality of life as a primary endpoint. For individuals with DS and other groups with cognitive impairments, the definition and subsequent measurement of efficacy may be contingent upon the baseline developmental level. However, this benefit may not be perceived as a benefit for individuals who are not employed or for individuals who have been placed in supportive settings like sheltered workshops. With such individuals, it is conceivable that drug-related improvements in cognitive function may not be noticed in the home and/or work setting. It is also conceivable that drugrelated improvements in cognitive function may decrease the individual’s quality of life and adaptive function in these settings. For example, an individual may become more expressive and assertive while taking the study medication. For school-aged children faced with daily academic challenges, these cognitive improvements may have a profound positive effect on quality of life, ability to perform within the classroom and at home, and with improved self esteem. However, for the adult who is living at home or working in a sheltered workshop setting, these same improvements in cognitive function may lead to increased confrontations with caregivers and co-workers and a breakdown of established social networks. Thus, the whole concept of efficacy and subject selection must be carefully considered in the study design. We have adopted a multiple outcome approach to efficacy analysis in our investigation of cholinergic therapy in an attempt to understand the drug effect on specific cognitive domains. The reason for this approach arose from the preliminary findings by Kishnani et al.  of improved adaptive function in some adults with DS treated with donepezil hydrochloride. The study was based on the hypothesis that acetylcholine plays a critical role in learning, We have adopted a multiple outcome approach to efficacy analysis in our investigation of cholinergic therapy in an attempt to understand the drug effect on specific cognitive domains. The reason for this approach arose from the preliminary findings by Kishnani et al.  of improved adaptive function in some adults with DS treated with donepezil hydrochloride. memory and mood [Nadel, 2003; Pennington et al., 2003] and that individuals with DS have lower levels of acetylcholine [Sacks and Smith, 1989; Florez et al., 1990]. These investigators reasoned that increasing the level of this neurotransmitter in individuals with DS may enhance cognition. Our subsequent work has investigated the effects of cholinergic treatment on multiple cognitive domains using multiple performance measures. While this approach can lead to errors in hypothesis testing, a secondary analysis of the types of measures that consistently approach and/or reach statistical significance can provide interesting insights into the drug effect mechanisms. For example, one can investigate whether improvement in one particular domain, such as attention or memory, appears to define the entire drug effect or whether processes such as associative or auditory processing improve across cognitive domains. Much preliminary work at this level is necessary to define the efficacy parameters before pivotal double-blind placebo-controlled trials are undertaken in the DS population. ARTICLE STUDY MEDICATION There are two important features to the selection of a study medication, which can be a pharmacological agent or a nutritional supplement. First and foremost, the study medication must be safe. Second, there must be some reasonable expectation (scientific rationale) that the medication will have a positive effect on the targeted behavior and/or condition. Safety is a relative concept based on the degree of risk versus anticipated benefit. A higher degree of risk can be tolerated when the benefit is life saving. When the targeted benefit is improved functional ability, the medication risks (side effects) must be relatively minor (i.e., greater than the physical and/or psychological risk ordinarily encountered in daily life or during the performance of routine physical or psychological examinations) and present the prospect of direct benefit to the individual subject [Duke University Medical Center, unpublished]. Due to ethical and liability issues, it is unlikely that a new drug would be directly introduced to a pediatric population, much less a pediatric DS population. However, it is reasonable to consider the off-label use of drugs approved for adult use, particularly when those drugs target cognitive function. When considering nutritional supplements or the off-label use of an adult drug in children, it is important to realize that too much of a good thing may be harmful. Megavitamins in large doses, for example, can be toxic to a child [Litovitz et al., 1994; Barrueto et al., 2005]. The DS condition adds another layer of complexity to the dosing question. Specifically, what is the effect of the extra chromosome and the genes in triplicate on the targeted therapeutic pathway? Animal models of therapeutic pathways, such as the mouse model of DS, may contribute to our understanding of the therapeutic pathway, but animal models in general cannot model the extent of human complexity. In order for an animal model to recapitulate the phenotypic consequences in DS, all interacting genes must be present in ARTICLE AMERICAN JOURNAL OF MEDICAL GENETICS PART C (SEMINARS IN MEDICAL GENETICS): DOI 10.1002/ajmg.c three copies. Segmental trisomy mouse models of DS that include the Ts65Dn, Ts1Cje, and Ts1Rhr [Sago et al., 1998; Costa et al., 1999; Olson et al., 2005], are trisomic for regions of mouse chr16, but not for all genes. Due to the lack of a mouse model with all features of DS, a number of studies need to be done directly in patients. Adequately designed, well-controlled pharmacokinetic (PK) studies of the target population provide critical data for dosing decisions. In the case of children with DS, the PK studies need to be completed with the same-aged children as targeted in the efficacy trials, because developmental changes influence absorption [Yaffe and Aranda, 1992]; distribution [Gilman, 1990]; metabolism [Brown and CampoliRichards, 1989]; excretion [Brown and Campoli-Richards, 1989]; and protein binding [Rane et al., 1971]. PK studies can determine tolerability and blood– serum levels. However, blood–serum levels are not necessarily predictive of medication levels in the brain due to the blood–brain barrier. Positron emission tomography (PET) provides the opportunity to document dose–response according to overall brain uptake and on targeted locations, such as the cerebellum, amygdala and hippocampus; where decreased volume has been considered the neuroatomical basis for functional deficits in individuals with DS [Aylward et al., 1999; Pinter et al., 2001a,b]. Functional MRI (fMRI) offers the potential for the ultimate dose– response analysis, that is, the analysis of the dose versus specific neuropathway function associated with the cognitive tasks under investigation. However, fMRI requires an extended period of interaction while the patient is positioned within a MRI scanner; a challenge that even many adults cannot tolerate. In our investigation of cholinesterase inhibitors (ChEIs) including donepezil hydrochloride [Heller et al., 2004] and rivastigmine tartrate [Heller et al., in press] for treating cognitive deficits in children with DS, we have found that the effective dose for children with DS can vary between subjects and that the relationship between dose and thera- peutic effect is not linear, that is, once an effective level is reached, the higher dose only produces more adverse effects, not higher performance. In our trial of rivastigmine tartrate, a significantly lower dose of drug (compared to recommended adult levels) with a slower titration schedule was better tolerated and produced fewer adverse effects. The availability of a liquid formulation of the study medication (rivastigmine tartrate) helped include children with DS in the study who have difficulty swallowing pills and also provided flexibility in dosing and titration. SUBJECTS In clinical research, the selection of the subject sample directly relates to the trial success and to the generalizability of the results. However, when investigating the pharmacological effects of a study drug on a vulnerable population such as individuals with DS, the generalizability In clinical research, the selection of the subject sample directly relates to the trial success and to the generalizability of the results. However, when investigating the pharmacological effects of a study drug on a vulnerable population such as individuals with DS, the generalizability of results must be secondary to safety. of results must be secondary to safety. Because of the variability in health status, comorbidities, and educational placements in children with DS, it is important to establish strict inclusion and exclusion criteria. To this regard, all subjects recruited into our series of trials investigating the cognitive effects of cholinergic agents have had a documented diagnosis of trisomy 21 and were 191 8 years of age or older, verbal and intelligible (capable of reporting pain or other adverse events) with IQs within the mild to moderate range of mental retardation. This focus on subject safety extends beyond subject selection criteria. In order to be enrolled in research, the subject’s family/caregiver must provide informed consent and be willing to participate actively in the study. This commitment includes a willingness to: (a) attend multiple visits; (b) store and dispense the appropriate daily dose of study medication; (c) clearly communicate any concerns that they may have about the study; (d) attentively observe and report changes in the health and behavior of their child; and (e) maintain open and frequent contact with the study team. This contact is particularly important when side effects occur or dose changes are made. Recruiting subjects and their families with these extensive enrollment criteria introduces bias into the sample. Individuals who are unintelligible, nonverbal, functioning below a moderate level of mental retardation or who are not fluent in English are excluded. Similarly, the requirements for caregiver support selectively include families with sufficient resources to attend multiple sessions and remain committed to their roles of observation and reporting. Recruiting efforts add additional bias. In an attempt to obtain a large and diverse pool of potential subjects, we advertise for subjects through local and regional DS clinics and support groups and through the National Down Syndrome Society. Our published work and presentations at regional and national conferences also serve to elicit family interest in participating in new studies. Through these recruiting efforts, there is a tendency to enroll subjects from families who are more highly educated, more sophisticated regarding patient advocacy, and more highly motivated to anticipate the possibly of change in their children. SAFETY The overarching goal in conducting clinical research in children is to protect 192 AMERICAN JOURNAL OF MEDICAL GENETICS PART C (SEMINARS IN MEDICAL GENETICS): DOI 10.1002/ajmg.c them from harm, but also give them the opportunity to participate in research from which they may benefit [Society for Adolescent Health Research, 2003]. Data establishing the safety of an investigational agent in the adult population cannot be extrapolated to children; therefore, there is a pressing need for studies involving children as research subjects. All aspects of the study design identify subject safety. The subject inclusion and exclusion criteria identify subjects capable of reporting pain or poor health, eliminate subjects at particular risk due a compromised medical condition, and reduce and/or eliminate potential drug interactions due to concomitant medications. Dose and titration schedules are designed to minimize adverse effects. In addition, regular clinic visits, caregiver training for adverse event reporting and an extensive monitoring schedule minimize the impact of adverse events should they occur. However, despite careful inclusion criteria, adverse events will occur. It is essential that all potential adverse events are recorded and considered in terms of the subject’s continued role in the study. The adverse event analysis is critical to determining the relative safety of the experimental medication. Capturing adverse events in an objective way is a difficult task. In regard to study design, objective thresholds for adverse events should be determined prior to conducting the study. An adverse event may be identified by abnormal laboratory parameters, symptoms showing toxicity, or significant changes in health or behavior. A serious adverse event is one that is life-threatening, causes a congenital defect, or that leads to hospitalization or a prolongation of hospitalization. Whenever a change in dose is made, careful monitoring of the subject must occur, especially following the initiation of the first dose. Expected side effects, based on the available study medication package insert, should be systematically reviewed with the caregiver and included in the safety data analysis for each visit. It is also important to complete a thorough review of systems to capture unexpected side effects that may or may not be related to the study drug. In addition to a systems review at study visits, adverse events must be captured also between study visits. There are standard approaches to collecting and analyzing adverse events, examples include the Adverse Drug Reaction probability scale [Naranjo et al., 1981] and the Common Terminology Criteria for Adverse Events version 3.0 [Trotti et al., 2003]. Individual institutional review boards also have rules as to how adverse events must be reported. The adverse event analysis completed in our research involves documenting each adverse event and then rating that event in terms of severity (mild, moderate, or severe), expected or unexpected and related or unrelated to the study. Our safety analyses include a detailed investigation of severe adverse events (should they occur) and the reason why subjects drop out of studies. Mild and moderate adverse events are documented and reported. Comparisons of adverse events across drug and placebo conditions are completed in double-blind, placebocontrolled studies. EFFICACY Efficacy, the critical element of a clinical trial that determines whether or not the study medication is beneficial, has two key aspects. First of all, does the experi- Efficacy, the critical element of a clinical trial that determines whether or not the study medication is beneficial, has two key aspects. First of all, does the experimental drug have an effect on the intended outcome? And second, if the experimental medication does have an effect, is this effect clinically relevant? ARTICLE mental drug have an effect on the intended outcome? And second, if the experimental medication does have an effect, is this effect clinically relevant? In terms of the study of cognitive function in individuals with DS, these issues boil down to: (a) whether the study medication improves cognitive function in some measurable way, and (b) whether or not this improvement in cognitive function will improve the individual’s quality of life. On the surface, these issues seem quite straightforward, but in practice they are not. In fact, a given study medication may indeed have an effect on cognitive function, but unless this effect can be measured, the medication effect will never be known. In order to demonstrate change across a group of subjects, the medication must have the capacity to effect change, the subjects must have the ability to demonstrate change, and the investigators must have the capability of measuring the change through structured interactions with the subjects and/ or the subject’s caregivers. An ongoing task of the Duke DS Research Team has been to assemble a test battery that is appropriate for the subjects’ developmental skill levels and sensitive to the effects of cholinesterase inhibitors. In order to minimize psychometric testing issues such as content validity, construct validity, concurrent validity, and test–retest reliability, standardized test measures are used to a large extent. It is important to note that most measures of cognitive function have been standardized on the general population. Age-matched individuals with DS fall on the extreme end of the performance scale on these tests, typically below the first percentile. The use of age appropriate scales for individuals with DS in clinical trials is not particularly useful because these scales are generally insensitive to performance change in the DS population. Rather than using a test battery composed of age appropriate scales, developmentally appropriate scales are used. Typically these tests have more items at or near the subject’s functional level and thus are more sensitive to change. Unfortunately, the standardized scores, adjusted for the age of the typically developing ARTICLE AMERICAN JOURNAL OF MEDICAL GENETICS PART C (SEMINARS IN MEDICAL GENETICS): DOI 10.1002/ajmg.c individual upon whom the test was normed, are most often not appropriate for individuals with DS. An alternative approach to measuring performance change is the use of raw scores (the total number of items correct) rather than standard scores. With the use of raw scores, the investigator links performance change over time to specific test items. This can provide additional insight into the study medication effect. However, raw scores (unlike standard scores) do not factor some psychometric test error, such as test– retest reliability, into the analysis. Therefore, when using raw scores in the efficacy analysis, it is particularly important to select test measures that are well designed, that is, contain minimal test error, and are extremely sensitive to the developmental level of the subject. In that way, the impact of psychometric test error on performance change is minimized. Additionally, the careful selection of tests that are highly sensitive to performance change in individuals with DS can eliminate test ceiling and floor effects which can cause an investigator to conclude that there is no medication effect when one really exists (Type II error) [Heller et al., in press]. When working with children with DS, it is also important to understand that while the performance measurement system may be exquisitely designed, the quality of the analysis is contingent upon the quality of the data. In this case, the old adage of ‘‘garbage in garbage out’’ holds true. If the subject is not motivated to provide his/her optimal performance at any given session, then the results of that session are compromised. Because determinations of medication effects are based on performance change across sessions, it is critical that these performance changes are not contaminated by subject motivational factors. It is an important and substantial challenge for a research team to elicit optimal performance from individuals with DS at each test session. Based on the experience of the Duke DS Research Team, there are a number of factors necessary to optimize the performance of individuals with DS at each test session (Table II). To this point, we have discussed the importance of establishing a test battery that is sensitive to performance change in individuals with DS and the importance of obtaining optimal performance from the subjects at each session to accurately quantify the effect of the medication on performance. However, we have yet to define cognitive function and how it should be measured. Cognitive function is a broad term encompassing multiple functional domains. The domains of interest in our studies of cholinesterase inhibitors are: Adaptive Function, Language (expressive and receptive), Attention, Memory, and Associative Processing. Within each domain, we have included several different measures or subtests in an effort to capture and better understand a potential medication effect. The inclusion of multiple measures also helps address the 193 obvious complexity inherit in each domain, such as the different aspects of language (receptive versus expressive) or memory (visual versus auditory, short-term versus long-term). As discussed in the Study Design section above, there is a trade-off between investigating the potential medication effect across multiple cognitive domains and the statistical power to demonstrate that the experimental medication has a statistically significant effect on the functional capability of individuals with DS. We would argue that in the current exploratory stage of research that it is is important to consider the broad potential effect of the study medication We would argue in the current exploratory stage of the research that it is important to consider the broad potential effect of the study medication in order to identify the possible mechanisms of action and to better understand the potential of the study medication for use in individuals with DS. in order to identify the possible mechanisms of action and to better understand the potential of the study medication for TABLE II. Factors Important for Eliciting Optimal Performance From Individuals With DS 1. A clinically trained research team with substantial experience in testing children and adults with cognitive disabilities. 2. Establishing a positive relationship so that the individual looks forward to additional visits. In this regard, we have learned to design a screening session into our trials where no efficacy data are collected. The purpose of this visit is to provide a second review to ensure that the subject and his/her family meet the inclusion criteria and are fully informed about the study and to establish the rapport between the examiners and the subject. The screening visit is particularly useful in reducing the anxiety level of the subject and the subject’s family. 3. Eliminate all distracters from the test environment, including interruptions, noise, and observers of the performance. 4. Insert as few changes as possible across sessions, that is, maintain the same examiner, maintain order of testing, test in the same room in the same orientation with the same decorating scheme, provide the same auxiliary services, that is, snack and/or lunch at the same time each session. 5. Communicate a positive expectation of optimum performance from the very start of the study and maintain that expectation throughout. 194 AMERICAN JOURNAL OF MEDICAL GENETICS PART C (SEMINARS IN MEDICAL GENETICS): DOI 10.1002/ajmg.c use in individuals with DS. However, it is also important not to lose sight of the goal of clinical relevance, because ultimately, clinical relevance will determine whether or not a particular medication is used. For example, in our research, we are eager to demonstrate that a cholinesterase inhibitor, such as donepezil hydrochloride or rivastigmine tartrate, improves the expressive language ability of individuals with DS. However, unless we can demonstrate how an increase in language function of this magnitude improves the quality of life for individuals with DS, it is unlikely that the clinical use of this medication with individuals other than the study subjects would be endorsed. To this end, we include multiple measures of overall function (i.e., adaptive function) as well as measures of specific cognitive domains in our research. We are particularly excited when study results demonstrate performance improvement in particular cognitive domains and simultaneous improvement in adaptive function. This type of result not only strengthens the hypothesis that cholinergic therapy can improve aspects of cognitive function in children with DS; it also suggests a potential mechanism for this result. Another challenge is the potential impact of floor and/or ceiling effects in the measures comprising a domain. At the very least, the investigators must be mindful of the subjects’ estimated mental age in choosing their measures. Including tasks that are too difficult can often lead to behavior problems, such as increased frustration and refusals to continue working. Tasks which are too easy for the majority of the subjects may eliminate the possibility of observing a drug effect. SUMMARY The issues of clinical trials for the enhancement of cognitive function in children with DS are challenging and complex, but they are not insurmountable. As with the development of leukemia research in DS, research successes are incremental and are achieved through the collaborative efforts of multiple stakeholders. Funding for clinical research in cognitive function in DS will increase when families, caregivers, The issues of clinical trials for the enhancement of cognitive function in children with DS are challenging and complex, but they are not insurmountable. As with the development of leukemia research in DS, research successes are incremental and are achieved through the collaborative efforts of multiple stakeholders. and individuals with DS aggressively seek private and public support for answers to optimizing functional ability and quality of life. New therapy alternatives will arise as private industry tackles comparable issues in conditions such as Alzheimer disease, stroke, and head injury. Similarly, new research ideas will evolve as the research community comes together to tackle these issues from many different perspectives. We hope that this discussion will raise awareness of the need for clinical research in the cognitive function of children with DS and the value of experienced multidisciplinary research teams who design and manage these types of trials. REFERENCES Alderman MH. 2005. The limitations of transferring evidence from clinical trials to the care of individual patients. High Blood Press Cardiovasc Prev 12:5–7. Aylward EH, Li Q, Honeycutt NA, Warren AC, Pulsifer MB, Barta PE, Chan MD, Smith PD, Jerram M, Pearlson GD. 1999. MRI volumes of the hippocampus and amygdala in adults with Down’s syndrome with and without dementia. Am J Psychiatry 156: 564–568. Baker ET, Wang MC, Walberg HJ. 1994. The effects of inclusion on learning. Educ Leadersh 52:33–35. ARTICLE Barrueto F Jr, Wang-Flores HH, Howland MA, Hoffman RS, Nelson LS. 2005. Acute vitamin D intoxication in a child. Pediatrics 116:e453–e456. Bennett FC, McClelland S, Kriegsmann EA, Andrus LB, Sells CJ. 1983. Vitamin and mineral supplementation in Down’s syndrome. Pediatrics 72:707–713. Berg JM, Kirman BH, Stern J. 1961. Treatment of mongolism with pituitary extract. J Ment Sci 107:475–480. Bidder RT, Gray P, Newcombe RG, Evans BK, Hughes M. 1989. The effects of multivitamins and minerals on children with Down syndrome. Dev Med Child Neurol 31:532–537. Brown RD, Campoli-Richards DM. 1989. Antimicrobial therapy in neonates, infants and children. Clin Pharmacokinet 17(Suppl 1): 105–115. Bumbalo TS, Morelewicz HV, Berens DL. 1964. Treatment Of Down’s syndrome with The ‘‘U’’ series of drugs. JAMA 187:361. Coleman M, Sobel S, Bhagavan HN, Coursin D, Marquardt A, Guay M, Hunt C. 1985. A double blind study of vitamin B6 in Down’s syndrome infants. Part 1—Clinical and biochemical results. J Ment Defic Res 29(Pt 3):233–240. Costa AC, Walsh K, Davisson MT. 1999. Motor dysfunction in a mouse model for Down syndrome. Physiol Behav 68:211–220. Duke University Medical Center. Pediatric Risk Assessment Form. Unpublished document. Eisenberg J, Hamburger-Bar R, Belmaker RH. 1984. The effect of vasopressin treatment on learning in Down’s syndrome. J Neural Transm 60:143–147. FDA modernization act of 1997. 1998. Guidance on medical device tracking; availability— FDA. Notice. Fed Regist 63:10640–10641. Florez J, del Arco C, Gonzalez A, Pascual J, Pazos A. 1990. Autoradiographic studies of neurotransmitter receptors in the brain of newborn infants with Down syndrome. Am J Med Genet 7:301–305. Gamis AS. 2005. Acute myeloid leukemia and Down syndrome evolution of modern therapy—State of the art review. Pediatr Blood Cancer 44:13–20. Gilman JT. 1990. Therapeutic drug monitoring in the neonate and paediatric age group. Problems and clinical pharmacokinetic implications. Clin Pharmacokinet 19:1–10. Harrell RF, Capp RH, Davis DR, Peerless J, Ravitz LR. 1981. Can nutritional supplements help mentally retarded children? An exploratory study. Proc Natl Acad Sci USA 78:574–578. Hassold TJ, Patterson D, editors. 1998. Down syndrome: A promising future, together. New York: Wiley-Liss. Heaton-Ward WA. 1962. Inference and suggestion in a clinical trial (Niamid in Mongolism). Br J Psych 108:865–870. Heller JH, Spiridigliozzi GA, Crissman BG, Sullivan JA, Eells R, Li JS, Doraiswamy PM, Krishnan KR, Kishnani PS. 2006. Safety and efficacy of rivastigmine in adolescents with Down syndrome: A preliminary 20-week open-label study. JCAP (in press). Heller JH, Spiridigliozzi GA, Sullivan JA, Doraiswamy PM, Krishnan RR, Kishnani PS. ARTICLE AMERICAN JOURNAL OF MEDICAL GENETICS PART C (SEMINARS IN MEDICAL GENETICS): DOI 10.1002/ajmg.c 2003. Donepezil for the treatment of language deficits in adults with Down syndrome: A preliminary 24-week open trial. Am J Med Genet Part A 116A:111– 116. Heller JH, Spiridigliozzi GA, Doraiswamy PM, Sullivan JA, Crissman BG, Kishnani PS. 2004. Donepezil effects on language in children with Down syndrome: Results of the first 22-week pilot clinical trial. Am J Med Genet Part A 130A:325–326. Hollowood TM, Salisbury CL, Rainforth B, Palombaro MM. 1995. Use of instructional time in classrooms serving students with and without severe disabilities. Except Child 61:242–253. Jones M. 1996. Down’s syndrome. J Audiov Media Med 19:178. Katz N. 2005. Methodological issues in clinical trials of opioids for chronic pain. Neurology 65(12 Suppl 4):S32–S49. Katzman DK. 2003. Guidelines for adolescent health research. J Adolesc Health 33:410– 415. Kishnani PS, Sullivan JA, Walter BK, Spiridigliozzi GA, Doraiswamy PM, Krishnan KR. 1999. Cholinergic therapy for Down’s syndrome. Lancet 353:1064–1065. Kishnani PS, Spiridigliozzi GA, Heller JH, Sullivan JA, Doraiswamy PM, Krishnan KR. 2001. Donepezil for Down’s syndrome. Am J Psychiatry 158:143. Litovitz TL, Clark LR, Soloway RA. 1994. 1993 annual report of the American Association of Poison Control Centers Toxic Exposure Surveillance System. Am J Emerg Med 12: 546–584. Lobaugh NJ, Karaskov V, Rombough V, Rovet J, Bryson S, Greenbaum R, Haslam RH, Koren G. 2001. Piracetam therapy does not enhance cognitive functioning in chil- dren with down syndrome. Arch Pediatr Adolesc Med 155:442–448. Lonsdale D, Kissling CD. 1986. Clinical trials with thiamine, tetrahydrofurfuryl disulfide (TTFD) in Down syndrome. J Orthomolec Med 1:169–175. Nadel L. 2003. Down’s syndrome: A genetic disorder in biobehavioral perspective. Genes Brain Behav 2:156–166. Naranjo CA, Busto U, Sellers EM, Sandor P, Ruiz I, Roberts EA, Janecek E, Domecq C, Greenblatt DJ. 1981. A method for estimating the probability of adverse drug reactions. Clin Pharmacol Ther 30:239–245. Olson LE, Tien J, South S, Reeves RH. 2005. Long-range chromosomal engineering is more efficient in vitro than in vivo. Transgenic Res 14:325–332. Pajonk FG. 2005. Clinical trial design in schizophrenia: Implications for clinical decisions. Curr Opin Psychiatry 18:692–699. Pennington BF, Moon J, Edgin J, Stedron J, Nadel L. 2003. The neuropsychology of Down syndrome: Evidence for hippocampal dysfunction. Child Dev 74:75–93. Pinter JD, Brown WE, Eliez S, Schmitt JE, Capone GT, Reiss AL. 2001a. Amygdala and hippocampal volumes in children with Down syndrome: A high-resolution MRI study. Neurology 56:972–974. Pinter JD, Eliez S, Schmitt JE, Capone GT, Reiss AL. 2001b. Neuroanatomy of Down’s syndrome: A high-resolution MRI study. Am J Psychiatry 158:1659–1665. Pueschel SM, Reed RB, Cronk CE, Goldstein BI. 1980. 5-hydroxytryptophan and pyridoxine. Their effects in young children with Down’s syndrome. Am J Dis Child 134: 838–844. Rane A, Lunde PK, Jalling B, Yaffe SJ, Sjoqvist F. 1971. Plasma protein binding of diphenyl- 195 hydantoin in normal and hyperbilirubinemic infants. J Pediatr 78:877–882. Sacks B, Smith S. 1989. People with Down’s syndrome can be distinguished on the basis of cholinergic dysfunction. J Neurol Neurosurg Psychiatry 52:1294–1295. Sago H, Carlson EJ, Smith DJ, Kilbridge J, Rubin EM, Mobley WC, Epstein CJ, Huang TT. 1998. Ts1Cje, a partial trisomy 16 mouse model for Down syndrome, exhibits learning and behavioral abnormalities. Proc Natl Acad Sci USA 95:6256–6261. Salman M. 2002. Systematic review of the effect of therapeutic dietary supplements and drugs on cognitive function in subjects with Down syndrome. Eur J Paediatr Neurol 6:213–219. Smith GF, Spiker D, Peterson CP, Cicchetti D, Justine P. 1984. Use of megadoses of vitamins with minerals in Down syndrome. J Pediatr 105:228–234. Trotti A, Colevas AD, Setser A, Rusch V, Jaques D, Budach V, Langer C, Murphy B, Cumberlin R, Coleman CN, et al. 2003. CTCAE v3.0: Development of a comprehensive grading system for the adverse effects of cancer treatment. Semin Radiat Oncol 13:176–181. Weathers C. 1983. Effects of nutritional supplementation on IQ and certain other variables associated with Down syndrome. Am J Ment Defic 88:214–217. Yaffe SJ, Aranda JV. 1992. Introduction and historical perspectives. In: Yaffe SJ, Aranda JV, editors. Pediatric pharmacology, therapeutic principles in practice, 2nd edition. Philadelphia, PA: Saunders. p 3–9. Yang Q, Rasmussen SA, Friedman JM. 2002. Mortality associated with Down’s syndrome in the USA from 1983 to 1997: A population-based study. Lancet 359:1019–1025.