DRUG DISPOSITION
Clin. Pharmacokinel. 23 (I): 10-29, 1992 0312-5963/92/0007-0010/$10.00/0 © Adis International Limited. All rights reserved . CPKl
Pharmacokinetic Studies in Paediatric Patients Clinical and Ethical Considerations Ralph E , Kauffman and Gregory L. Kearns Departments of Pediatrics and Pharmacology, Wayne State University School of Medicine, Children's Hospital of Michigan, Detroit, Michigan, and Section on Pediatric Clinical Pharmacology, Department of Pediatrics, University of Arkansas for Medical Sciences, Arkansas Children's Hospital, Little Rock, Arkansas, USA
Contents 10 12 12 15 19 22 22 23 15 26
Summary
Summary I, Practical and Theoretical Considerations 1.1 The Approach to the Patient/Subject 1.2 Methods and Techniques for Sample Collection 1.3 Methods for Data Analysis: Analytical and Pharmacokinetic Considerations 2. Ethical Considerations 2.1 History of Protection of Paediatric Research Subjects 2.2 Ethical Considerations in Drug Studies in Children 2.3 Special Populations 3. Conclusions
Important advances in paediatric clinical pharmacology have been made over the past 2 decades. However, there remains a reluctance to pursue pharmacodynamic and pharmacokinetic studies in children and, consequently, many important therapeutic agents have not been adequately studied in this population. Age-related pharmacokinetic/pharmacodynamic studies are not only essential to provide optimal drug therapy for children, but are quite feasible. Usually, paediatric pharmacokinetic studies are conducted in children receiving treatment for a specific medical condition. The approach to soliciting participation of paediatric subjects requires special sensitivity to the fears and anxieties of the child and the parents. Factors influencing subject enrolment and suggestions to enhance enrolment into study protocols are discussed. Pharmacokinetic/pharmacodynamic studies require repeated measurements over time and often entail obtaining multiple blood and urine samples. Techniques for reducing sample volume and number of necessary samples while minimising the discomfort and fear associated with obtaining multiple samples include the development of highly sensitive analytical methods to measure drug concentrations in small volume samples. The number of samples obtained from individual subjects can be minimised by using pharmacokinetic analytical approaches such as the nonlinear mixed effect model (NONMEM) which allows estimation of pharmacokinetic characteristics of a population using limited data from each subject. In addition, less invasive methods to measure drug metabolism/elimination such as salivary sampling, transcutaneous collection and breath analysis have been applied to the study of certain drugs, Children are a particularly vulnerable population because of their limited cognitive abilities
Pharmacokinetic Studies in Paediatric Patients
II
and dependence on adults. Thus, they must be afforded greater protection from exploitation as research subjects than that provided to adults.
The passage of the 1962 Kefauver-Harris amendments to the Pure Food and Drug Act in the United States provided a clear mandate for improving the safety and efficacy of new drugs for all patients, including paediatric patients. A decade after the amendments were passed, the American Academy of Pediatrics (AAP) Committee on Drugs (COD) published guidelines for the types of studies needed for approval of drugs for use by children (AAPCOD 1974). These guidelines provided scientific guidance and recommendations to all individuals [i.e. scientists and clinicians in private practice, academia, the Food and Drug Administration (FDA) and the pharmaceutical industry) concerned with any facet of therapeutic drug evaluation in infants and children. The AAP report contained 3 statements which are as relevant to paediatric clinical pharmacology today as they were almost 2 decades ago: first, that pharmacokinetic studies appropriate to the drug and its intended use are essential for (children in) each age group; secondly, that clinical investigations in paediatric patients are ethically defensible; and thirdly, that scientifically valid studies can be performed in paediatric patients. As a direct result of the technological advances which have occurred since the AAP report first appeared, important advances in paediatric clinical pharmacology have been made as reflected by the ever growing data base on many commonly used drugs (Besunder et al. 1988a,b; Boreus 1982; Gilman 1990; Green & Mirkin 1984; Kauffman 1985; Kearns & Reed 1989; MacLeod 1985; Milsap & Szefler 1986; Morselli 1983; Morselli et al. 1980; Prandota 1985; Rane 1980, 1984; Rane & Wilson 1976; Roberts 1984; Stewart & Hampton 1987; Triggs et al. 1984; Wilkinson 1978; Wilson 1972). Although these important strides have partially removed paediatric patients from their status as 'therapeutic orphans' (Kearns & Reed 1989; Shir-
key 1968), there remains a reluctance/resistance by the pharmaceutical industry and many clinical investigators to pursue pharmacokinetic and pharmacodynamic studies in infants and children. This aversion is fuelled in many instances by concerns regarding ethical constraints, a lack of awareness of the impact of development on drug disposition, the relatively small contribution of the paediatric age group to the profits from many drugs and products, the apparent lack of perception of need for these studies and a general lack of societal acceptance of performing clinical investigations in children. Additionally, in many cases, there is a basic unfamiliarity with the requirements and techniques for conducting pharmacokinetic/pharmacodynamic investigations in paediatric patients. The purpose of this review is to raise the level of awareness of the need for age-related evaluations of drugs and to demonstrate that such studies are not only necessary but quite feasible. Age-related pharmacokinetic/pharmacodynamic studies are essential to provide the data base from which optimal age-related doses of drugs may be derived. It is our hope that this discussion will provide an impetus for pharmaceutical company sponsors, responsible regulatory agencies and investigators to increase their respective efforts to ensure that drugs which are used in children are adequately studied in children. Only when this goal is achieved can children and their parents be assured that the drugs they are receiving are being administered under optimally safe and effective conditions. We first review some of the practical and theoretical aspects of performing clinical pharmacokinetic studies in infants and children. These include techniques for subject recruitment and retention of patients in the study, methods of sample collection, and analytical and pharmacokinetic techniques which are well suited for pharmacokinetic studies in infants and children. We then pre-
c/in. Pharmacokinet. 23 (I) 1992
12
Table I. Factors affecting patient recruitment (adapted from Spilker 1991) The nature of the appeal to participate The patient's/parent's perception of direct vs indirect benefits The specific requirements associated with the study (i.e. degree of difficulty, discomfort or compromised modesty) The degree of concern that parents/patients have regarding the disease or reason for hospitalisation The location of the study site relative to the patient's home Resources available to the investigator for recruitment The socioeconomic mix of the potential patient pool The place where recruitment occurs (i.e. home vs school vs hospital environment) The methods used for subject recruitment (i.e. public vs institutional advertising, physician referral, etc.)
sent the salient ethical issues to be considered when conducting clinical trials in children. Included are issues of informed consent, risk/benefit assessment and special protection of particularly vulnerable populations of children.
1. Practical and Theoretical Considerations 1.1 The Approach to the Patient/Subject In most instances, clinical pharmacokinetic studies in infants and children are not conducted in normal, healthy volunteers as is the general rule with many investigations performed in adults. Consequently, the 'subject' in paediatric studies is most often a patient who is being treated for a specific medical indication which may not be related to the study being performed. Since paediatric subjects are usually ill children, the approach used to solicit the participation of a paediatric patient in a pharmacokinetic study and the approval of their parents or guardians requires special sensitivity (Grodin & Alpert 1988) and should be in full conformity with specific national regulations governing research on minors (e.g. 45 CFR 46.401-46.409, 1983).
1.1.1 Factors Influencing Recruitment of Research Subjects Factors which may influence the recruitment of paediatric patients as research subjects are summarised in table I. The nature of the request for the patient and their parents/guardians to participate in a study may represent the most important determinant for the success or failure of recruitment efforts in a clinical pharmacokinetic investigation (Spilker 1991). Two factors, however, frequently dissuade the investigator or parent from freely entering into a trial. First, in adult clinical trials, investigations to determine the pharmacokinetic properties of a drug are generally conducted in phase I trials and may be viewed as phase I-type trials regardless of when they are conducted during the development of a drug (Spilker 1991). Accordingly, little direct patient benefit can be expected and/or promised, especially when initial single dose studies are performed. Secondly, for clinical pharmacokinetic studies to yield the type of information which the US FDA has deemed desirable (Peck & Collins 1990), it is often necessary that multiple biological specimens (e.g. blood, urine, saliva) be obtained (Svensson 1989). Frequently, these procedures may produce added pain and discomfort (e.g. the insertion of an indwelling cannula, multiple fingertip punctures) or may compromise modesty in the view of the patient or their parents (e.g. the supervised collection of multiple urine samples). Thus, clinical pharmacokinetic investigations in paediatric patients may be perceived as falling into a category of research involving children described by Rozovsky (1984) as entailing more than minimal risk to the patient and offering little or no direct patient benefit but only the prospect of 'generalisable knowledge'. When discussing this type of study with the patient and their parents/guardians they should be clearly informed that although the patient may not derive direct therapeutic benefit by participating in a clinical pharmacokinetic study, the new knowledge obtained as a direct result of their participation will not only make a unique contribution to understanding the pharmacology of a new drug in children but may also benefit other children who might
Pharmacokinetic Studies in Paediatric Patients
receive the drug in future evaluations of the safety and/or efficacy of the drug. Alternatively, in some instances, a pharmacokinetic study may be conducted as part of an efficacy study in which the child may directly benefit from receiving the drug, although the benefit is not directly related to the pharmacokinetic part of the protocol. In our experience, when the investigator clearly and concisely relates the scientific goals and specific aims of the particular investigation, outlines the required procedures and the safeguards to minimise both risk and discomfort to the patient, and educates the patient/parents regarding the scientific implications and importance of the information to be obtained, the consent process becomes truly 'informed'. Most importantly, the patient and their parents have a proper frame of reference on which to make a decision as to whether the potential benefits of participating in the study outweigh the perceived risks. Several additional practical issues which are of equal importance with regard to the appeal made for patient recruitment merit discussion. As expected, the hospitalisation of an infant or child usually produces a heightened level of anxiety and concern on behalf of the patient and their family. Consequently, an overwhelming desire which often permeates this scenario is that only those procedures which are 'medically necessary' be performed, that unnecessary discomfort be avoided and/or minimised and that no unnecessary intervention be introduced which might alter the planned course of treatment and/or prolong the period of hospitalisation. It is therefore of paramount importance that the investigator both understands and respects these desires and expectations which demand that the planned course of medical treatment for a given disease/condition not be altered or adversely affected by the subject's participation in any clinical investigation. In our view, this objective can be facilitated and achieved by the following actions. First, the investigator should be thordughly familiar with the patient's condition, carefully reviewing the medical record and discussing the patient with the primary care physician before mentioning the investigation to the patient
13
or their parents. Secondly, once a potential subject has been identified and the support of the patient's primary physician is obtained, the investigator should describe the study protocol to the patient and their parents and should carefully and completely review the scope, study procedures and expected outcome/relevance of the experimental findings. Thirdly, the investigator should demonstrate to the patient and their parents how the required study procedures will not alter the planned course of medical treatment and note the safeguards (i.e. monitoring procedures, 'rescue criteria', etc.) in place to insure that all risks associated with subject participation are truly minimal. Fourthly, the investigator must clearly relate that the ultimate decision and authority in allowing the investigation to begin and to continue resides with the patient and their parents. Lastly, and most importantly, the investigator needs to communicate clearly that his/her most important concern and that of his/her associates resides not with the outcome of the investigation but with the health, safety and welfare of the patient. Although this approach requires a considerable investment of time and resources on behalf of the investigator, it embraces responsibilities which should not be abdicated and, if performed properly, can result in a high rate of success in patient recruitment and retention. Thus, the success of recruitment is directly related to the degree of rapport the investigator has with the patient and parents. This is supported by our experience over the past 5 years which reflects successful patient recruitment (i.e. by signed consent form) in over 85% of all encounters for patients between the ages of 6 months and adolescence in 15 separate pharmacokinetic studies (Kearns et aI., unpublished data). 1.1.2 Techniques For Increasing Patient Enrolment There are several techniques which investigators can use to increase patient enrolment in clinical investigations (Spilker 1991). Those techniques which may be relevant to paediatric pharrnacokinetic studies are summarised in table II. As mentioned previously, the most important considera-
14
Table II. Techniques to increase patient enrolment (adapted from Spilker 1991) Assess patient motivation to enter study and specifically address patient/parental needs Use study procedures to broaden base of medical care by making the results from testing required by the protocol available to the parent. the primary physician and the medical record Enlist the services of a patient recruiter (e.g. nurse, pharmacist, lay person) Increase the level of patient benefits by providing financial resources to compensate the patient or their parent(s) for basic necessities (e.g. transportation, meals, child care for siblings of patient, etc.) Conduct relevant chart/record reviews (e.g. pharmacy or laboratory records, patient medical records) to determine total available patient pool Use targeted professional advertisements (i.e. colleagues, house officers, nurses) As far as possible (i.e. without compromising the safety of the study and/or the quality of the science), loosen the demands of the protocol (Le. eliminate some disagreeable tests/ procedures or modify inclusion/exclusion criteria) Expand the number of study sites (e.g. additional academic centres, use of academic private practice settings, etc.) Provide relevant communication (e.g. an 'outline' of the study and required procedures) and feedback to professional staff (i.e. nurses, physicians) who may assist with recruitment and/ or the study protocol per se Provide information regarding the status of the patient and/or relevant tests results to the primary care physician(s) Target research focus to subpopulations which historically are accessible and/or who have had previous good experiences with participating in clinical investigations
tions for increasing subject enrolment are to assess patient and parental motivation, completely discuss the experimental protocol and specifically address the expressed and perceived needs (or questions) of the patient and/or their parents. Other selected techniques for increasing patient enrolment (table II) which have proven to be successful in our experience and that of other investigators merit further consideration and discussion. Enlisting the services of an individual to specifically notify the investigator of the potential el-
Clin. Pharmacokinet. 23 (1) 1992
igibility and/or desire of a patient to participate in a clinical investigation may serve as an effective tool for enhancing recruitment. This activity may take several forms ranging from the hiring of a 'patient recruiter' to the use of professional medical staff(i.e. nurses, house officers, staff physicians) with general knowledge of the inclusion/exclusion criteria who agree to notify the investigator of potential patients. The patient recruiter may be a lay person whose primary responsibility is to identify potential study candidates and to ascertain from the patient and their parents whether they would be interested in speaking to the investigator about a given clinical investigation (Walson, personal communication). When medical residents (i.e. house officers) are used to identify potential study patients, it is often customary to offer some type of tangible incentive or 'finder's fee' (e.g. financial remuneration, educational stipends, textbooks, etc.). However, this practice has been condemned because of the potential for a conflict of interest or introduction of coercion into the recruitment process (Lind 1990). On the other hand, targeted professional education of nursing and medical staff regarding a specific area of research or clinical study can be very effective in increasing patient enrolment. This latter endeavour can produce a heightened awareness of a particular area of investigation and reinforce a spirit of cooperation and collaboration between the investigator and workers providing ongoing medical care to the study subject. Another successful technique for enhancing patient enrolment is to optimise the benefits associated with subject participation. As summarised in table II, this could include the provision ofmonetary resources to compensate the patient's parents for necessary expenses related to subsistence and transportation during the study period and expanding the patient's records by including the results of tests required by the study protocol [e.g. renal and hepatic function profiles, glucose-6-phosphate dehydrogenase (G6PD) screening, ECG, radiographs, etc.]. Although these tests may not be medically indicated for the patient's underlying condition, their costs are covered by resources sup-
Pharmacokinetic Studies in Paediatric Patients
porting a given study. This additional information may assist the primary care physician in the overall assessment of the patient. This is illustrated by our recent experience where G6PD deficiency was diagnosed in 2 children from prestudy laboratory data required for a single-dose pharmacokinetic study of a new, oral {3-lactam antibiotic (Kearns, unpublished data). It is important, however, that compensation or benefits not be of a nature or extent that they constitute a coercive influence on the patient or parents to participate in the study (see below). A final consideration of note for increasing patient enrolment resides with the site used for patient recruitment and study. If the patient must be transferred from a hospital room or an outpatient treatment area to a clinical research facility, care should be taken to insure that the environment is appropriately equipped and staffed for providing necessary medical care, is conducive to the needs and desires of the patient (e.g. a bathroom within the study room, contains materials to entertain/comfort the patient, has sufficient space to afford normal interaction between the patient and family members, etc.) and can accommodate the personnel and equipment necessary for conducting the study without producing a significant intrusion to the routine care of the patient. For clinical pharmacokinetic studies that do not involve elaborate equipment or labour intensive pharmacodynamic assessments, it is often possible to successfully perform the study in the patient's hospital room (Kearns et al. 1988). Additionally, the hospital room and nursing staff may provide a measure of'institutional familiarity' which is often times conducive to patient recruitment. This is especially true when the support staff are well educated about the study and are supportive of the work. Finally, certain types of investigations may allow the recruitment and enrolment of healthy children from environments which are characteristically not associated with clinical research. This has been recently illustrated by Relling et al. (1991) who conducted a pharmacogenetic study of debrisoquin oxidation in over 500 children who were
15
recruited from private elementary and secondary schools. 1.2 Methods and Techniques for Sample Collection To accomplish the goals of a pharmacokinetic study, it is necessary to obtain repeated observations of drug concentrations in a biological fluid which presumably reflects the concentration vs time profile (AUC) for the drug or its active metabolite at the site of action. Traditionally, the preferred and most accessible fluids for obtaining repeated assessments of drug concentration have been blood and urine (Kearns & Trang 1986; Matzke & St. Peter 1990; Svensson 1989). However, in paediatric patients, ethical, physiological and sometimes practical limitations may influence the availability of biological samples required for pharmacokinetic studies. There are several major considerations regarding sampling in paediatric patients and potential alternatives to the use of blood and urine specimens.
1.2.1 Repeated Blood Sampling Since most pharmacokinetic investigations are conducted by obtaining repeated blood samples (Svensson 1989), an invasive experimental procedure is introduced (i.e. venepuncture) which may carry risks that have been determined in some cases to be more than minimal (Janofsky & Starfield 1981). The risks associated with repeated venous sampling and/or venepuncture may include excessive blood loss (e.g. when the total sampling volume obtained over a 24h period exceeds 10% of the estimated circulating blood volume), pain, bruising, infection and possibly the anterior intraosseous syndrome (Saeed & Gatens 1983). When repeated capillary blood samples are obtained in infants by either fingertip or heel puncture (Kearns et al. 1985; Wilson et al. 1982), the potential exists for producing pain, increased anxiety (associated with restraining the infant while the blood sample is 'milked' from the extremity), bruising, infection, bony injury and, importantly, potential errors in drug measurement. Murphy and Ward (1991) re-
16
cently showed that discrepancies between drug concentrations in serum derived from capillary vs venous blood may result from contamination of the blood specimen with extracellular fluid and metabolic differences between central vs peripheral blood which might influence drug equilibrium. In our collective experience in evaluating the pharmacokinetics of antimicrobial agents (Kauffman et al. 1981; Kearns et al. 1982, 1989), and most recently midazolam (Tolia et al. 1991; Wells et al. 1991), we have found that indwelling, silastic venous cannulas or heparinised small gauge needles markedly reduce the discomfort associated with repeated venous blood sampling and, as stated by Svensson (1989), may also reduce the risks which accompany this procedure. When medically indicated, it may also be feasible to use indwelling central venous catheters to obtain repeated blood samples. The successful use of an umbilical artery cannula to characterise the disposition of gentamicin was reported by Landers et al. (1984). While these investigators found that pharmacokinetic parameters [e.g. clearance (CL), volume of distribution at steady-state (V ss)] for gentamicin were similar to those reported by previous investigators who used venous blood for repeated sampling (Assael et al. 1980), both the pharmacokinetics and pharmacodynamics of a drug may be markedly dependent on whether arterial or venous blood is collected (Chiou 1989a, 1989b). Another potential problem associated with using indwelling vascular catheters resides with possible contamination of the specimen by residual drug that is contained in the lumen of the catheter and/or has physically associated with the matrix of the catheter material (Umstead 1985). This particular phenomenon has been previously reported for aminoglycosides, cyclosporin and phenytoin (Franson 1987; Murphy & Ward 1991; Umstead 1984). It most often results from obtaining the blood sample for drug analysis through the same lumen used for drug administration or, with implantable venous infusion devices, from the reservoir which has not been completely flushed of drug-containing solution. For this reason, blood samples for pharmacokinetic analysis should al-
Clin. Pharmacokinet. 23 (1) 1992
ways be obtained from a site not used for drug administration. In our opinion, the most desirable, safe and scientifically valid method for repeated blood sampling in paediatric patients and that associated with the least discomfort is the insertion of a separate, indwelling venous cannula into the largest accessible vein, preferably in an extremity. Discomfort can be further minimised by placing the cannula in the forearm or hand, when possible, so the extremity used for sampling does not have to be immobilised. When the study drug is being administered intravenously, it is preferable to place the cannula in an extremity contralateral to that being used for drug administration to avoid 'downstream' contamination of the specimen. This technique provides a sound balance between sensitivity to risks and discomforts associated with research in children and the necessary rigor required for well designed investigations (Lascari 1981). The insertion of a peripheral venous cannula dedicated solely to experimental use may also provide a direct benefit to the patient as it may be accessible for therapeutic phlebotomy and, at the conclusion of the investigation, may be available to serve as a site for other intravenous therapies. 1.2.2 Urine Collection Given the marked developmental changes in renal function which may influence the pharmacokinetics of a drug in paediatric patients (Besunder et al. 1988a, 1988b; Kearns & Reed 1989; Morselli 1989), it is important to determine the renal clearance of compounds eliminated by the kidneys. Additionally, as described in a recent review by Spino (1991), the quantification of drug and/or metabolites in urine may not only prove invaluable in examining its clearance profile but, as in the case of theophylline in patients with cystic fibrosis, may show how specific hepatic biotransformation pathways may be influenced by a disease process. The accurate collection of urine specimens may significantly enhance the quality of a pharmacokinetic investigation in paediatric patients; thus, the investigator should attempt to overcome the inherent problems associated with this procedure.
Pharmacokinetic Studies in Paediatric Patients
Unless a patient has a medically indicated indwelling bladder catheter, the collection of timed, quantitative urine specimens must use spontaneously voided specimens. This is problematical in infants and small children who have not been toilet-trained, since externally applied collection devices frequently leak and the accurate recovery of a urine sample not contaminated by faecal matter from an infant's diaper is virtually impossible. With close attention, accurate urine collection can be obtained in male infants by attaching an external collection device and aspirating urine immediately after each voiding. A child who is toilet-trained can provide accurate quantitative urine collections, especially when voiding is supervised by a research nurse or a motivated parent who clearly understands the importance of sample collection. The investigator should also be aware that many children may not void on demand, that an erratic pattern of voiding may be normal for a given child and that it is not unusual for a normal, young child to have occasional nocturnal enuresis. In an attempt to overcome these difficulties, the investigator should, first, plan intervals for urine collection which will allow for the greatest chance of consistent specimen collection throughout the entire sampling period (i.e. collection intervals of 4 to 8h vs 13 years should be obtained unless there is an overriding reason why this should not be done. Also, children > 7 years of age should be provided the opportunity to assent to participation, unless doing so would clearly not be in their best interest. There may, however, be times when this guideline should be waived, either because of
24
the condition of the child or the circumstances surrounding the child's illness. 2.2.2 Risk/Benefit Considerations of risk assume that the study is well designed and has scientific merit. Enrolling children in poorly designed or scientifically nonmeritorious studies can never be defended as ethical regardless of how little risk is involved. IRBs occasionally debate whether or not they have the obligation or responsibility to judge the scientific merit of a study. To the extent that scientific merit relates to the assessment of risk to the participants, assessment of scientific merit does fall within the purview of the IRB. In most instances, the risk to paediatric subjects is reduced by commencing drug studies in children after animal studies and initial studies in adults have documented acceptable safety. However, in some cases, this is not possible; for example when the condition being studied only occurs in children or the drug is being studied for a uniquely paediatric indication and may be too toxic to study in healthy adult volunteers. In such instances the initial studies may need to be performed in children. With their vulnerability and inability to independently consent to participate as research subjects, children should be provided extra protection against undue risk. Both the US Federal and AAP guidelines have defined risk in terms of minimal risk and greater than minimal risk. Risk is relative and must always be defined in relation to possible benefits from participation. 'Minimal risk' is the degree of risk a child would encounter during usual life activities and routine medical care. Obviously, the type and degree of risk which would constitute minimal risk varies with age and environment. Circumstances which are not considered a significant risk to adults may constitute a significant risk to children and vice versa. Examples of risk factors which deserve special attention for children are anticipatory fear of a procedure, physical pain which they don't understand and separation from parents or familiar surroundings. In general, children should be permitted to participate in studies involving minimal risk if the
Clin. Pharmacokinet. 23 (1) 1992
study promises benefit to the larger society of children, to a specific group of children and/or to the individual child. Participants in such studies could include healthy children under certain circumstances. For example, single dose pharmacokinetic studies requiring urine collections and multiple blood samples might be conducted in healthy children if, in the judgement of the IRB, the risks associated with the test drug, urine collection and venipuncture do not exceed the risks to which the child was exposed in his/her daily life. However, inclusion of healthy children in studies which offer no direct benefit to the child must be reviewed and monitored by the IRB to insure that the child's rights are protected. Inclusion of healthy children
Clin. Pharmacokinel. 23 (I): 10-29, 1992 0312-5963/92/0007-0010/$10.00/0 © Adis International Limited. All rights reserved . CPKl
Pharmacokinetic Studies in Paediatric Patients Clinical and Ethical Considerations Ralph E , Kauffman and Gregory L. Kearns Departments of Pediatrics and Pharmacology, Wayne State University School of Medicine, Children's Hospital of Michigan, Detroit, Michigan, and Section on Pediatric Clinical Pharmacology, Department of Pediatrics, University of Arkansas for Medical Sciences, Arkansas Children's Hospital, Little Rock, Arkansas, USA
Contents 10 12 12 15 19 22 22 23 15 26
Summary
Summary I, Practical and Theoretical Considerations 1.1 The Approach to the Patient/Subject 1.2 Methods and Techniques for Sample Collection 1.3 Methods for Data Analysis: Analytical and Pharmacokinetic Considerations 2. Ethical Considerations 2.1 History of Protection of Paediatric Research Subjects 2.2 Ethical Considerations in Drug Studies in Children 2.3 Special Populations 3. Conclusions
Important advances in paediatric clinical pharmacology have been made over the past 2 decades. However, there remains a reluctance to pursue pharmacodynamic and pharmacokinetic studies in children and, consequently, many important therapeutic agents have not been adequately studied in this population. Age-related pharmacokinetic/pharmacodynamic studies are not only essential to provide optimal drug therapy for children, but are quite feasible. Usually, paediatric pharmacokinetic studies are conducted in children receiving treatment for a specific medical condition. The approach to soliciting participation of paediatric subjects requires special sensitivity to the fears and anxieties of the child and the parents. Factors influencing subject enrolment and suggestions to enhance enrolment into study protocols are discussed. Pharmacokinetic/pharmacodynamic studies require repeated measurements over time and often entail obtaining multiple blood and urine samples. Techniques for reducing sample volume and number of necessary samples while minimising the discomfort and fear associated with obtaining multiple samples include the development of highly sensitive analytical methods to measure drug concentrations in small volume samples. The number of samples obtained from individual subjects can be minimised by using pharmacokinetic analytical approaches such as the nonlinear mixed effect model (NONMEM) which allows estimation of pharmacokinetic characteristics of a population using limited data from each subject. In addition, less invasive methods to measure drug metabolism/elimination such as salivary sampling, transcutaneous collection and breath analysis have been applied to the study of certain drugs, Children are a particularly vulnerable population because of their limited cognitive abilities
Pharmacokinetic Studies in Paediatric Patients
II
and dependence on adults. Thus, they must be afforded greater protection from exploitation as research subjects than that provided to adults.
The passage of the 1962 Kefauver-Harris amendments to the Pure Food and Drug Act in the United States provided a clear mandate for improving the safety and efficacy of new drugs for all patients, including paediatric patients. A decade after the amendments were passed, the American Academy of Pediatrics (AAP) Committee on Drugs (COD) published guidelines for the types of studies needed for approval of drugs for use by children (AAPCOD 1974). These guidelines provided scientific guidance and recommendations to all individuals [i.e. scientists and clinicians in private practice, academia, the Food and Drug Administration (FDA) and the pharmaceutical industry) concerned with any facet of therapeutic drug evaluation in infants and children. The AAP report contained 3 statements which are as relevant to paediatric clinical pharmacology today as they were almost 2 decades ago: first, that pharmacokinetic studies appropriate to the drug and its intended use are essential for (children in) each age group; secondly, that clinical investigations in paediatric patients are ethically defensible; and thirdly, that scientifically valid studies can be performed in paediatric patients. As a direct result of the technological advances which have occurred since the AAP report first appeared, important advances in paediatric clinical pharmacology have been made as reflected by the ever growing data base on many commonly used drugs (Besunder et al. 1988a,b; Boreus 1982; Gilman 1990; Green & Mirkin 1984; Kauffman 1985; Kearns & Reed 1989; MacLeod 1985; Milsap & Szefler 1986; Morselli 1983; Morselli et al. 1980; Prandota 1985; Rane 1980, 1984; Rane & Wilson 1976; Roberts 1984; Stewart & Hampton 1987; Triggs et al. 1984; Wilkinson 1978; Wilson 1972). Although these important strides have partially removed paediatric patients from their status as 'therapeutic orphans' (Kearns & Reed 1989; Shir-
key 1968), there remains a reluctance/resistance by the pharmaceutical industry and many clinical investigators to pursue pharmacokinetic and pharmacodynamic studies in infants and children. This aversion is fuelled in many instances by concerns regarding ethical constraints, a lack of awareness of the impact of development on drug disposition, the relatively small contribution of the paediatric age group to the profits from many drugs and products, the apparent lack of perception of need for these studies and a general lack of societal acceptance of performing clinical investigations in children. Additionally, in many cases, there is a basic unfamiliarity with the requirements and techniques for conducting pharmacokinetic/pharmacodynamic investigations in paediatric patients. The purpose of this review is to raise the level of awareness of the need for age-related evaluations of drugs and to demonstrate that such studies are not only necessary but quite feasible. Age-related pharmacokinetic/pharmacodynamic studies are essential to provide the data base from which optimal age-related doses of drugs may be derived. It is our hope that this discussion will provide an impetus for pharmaceutical company sponsors, responsible regulatory agencies and investigators to increase their respective efforts to ensure that drugs which are used in children are adequately studied in children. Only when this goal is achieved can children and their parents be assured that the drugs they are receiving are being administered under optimally safe and effective conditions. We first review some of the practical and theoretical aspects of performing clinical pharmacokinetic studies in infants and children. These include techniques for subject recruitment and retention of patients in the study, methods of sample collection, and analytical and pharmacokinetic techniques which are well suited for pharmacokinetic studies in infants and children. We then pre-
c/in. Pharmacokinet. 23 (I) 1992
12
Table I. Factors affecting patient recruitment (adapted from Spilker 1991) The nature of the appeal to participate The patient's/parent's perception of direct vs indirect benefits The specific requirements associated with the study (i.e. degree of difficulty, discomfort or compromised modesty) The degree of concern that parents/patients have regarding the disease or reason for hospitalisation The location of the study site relative to the patient's home Resources available to the investigator for recruitment The socioeconomic mix of the potential patient pool The place where recruitment occurs (i.e. home vs school vs hospital environment) The methods used for subject recruitment (i.e. public vs institutional advertising, physician referral, etc.)
sent the salient ethical issues to be considered when conducting clinical trials in children. Included are issues of informed consent, risk/benefit assessment and special protection of particularly vulnerable populations of children.
1. Practical and Theoretical Considerations 1.1 The Approach to the Patient/Subject In most instances, clinical pharmacokinetic studies in infants and children are not conducted in normal, healthy volunteers as is the general rule with many investigations performed in adults. Consequently, the 'subject' in paediatric studies is most often a patient who is being treated for a specific medical indication which may not be related to the study being performed. Since paediatric subjects are usually ill children, the approach used to solicit the participation of a paediatric patient in a pharmacokinetic study and the approval of their parents or guardians requires special sensitivity (Grodin & Alpert 1988) and should be in full conformity with specific national regulations governing research on minors (e.g. 45 CFR 46.401-46.409, 1983).
1.1.1 Factors Influencing Recruitment of Research Subjects Factors which may influence the recruitment of paediatric patients as research subjects are summarised in table I. The nature of the request for the patient and their parents/guardians to participate in a study may represent the most important determinant for the success or failure of recruitment efforts in a clinical pharmacokinetic investigation (Spilker 1991). Two factors, however, frequently dissuade the investigator or parent from freely entering into a trial. First, in adult clinical trials, investigations to determine the pharmacokinetic properties of a drug are generally conducted in phase I trials and may be viewed as phase I-type trials regardless of when they are conducted during the development of a drug (Spilker 1991). Accordingly, little direct patient benefit can be expected and/or promised, especially when initial single dose studies are performed. Secondly, for clinical pharmacokinetic studies to yield the type of information which the US FDA has deemed desirable (Peck & Collins 1990), it is often necessary that multiple biological specimens (e.g. blood, urine, saliva) be obtained (Svensson 1989). Frequently, these procedures may produce added pain and discomfort (e.g. the insertion of an indwelling cannula, multiple fingertip punctures) or may compromise modesty in the view of the patient or their parents (e.g. the supervised collection of multiple urine samples). Thus, clinical pharmacokinetic investigations in paediatric patients may be perceived as falling into a category of research involving children described by Rozovsky (1984) as entailing more than minimal risk to the patient and offering little or no direct patient benefit but only the prospect of 'generalisable knowledge'. When discussing this type of study with the patient and their parents/guardians they should be clearly informed that although the patient may not derive direct therapeutic benefit by participating in a clinical pharmacokinetic study, the new knowledge obtained as a direct result of their participation will not only make a unique contribution to understanding the pharmacology of a new drug in children but may also benefit other children who might
Pharmacokinetic Studies in Paediatric Patients
receive the drug in future evaluations of the safety and/or efficacy of the drug. Alternatively, in some instances, a pharmacokinetic study may be conducted as part of an efficacy study in which the child may directly benefit from receiving the drug, although the benefit is not directly related to the pharmacokinetic part of the protocol. In our experience, when the investigator clearly and concisely relates the scientific goals and specific aims of the particular investigation, outlines the required procedures and the safeguards to minimise both risk and discomfort to the patient, and educates the patient/parents regarding the scientific implications and importance of the information to be obtained, the consent process becomes truly 'informed'. Most importantly, the patient and their parents have a proper frame of reference on which to make a decision as to whether the potential benefits of participating in the study outweigh the perceived risks. Several additional practical issues which are of equal importance with regard to the appeal made for patient recruitment merit discussion. As expected, the hospitalisation of an infant or child usually produces a heightened level of anxiety and concern on behalf of the patient and their family. Consequently, an overwhelming desire which often permeates this scenario is that only those procedures which are 'medically necessary' be performed, that unnecessary discomfort be avoided and/or minimised and that no unnecessary intervention be introduced which might alter the planned course of treatment and/or prolong the period of hospitalisation. It is therefore of paramount importance that the investigator both understands and respects these desires and expectations which demand that the planned course of medical treatment for a given disease/condition not be altered or adversely affected by the subject's participation in any clinical investigation. In our view, this objective can be facilitated and achieved by the following actions. First, the investigator should be thordughly familiar with the patient's condition, carefully reviewing the medical record and discussing the patient with the primary care physician before mentioning the investigation to the patient
13
or their parents. Secondly, once a potential subject has been identified and the support of the patient's primary physician is obtained, the investigator should describe the study protocol to the patient and their parents and should carefully and completely review the scope, study procedures and expected outcome/relevance of the experimental findings. Thirdly, the investigator should demonstrate to the patient and their parents how the required study procedures will not alter the planned course of medical treatment and note the safeguards (i.e. monitoring procedures, 'rescue criteria', etc.) in place to insure that all risks associated with subject participation are truly minimal. Fourthly, the investigator must clearly relate that the ultimate decision and authority in allowing the investigation to begin and to continue resides with the patient and their parents. Lastly, and most importantly, the investigator needs to communicate clearly that his/her most important concern and that of his/her associates resides not with the outcome of the investigation but with the health, safety and welfare of the patient. Although this approach requires a considerable investment of time and resources on behalf of the investigator, it embraces responsibilities which should not be abdicated and, if performed properly, can result in a high rate of success in patient recruitment and retention. Thus, the success of recruitment is directly related to the degree of rapport the investigator has with the patient and parents. This is supported by our experience over the past 5 years which reflects successful patient recruitment (i.e. by signed consent form) in over 85% of all encounters for patients between the ages of 6 months and adolescence in 15 separate pharmacokinetic studies (Kearns et aI., unpublished data). 1.1.2 Techniques For Increasing Patient Enrolment There are several techniques which investigators can use to increase patient enrolment in clinical investigations (Spilker 1991). Those techniques which may be relevant to paediatric pharrnacokinetic studies are summarised in table II. As mentioned previously, the most important considera-
14
Table II. Techniques to increase patient enrolment (adapted from Spilker 1991) Assess patient motivation to enter study and specifically address patient/parental needs Use study procedures to broaden base of medical care by making the results from testing required by the protocol available to the parent. the primary physician and the medical record Enlist the services of a patient recruiter (e.g. nurse, pharmacist, lay person) Increase the level of patient benefits by providing financial resources to compensate the patient or their parent(s) for basic necessities (e.g. transportation, meals, child care for siblings of patient, etc.) Conduct relevant chart/record reviews (e.g. pharmacy or laboratory records, patient medical records) to determine total available patient pool Use targeted professional advertisements (i.e. colleagues, house officers, nurses) As far as possible (i.e. without compromising the safety of the study and/or the quality of the science), loosen the demands of the protocol (Le. eliminate some disagreeable tests/ procedures or modify inclusion/exclusion criteria) Expand the number of study sites (e.g. additional academic centres, use of academic private practice settings, etc.) Provide relevant communication (e.g. an 'outline' of the study and required procedures) and feedback to professional staff (i.e. nurses, physicians) who may assist with recruitment and/ or the study protocol per se Provide information regarding the status of the patient and/or relevant tests results to the primary care physician(s) Target research focus to subpopulations which historically are accessible and/or who have had previous good experiences with participating in clinical investigations
tions for increasing subject enrolment are to assess patient and parental motivation, completely discuss the experimental protocol and specifically address the expressed and perceived needs (or questions) of the patient and/or their parents. Other selected techniques for increasing patient enrolment (table II) which have proven to be successful in our experience and that of other investigators merit further consideration and discussion. Enlisting the services of an individual to specifically notify the investigator of the potential el-
Clin. Pharmacokinet. 23 (1) 1992
igibility and/or desire of a patient to participate in a clinical investigation may serve as an effective tool for enhancing recruitment. This activity may take several forms ranging from the hiring of a 'patient recruiter' to the use of professional medical staff(i.e. nurses, house officers, staff physicians) with general knowledge of the inclusion/exclusion criteria who agree to notify the investigator of potential patients. The patient recruiter may be a lay person whose primary responsibility is to identify potential study candidates and to ascertain from the patient and their parents whether they would be interested in speaking to the investigator about a given clinical investigation (Walson, personal communication). When medical residents (i.e. house officers) are used to identify potential study patients, it is often customary to offer some type of tangible incentive or 'finder's fee' (e.g. financial remuneration, educational stipends, textbooks, etc.). However, this practice has been condemned because of the potential for a conflict of interest or introduction of coercion into the recruitment process (Lind 1990). On the other hand, targeted professional education of nursing and medical staff regarding a specific area of research or clinical study can be very effective in increasing patient enrolment. This latter endeavour can produce a heightened awareness of a particular area of investigation and reinforce a spirit of cooperation and collaboration between the investigator and workers providing ongoing medical care to the study subject. Another successful technique for enhancing patient enrolment is to optimise the benefits associated with subject participation. As summarised in table II, this could include the provision ofmonetary resources to compensate the patient's parents for necessary expenses related to subsistence and transportation during the study period and expanding the patient's records by including the results of tests required by the study protocol [e.g. renal and hepatic function profiles, glucose-6-phosphate dehydrogenase (G6PD) screening, ECG, radiographs, etc.]. Although these tests may not be medically indicated for the patient's underlying condition, their costs are covered by resources sup-
Pharmacokinetic Studies in Paediatric Patients
porting a given study. This additional information may assist the primary care physician in the overall assessment of the patient. This is illustrated by our recent experience where G6PD deficiency was diagnosed in 2 children from prestudy laboratory data required for a single-dose pharmacokinetic study of a new, oral {3-lactam antibiotic (Kearns, unpublished data). It is important, however, that compensation or benefits not be of a nature or extent that they constitute a coercive influence on the patient or parents to participate in the study (see below). A final consideration of note for increasing patient enrolment resides with the site used for patient recruitment and study. If the patient must be transferred from a hospital room or an outpatient treatment area to a clinical research facility, care should be taken to insure that the environment is appropriately equipped and staffed for providing necessary medical care, is conducive to the needs and desires of the patient (e.g. a bathroom within the study room, contains materials to entertain/comfort the patient, has sufficient space to afford normal interaction between the patient and family members, etc.) and can accommodate the personnel and equipment necessary for conducting the study without producing a significant intrusion to the routine care of the patient. For clinical pharmacokinetic studies that do not involve elaborate equipment or labour intensive pharmacodynamic assessments, it is often possible to successfully perform the study in the patient's hospital room (Kearns et al. 1988). Additionally, the hospital room and nursing staff may provide a measure of'institutional familiarity' which is often times conducive to patient recruitment. This is especially true when the support staff are well educated about the study and are supportive of the work. Finally, certain types of investigations may allow the recruitment and enrolment of healthy children from environments which are characteristically not associated with clinical research. This has been recently illustrated by Relling et al. (1991) who conducted a pharmacogenetic study of debrisoquin oxidation in over 500 children who were
15
recruited from private elementary and secondary schools. 1.2 Methods and Techniques for Sample Collection To accomplish the goals of a pharmacokinetic study, it is necessary to obtain repeated observations of drug concentrations in a biological fluid which presumably reflects the concentration vs time profile (AUC) for the drug or its active metabolite at the site of action. Traditionally, the preferred and most accessible fluids for obtaining repeated assessments of drug concentration have been blood and urine (Kearns & Trang 1986; Matzke & St. Peter 1990; Svensson 1989). However, in paediatric patients, ethical, physiological and sometimes practical limitations may influence the availability of biological samples required for pharmacokinetic studies. There are several major considerations regarding sampling in paediatric patients and potential alternatives to the use of blood and urine specimens.
1.2.1 Repeated Blood Sampling Since most pharmacokinetic investigations are conducted by obtaining repeated blood samples (Svensson 1989), an invasive experimental procedure is introduced (i.e. venepuncture) which may carry risks that have been determined in some cases to be more than minimal (Janofsky & Starfield 1981). The risks associated with repeated venous sampling and/or venepuncture may include excessive blood loss (e.g. when the total sampling volume obtained over a 24h period exceeds 10% of the estimated circulating blood volume), pain, bruising, infection and possibly the anterior intraosseous syndrome (Saeed & Gatens 1983). When repeated capillary blood samples are obtained in infants by either fingertip or heel puncture (Kearns et al. 1985; Wilson et al. 1982), the potential exists for producing pain, increased anxiety (associated with restraining the infant while the blood sample is 'milked' from the extremity), bruising, infection, bony injury and, importantly, potential errors in drug measurement. Murphy and Ward (1991) re-
16
cently showed that discrepancies between drug concentrations in serum derived from capillary vs venous blood may result from contamination of the blood specimen with extracellular fluid and metabolic differences between central vs peripheral blood which might influence drug equilibrium. In our collective experience in evaluating the pharmacokinetics of antimicrobial agents (Kauffman et al. 1981; Kearns et al. 1982, 1989), and most recently midazolam (Tolia et al. 1991; Wells et al. 1991), we have found that indwelling, silastic venous cannulas or heparinised small gauge needles markedly reduce the discomfort associated with repeated venous blood sampling and, as stated by Svensson (1989), may also reduce the risks which accompany this procedure. When medically indicated, it may also be feasible to use indwelling central venous catheters to obtain repeated blood samples. The successful use of an umbilical artery cannula to characterise the disposition of gentamicin was reported by Landers et al. (1984). While these investigators found that pharmacokinetic parameters [e.g. clearance (CL), volume of distribution at steady-state (V ss)] for gentamicin were similar to those reported by previous investigators who used venous blood for repeated sampling (Assael et al. 1980), both the pharmacokinetics and pharmacodynamics of a drug may be markedly dependent on whether arterial or venous blood is collected (Chiou 1989a, 1989b). Another potential problem associated with using indwelling vascular catheters resides with possible contamination of the specimen by residual drug that is contained in the lumen of the catheter and/or has physically associated with the matrix of the catheter material (Umstead 1985). This particular phenomenon has been previously reported for aminoglycosides, cyclosporin and phenytoin (Franson 1987; Murphy & Ward 1991; Umstead 1984). It most often results from obtaining the blood sample for drug analysis through the same lumen used for drug administration or, with implantable venous infusion devices, from the reservoir which has not been completely flushed of drug-containing solution. For this reason, blood samples for pharmacokinetic analysis should al-
Clin. Pharmacokinet. 23 (1) 1992
ways be obtained from a site not used for drug administration. In our opinion, the most desirable, safe and scientifically valid method for repeated blood sampling in paediatric patients and that associated with the least discomfort is the insertion of a separate, indwelling venous cannula into the largest accessible vein, preferably in an extremity. Discomfort can be further minimised by placing the cannula in the forearm or hand, when possible, so the extremity used for sampling does not have to be immobilised. When the study drug is being administered intravenously, it is preferable to place the cannula in an extremity contralateral to that being used for drug administration to avoid 'downstream' contamination of the specimen. This technique provides a sound balance between sensitivity to risks and discomforts associated with research in children and the necessary rigor required for well designed investigations (Lascari 1981). The insertion of a peripheral venous cannula dedicated solely to experimental use may also provide a direct benefit to the patient as it may be accessible for therapeutic phlebotomy and, at the conclusion of the investigation, may be available to serve as a site for other intravenous therapies. 1.2.2 Urine Collection Given the marked developmental changes in renal function which may influence the pharmacokinetics of a drug in paediatric patients (Besunder et al. 1988a, 1988b; Kearns & Reed 1989; Morselli 1989), it is important to determine the renal clearance of compounds eliminated by the kidneys. Additionally, as described in a recent review by Spino (1991), the quantification of drug and/or metabolites in urine may not only prove invaluable in examining its clearance profile but, as in the case of theophylline in patients with cystic fibrosis, may show how specific hepatic biotransformation pathways may be influenced by a disease process. The accurate collection of urine specimens may significantly enhance the quality of a pharmacokinetic investigation in paediatric patients; thus, the investigator should attempt to overcome the inherent problems associated with this procedure.
Pharmacokinetic Studies in Paediatric Patients
Unless a patient has a medically indicated indwelling bladder catheter, the collection of timed, quantitative urine specimens must use spontaneously voided specimens. This is problematical in infants and small children who have not been toilet-trained, since externally applied collection devices frequently leak and the accurate recovery of a urine sample not contaminated by faecal matter from an infant's diaper is virtually impossible. With close attention, accurate urine collection can be obtained in male infants by attaching an external collection device and aspirating urine immediately after each voiding. A child who is toilet-trained can provide accurate quantitative urine collections, especially when voiding is supervised by a research nurse or a motivated parent who clearly understands the importance of sample collection. The investigator should also be aware that many children may not void on demand, that an erratic pattern of voiding may be normal for a given child and that it is not unusual for a normal, young child to have occasional nocturnal enuresis. In an attempt to overcome these difficulties, the investigator should, first, plan intervals for urine collection which will allow for the greatest chance of consistent specimen collection throughout the entire sampling period (i.e. collection intervals of 4 to 8h vs 13 years should be obtained unless there is an overriding reason why this should not be done. Also, children > 7 years of age should be provided the opportunity to assent to participation, unless doing so would clearly not be in their best interest. There may, however, be times when this guideline should be waived, either because of
24
the condition of the child or the circumstances surrounding the child's illness. 2.2.2 Risk/Benefit Considerations of risk assume that the study is well designed and has scientific merit. Enrolling children in poorly designed or scientifically nonmeritorious studies can never be defended as ethical regardless of how little risk is involved. IRBs occasionally debate whether or not they have the obligation or responsibility to judge the scientific merit of a study. To the extent that scientific merit relates to the assessment of risk to the participants, assessment of scientific merit does fall within the purview of the IRB. In most instances, the risk to paediatric subjects is reduced by commencing drug studies in children after animal studies and initial studies in adults have documented acceptable safety. However, in some cases, this is not possible; for example when the condition being studied only occurs in children or the drug is being studied for a uniquely paediatric indication and may be too toxic to study in healthy adult volunteers. In such instances the initial studies may need to be performed in children. With their vulnerability and inability to independently consent to participate as research subjects, children should be provided extra protection against undue risk. Both the US Federal and AAP guidelines have defined risk in terms of minimal risk and greater than minimal risk. Risk is relative and must always be defined in relation to possible benefits from participation. 'Minimal risk' is the degree of risk a child would encounter during usual life activities and routine medical care. Obviously, the type and degree of risk which would constitute minimal risk varies with age and environment. Circumstances which are not considered a significant risk to adults may constitute a significant risk to children and vice versa. Examples of risk factors which deserve special attention for children are anticipatory fear of a procedure, physical pain which they don't understand and separation from parents or familiar surroundings. In general, children should be permitted to participate in studies involving minimal risk if the
Clin. Pharmacokinet. 23 (1) 1992
study promises benefit to the larger society of children, to a specific group of children and/or to the individual child. Participants in such studies could include healthy children under certain circumstances. For example, single dose pharmacokinetic studies requiring urine collections and multiple blood samples might be conducted in healthy children if, in the judgement of the IRB, the risks associated with the test drug, urine collection and venipuncture do not exceed the risks to which the child was exposed in his/her daily life. However, inclusion of healthy children in studies which offer no direct benefit to the child must be reviewed and monitored by the IRB to insure that the child's rights are protected. Inclusion of healthy children
