Systematic Review of Early Phase Pediatric Clinical Pharmacology Trials

Abstract

OBJECTIVE

Children have generally been excluded from early-stage clinical trials owing to safety concerns based in social expectations and not data. However, the repositioning of adult therapeutics for pediatric use and the increase in the development of therapies for pediatric only conditions require the participation of children in phase 1–2 trials. Therefore, the aim of this article is to systematically review the history and current state of early phase pediatric clinical pharmacology trials in order to understand safety concerns, trends, and challenges in pediatric trials.

METHODS

This review analyzed the nature of early phase pediatric clinical trials conducted for nononcology conditions through a systematic search that was performed for pediatric non-oncologic phase 1 or phase 1–2 drug and vaccine studies in MEDLINE.

RESULTS

The data show that the number of early phase pediatric clinical trials is still small relative to adults but has been on the rise in the past decade with relatively few serious adverse effects observed.

CONCLUSIONS

The widespread concerns about children's safety when they participate in early phase clinical trials seem disproportionate, based on our findings. The data confirm that these studies can be conducted safely, and that their results can contribute significantly to pediatric pharmacotherapy.

Introduction

The drug discovery, development, and marketing process is a highly regulated endeavor with well-defined stages, designed to evaluate drug safety and human pharmacology (i.e., phase 1 and phase 2 studies), efficacy (phase 3 clinical trials), and post-marketing safety (phase 4 studies). Drug clinical trials mostly involve adult volunteers; children, and pregnant and lactating women, are excluded from clinical research in general, and from early phase clinical trials in particular.¹ Many barriers prevent pediatric clinical trials in general: lack of monetary motivation, clinical trial design difficulties based in a developing anatomy and physiology, and ethical concerns are some of the examples. For early phase studies in particular, it has been historically claimed that this exclusion is mainly due to safety concerns, mostly based on the lack of information on potential adverse events associated with the studied medications.² However, in the case of the pediatric population, this perception of heightened risk is predominantly based on social concepts (e.g., “children are fragile”), as data supporting a higher risk of early phase studies to the pediatric population are lacking.³

Early phase pediatric clinical trials are scarce, especially phase 1 studies (i.e., those that evaluate safety and pharmacokinetics of new drugs), with the exception of pediatric oncology. Pediatric oncologists have a long-standing practice of systematically enrolling their patients in clinical trials, owing to the distinct nature of pediatric cancers that makes them (and their treatments) different from adult cancers. This makes clinical research to define effectiveness and safety in children unavoidable. However, outside pediatric cancer research, and some severe neonatology conditions, it is unusual for phase 1 (and, particularly, first-in-human studies) or even phase 2 drug studies to enroll children.² On the other hand, with new therapies for rare diseases that manifest predominantly, or even exclusively, in childhood becoming increasingly common, phase 1 pediatric clinical trials are expected to increase, because it may not be justifiable to expose healthy adult volunteers to medications that may have significant potential for adverse events, even if those would be acceptable in the context of a severe health condition in a child (i.e., in exchange for a possible improvement of that disease).

Furthermore, many medications initially approved in adults for a certain diagnosis are being studied in pediatrics for different diseases (usually referred to as drug repositioning, or off-label use depending on the circumstances).⁴ This transfer of a medication from an adult use to a pediatric application sometimes requires evaluation of safety and pharmacokinetics (i.e., akin to a phase 1 clinical trial). Often phase 4 trials are conducted in children to expand existing therapeutic indications, but there are many scenarios where this is not the best approach.

Understanding the current state of early phase pediatric trials can be useful in a number of ways. Informing policy makers on the impact of policies such as the US Best Pharmaceuticals for Children Act (BPCA) and Pediatric Research Equity Act (PREA) and bringing evidence to researchers on the real dangers for participants are the highlights. Therefore, the overall aim of this article is to systematically review the history and current state of early phase pediatric clinical pharmacology trials in order to explore safety concerns, trends, and challenges in pediatric trials.

Materials and Methods

A systematic search of the US National Library of Medicine bibliographic database MEDLINE, using PubMed, was conducted for peer-reviewed pediatric phase 1 clinical trials published in any year between 1964 and July 2020 (studies were not limited by start year in order to obtain the maximum number of publications in this already limited area). Studies were identified by using the filtered search option available in PubMed; the database was first examined by using the search strategy described in Table 1. In short, included studies were all phase 1 pediatric clinical trials that were conducted for non-oncology pediatric conditions. One investigator assessed the abstracts and full text articles to filter out studies that did not meet inclusion and exclusion criteria such as studying drugs for cancer, studies exclusively involving adult patients, or studies that were not phase 1 clinical trials. Any uncertainties in study eligibility were discussed and resolved with the input of the primary investigator. A list of all studies can be found in the Supplemental Material.

Table 1.

Search Strategy Used in PubMed

Initial Search

Additional Limits

Pediatric or child or infant, newborn or neonate or youth or adolescent) AND clinical trial

Humans “all infant (birth to 23 months)” or “all child (0 to 18 years)” or “newborn infant (birth to 1 month)” or “infant (1 to 23 months)” or “preschool child (2 to 5 years)” or “child (6 to 12 years)” or “adolescent (13 to 18 years)” English language (children or children - focused or “humans only (removes records about animals)”) “therapy (maximizes sensitivity)” or “therapy (maximizes specificity)” or “therapy (best balance of sensitivity and specificity)”

After this selection process (Figure 1), relevant data were collected from the selected studies by using a premade electronic data form for the following study information, if reported: study title and ID, author details, study publication date, type of trial (e.g., phase 1 or phase 1–2), type of intervention (i.e., drug, or vaccine), whether the trial involved blinding and/or was a first-in-human study, the condition and drug or intervention being tested, the study purpose and methods, and where the study was conducted. The following data about the study participants were also collected if available: sample size; participant ages, sexes, ethnicities; inclusion and exclusion criteria; whether healthy pediatric participants were included; and the occurrence of serious drug-related adverse effects and subsequent withdrawal of any participants.

Figure 1.

Flowchart of the study selection process.

Results

The systematic search, conducted in May 2020 and using prespecified search criteria (Table 1), initially produced 7173 findings, of which 317 met the preestablished eligibility criteria (Figure 1). These were further classified as drug studies (n = 225) and vaccine studies (n = 92) (Figure 1). No studies that tested a new drug in children without having tested in adults before were found.

Drug Studies. These studies included both pharmaceutical and biological drugs. Of the included 225 trials, 122 (54%) were strictly phase 1 (i.e., were concerned only with drug safety and pharmacokinetics), and 103 (46%) were combined phase 1–2 studies (i.e., besides safety and pharmacokinetics, they also evaluated drug effect in a small population of patients). The number of publications has been steadily increasing with time: 38 between 1988–2000; 68 between 2001–2010; and 119 between 2011–2019. The year of publication and therapeutic areas addressed in these are summarized in Figures 2 and 3, respectively.

Figure 2.

Publication year of included drug studies.

Figure 3.

US Pharmacopeia therapeutic category of included drug studies.³⁵

North America (n = 175) and Europe (n = 50) were the most predominant geographic locations, followed by Africa (n = 19), South America (n = 14), Asia (n = 24), and Australia (n = 8). A total of 151 (67%) studies were conducted in only 1 country, and the rest were multinational studies.

The median study sample size was 33 patients (range, 3–606). Only 42 (19%) used blinding when assigning study participants to treatments. All trials identified enrolled pediatric patients, but the minimum age varied; median minimum age was 1.2 years (range, 0–16).

The most common investigated therapeutic areas were antiviral drugs (27% of all studies), dominated by HIV-AIDS medications, followed by blood products and modifiers (10%), drugs for genetic or enzyme protein disorders (8%), immunologic agents (7%), and antibacterials (6%). The remaining 41% of trials were conducted for other conditions. Besides HIV-AIDS, which took up almost one-fourth of all studies identified (n = 56), no specific pathology was evaluated in more than 3 trials (Table 2).

Table 2.

Therapeutic Areas and Study Conditions

Drugs	Conditions Tested
Pharmaceutical
Antivirals (n = 60)	HIV-AIDS (56), La Crosse encephalitis (1), hepatitis B (1), cytomegalovirus (1), herpesvirus (1)
Blood products and modifiers (n = 23)	Neonatal sepsis (1), post-CPB organ dysfunction (1), bacterial sepsis (1), post-hemorrhagic hydrocephalus (1), severe chronic immune thrombocytopenia (1), Crohn disease (1), neonatal encephalopathy (1), severe meningococcal sepsis (1), Kawasaki disease (1), immune thrombocytopenic purpura (1), hypoxic-ischemic encephalopathy (1), primary immune deficiency (1), neonatal cardiac surgery (1), respiratory syncytial virus infection (3), venous thromboembolism (1), extremely low birth weight (risk of brain injury and neurodevelopmental problems) (1), hemophilia B (2), Shiga-like toxin-producing Escherichia coli infection (1), anemia of chronic renal failure (1), hemophilia A (1)
Antibacterials (n = 14)	Gram-positive infections (5), cystic fibrosis (1), acute bacterial skin and skin structure infections (1), Gram-negative infections (2), aerobic Gram-negative and Gram-positive and anaerobic pathogens (1), staphylococcal bloodstream infections (1), staphylococcal bloodstream infections, acute otitis media (1), tonsillopharyngitis (1), community-acquired bacterial pneumonia (1)
Other (n = 93)	Candida parapsilosis colonization (1), fetal alcohol spectrum disorders (1), hypoxic-ischemic encephalopathy and electrographic seizures (1), refractory partial-onset seizures (1), acute agitation associated with schizophrenia and/or bipolar mania (1), ADHD (6), amblyopia (1), asthma (5), autism spectrum disorder (1), bipolar disorder (3), Brazilian kala-azar (1), bronchopulmonary dysplasia (2), candidal fungal infection (1), cardiovascular surgery with CPB (1), childhood blindness due to Leber congenital amaurosis (1), chronic kidney disease (1), chronic pain (1), closure of the ductus arteriosus (2), corrective infant cardiac surgery (1), cystic fibrosis (2), Duchenne muscular dystrophy (1), emergence agitation after sevoflurane or halothane anesthesia (1), feeding intolerance (1), Fontan palliation (1), functional constipation (1), gastroesophageal reflux disease (6), glaucoma (2), growth hormone deficiency (4), Indian visceral leishmaniasis (1), intestinal failure (1), invasive candidiasis (1), invasive fungal infections (1), juvenile rheumatoid arthritis (1), long-chain fatty acid disorders (1), major depressive disorder and seasonal affective disorder (1), mechanical ventilation (1), medically refractory epilepsy (1), migraine headaches (1), moderate to severe pain (1), Mycobacterium avium complex bacteremia (2), mydriasis (1), necrotizing enterocolitis (1), neonatal opioid abstinence syndrome (1), neonatal respiratory distress syndrome (1), neutropenia (1), pain (2), pain following injury (2), Pneumocystis carinii pneumonia (1), primary focal segmental glomerulosclerosis (1), refractory epileptic encephalopathy (1), refractory partial-onset seizures (2), retinopathy (1), sedation (1), sickle cell anemia (3), transient hypothyroxinemia (1), traumatic brain injury (1), type 1 diabetes (4), type 2 diabetes (3), vitamin D deficiency in cystic fibrosis (1), Wilson disease (1)
Biological
Genetic or enzyme or protein disorder: replacement, modifiers, treatment (n = 19)	Infantile glycogen storage disease type II (1), mucopolysaccharidosis type IIIA disease (1), alpha-mannosidosis (2), mucopolysaccharidosis IIIB (2), Duchenne muscular dystrophy (2), mucopolysaccharidosis II (2), type 1 diabetes (2), mucopolysaccharidosis type I (2), Fabry disease (1), refractory epilepsy in tuberous sclerosis complex (1), spinal muscular atrophy (3)
Immunologicagents (n = 16)	Neonatal-onset multisystem inflammatory disease (1), Fanconi anemia and pancytopenia (1), ulcerative colitis (1) , systemic juvenile idiopathic arthritis (1), inflammatory bowel disease (1), immune suppression (1), atopic dermatitis (3), renal transplant (1), stable chronic renal failure undergoing dialysis (1), familial Mediterranean fever (1), breastfeeding supplementation (1), probiotics (1), severe sepsis (1)

ADHD, attention deficit hyperactivity disorder; CPB, cardiopulmonary bypass

A total of 22 serious adverse events deemed related to the investigational product were observed in 10 (4%) studies (Table 3). Half of the serious adverse events were infusion reactions (9 of them associated to intravenous recombinant human N-acetyl-α-d-glucosaminidase administration for children with mucopolysaccharidosis). Besides infusion reactions (which were observed exclusively with the application of biological drugs), most of the serious adverse events took place in patients with severe underlying conditions, which makes causality assessment difficult.

Table 3.

Serious Adverse Effects Reported in Drug Studies *

Drug Name	Adverse Effect
Recombinant human alpha-mannosidase (rhLAMAN)27	1–2 min episode of unconsciousness followed by seizures, 7 days after the last infusion
Intravenous recombinant human N-acetyl-α-d-glucosaminidase (SBC-103)28	Infusion-associated reactions† (9 events) considered to be an important medical event and related to treatment
Atorvastatin29	2 related toxicities in HIV-positive children with hyperlipidemia (1 participant had a Grade 3 serum creatinine elevation at wk 6 with no changes in antiretroviral regimen prior to wk 6; another had Grade 4 serum ALT and AST elevations at the final study visit and also received a diagnosis of drug-induced hepatitis)
Rituximab30	Three patients experienced serious adverse effects during the first 12 wk of the study. Two patients, both non-responders, had serum sickness; one, a 12-year-old male patient, presented with fever, fatigue, and rash after the second dose of rituximab (given on day 8 after first dose), and the other, an 11-year-old female patient, developed fever, joint pain and swelling, conjunctival hyperemia, and cutaneous rash after the second rituximab dose. Another patient (non-responder) developed CTC grade 2 infusion-related hypotension with his third dose (given on day 15 after first dose)
Indinavir31	Neutropenia in 1 patient (wk 16; 250 mg/m2 capsules), increased hepatic transaminases (grade 3 toxicity) in a patient with known fluctuations in liver enzymes (wk 12; 250 mg/m2 suspension), and recurrent episodes of hematuria (350 mg/m2 capsules, details below) in another patient
Tenofovir32	Low serum albumin concentration on day 2 of life in 1 infant
Risperidone33	Psychosis manic-depressive (n = 9), suicide attempt (n = 7), allergic reaction (n = 2), asthma (n = 2), bronchospasm (n = 2)
Drotrecogin alfa34	Serious bleeding events that occurred during the 28-day study (n = 6); one of these events was a fatal intracranial hemorrhage in a severely ill patient with meningitis that occurred post-infusion on day 8

ALT, alanine aminotransferase; AST, aspartate aminotransferase; CTC, common toxicity criteria

^* Note: Individual patient data were not always available, because most studies focused on reporting the therapeutic effects and/or the pharmacokinetics of the studied drug, and adverse events (especially those not serious) were reported summarily.

^† Infusion-associated reactions included pyrexia, vomiting, diarrhea, nasopharyngitis, and cough.

Vaccine Studies. All 92 vaccine clinical trials identified were conducted for infectious diseases, with 58% for viral infections, 25% for bacterial infections, 3% for both viral and bacterial infections, and the remaining 14% for malaria. The median study sample size was 90 with a range of 10 to 1278. The year of publication for each vaccine study is summarized in Figure 4. No vaccine-related serious adverse events were observed in any of these.

Figure 4.

Publication year of included vaccine studies.

Discussion

This systematic review was conducted to examine early phase clinical trials for pediatric conditions other than cancers. The collected data show that the number of phase 1 clinical trials conducted in children is small, as compared with the thousands of early phase studies in adults. However, this number has shown an increasing trend in recent years, especially in the last decade. Particularly, vaccine studies increased substantially from 2010 to 2015, which may reflect the studies conducted for rotavirus and pneumococcus vaccines (Figure 4).

Conducting early phase oncology trials is relatively easy to justify in pediatrics given that many types of cancers are almost exclusively found in children (and therefore, it would not make sense to test potentially toxic chemotherapy for these diseases in adults before doing the pediatric studies). Furthermore, pediatric oncology has a long-standing tradition of including most patients in clinical trial protocols, so that enrollment into clinical studies, including phase 1 trials, is usually streamlined (and relatively easier). But while oncology trials clearly have to be conducted separately in the pediatric population, that might not be the case for other diseases. For many years, conducting non-oncologic early phase studies in children was somewhat controversial, because many drugs can in fact be tested safely in adult volunteers first.

Actually, pediatric patients can almost always benefit from information gleaned from adult studies, such as safety profiles and patterns, reducing the chance of enrollment at a biologically ineffective dose or undue toxicity. Unfortunately, testing in adults is insufficient in many cases, because there are many adverse events that can only manifest in pediatrics (e.g., corticosteroid-induced growth-impairment leading to lower stature would mostly take place before adulthood, and not in older patients⁵; long-lasting brain maturation effects could lead to a lower-than-expected IQ in children exposed to some anticonvulsants, such as phenobarbital, but not in adults⁶). Also, conducting early drug studies exclusively in adults creates a delay in pediatric drug development. Finally, owing to variations in physiology and pharmacology in children, the same drug or vaccine already studied in adults may have vastly different effects in children, emphasizing the need for earlier, separate clinical trials in the pediatric population.⁷ Therefore, in many cases early phase studies in children are unavoidable even if adult data are available for some drugs. In particular, severe diseases (e.g., genetic diseases) or those that affect specific pediatric subpopulations (e.g., neonates) are difficult to appropriately study in adults, and thus early phase studies in the target population become vital.

The need for pediatric-specific clinical trials has been clear for many years, but these only became more frequent with the advent of legislation (initially in the United States and Europe, followed slowly by other areas of the world) encouraging studies in this population.^8,9 There are a multitude of rationales for early pharmacologic studies to be conducted in children instead of extrapolating data from adult clinical trials, as was the norm not too long ago. These include differences in drug pharmacokinetics between adults and children¹⁰ (and even among different childhood stages), the mentioned susceptibility to adverse events that would not be common or possible in adults, such as effects of medications on growth or development, and other issues.⁷ Regulatory agencies such as the US Food and Drug Administration have also proposed strategies to assess long-term safeness in pediatric populations,¹¹ such as the evaluation of safety information reported in the year following a labeling change (resulting from studies conducted under PREA or BPCA) and required specific pharmacoepidemiologic¹² or long-term studies for certain drugs. Unfortunately, these strategies are not always easy to implement, mainly because some adverse events (e.g., decreased final height in children exposed to medications that can affect growth, such as steroids) can take many years to appear or can be difficult to predict (products labeled on the basis of short-term studies but used on a long-term basis for chronic conditions such as asthma or diabetes). Funding for multiyear (or multidecade) cohorts is extremely difficult to obtain, and pharmaceutical companies are quite reluctant to plan such long-term studies.

As a consequence of all these difficulties, only a minority of commonly used drugs have been thoroughly evaluated or labelled for pediatric use. This has led to a widespread off-label drug use, that is, use of drugs outside the term of the Summary of Product Characteristics. In a hospital setting, 36% to 67% of children receive an unlicensed or an off-label prescription.^13–15 This proportion increases in the intensive care unit (up to 90% of children). Threats to efficacy and increased toxicity are the result.^{16, 9} For example, improper dosing of artemisinin combined therapy is an important determinant of inadequate efficacy and contributes to the development of resistance in malaria.¹⁷ Young children proved to be among the groups at highest risk for underdosing and corresponding insufficient drug exposure. Moreover, in a prospective pharmacovigilance protocol, off-label drug use was significantly associated with adverse drug reactions (relative risk, 3.44; 95% CI, 1.26–9.38), particularly when it was due to an indication different from that defined in the Summary of Product Characteristics (relative risk, 4.42; 95% CI, 1.60–12.25).¹⁸ Altogether, this suggests that early pediatric clinical trials can provide benefits to children, by helping improve dosing and detecting pediatric-specific adverse events.

However, despite the lack of pediatric clinical trials (and pediatric pharmacology data) described, comprehensive solutions for this problem are difficult to develop and implement.¹⁹ Certain strategies, such as the “carrot and stick” approach embodied by US and European legislation (e.g., BPCA and PREA) pushing for pediatric clinical trials, coupled to (sometimes significant) economic incentives for those pharmaceutical companies that do conduct these studies, have proven moderately effective but mostly for new drugs.²⁰ Older (and widely used) medications, frequently produced by generic companies, have been left in limbo, because their current manufacturers are loath to pick up the tab for conducting pediatric clinical trials for medications not very profitable³ (e.g., vancomycin,²¹ carbamazepine, and many others). A particularly problematic example of these pediatric medications is that of drugs for neglected diseases (e.g., leishmaniasis, Chagas disease, strongylodiasis), many of which are not even produced or sold in North America or Europe, and for which funding for clinical trials is close to non-existent. Some of these drugs did benefit from economic incentives after they were redeveloped and registered in the United States and Europe (e.g., benznidazole and nifurtimox for Chagas disease),²² which involved performing clinical trials in children, but these examples are a minority, and have in some cases involved practices that did not end up benefiting pediatric patients (e.g., miltefosine, for leishmaniasis, was studied for registration in the United States, but after obtaining a priority review voucher worth over 100 million dollars, the producer discontinued the medication²³). Research funding to carry out pediatric studies for neglected disease therapeutics has significantly dried up in recent years, and availability of pediatric clinical pharmacologists (especially with a focus on neglected diseases) is very limited. Unless these problems are addressed (e.g., by increasing training in pediatric clinical pharmacology and pediatric clinical trials in neglected diseases, and by ensuring abundant and reliable funding for such studies), the future will not look very different from now.

Most of the early phase clinical trials found in this review were largely done in high-income countries, demonstrating a disparity in how drug discovery and development is currently carried out. It is well known that those countries that develop and study novel treatments are among the first to benefit from these medications. This unequal geographic distribution of clinical research aggravates the vulnerability of children from lower-income countries who are not only underrepresented in clinical trials, but also experience an even longer delay in getting innovative, safe, and effective treatments. This has already been addressed by other authors. Marshall et al,²⁴ with a quantitative approach, showed that research priorities are not well optimized to reduce the global burden of disease.¹ They found that most randomized clinical trials are produced by highly developed countries, and the health needs of these countries have been, on average, favored. MacLeod et al²⁵ gathered suggestions for approaches that could help to achieve more effective ethical consideration when conducting trials in low-middle income countries. Despite these geographic disparities, we found—and even though we believe that most of these studies must have been published in English—the fact that we did not screen literature without an abstract in English is a limitation of our conclusion.

Fortunately, the idea that children must be protected from non–evidence-based interventions and from substandard treatments has been growing slowly but steadily in the past few years. The role of the regulators has been key in this shift.^{8, 9} There has been significant progress in recent years in terms of conducting clinical trials in children for an increasing number of disorders beyond oncology. Further analysis may be required to extrapolate these findings to other pediatric conditions and phases of clinical trials. However, even with the added difficulty of conducting clinical trials in children, the analyzed studies were largely successful and very few had serious adverse side effects.²⁶ The number of patients who withdrew during the studies was also surprisingly low. Specifically, none of the vaccine studies, which are conducted in healthy subjects, had any treatment-related side effects, emphasizing that clinical vaccine trials can be conducted safely in the pediatric population, a finding particularly relevant given the current COVID pandemic and controversies regarding delays in starting pediatric-specific COVID vaccine trials. This is especially critical when taking into account that potential stakeholders for vaccine studies, such as parents, are often very reluctant to accept enrolling healthy children in clinical studies that might be perceived as potentially dangerous.

Conclusion

The widespread concerns about children's safety when they participate in early phase clinical trials seem disproportionate, based on our findings. The data confirm that these studies can be conducted safely, and that their results can contribute significantly to pediatric pharmacotherapy.

ABBREVIATIONS

AIDS

acquired immune deficiency syndrome

BPCA

Best Pharmaceuticals for Children Act

CI

confidence interval

HIV

human immunodeficiency virus

LMIC

low-to-middle-income country

RCT

randomized clinical trial

SPC

summary of product characteristics

US

United States

FDA

Food and Drug Administration.

Funding Statement

Disclosures. The authors declare no financial interest in any product or service mentioned in the manuscript, including grants, equipment, medications, employment, gifts, and honoraria. The authors had full access to all the data in the study and take responsibility for the integrity of the data and the accuracy of the data analysis. Deejesh Subramanian received an Undergraduate Summer Research Internship from Western University to carry out this work.