Pediatric Antibacterial and Antifungal Trials From 2007 to 2017
BACKGROUND AND OBJECTIVES: The impact of the Best Pharmaceuticals for Children Act (BPCA) and the Pediatric Research Equity Act (PREA) on pediatric antibacterial or antifungal drug trials is unknown. Our objective was to identify and characterize trials conducted under the BPCA and/or the PREA.
METHODS: Pediatric antibacterial and antifungal drug trials with industry or US federal funding registered in clinicaltrials.gov from 2007 to 2017 were identified. Those conducted under BPCA and/or PREA were identified through US Food and Drug Administration and National Institute of Child Health and Human Development databases.
RESULTS: Of 17 495 pediatric trials registered on clinicaltrials.gov between October 2007 and September 2017, 122 systemic antibacterial or antifungal drug trials with industry or US federal funding were identified. Of these 122 trials, 98 (80%) involved antibacterials only, 23 (19%) antifungals only, and 1 (1%) both antibacterials and antifungals. These represented <1% (122 of 17 495) of pediatric trials. Neither pediatric antibacterial nor antifungal drug trials commonly enrolled neonates 0 to 30 days old (30% [30 of 99] vs 42% [10 of 24], respectively). Pediatric antibacterial and antifungal trials were commonly industry funded (79% [78 of 99] and 83% [20 of 24], respectively). In total, 65% (79 of 122) of pediatric antibacterial and/or antifungal drug trials were conducted under BPCA and/or PREA. Researchers in trials conducted under BPCA and/or PREA, relative to non–BPCA and/or PREA trials, more often collected pharmacokinetic data (70% [55 of 79] vs 26% [11 of 43]).
CONCLUSIONS: Although the majority of pediatric antibacterial and/or antifungal drug trials were conducted under BPCA and/or PREA, the overall number was low. Greater effort is needed to stimulate such trials.
- AACT —
- Database for Aggregate Analysis of ClinicalTrials.gov
- BPCA —
- Best Pharmaceuticals for Children Act
- CNS —
- central nervous system
- CTTI —
- Clinical Trials Transformation Initiative
- FDA —
- Food and Drug Administration
- ID —
- infectious disease
- IQR —
- interquartile range
- MeSH —
- Medical Subject Heading
- NIH —
- National Institutes of Health
- NLM —
- National Library of Medicine
- PREA —
- Pediatric Research Equity Act
What’s Known on This Subject:
Antibacterial and antifungal drug trials for pediatric indications are thought to be a major unmet health need, although firm data are lacking. Furthermore, the role that US federal legislation has had in impacting such trials is unknown.
What This Study Adds:
Federal legislation is likely having an impact on pediatric antibacterial and antifungal drug trials, although such trials still represent <1% of all pediatric trials in the study period. There were particular deficiencies in several areas, including enrolling neonates.
Pediatric drug development is complicated by logistical, technical, ethical, and economic barriers.1 These barriers, coupled with increasing antimicrobial resistance, have made pediatric antibacterial and antifungal drug development a high priority in public health. In addition to the need for new antimicrobial agents,2 we have an insufficient understanding of the efficacy, safety, and proper dosing of many antibacterial and/or antifungal agents in the pediatric population.3,4 To address this need, US federal legislation, including the Best Pharmaceuticals for Children Act (BPCA) and the Pediatric Research Equity Act (PREA), was enacted with the goal of increasing the study of drugs in children.5,6
BPCA was signed into law in 2002 and provided mechanisms for studying on-patent and off-patent drugs in children. For on-patent drugs being tested for pediatric use, BPCA grants an additional 6 months of marketing exclusivity that extends patent protection if studies in children requested by the US Food and Drug Administration (FDA) are performed. For off-patent drugs, the BPCA directed the US National Institutes of Health (NIH) to identify and prioritize drugs that need study and to conduct studies on priority drugs if the pharmaceutical industry sponsor declines to do so. Taken together, the on-patent and off-patent provisions of BPCA provide incentives for pediatric drug development.
PREA was signed into law in 2003. In contrast to BPCA, which is voluntary, PREA authorizes the FDA to require pediatric drug or biologic studies to ensure that the products are safe and effective in children.7 The FDA can require such pediatric studies in products for which a new drug application is submitted, and the drug and/or biologic is either likely to be significantly used in children or may provide meaningful benefit over existing treatments.7 Thus, BPCA and PREA together constitute a “carrot-and-stick” approach to stimulating pediatric drug trials.
However, the impact of BPCA and PREA on pediatric antibacterial and antifungal drug development is unclear. To assess this, we characterized the landscape of pediatric antibacterial and antifungal trials conducted under BPCA and/or PREA compared with other pediatric antibacterial and/or antifungal trials.
Creation of the Pediatric Antibacterial and Antifungal Drug Trials Data Set
The pediatric antibacterial and antifungal drug trial data set was created through a 2-step process. The first involved creating a larger infectious diseases (IDs) study data set, as previously described.8 We used the October 16, 2017, version of the Database for Aggregate Analysis of ClinicalTrials.gov (AACT) to identify trials that were registered on clinicaltrials.gov from October 1, 2007, through September 30, 2017. AACT is a relational database (PostgreSQL) that was developed and is maintained by the Clinical Trials Transformation Initiative (CTTI). It contains all information about studies that have been registered on clinicaltrials.gov since its inception in February 2000, and it is updated daily with content downloaded from clinicaltrials.gov. The database is publicly available in the Cloud with access information and documentation provided on the CTTI’s AACT Web site.9
For this study, we focused on trials that were registered after September 27, 2007, which marked the beginning of US requirements for registering many trials that are subject to FDA regulation.10 Using the registry’s “study type” field, we identified interventional and observational trials that were registered during the 10-year study interval. Using the registry’s “maximum age” field, we identified trials that were restricted to participants ≤21 years old.
Next, we grouped studies by clinical specialties within this derivative database.11 To identify trials relevant to IDs, we focused on 2 characteristics, condition and intervention, which were defined by data submitters or linked to Medical Subject Heading (MeSH) terms that were generated by a National Library of Medicine (NLM) algorithm based on the 2017 MeSH thesaurus. Of 10 466 MeSH condition terms reviewed, 991 (9%) were related to IDs. Not all sponsor-entered conditions could be linked to MeSH terms; therefore, free-text condition terms appearing in ≥5 trials were manually annotated for relevance to IDs. This identified 345 of 5140 possible free-text condition terms as relevant to IDs (7%). The second characteristic used to identify ID-relevant trials was the intervention term. We focused on intervention terms that were linked to MeSH terms generated by an NLM algorithm that appeared in ≥4 trials, and we reviewed these for relevance to IDs. With this, we identified 309 ID-related terms out of 2101 unique intervention MeSH terms reviewed (15%). We identified trials with at least 1 ID-relevant term in the NLM-generated MeSH condition field, the submitted free-text condition field, the NLM-generated MeSH intervention field, or the submitted intervention name field. Trials were manually reviewed (J.T.T., S.P.B., I.J., and W.E.Y.) to exclude non-ID studies, pediatric ID studies that were not focused on antibacterial or antifungal drugs, trials of nonsystemic drugs, and trials that were withdrawn before enrollment. Manual review was used to further specify the study parameters of interest, including infection type, outcome data type, safety data, and pharmacokinetic data (when applicable). For all identified antibacterial and/or antifungal drug trials, 2 investigators from the US FDA (S.N. and J.F.) used the FDA public listing of drug approval information,12 the NIH public listing of studies conducted under BPCA,13 and FDA administrative databases to determine whether trials were conducted as a required pediatric study under PREA, in response to a pediatric written request under BPCA, or were supported by the NIH BPCA program. To appropriately gauge the impact of BPCA and PREA on pediatric antibacterial and antifungal drug trials, only studies with industry or US government (NIH or otherwise) funding were included in this study. All other studies (eg, those funded by a university or non–US federal agency) were excluded.
Because clinicaltrials.gov does not require funding information, we derived the funding source on the basis of the “lead sponsor” and “collaborator” fields, as demonstrated in Supplemental Table 4.14 The enrollment number was reported for each trial. For active trials for which recruitment had not been completed, anticipated enrollment was recorded. Study duration was calculated as the time from the study’s start to completion (or planned completion) of the primary end point.
Reporting of results on clinicaltrials.gov was examined as of February 28, 2018. To determine whether trials had published manuscripts in the scientific literature, we used each trial’s clinicaltrials.gov identifier to search the PubMed database (www.ncbi.nlm.nih.gov/pubmed) and Google Scholar database (scholar.google.com) in February 2018. The time to results reporting was the time from completion of a trial’s primary end points to the reporting of results. The date of completion of the trial’s primary end points was defined on clinicaltrials.gov as the final collection date for primary outcome measures for all enrolled participants. Time to reporting of results on clinicaltrials.gov and the time to publication of results in the scientific literature were analyzed separately and as a composite. The date of results reporting was defined as the earliest date when results were either posted on clinicaltrials.gov or published in scientific literature. Evaluation of results reporting was restricted to completed or terminated trials.
Frequencies and percentages were calculated for categorical trial characteristics; means, SDs, medians, and interquartile ranges (IQRs) were calculated for continuous characteristics. The cumulative percentage of trials with completed or terminated follow-up was reported through 108 months from the study start by using Kaplan-Meier estimates. The cumulative percentage of studies in which researchers reported results was reported through 96 months from the primary completion date because researchers in no trials reported results beyond this 8-year cutoff. For the Kaplan-Meier analyses, follow-up time was censored at the end of the study period (September 30, 2017, for study completion and/or termination and February 28, 2018, for results reporting). Missing values were excluded from the analysis. Because the data set includes all studies in the population of interest, analytical methods were descriptive, and no statistical inference was performed. Analyses were performed at the Duke Clinical Research Institute in Durham, North Carolina, by using SAS 9.4 (SAS Institute, Inc, Cary, NC).
Pediatric Antibacterial and Antifungal Drug Trials Data Set
A total of 256 544 trials were identified in the AACT database on October 16, 2017, and 17 495 (7%) were pediatric trials registered on clinicaltrials.gov between October 1, 2007, and September 30, 2017 (Fig 1). Of these, 122 (of 17 495 [1%]) systemic antibacterial or antifungal pediatric drug trials were identified (Supplemental Table 5). Of these 122 trials, 98 (80%) involved antibacterial drugs only, 23 (19%) involved antifungal drugs only, and 1 (1%) involved both antibacterial and antifungal drugs. After comparing this list of 122 trials to the BPCA and PREA databases, we identified 79 (65%) that were conducted under those provisions. Of these 79 trials, 27 (22%) were conducted under BPCA only, 29 (24%) were conducted under PREA only, and 23 (19%) were conducted under both BPCA and PREA.
Source of Infection
Antibacterial drug trials most commonly involved pneumonia (20% [20 of 99]), intraabdominal infections (10% [10 of 99]), or skin infections (9% [9 of 99]), and they less commonly involved central nervous system (CNS) infections (3% [3 of 99]) or bacteremia (1% [1 of 99]; Table 1). Antifungal drug trials primarily involved candidiasis (42% [10 of 24]) or invasive fungal infections (other than candidiasis; 38% [9 of 24]). Among trials conducted under BPCA and/or PREA, the most common indications were pneumonia (14% [11 of 79]), intraabdominal infection (13% [10 of 79]), and candidiasis (13% [10 of 79]; Table 1).
Trial Enrollment Criteria, Size, Location, and Funding
When stratifying trials by enrollment ages, we observed that most studies enrolled infants (31 days to 2 years old). Specifically, 78% (77 of 99) of antibacterial drug trials and 92% (22 of 24) of antifungal drug trials enrolled subjects in this age range (though not necessarily exclusively in this age range; Table 2). Trials enrolling neonates (0–30 days old) were less common (antibacterial: 30% [30 of 99]; antifungal: 42% [10 of 24]; Table 2). Of the 39 trials that enrolled neonates (1 trial was both an antibacterial and antifungal drug trial), 26 (67%) were conducted under BPCA and/or PREA (Table 2). The majority of antibacterial and antifungal drug trials were industry funded (79% [78 of 99] vs 83% [20 of 24]; Table 2). Of the 79 pediatric antibacterial or antifungal trials conducted under BPCA and/or PREA, 71 (90%) trials were industry funded, with the other trials being funded by the US government through the BPCA off-patent program (10% [8 of 79]).
Timing of Trial Start Date and Completion
When pediatric antibacterial and antifungal drug trials were stratified by year of registration on clinicaltrials.gov, we observed a greater number of trials that were registered from October 2007 to September 2008. This is likely because the legislative requirement to register most FDA-regulated trials on clinicaltrials.gov was September 27, 2007 (Supplemental Fig 3A). After this initial time point, there were no upward or downward trends in the number of antibacterial trials, although the number of antifungal trials trended downward. Similarly, the number of trials conducted under BPCA and/or PREA remained stable over the study period, whereas the number of trials not conducted under these provisions trended downward (Supplemental Fig 3B). The median time to completion of antibacterial trials (27 months [IQR 16–41]) was shorter than that of antifungal trials (32 months [IQR 23–49]; Fig 2A). The median time to completion of trials conducted under BPCA and/or PREA (31 months [IQR 21–39]) was longer than that of trials not conducted under BPCA or PREA (23 months [IQR 11–46]; Fig 2B).
Trial Outcome Measures
The most common primary outcomes listed as part of clinicaltrials.gov registration were clinical efficacy and pharmacokinetics in pediatric antibacterial drug trials (36% [36 of 99] in each case) and pharmacokinetics (50% [12 of 24]) in pediatric antifungal drug trials (Table 3). Independent of the primary outcome, pharmacokinetic data were more commonly collected in antifungal drug trials relative to antibacterial drug trials (63% [15 of 24] vs 53% [52 of 99], respectively). All trials that collected pharmacokinetic data did so from the plasma, whereas a smaller number collected from additional source matrices, including cerebrospinal fluid (5% [6 of 122]) and urine (2% [2 of 122]). Trials that were conducted under BPCA and/or PREA, relative to those that were not, more commonly collected pharmacokinetic data (70% [55 of 79] vs 26% [11 of 43], respectively; Table 3).
Publication of Pediatric Antibacterial and Antifungal Trial Results
For antibacterial trials that were completed or terminated as of September 30, 2017 (62% [61 of 99]), 38 (62%) reported results either on clinicaltrials.gov or through the scientific literature by the end of the study period. The estimated cumulative incidence of results reporting at 96 months after the completion of a trial’s primary end points was 45% for results reported on clinicaltrials.gov and 56% for results reported through the scientific literature (Supplemental Fig 4A). Among the pediatric antibacterial trials that reported results either on clinicaltrials.gov or in a scientific publication, the median time to reporting was 13 months (IQR 11–27). For antifungal trials that were completed or terminated as of September 30, 2017 (63% [15 of 24]), 12 (80%) had reported results either on clinicaltrials.gov or in the scientific literature by the end of the study period. The estimated cumulative incidence of results reporting at 96 months after the completion of a trial’s primary end points was 64% for results reported on clinicaltrials.gov and 69% for results reported through publication in the scientific literature (Supplemental Fig 4A). Among the pediatric antifungal trials that reported results either on clinicaltrials.gov or in the scientific literature, the median time to reporting was 14 months (IQR 12–29).
Studies that were conducted under BPCA and/or PREA more frequently reported results by 96 months of follow-up (82%) relative to studies without BPCA or PREA involvement (56%; Supplemental Fig 4B). Although fewer non–BPCA and/or PREA trials had reported results, among those that did, the median time to results reporting from the completion of the primary end point (12 months [IQR 7–15]) was shorter than for trials that were conducted under BPCA and/or PREA (19 [IQR 12–28]). In general, there were no major differences in the type of outcomes reported on clinicaltrials.gov versus in the scientific literature.
The development of pediatric antibacterial and antifungal drugs is a well-recognized public health priority, and this study is the first to describe how BPCA and PREA are impacting such drug trials. Researchers in previous studies have used clinicaltrials.gov to explore the overall landscape of clinical trials14 and the subsets of pediatric interventional trials,15 ID trials,8 and pediatric antimicrobial trial designs.16 But the impact of US federal legislation in promoting pediatric antibacterial and/or antifungal trials has been unclear. This study revealed several important findings.
The overall number of pediatric antibacterial and antifungal drug trials that are industry or US federal government funded was remarkably low; they represented <1% of all pediatric trials. The small number is particularly striking relative to the global morbidity associated with pediatric ID; the World Health Organization estimates that 23% of child mortality stems from pneumonia and neonatal sepsis alone.17 Additional areas of particular unmet need include trials involving bloodstream infections and CNS infections.
Despite the overall low number of pediatric antibacterial and antifungal drug trials that are industry or US federal government funded, US federal legislation may be having an impact; 65% of the trials were conducted under BPCA and/or PREA. BPCA and PREA may also be having an impact on results reporting because 82% of trials conducted under at least 1 of the federal acts had reported results, relative to only 56% of non–BPCA and/or PREA trials. The timing of results reporting from the completion of the primary end point was relatively late in trials conducted under BPCA and/or PREA (median 19 months) and non–BPCA and/or PREA trials (median 12 months), however. We believe the overall low number of pediatric antibacterial and/or antifungal trials stems from multiple factors, including the low number of children who were eligible for such studies as well as the ethical and logistical challenges of enrolling children with acute medical issues.
Pediatric antibacterial and antifungal drug trials conducted under BPCA and/or PREA were most often industry funded. However, 8 trials were funded by the NIH through the BPCA off-patent program, indicating the importance of this program in the study of antibacterial and antifungal drugs in children. Trials that were not conducted under BPCA and/or PREA generally fell into 1 of several categories. They included industry postmarketing to evaluate a new indication, investigator-initiated studies funded through alternative means (eg, industry grant or NIH R01), trials requested by a non-FDA regulatory authority (eg, European Medicines Agency), or trials being done as part of the clinical phase of new drug development.
This study has several limitations. First, it is possible that trials involving both children and adults were inappropriately excluded because we limited our analysis to trials that only enrolled participants ≤21 years old. We do not suspect that a significant number of such trials were excluded, however, because the number of trials in this study (N = 122) is similar to that in a recent review of pediatric antimicrobial studies (N = 82) that was focused on a more limited set of clinical infection syndromes.16 Second, given that data from clinicaltrials.gov does not precede the passage of BPCA (2002) or PREA (2003), it is difficult to estimate the degree to which BPCA and PREA have directly stimulated pediatric antibacterial and antifungal trials. However, given the significant cost and resources necessary to conduct pediatric antibacterial and/or antifungal drug trials, we feel that the majority of trials conducted under BPCA and/or PREA would not have been done without the financial incentives of BPCA or the regulatory requirement of PREA. Third, this study was limited to pediatric antibacterial and antifungal drug trials that were either industry or US federal government funded. This was done to better examine the influence of US federal legislation because BPCA and PREA do not generally apply to trials outside these funding sources. A review of clinicaltrials.gov revealed 112 pediatric antibacterial and/or antifungal drug trials that were funded outside of industry or US federal sources. These additional trials often are focused on conditions and antibiotics that are typical of more resource-poor settings. Despite these additional trials, the conclusion that there are inadequate pediatric antibacterial and antifungal drug trials remains unchanged.
This study revealed a low number of pediatric antibacterial or antifungal drug trials over the study period that were funded by industry or US government sources. Approximately two-thirds of pediatric antibacterial and/or antifungal drug trials were conducted under BPCA and/or PREA, and researchers in such trials more frequently collected pharmacokinetic data and reported results that were relative to non–BPCA and/or PREA trials. Nevertheless, approaches to further incentivize such trials and improve the reporting of results are needed. These findings are important to clinicians because health care workers not only initiate and implement pediatric clinical trials but are the front line in stimulating enrollment into and advocating for such trials. Furthermore, clinicians should be aware of evidence supporting the use of antibacterial and/or antifungal agents in this population. And when the evidence is lacking, it is important to understand efforts to mitigate this deficiency.
We thank all the members of the CTTI Pediatric Infectious Diseases Trials team for their expertise and insight.
- Accepted June 15, 2018.
- Address correspondence to Ephraim L. Tsalik, MD, Division of Infectious Diseases, Duke University School of Medicine, DUMC Box 102359, Durham, NC 27710. E-mail:
FINANCIAL DISCLOSURE: The authors have indicated they have no financial relationships relevant to this article to disclose.
FUNDING: Supported by the US Food and Drug Administration (grant R18FD005292; cooperative agreement U19FD003800). Views expressed in this publication do not necessarily reflect the official policies of the US Department of Health and Human Services, nor does any mention of trade names, commercial practices, or organization imply endorsement by the US Government. Partial funding was also provided by pooled membership fees from the Clinical Trials Transformation Initiative’s member organizations.
POTENTIAL CONFLICT OF INTEREST: Dr Benjamin has received grants and revenue from AstraZeneca, Cempra Pharmaceuticals, Shionogi Inc, The Medicines Company, and Cidara Therapeutics, which are unrelated to the current study. Dr Smith has consulted on antifungal trials for Astellas Pharma Inc; the other authors have indicated they have no potential conflicts of interest to disclose.
- ↵Pediatric Research Equity Act of 2003, 21 USC 355c
- ↵Best Pharmaceuticals for Children Act of 2002, Pub L No. 107–109, 115 Stat 1408
- Vanchieri T,
- Butler AS,
- Knutsen A
- Clinical Trials Transformation Initiative
- ↵Food and Drug Administration Amendments Act of 2007, Pub L No. 110–185, 121 Stat 823
- US Food and Drug Administration
- Eunice Kennedy Shriver
- National Institute of Child Health and Human Development
- World Health Organization
- Copyright © 2018 by the American Academy of Pediatrics