David Siegel of the National Institute of Child Health and Human Development (NICHD) reviewed a list of pediatric-specific vulnerabilities relevant to exposure to CBRN agents and development of MCMs (Box 4-1). He expressed concern that children and pregnant women are often labeled as “special populations.” Children, Siegel countered, represent a very large segment of the population. They do not have “special needs,”even though they may be vulnerable. In addition, 10 percent of all women of child-bearing age are pregnant at any given time. This stereotype has resulted in research focused on adults, with studies on the “special populations” set aside for a later time, if funding permits. There has been a lack of funding at all levels of the MCM development process regarding children and pregnant women, Siegel remarked.
Another issue Siegel raised is off-label use of treatments. Regulatory issues have created special challenges for the deployment of appropriate biodefense medicines for children. In regular pediatric medical care, 50 to 75 percent of pediatric medications are used off-label. This lack of approved pediatric labeling for certain indications significantly affects federal jurisdiction and deployment of MCMs in the SNS, Siegel pointed out. He posed several questions for consideration: Should pediatric labeling be required for deployment of therapeutic agents to the SNS? Should Congress amend the EUA statute such that key MCMs lacking the requisite use labeling can be forward deployed?
Adding to the challenge is the widespread thinking that pediatric
- Children are known to be at greater risk following exposure to CBRN agents.
- As children are lower to the ground, they are exposed to an increased concentration of CBRN agents.
- With higher respiratory rates and lesser volumes than adults, a child will inhale a greater dose of agent.
- Children have smaller diameter airways, anatomic subglottic narrowing, omega-shaped epiglottic structure, relatively large tongue size, and less rigid ribs and trachea which make them more vulnerable to agent-induced pathology such as bronchospasm, copious secretions, and pulmonary edema.
- A child’s smaller mass alone reduces the dose of chemical agent required for toxic or lethal effects.
- Nerve agents penetrate the blood brain barrier more easily in children than adults, and children may only exhibit central nervous system (CNS) effects.
- Animal studies have shown that the lethal dose of nerve agent in an immature animal versus an adult animal is 10 percent.
- Young children, especially less than 4 years of age, are more prone to develop seizure disorders secondary to hypoxia or other CNS insult.
studies should not be done until all of the adult studies have been completed. In some cases, this is appropriate, but in other cases it is not, Siegel said. If animal studies are completed, it is not necessary to wait until phase III clinical trials are completed before starting juvenile animal studies. There are challenges to performing pediatric and obstetrical drug trials due to realistic (and nonrealistic) patient safety issues. In some circumstances, the pharmaceutical industry has been reluctant to produce therapeutic agents or devices for children because of a lack of profitability. This is being addressed to some extent, Siegel noted. There is also a shortage of researchers at all levels who are capable of performing studies with children or pregnant women.
Moving forward, Siegel supported efforts to establish an obstetrics/pediatrics section or working group at BARDA to examine the current contents of the SNS, be aware of up and coming MCMs, be part of the MCM prioritization process, and make obstetric/pediatric study recommendations for the necessary PK, efficacy, and safety data. FDA, he said, should be actively involved in this section, working proactively with academia and industry sponsors to expedite MCM development. Increased harmonization and sharing of data by FDA, industry, and academia is also
needed. Siegel called for expansion of the Best Pharmaceuticals Act for Children to cover MCM development for children. Other recommendations offered by Seigel included the following:
- Prioritize funding of a systems approach for pediatric formulation development.
- Encourage timely development of appropriate juvenile or pregnant animal models (and determine whether juvenile animals are really needed for a given specific indication).
- Increase development of pediatric virtual and organic modeling capabilities.
- Encourage development of pediatric biomarkers.
- Utilize the forthcoming central institutional review board (IRB) for disaster-related studies as a platform for prospective studies of children.
- Work closely with FDA to ensure that these studies are designed to collect the requisite information.
Pregnancy is the most dynamic period of human growth and genomic expression, making it a period of some vulnerability, said Christian Macedonia, medical sciences advisor to Admiral Mullen, the chair of the Joint Chiefs of Staff. Genomic expression, he noted, is not just about making proteins, and during pregnancy all aspects of the genome are in play. Pregnancy is also a period of immune modulation, including immune tolerance and immune suppression. However, immune suppression during pregnancy is not on par with, for example, that resulting from chemotherapy. While pregnancy is often referred to as a “delicate” condition, Macedonia emphasized that pregnant women are extremely durable. Pharmacokinetics in pregnancy is uniquely challenging. Increased volume of distribution, altered protein binding, increased glomerular filtration rate, and other maternal changes complicate dosing determinations.
From a more societal perspective, Macedonia pointed out that pharmaceuticals and vaccines are generally not tested in pregnant human patients. Drug safety in pregnancy is typically obtained through postmarket surveillance.
With regard to benefit-risk assessments, Macedonia noted that conventional wisdom has been that humans are rational creatures and make decisions based on what is the best outcome. However, studies in game theory and behavioral economics suggest that humans are not entirely rational and make decisions based on a calculus of what is not just the best outcome, but what is the most likely outcome. He described the “reg-
ulator’s dilemma” (a la the prisoner’s dilemma from game theory) as this:From which perspective does the regulator derive his or her assessment? Who are they regulating for (e.g., the physician, the maternal patient, the fetal patient, society, the manufacturer, the agency, themselves)? Regulatory decisions can have different risks and consequences for each of these stakeholders.
Macedonia offered several suggestions to help meet the MCM regulatory science needs of pregnant women. Open, transparent, and interactive public education on the benefits and risks of MCMs is needed. He called for leveraging the power of the social network, both as a means for education (e.g., though interactive apps) and for understanding the perspectives of the public. There is also a need for greater education of scientists and policy experts on the value of exploratory data methods (i.e., there are other valid options besides a randomized controlled trial). Finally, he recommended greater investment in point-of-use diagnostics to better define the at-risk population and the right dose of medication or vaccination needed.
In discussion, Pravin Jadhav of FDA’s CDER explained that FDA has a pediatric decision tree to guide decisions regarding what kind of data, and how much, is needed to approve a drug for pediatric use. There would be no difference, he said, between how the agency would handle drugs for pediatric MCM use compared to any other pediatric drug (although he acknowledged that MCM development may rely more on data from animals, as well as PK and systems biology data). As an example, he said that in 2009, the agency was writing the EUA for Peramivir for H1N1 influenza, and there were no data available on children from birth to 18 years of age. To address this, FDA reviewed data from other similar drugs that, like Peramivir, were eliminated through a renal route. Those data, along with what is known about the developmental biology of the kidney, allowed the agency to derive pediatric dosing recommendations. Jadhav noted that the EUA clearly states that the dosing was derived based on modeling and simulation. In 2010, data from children treated in Japan indicated that the prediction was quite accurate; in 2- to 9-year olds the dosing was off by 15 percent, and in ages 9 to 18 years it was on target.
Nancy Messonnier of CDC described the use of anthrax vaccines in children as a case study. Anthrax vaccine was licensed in a general use protocol for preevent vaccination. Following a bioterrorism event, it would be used in accordance with an EUA; however, as there are no data in children, the EUA would be limited to adults. The CDC Advisory Committee on Immunization Practices and the American Academy of Pediatrics Red Book Committee state that in the event of an anthrax event with inhalational exposure, a benefit-risk analysis should be made, and children should be given anthrax vaccine if warranted. CDC and FDA are
developing a strategic framework for how children would be vaccinated in an event, including how immunogenicity and safety data could be collected.
Messonnier expanded on the communication of benefits and risk raised by Macedonia, noting that in a public health emergency there is a need to communicate to large populations. There may not be opportunities to have one-on-one conversations with patients. When communicating to parents about the risks and benefits of vaccinating their children, how can we be clear and unambiguous, but also appropriately identify that there is a lack of data?
Lisa Mathis of CDER noted that due to the Best Pharmaceuticals for Children Act and the Pediatric Research Equity Act, among other developments, nearly 50 percent of products are now labeled for pediatric use, up from 25 percent. Unfortunately, she noted, there has not been that same level of advocacy in support of pregnant and lactating women. Mathis said that FDA’s proposed pregnancy and lactation labeling rule (currently being finalized) would remove the five letter categories from labeling and would provide more clear information about what is known about use in pregnancy, from both human and animal data.1
Robert M. Nelson of FDA said that the assumption that children are a vulnerable population, not only in the physiologic sense but also in the ethical sense, has resulted in additional protections for children who are enrolled in research. Unfortunately, he said, this has often resulted in a protection from research, rather than protecting children through research. The goal of research should be concurrent licensure for both adults and children so that children are not placed in the vulnerable position of receiving drugs off-label.
There are two general pathways in the additional protections for pediatric research. If there is not a prospect of direct benefit to the study subject, the risk of the intervention must be minimal. For higher-risk products under investigation, there must be a prospect of direct benefit to the subject. For example, much of the Pralidoxime data was generated through pesticide exposure (specifically, clusters in New York City where children ate rat poison imported from abroad, which was composed of organophosphates, not Coumadin). Data were generated by clinicians who were administering Pralidoxime off-label to these children. The decision to administer was based on clinical grounds, and the research (basically a blood test) was secondary, and a low risk to the patients.
During a public health event, there are many operational challenges to simultaneously conducting research: Where will the work be done?
1 See http://www.fda.gov/Drugs/DevelopmentApprovalProcess/DevelopmentResources/Labeling/ucm093307.htm (accessed June 9, 2011).
How do you pre-position assets around the IRB approval? These operational issues are solvable, Nelson said. Research in the preevent setting, where children are not at risk, is more challenging. There are some situations where neither pathway applies. In those cases, it was proposed that for studies that were scientifically sound and ethically appropriate, there could be a level of federal review. However, the decision whether to conduct the trial is a separate one from the parental decision whether to enroll their children in such a trial if there is no direct benefit to the child, Nelson noted.
Nelson added that although FDA does not have regulations governing research involving pregnant women, HHS human research regulations (45 CFR 46, Subpart B) provide for additional protections for pregnant women and fetuses. If an intervention is not for the direct benefit of the pregnant woman, then the risk to the fetus must be minimal. Again, according to Nelson, this regulatory framework constitutes a barrier to preevent studies of MCMs.
Participants discussed further the types of studies or data that would be needed for development of MCMs for children or pregnant women. Mathis said that “dose is everything.” A good deal of progress has been made for pediatrics because of the Best Pharmaceuticals for Children Act and the Pediatric Research Equity Act, but we need to commit more time and people to studying pediatrics, earlier on in the process. For pregnancy, it would be reasonable to assume that efficacy could be extrapolated, but PK data are most urgently needed. Nelson advocated for an open public deliberative process around both the science and ethics of conducting preevent trials.