Parents considering whether to enroll a sick son or daughter in a clinical trial often face a barrage of conflicting emotions. On one hand, they hope that the experiment will lead to a breakthrough in treatment. On the other, they must deal with the uncertainty and fear that come from willingly exposing their child to an unproven therapy that could turn out to be ineffective or even more harmful than standard treatment. What parents may not anticipate, however, is that their child's contribution to scientific understanding could also be squandered on a study that is never completed or that fails to be published in the medical literature.
Such wasted effort is a strong possibility, according to a 2016 study. Harvard University researchers reviewed the more than 550 pediatric clinical trials that were registered with the U.S. federal government over a three-year period between 2008 and 2011 and found that more than 40 percent were never finished or that they were completed but never published even five years later. In total, more than 77,500 children participated in studies that contributed little or nothing to advance treatments for their illnesses because the research disappeared from scientific view. Apart from the loss of time, money and resources, these failures are also “tragic because we have so little clinical trial information on children to begin with,” says Florence Bourgeois, an assistant professor of pediatrics and emergency medicine at Harvard Medical School, who co-authored the study.
Efforts to improve the situation are under way. Researchers at universities and medical centers are studying their own workflows to uncover hidden roadblocks to enrollment in pediatric clinical trials. At the same time, the National Institutes of Health is bringing together experts from academia, industry and government to try to make trial results more widely available. Although progress has been gradual, proponents believe they are on the right track.
Hitting a Wall
For decades physicians who wanted to prescribe a particular medication for a youngster adjusted the dosage downward from the amounts given to adults based merely on the child's weight. But youngsters are not just pint-sized adults. Their physiology changes rapidly as they move from infancy through adolescence, which means that various drugs may trigger unpredictable effects in their body. Infants do not metabolize an anesthetic called propofol, for example, at the same rate as adults, which leads to toxic buildups in babies. Newborns, children and adults also break down antiseizure drugs differently, making it difficult to estimate safe doses on the basis of weight and size alone.
Whereas clinical studies of adults have taken place in the U.S. for well over 100 years, pharmaceutical companies did not regularly test their products in children until the 1990s, when the federal government began offering financial incentives to do so. The Food and Drug Administration Modernization Act of 1997 provided companies six months of additional patent protection for every drug they tested in children, and after that, a nascent pediatric clinical trial infrastructure began taking shape. Additional laws expanded on those offerings and created programs for pediatric drug development at the NIH.
Nevertheless, the problem of mismatched physiologies remains, and many physicians are understandably nervous about calculating the right dosages. But their reluctance is not the only challenge. After excluding trials that were still recruiting participants or had not yet begun enrolling them, the Harvard team found that the biggest reason why studies failed was that they could not attract enough children to participate in the first place.
Lack of enrollment is often a problem with adult trials as well, but pediatric investigations face unique challenges. Pediatricians in community practice may not have enough time to participate in clinical research, especially if they work far from academic hospitals where supportive resources for clinical trials are readily available. Moreover, there simply are not as many children as adults. Just 20 percent of the U.S. population is younger than 14 years of age. Furthermore, children are, thankfully, much less likely than adults to suffer serious illnesses. “So by definition, most childhood diseases are rare,” says Danny Benjamin, a professor of pediatrics at the Duke University School of Medicine.
Ronnie Guillet knows all about enrollment challenges from firsthand experience. A professor of neonatology at the University of Rochester, Guillet had to shut down a clinical trial in 2014 that would have clarified how long newborn infants should be treated if they develop seizures. About one in 200 babies suffers seizures within days of birth, and doctors disagree on how long to give those infants antiseizure medications. Some continue treatment only until the seizures stop. Others prolong the drugs for several weeks or even months in an attempt to keep the seizures from recurring.
Guillet worried that extended treatment might harm an infant's brain, although she also knew that repeated seizures can cause neurological damage. The only way to know how to balance potential benefits and risks was to conduct a clinical trial. Thus, after randomly assigning infants to one approach or the other, she planned to monitor their neurological development. She needed to enroll 250 babies within two years to have a large enough sample size to feel confident that her results were statistically reliable.
But parents—and, somewhat surprisingly to Guillet, neonatal nurses—were skeptical. “Some worried about putting kids on the extended therapy, and others worried about the placebo,” Guillet says. “They had similar fears but for opposite reasons.” She had to cancel the study when only 13 children had enrolled after a year, and her underlying question remains unanswered.
Attracting the required number of participants for a study does not, however, guarantee it will be published or that the results can be widely shared. Drug companies, for example, have a good track record for enrolling enough children to finish their investigations. But fewer of their results actually appear in a scientific journal by comparison with academic trials paid for by the NIH or other independent organizations. “Drug companies care about publication only if the results support their commercial goals,” Harvard's Bourgeois says. In other words, they tend to publish only the studies with positive results. As she and her colleague wrote in their 2016 paper, they consider nonpublication a “violation of the ethical imperative to share results of trials that involve human subjects.” Researchers at universities, on the other hand, live or die by their publication record, Bourgeois explains, so “you'd expect that the academic trials would generate more published studies, and that's what we saw.”
A Matter of Trust
Making improvements in a complex process such as a clinical trial generally requires greater coordination and collaboration among several groups of people. Last fall the NIH launched a new Trial Innovation Network that brings together industry, regulatory agencies, universities and other organizations to increase the number of underserved populations—including children—who are helped by medical studies. Among the challenges the network has identified is the need for better communication among the professionals who try to enroll new participants in a trial—known as recruiters—and the children's caregivers.
“Recruiters generally don't take enough time to understand the day-to-day circumstances and motivations of the populations they're trying to enroll,” says Yvonne Joosten, who directs one of the network's newly funded programs, based at the Vanderbilt University Medical Center. Together with colleague Tiffany Israel, Joosten has begun coordinating community meetings on behalf of clinical trial investigators before their studies are launched. During these “listening sessions,” as she calls them, Joosten talks with families to learn more about the barriers that might keep them from participating. Transportation to and from study sites could be a deterrent, or parents might feel there is no value in getting involved should their children be assigned to a control group. “Researchers aren't obligated to take advice from community members, but those that do are often happy with the results,” she says.
On the publication front, federal agencies are flexing some regulatory muscle to ensure that investigators make their results available for others to see—whether or not they appear in a scientific journal. Investigators in the U.S. are required to register new trials and post final results on a searchable NIH database found at ClinicalTrials.gov. And those that do not can face stiff penalties under a law known as the Food and Drug Administration Amendments Act of 2007. The FDA can fine noncompliant drug companies $10,000 a day, and the NIH can pull support from its sponsored scientists. But in practice, not one company or NIH-funded scientist has ever been penalized. According to Jerry Sheehan, assistant director for policy development at the National Library of Medicine, the 2007 law is full of ambiguities about which clinical trials had to register with the site and what kinds of results they had to report.
In an attempt to eliminate any loopholes, the Department of Health and Human Services (of which both the NIH and FDA are a part) issued a lengthy ruling that clarifies what is required in exhaustive detail. That rule became enforceable in January. Whether it provides clinicians with enough information to prescribe new medications—or to prevent unnecessary or even dangerous treatments—for children remains to be seen. But this and other efforts should, at the very least, reduce the chances that the contributions made by the youngest participants of scientific studies will be lost.