New study raises questions about early-stage cancer therapies for children

The bar for clinical trials involving children, by design, is set higher than it is for adults. Before a drug can be given to children in a Phase 1 trial, it must have already been tested in adults and shown some signals about its safety and clinical promise.

That, however, does not mean the outcomes will be any better, according to a new study.

Scientists conducted a systematic review and meta-analysis of Phase 1 clinical trials involving children with cancer and found, on average, only 1 in 10 will see their condition improve. About 1 in 50 children will die from complications related to the drug.

Those outcomes are roughly in line with the outcomes for adults who volunteer for Phase 1 trials in cancer, a result the researchers didn’t expect.

“The prospects of major benefit in pediatric Phase 1 studies are very low and the prospects of major harm are nontrivial,” said Jonathan Kimmelman, senior author of the study, which was published Tuesday in PLOS Medicine. “We shouldn’t take too much comfort in the fact we tested these drugs in adults [because] we’re not getting that much better of an outcome.”

Dr. Robert Truog, director of Center for Bioethics at Harvard Medical School, wasn’t surprised that children and adults fare much the same.

“I think it’s reassuring to know these [rates] are not substantially different from what we see in adults,” said Truog, who was not involved in the study. “It will also be useful for IRBs [institutional review boards] who review these studies, and it can now be included in the conversation that oncologists have with the parents of children with cancer. It will help them to make a more informed decision.”

Phase 1 trials mark the first time drugs are tested in people. Researchers seek approval from IRBs at hospitals or universities based on evidence from animal experiments in this first of three phases. The prospect of any benefit to the patient is the lowest at this point in the life cycle of the drug and uncertainty about the potential for harm is the highest.

Regulations from the Food And Drug Administration are in place to protect children in all phases of research, in general requiring risk to be offset by potential benefit, with finer adjustments to justify gradations on either side of the equation. That allows “more than minimal risk,” for example, if there is a potential for “direct benefit.”

Faced with making a decision about a Phase 1 trial, a parent’s desire to do something is very natural and very real, said Truog.

“It’s such a common part of our human nature. When our back is against the wall and somebody says, ‘Here’s something new that might work,’ the immediate reaction is to say, ‘Let’s do it,’” said Truog, who is also a specialist in critical care medicine at Boston Children’s Hospital.

At this early stage, the potential for harm is real while the benefits are only surrogate measures, said Kimmelman, an associate professor of biomedical ethics at McGill University. One such measure at Phase 1 might be tumor shrinkage, which does not necessarily translate into what people really care about: survival.

“The risks are real. When you have drug toxicity, that is not a surrogate endpoint,” he said. “At the end of the day, when you’re testing a new drug in any population, even in children, the prospect of it being beneficial and safe is extremely remote and I think we need to come clean with that, that even pediatric Phase 1 studies where we have these extra layers of protection really don’t have great risk-benefit balance.”

Kimmelman and other researchers at McGill scrutinized 170 studies involving 4,604 patients.

They do not take a position on what impact their results should have on regulations. Nor do they deny the need for research in children.

They do note several limitations in their study. Certain blood cancers, such as leukemia, had a better response to experimental drugs than solid tumors did, so the average might not be reflected in any one type of cancer. And the researchers did not conduct their own analysis of Phase 1 studies in adults, relying on other data to make their comparisons.

Still, they hope their analysis will help parents and doctors while also sparking discussion about the risks allowed under national and international regulations.

“Our findings could be used in consent forms to allow patients and their guardians to make more informed decisions about participation in Phase 1 trials,” said Marcin Waligora, an associate professor of philosophy and bioethics at Jagiellonian University Medical College and first author of the study. “In addition to being useful for parents, the study results could help researchers, ethics committees, and oncology trial designers.”

Truog of Boston Children’s said the study could be helpful in another way. It could change how people think about palliative care for a child who has exhausted treatment options and whose parents are weighing a Phase 1 trial. The study found that, on average, patients developed more than one serious, dangerous side effect from the drug being studied. While not fatal, these adverse events could diminish the child’s quality of life, he said.

“When would it be better for children to have good palliative care focusing on the life the child has left versus continuing to pursue experimental treatments that may introduce significant adverse events?” Truog asked. “These are all things parents need to take into account.”