No matter how Gretchen Stewart held her 3-month-old son, she couldn't console him. Gage Stevens fussed, cried and vomited often. Nothing his pediatrician recommended, including switching his formula, worked either. Finally, Gage's doctor told his mother about a specialist at Pittsburgh Children's Hospital who was testing the anti-heartburn drug Propulsid on children with Gage's condition. Maybe that would help, she recalls him saying.
Gage was enrolled in the study on May 11, 1999. Six months later, his grandmother gave him his nightly dose of medication and tucked him into bed. By morning, he was dead.
His family believes his participation in the drug study killed him. The local coroner seems to agree. Last month, Gage had the unfortunate honor of receiving a second death certificate. Instead of sudden infant death syndrome, originally listed as the cause of death last November, it now reads "cisapride [Propulsid] and cimetidine [Tagamet] prescribed therapy" and "trial drug study." After reexamining the case, Cyril Wecht, the Allegheny County, Penn., medical examiner, concluded that Gage died from cardiac arrhythmia -- or heartbeat abnormalities -- a side effect of Propulsid.
The death of Gage Stevens, who suffered from a relatively common condition, gastroesophageal reflux disease (GERD), has reopened the issue of whether it is right to test powerful drugs on children. While the death of 18-year-old Jesse Gelsinger brought criticism to the field of gene therapy -- and prompted Department of Health and Human Services Secretary Donna Shalala last week to announce increased federal monitoring of medical research -- Gage's death takes the debate to a deeper, more troubling level.
Should babies who can't even say "Mama," let alone decide to participate in a medical experiment, become pawns in the high-stakes game of drug research? Or is the death of a child simply the price that must be paid to determine if a drug is safe and effective for widespread use?
"One way we feel comfortable about research is telling each other, these people know the risks and they choose," says Arthur Caplan, director of the Center for Bioethics at the University of Pennsylvania Medical Center. "We can't feel as good about allowing experiments on children."
Most drugs are tested extensively on adults before the Food and Drug Administration approves them for general sale. Once a drug is on the market, a doctor may prescribe it for any patient, even a child. It is generally believed that only about one-fourth of medicines prescribed for children have been approved specifically for kids.
Amid growing concerns over the inadequacy of pediatric drug testing, federal legislation was passed three years ago to promote studies on kids. Since then, the number of vaccines and pediatric medicines in development has skyrocketed -- and so have the number of kids on the frontlines of research. Gage Stevens may be the first casualty of this new era.
There is no way to prove posthumously that Gage died because of heart problems, as the medical director of Children's Hospital and other experts point out. His case is also complicated by problems in the study protocol. The researchers led Gage's parents to believe that Propulsid was FDA-approved for kids, when it wasn't. According to the family's lawyer, the second page of the "informed consent" document noted that there had been deaths as a result of taking this medication, but said most of those occurred when Propulsid was mixed with an antibiotic. In fact, Propulsid had been linked to at least 80 deaths and 341 heart rhythm abnormalities. Nineteen of the deaths were patients under 19 years old.
"If there had been anything that had been conveyed to Gretchen to show that there was remotely a risk that her baby would have died, she would have not participated in the study," says Anthony D'Amico, an attorney for Gage's family. Children's Hospital has suspended the Propulsid study. The manufacturer has pulled the drug off the market.
Since the FDA Modernization Act was passed in 1997, the number of vaccines and medications in development for children has jumped 52 percent, from 146 to 217, according to a recent survey by the Pharmaceutical Research and Manufacturers of America. An additional 52 drug trials using children are set to begin shortly.
To determine if a drug is safe and effective, it must be tested in clinical trials, or rigorous scientific studies involving large groups of patients. The gold standard of drug testing is the controlled double-blind study: Volunteers are randomly sorted into a group that receives the experimental treatment, and a group that gets a placebo. Neither the subjects nor the researchers know who is in which group, until the end, which is why these studies are said to be "double blind."
Most drugs prescribed for children have been tested on and approved for adults, but not kids. The medical community has pushed hard to change this. Drugs affect small, young bodies differently than grown ones. Doctors are allowed to prescribe medication for patients "off-label" -- for purposes that have not specifically been approved. But many doctors say it is bad medicine simply to halve the dose of an adult drug and give it to a child. Inevitably, they say, kids will be put at risk. Either we limit and monitor the danger in a clinical trial, or experiment on the pediatric population at large through widespread off-label use.
"One could speculate that there are more deaths through the use of Propulsid off-label than through the use in research studies," says Dr. Robert Nelson, associate professor of pediatrics at the Medical College of Wisconsin in Milwaukee. "If doing research could mean that we could reduce the amount of deaths outside the studies, then that could be for the good for the children."
It was the widespread off-label prescription of Propulsid that prompted gastroenterologist Susan Orenstein to study the drug. She was skeptical of the pediatric community's embrace of the medication, which was routinely prescribed to children with GERD. She recruited 100 infants to be in a double-blind study: One group would get Propulsid; one group would get Tagamet, a top-selling drug for adults that blocks the production of acid; one group would get both drugs, and one group would receive a placebo.
"Rather than an investigator like this being portrayed as pushing the envelope, at times they can be pushing the brakes on things," says Dr. Donald Fischer, the medical director of Children's Hospital at Pittsburgh who oversees research there.
Gage initially ended up in the placebo group. When he failed to improve, his mother told researchers. In July, three months after he entered the study, they put him on Propulsid and Tagamet.
Gastroesophageal reflux disease can range from a minor annoyance -- a "spitty" baby -- to a serious illness. Stomach acid can come up to the esophagus and spill into the trachea, narrowing the airways and making it difficult to breathe. Children's Hospital officials say Gage was very sick, while his family contends that he would have eventually outgrown the condition, as many children do without prescription drugs.
In March, Jaansen Pharmaceutical pulled Propulsid off the market. "If [Orenstein] hadn't been doing this, and this child hadn't been in a study and was one of the thousands of patients getting it in the community, I'm not sure a red flag would have been raised," Fischer says.
But it is hard to imagine that Gage's family would find solace in this. On Mother's Day, almost a year after her baby was enrolled in the study, Gretchen Stewart struggled painfully to make sense of the loss, the family's attorney said.
The need for pediatric research is clear -- until it's on the back of your child. "Do you want to sacrifice your son or daughter for the next two or three generations?" asks Adil Shamoo, vice president of Citizens for Responsible Care and Research and professor of biochemistry at the University of Maryland School of Medicine in Baltimore. "I wouldn't. I would do a lot of things in my life but not this, not precious children to be used as guinea pigs."
Researchers are supposed to meet a high ethical bar in experimenting on children. With adults, a scientist may conduct a study that has tremendous risk and little therapeutic benefit, as long as the research is expected to yield valuable scientific information. Under federal regulations, children generally must benefit from participation in a study. If they don't, the study should pose minimal risk. In most cases, the experimental drug should be no more dangerous than other treatments for the child's condition.
"The general rule is that most trials on children ought to help those in the study," says Dr. Joel Frader, chairman of the institutional review board at Children's Memorial Hospital and associate professor of pediatrics and medical ethics at Northwestern Medical School. "But there are some problems because if you make that the exclusive way to give drugs to children, then you have a problem with the introduction of drugs that might help others but are very unlikely to help the child."
Phase-one studies of experimental cancer drugs are an example of this. These are the first tests on a small group of patients, designed to find out if the drug is toxic or causes serious side effects. Participants have a slim chance of improving, and may be hurt by the sort of drug-related complication the study is intended to discover. Typically, children in phase-one cancer trials have failed all other treatments. Many are dying. Their parents often enroll them in the study in the desperate hope of a miracle.
"Some of the most difficult questions are with those who are very ill or are dying because their parents will agree to do almost anything to save their lives," says Leonard Glantz, associate dean of Boston University's School of Public Health and co-editor of the book "Children as Research Subjects." "It's very understandable but you want to be very cautious."
By far the most controversial pediatric study in recent years was not on severely ill children, but healthy kids. It was conducted from 1994 to 1995, on 34 boys, ages 6 to 10. All were black or Hispanic.
Evidence suggests that in adults, there is a link between lower levels of the brain chemical serotonin and aggressive behavior. Researchers at the New York Psychiatric Institute wanted to find out if this held true for kids. They recruited boys whose brothers were juvenile delinquents. Each subject received a single dose of fenfluramine, which helps researchers measure a hormone indicative of serotonin levels.
The study drew outrage and accusations of racism. Some parents say that they enrolled their sons in the research only because they thought it somehow might help the legal case of their older son. Clearly, the study was not intended to benefit the children, much less help the cases of their brothers.
But institutional review boards, or IRBs, the committees that approve human studies at any medical center, may give the green light to research without any therapeutic value -- so long as it doesn't pose more than minimal risk for the child. The psychiatric institute's IRB decided this study fit that bill -- and would produce important data on a pressing societal problem.
"The development of antisocial behavior in youth is a problem that we have few answers about the cause and no good solutions," says Dr. Timothy Walsh, who was co-chair of the IRB. "If we're going to figure out ways to help parents help their kids grow up in the way all parents want their kids to grow up, we've got to do research."
The federal government investigated and absolved the institute of wrongdoing. But years later the study is still heralded as an example of what happens when research on children is allowed to go too far. The mother of one of the boys in the study filed a lawsuit last December against the researchers and Columbia University, which is affiliated with the Psychiatric Institute.
"About two weeks after he was given the drug he started having sharp, painful headaches," Charisse Johnson told a congressional subcommittee holding a hearing in December on the adequacy of protection for subjects in medical research. "Then as the headaches became more unbearable, he started having anxiety attacks and hyperventilating. He would start gasping for breath as if he couldn't breathe, as with someone who was having an asthma attack."
When other medical explanations were ruled out, and she learned about fenfluramine's side effects, including heart valve damage, she began to suspect the study was to blame. Isaac has since been diagnosed with post-traumatic stress disorder.
"To this day my son continues to suffer the severe consequences of the reckless disregard for him as a human being by these experimenters. To them, he was just another guinea pig," she said. Fenfluramine, famous for its use as the popular anti-obesity drug fen-phen, was pulled off the market in 1997.
Walsh declined to comment on the case, but he says there is no evidence that a single dose of fenfluramine poses significant risk.
When children become the essential ingredient of adult goals, says Boston University's Leonard Glantz, the question arises: Are the researchers and the parents, for that matter -- acting in the best interest of the child? "Pure research is a gift that subjects give to investigators, saying 'You can use my body as a laboratory so you can get information,'" Glantz says. "In order for something to be given, you have to give your consent. Young kids can't do that."
Gage Stevens suffered one of the cruelest ironies of modern medicine -- a treatment intended to help him probably harmed him instead. Like Gage, all kids need medicine at some point. The ethical quandry for parents is as personal as a child's life. Nobody wants to give children a drug that has not been thoroughly tested. But what parent wants to gamble with her own child by enrolling in a study? As long as scientists keep churning out powerful, promising medications, a child's death may be the only way to hoist a red flag about a drug's dangers.