How unbelieveable is this? Parents be warned...

DANGEROUS DOSAGE

To make pediatric medicine safer, thousands of children are being used to test drugs originally designed for adults. Tragically, the side effects can sometimes prove deadly.

By Alice Dembner, Boston Globe Staff, 2/18/2001

First in a series of occasional articles on medical trials involving children.

Gage Stevens might have taken his first steps that bleak morning in November 1999. And by his first birthday, he would probably have outgrown his one persistent ailment - acid reflux. Instead, he lay in the Allegheny County, Pa., morgue.

His mother, Gretchen Stewart, had taken him to a specialist at Children's Hospital of Pittsburgh the spring before he died, seeking to ease the baby's crying as the acid irritated his esophagus. Dr. Susan Orenstein recommended Gage enter a study she was running to find ''the safest, most effective treatment for esophagitis in infants'' using the ''best available'' medications, according to the consent form Stewart signed.

But the experimental treatment triggered a heart rhythm disturbance that killed the 9-month-old, the coroner found. By that time, the key drug, Propulsid, had also been linked to many adult deaths.

''They told me they were just trying to see how effective it was. Had I known it was a dangerous drug, I would never have let him take it. Who in their right mind would?'' said Stewart of Homestead, Pa., who has filed a wrongful death suit against the doctor, the hospital, and the drug's maker, Janssen Pharmaceutica.

Medical experiments across the country appear to have killed at least eight children and subjected hundreds more to harmful side effects in the last seven years, according to an investigation by The Boston Globe. These tragedies, ethicists suggest, are a harbinger of troubles ahead as a federal push to test more medicines in children kicks into high gear this year.

An estimated 45,000 children are participating in industry-sponsored testing of new drugs, up from about 16,000 in 1997, according to Christopher-Paul Milne, senior research fellow at the Tufts University Center for the Study of Drug Development. ''Although researchers are looking pretty hard at the ethical issues, there's greater risk if you increase the number of children being studied,'' he said.

The rush to experiment on children has been triggered by federal initiatives intended to improve the safety of drugs for children. Three-quarters of the medicines used in children have never been fully tested for safety and effectiveness in them, including heavily prescribed drugs such as the asthma medication Albuterol. And children have been hurt when doctors lacking that information prescribed inappropriate doses of drugs marketed for adults.

In response to appeals from pediatricians, since late 1997, Congress has offered drug companies a lucrative six-month extension of patents on drugs already on the market if they test them on children. Also, as of last December, Food and Drug Administration policy requires that any new drugs be tested on children as well as adults.

Drug companies are already pouring an estimated $1 billion a year into pediatric testing, according to CenterWatch of Boston, which tracks the industry. The additional sales of a single blockbuster drug during an extra six months without competition from cheaper generics could offset that entire investment.

The increase in pediatric research, which builds on a longer tradition of testing children in cancer and AIDS research, has brought some benefits. Since 1997, 14 drugs have been newly labeled with instructions for use in children as a result of the push, while studies on dozens more have been completed and are awaiting review by the FDA. Many of the tests are safe and ethical.

But an examination of research in children since 1994 shows that the potent combination of vulnerable children, ambitious researchers, potential profits, and weak oversight can hold great peril for those children. From deadly infections in Memphis to a troubling misdiagnosis in Arkansas, children have suffered in the name of science.

In the case of Propulsid, the manufacturer, the doctor, and Children's Hospital deny that the drug caused Gage's death.

''We believe it's not possible to establish a direct causal relationship with the administration of the drug. We learned that the infant had been brought to the ER several days prior [to his death] with serious respiratory illness,'' said Janssen spokesman Greg Panico.

However, county coroner Cyril Wecht is convinced that a reaction to the drug - not a respiratory problem - killed the child. Although autopsy can never prove arrhythmia, he said he ruled out all other possible causes.

''We're satisfied that ... we are not acting prematurely, unwisely, unfairly, or unscientifically,'' said Wecht, who is also a nationally known commentator on high-profile cases.

Moreover, Propulsid was linked to potentially dangerous heart rhythm problems even before it was approved for adult use in 1993. By 1997, as doctors increasingly used their discretion to prescribe Propulsid to children despite the lack of federal approval, the FDA stepped in. Armed with reports of cardiac deaths of three young children taking Propulsid, regulators suggested that Janssen test the drug in children or provide a stronger warning label.

Janssen provided the warning label but held off on its own large trials, instead providing the drug free to researchers, including Dr. Orenstein in Pittsburgh. Orenstein enrolled about 100 children in her study, which compared Propulsid given with and without another drug, Tagamet, against a placebo.

Neither Orenstein nor her lawyer responded to requests for interviews and hospital officials declined to explain why the study was conducted in a way that seemed contrary to the FDA warning that Tagamet might increase Propulsid's heart risk.

In the months after Gage's death, the FDA received reports of more than 100 cardiac deaths in people taking Propulsid, including 19 children whose doctors had prescribed it. Janssen pulled it from the market last July.

''The country is still learning how to study children ethically,'' said Robert Ward, chairman of the American Academy of Pediatrics Committee on Drugs, stressing that he could not comment directly on the Propulsid case.

Cancer centers have been at the forefront of clinical testing in children, with long-standing research programs for all forms of the disease. But even they may not set the best example, despite good intentions.

Three children died of complications in an experiment at the internationally known St. Jude Children's Research Hospital in Memphis in 1998 and 1999, as researchers tried to improve the five-year ''cure rate'' for acute lymphoblastic leukemia, normally one of the most treatable forms of the blood cancer.

The children, whom the hospital declined to identify, died of infections or seizures apparently caused by the first phase of the multifaceted drug and radiation experiment, according to Dr. William Evans, the hospital's executive vice president. The death rate in the first phase of the trial, three of 53 children, was double the usual rate.

Inspectors with the federal research protection office, acting on a complaint, faulted the hospital for failing to report the deaths promptly both to the government and to the hospital's ethical review board, so they could decide if other children's lives were at risk.

An independent review, required by the government, later concluded that the hospital did not unjustifiably delay reporting the deaths and tried to adjust the experiment to make it safer. Nonetheless, with the deaths still unexplained, the hospital abandoned the study and agreed to hire an ombudsman for patients in future studies.

''We have since modified the treatment and had no problems,'' said the lead researcher, Dr. Ching-Hon Pui, who declined to talk in detail about what happened, saying, ''I still feel bad and I don't want to refresh my memory.''

Even when cancer researchers know that an experiment may increase the child's risk of death, it's not clear that they always fully inform the children or their parents.

Two children died suddenly and two more suffered life-threatening infections in the first phase of a leukemia experiment conducted in 2000 - nine years after researchers at Dana-Farber Cancer Institute in Boston warned of an unusually high rate of death and infection in similar chemotherapy.

Among the first 32 children enrolled in the Pediatric Oncology Group study at sites across the United States, Canada, and Europe, a 10-year-old girl died from a fungal infection and a 5-year-old girl succumbed to a septic infection. In addition, a 12-year-old went into shock and a 19-month-old suffered a brain inflammation, according to a study report.

Researchers at Dana-Farber had warned about just such a tragedy from using the same drug, dexamethasone, in a four-drug combination during the first stage of treatment for acute lymphoblastic leukemia. ''The toxicity associated with dexamethasone used during induction therapy outweighed its potential benefits,'' the Dana-Farber researchers wrote in the journal Cancer last April, spelling out their earlier warning.

But doctors at the Pediatric Oncology Group, with the approval of the National Cancer Institute, believed that a change in an antibiotic prescribed along with dexamethasone would prevent a recurrence of the toxic reactions in Boston.

Looking back, Dr. William Bowman, the study chairman, said he wasn't sure whether the children and their parents were told of the specific risk suggested by the Boston study. They were warned, he said, about the usual risk of death during the induction phase for this type of leukemia - about 2 percent. In this case, the death rate was three times higher.

Ultimately, the Pediatric Oncology Group substituted another drug for the dexamethasone in the first phase. ''We didn't want to take the risk any longer,'' Bowman said.

Today, other researchers are reluctant to criticize Bowman's team, since dexamethasone shows such promise against cancer. ''Part of our task is to fine-tune how we use these drugs in an attempt to improve outcomes. Sometimes very small changes have adverse outcomes and sometimes they have positive outcomes,'' said Dr. Stephen E. Sallan, chief of staff at Dana-Farber.

But Boston University medical ethicist Leonard Glantz said the doctors should have spelled out the risks and alternatives for the parents. ''Certainly parents should be told when you have a responsible medical authority who thinks the risks of a particular regime are especially high. It appears there were some alternatives that reasonable scientists would argue were less risky for the children.''

There is strong reason to believe that deaths and injuries in research involving children are more widespread than what is reported to government offices that handle complaints. For one thing, parents of children with serious diseases often see drug trials as their best option and are less likely to complain later if something goes wrong. In addition, many parents erroneously believe signing a consent form detailing the risks of the experiment means giving up their right to a remedy.

''They think if something negative happened, `Well, I accepted it and I have nothing to complain about,''' said Abbey Meyers, president of the National Organization for Rare Disorders, a patient advocacy group.

Adding to the difficulty of assessing the scope of research injuries, neither drug companies nor the FDA typically release information about problems in drug trials, unless the drug is approved. And even after approval, the FDA drags its feet in releasing details of the trials. Moreover, there is no national effort to track the testing of children or any central review of the reports of side effects.

''It's likely we wouldn't be aware of it unless a patient or their parents brought it to public attention or there was an investigation by [the US Department of] Health and Human Services. Things could happen very easily and no one would know about it,'' said Dr. Ralph E. Kauffman, director of research at Children's Mercy Hospital in Kansas City, Mo., and a leader of the movement to test drugs in children.

Kauffman, like many researchers, believes that the new testing program is much safer than the current system in which doctors prescribe drugs without knowing their effects. ''The risk to a child in a well-run clinical study is much less than the risk of a child receiving that drug in a doctor's office without specific information on dosing,'' he said.

For instance, the antibiotic chloramphenicol was widely used for infections in premature infants until a study in 1959 discovered that babies were unable to metabolize the drug and it was killing them. The drug had only been approved for use in adults for infections resistant to penicillin.

Such problems led the American Academy of Pediatrics to campaign for testing of drugs in children in the late 1970s. But drug companies resisted, citing the difficulties of finding qualified researchers and parents willing to volunteer their children as well as the legal liabilities. While a few companies pursued the pediatric drug market, many saw no financial reason to do so since drugs approved for adults could be legally prescribed to children.

But an act of Congress in 1997, giving companies financial incentives to test drugs in children, has led to a flood of testing. Milne, of Tufts, estimates 600 industry-funded studies are underway, double the number in 1997.

The initiative has spurred the creation of a testing infrastructure, including government-supported centers around the country. In New England, Yale University houses one center, where 10 studies are ongoing. New England is also home to two other big players in the private sector - Parexel International of Waltham, a contract research organization that recently established a pediatric division, and Kelson Pediatric Partners of Hartford, which says it has developed a network of pediatric researchers with access to a pool of 120,000 children. Boston's academic medical centers have seen a small increase in testing.

So far, there are no indications of abuses of the magnitude of those at the Fernald State School in Waltham during the 1940s and 1950s when retarded children were given radiation-spiked cereal in a test sponsored by MIT and Quaker Oats.

But there are plenty of troubling examples of research that demonstrate weak oversight of the burgeoning wave of studies.

In the case of Propulsid, the FDA finally took Dr. Orenstein to task for numerous failings in her study. But its review came five months after Gage Stevens died.

FDA investigators found she failed to report to her hospital review board a number of earlier problems resulting from the drugs, including the case of a baby who had to be hospitalized when he briefly stopped breathing.

In addition, the FDA noted egregious errors in the consent form that the Stevens family signed, including the incorrect claim that Propulsid had been approved for use in children and a failure to spell out known risks.

Stewart and her husband, Scott Stevens, filed a wrongful death suit last September that also accuses Janssen of negligently designing, testing, and marketing the drug.

Ironically, Gage may not even have qualified for the study. Although the consent form says the baby had been diagnosed with esophagitis, a pathology report done just before he entered the study found no evidence of ''significant inflammation.''

There's no doubt that cancer victim Tyler Shelton qualified for the study he participated in. And there's also no doubt that safety mechanisms failed him, largely because his doctor overlooked them, according to regulators.

In May 1997, the 3-year-old was rushed to Arkansas Children's Hospital in Little Rock for emergency surgery to remove a stomach mass. It was cancer, Wilms tumor, that had overtaken a kidney and spread into the adjoining tissue. At the suggestion of Dr. David Becton, a pediatric oncologist, Annette Shelton entered her son in the local branch of the National Wilms Tumor Study, consenting to have samples of Tyler's tumor analyzed for biological and genetic markers and to chemotherapy considered standard for the disease.

''I remember feeling relief,'' said Shelton. ''Tyler was at Children's Hospital, the best in the state, the head guy was seeing us,'' and he said Tyler had a great chance of full recovery.

But the treatment Tyler got appears anything but standard.

According to a lawsuit filed by Shelton and the findings of the federal research protection office, Becton overlooked key test results and assigned Tyler to a study treatment designed for a less extensive and aggressive form of the cancer. Compounding the problem, Tyler's pathology slides weren't sent until months later to the national study office for a review designed to catch errors in diagnosis.

That office notified hospital officials of the mistaken diagnosis in the fall, but regulators say Becton still didn't report the problem as required to oversight agencies.

Shelton said she didn't learn about any of this until months later, after the family had moved to Chicago and the cancer spread to Tyler's liver. There, a new doctor reviewing Tyler's medical records discovered the mismatch between Tyler's condition and treatment.

By then, even the most aggressive treatment couldn't save Tyler, who originally should have had an 80 percent chance of surviving with the proper treatment. ''By November 1998, he was looking like he was three to four months pregnant,'' Shelton said. Sometimes blood poured from his nose and mouth.

Shelton turned to her church for support, and Tyler turned to his identical twin. When Tyler lost all his hair, the barber shaved Tanner's head to match. Most days, dressed alike, Tyler rallied to accompany Tanner to school. In the end, Tyler closed his eyes for the last time with a kiss from Tanner and a hymn from Shelton.

In an affidavit filed with the court, Becton said, ''I deny that I was guilty of any negligence in the care and treatment I rendered to Tyler Shelton in 1997 which was the cause of Tyler's death.'' He said that he corrected the diagnosis a few weeks into Tyler's treatment after reviewing medical records, then made ''a medical judgment'' to leave Tyler on the same treatment regime. He said he explained both the less favorable diagnosis and the treatment plan to Annette Shelton in May or June 1997.

But Shelton says there was never any such conversation, and the family has found nothing to document Becton's version in the medical records. In fact, Becton contradicts his own story in a letter to one of Tyler's Chicago doctors.

Recounting the tragedy, Shelton still wonders why Becton didn't admit his mistake and change the treatment. ''Why,'' she wails, tears staining her cheeks, ''Why did you not say anything? Why did you not do anything? This is not a car that got the wrong motor. This was a child's life at stake.''

Globe librarians Lisa Tuite and Rosemarie McDonald contributed to this report. Alice Dembner can be reached at ADembner@globe.com.

-- Aunt Bee (Aunt__Bee@hotmail.com), February 18, 2001