Combating childhood cancer is one of oncology’s greatest success stories. While cancer in children remains, thank goodness, rare, it still strikes about 14,000 kids a year and kills about 3,000, making it the second leading cause of childhood mortality after suicide.

But the field has made tremendous strides in recent years. While some tumors of the central nervous system remain stubbornly resistant to treatment, many forms of childhood cancer like acute lymphoblastic leukemia are approaching an 80 percent cure rate. That disease’s cure rate was near zero in the 1950s.

What accounts for this progress? Every form of childhood cancer is, by definition, a rare disease. Therefore, it usually receives less than adequate attention from the pharmaceutical industry. Clinicians who want to investigate potential treatments usually turn to the National Cancer Institute’s Children’s Oncology Group (COG), which knits the nation’s 200 children’s hospitals into an effective network for carrying out clinical trials.

The COG has had remarkable success over the years in encouraging community oncologists to enroll their underage cancer patients in clinical trials. Fully 94 percent of the children diagnosed with cancer in the U.S. are treated at institutions affiliated with COG, and more than 60 percent enroll in trials.

But much remains to be done. At a meeting last fall of the pediatric subcommittee of the Food and Drug Administration’s Oncology Drugs Advisory Committee, officials laid out the dimensions of the problem. There are the 169 generic cancer drugs on the market that, although they are still in widespread use in adults, have no pediatric labeling and no pediatric dosing information because they were never tested in kids. And since they’re off patent, no drug company has a stake in testing them, even with the incentives created in the 2002 Best Pharmaceuticals for Children Act (BPCA).

That bill grants six months of additional exclusivity if a company tests its drug in children. While that’s great for drugs that have been on the market for a while and are still on patent, it does nothing for generics.

New cancer drugs coming to market are another matter entirely. Since 2003, they’ve come under the Pediatric Research Equity Act, which requires drug companies to conduct pediatric tests for every new drug brought to the FDA as long as there is a potential use in the underage population. “But most of the blockbuster cancer drugs are for diseases that don’t exist in children – cancers of the breast, colon and lung,” said Gregory Reaman, chairman of the pediatric subcommittee and of COG. “The existing incentives are not enough.”

Indeed, the existing incentives can be abused. Earlier this year, the FDA sent out a press release marking the 100th drug that had been given six months additional patent life under BPCA. One of the seven cancer drugs on the list was fludarabine, which had been approved in 1991 for chronic lymphocytic leukemia, a cancer that does not strike anyone under 30.

How did that happen? In 2003, fludarabine was approaching the end of its patent life. So Berlex, its maker, conducted a trial in 62 children, according to FDA documents, thus meeting the requirements of the law. When the company got its six-month extension, the FDA felt compelled to note that the data was insufficient “to establish efficacy in any childhood malignancy.”

Officials at COG are hoping the National Cancer Institute will step up its efforts in testing off-patent cancer drugs in kids. At last October’s meeting, the agency promised to develop a priority of list of 15 to 20 drugs and invite proposals from pediatric oncologist specialists around the country. Dosing trials are already underway on vincristine, dactinomycin, and methotrexate. Other generics high on COG’s wish list include cisplatin, etoposide, and cyclophosphamide.

The focus on older drugs defies conventional wisdom, but conforms to the lessons learned over the years at COG. Earlier this year, the British Medical Journal published a survey of 126 randomized, controlled clinical trials carried out over the past 45 years by COG. The reviewers, led by Dr. Benjamin Djulbegovic of the University of South Florida, found that a new treatment or regimen in the trials was just as likely to be inferior to standard treatment as superior to standard treatment.

While the authors were primarily concerned with shoring up the ethics of putting children into randomized clinical trials (if you knew a new drug or regimen was going to be better, it would be unethical to put someone on the standard regimen), it concomitantly proved that new isn’t always better. Indeed, the reviewers found that “industry sponsored trials are associated with increased likelihood of outcomes favoring sponsors, most likely due to selective reporting of favorable outcomes and violation of the uncertainty principle in the design of the trials.”

This meta-analysis of childhood cancer testing is a powerful argument in favor of increasing the amount of government funding for objective clinical trials. Conversely, the early experience of the Best Pharmaceuticals for Children Act calls into question the wisdom of incentives like patent extensions, at least when it comes to cancer drugs.

— By Merrill Goozner

Posted on June 1, 2006 06:00 AM