Pediatric drug testing may be unhealthy

Even by Capitol Hill’s standards, last Tuesday’s press conference was an exceptionally self-congratulatory event. The occasion was the unveiling of the Pediatric Research Equity bill (S. 650) by a bipartisan group of senators.<?xml:namespace prefix = o ns = “urn:schemas-microsoft-com:office:office” />

Accompanying them were representatives from the <?xml:namespace prefix = st1 ns = “urn:schemas-microsoft-com:office:smarttags” />American Academy of Pediatrics and the Elizabeth Glaser Pediatric AIDS Foundation, with a supporting handout bearing the signatures of 20 other medical associations.

Last fall a federal court had ruled that the Food and Drug Administration (FDA) lacked the authority to mandate that pharmaceutical manufacturers test on what are known as “off-label” uses in children. This group intended to give FDA that authority. What could conceivably be wrong with that?

There’s quite a bit, for anyone concerned about the already-slow pace of new drug development.

“Off-label” use involves using a drug for something other than its official “label” indication. When a drug firm submits a new drug for approval, it must prove to FDA that the drug is safe and effective for the purpose described on its label. But once a drug is approved, a doctor is free to put it to any appropriate use. Such off-label use may sound fishy, but in many areas of medicine, such as cancer treatment, it represents the highest standard of care because it’s based on the latest research.

The issue addressed by S. 650 isn’t whether drugs for children are being adequately tested on children; they already are tested under current law. S. 650 deals with whether adult drugs that pediatricians have found useful must be tested, by their manufacturers, for such off-label pediatric use.

Having more data on drugs sounds wonderful, but there’s an inevitable trade-off between developing more data and developing more drugs. FDA’s drug-approval process is already incredibly long and cumbersome; on average, it takes about 15 years and $800 million to bring a new drug to market. While some argue that FDA is approving drugs too quickly, that view isn’t shared by the medical profession. A poll of oncologists released last year found that over 60 percent believe FDA is too slow in approving new therapies.

S. 650’s mandates would impose yet more regulatory hurdles for new drugs. These might be trivial for some candidate drugs, but they could delay, or cancel altogether, the approval of others. Moreover, children are not the only special patient population; claims of additional pharmacological tests can validly be made for the elderly, the immune-compromised, and pregnant women. If testing mandates are imposed for every special population, we will get more data, but almost certainly fewer drugs as well. We may even lose some of the drugs we use today; S. 650 permits FDA to block an already approved drug if its manufacturer doesn’t agree to undertake pediatric testing requested by the agency.

Advocates of S. 650 ought to be able to demonstrate serious harm under the current system. They haven’t. Last year, the “poster child” incident for their cause involved a Tennessee hospital’s use of erythromycin in 1999. Some 200 babies were given the drug after they were exposed to whooping cough by a hospital staffer. Seven infants subsequently developed a stomach obstruction, pyloric stenosis, which required surgery. All of them recovered, but the claim, repeated in several newspaper editorials, was that pediatric testing might have prevented these adverse reactions.

The claim is wrong. It turns out that the pyloric stenosis occurred only in those infants who were given the drug when they were under three weeks old. Testing on babies would not have discovered this side effect, unless the tests happened to be focused on a large enough number of infants younger than three weeks. But the notion that a researcher could have ethically administered erythromycin to such young infants purely for research purposes is preposterous.

If there is a serious problem regarding off-label pediatric data, there are better ways to deal with it. Congress has already enacted testing incentives via six-month patent extensions for firms that undertake them, and it has established a direct federal funding mechanism for such tests as well. Those provisions involve some pretty obvious costs, such as temporarily higher prices for drugs that remain on patent, but at least those costs are calculable. S. 650’s testing mandates, on the other hand, have the far less visible effect of imposing new burdens on drug development. That may be politically attractive, but if the end result is a slowdown in the development of medicines, the real cost will be incalculable.