I Want a New Drug
Why price controls will stop pharmaceutical progress.
Jun 16, 2003, Vol. 8, No. 39 • By WILLIAM TUCKER
Efficacy testing adds years and hundreds of millions of dollars to the approval process. Desperate patients wait indefinitely while FDA regulators chew their pencils and scratch their heads, looking for more convincing evidence. Meanwhile, with the spread of information on the Internet, clinical trials for efficacy are becoming more and more difficult to complete. Say you're dying of cancer. Would you be willing to participate in an FDA trial where there is a 50 percent chance you will be receiving a placebo? "An increasing number of trials are now falling apart as soon as there are perceived results," says Tom Miller, health policy analyst at the Cato Institute. "It's also getting harder and harder to recruit volunteers."
Rather than allowing an orderly progression of new products at reasonable prices, efficacy testing has turned the industry into a casino. For every 5,000 new compounds the industry screens, 250 are chosen for preclinical testing, according to Pharmaceutical Research and Manufacturers of America (PhRMA). Five of these will enter long-term clinical trials. Only one will be approved, says Joseph DiMasi of the Tufts Center for the Study of Drug Development. Thus, each marketed drug must earn back on average $1 billion in FDA testing costs. But only three of ten marketed drugs earn back even their own investment.
The syndrome has left even the most successful drug giants highly dependent on one or two blockbusters. Merck has Vioxx, a $3-billion-a-year painkiller; Pfizer has Viagra; Eli Lilly has Prozac. But few of their other products make money. As patents on these meal tickets approach expiration, investors get jittery. "Clinical trials have become so expensive, it's very difficult to be a mid-sized player," says Ben Bonifant, a life sciences specialist at Mercer Management Consulting, Inc.
Profits have become even more tenuous because of generic imitators, which have prospered since the 1984 Hatch-Waxman Act. Before 1984, generic drugs also had to undergo testing to meet FDA requirements, which effectively extended the original drug's patent far beyond the statutory 17 years. Hatch-Waxman struck an artful compromise, allowing generic imitators to use the patent holder's results for market approval. In exchange, the patent life on original drugs was extended up to five years. Since 1984, generics' share of the market has risen from 19 percent to 50 percent--which has done wonders for drug prices. But it also makes the major research giants even more dependent on highly successful new products.
Making testing even more complex is the emerging pattern that different drugs work differently for different groups of people. One recent drug, for example, proved to have a significant effect in treating AIDS in African Americans but not in whites. Should the drug be approved only for use by blacks? Should it not be approved because it won't help whites? The FDA hasn't yet decided.
Another major impediment to drug development is that the FDA requires efficacy testing for each separate use. Often a drug marketed for one disease turns out to be effective in treating another. Yet the FDA still requires another trip through the bureaucracy. Bayer, for example, is not allowed to market aspirin as a prevention for heart disease, even though studies have shown it cuts the risk of coronary by half.
What the FDA does allow is for doctors to prescribe drugs for non-approved use on an informal basis. Yet this has produced a whole new round of drug scandals. Warner-Lambert is currently being sued by a "whistle-blower" who has accused the company of paying doctors to publicize Neurontin, an epilepsy drug, for a dozen other conditions, including pain, bipolar disorder, and restless-leg syndrome. Schering-Plough is facing criminal charges for marketing several of its drugs for unauthorized uses.
Rare "orphan" conditions such as restless-leg syndrome are underserved by pharmaceutical research, because the small market for a treatment cannot support the cost of FDA testing. Can Warner-Lambert be expected to spend millions of dollars to get Neurontin approved for each new condition? If a drug is already proved safe for one condition, why not allow its use in others?
One proposal has been to do away with the FDA's efficacy testing. Doctors and patients could figure out for themselves whether a drug works--once its long-term safety has been established. The problem is this might open doctors to liability, which is something they don't need right now. Another idea would be to allow private organizations to certify safety and efficacy, the way Underwriters Laboratory certifies electrical devices. This would at least break up the FDA's monopoly mentality and alert it that there are people out there waiting for it to act.