View all Articles
Commentary By Paul Howard

What policymakers can learn from a $21 billion failure

Health Pharmaceuticals

William Osler, one of the founding faculty members at Johns Hopkins Medical School, once remarked: “If it were not for the great variability among individuals, medicine might as well be a science and not an art.” A century later, medicine is, despite its technical prowess, in many ways still an art — albeit an expensive one.

Pfizer learned that lesson the hard way in early December, when they halted development of their much anticipated “good cholesterol” boosting drug, torcetrapib. A late-stage clinical trial revealed excess deaths and cardiovascular problems in patients taking the drug together with Lipitor, compared to Lipitor alone.

In the 15,000-patient trial, there were 82 deaths for the torcetrapib combo vs. 51 for Lipitor. The failure cost Pfizer $800 million in research costs, $20 billion in market capitalization and 15 years of research.

The upside: Torcetrapib’s demise may mute congressional plans to institute onerous new safety regulations when the Prescription Drug User Fee Act comes up for reauthorization early next year.

Thanks to Pfizer’s woes, Congress has a bird’s-eye view of how difficult and expensive it is to develop new medicines, and why better science — and not more regulations — is what is really in the public’s best interest.

The exact source of the problems that halted torcetrapib’s development is unknown. Indeed, Pfizer may have been tripped up by its commitment to innovation. Raising “good cholesterol” seemed like a promising new gamble, but it ultimately had unforeseen risks. Other companies developing similar drugs have billions more dollars wagered on their success; as they explore new treatments and cures, setbacks are inevitable.

Sound like a risky business? It is. The ancient Greeks used the same word for medicine — pharmakon — as they used for poison, and with good reason. Evolution loves conservation, and the same biological mechanisms that can run amok and create tumors or arthritis are utilized in other organ or tissue systems to repair cells or fight infections.

Disease is often the result of one of these mechanisms going awry, but it is frightfully difficult to target the culprit without disturbing another, perhaps vital, function. Like dropping a rock into a pond, sending a single chemical into the body produces ripple effects that can generate serious side effects. To complicate the problem, some effects only show up very rarely, or only as the result of interactions among multiple drugs.

Through extensive testing, companies do a good job of weeding out dangerous drugs well before they reach the late-stage testing that torcetrapib did, let alone the market. Still, no tests are perfect and rare problems can slip through the system, emerging in late-stage trials or even after the drug is approved. Some adverse effects might not become detectible until 100,000 or more patients take a drug.

Ironically, while the industry gets better at weeding out problems, the drugs that do make it to market are held to increasingly higher standards. Rare reactions that might be tolerated for older, more familiar drugs, like aspirin or Tylenol, now elicit congressional outrage, drug withdrawals, and multi-billion dollar lawsuits.

The key to safer — and less expensive — drug development is not increased legislation; it already costs about $1 billion and a decade of research to bring a single medicine to market. Rather than adding more costs, we need to take Osler’s warning to heart and focus on technological solutions to help unravel the mystery of individual variation.

Industry, the Food and Drug Administration and the National Institutes of Health are already working on methods to target drugs to a patient’s unique DNA profile, turning individual variation into an advantage rather than an Achilles’ heel.

The FDA is also working with researchers to validate safety biomarkers that can serve as early warning signs for drug toxicities. Eventually, these programs will help to develop safer, more effective (and hopefully less expensive) drugs. Ultimately, scientific standards should obviate the need for the FDA to monitor the minutiae of drug development.

Still, innovation can move only as fast as science. If Congress is serious about promoting long term drug safety, and not just scoring cheap headlines, the best thing it can do is fund the FDA’s Critical Path project. This project aims to develop scientific tools that can transform the hit-and-miss process of drug development into one defined by more empirical standards.

Better, safer drugs are within our reach but this requires patience from policymakers. By encouraging innovation, Congress can help to make drug safety what it should be: not an art, but a science.

This piece originally appeared in

This piece originally appeared in