To the editor:

Laura DeFrancesco's book review in the June issue (Nat. Biotechnol. 22, 663–664, 2004) exposes many of the biases and flaws of Merrill Goozner's “The $800 Million Pill: The Truth Behind the Cost of New Drugs,” but her criticisms do not go far enough. The book is just one more exercise in bashing the drug industry by an organization that devotes itself to that pastime.

It is misleading to observe, as Goozner does, that US consumers pay twice for their drugs—once with tax dollars for research that provides the scientific substrate for pharmaceutical R&D, and again to purchase the marketed drugs themselves. In fact, the US National Institutes of Health (NIH; Bethesda, MD, USA) funds primarily pre-commercial fundamental research into the biochemistry, physiology and molecular biology of cells and organisms, in health and disease. A perfect example is the invention of recombinant DNA technology, an indispensable tool in contemporary drug discovery, which emerged from the confluence of NIH- and National Science Foundation (NSF, Washington, DC, USA)-funded basic research on enzymology, molecular genetics, microbiology and analytical techniques. Who could have foreseen that the research on the hugely complex biology of bacteriophage lambda or the sex life of Escherichia coli would eventually pay such huge dividends?

Goozner's proposal to shift much of the responsibility and expense for pharmaceutical development from industry to “a more cost-effective government agency” is pure folly. In Milton Friedman's delightful phrase, that is like trying to design a cat that barks. History has shown repeatedly that the private sector, not the government, is better suited for product design, testing and manufacture. It has been a long time since the government itself produced munitions, for example. One impetus for its getting out of the business was the catastrophe that occurred in 1845, when the USS Princeton undertook the initial tests of new 12-inch guns. The first of these, which had been designed by a navy captain, exploded, killing the secretary of state, secretary of the navy, a naval captain and a Maryland congressman, among others. Had President John Tyler not been delayed below decks, he, too, would likely have been killed.

The history of government manufacture of pharmaceuticals is far from encouraging. Consider, for example, the decades of production of human growth hormone for short children by the US National Pituitary Agency (Bethesda, MD, USA). This program, conducted from 1963 to 1985 under the auspices of the NIH, was a haphazard operation. The hormone was prepared from human pituitary glands recovered from cadavers, and the absence of rigorous collection guidelines and purification procedures permitted contamination of the formulated drug with the prions that cause Creutzfeldt-Jakob disease, the human equivalent of bovine spongiform encephalopathy, or 'mad-cow disease.' As a result, several dozen recipients have died a lingering and gruesome death.

If this had been a private operation, competition and the threat of liability would have compelled frequent updating of the drug's purification and formulation with state-of-the art technologies and would have required rigorous adherence to government regulation. But when the government itself is the manufacturer these forces are attenuated, and the shield of government safety regulation is weakened. The nation's drug regulator, the US Food and Drug Administration (FDA, Rockville, MD, USA), is a sibling agency of the NIH, and the two agencies' common political interests and 'professional courtesy' appear to have compromised vigorous oversight over the NIH's production of human growth hormone.

This example underscores the importance of disinterested oversight that effectively neutralizes the conflict of interest that is inherent in essentially the same entity testing and marketing a drug. FDA, perhaps the most stringent regulatory agency on the planet, now provides that check on industry (a factor that Goozner overlooks), but having FDA regulate a sibling government agency charged with developing new drugs would create myriad new problems.

DeFrancesco is right that Goozner's book offers no “easy answers.” She might have added that readers won't find any sensible ones there either.