Healthy Skepticism Library item: 1758

Warning: This library includes all items relevant to health product marketing that we are aware of regardless of quality. Often we do not agree with all or part of the contents.

 

Publication type: news

Pollack A
Despite Billions for Discoveries, Pipeline of Drugs Is Far From Full
The New York Times 2002 Apr 19
http://www.nytimes.com/2002/04/19/business/despite-billions-for-discoveries-pipeline-of-drugs-is-far-from-full.html?pagewanted=1

Full text:

This should be the golden age for scientists. The deciphering of the human genome is laying bare the blueprint of human life. Medical research has increased understanding of disease. Robots and computers are turning drug discovery from a mixing of chemicals in a test tube to an industrialized, automated process.

Yet if industrialization normally means higher speed and lower costs, the pharmaceutical industry has been experiencing the opposite – a “clear fall in productivity,” according to Dr. Frank L. Douglas, the chief scientific officer of Aventis. Instead of narrowing the list of compounds that might be useful in drugs, automation has broadened it – greatly increasing the number of formulas tested without yet delivering commensurate growth in safe and effective drugs. The industry’s output of new drugs has risen only modestly in the last two decades despite a more than sixfold increase, after adjusting for inflation, in research and development spending, to more than $30 billion annually. In the last few years, the output has actually declined. “It makes you wonder: what are they doing?” said John Borzilleri, a health sciences portfolio manager at State Street Research and Management in Boston. “Are they spinning their wheels or is it just a matter of time?” The perceived paucity of new drugs in company pipelines has become a preoccupation of the industry and of Wall Street. Bristol-Myers Squibb, Merck and other drug makers have said earnings will be lower than expected this year, in part because there are not enough new products to offset declining sales of old ones that lose patent protection and face competition from generic versions. The industry mantra is now “it’s the pipeline, stupid,” said Dr. Steven M. Paul, group vice president for discovery research at Eli Lilly in Indianapolis. But no one has yet found a reliable way to fill the pipeline. With drugs not coming fast enough to sustain the double-digit growth in earnings and revenue that Wall Street has come to expect, more companies might merge, to bolster earnings through reduced costs, analysts say. Companies are also trying to increase revenue from the drugs they do have by raising prices, advertising heavily to consumers and scrambling to extend their patents, actions that have embroiled the companies in controversy. Over the last 25 years, a parade of technologies has promised to transform drug development: genetic engineering, rational drug design, combinatorial chemistry, improved screening – and now, genomics. Yet the time spent to develop a drug, not counting the months consumed by government review, has lengthened to more than 11 years from about 9 years in the 1980’s, according to the Tufts Center for the Study of Drug Development, and the cost has more than doubled, after adjusting for inflation, to $800 million. Critics, who note that the Tufts Center gets money from drug companies, say those figures are inflated to justify high drug costs. Still, the industry’s failure rate, a big part of its costs, has not declined. Only one in about 5,000 early drug candidates and only one in five drugs that enter clinical trials ever make it to market, experts say. The remainder drop out because they do not work or are toxic. “The odds are just dreadful, and they seem to be getting worse,” Julie A. Olson, a vice president for licensing at Pfizer, told a recent biotechnology conference. Some executives say that given the long development period, genomics and some other technologies are too new to have made a difference. Recent increases in research spending should lead to more drugs 10 years from now, they say. “We’re beginning to tackle all sorts of diseases we couldn’t before,” said Dr. Goran Ando, head of research and development at Pharmacia in Peapack, N.J.. But, he added, “it won’t happen overnight.” To be sure, looking at just the number of drugs getting to market can be misleading because the companies are producing better-selling drugs. C. Anthony Butler, an analyst at Lehman Brothers, said the industry’s pipeline in 1995 contained 450 drugs, of which he projected that only 15 would have peak annual sales exceeding $800 million. In 2001, he said, the pipeline had about the same number of drugs but 92 of them were potential $800 million products. Maintaining that pace, however, will not be easy. “In some ways the easy drugs have been done,” said Dr. Robert H. Rubin, a professor of health sciences and technology and of medicine at Harvard. Drug development is a cumbersome process. Companies usually start by identifying a target, often a protein in the body that is thought to play a role in some disease. Then they try to either design or find a compound that can attach itself to the target protein, thereby changing the course of the disease. They must make sure the compound is otherwise suitable – that it can be made into a pill, for example. It is then tested in animals for toxicity. Only then can it be tested on humans. Technology has helped with the early part of the process, the discovery of compounds, but not as much with the costliest and most time-consuming portion of drug development – clinical trials. Scientists still cannot tell whether a drug will work or be toxic until they test it. “The slowest parts of drug discovery and development are pretty much the same,” said Peter S. Kim, the executive vice president for research and development at Merck. But even in the early part of drug development there has been disappointment. In the 1990’s, a new technology called combinatorial chemistry allowed companies to create hundreds of thousands of compounds by mixing chemical building blocks in different combinations. Drug makers then developed robots to screen this wealth of compounds. But many of the compounds created this way lacked characteristics that would make them suitable for use as drugs. Still, there has been progress. Bristol-Myers, for example, increased the number of compounds coming out of its early discovery stage to 14 a year late in the 1990’s from 6 a year early in the decade after spending $50 million to install an advanced “screen machine.” The newest technology, genomics, could increase the number of targets in much the way combinatorial chemistry increased the number of chemicals. Until now, virtually all drugs have been directed at an estimated 500 proteins in the body. But by sifting through the human genome, companies are finding thousands of genes that produce previously unknown proteins that might be involved in the disease. In the long run, that is expected to open vast horizons, and perhaps even let companies reduce failure rates in animal testing and in clinical trials by enabling them to predict toxicity and effectiveness by studying how a drug affects genes. But in the short run, it has left the industry inundated with targets and data that may increase the failure rate by leading companies to start trials before they fully understand what the new data are telling them. As Jerry Karabelas, a former head of pharmaceuticals at Novartis, once put it: “Data, data everywhere, and not a drug, I think.” To bolster their output in the meantime, big drug companies are turning to biotechnology companies for products and technology, typically spending about 30 percent of their research budgets on outside collaborations. There can be fierce bidding and rising prices for drugs that are close to reaching market, raising the risks for drug companies. Last fall, Bristol-Myers agreed to pay $2 billion to ImClone Systems for an ownership stake and the rights to market a cancer drug. But approval of the drug has since been delayed. By contrast, some analysts and executives say, the drug companies are becoming more cautious about paying for basic technology because they are disappointed that what they have acquired so far has not led to more drugs. Some analysts say the drug industry is undergoing a transition similar to the computer industry’s move from vertical integration, exemplified by I.B.M., to a horizontal structure – with Intel making chips, Microsoft making software and others specializing in manufacturing or sales. Drug companies, these analysts say, will increasingly become the marketers and coordinators of work done by others. “We think the old model of having everything under your own roof, a completely integrated monolithic organization, is not feasible,” said Pradip K. Banerjee, a partner at Accenture, the consulting company. Because biotechnology companies are smaller and more focused, they can often move faster than the big drug companies. Vertex Pharmaceuticals, based in Cambridge, Mass., said it spent about $50 million on each of seven drugs to get them into the second phase of clinical trials, a fraction of the usual costs. Joshua S. Boger, the chief executive of Vertex, said his biotechnology company was organized from scratch to take advantage of new technologies. If new technology is just put into the existing process, “you’re just going to move the bottleneck to another place,” he said. But as biotechnology companies have moved from making well-known compounds like insulin to more complex challenges, many of them are experiencing failures in clinical trials or at the Food and Drug Administration. Amgen, the largest biotechology company, went 10 years without a new drug until last year. It is now growing by using the tactics of bigger drug companies – introducing improved versions of its drugs and buying another company, in its case Immunex. In any case, executives say they had little choice but to try new technology. “Had we not had these technologies,” said Dr. Douglas of Aventis, “I think the situation would have been much worse.”