The Other Side of Drug Discovery, Part 1

By Jason Socrates Bardi

The shop seemed to be full of all manner of curious things—but the oddest part of it all was, that whenever she looked hard at any shelf, to make out exactly what it had on it, that particular shelf was always quite empty: though the others round it were crowded as full as they could hold.

"Things flow about so here!" she said at last in a plaintive tone, after she had spent a minute or so in vainly pursuing a large bright thing, that looked sometimes like a doll and sometimes like a work-box, and was always in the shelf next above the one she was looking at.

—Lewis Carroll, Through the Looking Glass, 1863

"What do you suppose is the most successful drug in the last 5,000 years?" Tamas Bartfai asks me. This question almost seems like one of those obvious ones that you should be able to guess, even if you've never thought about it before. I don't.

After a moment's demur, I offer, weakly, "aspirin." (Antibiotics, I figure, would be too obvious).

"Aspirin?!" says Bartfai. "Aspirin wouldn't even be approved by the FDA [U.S. Food and Drug Administration] today [because intestinal side effects would complicate its clinical trials]. No it's morphine."

Opium, the poppy extract that contains morphine, has been used for 5000 years, he tells me. It was a huge breakthrough for treating the indications of diarrhea and pain. Significantly, morphine was isolated from opium by F.W.A. Serturner in the early 1800s and used by doctors for nearly two centuries before its endogenous ligand and receptors (only cloned in the last 10 years) were known.

"If you can discover a [disease modulating] effect of a compound, even without knowing how it works," says Bartfai, "if the compound does something that no other drug does, then people will buy it."

This is Bartfai's first lesson in drug discovery—that drugs were not always discovered the way we discover them today. Today, when scientists set about designing drugs, they often have structurally known, cloned receptors and enzymes to target that play crucial roles in the diseases. They have access to genomic information, and many times the protein structure as well as sophisticated animal models for many of these diseases. They have chemical libraries of tens of thousands of potential lead compounds to test, and work in a society that recognizes and embraces the need for new and better drugs to fight any number of ailments.

And yet, despite all these advances, the diversity of chemical entities that make up the modern drug lexicon has yet to explode, though Bartfai believes there will be an explosion in the next 15 years, if society permits it. After all, he says, drug costs are only 15% of the total health care costs.

A Primer on Drug Discovery

While there are around 10,300 FDA-approved drugs in the United States today, most of these are made up of some combination of only 433 distinct molecules. Acetaminophen, for instance, is an ingredient in hundreds of separate drugs. Half of the 433 were approved before 1938, and at least 50 are "me too" drugs, a slightly modified form of compound already on the market. Finally, there are only eight major, chemical "scaffolds" upon which all the 433 molecules are based.

"All this says," says Bartfai, "is that it is darned hard to make a new molecule that is going to be a drug these days."

Over the next three weeks, Bartfai is giving a series of lectures on drug discovery sponsored by the Skaggs Institute for Chemical Biology at The Scripps Research Institute (TSRI) in order to communicate his decades of experience as a consultant and executive in the Pharma industry to San Diego's scientific community.

The steps that lead to success in bringing a drug to the market are not always the most obvious ones to a scientist, says Bartfai, who came to TSRI in 1999 to join the Department of Neuropharmacology and was soon after appointed director of the Harold L. Dorris Neurological Research Center.

Prior to 1999, Bartfai served as senior vice president in charge of central nervous system research at Hoffman-La Roche, a department most famous for the drug Valium, and for its Parkinson's disease drugs. He was brought there to develop a major human genetics effort to aid discovery of new treatments for schizophrenia and Alzheimer's disease. Before that, he was involved in the development of Zimelidine, the first selective serotonin reuptake inhibitor (SSRI) and two anti-psychotic agents used in the treatment of schizophrenia as a consultant for Astra (now AstraZeneca). He has been a long-term consultant for several major Pharma companies.

From these experiences, Bartfai has an almost unique cache of information. He knows the reasons behind decisions made by the FDA and companies in the pharmaceutical industry. He knows how a small start-up biotech company should interact with the major pharmaceutical firms. And, most importantly, he knows how to get a drug approved by the FDA—a point which he says he thinks backwards from. In any scientific project aimed at producing a therapeutic agent, after all, if you do not know how you will do the clinical trials and how you will get it FDA approved, it is not going to be a drug, no matter how great the discovery.

The lectures are not necessarily designed to present new information, but rather to present a few themes that will help organize a person's understanding of the drug design process—adding historical and economic perspectives.

The historical perspective is in part the lesson of morphine—that successful drug discovery was not always done the high throughput way it is done now. The economic perspective is more complicated, but nevertheless essential when dealing with the pharmaceutical end of the drug-discovery business.

"Nobody teaches this," he says.

The Anatomy of an Industry

"The stakes for the pharmaceutical companies are enormous," says Bartfai. "Twenty-six billion dollars was spent on developing drugs last year, and nine new chemical entity (NCE) drugs were approved." (The FDA defines an NCE as a drug with no previously approved active chemical moiety).

This stakes are reflected in the typical price-to-earnings (P/E) ratios for companies involved in drug design.

The P/E is the ratio of the market value of a single share of stock divided by the total earnings for that share. So a P/E of 10 means that every $10 of stock returns a dollar of profit. According to Charles Schwab Company, the historic average P/E ratio for all the fortune 500 companies is around 16 and, of course, varies with the economy.

Blue chips stocks, the shares of those industries like car manufacturers or utilities companies tend to have smaller P/E ratios—under 10. The low number reflects the fact that these industries are time-tested and established as long-term, secure earners.

The P/E of growth industries, like those in the health sector, tend to be larger. The market average for major pharmaceutical companies is around 20. Biotech companies tend to be even higher. The 2001 P/E average for the largest biotechnology companies listed on the New York Stock exchange was nearly 40, according to Forbes magazine. Small publicly traded biotech companies can be higher still.

The high P/E ratio of pharmaceutical companies and biotechs means that the earnings per share is low with respect to the cost of the stock, but, more importantly, that investors have high expectations for the companies to show significant growth and yield large profits in the future.

Another characteristic of the pharmaceutical industry is a lack of customer brand loyalty. There are no "Upjohn people" or "Roche people" the way there are "Pepsi" or "Coke people." Unlike other industries, where products may be selected on the basis of manufacturer, pharmaceutical companies rely wholly on the effects of the drugs they produce and market.

"If you succeed, nobody asks [whether you] you have made this kind of drug before," says Bartfai. "If you make a truly original drug that has a health benefit, people will eventually buy it."

The drug doesn't even have to be the first one on the market to fight a particular indication. The thing that counts is that it is significantly better than any other drug out there.

As an example, Bartfai points to Astra, which entered the billion-dollar-a-year heartburn market with the drug Losec (also called prilosec), which blocks a proton ATPase responsible for pumping acid into the stomach. Astra was at the time a tiny company entering a huge pond. But Losec was more effective than its competitors, Zantac and Tagamet, both well known labels, and is one of the reasons why Astra (now merged to form AstraZeneca) has emerged as the largest pharmaceutical company in the world.

"Nobody even knew how to spell [Astra]," says Bartfai, who was a consultant for the company at the time "But we made a significantly better drug, [which] so far has had $36 billion in sales."

The health benefit doesn't even have to be as tangible as reducing acid reflux and stopping somebody's heart burn.

"You can sell something which does not cure a disease if you have a good enough argument that it can prevent a disease," says Bartfai. "High cholesterol is not a disease, but six billion dollars is spent each year on cholesterol-lowering drugs."


NEXT WEEK: Part 2—Drug Development is the Single Most Regulated Human Activity



The Lectures:

DEVELOPMENT OF DRUG DISCOVERY PARADIGMS IN BIG PHARMA OVER THE LAST 100 YEARS, on Thursday, April 25. This overview will cover drug discovery paradigms, physiology-based drug development, 1880–1980; high-throughput screening-based drug development, molecular/cellular drug discovery, 1980–2000; and "reverse pharmacology," based on genomics and proteomics, 1999–present.

TARGET-BASED DRUG DISCOVERY, on Thursday, May 2. Topics to be discussed include validated drug targets: what they are and for whom, a determination of their value, and comparison of targets for the same clinical indication.

SELECTION OF CLINICAL CANDIDATES: MULTIPLE PRESSURES, on Thursday, May 9. The presentation will focus on the key milestones of preclinical drug development—timing, expenditures, backup strategies, outside validations, and orphan drugs—as these factors play out in big Pharma decision making.

All lectures will be held from 5 to 6:30 PM in the Valerie Timken Amphitheater of Green Hospital.




Go back to News & Views Index


This comparison between the categories and numbers of drug targets in the years 2001 and 2005 shows the influence of genomics on drug discovery-the addition of new drug targets.