|
Figuring Out Pharmacogenomics
STACY LAWRENCE
In an age of increasingly circumscribed doctor-patient relationships, a new ideal of patient treatment is emerging. Pharmacogenomics, the application of information about genetic variation to enhance the understanding and manipulation of drug response, could ultimately enable diagnosis and medication tied to a patient’s genetic predispositions.
In late 2002, in the wake of the complete mapping of the human genome, another similarly massive task was undertaken—the International HapMap Project. Hap is short for haplotype, a pattern of sequential variations in the genetic code. This project aims to understand fundamental variations in this pattern that are found within populations that are highly similar genetically, including the Yorubas in Nigeria, the Japanese, the Han in China, and U.S. residents of northern and western European heritage.
Differences of just a single letter in genetic sequences are single nucleotide polymorphisms (SNPs, pronounced “snips”). The particular genetic phrasings and the traits they are associated with are fundamental to identifying predisposition to disease and differences in reaction to treatment.
Genetic variations determine all sorts of characteristics, including the extent to which individuals react differently to the same medication—some patients don’t respond or even experience an adverse reaction. For any given drug, an estimated 35% of patients fall into these two categories. In the United States, more than 2.2 million serious cases of adverse drug reactions and more than 100,000 related deaths were reported in a frequently cited Journal of the American Medical Association study of 1994 statistics. That makes adverse drug reactions one of the leading causes of hospitalization and death in the United States.
Pharmacogenomics has played a relatively minor role in the drug discovery process so far. Still, over the last 20 years, more than 50 submissions to the U.S. Food and Drug Administration for drug approval— investigational new drugs (INDs) and new drug applications (NDAs)—have incorporated pharmacogenetic data on people who metabolize the drug in question poorly and are thus at greater risk of side effects. The FDA has announced that this year it will issue official guidelines on the use of this type of data in the drug approval process. One successful example of pharmacogenomic research is the breast cancer treatment Herceptin (trastuzumab), developed by Genentech and Roche for the 30% of women with breast cancer tumors that are genetically inclined to have the Herceptin receptor. In 1999, the first year it was available, Herceptin made almost $200 million in the United States, the highest first-year sales of any cancer drug in the country. Revenue has continued to grow, reaching almost $400 million last year.
That’s a respectable figure, though by no means does it approach the $1 billion annual revenue level associated with blockbusters. The pharmaceutical industry is highly dependent upon adding to the ranks of blockbuster drugs. Only 3 out of 10 pharmaceuticals marketed ever generate enough revenue to cover the average cost of research and development for a single drug, which has almost tripled in the last five years from $318 million to $897 million.
The ability of pharmacogenomically derived drugs to become blockbusters may be circumscribed, since they are designed to target a subsegment of any given disease population. But, as confirmed with Herceptin, the prescription rate within that subpopulation can be expected to be substantially higher than for other treatments. Pharmaceutical companies also anticipate charging a slight premium, estimated at 10%, for these personalized medications.
Additionally, within such a tightly focused target population, researchers can anticipate a higher success rate than within a more general population, further enhancing the fiscal feasibility of pharmacogenomic drugs. This could mean that associated R&D costs get substantially trimmed. One study by the Tufts Center for the Study of Drug Development suggests that more efficient preclinical screens that boost clinical success rates from the current 20% to 33% would lower the average cost of a new drug to $560 million.
 (Subject to Terms and Conditions)
Additional charts accompanying text include:
Orphan Drug Approvals
Market Share of Pharmacogenomics Tools, by Company
Pharmaceutical Licensing Agreements, by Technology Type
 |
