In the near future - maybe five years, maybe ten, but definitely within my lifetime - it will be technologically possible to make medical decisions based on an individual's specific genetic makeup. I was convinced of this at a recent symposium, "Individual Genetic Variation: Implications of the Coming Transformation of Medicine," sponsored by the Stanford University Program in Genomics, Ethics, and Society. It was an intellectual journey exploring the possibilities of what the future might hold in the upcoming genomic age.

However, although any number of technologies might be possible, other factors determine whether they become common practice. Aside from the obvious ethical and legal issues associated with genetic testing, the "so what" factor could prevent doctors and patients from enthusiastically embracing the practice. Simply stated, it does patients little good to know they carry a susceptibility gene for Alzheimer's or a predictive gene for Huntington's if they can do nothing about it except look forward to a dismal future.

The so-what factor could be decisively eliminated through pharmacogenetics - the development of drugs tailored to the genotype of the patient (or the patient's tumor or pathogen). Because such drugs are targeted very specifically, they hold great promise of being both highly effective and safe. Most major pharmaceutical companies have initiated pharmacogenetic drug development programs in the last few years. The pharmacogenetic approach not only targets clinical trials to the most responsive patient population, but also may resurrect drugs that have failed clinical trials because of low response rates in the general population.

One drawback, however, to the pharmacogenetic approach, generally less publicized than the benefits, is that pharmacogenetic drugs usually are targeted to smaller markets than are traditionally developed drugs. Given the high cost of bringing a drug to market, estimated at $500 million per successful drug, pharmaceutical companies typically analyze whether a drug will reach certain sales goals before entering human clinical trials. If the market for a drug is perceived as being too small, the company often opts to discontinue development, even if things are going well in the lab. This roadblock can be cleared if the drug with a small market is able to receive orphan-drug status.

The U.S. Food and Drug Administration (FDA) created orphan-drug status to encourage companies to develop certain drugs for rare diseases - diseases or conditions afflicting fewer than 200,000 people. Companies that develop these "orphan" drugs receive such generous perks as tax breaks on clinical trials and seven years of marketing exclusivity. As noted, the market for a drug developed pharmacogenetically could fall into this category.

Herceptin is one such drug. Discovered by Genentech and just recently approved by the FDA for the treatment of a subset of patients with metastatic breast cancer, Herceptin is a monoclonal antibody directed against the HER2/neu protein (a cell-surface receptor) that is present in all normal tissue and overexpressed in some breast tumors. Before a woman receives Herceptin, her tumor must be tested to ensure that it is a HER2/neu overexpresser and that she is thus eligible for treatment. An estimated 30 percent of all breast cancer patients have tumors that overexpress the HER2/neu receptor, and it is presumed (although not proven) that the drug will be effective only in patients with HER2/neu overexpressing tumors.

Approximately 165,000 people in the United States have metastatic breast cancer. Of these, 30 percent overexpress HER2/neu - clearly under 200,000. Yet Herceptin was denied the orphan-drug designation. One does imagine that treatment of breast cancer was not exactly the scenario the FDA had in mind when creating this special class. Indeed, as more companies take a pharmacogenetic approach to drug development for traditionally large disease areas and these drugs advance in the pipeline, a dramatic increase in requests for orphan drug status is likely.

While declining to comment directly on the Herceptin case, Michael Dreis at the Office of Orphan Products Development (OOPD) explained that products are most commonly denied orphan-drug status because of disagreements over how the target population is defined. In other words, companies tend to carve markets up creatively so that they can claim fewer than 200,000 patients, and the OOPD scouts for reasons why the population should be larger. Nevertheless, at first glance, Herceptin seems to qualify without question as an orphan drug.

The OOPD defines a drug's patient population as the total expected treatment population, not as only those patients whom the company identifies as eligible for clinical trials, or for whom they can get an FDA indication. For example, companies are often able to enroll in trials only patients for whom standard therapy has failed; thus the drug is only approved for use with these refractory patients. There is no scientific reason, however, why the drug could not be used as first-line therapy. Trying for an orphan-drug designation, companies may attempt to define the market as refractory patients only, but the OOPD would include the entire potential patient population.

In the Herceptin case, the OOPD may have questioned whether Herceptin legitimately could be restricted to metastatic patients. In fact, HER2/neu is overexpressed in multiple cancers, not just in breast cancer. A potential patient population thus enlarged may have resulted in the denial of the orphan-drug designation. In any case, Dreis emphasized that the OOPD does not treat pharmacogenetic drugs differently from other drugs. If a pharmacogenetic drug were to be submitted that was scientifically appropriate for use only in a small subset of a larger population, it would have no trouble achieving the orphan-drug designation.

It would appear, therefore, that there is no immediate cause for concern regarding the reluctance of the OOPD to grant orphan status for drugs targeted to subsets of large populations. However, long-term questions remain. It is hard to imagine that there would be no legislative response if within ten years pharmacogenetic techniques take off, and half of all drugs developed are "orphans." The waters could be further muddied over definitions of eligibility. If a drug is found to be significantly more effective in one sub-population, but it also has minimal usefulness in the general population, which group would the OOPD use in its calculations? There has been speculation that HMOs are likely to limit coverage for such drugs to the optimal population - one wonders if the OOPD would follow suit.

Another possibility is that the orphan drug designation will matter less as pharmacogenetic approaches bring down the cost of drug development. It has been hypothesized that the use of pharmacogenetics will increase pharmaceutical companies' success rates in drug development. The $500 million price tag for bringing one drug to market incorporates development costs for 4,999 other drugs that fail. The actual cost of developing a single drug is closer to $75 million. With pharmacogenetics, although the market for each drug may be smaller, companies may be able to produce more drugs for less money. However, the opposite side of this coin is that pharmacogenetics may actually increase the cost of running clinical trials, for example requiring enrollment of more participants to sort out multiple combinations of relevant genotypes.

Whether drug development through pharmacogenetics will be more or less expensive than traditional techniques remains to be seen, as does whether these drugs will be safer or more efficacious. Companies undertaking this new approach to drug development are assuming a great risk. Nevertheless, drug discovery through pharmacogenetics is logical and promising. If the advantages created by the orphan-drug designation can encourage pharmaceutical companies (and biotech companies) to take a pharmacogenetic approach to drug development, I believe we will all benefit down the road as consumers of highly directed therapeutics. If we are to eliminate the "so what" of genetic testing, we need more Herceptins. If the orphan-drug designation can be a means to that end, I hope it is used often.