Should “race” be a criterion for inclusion in a clinical trial—and, by extension, is it appropriate for drug labeling to mention it specifically? For instance, should a drug’s label say that only “white people” should take it? Those are complicated questions, but the simple answer is that you go where the data take you.
Clinical trials are not designed to show the effectiveness of a treatment (drug, medical device, or other intervention) in a completely random sample of people in the general population. Instead, researchers use a variety of strategies to select a subset of patients in whom the intervention will likely be easier to demonstrate. This concept is called “enrichment,” which is defined by the FDA as “the prospective use of any patient characteristic to select a study population in which detection of a drug effect (if one is in fact present) is more likely than it would be in an unselected population.”
As the FDA observes in an agency guidance document on enrichment strategies, “Some of these selection strategies are obvious (e.g., patients are enrolled only if they have the disease that the drug being studied is intended to treat), but there are many more ways in which patients are typically chosen to make detection of a treatment effect more likely.” An increasingly important approach is the use of biological indicators, or "biomarkers"—such as certain DNA sequences or the presence or absence of drug receptors—as an indicator of the likelihood that the intervention will be effective and, therefore, whether a given patient should be eligible for the clinical trial.
This concept is not new. It has been known for decades, for example, that persons genetically deficient in an enzyme called G6PD can experience severe and precipitous anemia if they are exposed to certain drugs, so these patients should be excluded from clinical trials of that class of drugs. Several years ago, prognostic biomarkers began to make a big difference in cancer therapy. Drugs such as Erbitux and Vectibix only work in tumors containing the normal version—but not the mutated variant—of a gene called KRAS. If mutations of KRAS are present, the drugs are ineffective.
Once such phenomena become known, they have obvious implications for clinical trials: If you know which patients to exclude, you protect them from possible side effects from a drug that will not provide benefit and you avoid delays in beginning interventions that might work.
Such "enrichment" for patients who are likely to experience a beneficial effect in clinical trials enables drug companies to perform smaller, better-targeted clinical studies in order to demonstrate efficacy. The reason is related to the "statistical power" of clinical studies: In any kind of experiment, a fundamental principle is that the greater the number of subjects or iterations, the greater the ability to detect small effects and the greater the confidence in the results of the study. Small studies generally have large uncertainties in results, unless the effect of the intervention is very potent. And that is where enrichment makes a difference, helping researchers to design clinical studies that will show a high "relative treatment difference" between the drug and whatever it is being compared to (often a placebo, but sometimes another treatment).
Genetic markers are obviously useful for obtaining enrichment, but how about more subjective factors such as “race” or ethnicity?
Dr. Abigail Zuger, a professor of medicine at Columbia University in New York, has strong opinions about the subject: “It has been clear for decades that race has minimal relevance to the body's inner workings. Research has repeatedly shown that the biologic variations among individuals of the same race are reliably great enough for race to retain little utility as a biologic predictor. You might as well sort people by height.”
Zuger’s disdain for sorting by height is misplaced: If you were testing a growth-promoting drug in children of short-stature, for example, you would select only subjects of less than average height for the trial. (In fact, as an FDA medical officer, I reviewed such a trial to test the safety and efficacy of genetically engineered human growth hormone.) Perhaps belaboring the point, Zuger goes on to quote from a 2005 editorial in the journal Nature Biotechnology, “Pooling people in race silos is akin to zoologists grouping raccoons, tigers and okapis on the basis that they are all stripy.”
Such uncompromising, sardonic statements appear to owe more to political correctness than to human biology—or to the actual practice of medicine. In fact, because they do sometimes reflect “the body’s inner workings”—that is, differences in physiology—diagnostic tests and therapeutic interventions targeted at certain ethnic or racial categories can be useful.
In order to screen for a number of genetic diseases that occur predominantly in Jews of Ashkenazi, or Eastern European, descent, for example, there is a group of genetic tests called the Ashkenazi Jewish Genetic Panel (AJGP). Similarly, sickle-cell anemia is found disproportionately frequently in blacks, and a hereditary enzyme deficiency that causes sensitivity to fava beans and certain drugs is found primarily in Africans and persons of "Mediterranean" descent; after exposure to quinine-like drugs, as many as 10 percent of black men develop a serious condition in which red blood cells lyse, resulting in severe anemia.
Thus, race or ethnic origin—although far less precise than molecular markers—can sometimes serve as a useful surrogate for more precisely defined genetic differences.
There are other kinds of groupings that may have diagnostic or therapeutic implications. Although not due to genetic differences, some of the most important variation seen in response to drugs is due to aging; for several reasons, older patients are far more likely to experience adverse drug reactions. For one thing, clearance by the kidneys and liver—the two most important routes for the elimination of drugs—is reduced; as people age, these organs get less blood flow, and there is diminished activity of the hepatic enzymes that metabolize drugs.
Another interesting age-related anomaly concerns the decrease in total body water and the relative increase in body fat seen in older people. Because of these changes, water-soluble drugs become more concentrated in the blood and fat-soluble drugs have longer half-lives. Moreover, serum protein levels are decreased in the elderly (especially if they're sick), which reduces the protein binding capacity of the blood, and leaves more free—that is, active—drug circulating. (And yet, surprisingly few physicians routinely reduce drug dosages in older patients.)
But let us return to Dr. Zuger’s comments, made in the context of reviewing a book about a cardiac drug called BiDil, which was approved in the United States in 2005 specifically for black patients (although it can be prescribed off-label for anyone). Due to the lack of evidence of efficacy in early clinical trials, the drug, which is actually a combination of two proven cardiac medicines (the vasodilators hydralazine and isosorbide dinitrate), had been rejected by the FDA nine years previously for approval for patients of all races. But because analysis of the data in various subgroups revealed a suggestion of benefit to black patients, another trial was performed on 1,050 self-identified black patients with severe heart failure who had already been treated with—but had not responded to—the best available therapy.
The results were so striking—43 percent reduction in mortality and 39 percent decrease in hospital visits among patients who received BiDil—that the study was stopped early and the drug was approved. This is an enrichment success story. Since its approval, BiDil has not been a great commercial success but it remains on the market.
Commentators have expressed a wide spectrum of views about the appropriateness of a therapy designated for one racial group, some even calling it discriminatory. Francis Collins, then director of the U.S. National Human Genome Research Institute (and now head of the National Institutes of Health) said at the time BiDil was approved that "we should move without delay from blurry and potentially misleading surrogates for drug response, such as race, to the more specific causes."
He was correct, of course. But you go to war against illness with the data you have, not the data you wish to have. Political correctness notwithstanding, drug testing, approvals, and labeling must go wherever the evidence leads.
Henry I. Miller, MS, MD, is the Robert Wesson Fellow in Scientific Philosophy and Public Policy at the Hoover Institution. His research focuses on public policy toward science and technology encompassing a number of areas, including pharmaceutical development, genetic engineering in agriculture, models for regulatory reform, and the emergence of new viral diseases.