Drug efficacy: tough new targets

Most drugs work in fewer than one out of two people;
wouldn't it be nice to know who's who?

By Susan Usher

Meet the P450 family: Coming soon to a lab near you The P450 group of drug-metabolizing enzymes is produced in the liver and oxidizes foreign chemicals. Two genes of the 57 for P450 enzymes play an important role in drug metabolism. The polymorphic CYP2D6 is involved in the metabolism of 25% of prescribed drugs, including beta-blockers and antidepressants. The polymorphic CYP2C9 is involved in the metabolism of 5% of drugs. Both have been linked to the failure of commonly prescribed drugs such as codeine (which depends on oxidization into morphine by CYP2C9 and is ineffective in about 10% of the population), and potentially Prozac. Warfarin doses safe in most patients can induce life- threatening bleeding in the small group with CYP2C9 polymorphisms that reduce the activity of the metabolic enzyme. Pharmacogeneticist Magnus Ingelman-Sundberg from the Karolinska Institute in Stockholm, Sweden, estimates that 80% of serious drug reactions involve drugs metabolized by polymorphic P450 enzymes. Even now, genomics companies are manufacturing DNA microarrays to identify common SNPs that effect activity of the P450 enzymes. Some hospitals are now starting to use them to identify people who may suffer serious adverse reactions, especially to cancer drugs. As the speed of analysis grows and the cost falls, we can expect genotyping for this set of SNPs to become routine in drug trials, diagnostics and, of course, prescribing.

It is a familiar phrase, and something that many physicians might have already been aware of -- not all drugs work. And with GlaxoSmithKline senior vice-president of genetics research Dr Allen Roses going public with a similar opinion, more and more patients are becoming aware of this situation. In late 2003, Dr Roses announced to the world that most prescription medicines don't work on most people who take them. This statement got him into hot water -- and onto the front pages of newspapers worldwide.

While it made for sensational reading and may have come as a surprise to a few, undoubtedly healthy citizens, the fact that drugs don't work for everybody is hardly news. The exciting thing -- and what Dr Roses was actually discussing when he made the remark -- is that we are heading into an era of genetic profiling. This will allow doctors to identify which patients are most likely to respond to a given therapy and find alternatives for those who carry genes that interfere with the metabolism of the medicine.

WHERE IS THE RESEARCH GOING
Here in Canada, Dr Pavel Hamet, professor of Medicine and director of Research at the Université de Montreal Medical Centre, is studying the pharmacogenetics of hypertension in an attempt to refine both the diagnosis of the disease and the therapeutic strategies used to treat it.

"The way we do clinical trials today," explains Dr Hamet, "is to study drug effect in thousands of people and if the drug works as a mean, then we apply it to everyone. But if you look at the details, up to 50% of subjects do not respond to these drugs."

Knowing who is who would improve the effectiveness and cost-effectiveness of therapy. "Right now, in hypertension, my only recourse is to give all my patients Drug A for three months, then if that doesn't work, either add Drug B or switch to Drug B," Dr Hamet states. "And if I add Drug B, I'm unlikely to stop Drug A because I don't know whether it's the combination that works or Drug B on its own." In the future, he anticipates being able to obtain a blood sample analysis that would tell him which drug a patient would respond to.

Dr Hamet's research team has studied a population from the Saguenay-Lac St Jean region in Quebec, which is a relative genetic isolate to identify phenotypes that predispose to hypertension. They are now completing a prospective pharmacogenetics study where patients with hypertension are assigned drug therapy according to their alleles. "We have to test genetic individualization with the same scrutiny we do a new drug," he says. "The test needs to be validated and then put to prospective trial."

Dr Hamet represents a small group of clinician researchers around the globe who are enthusiastically pursuing pharmacogenetics research. Governments, pharmaceutical companies and many doctors are still on the fence about investing in individualizing therapy. Dr Jeff Johnson, associate professor at the University of Alberta, Canada Research Chair in the Department of Public Health and fellow at the Institute for Health Economics thinks we have ample evidence about the effectiveness of many drugs and the key now is to make sure that people are getting them. He sees significant treatment gaps because of health system barriers and because people go to their family physician for acute concerns rather than prevention and screening of chronic conditions.

"We have a lot of cleaning up to do in implementing what we already know," he says. His research team has identified large treatment gaps among people with diabetes who have high blood pressure or elevated cholesterol. "They're not getting diagnosed and they're not getting the drugs we have. The genetic factor isn't even an issue."

Dr Johnson does admit, however, that knowing which drug to start with could help, especially when physicians are so overwhelmed with acute problems that follow-up to monitor treatment effectiveness and side effects suffers. "A three-month trial of a drug can easily stretch to six months or more before a follow-up visit is arranged," he says.

GOVERNMENT SUPPORT
Dr Hamet attributes government timidity about pharmacogenetics testing to short-term cost concerns and the immature stage of the tests themselves. However, the profiling of the single nucleotide polymorphisms (SNPs) used to highlight polymorphic genes that influence our response to individual drugs is a fast-growing sector (annual expenditure on SNP research is predicted to grow from $158 million US in 2001 to more than $1.2 billion in US 2005, according to a report in Nature in October 2003) and the costs of tests are falling fast. Add to that increasing concerns about drug safety and governments are starting to respond. Dr Hamet senses greater enthusiasm for genetics from Quebec's new health minister, Dr Philippe Couillard. The US National Institute of Health has established a Pharmacogenetics Research Network and the UK Department of Health has earmarked $6.7 million US over the next three years for pharmacogenetics and is also looking into the ethical issues of pharmacogenetics, a much needed examination as we stand to see important changes to the way drugs are studied, labelled (The US FDA is already considering labelling some drugs as suitable only for people with a defined genetic profile) and covered.

As far as industry goes, companies have been collecting DNA to use in retrospective genetic analyses of clinical trials, but have not yet introduced them prospectively. There is concern about limiting market size to certain genetic subpopulations and about increased costs of clinical testing. Within this environment, Dr Roses stands out as a bold pioneer. GlaxoSmithKline has several large pharmacogenetics projects underway and has used retrospective analysis of genetic profiles of participants in clinical trials to help identify patients likely to have severe adverse reactions to the AIDS drug abacavir. He is confident they are on the right path and that it will, in the end, produce more drugs that work and are safe to use.