Rx4U: More Drug Firms Are Trying to Get Personal

By Gautam Naik, The Wall Street Journal

 March 25, 2002

It is one of the biggest dilemmas in modern medicine: Every day, doctors diagnose patients and prescribe the most appropriate drug, even though in many cases the drug won't work or will trigger miserable side effects.

In a perfect scenario, a doctor could perform a genetic scan of a sample of a patient's DNA, determine if a drug will work and tailor the medicine to suit that individual.

"Personalizing" drug treatment has been one of the hottest areas of research since the human genome was mapped nearly two years ago. Drug companies are making big advances, with laboratory experiments starting to shed light on how tailored medicines could help people with ailments like cancer, AIDS and obesity.

Plenty of hurdles must be overcome before this new method of prescribing drugs is accepted. It isn't clear how expensive such tests will be, or whether they will be covered by insurance. The use of patients' DNA also raises questions about privacy. And targeting patients so narrowly isn't a pure blessing for the large drug makers, which have prospered by selling blockbuster drugs with a one-size-fits-all approach. They may find it less profitable to develop variations to fit different genetic profiles.

Nonetheless, many are forging ahead. Genetic tests are already used to predict which patients will respond well to certain drugs. The Mayo Clinic in Rochester, Minn., uses a routine blood test that can identify the 10% of childhood-leukemia patients who, because of certain genes they carry, find it difficult to metabolize a particular drug. Roche Holding AG of Switzerland and Genentech Inc. of San Francisco offer a similar test for their drug herceptin, used to treat aggressive breast cancer. The companies found that 25% to 33% of breast-cancer patients carry a "switched on" version of a gene that makes them susceptible to more aggressive tumors. For others, herceptin wouldn't work. Without such targeting, Genentech says, herceptin would have failed in clinical trials.

Newer techniques under development involving more sophisticated screening of DNA may take three to five years to show up in clinics. But early results are encouraging.

Scientists at GlaxoSmithKline PLC have long known that about 5% of the patients who take its AIDS drug abacavir suffer a severe allergic reaction. Recently, they combed the DNA of patients and identified two variant genes that seemed to be associated with the side effect. The upshot: One day, doctors may be able to do a genetic test that shows which patients are likely to suffer the adverse reaction.

Glaxo's genetic test is a long way from being commercially available, but "we have proof of principle" that it works, says Allen Roses, head of genetics at the company.

Lured by the idea of personalized medicine, many large drug makers now routinely collect DNA samples from patients enrolled in clinical trials. By being able to target more "responsive" patients, companies can show increased effectiveness for their drugs. That, in turn, could help them get speedier approval from regulators.

In January, Gennaissance Pharmaceuticals Inc. completed a pilot study of 100 German patients with advanced heart disease. The patients were treated with several cholesterol-lowering statin drugs, including Pfizer Inc.'s Lipitor and Bristol-Myers Squibb Co.'s Pravachol. While these drugs are known to lower "bad" cholesterol, their effect in raising levels of "good" cholesterol is less clear. Gennaissance wanted to see if it could find a subset of patients who shared some common genetic variation that helped them benefit from a statin "double effect" -- a higher level of "good" cholesterol and a lower level of "bad" cholesterol.

The New Haven, Conn., company says it studied 10 genes associated with how cholesterol is transported in the bloodstream. One such gene was shared among 25 or 30 patients -- those who benefited from the double effect. The small study was far from conclusive, but the company plans to announce its results at a meeting of the American Heart Association in Hawaii next month.

Targeting Your DNA

Some experiments in personalized drugs:

(1) These drugs are commercially available, but efforts to tailor them based on genetic profiles are at an experimental stage only.

(2) This drug was discontinued after failing efficacy test in clinical trials.

Source: WSJ reporting

This week, the company is expected to release initial results of a yearlong study involving 800 patients and 200 genes, which may show stronger links between genetic variation and the double effect of statins. Gennaissance hopes to license such tests for use in prescribing drugs.

Genes can play a vital role in determining how people respond to medicines. They make receptors -- molecules that sit on the surface of cells -- and many drugs need to attach themselves to these receptors to work properly. Genes also make enzymes that determine how much of a drug is absorbed or eliminated by the body. Someone who carries a defective gene might produce too little of an enzyme and therefore experience an adverse reaction to a drug.

About 1% to 3% of all patients are known to be "poor metabolizers" of common medicines, and that has prompted Orchid BioSciences Inc. to build an entire business around the idea. The Princeton, N.J., company is focused on five enzymes known to interact with 300 of the most commonly prescribed medicines. Orchid is developing a panel of DNA tests for these medicines that would let a doctor quickly determine how well a patient metabolizes a drug. The doctor could use that information to establish the proper dose. Orchid says it will begin experiments in a few months and expects the test to be ready next year. It will cost less than $200 to administer, the company says.

But the prospect of tailoring drugs has also triggered a debate at some companies that are cautious about straying from the one-size-fits-all model. In recent human clinical trials of a promising obesity drug, GlaxoSmithKline found that patients who carried three genetic markers were more likely to benefit from the drug than those who didn't carry those markers. Now Glaxo is trying to figure out whether it should kill the drug or risk hundreds of millions of dollars to keep developing it. "The question is whether you want to go to the next step in development for something that has a 35% to 40% success rate," says Glaxo's Dr. Roses.

In a recent paper in the Pharmacogenomics Journal, two senior officials of the Food and Drug Administration state that the FDA is keen to receive more drug applications based on genomic profiles. "We now have some support from regulators," says Dr. Roses. "It's a major, major change in the landscape."

Write to Gautam Naik at gautam.naik@wsj.com

Corrections & Amplifications:

Genaissance Pharmaceuticals Inc. of New Haven, Conn., was incorrectly spelled as Gennaissance in the above article.