DNAprint Genomics (DNAG)

[falon]

Doctors' Prescriptions Get Personal

Genetic Testing Helps Predict Whether A Drug Will Be Effective, But Brings Hard Choices




http://www.ctnow.com/news/cusom/newsat3/hc-personalizedmed1019.
artoct19,1,4651057.story?coll=hc-headlines-newsat3



October 19, 2003
By WILLIAM HATHAWAY, Courant Staff Writer

A desperate parent of a child with attention deficit disorder now can get a genetic test that may help a doctor determine the optimal dose of a promising new drug.

Hartford Hospital is developing a similar test that will help predict whether an asthma patient is likely to be helped at all by commonly prescribed medications.

Oncologists already can test the genes of some cancer patients to predict who will benefit from powerful new drugs and who won't.

The long-predicted era of personalized medicine has arrived. Even simple new DNA tests have the potential to save thousands of lives lost to adverse drug reactions. Limiting prescriptions of drugs that don't work and identifying potentially dangerous drugs early in their development could save billions of dollars in health care costs each year.

Such a revolutionary change in medicine poses a host of questions in search of answers. Should new drugs be denied to desperate patients who, because of their genetic makeup, are unlikely to benefit from them? Are small risks of adverse side effects from some drugs great enough to justify the cost of genetic tests? Will the development of new drugs be delayed because such tests might not be available to identify those who might be harmed by them?

"Everybody is watching to see who can make it work," said Gillian Woollett, vice president for science and regulatory affairs at the Biotechnology Industry Organization, which has followed the issue closely.

Doctors have long known that people exhibit a wide variety of reactions to the same drugs. The revolution in genetics holds the promise that one day medical therapies might be customized to individual patients.

However, major advances in personalized medicine have been slow in coming, in part because scientists still are struggling to understand the root causes of many diseases. The goal is not only to identify the individuals who will respond best to existing drugs, but also to design new drugs that will work best given the genetic makeup of individuals.

For instance, doctors still do not understand exactly why existing drugs seem to alleviate symptoms of many brain diseases, such as schizophrenia, depression or attention deficit/hyperactivity disorder. To their great frustration, doctors can spend months or even years trying to figure out the right drug and dose for a patient.

"It's a huge problem. I can't tell you which person will respond better to certain medications," said Dr. Robert Sahl, assistant medical director of child and adolescent psychiatry at the Institute of Living in Hartford.

Sometimes the variation in the response to drugs costs patients more than misery, money and inconvenience. Adverse drug reactions cause more than 100,000 deaths in the United States annually.

These adverse reactions have various causes, but doctors and drug companies have known about one factor for decades: People metabolize drugs differently.

Metabolic Differences

Identifying those individual metabolic differences has become the first wave of personalized medicine to hit medicine's mainstream.

People have variations in their liver enzymes that determine how much of a drug will reach its intended target. For instance, among a large family of liver enzymes known as cytochrome P-450s, one - CYP2D6 - is involved in metabolizing about a quarter of all drugs.

Most people are "extensive" metabolizers - they process most active ingredients of a drug in the liver, allowing only small amounts to escape and reach their intended target. However, about 7 percent of Caucasians lack the CYP2D6 enzyme and are known to be poor metabolizers: They let much more of a drug's ingredients pass intact through the liver. Members of this group in theory are at higher risk of adverse side effects because they receive a higher active dose of the medicine.

Poor metabolism has been implicated in adverse side effects of some anti-psychotic drugs and in other effects, such as nullifying the benefits derived from codeine. Also, many anti-depressants inhibit CYP2D6 action, in effect turning "extensive" metabolizers into poor metabolizers, a fact that has led to at least one lawsuit. The suit claimed that genetic variations led to an overdose of Prozac and the suicide of a depressed adolescent.

There are also a smaller number of "hyper-metabolizers," who allow little or none of a drug to reach its target. These patients may receive little or no benefit from many drug treatments.

With the approval of Eli Lilly's drug Strattera in November 2002, the U.S. Food and Drug Administration for the first time ordered that metabolic information should be shared with the public. Billed as the first non-stimulant to treat attention deficit/hyperactivity disorder, Strattera is also the first drug packaged with information about how people with different metabolisms respond to the medication. Also included with Strattera's labeling is a note that genetic tests are available to identify poor responders.

Clinical trial results on Strattera did show that compared to extensive metabolizers, poor metabolizers had an elevated risk for several adverse side effects, including decreased appetite, insomnia and depression.

So why wouldn't patients and doctors demand genetic tests that identify metabolic status? asks Dr. Gualberto Ruano, the scientific founder ofGenaissance Pharmaceuticals, a New Haven pharmacogenomics company, and founder of Genomas, a new biotechnology company.

"I submit that if you want the state-of-the-art treatment for attention deficit disorder, you should have the metabolic state of your child analyzed," Ruano said.

However, the value of such genetic tests is not cut and dried, say Eli Lilly scientists who developed Strattera.

No statistically significant link between poor metabolizers and a serious medical ailment was found during trials of Strattera, said Dr. David Michelson, senior medical director of Lilly Research Laboratories at Eli Lilly.

Requesting genetic tests for metabolism might make sense when the risk of a serious side effect is high, but would be of limited or no value with a safe drug like Strattera, Michelson said.

In other words, is it worth paying $100 for a test to see if a child with attention deficit disorder has a slightly higher chance of having decreased appetite?

Officials at Hartford Hospital believe that as a new generation of drugs is approved, and more is learned about genetic influences on their safety and efficacy, more doctors will demand that their patients get such tests.

These new genetic tests will not reveal underlying health risks that might become a privacy concern, but will home in on specific physiological responses, said Gregory Tsongalis, director of molecular pathology at Hartford Hospital.

"I can't stress enough that we aren't talking about tests that will tell you anything about the disease or whether you are predisposed to disease," Tsongalis said. "If it is on the label, we think it will be almost a requirement that doctors ask for these tests."

Relief Or Crushed Hopes

Hartford Hospital is part of a consortium of hospital and university laboratories that are collaborating to develop new tests capable of detecting important genetic variations among patients. The tests will help predict adverse drug episodes and the potential efficacy of the drugs.

"We want to develop some simple tests to prepare for what we think will be a big transition in how people prescribe medicine," Tsongalis said.

For instance, he said, Puerto Ricans are more likely to have a genetic variation that makes them respond poorly to a family of drugs known as beta agonists, which are commonly prescribed asthma medications.

The savings from eliminating worthless prescriptions of drugs to non-responders could "be gigantic," he said. "Why take it if it won't do any good?"

However, that question may not be so easy if the disease is cancer.

Patients with some types of cancer are already among the first to experience the relief - or crushed hope - of having a genetic test that can determine whether or not they will benefit from a lifesaving drug.

The breast cancer drug Herceptin approved in 1998 was designed to combat metastatic breast cancer, which overproduces the HER2 protein. Genetic tests can identify the 20 percent of breast cancer patients who are good candidates for treatment with Herceptin - and the 80 percent who aren't.

New cancer drugs also are being developed that target known genetic abnormalities in tumors - which means new tests can be developed to determine who will best benefit.

But those tests cannot say with certainty that a patient with a genetic makeup that is a poor match will receive no benefit from a new cancer therapy.

"If there is a lower probability of success, but no other therapies [are] available, do you deny the patient treatment?" said Woollett of the Biotechnology Industry Organization.

New Era, New Rules

That is only one of many questions that have drug companies and biotechnology firms both hopeful and leery about advances in personalized medicine.

Pharmaceutical companies in theory at least can save tens of millions of dollars in drug development costs by using genomic data to quickly identify groups of people with gene types that put them at high risk of having adverse side effects or unlikely to respond to the medication.

On the other hand, drug companies are also concerned about being forced to share genomic data with competing companies seeking to develop similar drugs, Woollett said.

Pharmaceutical companies also want to make sure that approvals of some drugs not be held up because genetic tests aren't yet available to identify poor responders. It is also not clear whether insurance companies will pay for the new generation of genetic tests, she said.

The FDA has scheduled a November meeting in Washington to begin to formulate guidelines for sharing genomic data and other questions raised by drug companies.

In many ways, the regulatory issues that arise from the advent of personalized medicine are no different than those already facing federal regulators, said Dr. Janet Woodcock, director of the FDA's Center for Drug Evaluation and Research.

With or without genetic information, regulators must balance the potential benefits of new therapies against potential risk, she said.

"I don't believe this will be as challenging as some people believe," Woodcock said. "We've been doing this for a very long time."

No matter what guidelines the FDA eventually adopts, personalized medicine is here to stay, she predicted.

"I think the curve will be slow and flat at first and then really take off," Woodcock said. "And we're on it."