Share |





The New Frontier for Drug Safety
Personalized Medicine

Robert Goldberg, Ph.D., Gualberto Ruaņo, M.D., Ph.D.
February 17, 2005

It is now clear that the needs of the pharmaceutical industry, the FDA, and the public at large for improved drug safety can be addressed with personalized medicine in ways that the current "one size fits all" model could not even begin to reach.

Someday in the not so distant future, doctors will be able to offer patients medical treatments that are tailored to their own unique DNA profiles. This approach - known as personalized medicine - can keep valuable drugs that are safe and effective for the vast majority of patients on the market by pinpointing those at risk of side effects and underscoring the reality that no drug could possibly be safe for everyone. The rush to withdraw drugs because of side effects and launch multi-billion dollar liability suits to recoup patient damages would become a thing of the past.

Personalized medicine approaches drug safety with two simultaneous approaches: pharmacokinetics and pharmacodynamics, technical terms which can be summarized, respectively, as "what the body does to the drug" and "what the drug does to the body."

We now have a very mature understanding how genomic variability affects how the human body reacts to drug treatment. Cytochrome P450 (CYP) is a family of enzymes representing a major part of the body's powerful detoxification systems. The CYP enzymes also metabolize most of the pharmaceutical compounds given as therapeutic drugs.

Genomics has also supplied researchers with databases of functional DNA variability in the CYP family. The CYP variants collectively produce hundreds of DNA variations - meaning that patient populations have a wide spectrum of reactions to specific drug compounds.

The good news is that it is now a routine part of drug development to assess the main CYP pathways for drug metabolism and to help predict patient response and potential adverse effects. The bad news is that none of this wealth of genomic information has been translated into DNA diagnostics that doctors can use to help prescribe drugs for their patients.

Is this a problem? Yes. A very serious one.

Clinically, the population can be divided into three broad groups according to how they are expected to react to drug treatments. Extensive metabolizers represent 50% of the population and are thus the "normal" group, which can be administered drugs following "average" dosing practices. In contrast, poor metabolizers represent 10% of the population with a "deficiency" in drug metabolism, which could lead to life threatening drug accumulation and severe adverse reactions. Intermediate metabolizers represent 40% of the population and also face safety risks, because drug interactions can turn them into poor metabolizers by inhibiting their CYP gene expression. If doctors could identify poor and intermediate metabolizers ahead of time they could reduce drug dosages and increase patient monitoring.

With genotyping, patients and drug companies will be able to easily detect and avoid all too common reasons drugs fail in development or fail to work at all: toxicity and side effects due to the different ways drugs are metabolized. Translation: CYP diagnostics could be used to personalize drug treatment to ensure that every patient gets the most safe and effective course of treatment we can deliver.

Up to now, treatment has been driven by almost blind optimism. Doctors and researchers tend to think that it is the drug that dictates the effect on a patient. In reality, we now know that it is the human body that dictates the effect of the drug.

Consider, for example, the interplay of "atypical" antipsychotic drugs (AAPs) and metabolic syndrome.

Metabolic syndrome is a combination of abdominal obesity, abnormal cholesterol, high blood sugar, and high blood pressure - a recipe for diabetes and heart disease. AAPs are associated with drug-induced metabolic syndrome (DIMS) after only three months of therapy. However, AAPs are recognized as valuable treatments not only for schizophrenia, but also for bipolar disorder, depression and obsessive-compulsive disorders.

In September 2003 the FDA asked all manufacturers of AAPs to add a warning statement to the labeling of these drugs describing the increased risk of hyperglycemia and diabetes in patients.

Still, DIMS develops in some but not all patients. Genomic differences in patient metabolism could account for the variability in drug-induced weight gain. Better diagnostics would allow personalized drug treatment to minimize potential side effects and improve long term compliance with drug therapy that is essential to patient well-being and recovery.

Possible interventions include adjustment of dosage, nutritional counseling, and close monitoring for individuals at risk of DIMS. Other drugs could be brought to bear that counteract the risk of developing diabetes and heart disease if patients fail to respond to other psychiatric treatments.

Patient responses to psychiatric drugs are highly variable, and many patients will experiment with several drugs before finding one that works - if they are lucky. Patients benefiting from these drugs are mostly young, and will require life long pharmacotherapy. Consequently, psychiatry is one of the areas of medicine that has the most to gain from personalized medicine.

But if we do not give doctors the tools to better prescribe AAPs now, these drugs could easily morph into the next "Vioxx" and result in a flood of lawsuits, public acrimony, and eventual withdrawal of these valuable medicines.

Positioning each individual along a continuum of risk based on physiological and DNA markers would be an enormous gain for physicians as side effects are prevented or treated, and catastrophic complications are avoided. Industry can help to advance personalized medicine through integrated physiological genomics research on patient outcomes, the real world of clinical practice after the controlled trials of drug development.

Meanwhile, the FDA and its stakeholders need to re-evaluate their entire approach to product safety, liability and surveillance. Until now, there has been an understandable tendency to rule out side effects as necessary by-products of therapy. In personalized medicine, every side effect will be scrutinized and the paradigm that they are unavoidable will simply become unacceptable - analogous to building and selling a new car without seatbelts.

Public opinion will have to adapt as well. Even extremely rare and catastrophic side effects that now require drug withdrawal from the market can be identified and detected. Adverse events can be linked to genomic differences before drugs hit the market or before they are prescribed. That means drug withdrawals - now the cause for rampant public skepticism of industry and the FDA and an attractive target for class action lawsuits - will become a thing of the past.

By focusing on both common and rare side effects, the practice of personalized medicine can make drugs safer and make drug development more efficient and less costly. Companies, insurers and patients can all benefit from treatments patients know fit their own unique DNA profile.

Personalized medicine might inaugurate a golden age for the pharmaceutical industry, as new genomic technologies expedite regulatory approval and launch of new drugs that are safer and more effective, validating market pricing and ensuring the gold standard for successful products: consumer confidence.

Dr. Gualberto Ruaņo, M.D., Ph.D., is the President and CEO of Genomas (, which develops novel diagnostic products for personalized medicine. Robert Goldberg, Ph.D. is director of the Center for Medical Progress at the Manhattan Institute for Policy Research.

home   spotlight   commentary   research   events   news   about   contact   links   archives
Copyright Manhattan Institute for Policy Research
52 Vanderbilt Avenue
New York, NY 10017
(212) 599-7000