You are here

Medicine is getting personal

In the future, more targeted and accurate prescriptions can be made with progress in the field of genetics

Soon, doctors may be able to predict who should take certain medications and who will suffer side effects. With personalised medicine, physicians may be able to use genetic profiles to make better treatment choices.

THE year is 2028 - you wake up with an awful headache and decide to pay your doctor a visit. In the doctor's office, he looks you over, listens to your symptoms, thoroughly examines you and prescribes you some medication. But first, the doctor takes a look at your genes.

That's right, your genes. Medication in 10 years' time may not look or work differently from those you are taking today, but they will be tailored to your genes. Many doctors expect that genetics will soon pervade medical treatment. This study of how genes affect a person's response to drugs is called pharmacogenomics. This new field combines pharmacology (the science of drugs) and genomics (the study of genes and their functions) to develop effective, safe medications and doses that will be tailored to a person's genetic makeup. This is the essence of personalised medicine.

Many drugs that are currently available are "one size fits all", but they may not work the same way for everyone. Your doctor figures out what drug and how much to give you based on your age, weight and general health factors. However, depending on your genetic makeup, some drugs may work more or less effectively for you than they do in other people. Similarly, some drugs may produce more or fewer negative side effects (called adverse drug reactions) in you than in someone else. Adverse drug reactions are a significant cause of hospitalisations and deaths.

Pharmacogenomics may also help save you time and money. By using information about your genetic makeup, doctors may be able to avoid the trial-and-error approach of prescribing medications. The "best-fit" drug for you can be chosen from the beginning.

Market voices on:

Imagine this scenario...

Your doctor ordered a pharmacogenomic test to decide how best to treat your high cholesterol. He might find a variant indicating that a commonly used drug would have little effect on reducing your cholesterol levels, but it would increase your risk of a heart attack or stroke. Luckily, you would not be given that commonly used drug due to this genomic information. Instead, another drug would be provided to you that lowers your cholesterol without negative side effects.

This is not a far-fetched scenario. Pharmacogenomic information is already being used today for a few health conditions. Take, for instance, abacavir, a commonly prescribed drug therapy for HIV, the virus that causes Aids. When first used, some patients developed severe rashes, fatigue, and diarrhoea - symptoms of a possible immune system reaction. Scientists looked at genomic variants associated with the immune system and finally identified one - called HLA-B*5701 - that causes the overreaction. Now doctors routinely test for the variant to find out who should avoid the drug.

Trastuzumab is used to treat metastatic (spread) breast cancer. It is effective against tumours that overexpress the HER2 gene. If the patient's breast cancer do not have high levels of HER2 protein then trastuzumab will not be effective. The drug is expensive and has known side effects on the heart. Therefore, selecting appropriate patients with pharmacogenomic testing to receive trastuzumab is important.

Genetic testing

The US Food and Drug Administration (FDA) now recommends genetic testing before giving the chemotherapy drug mercaptopurine to patients with acute lymphoblastic leukaemia. Some people have a genetic variant that interferes with their ability to process the drug. This processing problem can cause severe side effects and increase the risk of infections, unless the standard dose is adjusted according to the patient's genetic makeup.

Given the field's rapid growth, pharmacogenomics is soon expected to lead to better ways of using drugs to manage heart disease, cancer, asthma, depression and many other common diseases. A useful example in the field of heart disease to illustrate is the common blood-thinning drug warfarin. Most commonly prescribed for patients with atrial fibrillation, it is also used to prevent clotting events in patients with mechanical heart valves or deep vein thrombosis. It has excellent efficacy but also a narrow therapeutic window requiring fine titration of dosing. Underdosing puts the patient at risk of clotting complications while overdosing results in bleeding events. Variation in the CYP2C9 gene can reduce warfarin breakdown in the body putting patients at risk of bleeding during treatment.

In 2007, the FDA revised the label on warfarin to explain that a person's genetic makeup might influence response to the drug.

The FDA also is considering genetic testing for another blood-thinner, clopidogrel, used to reduce the rate of secondary heart attacks and stroke after a first event. Approximately 27 per cent of the population have a genetic variant of the CYP2C19 gene that makes the drug ineffective, putting them at risk of suffering from a second heart attack or stroke while on medication. If the information is available to the doctor, an alternative blood-thinner could be considered from the onset, reducing the risk of a second heart attack or stroke while on treatment.

Pharmacogenomics may also help to quickly identify the best drugs to treat people with certain psychiatric disorders. For example, while some patients with depression respond to the first drug they are given, many do not, and doctors have to try another drug. Because each drug takes weeks to take its full effect, patients' depression may grow worse during the time spent searching for a drug that helps.

Atomoxetine is a medication used to treat attention deficit disorder (ADHD). Those with genetic variation of CYP2D6 gene has been found to poorly metabolise the drug and are at risk of adverse drug reaction from overly high concentrations of the drug in the body with the resultant increased suicidal tendency.

The field of pain medication is another area where pharmacogenomic testing has potential usefulness. Codeine is a commonly prescribed medication for pain control. It is largely a prodrug, and its activity is primarily dependent on its conversion to morphine. Patients who have little CYP2D6 activity, the enzyme activity that converts codeine to morphine, are likely to have little response to codeine. The number of people with low CYP2D6 activity is substantial, making the drug ineffective for their pain control. The more dangerous situation, however, occurs when ultrarapid metabolisers take codeine. They may develop severe adverse effects from excessive morphine concentrations in the blood. In one tragic case, a healthy breast-feeding newborn infant developed fatal morphine toxicity; his mother was an ultrarapid metaboliser who was taking codeine, and her milk contained toxic amounts of morphine. Pharmacogenomic testing could potentially guide the dosing of codeine or a switch to a more appropriate alternative.

With the complete mapping of the human gene via the Human Genome Project (HGP) in 2003, anticipation was high that genetic information would radically improve medicine; that side effects would be more predictable, and that patients could be screened for likely drug responses. Thus far, progress has been much slower than what initial excitement anticipated.

Gradual progress

One barrier to widespread clinical implementation of pharmacogenomic testing is the lack of clear, curated, peer-reviewed guidelines that translate laboratory test results into actionable prescribing decisions for specific drugs. Several international consortia have come forth to develop guidelines, of which the Clinical Pharmacogenetics Implementation Consortium (CPIC) and the Dutch Pharmacogenetics Working Group (DPWG) are the most widely recognised. Currently, these consortia have evaluated over 100 gene-drug interactions.

The FDA also recognises that pharmacogenomics can play an important role in identifying medication responders and non-reponders. It mandates pharmacogenomic information appear be indicated on the labels of more than a hundred medications that are currently on the market. Before the Human Genome Project started, only four drugs carried such a label.

As DNA sequencing costs continue to decline and our knowledge increases, tailoring drugs to your genomic profile will become more common in medical practice. The eventual goal is to have pre-emptive broad-based pharmacogenomic testing for everyone early in life. The records will be made available through individual electronic medical records accessible by whichever doctor you chose to see, guiding therapy.

That will be when doctors can finally offer with certainty the right dose of the right drug to the right patient. I will be looking forward to that day.

This series is produced on alternate Saturdays in collaboration with Singapore Medical Specialists Centre