The challenge

Each of us metabolises drugs in different ways. The liver contains proteins called CYPs (short for cytochrome p450, pronounced ‘SIPS’) which process a range of chemicals and toxins and makes them less harmful. Variations in the genes encoding CYPs can impact how the proteins process drugs and in turn influence how different people respond to the drugs when treated. For example, the painkiller codeine is a ‘prodrug’ and is actually inactive until metabolised by the liver into an active form, morphine. The CYPs in some people cannot effectively convert codeine into its active form and so it is largely ineffective. More concerningly, some people metabolise codeine extremely efficiently and may experience toxicity, almost equivalent to a morphine overdose. Introducing CYP genetic tests in clinical practice could improve patient safety, treatment effectiveness and potentially cost by helping clinicians choose medicines and tailor medicine doses to match individual metabolism. However, there are technical and ethical hurdles to overcome before this can happen.

The solution

Through Better Care, a team led by researcher and practicing GP Professor Bruce Guthrie will investigate how pharmacogenomic testing could be implemented in live prescribing. This will involve examining large, linked datasets, including dispensing prescribing data on around 400,000 people in the Tayside area. These patients have also had genetic tests performed on some of the genes encoding CYP proteins which influence how they respond to the different drugs. This linked data will enable the team to look at prescribing pathways and the potential benefit of using pharmacogenomic testing for commonly used medicines. In addition, the team is conducting qualitative interviews with GPs, other prescribers and members of the public to understand attitudes towards, and barriers to implementing this testing.

Professor Guthrie comments:

“It is unlikely that GPs will be actively against the idea of using pharmacogenomic information; but there are concerns and challenges. For example, GP computer systems routinely trigger alerts during prescribing, most of which are irrelevant. Too many alerts slow down prescribing and can create alert fatigue where important messages are missed.”

“And similarly, patients may have concerns about who will be able to see their genetic information. Implementing a new kind of pharmacogenomics alert system therefore needs careful design to ensure it addresses clinician and patient concerns.”

Impact and outcomes

A fully developed, live point-of-care pharmacogenomics capability in general practice could bring considerable benefit in terms of patient safety, treatment effectiveness and potentially cost. It would allow a more a personalised and tailored approach to prescribing, where patients are given treatments that are more likely to be effective and less likely to cause serious side effects. This would save time and resources, for example avoiding patients starting a treatment, only for them to stop because of side effects or lack of any benefit.