In 2005, the Royal Society’s report Personalised Medicines: hopes and realities looked at the potential and limitations of pharmacogenetics, the use of genetics to predict an individual’s response to a drug. The report came at a time when the Department of Health was entertaining the idea of implementing pharmacogenetics within the NHS (see the 2003 White Paper, Our inheritance, our future – realising the potential of genetics in the NHS for more details). The RS Personalised Medicines report concluded that the use of pharmacogenetics was unlikely to change clinical practice overnight. Instead, it predicted that it would take 15-20 years for the true potential of personalised medicines to become apparent.

A good illustration of the potential of pharmacogenetics is the application of genetic diagnostics in warfarin dosing. Warfarin is a routinely-administered blood thinner that is processed in the liver by the enzyme CYP2C9. Patients with certain variants of CYP2C9 are thought to break the drug down at a slower rate (up to 20-fold variability between patients) and therefore require a reduced dosage. The identification of slow-metabolising patients is clinically advantageous as they represent a group at risk of being given more Warfarin than necessary and could suffer from undesirable side-effects as a result. Out of over 75 000 patients in the UK who currently receive warfarin, it is estimated that 8-26% suffer from severe side effects, including dangerous bleeding.

The strategy to shift away from a one-size-fits-all approach to medicine and towards grouping patients based on treatment efficacy is known as stratified medicine. In May 2012, the Technology Strategy Board announced a national roadmap to accelerate the development and uptake of stratified medicine in the UK, which will includes around £200m worth of investment over five years (see the upcoming Academy of Medical Sciences report, due to be published early next year, for their recommendations concerning regulatory, economic and infrastructure barriers to the implementation of stratified medicine).

Unfortunately, useful stratification may be less straightforward than it seems, as much of the variation observed between patients is due to factors other than genetics, such as age, lifestyle, presence of other illnesses and use of other drugs. For example, variation in warfarin metabolising enzymes only accounts for 30% of warfarin dosing variability, throwing into question current clinical benefits and cost-effectiveness of routine warfarin pharmacogenetics.

Encouragingly, recent advances in sequencing and collaborative data collection could help to improve stratification. Genome sequencing is becoming less expensive. In 2008, next generation sequencing technologies dramatically decreased the cost per Mb of DNA sequence, down from £63 in Jan 2008 to £2.4 in Oct 2008. In 2012, the cost is now down to 56p per Mb. On March 30th 2012, the UK Biobank, funded by UK research, charity and government institutions released a unique data resource of the health and lifestyles of half a million Britons – including 26 000 people with diabetes, 50 000 with joint disorders, 41 000 non drinkers, and 11 000 heart attack patients. In addition, data from the 1000 genomes project has been published online. The 1000 genomes project is an international collaborative endeavour which represents the largest collection of human genetics freely available to researchers worldwide.

Using these large databases of patient history and genomic data, researchers hope to untangle the complex interplay of factors behind clinical outcome variability, through the identification of medically useful signatures called biomarkers.

Translated into accurate diagnostic tests, these biomarkers could truly help to unlock the potential of personalised medicines. However securing this success will require supporting strong links between basic research, industry and health care practitioners.