21 March 2016
A recent update published in BMJ of an earlier review of studies examining the communication of genetic risks of disease and impact on risk-reducing behaviour concluded that telling people about increased genetic predisposition towards disease did not encourage them to change their lifestyles to mitigate that risk.
A piece published in the Independent newspaper reporting on the paper said the findings ‘could change how personalised medicine is used’.
We disagree. The PHG Foundation has been working on the health applications o f genetics and genomics for almost twenty years and, in common with most people working at the cutting edge of personalised medicine, we are well aware of the general lack of eviden ce to suggest that any form of communication about risk - genomic or otherwise - motivates sustained behavioural change.
Since motivating health-related behaviour change is extraordinarily difficult in general, it is not really surprising that genomic information does not confer greater influencing power than other types of risk information. The public has more sense than to react differently just because of the ‘genomic’ origins of the information when the absolute risk estimates are very similar. Re-stating and updating the poor quality evidence in this area is certainly not grounds for radically changing our ideas about how personalised medicine will be used.
Genomics has an important role to play in health improvement, including personalised disease prevention, as one of a number of relevant biomarkers of common, complex disease susceptibility – that is, the many diseases influenced by multiple contributory genetic and environmental factors. The PHG Foundation believes that using combinations of these biomarkers will enable us to understand disease at a more basic molecular level and drive the development of interventions to be offered to population subgroups and individuals that are tailored to the underlying cause and targeted according to the specific level of risk.
Thus we agree with the authors that DNA tests ‘may have a role in stratifying populations by risk, to enable clinical and behavioural interventions to be targeted at those at increased risk’. However, we stress that for common complex diseases, genomic information alone is not (and should never have been expected to be) enough to motivate behaviour change, or even to stratify populations into subgroups for more targeted interventions.
The BMJ paper addresses an outdated research question and as such does not contribute anything particularly new or useful to debate. Instead of dismissing the role of genomics and personalised medicine on the basis of poor quality evidence supporting a largely irrelevant question, researchers and policy makers should focus instead on generating robust evidence on when and how genomic and other forms of personal information can be best employed to guide clinical and behavioural interventions. This will necessitate acknowledging the potential impact of both genomic and other risk information in empowering individuals to improve their health.
Another useful step would be to demand greater transparency around genomic testing, especially direct-to-consumer tests, in terms of demonstrating validity and clinical utility. Currently, a product may correctly identify a genetic variant, and that genetic variant may have at some time and in some place been linked with the risk of a given disease, but the presence or absence of that same variant may have no value whatsoever for disease risk prediction.
Progress towards effective personalised medicine is dogged by a determination amongst both evangelists and sceptics to present genomic analysis in polarised ‘all or nothing’ terms. As we learn more about the widely varying contribution of genomics to disease, it becomes clearer that a more nuanced approach to its role in prevention is essential. Creating and demolishing ‘straw man’ arguments as in the BMJ article does not encourage a balanced approach.
Rather, health policy researchers and advisers should, like those at the PHG Foundation, be critically evaluating evidence on the value of new personalised medicine approaches to major health problems such as cancer and rare diseases. Where the evidence warrants it, we should be supporting the use of genomics in ‘personalising prevention’ for common chronic diseases – but there is no point trying to make these decisions until proper evidence from biomedical and clinical research is available. All it serves to do is to undermine these efforts – potentially delaying the benefits for patients.
Our formal rapid response to the BMJ article can be accessed here.