Two companies have recently launched direct access whole genome scans, and more are planning to introduce similar services in early 2008.
Icelandic company deCODE Genetics was the first into the market with deCODEme, launched on 16th November 2007. This was closely followed by the Californian company 23andMe, which together with its influential and somewhat ubiquitous partner Google, announced a similar service just 3 days later. Both companies use a microarray to analyse hundreds of thousands of single nucleotide polymorphisms (SNPs) in DNA taken from a saliva sample, in order to predict susceptibility to dozens of different traits and diseases. The service costs around $1,000, which includes a password-protected internet site for customers to explore their data, and the companies emphasise that they are providers of genetic information, not medical diagnoses. Another two US companies, Navigenics and Knome, are preparing to launch similar but more medically oriented services early next year.
Consumers will learn how their lifetime risk of contracting a number of complex diseases – including Alzheimer’s disease, coronary heart disease, type II diabetes and prostate cancer – varies from the baseline population risk. In theory, this information will empower the individual to take control of their own health by making smart lifestyle choices. However, whilst these developments undoubtedly represent a huge leap forward in personalised genomics, there are serious questions regarding translation of the information directly into personalised medicine or individualised health plans. Estimating the individual risk of a complex disease, which involves the interplay of multiple genetic, epigenetic and environmental factors, is currently near impossible based on genetic data. The network of gene-gene and gene-environment interactions that moderate the effect of individual gene variants are largely unknown, and many of the correlations between specific SNPs and disease are far from being proven at an individual level. Whilst some correlations are weak, derived from small studies with insufficient power to draw statistically significant and reproducible conclusions, many of those that have been validated by large studies or meta-analysis make only a very small contribution to the overall risk of getting a disease.
As a result of these complications, European guidelines were recently unveiled stipulating that genetic tests should only be used under medical supervision (see previous news article). But how many doctors understand how to interpret the results of these tests for an individual? And whilst the companies promise to accept customers anonymously, issues of informed consent and privacy still loom large over this entire field.
Although much good may come from these new services, all but the most savvy consumer should be wary of accepting false truths. Without valid interpretation, personalised genomics could be, at best, a waste of time and money, and at worst, wholly misleading and potentially damaging.