2nd quarter 2004

Biological databases facilitate research int frequently occuring disorders

Great was the enthusiasm among geneticists and greater still in the media when the human genetic code was identified in 2001. Optimists expected the genetic causes of common diseases, heart attack, cancer, stroke, depression, etc., to be discovered in the not too distant future. This would also have impacted our ability to predict such diseases, which would in turn have opened the door to large-scale antiselection in life, and health insurance.

It is now apparent that those forecasts were far too optimistic. Common diseases have highly complex origins; they generally result from a combination of a number of weak genetic and non-genetic factors. In 2004, there are virtually no reproducible genetic factors for common diseases whose risk would be greater than that of smoking.

At the same time, many genetic factors have now been identified which are known to increase the risk of disease in the order of 1.5. However, these genetic variants sometimes increase the risk by a factor of only 1.2, and we have no idea how the factors interact. There are both variants that decrease the risk of cancer in smokers and those that increase it.

There are variants whose risks add to the total risk and others that multiply it. There are genetic changes that simultaneously increase the risk of contracting one disease and reduce the risk of contracting another. In addition to which, genetic risk factors interact, of course, not just with each other but also with environmental factors.

What concerns geneticists today is, "How do we research this chaos?" One answer is - biobanks.

Biobanks are currently being set up in various locations throughout the world.

The veteran among them is the Iceland genome project which is pursuing the strategy of tracing genetic variants that cause common diseases. Iceland has a small population that can be traced back to a handful of Viking immigrants. Furthermore, Iceland traditionally has exceptionally well-documented family trees and a sophisticated health system.

Just under half the total population has now been genetically examined. Gene mutations found include those responsible for heart attack, stroke, osteoporosis and schizophrenia. The Swiss pharmaceuticals concern, Hoffmann la Roche, has acquired the rights to develop new medicaments using this knowledge.

Estonia has long had plans to test even more probands than Iceland but these have not got off the ground owing to financial problems.

Since 2002, the United Kingdom has been building up the world's largest biomedical database. The UK Biobank plans to collect medical data, life-style details and biological material (blood samples for DNA, i.e. genetic material; where applicable, surgically removed tissue for RNA, genetic activity) from half a million participants in the 45-69 age bracket, who will be monitored over a continuous ten-year period.

Unlike the Iceland genome project, the UK Biobank is aimed not so much at discovering new gene variants as at casting light on the links between the different risk factors. The UK Biobank has clear potential to classify risk factors in isolation and in combination more precisely. This would go hand in hand with the potential to predict diseases more accurately.

Since the Iceland project information is not public, but only available to companies (directly to the Icelandic firm Decode Genetics, indirectly to its clients, e.g. Hoffmann la Roche), the Human Genome Organisation (HUGO) has called for public access to genetic databases of this type. In the meantime, efforts have been made to combine individual projects/databases. The Public Population Project in Genomics (P3G) aims to combine information from three separate biobanks:

• —
  Estonian genome project;
• —
  Canadian CARTaGENE (universities in Quebec plan to investigate the genetic variability of 1% of the province's population aged between 25 and 74 (60,000 subjects);
• —
  European EuTwin (Scandinavian universities will carry out genetic investigations into some 800,000 sets of twins with regard to physical size, corpulence, migraine, coronary heart disease, stroke and longevity).

The P3G database would thus be based on a population of 2.3 million. The data would be available free of charge to scientists meeting specific scientific and ethical criteria.