Encore une fois

WHEN the DNA sequence of the human genome was revealed in 2000, many people expected it to start a revolution. Researchers would be able to discover the genes that caused or influenced diseases. And drug companies would be able to use that knowledge to create better medicines. Until recently, though, it has been a case of “revolution postponed”. The flood of promised discoveries has been more like a trickle.

Much of the reason for the unfulfilled promises was naivity about how straightforward the link between different versions of genes and particular diseases would be. But that naivity has gone, and the fact that complex illnesses often have contributions from large numbers of genes is now recognised. This recognition, plus better computing and sequencing power, mean researchers are indeed beginning to pick the relationships between genes and disease apart.

In January, for example, a group at the Broad Institute in Cambridge, Massachusetts said they had homed in on most of the genes involved in schizophrenia, and thus had a hypothesis for a mechanism that might be causing it. This week it is the turn of breast cancer, as the most comprehensive analysis yet of mutations related to this condition is published in Nature.

The team that did the analysis, led by Serena Nik-Zainal of the Sanger Institute in Cambridge, England, sequenced the genomes of cells from 560 tumours. These, says Dr Nik-Zainal, proved vastly different from the genomes of healthy cells from the patients involved, for they had generally acquired thousands of mutations. That is not surprising. Early mutations in the development of a tumour often involve genes involved in DNA repair. Once this has happened other mutations accumulate. But all these secondary mutations make it hard to sift out the ones which are clinically relevant. To do so you need to compare lots of samples from different people, and thus see which mutations some of them have in common.

That, Dr Nik-Zainal and her colleagues did. They also found mutations that were known to indicate damage from exposure to ultraviolet light (perhaps from too much sunbathing), smoking and other risk factors. This study makes it possible, on the basis of genetic similarities, to cluster breast cancers into different types. In essence, even though their symptoms are similar, these groups are different diseases. It is therefore likely that they will respond differently to particular treatments. The next stage is to see which treatments work best on which clusters.

Genomics is already having an impact on the treatment of cancers such as this. Foundation Medicine, in Cambridge, Massachusetts, offers a test for 300 genes that are often mutated in solid tumours. This lets oncologists try to match treatments with patients. In the future, DNA from patients’ tumours is likely to be sequenced completely for diagnostic purposes.

The arrival of genomics as a science of clinical significance was further underscored by AstraZeneca’s announcement on April 21st that it would form a collaboration to sequence 500,000 samples taken during clinical trials it has conducted over the years. This British-Swedish drug firm says it plans to see 2m genomes so studied over the next decade.

One of its main partners in this endeavour is Craig Venter, a man who led one of the human-genome sequencing projects that resulted in the announcement in 2000. Dr Venter, now the boss of a genomics firm called Human Longevity (HLI), which is based in San Diego, said at the launch of the collaboration, “you can’t tell much from a handful of genomes. We needed large numbers of genomes to find small differences.”

He should know. HLI has compiled a list of the common variants in the human genome. The firm needed 10,000 genomes to pick these out. Larger numbers than this will therefore be required to notice rare and ultra-rare variants that may or may not be associated with diseases. Dr Venter reckons that to discover all of the genetic variation which human beings display, 10m genomes will need to be sequenced.

Variations on a theme

Of course clinical data, too, are needed to work out what those variants all mean.AstraZeneca has these for its trial participants. Genomics England, part of Britain’s Department of Health, has clinical data too. It intends to sequence 100,000 whole genomes from consenting patients. And other pharmaceutical companies are also looking to form collaborations with sequencing groups such as HLI and Genomics England.

It all bodes well for the future of genome-based diagnosis. A report published in 2013 found that nearly a third of drugs in clinical development are associated either with a known DNA variant or with a variation in the structure of a specific protein, ultimately traceable to DNA. The presence in, or absence from, a patient of such a variant allows drugmakers to know whether their products are likely to work in that individual. A new era of genome-based medicine is set to arrive. Again.