Imagine being told by a doctor that you have a high risk of breast cancer and that the only way to greatly reduce that risk is to have a double mastectomy. This is the awful choice some women are faced with, most famously Angelina Jolie.
But what鈥檚 even worse is not to have this choice at all. At present genetic tests can only identify a tiny fraction of the women at high risk, but there鈥檚 now a way to identify far more. Deciding what to do about the results, though, will be even harder.
The first gene variants linked to breast cancer were identified by studying families with a history of the disease. These families typically had mutations in genes called BRCA1 and BRCA2.
Advertisement
These genes code for proteins that help prevent cancer by repairing DNA damage. If either protein is disabled, the risk of breast and ovarian cancers rises greatly.
But not all BRCA mutations disable the proteins. Some are completely harmless. The hard part is telling which is which.
Rare genes
Most gene variants are rare, and if only a few people in the world have a particular mutation, even studying all of those people would not be enough to tell you what it actually does. So far doctors have studied around 7000 mutations in the BRCA1 gene, yet most are classified as 鈥渧ariants of uncertain significance鈥�.
This is why biologists have embraced CRISPR genome editing so enthusiastically. They can now make specific changes to the genomes of cells and animals to find out what effect gene variants have.
A team led by Jay Shendure at the University of Washington, Seattle, has used CRISPR on a grand scale to study 4000 variants in the BRCA1 gene. Of the 4000, 169 have already been shown to be dangerous by large clinical studies and 22 to be harmless. The effects of the rest are unknown.
The team created human cells with any one of the 4000 variants. They let the cells multiply in a dish for 11 days and then counted the number of cells with each mutation. Those with mutations that impaired BRCA1鈥檚 DNA repair activity did not multiply as much as other cells 鈥� in theory revealing which of the 4000 variants raise the risk of cancer.
The gene-edited cell screen correctly identified 162 of the 169 dangerous variants and 20 of the 22 harmless one. 鈥淎ssuming the current clinical classifications [are right], our method shows near-perfect accuracy for determining which mutations will confer elevated risk of early-onset breast cancer,鈥� says Shendure.
What鈥檚 more, similar approaches could be used to assess the effects of variants in the 100-odd other genes linked to cancer, and perhaps other diseases too. This could be a huge advance for genetic testing. But it also creates new dilemmas.
Should we start telling people about their cancer risks based on this kind of indirect evidence, given it is less reliable? Yes, says Shendure. 鈥淚t will of course not be the only piece of information that feeds into a major decision such as surgery, but we believe that it can certainly add value to that decision-making process.鈥�
But the nature of this evidence is hard to understand. For those at high risk, making decisions will become even more complex and difficult than it is now.
On the flip side, women told they do not have a high risk after genetic testing will be able to be much more confident in the result. At present, tests can give a false sense of security because they look only at a few BRCA variants.
Nature
