Angelina Jolie increased breast cancer risk: would a consumer genetics test have picked it up?

Angelina Jolie tested with Myriad Genetics BRACAnalysis, a clinical genetic test ordered and interpreted by a doctor. Her lifetime risk was predicted at 87% for breast and 50% for ovarian cancer. Angelina Jolie underwent a preventive removal procedure for these fortunately non-vital organs.

More and more, individuals can order and freely use their own genomes with services such as @23andMe and the question I am trying to answer is “what is the likelihood that the same outcome would have occurred had Angelina done such a test on her own instead?”.

All together, hereditary breast cancer resulting from BRCA1/2 gene mutations represent 2 to 5 percent of total breast cancer cases (source). Around 1 in 8 women will develop breast cancer in their lifetimes.

Since Angelina was in the highest risk category which is only observed in BRCA1, not BRCA2, let’s narrow the question to BRCA1 mutations.

The BRACAnalysis test determines the full sequence of the BRCA1 gene and could therefore assess the more than 500 single-nucleotide variants that are known to be causal in breast cancer, but I can’t find precise information on the actual variants assessed (the test also analyses more complex gene alterations). The risk of all BRCA1 mutations in aggregate increases 11% and mathematical models that allow to estimate a risk for any combination of BRCA1 mutations can be built.

Out of these more than 500 variants, only two BRCA1 are assessed by the 23andMe test: 185delAG and 5382insC (source). However, these mutations are by far the most frequent and increase risk the most. The 185delAG mutation confers a lifetime risk of 80–90% of breast cancer, and a 40–50% risk of ovarian cancer (source). The 23andMe website reports identical risks for both variants assessed (though the reported risk is lower at 57%).

Therefore, we can induce that either:

• Angelina had several mutations which combined risk was very high, but this is unlikely since the average risk is 11% and she would have to harbour many independently pathogenic mutations (when I get around to analyse the risk by individual variant and the combined risk models I may be able to rule this out altogether), or
• more likely, she had one of the frequent mutations assessed by 23andMe and would have arrived to the same “diagnosis”of “lifetime risk of 87% for breast and 50% for ovary cancers”.

To be fair there are a number of inconsistencies in the data available and I may have been biased in selecting the sources that allowed me to answer the question. Answering this simple question proved more difficult than I had anticipated.

As millions of people start ordering these tests, it would be interesting to know what proportion would benefit from such a diagnosis. But to answer this question I need to know what proportion of the population at risk of hereditary breast cancer harbour the common variants. Unfortunately, this will depend on the specific populations as for example the frequency of the common BRCA1 mutations is much higher in Ashkenazi jews than in the general population, and some dutch women carry a risk mutation that is specific to them.