This is amazing. BRCA mutations have been linked to increased risk of breast cancer and ovarian cancer. Yet these mutations are not that rare. Hasn't natural selection been doing its job? Or is there some benefit that balances the risk?
The authors of "Effects of BRCA1 and BRCA2 mutations on female fertility," recently published in Proceedings of the Royal Society, hypothesized that women with a BRCA mutation might have more children, even if they don't live as long, on average.
Today, enough couples use birth control that the number of children born depends on preferences for family size, not just innate fertility. So the authors compared women (with and without the mutation) born before 1930. They used the Utah Population Database, which has data on births, deaths, and family relations for large numbers of Utah residents. Most of those women are no longer alive, however, so how can we know whether they had the BRCA mutation or not?
The authors used a variation on ancestral state reconstruction. When two women had the same BRCA mutation -- apparently there are various versions -- the authors assumed that their most-recent common ancestor had that mutation also. They also identified a number of controls -- women who presumably did not have a BRCA mutation, because none of their descendants did -- from the same time period.
So, was there any difference in fertility between women with versus without a BRCA mutation?
Yes. For women born before 1930, those with a BRCA mutation had an average of 6.2 children, while controls had 4.2. There wasn't much difference in survival up to about age 50. By age 75, however, about half of the women with a BRCA mutation had died, while 80% of those without the mutation were still alive.
How does birth control change this picture? More recently, women with a BRCA mutation had an average of 4.1 children, versus 3.4 for controls. Is that difference small enough that we can expect the BRCA mutation to become less common over generations?
Natural selection increases the frequency of whatever genes lead to the most descendants, whatever the cost in human misery. Natural selection will tend to preserve the health of postreproductive women if they contribute to the survival and reproduction of their relatives, particularly their own children and grandchildren. But it must take a lot of grandmothering to make up for having two fewer -- or even 0.7 fewer -- children.
I've written about trade-offs between reproduction and longevity before, including our theoretical work showing that natural selection can favor mechanisms to switch between reproduction and longevity, depending on whether overall population size is increasing or decreasing.