Personalized medicine won’t work, but race-based medicine probably will

When the Human Genome Project got started, it was sold as promising to revolutionize medicine. Once we could easily sequence people’s genes, it was said, we could tailor drugs and treatment to their special needs and correct for minor differences in physiology that influence disease and its cure.

As Greg Cochran has pointed out, this probably isn’t going to work. There are a few genes like BRCA1 (which makes you more likely to get breast and ovarian cancer) that we can detect and might affect treatment, but an awful lot of disease turns out to be just the result of random chance and deleterious mutation. This means that you can’t easily tailor disease treatment to people’s genes, because everybody is fucked up in their own special way. If Johnny is schizophrenic because of 100 random errors in the genes that code for his neurons, and Jack is schizophrenic because of 100 other random errors, there’s very little way to test a drug to work for either of them- they’re the only one in the world, most likely, with that specific pattern of errors. This is, presumably why the incidence of schizophrenia and autism rises in populations when dads get older- more random errors in sperm formation mean more random errors in the baby’s genes, and more things that go wrong down the line.

But while personalized medicine won’t work, race-based medicine most likely will. The differences in physiology for different ancestral populations weren’t random- they evolved because they helped those populations survive in their ancestral environment. People acknowledge this for sickle-cell anemia, which (if you just get one gene for it) helps you resist malaria, but if you get two gives you the disease. My son has two mutations in the genes for the alpha protein for hemoglobin, also called alpha-thalassemia trait- this might make his red blood cells appear smaller than normal, but also would probably give him some resistance to malaria, if he ever needs it. If he had three mutations he’d be very sick, if he had four mutations he’d be dead. Alpha-thalassemia isn’t random- people are likely to have the trait if their ancestors lived in somewhere with a lot of malaria, they don’t if their ancestors didn’t.

The point is, that because these patterns aren’t random, you can actually do something with them; you can probably develop drugs that accommodate these mutations, because you could get enough people to test them with. This is true for genetic racial differences that we don’t recognize as disease, too. People with sub-Saharan African ancestry have a higher inflammatory response and faster glomerular filtration rate in their kidneys. Both characteristics are probably helpful in some circumstances and harmful in others, but they’re regular and predictable, and so doctors can and should respond to them. Black women have a much higher risk of preterm labor, and doctors should know that, too, and be ready to prescribe terbutaline if it looks like their patients are having contractions. Getting more women to carry their babies to term is a much more promising way to reduce the black-white achievement gap than worrying about minor differences in within-district school composition by race.

None of this stuff is magic, and it’s not like knowing it is going to add ten years to any group’s average lifespan. But trying to guilt doctors into pretending any physiological differences are due to racism, as some influential people have been trying to do, will be harmful.

The same early 2000s PBS-Frontline documentary about the Human Genome Project that I used to show my students, where the principal investigators said we were all going to have personalized medicine, also had someone come on to say that you couldn’t tell race from people’s DNA; this is obviously false, as the whole AncestryDNA and 23andme business model shows.

2 thoughts on “Personalized medicine won’t work, but race-based medicine probably will

  1. This is probably going to become less and less useful as races mix, though new patterns may emerge. (I’m curious to see if we get any new illnesses arising from the mixture of East Asian and Ashkenazi genes, for instance.)


Leave a Reply

Fill in your details below or click an icon to log in: Logo

You are commenting using your account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s