[Lecture] Genetic testing

[magid]

Last night I went to the first of four Longwood Seminars, talks about medical topics at Harvard Medical School that are open to the public. I hadn't been anywhere on that campus before; Avenue Louis Pasteur is an interesting mix of classical (including Boston Latin) and modern (the building with the circular lecture hall we were in). I don't know if the street lights are specially designed for something in particular (avoiding light pollution? energy efficiency?), but it wasn't lighting the street, which was noticeably dark when there weren't building lights to supplement.

Anyway. Three people spoke. Dr. Raju Kucherlapati gave an introduction to genetics, discussing how the Human Genome Project has opened the way for genetic testing.

Dr. Mira Irons (a pediatrician as well as a geneticist) went into more detail about DNA testing, how there are hundreds of gene abnormalities (but only some of which are disease causing, and not all disease-causing mutations are detectable), about what is done exactly (a specific test on isolated white blood cells), reasons for testing (for medical diagnosis, but also for decision making, such as testing for carrier status or prenatal testing), and some discussion of choice and risk. For instance, if there is a medical condition that is only adult onset (without treatment/lifestyle changes that could affect it before then), don't test anyone under 18, so there aren't issues of telling/not telling a minor. There's not one general test for abnormalities, so any sample sent out for testing is going to be for something specific, and if there's a family member who already has X, they're going to be tested first, for easier comparison. It surprised me to hear how quickly things are changing, with new tests becoming available all the time. There can be a lot of emotion clouding people's evaluation of risk (, especially with prenatal testing). (Plus I suspect USians aren't well versed in understanding probabilities of risk.)

Dr. Michael Murray spoke last, talking about the decision to be tested, with the possibility for genetic fatalism ("I have the gene for X, so it doesn't matter what I do."/"I don't have the gene for X, so it doesn't matter what I do.") (I hadn't realized that only a fraction (far less than half) of people who smoke get lung cancer; the current theory is that smoking + genetic predisposition = lung cancer.). The field is so new that there's a concern about DNA testing becoming a commodity (he showed some commercial sites).

During the Q&A session (questions submitted on slips of paper to the moderator) other issues were raised. While health insurance is not supposed to allow genetic testing as the basis for pre-existing conditions, life and disability insurance are not similarly constrained, so this might become a problem. It's unclear what will happen vis a vis genetic testing and employment (hiring/firing), though President Clinton said federal employees should not be discriminated against on the basis of genetic testing, and IBM has made a similar statement. Pharmacogenetics were touched on briefly, not enough for me to learn something new. And of course there were anecdotes.

A number of informational sites:
GeneTests has information about gene testing and labs that are available (designed more for medical people than laypeople).
Educational materials are available through DNA Interactive.
Guide to gnetic disorders at Your Genes, Your Health.
Make a family medical history that you can share with family members and/or doctors.

Next topic (March 15): Maintaining Mental Agility: From brain aerobics to repair

After the talk, I estimated that taking the T would take me about an hour to get home, with wait times and walking to and from. So I decided to walk instead, and that also took me an hour. I win :-).

Comments

[queue.livejournal.com]

I forget if I told you or not, but my sister got tested a while ago for cystic fibrosis. Not surprisingly, she is a carrier (statistics question: would you say that there was a 50% chance she was a carrier, or a 2/3 chance because you can rule out the case that she actually has the disease?). Her husband then got tested, and he is not a carrier. They had wanted to do different things if they had both been carriers, so I guess it's a good thing that's not the case.

[fetteredwolf.livejournal.com]

Oh, so it's the conference hall in the NRB, right? I'm wondering if I should register for next week. What will they do to people who register but don't turn up? I have a flow cytometry appointment till 5, and not only does it always end up taking longer, but then I have to prepare the cells we get for DNA extraction. But then again, it's right next door.
Hmm.

[magid.livejournal.com]

I don't remember if you told me that before.

Statistics question: since both parents have to be carriers, there are four possible outcomes, one with, one without, two carriers. But given that she doesn't have it, I'd assume 2/3 probability of carrierhood (-ship?).

I'm glad her husband's not a carrier; it makes things a lot simpler. (Does it mean that their kids should be tested for carrier status when they're ready to reproduce?)

[fetteredwolf.livejournal.com]

If either one or both of your parents have the CF mutation, it's 50%. It would be 100% if one of your parents was homozygous, i.e. had CF.
CF/+ x CF/+ or CF/+ x +/+ = the probability that she has gotten one copy is 50%.
If CF/+ x CF/+, the probability of getting two copies is 50%x 50%, or 25%.
If neither of your parents have been tested, then the probability of her having the mutation is about 1:25, if she's an Ashkenazi Jew.

Edited to correct frequency of mutation in Caucasian/Ashkenazi population

[magid.livejournal.com]

NRB? Not sure. 77 Avenue Louis Pasteur, the conference room just to the left when you go in the leftmost entrance.

You can register for them all; there's no penalty for not showing up. In fact, enough people who hadn't registered showed up that they ran out of packets to hand out (those who didn't get can have them mailed), and the place was pretty full. Also, some people came in a bit late; it wasn't a big deal.

[magid.livejournal.com]

(a) Queue's brother had CF.
(b) Neither Queue nor his sister is Jewish (if they were, I assume they'd be Ashkenazi, though).

[hrafn.livejournal.com]

Ah - it's good to know they aren't both carriers.

[fetteredwolf.livejournal.com]

But given that she doesn't have it, I'd assume 2/3 probability of carrierhood (-ship?).

But you still have to take into consideration that there is a 50% chance of inheriting the mutated allele, from either parent.

Queue,

[fetteredwolf.livejournal.com]

I'm sorry if it sounds like I'm treating you like a genetics class problem. I don't mean to be insensitive.

[magid.livejournal.com]

OK, this is my reasoning:

Given that his brother had CF, both his parents were carriers (neither of them had CF). Therefore, there's a 25% chance of Queue's parents' children having CF, a 50% chance that the kid is a carrier, and a 25% chance of neither having nor carrying. In this case, we know that she doesn't have CF, so there are two options left, one with twice the probability of the other, so a 2/3 chance of being a carrier.

What am I missing?

Re: Queue,

[queue.livejournal.com]

Not at all. I'm all about the math/science problems. Now, if you were treating me like a philosophy problem, I might get offended ;-)

I was going to say more, but magid covered my point about the statistics question. I didn't realize that the rate of the mutation was so high in Ashkenazim.

An interesting tidbit that I learned from talking to my sister: there are apparently a number of mutations that result in cystic fibrosis. Now that we know (well, I don't happen to know it, but it's find-out-able) what mutation she has, that will make my testing simpler (and maybe less expensive?) if I ever feel the need to get tested. And I think she may have had my mom get tested, so that my sister's test would be easier, but I could be misremembering the details.

And while I'm on the subject, wouldn't it have made more sense to test her husband (who has no family history of cystic fibrosis) first? If you're going to have to do at least one test anyway, doesn't testing him first reduce the likelihood that a second test will have to be performed?

[queue.livejournal.com]

I think this means I get to be an uncle within the next couple of years. All the fun of having my own kids without any of the responsibility!

[ichur72.livejournal.com]

Have you heard of Dor Yesharim, which does genetic tests focusing on Tay-Sachs and a few other conditions? If not, more info here:

http://www.yoatzot.org/article.php?id=123

[fetteredwolf.livejournal.com]

OK, you're right. I was thinking about it in terms of single alleles.

Re: Queue,

[fetteredwolf.livejournal.com]

1. Test husband first, and he's negative. Your sister will still want to be tested since she's aware of the aforementioned 2/3 probability of being a carrier, so that her children will know if they have the possibility of being carriers. => 2 tests performed.
2. Test husband first, he's positive. Obviously, your sister will want to be tested. => 2 tests performed
3. Test wife first, she's positive. Same as 2, => 2 test performed.
4. Test wife first, she's negative- no need to test husband, since he has no history of CF in family and probability that he has it is same as probability in rest of population. Kids will eventually make their own decisions regarding testing.

The rate of mutation in Ashekanzim is only slightly higher than it is in non-Jewish Caucasians.

[magid.livejournal.com]

Oh, interesting. No, I hadn't heard of them.

At least six years ago now, I had a doctor who kept wanting me to be tested for Tay-Sachs and one of the others; it felt odd, because I was not even dating someone at the time. I suppose it would've been useful, though.

Re: Queue,

[magid.livejournal.com]

And I think she may have had my mom get tested, so that my sister's test would be easier, but I could be misremembering the details.

According to what I heard yesterday, that sounds right: test the one who's known to be affected, to pinpoint the exact mutation because there are so many possibilities, then go to the unknown person's test.