NPR Story
10:44 am
Fri June 7, 2013

Whole Genome Scans Could Reveal Too Much

Originally published on Fri June 7, 2013 12:33 pm

Transcript

IRA FLATOW, HOST:

This is SCIENCE FRIDAY; I'm Ira Flatow. If you're thinking about getting married or having children or just contemplating your health care options, you or your doctor may decide to have your DNA analyzed, looking for genes that may indicate possible trouble ahead. Maybe there's a telltale mutation hiding there or a recognizable pattern of genes.

It is possible now to order up a scan of your genome and see where trouble may lie or may not. Problem is, you might wind up finding more than you bargained for. Your doctor maybe looking, for example, for clues about colon cancer and find an increased risk for breast cancer instead, or Alzheimer's or heart disease.

And at that point is your doctor obligated to tell you about those risks? What if you didn't ask for it, you don't want to know? Is the doc obligated to tell you anyhow? And do you know enough about genome scanning to even know what you want to know? It's not just a patient who might worry about the results. If it's a faulty gene, there's a chance the patient's family members might share the same dangerous mutation. Do you tell them? What is the ethics of that?

The conundrum gets even more complicated. You don't even have to be born yet to get a genetic scan. We're already doing it for fetuses and human embryos produced by in-vitro fertilization. What are parents to make of these test results? Will the results influence their desire to have the baby? Could that knowledge be abused? And where do you go to get genetic counseling anyhow? Are there such genetic specialists?

The ethics of genetic testing and some advice on how to handle it, that's what we're going to be talking about for the rest of the hour. Give us a call. Our number is 1-800-989-8255, 1-800-989-TALK. You can also tweet us, @scifri, S-C-I-F-R-I, or go to our website at sciencefriday.com and join the discussion going on there.

Let me introduce my guests. Hank Greely is law professor at Stanford University in California. He's also director of the Center for Law and Biosciences there. Welcome back to SCIENCE FRIDAY, Dr. Greely.

HANK GREELY: Thanks, Ira, nice to be on the show.

FLATOW: You're welcome. Susan Wolf is a professor of law and medicine at the University of Minnesota in Minneapolis. Welcome to the show.

SUSAN WOLF: Thanks, Ira.

FLATOW: Kelly Ormond is a certified genetic counselor, director of the Genetic Counseling Training Program at Stanford. Welcome back to SCIENCE FRIDAY.

KELLY ORMOND: Thanks so much for having me.

FLATOW: Let me begin with you. Let's talk first here. You were an author on a recent paper from the American College of Medical Genetics, which gives guidelines on how to deal with incidental things doctors might find, as I mentioned before, in a patient's genome scan. For example, you scan for colon cancer, you find maybe Alzheimer's risk genes. What do you do about it? Tell us about what happens when you find all these different things?

ORMOND: Yeah, you know, I think the new genetic testing that we're doing through sequencing has been a game-changer in the sense that traditionally if you came for a genetic test, I would be doing a very specific test either for a single gene or a handful genes that were all around the same condition.

So you brought up the example of colon cancer. If you had a family history or you yourself had colon cancer, we would just look at those things, and we would try to find a genetic explanation for the condition that you or your family had.

And now that we're using this sequencing as a tool to make diagnoses, it's possible that we could actually get a glimpse at all of these other things that you have alluded to. And so like you said, that's not the point that you're originally doing the test for, for most people, and it raises the question of what should we do with all of it.

And the ACMG, among many other professional organizations, said gosh, we really need to think about this. And they convened a group of people who work in all different clinical areas and laboratory areas and ethics to kind of think about, first, is there anything that we think sort of supersedes patient preference in the sense that we really think these are conditions where we're sure enough about the genetic prediction, we think they're serious conditions, it's identifiable in you, and if you know about it in advance, you can do something to change the course, you can become healthier either because you're getting screened or you're doing something differently.

And so the committee decided that for a very small handful of conditions there were things that met that criteria and we felt should actively be disclosed. And I do want to be very clear that the committee still thinks that there needs to be a consent process about this, that people need to know that that's going to happen and that they could be getting these pieces of information.

FLATOW: So you should be telling patients - I understand there are 57 mutations involved here, that the lab should disclose the patients, and there's something that you can do about those things, but not genes - not problems that you can't do anything about, if I hear what you're saying.

ORMOND: Right, we were very clear about that. So as an example, there are 15 hereditary cancer conditions and 30 cardiovascular conditions and then two that have to do with taking anesthesia and having a very bad response. So they're all things that there are surveillance for and there's treatment for and that if you have a change in one of these genes, there is what we would consider to be a significantly high chance that you'd actually develop cancer.

We very clearly did not put conditions like Alzheimer's disease or psychiatric illness or Huntington's disease or any of those sorts of conditions on because it's much clearer that people are going to have really personal responses to whether or not they'd want that information.

FLATOW: Absolutely. Dr. Wolf, you published a report disagreeing with these guidelines.

WOLF: Yes, we did. You know, for about 50 years, Ira, we have all been working to make sure that medical care, testing, the information given back to patients, is governed by patients' preferences. People don't give up their rights when they go to see the doctor. They get to decide what tests are done. They get to decide what information is given back to them.

And in genetics that's been a particularly strong message. We don't inflict on people bad news, even news they could do something about, because we think it's a good idea. They get to decide. And people have different preferences about this. People who are undergoing genome sequencing because they have a cancer that hasn't responded to any other treatment and now we're last ditch looking for something to do for them, some clue in the genome that would tell us what to try next, may really not want to hear about this extra bad news from other quarters of the genome.

So the problem with this report is that it says that people really don't have those rights, that those rights don't apply for gene sequencing, that yes, they should be told when they show up, OK, we want to do this sequencing because of your cancer, for example, or for your kid because your kid has some severe syndrome and we can't figure it out, we need to the sequencing.

But here's the deal: If you're going to have the sequencing for that, we're going to have to look for the 57 extra. And if you don't like that, you can walk. That's the deal: The only terms on which you can get the sequencing is if we do this 57 extra. And the committee, to its credit, was very blunt and said that's it for the right not to know.

The stuff goes back, it goes back to the patient, even if they don't want it, and even if they're a kid, where for a long, long time we have said no, the only genetic information that should go back on kids is information that's really medically necessarily to go back while they're still a kid.

If it's adult onset, wait. Let them decide themselves, using their own values when they reach 18 or older. So in a couple key ways, this report really abandons the prior ethical consensus in a way that needs to be carefully, carefully considered.

FLATOW: Dr. Greely, where do you weigh in on this?

GREELY: I feel strongly both ways.

(LAUGHTER)

GREELY: As a practical matter, I don't think this disagreement is going to turn out to be very important, because as a practical matter I don't think very many people, when you say, Ira, we're going to do a whole genome scan on you, we may find one of these Heinz 57 varieties of things that we aren't looking for, that is very powerful genetically, very high risk it's going to happen, it's a very serious disorder and for which we have a good intervention - I don't think very many people are going to say don't tell me about that.

Some may, but I don't think very many are. I think that's going to be a very, very small minority. So as a practical matter, I don't think that side of it's very important. As a matter of principle, I generally agree with Susan on this. I think as a matter of principle people generally should have a right to control their medical care, although I would note they don't have a perfect right to control their medical care.

I don't have the right to go into my doctor and say, hey, Doc, I'd really like some Oxycontin for the weekend because I want to have a really good time. You know, there are limits to what patients can request, but I do think what kinds of diagnostic information they should get back should be among the things patients should request.

So as a matter of principle, I think I am somewhat on Susan's side. It's a close question; there are good arguments both ways. As a practical matter, I don't think it's going to be very important.

FLATOW: Let me bring in - let me bring in another participant, someone with personal experience here, who has actually been participating in this whole process, and she is Pam Widick.

She's a registered nurse. When she was diagnosed with cancer, she became a patient at the Mayo Clinic Center for Individualized Medicine, and had nearly all her genes sequenced as a result. Welcome to SCIENCE FRIDAY, Ms. Widick.

PAM WIDICK: Thank you, Ira. Good afternoon.

FLATOW: Tell us about how you ended up getting your genome scan.

WIDICK: Well, after two years of just other treatments that weren't working for my cancer, which I have cholangiocarcinoma, which is - where it started from going from the liver into the bile duct, and then had also spread to both of my lungs by the time they found it. I went through three other kinds of chemo and was unsuccessful with all three of those. So we decided that this might be the best bet, because they really didn't have any other ideas of what else to do at the time.

FLATOW: So they would scan your genome, looking for a way to combat the cancer, maybe find a little Achilles' heel, and go in that way.

WIDICK: Yes, yes.

FLATOW: How did they prepare you for that? Did...

WIDICK: I met with a genetic counselor, and she tried to explain it as best she could with her doing what she does, and me being all new to this with dealing with the cancer and...

FLATOW: Did they warn you they might find other things during the scan?

WIDICK: Yes. Yes.

FLATOW: They did.

WIDICK: Yes, they did. They were very in depth of what the possibilities were, and wanted me to make sure that I spoke to my family. And, of course, I'm the oldest of six children, and they were all, like, yes, you know, we don't care what you find. Hopefully, we can find something that will help with your cancer.

FLATOW: So you were open to that idea, finding other things?

WIDICK: Yes.

FLATOW: And did they find anything else that worried you?

WIDICK: No, they did not.

FLATOW: So that's...

WIDICK: Nope.

FLATOW: ...the good news.

WIDICK: Yes.

FLATOW: Do you think that you would have any recommendations for people who are going through this process? What could they learn from your experience?

WIDICK: I think for anybody who - you know, I just think it's great that they can do this and possibly find what I had, a certain gene mutation that the chemotherapy that I'm on right now, it showed that - it has been shown to target this particular gene mutation. So I think if they can find that kind of stuff before it gets advanced, that, I mean, that I think the death rate would go down in our cancer patients because it would be treated, hopefully, earlier, caught earlier and be able to do some treatment before it gets bad.

FLATOW: Did it help you?

WIDICK: I go for scans next week. So I'm not sure yet.

FLATOW: All right. Well, good luck to you, Pam. Thank you for taking time to be with us today.

WIDICK: Thank you, Ira. Have a good day.

FLATOW: Pam Widick, who was a registered nurse and went to the Mayo Clinic Center for Individualized Medicine. This is SCIENCE FRIDAY, from NPR. I'm Ira Flatow. We're talking about individualized medicine, the human genome, what you should know, what you should not know - Kelly Ormond, Susan Wolf, Hank Greely. Hank, what do you think of people actually getting whole genome sequencing today? Should people go in and - just as something to do and find out on their own?

GREELY: I wouldn't do it. I had an opportunity to do it and decided not to, but it's a close call for me. I think right now, there are real questions about the accuracy of the process, really, at two different levels. One is the accuracy of the actual sequencing, how accurately it tells you what the A's, C's, G's and T's of your genome sequence are. And, you know, even it only makes one mistake every million times, you've got 6.8 billion base pairs. So that's 6,800 mistakes, anyone of which could lead you down a real bad blind alley.

So we need to be confident about the accuracy, and there are some areas where the machines are pretty accurate. There are other things we know that they're not very good at. There are diseases that are expanding repeat diseases, like Huntington's disease. It's a CAG-CAG-CAG. If you've got 36 or fewer, you're safe. If you've got 37 or more, you're not safe. The sequencing machines are really bad at counting repeats. So they're not, I think, easily accurate enough. You can follow up positive results to double-check them.

It's harder to follow up the negative results. But I think the bigger problem's going to be figuring out what the sequence means. Even if I had a perfectly accurate sequence for you, there would be a lot of it where it was different from the average human, but I can't tell whether it's different in an important way. And I worry a lot that we'll end up in a situation where different companies will do, through software, the analysis of the sequence, and they'll give you results, but you won't know whether the results are right.

It's kind of like the SNP chip, direct-to-consumer companies - 23andMe and others - that are doing a limited genomic analysis, telling people their results. A few years ago, the GAO did a study. They sent samples from the same people to three of the companies. The companies got all of the SNPs almost perfectly right. They called the actual DNA variations correctly. But a third of the time, one company said you're at high risk for this disease, and the other company said you're at low risk for this disease.

They were reading the same literature, but making different calls. Right now, it's really hard to make good calls for many, many things you find in the genome. Some of them are simple. Some of them are hard. So, for me, the accuracy issues are a real - a real discouragement from trying to do this just for the heck of it. There may be some circumstances like people facing serious cancer problems, or like children with syndromes that are of unknown origin, but look like they're genetic, where it makes sense today. I think we need to get better at understanding it before it makes sense more broadly.

FLATOW: Kelly Ormond, as a genetic counselor, would you agree with that?

ORMOND: I absolutely agree, and I think that Hank nailed it. And I'll give you a couple of examples to illustrate that even further. I've been involved in helping to look at some of the variants that come out of genomes from what we would sort of consider to be healthy individuals doing this for lots of different reasons, including curiosity. And out of those millions of base pairs we look at, there's, on average, somewhere between about 100 and 200 that we decide to look more closely at and then figure out - try to figure out: Do they mean something or not?

And as he said, we do come up with different answers. When we've got four or five different people looking at the same information in the medical literature, only about two-thirds of the time are we coming up with the same answer. And laboratories - when we've sent clinical samples to more than one laboratory - they will get the sequence right. They'll list the variants, but they will have different decisions about if they're going to decide this is important enough for me to tell you or not. And I think there is a perception that everyone is getting the same information, and it's really not true.

And then the second thing that I think is a challenge is that even if we decide this is a change that's important for you to know about, we don't actually know how likely you are to get a disease in a prospective state, because most of the studies we have that we're basing that on are coming out of very biased families that were identified because they had a lot of disease in the family. So...

FLATOW: OK.

ORMOND: ...probably we're going to have overestimates.

FLATOW: Yeah. All right. We're going to come back and take your calls. 1-800-989-8255. You can tweet us @scifri. Talking about the genome at this hour on SCIENCE FRIDAY. So stay with us.

(SOUNDBITE OF MUSIC)

FLATOW: I'm Ira Flatow. This is SCIENCE FRIDAY, from NPR.

(SOUNDBITE OF MUSIC)

FLATOW: This is SCIENCE FRIDAY. I'm Ira Flatow. We're talking this hour about getting your genomes scanned, and what happens if your doctor finds some unexpected mutations that might increase your risk of disease. Do you want to know? Should you be told? What should you be told? Hank Greely, law professor at Stanford University, also director of the Center for Law and Biosciences there. Susan Wolf, professor of law and medicine at the University of Minnesota in Minneapolis. Kelly Ormond, certified genetic counselor, director of genetic counseling training program at Stanford. 1-800-989-8255.

One question, let me talk about the doctor, Susan, because it seems like the doctor is now in the middle here, caught in a conundrum of do I - can I get a report back? I've got all these things that might be wrong. Do no harm. Do I have my patient's, you know, best interest, right? Wow. What a decision the doctor has to make.

WOLF: I think you're right, Ira, and it's made even more difficult by the fact that we know a lot of doctors don't understand much genetics, much less the whole genome. So that's a problem, out of the gate. There are not enough genetic counselors. Kelly's in rarefied company there, and we need a lot more genetic counselors and medical geneticists. But I think you're right, that physicians may feel - on the one hand, they want to do everything they can for their patient to prevent harm, to warn of potential medical problems and try to head that stuff off at the pass.

On the other hand, physicians also have these strong, clear, ethical, legal obligations to really talk to their patient and make sure that the patient's testing and medical care is governed by the patient's own values and the patient's preferences. You know, this is something that's come up forever in end-of-life care, where there's all sorts of things the doctor can do to extend life, but the patient says: You know what? That's not me. That's not what I want.

We're right here at the dawn of genomic medicine. This whole problem's going to get bigger, not smaller. That's why I think it's so crucial to get the ground rules straight, to say this is like other parts of medicine, where it's really the patient's body and life and values that should govern what happens.

FLATOW: So shouldn't the patient, then, be given all the information that you can to make a decision, even stuff that she - he or she did not want to know?

WOLF: Well, the problem is that information isn't benign. You know, it sounds like, oh, sure, that would be the best, if you've got all the information. But what we've discovered is that the information - particularly when it's predictive - when it says, well, we think you have a 75 percent chance of getting such and such horrible cancer or something we can't even treat, like Alzheimer's disease, some people will want that information, but some people will feel overwhelmed and terrible and don't want that information. That's part of the job of genetic counseling, is talking deeply with these people so they can make good decisions about what burdens to undertake.

FLATOW: Let me ask a question to all three of you, then: Does age matter? You've talked about end of life, but what if we're at a stage at the beginning of life, or a young child or a fetus even, where we can tell. Does that change the plan about what you tell people?

WOLF: Well, we've long said it should drastically change the plan. The new ACMG guidelines challenged that. But for at least 20 years, we've had in place a consensus that if we're talking about genetic information that's not medically needed in childhood - for example, a BRCA variant that predicts increased risk of breast and ovarian cancer later on in the person's life - that they should be able to make this tricky, values-driven decision for themselves. They may not have the same values as their mom and dad, and they should - you know, we should wait, and then offer them that information, that testing when they've come of age and can make the decision themselves.

FLATOW: Hank?

GREELY: I think the child issue is really a tough one, here. In the old days, when you were testing a kid, it was almost always because there was a family history that would lead you to be worried that the child might have that. What makes this a little bit different, I think, Susan, is if it's one of the 57 varieties that the ACMG has listed, it's something that you didn't necessarily expect.

It's probably something that the child got from one of his or her parents, and this could be important medical information with respect to the parents. So I think that makes the situation a little bit harder than it was in the old days.

WOLF: And that is something that was commented on, Hank, in the recommendations. And I do want to point out also, I very specifically looked at the different conditions yesterday, and virtually all of them do have the potential for childhood onset; either as enough of a percentage, which was what led us to lead, you know, to include these on the list, or that it is at least feasible.

FLATOW: When people talking so much about Alzheimer's disease, why is that not on the list?

WOLF: It's purposely not on the list.

FLATOW: I understand that, but don't you think that people might want to know if they're going to get Alzheimer's disease?

ORMOND: Well, some people do and some people have various strong feelings in the opposite direction. So that's the first piece, is that there's so much disparity. The second piece is there's not something that we could at least medically do to change one's outcome if they knew they were at risk. And, you know, over the years, that may change. We may have things that one can do early on or before the condition even onsets that could change that piece. But then the third piece is that the predictability of many of the dementia genes is just not as good as the genes that are on this list.

GREELY: So...

WOLF: Ira, if I can?

FLATOW: Yes.

WOLF: I think you raised a really important and deep question. You know, in the business, we've all been talking about what are the actionable variants, the things that the doctor could do something about. But the way we've been thinking about actionable is very doc-centric. I think you are totally right, that from the standpoint of people, real people and patients, some of these things that the doc can't fix, they're going to be people who very much want that information and regarded as actionable in their lives. They're going to do different estate planning. They're going to say, you know, I'm going to seize the moment now and travel with my partner or do something with my kids.

You know, NIH has been funding a ton of research on all of these questions, to their enormous credit. They've even created, now, something called the Return of Results Consortium to link investigators all around the country so that we can make progress and gather data on these questions. One of my deepest concerns about the ACMG guidelines is they're premature. We need data.

GREELY: I think it's important, though, to remember exactly what the ACMG guidelines are about. They're about what labs have to tell doctors, and what doctors should tell their patients. Now, I disagree a bit what the doctor should tell patients, but it's a very limited sort of question. It doesn't say anything about whether a patient can say to the doctor, I'm really interested in finding out my Alzheimer's risk, and what the doctor should say to that. And I think that would be perfectly appropriate if the patient says, you know, I'm really interested in Alzheimer's.

There's nothing in the ACMG guidelines that says you don't return that. Or even if the doctor doesn't say, you know, among the other things we might get from this, we might get some information about your Alzheimer's risk. Is that something you'd be interested in finding out?

FLATOW: Mm-hmm. Mm-hmm.

ORMOND: Yeah. To me, this is about the conditions where there's medical equipoise and where there isn't. And if there's a real clinical reason to give the information, those are the kinds of things where committees can look at it and make decisions about if that's something that should be disclosed or not, and for the rest that absolutely comes back to what you're talking about, Susan, which is patient preference. And as a genetic counselor, autonomy is near and dear to my heart, so I completely support that.

FLATOW: Hank, one topic you're writing about now is sort of the brave new world of genetic testing, something called pre-implantation genetic diagnosis or PGD. Can you tell us about that?

GREELY: Sure, although, PGD isn't so much new. The first baby born after PGD was born in 1990, so he would be about 23-years-old now, but it hasn't been used very much. PGD takes a three-day-old embryo, an embryo three days after fertilization when it's about eight cells, plucks one cell out of it, does a genetic test on that cell, and uses the genetic test on that one cell to provide information about the rest of the embryo.

FLATOW: Mm-hmm.

GREELY: Unfortunately, you need to do IVF in order to do PGD. You need to do in- vitro fertilization because if you get pregnant the old fashion way, day three after fertilization, that embryo is halfway down one of two fallopian tubes and good luck trying to find it. If you do fertilization through IVF, that embryo is in the Petri dish that you put it in. So there hasn't been very much use made of PGD, still about 4,000 babies were born last year in the United States after PGD.

What I think is going to transform this - probably not for another 20 or 30 years - is we'll be able to make IVF much, much simpler. IVF is a pain, it's expensive, it's unpleasant, and it's dangerous. At least it's unpleasant and dangerous for the woman. It's one of those areas where life is unfair. About one percent of women who go through the egg harvest procedure end up hospitalized as a result every year, which is - it's not a terrible risk, but it's not an entirely benign procedure, and it's not fun. It's not easy.

I think what we're going to be able to see - and we're beginning to see research with mice and beginning to research with humans on this - is the stem-cell technology transforming this. We'll be able to take skin cells from you, turn them into what are called induced pluripotent stem cells, cells that can make every other cell type. We can already do that. The next step, we'll be able to - we'll be turning those cells into, say, sperm. It will take a long time before we can do that in a way that is safe and efficient. But at the point we can do that - and I think the eggs will be more important than the sperm - it makes IVF a lot simpler.

If we could take skin cells from Kelly, turn them into IPSCs, these induced pluripotent stem cells, turn those into eggs, Kelly could then make a 100 embryos - along with sperm probably provided the old fashion way - and do PGD on each of those 100 embryos. And here the advances in sequencing come in, instead of just learning about one trait or a handful of traits, Huntington's disease, male or female, HLA match for an existing sick sibling, you'll be able to do the entire genome of the three-day embryo from one cell. And I think our children are going to be confronted with checklists. They'll make 100 embryos and they'll be given a list of the traits of the 100 embryos and be asked, OK, which ones do you want for your babies?

FLATOW: 1-800-989 8255 is our number. This is SCIENCE FRIDAY from NPR. Susan, what do you think of that choice?

WOLF: I think it's going to be a very difficult capability to manage. You know, really, because we've been doing PGD, we are already facing this question of what are the things we shouldn't do PGD for? I mean, should do PGD to tell whether the embryo has an ApoE-4 variant, that means they are more likely to get Alzheimer's, you know, in their 80s. Is that really a legitimate ground on which to say no, we're not going to transfer that embryo? Let's find a different embryo.

Some of the same questions coming up with another technology, which is coming into use, non-invasive fetal testing. That's further along. You don't need IVF to do it. You got a pregnancy in process. But pretty early in the pregnancy, we're able to find fetal DNA. So that's from the fetus - genetic material from the fetus that's floating around in the mom's blood, in her plasma, get it out and perform genetic testing on it. Potentially, all the way up to genome sequencing. So these all ask variations on the same question that Hank put forward, which is, is it legitimate for the adults procreating to have that much decisional control over the next generation? And I don't think we have, now, the wisdom to that wisely.

GREELY: Just a second, Susan, on the non-invasive prenatal testing. There are already four companies in the United States offering this service today. Several hundred thousand American women will have their pregnancies tested with this now. Right now, the tests are limited only to chromosomal problems, Down syndrome, trisomy 13, trisomy 18, some sex chromosome issues. But that's going to expand.

I think the difference between the prenatal testing and the pre-implantation testing, though, probably will be that parents would be more tempted to choose between embryos for traits than they would to abort an actual pregnancy. Aborting a - I don't think very many parents would abort a pregnancy because the hair color was wrong, but parents, given a choice of 100 embryos, might choose between number 28 and number 47, based on hair color.

FLATOW: Mm-hmm. Let me - you have a minute or two left and an important question that people, I'm sure, are asking. What about the companies - the insurance companies - how are they going to react to all of this? Does that knowledge get out to them? Can we be denied coverage, or how is that going affect them?

GREELY: Well, the good news right now is that under the Genetic Information Nondiscrimination Act, at least with respect to health insurance, insurance companies are not allowed to use this information against you. Now, it doesn't apply to life insurance, disability insurance, long-term care insurance, and people are talking about whether GINA should be expanded to that. Each state has its own laws on genetic discrimination. Some of them are broader, some of them are narrower.

But the health insurance side, you know, no law is perfect. Murder has been illegal for a long time, but it still happens. But the health insurance side, we can feel fairly comfortable about. The other insurances are still up for grabs.

WOLF: And Ira, I think that means it's very important for people who are faced with this choice about whether to undergo sequencing, have a deep conversation with their care providers about how much information are you generating about me. Where is that information going to be recorded? Because as Hank says, we only have partial protection now against the misuse of that information.

FLATOW: All right. We're going to stop there because - this is all great conversation, and I'm sure it's a topic we're going to revisit many times. I want to thank all of you for taking time to be with us today. Susan Wolf, professor of law and medicine, University of Minnesota in Minneapolis, Kelly Ormond, certified genetic counselor, director of the genetic counseling training program at Stanford, Hank Greely, law professor at Stanford also, director of the Center for Law and the Biosciences there. Thank you all.

WOLF: Thank you.

ORMOND: Thank you.

GREELY: Thank you.

FLATOW: You're welcome. Transcript provided by NPR, Copyright NPR.