eMERGE & CSER Session 1 Panel Discussion

[0:00] We could get all the speakers back up. And if you have questions for the speakers or comments, We find a microphone. Thank you. So just to answer Les' question now that I thought about it more, I'd be happy with much lower sensitivity and much higher specificity. for incidental findings. Amen. I think, you know, as a community, I think you can decide. I think actually your answer might be just on the left, right?

[0:32] My view is-- Male Speaker: Making an error with malignant hyperthermia is different than a hereditary cancer syndrome. And so you could -- I agree you can set a threshold, but they might be different for different conditions. Dr. Right. But I would say my main -- the main point I was trying to make is I actually think for each gene on the list, there needs to be a guide. There needs to be a guide for the lab. and there needs to be a guide for the clinician. This is what the lab needs to look for or report.

[1:04] And this is what the clinician should know. I mean, really, that's My main point. Unless just to get to your point, when there's a way to orthogonally confirm this, I think that changes the threshold. Yeah. Robert. ROBERT GREEN: Yeah, hi. Robert Green from Boston. I might surprise you that I actually agree with almost everything that Sharon said. Although The notion that you can somehow presage all of this and preplan it ahead of time for the individual patient is I think where this breaks down into untenable complexity. And so what I would urge us to think about is the difference between molecular diagnosis in the traditional genetic sense, where there's phenotype first.

[1:51] And then we go in and we make an actual diagnosis. And the notion of using genomic screening. as a way of setting a first step prior probability that can then be contextualized with that individual. through the traditional practice of medicine, gathering other historical pieces of information, focusing in on the physical exam. I think if you think about Secondary or unanticipated findings as a diagnosis, it is completely untenable.

[2:24] to use them in unaffected individuals. The prior probability and the the false positives are way too high. But if you think about them differently as a way of directing the clinician's attention, to refocus on the family history. to refocus on a particular part of the physical exam And then -- now that is a sophisticated way of handling genomic information. Not even every geneticist is ready for that, much less every practitioner in society. But that's the direction I think we're going to have to go, because I don't think we can put this genie back in the bottle.

[3:02] I don't think we can convince the world that it's too complicated for us to share this information back with them again. I'M GOING TO GO AHEAD AND THEN I'LL ASK THE QUESTIONS. My only brief comment is from the cancer perspective, it's very hard to contextualize. So if you have a woman, who was, you know, adopted, whatever, doesn't have a family history, cancer, and she has an incidental finding in one of these genes, you have now told her that she's going to be having MRIs. Like, until she gets cancer, you won't know, you can't contextualize it. So...

[3:35] And that's the VHL, that's the reason that comes up, because of all of the screening those patients then undergo. So I do, I agree with you in principle, I think it's hard to contextualize for many of those genes. And I do think part of what we're being judged on is what's the cost also associated with this downside? So the cost to the person in terms of the provider, the patient, the burden. But also, I mean, we're also worried about third party payers and what they're looking at this. And I don't want to shoot ourselves in the foot with what we want with the main issue in terms of why we're doing this to get distracted in terms of some of these other issues.

[4:09] So I completely agree, Sharon, with the notion that we need to create better resources to support labs in the interpretive process. And I think, you know, defining gene disease validity, what are the diseases these genes are associated with, what -- you know, more direct support for how do you decide what's a true loss of function variant, all those things, 100% behind that. But as you and I know well, it does take time to get through this, and it will take us, you know, even if we prioritize it within ClinGen, it will take several years probably. So, you know, in health care we're often faced with, you know, you do the best with the knowledge you have and sometimes you go forward and then you realize you were wrong and you have to backtrack whether it's, you know, replacement therapy or not or things like that. So I do want to push back and say, Maybe we do need to recognize that we're not perfect with respect to positive predictive value and sensitivity, but we need to sort of figure this out. And one of the things that I think Les has brought up in the future we've had offline discussions about is whether we should create a mandatory sort of database of the people who are getting incidental findings back and track what's happening not only with their phenotypes but with do these variants get reclassified.

[5:31] that lived with VHL. That was not an incidental finding. That was actually a diagnostic case. I would argue that the over-interpretation of variants is worse in diagnostic cases, particularly variants interpreted a long time ago, than it is in incidental findings today. I'm not saying it's perfect, but, you know, we have to— keep that in mind that there was a lot of, you know, oh, he must have this disease, he's affected, and the variant must be pathogenic. So, but to get back to my question, should we create some sort of registry of individuals and make the labs who report these have to put this on the reports that, you know, and request that these individuals register and we can track what happened to them over time. It's just an idea that a lot of us have been discussing.

[6:13] I personally think we should have the registry. I think the compliance with the registry is likely to be very low because of the diffuse, you know, nature in terms of who's getting these reports, but absolutely. Thank you. I'M NOT SURE. Yeah, two brief comments. One is I think that, you know, population-based sequencing programs are going to be able to also help to answer some of the questions related to the contextualization. So I think we'll have another way to try and answer that question. But since payers were brought up, I wanted to relate an interesting experience we had when we went to our provider-owned payer about coverage of exome sequencing for clinical. When we actually presented the incidental finding things, to our shock because we were prepared to prepare for the arguments about, you know, defending the reason for doing that because of cost, they said, "This is a great added benefit. We really think this is a good thing because if this is something that can enhance the health of not only the individual but the family, this is good." Now I seriously doubt that all payers are going to respond that way, but I think we shouldn't necessarily go in assuming that everybody's going to look at this as a potential negative.

[7:23] Thank you. I would like to ask the panel, Specifically-- Ian about where do you think the returning clearly pathogenic results to to pediatric cases. will be for clearly adult onset diseases in about, say, five years when we have a lot more experience with this. Do you think we're going to go in the direction of returning those results, or do you think-- We're going to be more historical. and say, no, you can't have it. We're going to trust the system.

[7:56] to get that information at the right time. to those people. Thank you. Well, if anyone knows basketball, coming from Philadelphia and telling us to trust the system is a weighted topic. I think-- I mean, my gut feeling is we should not be returning those results to pediatrics, but in order To not do that-- there are so many layers to that. One, if you know that information are you going to be obligated to return it or give the option to return it once that child is an adult?

[8:31] If it's impacting the parents, which is the argument on the ACMG list, is there a mechanism that can be put in place where any trio sequencing The parents are automatically given the option or I guess given the option nowadays to do those 59 gene lists and have a separate report that's not reflective of the child, but with the caveat saying once your child is 18, they should consider doing that. So that involves, that's really complex in terms of doing that kind of follow-up and preserving health, which is what the intent of many of this is.

[9:07] Um. I think in pediatrics it's very complicated, and I think that's been argued. in many different forms, but I feel strongly that adult onset conditions should likely not be returned to the child, with the best argument being that it could impact the parents. and there should be different mechanisms. put in place to respond to that rather than just returning the child's result. We're getting close to the answer, but there are less questions above. So I have a comment and question. So with regard to Um.

[9:38] the big discussion here in terms of the positive predictive value and so forth, I think we have to think about the context in which this is applied in medicine in general and health care in general, which is not precise. That doesn't mean that we shouldn't try to make things better and better and better. And the anecdotes of, you know, getting it wrong are useful sometimes, in terms of driving us to figure out what happened and how we can do better. I don't think they're useful to say that the entire-- system should be dismantled. And if you think about this in terms of prevention, which primarily is what we're talking about with pre-symptomatic results, is it starts to blend over into public health, where a different kind of standard and a different approach is taken. So if I decide that I'm not going to plant my crop in Africa, my wheat crop because there might be a few grass seeds in there, I'm not doing myself a favor.

[10:37] So you have to think about that balance, I think, going forward. The question I had was -- well, actually, it was sort of a double question -- is, why do you think it is that the pediatric things which us pediatric geneticists think are clearly actionable and should be here, are less represented in that. And then, I think actually you answered the other one about the BRCA example. Yeah. Well-- I just wanted to comment on your first comment. The reason we're bringing up these anecdotes is we have no idea what proportion of the patients it actually represents. This is also new.

[11:12] And I do think we have to have equally strong data, because it is in the preventive field, as to how much good you're going to do. That's not clear from the list. So just in general, many of the pediatric conditions, although I would love to have nominations for more, are symptomatic. And so we were thinking most of those were going to be in the diagnostic space and or, as people have said, on newborn screening already in terms of the things that have intervention. But certainly if people can come up with other examples, you know, there's no reason to exclude them just because they affect kids.

[11:46] Okay, we're going to have a short break here to grab some more coffee before we move on to cascade testing. If I can have Jim Evans, Kathy Leppig, Sharon Plon, and Bob Wilden up in front for just a minute, please. Thank you. part of the challenge. Thank you. I'll share.