 Thanks, everyone, for sticking around till the end. By this time in the day, you might be thinking to yourself, how exactly do I consent participants and return results to participants in the kinds of exome sequencing protocols that we've been talking about today? My name's Julie Sapp. I'm a genetic counselor here in the Genetic Disease Research Branch in NHGRI. And you may have already come to the same conclusion that we have about answering this question. It's a relatively simple question, but it's a bit of a less than straightforward answer, and so I'm hoping to shed some light on that today. So because you're here, and of course, after Sarah's talk, you shouldn't have any questions about the fact that the informed consent process plays a central role in the participant interface in many clinical research protocols. In fact, one of the most common questions we get about our exome protocols is, can we see your consent form or can we use your consent form? We are more than happy to share our form with anyone who wants it, but as any of our participants can tell you, consenting to our protocols is a bit more involved than just signing a consent form, and maybe by the end of my talk today, you'll understand a little bit why. I think that the reason why it's very tempting for us to fixate or focus on the consent form when we're thinking about these protocols is because we think of this as a very part of a very linear process, and I think it certainly looks that way to our participants. We scientists design a study, we ask people to be in the study, ask them to donate their time or their samples to the study, and then if all goes according to plan, we're in a position to return results to them and to the broader scientific community. I think that when people ask us to see our consent form, the reason why that turns into a 45-minute or hour-long conversation in many instances is because we think of this as a bit more iteratively, and we think of some of these relationships really as not so linear, but as informing one another. I think that part of the reason that is, as Sarah mentioned, that this question about whether or not we should return results is really not something that has really come up for our group. We've really decided on a very active results disclosure policy, and Jennifer started to outline some of those reasons, and I'll go into a little bit of that more. That's part of the reason why this has become a bit of a more involved process for folks in our group. These are some of the questions that I think have come up for us as we have thought about how to go about conducting these kinds of protocols from the human subjects perspective, and questions that we talk about other interested investigators with all about. Questions like, what do I need to keep in mind as I approach my perspective participants? How exactly do I go about getting their consent? I'll go over some very specific strategies that we've employed for our protocols. What about, and by what about, I mean, what about enrolling kids? What about enrolling healthy people? What about enrolling people with intellectual impairments? These are some challenges that we've encountered, and I'll spend some time talking about how we've overcome some of these challenging situations. How exactly do we return results to participants if we've decided to do that? In an outline format, I'm going to go over some of the general considerations and the rationale that we've employed in designing our informed consent procedures, give you a specific example of one approach to informed consent that we use in our protocols, review again how we've addressed some of these challenging situations and patient populations, and talk to you about some findings we have in our experience in returning results. I think that for whole XOM sequencing protocols in particular, as I mentioned, we've found it really helpful to think of informed consent not so much as a one-time encounter but as an ongoing process and as an opportunity for ongoing researcher-participant dialogue. So it's a two-way communication where participants can talk to us about their own goals and expectations and plans from participating in our research, and we, as investigators, can share our goals and our expectations and kind of what we expect to find with participants. Participants who are actively engaged or who feel that they're actively engaged in a research partnership are more likely to remain engaged in an ongoing research protocol. So we'd like to try to keep that in mind as well. So again, I think that we view this as a really good opportunity to share with participants the fact that our research goals drive how we think about returning results and our informed consent process, and it's a great opportunity for us to talk with them about the intersection of their own goals and expectations with ours and how closely those match up. So as pretty much everyone has spent time talking about today, the data are the inherent challenge in these kinds of protocols, and this is what we spend some time thinking about, and we certainly spend a lot of time talking about this with our participants. So in terms of volume, as you know, an immense number of variants can be generated for each participant, and these variants can fall anywhere on a continuum from completely novel and never before described to extremely well characterized, and for those that you can interpret, they also fall on this continuum from completely benign to highly penetrant and likely to be very deleterious. As well, the data are generated and interpreted iteratively, and so what you, our knowledge of the genome is expanding, and there can be further downstream uses and interrogations that can take place with any individual exome sequence. And so I think that there is an inherent uncertainty in the data that we're generating from whole genome and whole exome sequencing protocols. Again, what we know about the genome is a moving target, that's kind of why we're all in this, right? We'd like to know more about it. And so I would like to argue that fully and adequately conveying the scale and scope of genomic interrogation to participants in a really scientific way is neither practical nor possible, so I think that it's important to kind of think about that and to think about how the impact that the data that we learn and the impact that it has on individual participants varies tremendously, and that it may vary and be non-static for us as investigators. And so again, as we have these exome data and we think about new ways to interrogate existing data sets, our involvement with the data are maybe not static and not trivial over time. So with that in mind, let me talk to you about our specific approach to consent that we've employed with one of our particular protocols. So we have two exome protocols that are actively going right now in our branch. ClinSeq, which many people have mentioned here, it's a hypothesis-generating protocol. I'm actually not going to spend a lot of time talking about that. Instead, I'll talk with you about this protocol, our gene discovery protocol. I have a sense that that is going to have, that this protocol has more in common with many of the protocols that maybe you're designing in your heads as you're listening to the talks today. So to tell you a little bit about the study, the goal is to find genes for disorders that are very broad categories. So we enroll pro bands with rare conditions, and we have very broad eligibility criteria. So we have enrolled participants as young as one day old. Our oldest participant is about 83 or 84 years old. And the spectrum of conditions that we approach using this protocol ranges from severe neonatal onset congenital anomaly disorders to more adult onset endocrine disorders. So we often employ a trio approach, but not always. It's not always possible. It's not always practical. But we generally think about a family participating in our study, and we generally think of enrolling families. And that's because we're enrolling moms and dads and kids. The kids are usually affected, the affected pro bands. And we do make comparisons across pro bands, again, as part of our filtering strategies. And again, this is whenever possible. But that's not always possible, and it's not always necessary. As Les mentioned earlier, for one of the disorders that we studied, we were able to find the causative variant with using just one trio. And we employ a qualified results disclosure policy. I do think that this protocol would fall more in that third category that Sarah was mentioning, where we do have a fairly robust results disclosure policy. So in terms of what it looks like, very practically speaking with participants. So if this is, we initiate contact with participants. And then I generally have a phone call with participants describing the study. And the goal of that conversation is to alert them to the fact that we'll be asking them to participate in a very comprehensive gene sequencing study that's kind of unlike maybe anything that they've thought of before. And to alert them about the fact that they may be in a position to make decisions about learning things that they didn't necessarily anticipate. And then we send them the consent form and a one page summary of the study that's written in understandable language, we think, or we hope. Then if they're interested, they call us back and let us know that they'd like to participate. And if that is true at that time, we schedule a visit to the NIH for further phenotyping and to actually have the conversation about consenting to the protocol. So we think about the timeline here where the orange arrow here is the timeline of the protocol. For participants, it looks something like this. We call them or they contact us. We have them come to NIH where they sign our consent form, talk with us, and we maybe do some additional phenotyping. Importantly, that is not concurrent with their decision about what kinds of results they'd like to receive from participating in the protocol. But at some point, they make some kind of decision about what results they'd like to receive. We clear validate those results and return them to the participants at another NIH visit in person. And then there is this ongoing, we leave the door open for ongoing periodic re-annotation of their XM sequence. And that may include some possible future re-contact. And of course, while this is going on, this period of time between consent and then this undetermined endpoint, hopefully we figured out what caused the disorder of interest. We've annotated our secondary variance in a way that we think makes sense to us and to our participants. And we maybe have designed and implemented some additional research questions that we can use their XM sequence for. So because this is really kind of the money point here, that's what I'll spend some time talking about. Sarah mentioned many of the concepts that we review with participants in the informed consent session and during the process are not really that revolutionary to genetic studies. Every consent session contains some information about what we're asking people to do and what their risks and benefits are. But really what we're focusing on here is the fact that they can learn a whole lot of information. So let me kind of break down how we go through that with our participants. So I think that for most participants, the analogy that I find the most helpful to describe the data that we generate from a family's XM sequence is a torrent. Some people have used the analogy of it's like getting a drink of water from a fire hose. I think water really lends itself very well to these kinds of analogies. So I think that we have an absolute downpour or torrent of data. And that our goal is to try to sift through this vast amount of data to find something that we're interested in. So to make sense of all of this information, we frame what we could learn in terms of the goal of the study. So we broadly characterize and categorize our results for ourselves and our participants using this terminology that many people here have used, this idea of primary and secondary variants. So when I'm using this analogy with participants, I talk about how what we're really after here is this nugget of gold, the genetic cause for the disorder under investigation. So the disorder that's affecting themselves or their children. So this is us in the Made of the Mist in Niagara Falls, panning for gold in this enormous river trying to find this one gold nugget. And we find other things. So in addition to that nugget of gold that shows up when we pan through this enormous amount of data, there is other information that we may find. And we think of that as secondary variants. And when we talk to participants, we're pretty careful to let them know that this primary and secondary terminology is really because of the result, because of the goal of the study. It's not that these results in and of themselves are of secondary importance. In fact, for some families, the clinical significance of a secondary variant may far outweigh any of the clinical significance of a primary variant. But we really define this as everything else. And we're very careful to let them know that finding these things is not the goal of the study. That's not really why we're in this business. We're not in this business to find out all the other stuff that's lurking in their genome. We're really after one thing in particular. And the fact that we find these things is inherent to our methodology. As Jennifer mentioned, these data are present. And we don't think that the way that we describe it to participants is that we really can't do this kind of averting of our eyes, that this ends up in our gold pan. And we have to make sense of it for participants. So then we go over a little bit more detail about what exactly these things are. We do have a couple of examples that are present in our consent form. And I usually do review some examples with participants. But I think that the reason why we give examples and the reason why we categorize them instead of coming up with a list is that it is not possible to give people a list of all the stuff that we can find. So we have to categorize it some way. So we talk about autosomal recessive disorders or carrier status for autosomal recessive disorders as pretty much the only kind of genomic variation that we can be reasonably certain that we will find for any individual that we interrogate. And many people are familiar with some examples of the cystic fibrosis, Tasex disease, sickle cell anemia, deafness, these kinds of things. That's how we review this with participants. And then we talk about genetic variants that are actually known to cause or contribute to human disease. And we talk about these things in three general ways. So we talk about the onset of this disease. Is it might we diagnose you with some condition that you maybe haven't been worked up for or some symptom that you haven't really brought to your doctor or that your doctor has dismissed as not genetic by this kind of interrogation? And I'll tell you that has happened in the ClinSeq cohort. Is this disorder treatable or preventable or manageable in some way or not? And then given your family history, is it something that you as the participant might find surprising or more or less expected? So again, it might really not be very surprising for someone who comes from a family where there's a really large proportion of people with hypercholesterolemia to learn that they have a variant that increases their chances to have high cholesterol. But for someone who has a completely negative family history, that same variant might be very much surprising. And we also talk about how a lot of the variant results that we generate, we're not necessarily able to interpret right away. And that many, many of the things that we find fall on the spectrum of normal variation. So we give participants the choices to learn the first two categories. Our thought here is, again, because our goal is to responsibly deliver meaningful results to participants that fall into the secondary variant category. If we can't make sense of it, we're not really in a position to return it to participants at this time. And if it's on the spectrum of normal variation, we really don't explicitly tell them that. So I think one corollary that we try to make very explicit in our informed consent procedures is that going through this protocol is not the same thing as getting a clean genetic bill of health. So again, I think that even though we go through some specific examples of the secondary variants, the most important part of our conversation is, again, some of the general characteristics about secondary variants and how we generate them. So I think participants really resonate with the idea that this is ancillary to our research goal, that we can't help but find these things, and we want to engage with them about how to get this information back to them. I do generally let them know that annotating an XOM is time consuming and that it's done by people, like Jennifer, and not entirely by computers. And it's an ongoing process. As I mentioned, it does represent a departure from traditional paradigms. So many of our families have had genetic testing. They haven't had exomergenome sequencing. And so it's kind of a new idea for them. And that the impact will vary hugely across participants. And then I think one really key point here is that as we're studying a particular disease cohort, again, our goal is to generate these kinds of data for many reasons. It's expensive. But because we can find this other stuff as well, we want to generate these data for as few people as possible. So if we can figure out the cause of whatever disorder it is we're studying by looking at one trio, we're not going to proceed to exome sequencing for every single person that we have in our cohort. And so even though participants consent for whole genome sequencing, what we actually may do with their sample ranges from nothing all the way through genome sequencing, we usually start with exome sequencing. So even though we're having this long and drawn out conversation with participants, they may not even be in a position to make these kinds of decisions. So I think that the main point here is that secondary variants really defy complete a priori delineation and categorization. And I think that participants do understand this once we kind of go through the process of talking with them about what they are and how we go about finding them. So participants in our protocols can choose to receive results or not. They can choose whether or not they want to receive the primary variant and or the secondary variants by category. So they can tell us, you know, I'm really not interested in learning anything where there's no treatment and it would be totally surprising to me. And each participant who signs the consent form and each participant who participates in the study is an independent actor. So even though we enroll moms and dads and kids, you know, a couple can make completely different choices about what kinds of results they would like to receive. And I'll go into a little bit about why that's a challenge for us. And then we usually talk about the duty to warn as this idea that we want to leave the door open for some kinds of variants that we would really like to return to them, regardless of whether or not they've given us our permission to do that. So I think that it's a really fine balance here talking about these kinds of variants and I'm talking about, you know, the breast and ovarian cancer variants here and things that are on that continuum with participants because the goal is not to alarm them. So I do let them know that these variants are rare overall. It's of course, again, not our intent to discover this kind of information, but it's really hard, particularly with some of the cancer pedigrees that Jennifer showed. It's sometimes really impossible for us to predict or make an accurate prediction of who these kinds of variants are gonna show up in. And so it's important to alert participants to this possibility. And again, we do talk about how this is not why we're doing this, but it could be a potential benefit for some of the people who participate in our studies. So again, I think most participants really identify with this intent. They're very comfortable with the idea that we've left the door open to pick up the phone, contact them about a medically actionable variant that could really make a big difference for their healthcare. So I mentioned this earlier. I think that any time a protocol employs any kind of genetic testing, it's important to consider the familial implications. We go beyond some of the misattributed parentage, boilerplate language that's included in all consent forms and that most people use when consenting participants to go into a little more detail. So not all family members may undergo the same level of genomic or exomic interrogation. So again, we usually enroll a trio at the time of consent, but we really may only send the pro band off for sequencing. And as we talk about secondary variants, we do spend some time just asking participants whether or not they've mentioned to their extended family that they're having their genome sequenced because they may be in a position to share important health information with other family members. And then some approaches require communication among family members. So as David mentioned, sometimes it's really critical to have a complete trio or to have a trio and another sibling for a spouse that it has sole custody of her child and maybe dad's not involved. We really maybe not be in a position to be able to proceed unless we can get the paternal sample. Similarly, it's possible for any two members of a trio to triangulate the genetic information of the third member of the trio. So again, even though everyone who participates makes their own decisions about what results to receive, if a mom and a child elect to receive their results and they compare notes, then they automatically know what's going on with dad. And again, we do employ this protocol with minor children and I'll get into that a little bit more specifically in a minute. So why do we go into so much detail with our participants and why am I going into so much detail with you today? Again, because our goal is to return meaningful results to participants and because we want to enable our participants to make informed decisions about results. And I think that the theme that kind of underlies these two points is this idea that I do discuss with most participants is that once you choose to know something, it's really impossible to return to a state of ignorance. And I think that's something that's important for both participants and investigators to keep in mind as we're thinking about generating these kinds of data. So withdrawing from this protocol is maybe not so straightforward. This is not like other genetic testing that people have had done. Even though it's clinically available, we do this in a, so there are some alternatives. Again, this is currently only available on our research basis. And withdrawing from the protocol may not be a straightforward process. So we talk with, we alert participants to the idea that if they change their mind and we've already generated their sequence data, then we need to have a conversation with them about how we're gonna manage those data. And I think another interesting point to convey is that we're kind of all hoping that this enters clinical practice in the future. So even though we're making a big deal about all of this information that we can find now, I think that within the lifetimes of many of our participants, this kind of testing is something that their clinicians will order for them on kind of a routine basis. And I think many people understand that, but are yet excited to be getting in on the ground floor of these kinds of technologies. So some challenging populations and situations that I just wanted to review briefly. A parent may consent on behalf of their minor child for this protocol. And we do employ a specific consent form, specifically for parents to sign on behalf of their children. And we will return some results for minor children under certain circumstances. So as Jennifer mentioned, there are guidelines about what kinds of results to return for people who can't make their own decisions about what kind of genetic information they wanna know about themselves. So really, ideally, the only kind of result that we feel really good about returning are medically actionable variants that are actionable in childhood and that could really alter the health of a pediatric population. But sometimes we find ourselves in a situation, like I said, where we only have exome data available on a minor pro band. And so because our families are typically of reproductive age, we are okay with returning carrier status for minor children because it could have health implications for that family if the parents would like to receive that information. And it's the same thing with variants that are only actionable in adulthood. Again, there are very special circumstances under which we would prefer to return that information even though it is on a minor. So if we find a BRCA one or two mutation in an eight year old and we don't know which parent it was inherited from, we kind of wanna pass that information onto the family so that they can make good decisions about their medical care. But in general, we talk to parents about how we would like for their children to recontact us at the age of majority. So when they turn 18 or when they get close to 18 to talk to us about the fact that their parents enrolled them in the sequencing study when they were five or six or 10 or however old they were and that we can find genetic information about them and would they like to learn this information? So that is the ideal circumstance. As with most protocols, we can enroll people with intellectual impairment and because many of our disorders are severe congenital anomaly phenotypes, this happens a fair amount. And we do require a legal guardian or surrogate decision maker and we require proof of that prior to consenting participants and we have drawn on the services of our ethics colleagues to help us sort through this when appropriate. The case of intellectually impaired minors is another circumstance that we have found ourselves in and it involves a thorough and nuanced discussion at the time of consent because really the rationale for withholding some kinds of variance for pediatric populations is very closely tied to a child's decision making capacity and wanting them to achieve their own decisions at an appropriate time. But for some severely intellectually impaired minors, this is maybe not necessarily going to be in their future and we talk with parents about how we are willing to share results for their intellectually impaired minor child of really any kind according to their preference but it's not something that we usually specifically state to parents but we do wait for parents to raise with us their child's intellectual impairment and future decision making capacity. So I think that this whole conversation should hopefully make you think that this kind of research is maybe not appropriate for everyone and indeed some participants are kind of turned off by this just from our initial conversations. So we're really hoping and looking for people who are willing to engage with us over a period of years where there isn't a family structure that will impose additional burdens on participants as they're trying to make decisions about what kinds of results they would like to receive and then where there are some discussion of medical and social resources that are maybe available to these folks or it's at least important for us to consider. So again, how exactly do we get results back to folks? So as Sarah mentioned, there are many different results disclosure policies so there's this idea that you can not give people any results, that you can give them everything that you find or that you can choose some results disclosure and I would say that that's what we employ. So again, this importantly happens not here but here. So during the consent session we again review all of these possibilities with participants and we ask them to think about what they might say at that future conversation but they don't commit to a results decision at the time of consent and there are no boxes that they check on our consent form saying that they want one category or another of results. So again, the process here is that we review their results preferences that we kind of come to together during the informed consent process and we note those in our charts. There's no commitment to a preference at the time of consent because we recognize that people's perceptions about what kinds of results they'd like to receive may change over time. And so after they leave, annotation proceeds per the study goals so we do, again, the business is to try to uncover that primary variant and participants are recontacted when some results are available. We review the broad categories that we talked about at the initial consent session at that time. They make an election about what kinds of results they'd like to receive. We validate those in our clinical lab and then we ask them to come back to NIH for an in-person review of their results and this may happen more than once as we re-annotate their sequence. So just a few anecdotal quotes I have up here from our participants. So one father of a child with a rare disorder said about secondary variant results for himself and for his child. I think knowing is better than not knowing. I absolutely want to know whatever you find. So I think that for some people it's deeply unsettling that we might have this kind of information and that they are not privy to it. Another participant actually asked, do you have to look? That sounds like a whole lot of effort. Do you have to go through all that effort? And I thought that was a great question. Another mom said, I'd probably want to know. I have a great fear of knowing, but I'd want to know. And so I think that, again, these are the types of responses that we get from participants when we talk to them about this information, but really at the end of the day, most people say that they want to learn absolutely everything and that includes parents wanting to learn any and all results that are available for their children. And some parents have been uncomfortable with some of the ethical distinctions that we've drawn between different kinds of variant results and this idea that we want their children to re-approach us when they can make their own decisions. So I'd also say in conclusion that I think that we feel very fortunate because during these conversations, it becomes pretty clear to us that the participants' goals and our goals are really closely aligned. So these are parents or people affected with rare disorders whose children are affected with rare disorders or people who are affected with rare disorders and they are very motivated to help us find the molecular etiology for these disorders. I think in general, patients understand the complexities involved in this process and their preferences about what they'd like to receive or the way that they approach those results or the meaning that they ascribe to results that are available varies tremendously across participants. So I hope that today I've left you with the impression that these can be iterative relationships, that there can be relationships that go both ways between the design of our protocols, how we communicate with our participants about being in our protocols and how we return results and that in general, more transparency about this process can go a long way towards accomplishing our research goals. And with that, I'll thank many people, both in our lab, particularly my colleague Flavia who's the ClinSeq expert, all my conversations with people from our group, including Jennifer and Sarah and of course the participants in our rare disorder protocol and the ClinSeq participants. Thank you. Thank you.