 Lainey Friedman Ross is the Caroline and Matthew Buxbaum professor at the University of Chicago, where she also is a professor in the departments of pediatrics, medicine, surgery, and the college. Lainey is an associate director of the McLean Center for Clinical Medical Ethics and also directs the Research Ethics Consultation Service at the University of Chicago Medicine. Lainey is also the co-director of the University of Chicago's Institute of Translational Medicine. Lainey is a member of the Department of HHS, Secretary Advisory Committee on Human Research Protections, and is also a member of an Institute of Medicine Committee on the Assessment of Studies of Health Outcomes related to recommended children immunization schedules. Lainey is the author of three books and hundreds of articles, peer-reviewed articles, in the medical literature. The three books that I mentioned, one is called Children, Families, and Healthcare Decision Making, published by Oxford. A second one is called Children in Research, Access Versus Protection, published by Oxford. And the third book, which is forthcoming but completed, is Transplantation Ethics Second Edition that Lainey and Robert Beach worked on collaboratively. Today, Lainey is going to talk to us on the topic of, I'm not sure what it means, but I'm going to tell you the topic, children are not canaries, and then we'll moderate the Ann Dudley program. When Stan and Mark invited me to give this talk about ten months ago, am I still in the way? No. Okay. When Stan and Mark invited me to give this talk about ten months ago, I immediately blurted out, children are not canaries, because I knew my talk had to deal with children and animals. And let me explain why. Why I talk about children and canaries in honor of Ann Dudley. So first, why children? So let me tell you, when I first came here to the McLean Center, I came and interviewed in September of 1993. And I had an interview with John Lantos and with Mark Siegler. And both of them asked me within the first five minutes, so when are you going to have kids? I'm a New Yorker. I looked at them, gave them a dirty look, and just ignored the question. So then I went on my interview with Ann Dudley, and she waited about six minutes before she turned around and said, so when are you going to have kids? So I looked at her incredulously. I said, you're a lawyer, you know you can't ask me that question. And she started laughing. And then I got a little bit more flustered, and I said, it's again, some law, I've spent enough time at Yale Law School to know that you're violating my rights. And she kept laughing. And then she turned to me and she said, we've all talked about you. We know you're going to be successful academically. We just want to make sure that you remember the importance of balance in life. So we all want to make sure you're going to have kids. That threw out all my New York suspicions out the window. Thank you Ann. My kids owe you a big kiss. I'm also the pediatrician of Ann Dudley-Goldblatt's children, and I want to thank her for entrusting me to the care of her grandchildren. I know it's a hip of violation, except for the fact that I actually ask for permission, and I know many of them are in the audience today, Zanna and Aram and Jeremy and Carrier here. And again, thank you for letting me take care of your children. Why did I want to talk about canaries? So anybody who knows Ann Dudley-Goldblatt knows how much she loves Noah's Ark. We all know her beautiful, beautiful clothes with animals on them, particularly tigers, although I'll never understand exactly what her fatuation with tigers are. When my children, when my oldest was about two years old, for the first two years, we actually flew to New York about every four months to get pictures of our kids. And at some point my husband turned to me and said, I think we have to acknowledge we're staying in Chicago for a while. We need a photographer. So I went up to Ann Dudley and I said, I need a photographer for my children. And she said, well, I don't know about children photographers, but I know about animal photographers. I have a great dog photographer, and I'm sure if you can take good pictures of dogs, you can take good pictures of children. And so, Bob LeVon became the pediatric photographer of my children, and has actually followed them for the past 18 years. Again, thank you, Ann Dudley. So why children are not canaries? And the answer is the whole concept of canaries in the coal mine, and that children should not be, in a sense, used to find out information about other people, including their parents. It's also interesting to think about the genetics of canaries. So there is the fundamentals of color genetics in canaries. And like canaries, children come in all colors and sizes. But that's where the similarity ends. Canaries are known to be small, quiet, easy to care for, and inexpensive pets. So instead today, I'm going to talk about genetic, targeted testing, as well as population screening of children. My objectives are fourfold, first, to review the historical justification for genetic newborn screening of children, second, to examine two new proposals to increase genetic testing in screening of children on the grounds of family benefit. Third, to examine the concepts of voluntary consent to mandatory testing, and fourth, to consider alternative approaches to these two new proposals. So let me give you some background information. I'm going to give my disclosures that I was a member of the AAP Committee on Bioethics, as well as a member of the American College of Medical Genetics and Genomic Social Ethics and Legal Issues Committee, when the two committees worked together to write a policy statement and technical report entitled, Genetic Testing of Children. I take the credit or blame for mentioning at the Committee on Bioethics that we needed to write a statement, and I said the same thing at the American College of Medical Genetics, and at the Committee on Bioethics, I said, wouldn't it be great to collaborate with the American College of Medical Genetics? And they said, sure, ask one of the members whether they agree. So I asked myself, and I agreed, and so we wrote the policy. Now, writing policy involves compromises, and while the committee members came to consensus, I have to confess that all the members didn't get along as well as Lainey Ross from the Committee of Bioethics got along with Lainey Ross on the Committee of ACMG, and so not all members agreed with each of the recommendations. So it should be known that the AAP hadn't revised their statement since 2001. The ACMG didn't have a statement written in 1995 with the American Society of Human Genetics. I actually invited the American Society of Human Genetics, but they stated they were no longer writing policy statements. So it was just the AAP and the ACMG writing it. And both groups came to the agreement that the moral justification had to be, quote, the best interest of the child. Now, in the pediatric bioethics literature, there's actually no consensus on what best interest means, what interest should be included, whether to include only self-regarding interest or other regarding interest, how to balance short-term and long-term interest, and whether to include child autonomy as a component of best interest. That is incorporated in the UK and in the Universal Declaration of Rights of the Child, and for those who don't know, we're one of two countries that are non-signatory. We can be proud that it's us in Somalia. But anyway, we all agreed, though, it had to be the best interest of the child, and that's going to become very important, particularly as we think about the fact that children really ought not to be canaries. So the first case study I want to look at is expanding newborn screening. The history of newborn screening begins with Bob Guthrie back in the 1960s when he developed both the blood inhibition assay as well as the filter paper on which to collect the blood, and so begins newborn screening, looking to find a preventable cause of developmental and cognitive disabilities. Despite high uptake, Guthrie and what was then known as the NARC, the National Association for Retarded Children, now known just as the ARC, they lobbied for mandatory PKU screening. And that was actually very controversial at the time. The first question asked by the physicians was, should government tell physicians how to practice medicine? And they were also, from the physician perspective, some clinical concerns. The first being that did we know enough about the natural history of the disease? We had actually never done any type of randomized testing to see if the treatment was effective or safe and things of that sort. But despite that, Guthrie and the NARC at the time were very successful and within a decade had 43 states had mandatory testing statutes. Now the traditional focus then when you think of PKU was to take a condition for which early treatment was expected to prevent morbidity. And other early newborn screening conditions also were either to prevent cognitive disability or mortality. But one has to ask the question, is that adequate justification for mandatory program? If you tell a parent that you can do a test and then prevent disability or death, how many parents really are going to refuse? And Guthrie was really less concerned about parental refusals and more concerned that doctors weren't going to offer the test. And that was why he was pushing it. But it's actually only six years after Guthrie developed his test and his filter paper before we even have any standards of what are the criteria that should be included in order to perform population screening. So screening on a universal basis in asymptomatic children in order to diagnose them to prevent disability in the future. And so in 1968, the World Health Organization developed these 10 criteria, the Wilson and Younger criteria. And basically the gist of the 10 criteria are it needs to be an important health problem for which we can diagnose and then treat. And it's critical that there was both diagnosis treatment and again screening is supposed to pick up a lot of individuals and then you have to do follow-up diagnostic testing. So you had to have both the screening test, the diagnostic testing and the treatability and there needed to be a case management over the course of the disorder. The moral controversies are really focused in two areas. One is what do we mean by benefit and what do we mean by consent? So the traditional conception of benefit in the population screening world was that early treatment was meant to prevent morbidity and mortality. But there's been a real push recently, particularly in the genetics community, to expand the conception of benefit, to say that knowledge is good, even when no treatment is available, that by giving a diagnosis early will prevent the diagnostic odyssey. The parents won't have to keep going back and forth to doctors. They'll have a diagnosis even before the child develops symptoms. It also has the advantage and most of the conditions we're gonna pick up in the newborn period are genetic, that it'll be important for parental reproductive decision making and also by picking it up early and in fact in the asymptomatic period we can allow greater scientific study both of the natural history but also potential experimental treatments. The problem is that it's all being done without consent. Again, Guthrie within 10 years had 43 states. Today all 50 states have mandatory testing and no longer do any actually require consent and in one in Nebraska parents don't even have the right to refuse and most of the other states you have the right to refuse on the basis of religious and sometimes of philosophical grounds. The problem with not having consent is that not everyone wants to know about health conditions prior to the onset of symptoms, particularly if it's not treatable. Not everyone wants to know their reproductive risks and finally not everyone wants to be a research subject. So the presumption that we can test these children without their parents permission and then conscript them into research studies is a little disturbing. So why the pendulum shift? Why did we move in a sense from everybody had to give informed consent to the move of universal mandatory screening and part of it and why the pendulum shift to expand from conditions like PKU for which early treatment is helpful to now when we're screening most children for over 45 conditions some of which are non-treatable part of the shift became because of technology. We had this technology called tandem aspect geometry which allows us to test from many many conditions with one blood sample and so moving away from one condition one test allowed for testing of a lot more many more tests than we used to do and basically tandem aspect is now has been implemented in all states. In 2002 it turns out that in some states like West Virginia we were testing for four conditions in Massachusetts we were testing for 40 and so parent advocacy group started to scream that why should a child's health status in a sense be dependent on geography and so the American College of Medical Genetics with HRSA decided to do a study or to get experts together in order to develop guidelines for state newborn screening programs and in 2005 they actually published their recommendations which included 25 conditions and 29 secondary conditions had strong support from parent advocacy groups and strong criticism from some scientists and ethicists. So what were the criticisms? Well first they said they were gonna do a two-part methodology. The main component was a survey of all stakeholders for 84 potential conditions to be included in a uniform newborn screening panel. The invitation was sent by email to the genetics community as well as to parent and patient advocacy groups. Although three of the conditions included were infectious diseases, the email was never sent to individuals with this type of expertise and so the committee in the end just ignored these conditions and in fact there were no invitations sent out to general pediatricians who was sort of if you were a friend of Rod Howells you got an invitation to tell us what you thought of these 84 conditions. The ACMG survey had no coherent analytic framework. Some questions asked about objective facts. How frequent do you think this disease occurs as if that was something that a democratic vote would be able to determine. Other questions mixed fact and value. For example, one question asked the respondent to rank the burden of the untreated condition on a five point scale which made assumptions about the consequences for the infant with his or her normative judgment about how serious a burden those consequences represent for the child and for the child's family. There was no justification for the waiting on the survey and in fact 200 points were awarded to a test if it was known to be associated with a life saving treatment but also 200 points were given if you could actually treat the condition and so after some of the scores came in they didn't like how the rankings were coming so they actually changed the waiting but they didn't actually tell the respondents who had already responded to the questionnaire. The second part of the methodology was supposed to be a systematic literature review. This was never completed. I was at the meeting where the ACMG HRSA vote came about and Rodney Howell literally said to his committee, I wrote the report, trust me, the data are good in the systematic review, trust me and they all voted to support sending it off to the Department of Health and Human Services Secretary without actually anyone having read the report. The problems with the platform technology is that tandem mass spec diagnosis a lot of variation but that doesn't mean that variation is disorders and in fact many of the conditions do not meet the Wilson and Youngner and so we also may be doing a lot of over diagnosing. Some conditions that it does diagnose there are currently no available treatments. Some conditions may actually not be conditions. My favorite is two MBAD before tandem mass only five cases had ever been identified in the first six years in Wisconsin, 27 additional infants were diagnosed and all were from Hmong families. They were told to put their children on these very restrictive diets. They actually didn't follow medical advice and virtually all are asymptomatic despite that. So with diagnosing conditions starting treatment and had we done the treatment who knows whether we would have harmed them. So what's the fate of the uniform panel? As I said it was endorsed by the Secretary Advisory Committee at the time despite the fact that the report actually hadn't been published and therefore not read by the committee that endorsed it. The systematic literature reviews eight years later have still not been performed. Despite that after the dust settle the committee created a new set of criteria for adding conditions. They also said that the same criteria could be used for subtracting conditions but as of 2014 no condition has been subtracted including that two MBAD as I mentioned before. So again one of the commentators on this whole switch to expanding universal newborn screening has actually argued we moved away from a public health emergency to in a sense newborn screening as a public health service, right? So with PKU you need to start diet within the first two weeks of life in order to prevent cognitive disabilities but now we're diagnosing things for which treatment may never need to be performed. This then reduces the arguments in favor of mandatory screening. If we no longer, if it's no longer an emergency we could even argue whether it's needed for emergencies. I actually believe that parents are smart enough to know that they would want their children tested for conditions that are treatable and preventable in the newborn period but even if you don't think parents are smart enough to do that clearly as we take conditions that may not actually be conditions the arguments in favor of mandatory screening becomes much less, much weaker. It also then conflates the fact that we're doing PKU screening and two MBAD screening and so therefore you can understand why some parents get confused and may refuse all of it because it's often, it's only offered as a bundle. You take the whole kit and caboodle or you get nothing. It also ignores the other costs of newborn screening which includes false positives. So as I said, you do a screening and only if the screening is positive do you follow it with the diagnostic test. It's gonna identify mild and atypical cases. We've actually seen that with MCAD where we thought that anyone who had MCAD, medium chain acid dehydrogenase, we thought anyone with MCAD would be at risk for really bad outcomes if they dehydrated. And now as we've been doing newborn screening we've come to realize that it really is a spectrum of conditions and so that we're over diagnosing and therefore driving these parents crazy that if their kids have one episode of acute gastroenteritis that they need to come to the emergency room for IV hydration. It also ignores the cost of newborn screening beyond their blood test. So newborn screening is known to be a system, it starts with universal screening and then it moves to diagnostic and then it moves to case management and follow up. And so to that extent, this whole expansion really starts threatening the whole notion of a universal program. So what did the AAP and ACMG joint statements say about newborn screening? We actually came out in support of holding newborn screening to public health screening criteria for emergencies and not just as a public health service. And to quote, despite variability in how the criteria are interpreted there's broad consensus that to qualify for population-based the natural history of a candidate condition should be understood. Unacceptable intervention for affected individuals should be available and the cost-effective screening and confirmatory testing should be available. We also went on to support mandatory offering of newborn screening that every child should be offered this test but following education and counseling about the substantial benefits of newborn screening, its remote risks and the next steps in the event of a positive screening results, parents should have the option of refusing the procedure and an informed refusal should be respected. So what's the problem? Well first the uniform panel as currently mandated is not limited to the Wilson and Youngner criteria just because we can't include in many of these conditions doesn't mean that we ought to. The second is for those who really argue that the big benefit of this expanded screening is for reproductive benefit, I wanna point them to an article written in 2011 by Bell and Carrius who had created a chip that was gonna cost under $100 which would allow parents to get carrier testing for 450 autosomal recessive conditions. They're actually about 2,000 autosomal recessive conditions. I'm sure they're working on adding the other 1,500 but to the extent that parents wanna know what risks they have for having a child with a genetic condition, the answer is if you wanna know about mom or dad, test mom or dad, don't use the child as the canary. The second case is whole genome sequencing and this is usually done to explain why an individual has a specific problem that hasn't been explained by other more routine type of tests. Now your whole genome has a lot more information than just finding out whether you have a particular disorder. If I look at your whole genome, I'm not only going to be looking at possible causes for developmental and cognitive disabilities, but I'm also going to find those 450 autosomal recessive conditions that Bell has identified as well as many other genetic dominant conditions from breast cancer to colon cancer, et cetera, et cetera. But to learn that additional information actually takes a lot of resources because it's not just like when you put persons, chromosomes into a whole genome sequencing scanner that all three billion base pairs get read out. There's actually a lot of specific analysis that has to be done, expansion of different parts. And when you do that, you're actually gonna realize how much variation is in the human genome and so we're gonna get a lot of variants that we don't know if they're normal or if they're abnormal and so we call that variations of unknown significance or VUS. So whole genome sequencing. In February 2013, that AAP ACMG statement that I told you I'd been working on for a genetic testing of children was published, it turns out that the ACMG had asked us not to include whole genome sequencing in the statement because one, it was still considered research and two, that they were writing a more comprehensive statement. And so our joint statement actually reads genetic research including the use and retention of dried blood spots and whole genome sequencing is beyond the scope of this technical report and accompanying policy statement. What we didn't know was that there was another committee on the ACMG who was actually working on that statement and was gonna publish it within 25 days of the online publication of the AAP ACMG joint statement and they issued their recommendations about the return of incidental findings from whole genome sequencing, as I said, within three and a half weeks. And the gist of the new policy was if sequencing is performed for any clinical purpose, whether in children or in adults, there was a specific list of 56 genes associated with 24 conditions should be assessed in addition to the genetic information for which testing was sought. So if you're looking for cognitive disabilities, by the way, look for these other 56 genes associated mostly with cancers but also with cardiovascular disease, most of which were adult onset. And in fact they added neither patient nor ordering physician consent is needed for whatever test you order, just do these additional 56 genes. Most of the 56 genes are highly penetrant in high risk communities but and most of the genes as I said referred to adult onset cancers are cardiac. And so in a sense then you can say that this ACMG report is actually arguing in favor of predictive genetic testing for adult onset conditions even when no treatment is needed and no prevention is necessary in childhood. Now remember our AAPACMG statement, well it didn't look at whole genome sequencing, did look at predictive genetic testing and here's what we said. Predictive genetic testing for adult onset conditions generally should be deferred unless an intervention initiated in childhood will reduce morbidity or mortality. An exception might be made for families for whom diagnostic uncertainty poses a significant psychosocial burden particularly when an adolescent and his or her parents concur in their interest in predictive testing. So in general we were like you shouldn't be doing predictive genetic testing for adult onset conditions and yet in some families it becomes so stressful that parents and child all agreeing to it we could imagine that there would be exceptions to the rule, we're not the ones living with this disorder, it was to show respect for patients and families. But there was no way that that was suggesting that we would support a mandatory hunt in low risk populations. If I came and did this test and all of a sudden realized that there was a BRCA, so a mutation at risk for breast cancer in a family, it's not like one they didn't even consent for my ordering it, two it wasn't like it was causing psychosocial stress because no one in the family even knew that that ran in the family. So this notion of a mandatory hunt or as the supporters would call it opportunistic screening since you already have the blood sample just do additional tests. Doesn't require consent of ordering physician, doesn't require consent of patient and while the genes are known to be high risk in high risk populations, it's really not known what type of risk it causes in low risk populations. So BRCA again the gene for breast cancer is a good example because actually in high risk families we know it's penetrant about 50 to 80% of the time in low risk families it can be penetrant only 30% of the time. And so we're giving these people these genetic information and telling them that they're at very high risk and they are at increased risk over the general population. It's only about 10% of women will get breast cancer over a lifetime and now these women have a 30% risk but that's a lot lower than the 80% risk. And so we're not sure how women and their physicians are gonna deal with this type of information. The same month that they were publishing their report on the mandatory screening for these high risk genes there was actually a report coming out in Europe showing that many of the cardiac genes that the ACMG thought were so highly penetrant were shown not to be highly penetrant in the European communities. And so lots of reasons to be suspicious of this list of 56. There was a lot of pushback to this second ACMG report some from the pediatricians who were really annoyed not just myself but really annoyed that in a sense they seemed to be flip-flopping one month to the next about predictive genetic testing of children and because of that the ACMG then came out the next month with a clarification reaffirming that testing had to be based on quote the best interest of the child. They affirmed its recommendation not to perform diagnostic testing for an adult onset condition in children and yet here's what they wrote. But you should disclose incidental breast cancer finding because if the child carries a pathogenic mutation there's a high probability that the parent does as well. And that's actually they argued in the child's best interest because it potentially prevents a severe adverse health outcome in a parent. They also argued that it was not inconsistent with their report from February. They said the AAP ACMG when they were saying don't test children for adult onset conditions the AAP ACMG report was focused on high risk families and here we have no reason to suspect and so therefore this was new information for these families to which of course the critics responded we don't have data to show that these are really high-riskly penetrant in these communities either. And their clarification continued in the fact that they had said that no physician nor patient or parent consent was necessary they conceded that patients and physicians should have the right to refuse mandatory hunts and so they began to permit an opt out. But they still argued that the ACMG still assumes that parents should get predictive information about their child's risks which may inform them about their own risks. Despite this clarification I'd argue there's still contradictory between the AAP ACMG joint statement and this new statement. So how big a problem are these incidental findings in the whole genome sequencing? Just want to point out two studies. This is a study which will show you that basically hundreds of incidental findings are gonna be picked up and we're not really gonna know what to do with them. And my favorite example is that there is an individual who put his whole genome on the web and I circled two that are homozygous which means that this individual has these disorders but if you look at the asterisk you'll see that they thought that these would both produce benign phenotype so even though the individuals that risked genetically they didn't expect any health problem. This individual also had two more homozygous conditions and I'm circling them and explaining that cocaine syndrome would lead to failure to thrive, microcephaly, cognitive disability and that Usher syndrome would lead to deafness and progressive vision loss. Now if I were given this genotype about my own child the first thing I would be doing is looking into special ed programs and getting physical and occupational therapy started as soon as possible. The problem is this was done on an adult by the name of James Watson. So clearly the genes didn't predict the phenotype. He clearly doesn't have cognitive disabilities or deafness or progressive vision loss. So what's the problem? There is a consensus against predictive genetic testing of children for adult onset conditions. We wanna know parental risk, test the parents and pediatrics and probably in all of adult medicine mandatory hunts for high risk alleles in the low risk population should only be done as research. So under a research protocol with IRB approval. So let me end by saying that children are not canaries. Children are also not tomatoes. And let me explain why there was an article in the New York time which explained actually that the tomato has 31,760 genes which is about 7,000 more genes than a child has. So either tomatoes are more complicated than our children or it's possible that children are more than their genes. And on that note, children are valuable and vulnerable. Policies need to focus on the child's best interest, mainly focusing on their self-regarding clinical interests but this is not to ignore the children do have non-clinical and other regarding interests but only to say that healthcare policy should focus primarily on the child's clinical interest. So on that note, let me end by saying I love you Ann Dudley and we're gonna have a panel where we're all gonna get to say we love you Ann Dudley. So let me invite up my fellow panelists, John Lantos, Paul Helf, Naranjan Karnak and Emily Landon for a panel discussion of how Ann Dudley has influenced all of us in our careers.