Jeffrey Gruen: The Genetic Roots of Learning Disabilities

Harlan Krumholz: Welcome to Health & Veritas. I’m Harlan Krumholz.

Howard Forman: And I’m Howie Forman. We are physicians and professors at Yale University. We’re trying to get closer to the truth about health and healthcare. This week, we’ll be speaking with Dr. Jeffrey Gruen. But first, what’s got your attention, Harlan?

Harlan Krumholz: I want to talk a little bit about an article that came out that was taking a look at physical activity and risk. You exercise every day, don’t you, Howie?

Howard Forman: I do.

Harlan Krumholz: There’s a lot of evidence that people who exercise, stay active, do better than people who don’t. But there’s another thing that you can explore, which is among people who don’t exercise very much, does it make any difference whether they take a few minutes to run up the stairs?

Howard Forman: Bursts.

Harlan Krumholz: Bursts of activity. People are talking a lot about this. You may know that there’s a great interest in this high-intensity training idea, and that’s really more about 10 or 15 minutes where you’re doing interval training, what they call Tabata, where you’re going really hard and then taking a brief rest and going really hard.

Howard Forman: For a long time, you see people making these pitches about, “All you need to do to get rid of all your belly fat is three minutes a day, or five.” It always sounded like just charlatans on TV, so I’ve been very excited about this too.

Harlan Krumholz: I think this HIT thing, this high-intensity training, actually might be real. Listeners know that I’m a big fan of this UK Biobank, which is this large study of, ultimately, about 500,000 people in the UK. These people volunteered and provided a ton of information and, in many cases, blood and other samples that are going to go on imaging. It’s created this large database which is available to researchers around the world. People continue to mine this database to try to create a new knowledge that can help us.

A group of investigators took a look at the UK Biobank group and said, let’s take a look at people who don’t exercise. Let’s try to focus on them. Let’s ask this question, whether or not these short bursts of activity—but they’re not, like, exercise, running up the stairs for a minute or two, or if you’re walking across the street and you want to get across the crosswalk and then you zip across, you have this extra bounce in your step to get that done—does that make any difference?

And what they did in the UK Biobank was at some point after they had already started, they gave people wearables. A sensor that would go on their wrist and they did that, had them wear it for about a week. These individuals were being monitored. Many studies asked people like, “Hey, do you do this? Do you that?” And people have trouble recalling or maybe they’re a little ashamed to say they’re not moving or whatever they’re doing. It turns out that this is really elevating our research in this area because we’re actually using direct measurements of people using basic devices that they’re wearing on their wrists and those are giving us some sense of what’s going on. This was pretty interesting. And then the other thing about the UK Biobank is they collect a ton of information.

We know that somebody who might zip across the street might have other factors, other things about them, including their age and other conditions about them, which might be influencing their ability to do something like that. When you’re going to study and try to isolate something like that activity, you’ve got to take that all into account. They were able to do that. They studied about 25,000 people and then followed them up for about seven years. And during that time there were 852 deaths that occurred. 25,000 people, 850 deaths...overall, still a minority of people had that happen, but they took a look and they said, compared to people, again, these are non-exercisers generally, but among those people who had none of these intermittent, rapid one- or two-minute bursts of activity, they looked and compared those who were engaged in it to those who didn’t engage in it, maybe saying like three bouts a day, they had a 38 to 40% lower risk of all-cause and cancer mortality and almost a 50% lower cardiovascular mortality.

Now again, sometimes in papers like this, they say “it reduced.” We don’t know if it reduced, but we can say that people who had this had much lower risks of mortality. And then they looked at it in other groups, they said, let’s look at people who actually do exercise and how does that correlate? And again, they saw that small amounts of vigorous, what they were calling non-exercise physical activity—this is like I’m running across the street, like I said, or you’re taking the stairs up two flights instead of the elevator or escalator—and for those individuals, again, this non-exercise physical activity was associated with a substantially lower amount of mortality as a result. I was very interested in this. As they say, “Shut up and eat your medicine.” It was, I thought, an important one that was reported in The New York Times, lots of people talking about it.

Howard Forman: That’s fascinating. Thanks for summarizing that.

Harlan Krumholz: Howie, let’s pivot to our guest today. Something really interesting. We talk about reading disabilities, and this is really a gem that you found Jeff Gruen to join us today.

Howard Forman: Dr. Jeffrey Gruen is a professor of pediatrics and genetics at the Yale School of Medicine. His research focuses on learning disabilities such as dyslexia, language impairment, and speech sound disorder and particularly the genetic foundations or origins of these conditions. His lab has identified DCDC2, a dyslexia gene cited by the Journal of Science as the fifth top breakthrough of 2005. And he leads many projects including the Yale “Genes, Reading, and Dyslexia” or GRAD study, which studies the genetics of dyslexia in Hispanic-American and African-American children. He also started the New Haven Lexinome Project, a six-year project investigating the reading and cognitive abilities of first-grade students to study aspects of genetics in reading and learning disabilities. And he now leads the new cross-disciplinary program on learning disabilities, known as the Yale Program for Learning Disabilities Research. Professor Gruen is a member of the Learning Disabilities Research Center’s consortium funded by the National Institutes of Health.

He received his bachelor’s and medical degree from Tulane University. He started his training at Yale in 1981 with an internship in pediatrics followed by subspecialty training in neonatology and then research training in molecular genetics. First, let me just say welcome to you to the Health & Veritas podcast, and I want to start off by asking you to explain to our listeners how it is that you are able to connect genes to a condition and then a condition to individuals. Tell us about how that work occurs.

Jeffrey Gruen: Sure. First of all, thanks for inviting me. It’s really a pleasure to be here. That’s a great opening question. And let me bring it down to the family level. Families that have children with learning disabilities know that it runs in families. When they contact me and they call me, one of the questions I’ll ask them is, “Who else in the family is struggling with reading or struggling with language?” And inevitably they’ll say, “How did you know?” But it’s been known that learning disabilities run in families for at least a hundred years. But the connection to genes really wasn’t made until really the seventies when folks began to look at twin studies. And preceding molecular genetics or twin studies where people would compare outcomes between non-identical twins and identical twins. And intuitively you can understand that identical twins share 100% of their genome and non-identical twins share about half.

And if concordance is higher—that is, if both kids are affected more frequently in identical twins than non-identical twins—then it makes sense that the reason is it’s because of their genome, it’s because of their genes that they inherit. What’s not so clear to families is that it’s not the kind of inheritance that you would expect that we learned in high school. It’s not Mendelian genetics. It’s not dominant, it’s not recessive, it’s actually all the above. And the reason is because reading disability or dyslexia or language impairment are polygenic. That is they’re caused by multiple genes and an environmental factor. And it is such that in fact from genetics you can determine risk, but you can’t determine diagnosis. The best we can do is risk, but the data and the foundation are those twin studies.

Harlan Krumholz: I wanted to break this down. I’ve got some rapid-fire questions for you, just to set the tone because maybe Howie’s more sophisticated than I am about this stuff. And I want to just get to some of the basics. What is dyslexia?

Jeffrey Gruen: It’s simply defined as difficulty learning to read in the presence of normal brain functioning. Normal IQ, no underlying brain disorder, an adequate opportunity to learn to read.

Harlan Krumholz: Give me an example of some famous people who’ve had dyslexia.

Jeffrey Gruen: I think the most famous for our group and most relevant would probably be Charles Schwab.

Howard Forman: Really interesting.

Jeffrey Gruen: I’ve never met him, but I think he’s done pretty well. And people say he’s pretty highly effective.

Harlan Krumholz: Aren’t there some historic figures also who’ve said they have dyslexia?

Jeffrey Gruen: There are. You can see them on the web. I don’t generally cite them because I don’t think they’ve been formally tested.

Harlan Krumholz: What’s the prevalence of dyslexia? How many people have it? What’s the rate of it?

Jeffrey Gruen: The range is depending upon what study you look at, somewhere between 7% and 20%. But it’s really quite high. It’s very high. I usually quote around 10% as an average. Now the reason why there’s such a wide range is because it depends how you set the cutoff. If you look at reading, if any of us looks at reading performance across a large number of children, let’s say we look at across the 2,300 fourth graders in New Haven County and we did any standardized reading assessment, let’s say we did comprehension in these fourth graders, we would get, we’d get a bell curve, a normal distribution. And what we’re really looking at is the tail end, the poorer performance. It depends where you set your cutoff. You can set it at 5%, you can cut it off at 7%, 10%, et cetera. In the United States, it depends what state you live in.

Harlan Krumholz: A lot of people might ask you though, to what extent is this, and I want to, going to get to this in a second, your connection with genes. But a lot of people might think, “Isn’t this a lot about what your home environment is now?” Lots of people are on screens. I’ve got kids who are reading later now. Then kids are having trouble because they’re missing a window when some of the formative connections are made about reading because they’re watching a lot of videos. How is it that you’re able to identify this as what’s actually mechanistically inherited or biologic versus you’ve got all other things. Some people may have trouble with eyesight or other causes that are making it hard for them to read. How do you disentangle that? When you come up with those numbers, what percent do you think is actually medically driven that is biologically driven as opposed to the environment and behavioral?

Jeffrey Gruen: What we say is that the heritability is ranged between roughly 47% and 82%.

Harlan Krumholz: That’s pretty precise, by the way.

Jeffrey Gruen: This is rapid-fire. And those are consistent with Alzheimer’s, schizophrenia. In fact, the only thing really probably greater than that are autism, which is probably amongst the 90%. We’re talking about heritability, that’s across a population. What I’m telling you is, and I generally think of it in somewhere closer to 80%, is that 80% of our reading performance is because of our genes and only a small part environmental. How do we know that? Because of those twin and family studies that I was talking about. And because you look at those family and twin studies and especially in twins, when you can control for environment, not perfectly but pretty good, that’s how people finally determine that environment is actually a smaller piece in the total genetic piece.

Harlan Krumholz: I never imagined that. Let me get one more rapid-fire, Howie, then going to bounce back to you because it’s just one other question I had that was on the foundational side. Does this vary by language? You’ve got Chinese, which is more pictorial; you’ve got letters. Does dyslexia vary in different kinds of languages?

Jeffrey Gruen: People thought that’s the case, but it actually is not. And if you think about it intuitively it makes sense. Homo sapiens, we use the same mechanism and pathways in our brains to process both verbal and written language. It is true that the eastern languages are more iconographic or pictorial, as you say. That’s absolutely true. But we spent a lot of time in China and the prevalence of dyslexia is about the same as everywhere else in the world. And in fact, in a recent paper from Hong Kong, they also identify DCDC2 as associated with reading performance.

Harlan Krumholz: Fascinating.

Jeffrey Gruen: And those children speak Japanese. In the early reports, I think in the seventies the thought was there were different pathways, but in fact they’re not. And as we see the imaging studies come through now of the last five years, we’re using the same exact pathways, the same parts of our brain.

Harlan Krumholz: And their speaking is normal? It’s just the reading?

Jeffrey Gruen: Typically, to be in our studies, you’ve got to be smart. We use an IQ cutoff. This is not intellectual development problems. These people are smart; these kids are smart. In fact, girls are really smart. And that’s the problem with identifying dyslexia in girls. Because they hide. Especially in the early grades, boys in early grades, they act out.

Howard Forman: You find it.

Jeffrey Gruen: But girls hide.

Howard Forman: One of the things from the Lexinome Project that was striking to me was that over the years, the students continue to track. Once they have the gene, they don’t suddenly improve; they continue to track. But one of the objectives of the project is to figure out what interventions might work for these populations. Can you tell our audience a little bit about what are the interventions that might work, that do work, and what’s the hope for these populations?

Jeffrey Gruen: That’s a great question. Interventions do work, Howie, but there’s a narrow window. And the window is somewhere between first and early fifth grade, maybe as late as sixth grade. If you find a child, you can identify a child who’s struggling with reading in the early primary grades and you offer them intense, long intervention. I’m talking about 90 to 110, 120 hours over the course of a single year. That means pulling them out of class for probably every day for an hour in small groups. And do that for a year. About almost 80% of first graders, second graders, and third graders can be brought up to grade-level reading. And if you retest them two years after completing the intervention, it’s sustained. They’ll still be reading at grade level. But if you misidentify, if you don’t identify the kids and you pick them up in high school, then the response to intervention hits maybe 25%, maybe 30%.

The window of plasticity is early on. And that’s why it’s so important to identify these kids early. And most of the paper/pencil- or computer-based ways that we do this are not very sensitive. You think about it for a minute; you have to do some reading. You have to have some reading skills in order to be tested. Kindergarten is out; they don’t do so great. You can get some language stuff but not really reading stuff. First and second grade, not so great. They get better in fourth, fifth grade. But you’re moving out of the window when intervention works best. The idea here with genetics would be to deploy, on a large scale, a genetic screen to identify children at risk and get them into intervention early, when it really works. Instead of wasting money and time when finally picking the kids up in high school. And there’s some estimates that maybe 50% of kids with reading disability or dyslexia are not picked up even until they hit high school, if they make it.

Harlan Krumholz: I was going to ask you about that, what percent you thought, so 50%. My hunch is there’s also health equity issues in here too. Who gets picked up? And it’s an added disadvantage of you’ve got social determinant issues and then if kids have reading... Is that what you found in the research?

Jeffrey Gruen: What we have always strived to do is to be in the community. As you know, Yale, Harlan, doesn’t have a lab school. We’ve done our research in New Haven Public Schools and because—that’s going to be pretty representative if you look at the demographics, pretty representative broadly of the United States. I think that’s a good place to go. We didn’t really address that, but we did address that it’s well known that all the genetic studies of reading and language, et cetera, have always traditionally been done in European-American kids, in kids of European ancestry, White kids. And there really was no genetic data. That’s why, as Howie mentioned in the opener, the grad study really hit only Hispanic-American kids and African-American kids. And our Lexinome Project is more heavily weighted towards that because that’s represented in New Haven Public Schools. And that’s what we have. Is there an equity issue? Absolutely, there’s an equity issue. It has to do with economics, has to do with location, it has to do with the schools are structured but it’s real and it’s there. Yes.

Harlan Krumholz: Is there a clue anyone listening should be thinking about that tips them to say maybe we should be referred to someone like you?

Jeffrey Gruen: They don’t want to come to me. Last thing anybody ever needs a neonatologist. But what I would say is I think one of the biggest clues is language. We know that children have a very regular, consistent developmental timeline for acquisition of language. We know for example that by 12 months most kids are doing single words, attaining that single word to a meaning. By two years, 24 months, they’re putting two words together. By two and a half, it’s actually more complex and maybe whole sentences or word phrases, et cetera, et cetera, et cetera. There are a class of kids and the frequency is about 6% that have something called “developmental language impairment.” We used to call it “specific language impairment.” Basically these are kids that otherwise attain their developmental milestones but don’t talk. And they may put off speaking until they’re two, three, four, sometimes five years of age.

By the way, speech pathology is really, really helpful for this for families that have it. But what’s interesting is about a third or so of those kids will go on to have a true reading disability or a dyslexia by the time they hit a complex curriculum late in first, early in second grade, a third. And if you go backwards, if you go to kids that you’ve identified say in third, fourth, fifth grade, and you ask their parents, “What was the language history?” They’ll tell you about a third of those kids actually had a history of delayed onset in language.

Harlan Krumholz: That’s fascinating, because you were saying before these are separate, the “speaking” and the “reading” [disabilities], but these kids who are delayed speakers actually then end up having reading. But the kids who are delayed readers, maybe that’s not as much an issue.

Jeffrey Gruen: What I’m saying is that there’s shared genetic effects. And we know that. A third of the kids with language impairment actually going to have reading. And if you look at, there are some people who said that language and reading are almost on a continuum. They’re separate processes and there are separate genes. It’s not all the same genes, but they do share some. And there is a very strong relationship between auditory processing, how we understand each other, how we’re having this very complex discussion right now, how we understand each other and how we read and how we decode and understand what we are reading and process that, they are definitely related. They’re not the same.

And that’s totally different to how we speak. How we express ourselves, the physical motion of how we use our cranial nerve to actually articulate words; that’s separate, again. The other problem is that a lot of kids will be comorbid. About a third of kids will have more than one learning disability. They may have “attention in reading” or they may have “language in reading” or these sorts of things. And it makes it a very, very challenging thing for schools and for parents.

Harlan Krumholz: Wow. So interesting.

Howard Forman: Can I ask you? Before we started the podcast, we had a conversation—and you’re welcome not to answer this, by the way, but can I ask it anyway?—you faced an existential question at the beginning of your research career about what direction to go and you really weren’t sure of what to do, and you asked the person who preceded you. Can you give us a little quick summary of that conversation? Because I found that fascinating, and I think people should know this.

Jeffrey Gruen: Sure. Quickly, as you mentioned in the opener is that I came to Yale to do residency training in pediatrics. And as a young male I became interested in neonatology. Neonatology is intense. It’s fast. You get to do procedures; it’s nice. But along the way I learned that actually this whole idea of doing science, it’s really a cool thing. And I decided during fellowship I was going to incorporate science. And I went into Jim Jamieson’s lab for two years in cell biology. And while I was in Jim’s lab, I happened to meet some people from Sherm Weissman’s lab and was very interested in the things in cloning that they were doing. It was relatively new, and I was really interested in it. I went to talk to Sherman, and I had no idea what he was talking about because Sherman was inventing many of the methodologies that were ultimately going to be used in the Human Genome Project.

But clearly I respected who he was and the accomplishments that he’d been doing by that time. The person who preceded me in the lab at that time was a guy by the name of Francis Collins. And as you know, Francis became ultimately the head of the Human Genome Project for the NIH and the head of the NIH. But in those days he was just Francis Collins who had come here to do fellowship training, go to medical school, do residency and fellowship and genetics. And I knew him clinically, and I knew he preceded me in the lab. I gave him a call. He was at his first job. I think he was at University of Michigan. And he kindly answered of course, because he was just a regular guy then (he wasn’t). It was before he cloned cystic fibrosis.

Harlan Krumholz: He’s still a pretty regular guy.

Jeffrey Gruen: He still plays the guitar. I don’t know if he rides his motorcycle anymore, but I hope not. And he said, “Jeff, there are lots of good labs at Yale you might think about, but I think Sherm Weissman is the smartest guy I ever met.” I didn’t really know what I was getting into, but I did know I was going to jump into the deep end of the pool, and I always thought that if I’m going to do something, I might as well go all out. And that’s what we did. And I stayed there for seven years. I didn’t really have a science background, and I was at the bench for seven years. Sherman was wonderful. It was a great environment. He was very open. He’s a very creative guy, as you probably know if you’ve ever spoken to him. And it worked out really, really well. I didn’t then go into reading disability, he put me on more of a conventional project, eventually got scooped by a private company.

But I had advantages that we were able to use and begin to address this whole idea of learning disabilities and the genetics of learning disabilities. That’s how when people meet me, this is what they think. They think, “Yeah, pediatrician, he likes kids. Reading’s really important.” That’s all true. But that’s not how it happened. It happened [that there was a] competitive edge in molecular genetics. By that time working with Sherman, I had cloned the entire short arm chromosome 6. This is before the Human Genome Project. And when we got scooped for hemochromatosis, which happened by a private company, I said, “What am I going to do now?” And I looked around, and there was the reading disability stuff. And that’s how we started going after it. Yes, I care about kids, it’s absolutely true, but that’s not my clinical background. It was because of the science. And the science is very cool.

Harlan Krumholz: Appreciate you taking the time to be with us.

Jeffrey Gruen: Thank you so much.

Howard Forman: This is amazing. And hopefully we’ll have you back. If you can spare more time, we’d love to have you back.

Harlan Krumholz: That was good. I really learned a lot in that discussion, Howie. But let’s pivot to your side. What’s on your mind this week?

Howard Forman: I want to actually stick to that theme a little bit. And I’m going to talk about my own genetics, which I’m an open book and you and I have been friends and colleagues for most of my professional career and you know a lot about my own medical history. You’ve seen me suffer two catastrophic surgical emergencies that were at least partly, if not fully, due to what I have believed to be one of the Ehlers-Danlos syndromes. And I have a vast array of symptoms that seem to align with one or another of the subtypes of Ehlers-Danlos. And I’ve known that I’ve had some type of collagen problem since I was about 18 years of age. But this great story was published in Stat News a few days ago. And we’ll link that to the website that describes other similar people who have suffered from one of these subtypes.

It resonated a lot with me. Now the title is “Revenge of the Gaslit Patients.” And I personally have never felt particularly gaslit by this, maybe because I’m in medicine, but it was always difficult to tell people that I had this constellation of symptoms. And by the way, many of them seemed completely unrelated to collagen, but that it was likely due to this disease. But I had no proof of that, and I’d been tested for many genetic diseases. They always came back negative. I had specialists at Yale, at the Mayo Clinic, at Johns Hopkins. Those specialists were always very encouraging, reminding me that most such patients don’t fit neatly into a genetic profile. But it nonetheless was challenging to explain to other doctors or other individuals. About six months ago, for the first time, my decoded genome started to actually show positive findings that seem to put me closer to an actual diagnosis, much as is alluded to in the article that I referenced.

At the moment, the best that they can say about me is that I likely have something that we call Ehlers-Danlos classical-like syndrome, which is one of the 13 subtypes that are now identified. There are sufficient other abnormal genes that they’ve identified that could help us refine this further in the years ahead. I bring this up not so much to tell people my own personal medical history but to say that many, many of our listeners have unique symptoms or arrays of symptoms or are predisposed to certain conditions or one or another or even predisposed to be well treated or poorly treated by certain medications that are often written off. People can be described as psychosomatic or symptoms that are psychosomatic. But as we heard today, the genetic revolution is just revving up. And I hope it brings comfort and most of all diagnosis and treatment for so many people who live with chronic conditions and may or may not feel like they are gaslit.

Harlan Krumholz: My hope is that we’re on the cusp of being able to illuminate the underlying cause and help to categorize and affirm what a lot of people have, they know they’ve got something important, they just don’t know what it is. It hasn’t met the patterns that we’ve seen of disease in the past that are our descriptions of disease I think will involve substantially. Right now we’re inherited a taxonomy description of disease that’s a hundred years old probably, in many cases. And we’re, I think, about to reform our understanding of disease in the way we describe it. One thing I wanted to ask you though, Howie, is, because some of our listeners still may not be familiar with this syndrome. Who is Ehlers-Danlos? Who are? What is that? And why is it called that? And do you want to explain that a little bit?

Howard Forman: Sure. It’s a collection of diseases. Like I said, there’s 13 genetically identified subtypes and then there’s another catchall category. The two names are two different doctors that started to describe the syndromes many, many decades ago. The worst of them is lethal at a relatively young age. I in fact have several friends with children who have the more lethal variants of this, it can present with aortic rupture. It can look like Marfan syndrome, it can have very fragile skin, dislocating lenses and so on. I always believed that I had the hypermobile type, which is what was described in the article. And only now as—

Harlan Krumholz: “Hypermobile” means?

Howard Forman: Means that I dislocate my shoulder very easily. I have flat feet, knock knees. I have a lot of other, what we call stigmata or signs of this hypermobility. You know me, I’m a very lanky person. That’s a classic feature.

Harlan Krumholz: Very good-looking lanky person.

Howard Forman: Thank you. I have a high-arched palette. I have dental crowding. These are all very specific features of the hypermobile-type Ehlers-Danlos syndrome. But we’re learning so much about it. And much like Jeffrey Gruen said today, it’s not typical Mendelian genetics. It’s multiple different genes interacting on different chromosomes that ultimately lead to the syndrome. And I think that’s what we’re getting closer to.

Harlan Krumholz: Wow, that’s fascinating. And you know what, that’s so nice of you to share that with us. It’s not easy to talk about your own medical history, but I think it’ll help others who are both experiencing what you may have experienced, but also to know that we’re on the cusp of really some advances which may give people... By the way, I’m doing a lot of work with long COVID and Post-viral Syndrome.

Howard Forman: Yes, exactly. That’s what I was thinking of.

Harlan Krumholz: I believe this is a similar area where we would begin to understand more and be able to help people.

Howard Forman: And even long COVID is probably some combination of not just exposure to a virus but some other factors that may include genetic predisposition.

Harlan Krumholz: Absolutely. I think it’s going to be a combo. You’ve been listening to Health & Veritas with Harlan Krumholz and Howie Forman.

Howard Forman: How did we do? To give us your feedback or to keep the conversation going, you can find us on Twitter for now.

Harlan Krumholz: I’m @hmkyale, that’s HMKYALE.

Howard Forman: And I’m @TheHowie. That’s @T-H-E-H-O-W-I-E. You can also email us at health.veritas@yale.edu. Aside from Twitter and our podcast, I’m fortunate to be the faculty director of the healthcare track and founder of the MBA for Executives p rogram at the Yale School of Management. Feel free to reach out via email from more information on our innovative programs where you can check out our website at som.yale.edu/emba.

Harlan Krumholz: Health & Veritas is produced with the Yale School of Management. Thanks to our researcher, Jenny Tan, and to our producer, Miranda Shafer. They are amazing. Talk to you soon, Howie.

Howard Forman: Thanks very much Harlan. Talk to you soon.