Genetic Testing in EDS with Dr. Paldeep Atwal

[00:43] Dr. Linda Bluestein: Welcome back, every bendy body, to the Bendy Bodies Podcast with your host and founder, Dr. Linda Bluestein, the Hypermobility MD. Today we'll be talking to geneticist Dr. Paldeep Atwal. I am so excited to talk to Dr. Atwal again after our last conversation that was about 4 and a half years ago. Dr. Atwal and I have a couple of patients in common, and we're going to hear about one particular case that's so interesting because the patient presented really looking very much like they had hypermobile EDS, but Dr. Atwal did genetic testing on this person and discovered that they actually really had a very rare type of Ehlers-Danlos syndrome or EDS.
[01:21] Genetics is so important. It's so complicated and so nuanced. So this is going to be a really important conversation. Dr. Atwal is a board-certified clinical and medical biochemical geneticist. He formerly served as Mayo Clinic's medical director for the Center for Individualized Medicine and clinical lead for the Department of Clinical Genomics at the Jacksonville campus. Dr. Atwal's clinical interests include undiagnosed diseases following lengthy diagnostic odysseys and inborn errors of metabolism, including mitochondrial diseases. Through his work, he helped discover two new genetic connective tissue syndromes that result from defects in the FLNA and AEBP1 genes and has published extensively on human genetics with over 100 publications to date.
[02:05] Genetics is a very exciting and very fast-moving field, so this is a really important conversation. As always, this information is for educational purposes only and is not a substitute for personalized medical advice. Stick around until the very end so you don't miss any of our special hypermobility hacks. Let's get started.
[02:29] Okay. I am so excited to be here with Dr. Atwal. I spoke to Dr. Atwal back in episode 8 called Demystifying Genetics. That episode was released over 4 years ago. So, in the world of genetics, I'm sure a lot has changed since then. Am I correct about that?

[02:47] Dr. Paldeep Atwal: You are. And I can't believe that was 4 years ago.

[02:49] Dr. Linda Bluestein: Wow.

[02:49] Dr. Paldeep Atwal: That's so long ago. It feels like maybe 2, but not 4.

[02:54] Dr. Linda Bluestein: Right, right. Actually, Rob looked it up for us. It was actually 4 years and 4 months ago that the episode was released. So we had the conversation, of course, before that. So it's pretty crazy to think about that.

[03:06] Dr. Paldeep Atwal: It is really crazy. Time flies, as they say.

[03:09] Dr. Linda Bluestein: Yeah, for sure. So I am so excited to chat with you. Of course, so many people have questions when it comes to genetics in this space. So this is such an important conversation. And maybe you can start by giving us a little Genetics 101?

[03:26] Dr. Paldeep Atwal: So when you think genetics, we really understand genetics very intrinsically. When we start giving examples such as why do we look like our parents, and why do we have similar hair color, eye color, skin color — all of these physical features — we know what genetics is. It's these units of heredity, if you will, that are passed down from one generation to another. And it's what makes us look what we do, think how we do, and grow and develop normally.
[04:05] So most of the time you think genetics, you think DNA, right? DNA is the smallest piece of genetic material that's housed in chromosomes. And then those chromosomes are what's in your cell that really tells your body how to grow and develop. Now, there's a very simple 46 pairs of chromosomes. It's surprising for some people to learn that chromosomes 1 to 22 are numbered based on size. So there's nothing more complicated than that — it's just how big they are. People think there's some weird, interesting formula to figure out why a chromosome has a certain number. It's just because it's the biggest one, and then 22 is the smallest. And then X and Y, the 23rd group — two Xs, you're female; having a Y chromosome means you're male. And really, that's the basics of genetics. Obviously there's much more to that that we'll talk about, I'm sure, but that's the basics.

[05:09] Dr. Linda Bluestein: And of course we're going to talk about hypermobile EDS today and related conditions, because we know that there's a lot of differences, right? But there's also a lot of phenotypic overlap. So could you explain what phenotype and genotype are and why that's important?

[05:27] Dr. Paldeep Atwal: Yes. So genotype, if we turn the word around a second, is the type of gene — what version or what type of gene do you have at that particular locus of that particular gene. And then the phenotype — "pheno" is the term used to describe the actual physical clinical presentation in someone. So when we use the term phenotypic expression, it's the expression of the findings in a physical way.
[06:06] So if the gene — let's say the genotype causes, think of something ridiculous like green hair — then the phenotype would be green hair. And so that's the phenotypic expression of that genotype. And then maybe another one causes bright blue hair. We use the term genotype-phenotype correlation. And that's very important when they correlate, but also when they don't correlate. So if your genotype says you're supposed to have blue hair and you have blue hair, great, they're correlated. But if it doesn't, that's also interesting. So what happened? When it makes sense, it's useful, but also when it doesn't make sense, it's useful as well. But that's the difference between genotype and phenotype.

[07:02] Dr. Linda Bluestein: So that's really interesting. So you're saying that sometimes you can actually look at the genotype — the type of gene — and you expect to see a certain type of phenotype from that genotype, or the expression of the gene, and it doesn't correlate like you would expect it to. Okay.

[07:17] Dr. Paldeep Atwal: And this happens regularly in my world, of course, because we have — let's say we have a genetic finding that the lab isn't too sure about. So it's my role as the clinical geneticist to look at this and decide if it's clinically meaningful for that patient. So I'll look at the genetics and look at what is expected to happen, right? A certain thing is supposed to happen in the body based on that genotype. And the term we use, and you're obviously familiar with this as a physician, is clinical correlation, right? So is there clinical correlation or is there phenotypic expression of that genotype in the patient?
For example — and this happens often in Ehlers-Danlos — I'll do a large panel and they'll find a variant of uncertain significance in one of the genes in that panel. And the gene is responsible for a very severe skeletal dysplasia that's obviously manifesting from childhood. And so the patient naturally is concerned: what does this mean? Is this — "Do I have this? I looked it up, it looks horrible." All of that is normal. So you look at it and I'll explain that this gene causes this condition, and this condition is obvious from birth. They're sitting there 40, 50 years old, not having any of those symptoms. So they don't have clinical correlation, and they're not going to suddenly develop clinical correlation from a condition like that — you would have it from birth. So I tell them there's no clinical correlation here, there's no genotype-phenotype correlation, and therefore you don't have to worry about that.

[09:14] Dr. Linda Bluestein: Okay, so that's really, really interesting. And we're going to for sure talk about variants of uncertain significance, and we are definitely going to talk about some real challenges in this space, like direct-to-consumer testing and people who don't have a clinical geneticist like you on their team, which is really challenging. So probably a number of people are going, wait — I didn't think that you could have a genotype and not have the phenotype. So how could that possibly work?

[09:44] Dr. Paldeep Atwal: Yeah. Most of the time we think a single gene change, you expect it to produce the phenotype of that gene change. So let's say you have a disease-causing or pathogenic change in a gene — that's the genotype — and you'd expect the phenotype to be whatever disease that was. And that's pretty clear. It may be surprising for some people that in some conditions, that happens all the time.
[10:10] For example, if you take a condition like Huntington disease, a neurodegenerative condition, if you have the genetic change that causes that condition, you're going to get the condition. It's only a matter of time. Now contrast that with something like hemochromatosis, a condition of iron overload in the liver. You can have the genotype for hemochromatosis, meaning 2 abnormal copies of the HFE gene. Only 5% of the time, though, do you actually have phenotypic expression of full true hemochromatosis. So around 95% of the time, you have the genetic change but you don't have hemochromatosis.
[11:01] Most genetic conditions are somewhere in between those two. And people quite rightly will ask, well, why? This doesn't make any sense. It doesn't make any sense in isolation, but those genes do not live in isolation. Those genes live amongst 20,000 other genes that are having interactions and relationships and effects on each other. And so that's why sometimes you don't see the condition. We're only starting to figure out the complexities between these gene-gene interactions.
[11:39] What we really practice currently in medical genetics, for the most part, is Mendelian genetics, named after Gregor Mendel — you probably learned about him in high school biology textbooks, the peas, dominant and recessive inheritance, where a single gene or single trait is passed down in a certain fashion. And that obviously is true and we see many conditions like that, but there's much more to genetics that we're only starting to uncover: complex traits, complex disease genetics, gene-gene interactions, and so on and so forth.

[12:13] Dr. Linda Bluestein: Wow, that's really fascinating. And I like that you gave two examples that are really on opposite ends of that spectrum. Those examples really help to illustrate the challenges that we can see in genetics.

[12:24] Dr. Paldeep Atwal: That kind of surprised me, to be honest. I didn't realize that with hemochromatosis, you could have 2 abnormal copies of the gene and only 5% of people would actually get hemochromatosis. I did not know that.

[12:43] Dr. Linda Bluestein: Yeah, it's such a common trait that, thankfully, that's the case.

[12:50] Dr. Paldeep Atwal: Right, right, right. So what about — you gave those 2 examples, very opposite ends of the spectrum — what about the inheritance patterns for the other types of EDS? We'll put hypermobile EDS aside for a minute because of course we don't know the genetic marker for that yet, but for the other types of EDS, where do they fall in there, do you think?

[13:12] Dr. Linda Bluestein: So most other types are dominant. There are some recessive, but most are dominant, meaning you need one abnormal copy of the gene to manifest. And that being said, even within a family with classical EDS or vascular EDS or something like that, you'll see a spectrum. If you have the genotype, the genotype-phenotype correlation is high, meaning that if you have the gene change, it's highly likely — much closer to 100% — that you're going to see changes.
[13:54] And just to clarify, since we're really diving deep into genetics here, there are two genetic concepts that are often confused, and I think it'd be great to have an explanation from a geneticist on this. I explain this to patients all the time as well.
[14:20] So there's the concept of incomplete penetrance — meaning the gene does not always manifest the disease. It's incomplete penetrance into the physical spectrum or the clinical spectrum, what you actually see in the patient. Now, that's different from a similar but distinct genetic concept: variable expression. You could have a gene where, let's say, in the same family with the same gene change, a sister is more affected than a brother, or a parent is more affected than a child, or vice versa. The same gene, yet in one individual it's affecting them more — the phenotype is more severe — than the other. That term is called variable expression.
[15:26] So they're both penetrant, but one person has it milder and the other has it much more severely. Yes, they both have it, but one has it much more than the other. So there are layers and layers to these things.

[15:42] Dr. Linda Bluestein: This seems like it also makes it really hard for a condition like hypermobile EDS to find the gene or genes that might be responsible.

[15:52] Dr. Paldeep Atwal: Absolutely, absolutely. And I think there are actually clusters of people within hypermobile EDS that are distinct in their own way. You know, you see people that tend to have severe autonomic features — POTS, GI motility issues, gastroparesis, temperature dysregulation. And in some individuals I've seen, the dysautonomic features are their most challenging part of their disease burden, more than the joints — sometimes significantly more. And then there's a subset where it's really the mast cell issues, and some don't really have mast cell issues. So we've got distinct groups within the group. And you see that in many conditions. I think a lot more work needs to be done to characterize that.

[16:53] Dr. Linda Bluestein: Yeah. And that paper — I'm trying to remember if you were one of the co-authors on that paper about phenotypic clusters that came out recently — I think they identified like 3 or 4 different clusters. Do you remember seeing that?

[17:00] Dr. Paldeep Atwal: Like 3 or 4 different clusters? Yeah, yeah. I agree with what they found — it's very true. And I see it in clinical practice every day that there are absolutely phenotypic clusters. And maybe 3 or 4 is a good start. Maybe that's accurate, and maybe there are more. We're just starting to look at that.

[17:23] Dr. Linda Bluestein: Yeah, some had a more predominant mast cell picture, like you said. Some had a more predominant neurologic picture, etc. But that could be gene-gene interactions, influence of the environment, incomplete penetrance, variable expression.

[17:37] Dr. Paldeep Atwal: All of those phrases, all the buzzwords.

[17:41] Dr. Linda Bluestein: Yeah, I just thought I'd throw that all in there, make a little soup out of it.

[17:46] Dr. Paldeep Atwal: Which is what it is.

[17:47] Dr. Linda Bluestein: Yeah, wow, fascinating. Okay, so when it comes to the inheritance of hypermobile EDS, can we really actually say anything about that if we really don't know what the genetic marker is?

[18:01] Dr. Paldeep Atwal: That's a great question. And the answer is we can, based on — if we take a step back and think historically — prior to the ease and advent of large-scale molecular genetic testing. The geneticists before my generation didn't have access to genetic testing like we do. They weren't sending exomes, they weren't doing 50+ gene panels and getting them back in a couple of weeks. They would maybe have access to a single gene they could test for if they were lucky — or not even that, if they didn't know the gene.
[18:36] So they had these big textbooks of human malformation that we had to learn through training. You would look at these textbooks for patterns, and then there was this resource called Mendelian Inheritance in Man — used to be textbooks, now it's online. These are catalogs of different syndromes. And once you identify someone has one of these, you can follow what happens in those families over generations. You can see it is genetic, recurring in a certain pattern.
[19:24] Now we know different patterns infer different inheritance patterns. For example, if you have generation to generation, that's different from only siblings being affected with no generation to generation. So we see it's generation to generation, and we also see the hallmark of dominant inheritance, which is male-to-male transmission — a father can pass it on to a son. Now, that doesn't happen in X-linked inheritance, it doesn't happen in mitochondrial inheritance, and obviously it doesn't happen in recessive inheritance. So because of those features that have been documented throughout the past few decades of describing these kinds of Ehlers-Danlos hypermobile and other types, we know it's dominant.
[20:17] Now, that being said, back to the point you made earlier on genotype-phenotype correlation — often the expression is highly, highly variable within a family, such that it's questionable whether someone even has it. And I get this question all the time from patients: "How do I have it? My parents aren't affected." It can be quite challenging.
[20:42] I think it's even more challenging now because we're used to every genetic condition having a gene that we can test for. But that's normal now, which is a huge testament to the success of genetics. Four years since we did this podcast is like 15 years of medicine in terms of genetic advancement. Clinical practice has changed in terms of what testing we're doing, what genes we're looking at, and it'll change another 4 years. The rate of change in a specialty like this is remarkable to think about.

[21:27] Dr. Linda Bluestein: I think maybe we should book our next interview now so that we can do it in a year instead of in 4 years.

[21:35] Dr. Paldeep Atwal: I agree.

[21:38] Dr. Linda Bluestein: And when it comes to hypermobile EDS — we have the 2017 criteria for hypermobile EDS, and then if you have symptomatic joint hypermobility but you don't meet the criteria for hypermobile EDS or have another condition to explain your symptomatic joint hypermobility, that's the current definition of hypermobility spectrum disorder, correct?

[21:58] Dr. Paldeep Atwal: That's correct, yeah.

[22:00] Dr. Linda Bluestein: So some people are saying, well, are these really the same condition? Are they different? What are your thoughts on that?

[22:08] Dr. Paldeep Atwal: Yeah, I think we need to do better with the criteria. We need to look at a more holistic view of what's really affecting people in terms of clinical criteria. There's probably not enough — certainly not enough in my opinion — when it comes to the dysautonomic features and mast cell involvement, those kinds of things. We're not adding enough of that.
[22:36] I also never like the idea of a very complex diagnosis — a genetic condition with multiple overlapping areas and multiple systems involved — simplified down to ticking a couple of boxes with A's, B's, and C's. I think it's hard to do that, and while it's not to say it's not useful, if we do that, we miss things.
[23:03] I think with the hypermobility spectrum, it's not my favorite term, if I'm honest. I think there are problems with labeling people with hypermobility spectrum disorder without giving them clarity. People feel a bit in limbo with that diagnosis. I tend to try not to do that and look one way or the other.
[23:33] And there's also — just on that note, bring me back if I tangent too far — there's also a subset of patients I've seen that's very interesting. They are not hypermobile, yet they have all the other features: POTS, GI motility issues, mast cell, temperature dysregulation, even bruising issues and other skin manifestations. But they don't have joint hypermobility, though they do have joint pain. So I think there's a lot more work we need to do to figure out these different groups. There are likely many categories we don't even realize are there.

[24:14] Dr. Linda Bluestein: Yeah, I definitely have observed that as well. And sometimes it's really challenging because the criteria start with generalized joint hypermobility and since we don't have a lab test for hypermobile EDS at this point in time — but I'm going to put you on the spot and ask you, if you were to take a guess about prevalence — and we'll come back in like 4 years and look back at this episode and say, "Okay, how close were you?" — what would you think?

[24:49] Dr. Paldeep Atwal: Well, can I qualify it as an educated guess?

[24:55] Dr. Linda Bluestein: Yes, yes, of course.

[24:57] Dr. Paldeep Atwal: With some history. So there's another genetic condition called Fabry disease. It's a lysosomal storage disorder. Prior to the advent of molecular genetic diagnosis, the prevalence was — I don't recall the exact numbers, so let's just give an example — let's say it was 1 in a million. And then we were able to test for it genetically. But the biggest thing that came was a treatment for it. Once there was a treatment, everyone started getting tested for it, and suddenly the prevalence became much, much higher. So it went from 1 in a million to, let's say, a much lower number. And then it was put on newborn screening in certain states, so now we're looking at it at every child at birth. The numbers have dropped thousandfold in magnitude in terms of the diagnosed prevalence.
[26:08] I think if you look at what people historically thought about hypermobile Ehlers-Danlos, something similar may happen. I think the true prevalence is closer to 1 in 1,000, or even less than that — so 1 in 500 to 1 in 100, right? We're on par with things like type 1 diabetes. It's much more common than we realize and very underdiagnosed, very underrecognized.
It's a specialty issue. The specialty for the diagnosis is mainly medical genetics, but most geneticists reside in children's hospitals and don't see adults. Many patients manifest or are already in adulthood. So we have these compounding problems, and there are hardly any geneticists in the country to begin with. And this is just the US — I see patients all over the world and they don't have a geneticist in their country to even make a diagnosis. I'm not talking about small countries; I'm talking about big, billion-plus population countries that are struggling with these issues.
[27:16] So for all those reasons, I'm going to say the true prevalence is much closer to 1 in 1,000 or less.

[27:26] Dr. Linda Bluestein: And when you say "or less," you mean—

[27:28] Dr. Paldeep Atwal: Oh, sorry — 1 in 500 to 1 in 100.

[27:30] Dr. Linda Bluestein: Right. Yeah, I just wanted to make sure I understood and to make it clear.

[27:35] Dr. Paldeep Atwal: Thank you for clarifying.

[27:36] Dr. Linda Bluestein: Yes, of course. Okay. So we're going to take a quick break and when we come back, we are going to cover a bunch of other fantastic genetics topics with Dr. Atwal, including direct-to-consumer testing. We'll be right back.

[28:51] Dr. Linda Bluestein: We are back with Dr. Atwal. I'm so excited to chat with you some more about EDS and all of these complicated genetics. I think nowadays part of the confusion is, you know, you can just shop online like you're shopping for clothes and purchase your own genetic tests. And I think that people might really be blown away by the first part of this conversation — at least I feel like I am — to realize that whether it's variable expression, incomplete penetrance, or gene-gene interactions, A doesn't always equal B.

[29:28] Dr. Paldeep Atwal: Yes. And I think as well, there's the issue of the quality of the testing in question. I've worked and helped with a number of companies looking to facilitate genetic testing, which is a very good thing to facilitate. In that capacity, we almost became like mini regulators — we would only look at supporting tests if they were clinically valid and clinically meaningful.
[30:01] I think the average member of the public would think that if a test is available, there must be some good regulation that has gone through to ensure that it's clinically relevant, clinically meaningful, and has clinical utility. They may be surprised to hear that in some cases, that's not the case. For example, I remember in that capacity, I had a test come to me where they were offering genetic testing to see what position in a football team you would be best at.

[30:38] Dr. Linda Bluestein: Are you serious?

Dr. Paldeep Atwal: Yeah, I'm serious. Obviously this is nonsense, right? And the other thing people do is they take odds ratios and try to apply them to create big effect sizes when they don't exist. For example — and this is another thing that's difficult to explain, but when you explain it this way, it's not — if someone tells you your risk of something doubles, that's kind of alarming, right? If it's a cancer or something else, then of course there's more alarm. Now, if I told someone their risk of a particular rare cancer is 1 in 10 million, and I said your risk is now 1 in 5 million, that's not really a meaningful change. But I could also say your risk is doubled for this, right? Taking things into context is critical.
The second thing is quality. There's quality in all aspects of the genetic test process, just like any other test process. There's the quality of the collection, the extraction of the DNA, quality control in that extraction to ensure there's enough good-quality DNA to run the report. Then there's the actual quality of the sequencing. The sequencing, for the most part, is actually generally done well by most labs — it's very standardized, often on the same platforms.
[32:42] The biggest issue comes after the sequencing, in interpretation and sign-out of the findings. The human genome is very much like reading hieroglyphics in the sense that seeing it is the easy part — we can all look at it — but understanding what it means is the complex part. Creating clinical relevance from that understanding is an even further challenge. That part takes a lot of time and effort and skill by laboratory-based molecular geneticists, molecular pathologists, and other specialists.
[33:33] So offering a sequence and then giving someone a very rudimentary filtering algorithm to play about with on their computer is not good in terms of quality interpretation. I get these reports sent to me all the time, almost every day. I call it the traffic light sign — whenever you see a test that qualifies your genetic variants in red, yellow, and green, I'm not saying the test is necessarily bad, but that is a bit of a red flag. Proceed with caution. You often see that traffic light sign in many of these less accurate reports.

[34:40] Dr. Linda Bluestein: I see that all the time. I just saw that yesterday, literally — somebody brought me their report. So that is really interesting. I feel like, as you said, we think that if it's being allowed, it must be regulated in some way and be high enough quality. And then if we get a report back that says we have certain genetic markers for things, especially if it's got the red light, green light, yellow light — it is hard to think it might not be accurate.

[35:21] Dr. Paldeep Atwal: Of course. So let me give you an example. This is not one isolated incident — I've seen this a number of times. So a patient thinks they have Ehlers-Danlos, they've researched online, looked at it all, and they decide to order a test by themselves. In good conscience they think this is the right test, this is a good test. And so they order the test and it comes back — they have a pathogenic variant in collagen 3, or COL3A1, for vascular EDS. Now they're freaking out, right? They're worried they have vascular EDS. They're worried about themselves, they're worried about their kids.
[36:19] So they eventually make an appointment to see me or another geneticist, and I'll do a proper history and clinical assessment. There are some facial features and other features of vascular EDS that they don't have. But what am I supposed to do? They have this report saying they carry this variant.
[36:43] So I look the variant up. There are actually two scenarios that can happen here. The first scenario is that the variant, by every other reputable clinical lab — you can look it up in ClinVar, the Clinical Variation database — all of the good clinical-grade labs will report their interpretation of that variant when they've seen it. And if all of them are saying it's benign and this consumer lab is saying it's disease-causing, I'm going to trust the clinical-grade labs. So I'll say that to the patient. Hopefully that reassures them. But they've also gone through a lot of psychological stress — a huge amount of stress, unnecessarily.
[37:34] The second scenario is that the variant that has been found is indeed reported as disease-causing in ClinVar and by other labs. So the variant is known to cause disease. In that case I'll say, let's confirm whether you truly have this variant by a clinical-grade test. I'll order a test from a lab that I feel comfortable standing behind those results, because currently I'm not comfortable standing behind the consumer report. And I'll specifically ask the lab — I'll send them the whole report too — this is the one variant you must look at. And they come back and say, no, that wasn't there. It was the wild-type — a normal variant.

[38:34] Dr. Linda Bluestein: And this has actually happened? Multiple times? Really? Wow. That's scary, because people are going through incredibly stressful things. Like you said, we have a shortage of geneticists, so some people can't get in to see a geneticist and therefore they're purchasing testing online. And they may make decisions about having a family based on that information.

[39:02] Dr. Paldeep Atwal: It's frightening, very frightening, actually. When it comes to things like family planning and recurrence, that's when you really start thinking. We have this thing in genetics — we don't just treat patients, we treat families. Really, that's what we do. The person is there ostensibly for themselves, but actually, when you really get into it and start talking about family history, they'll tell you, "I actually have 3 kids, and I'm starting with me, but I'm really more concerned about my kids than me." So they're actually more interested in their family than just themselves. And that's what we deal with all the time in genetics. That's the nature of genetics — we look like our parents, we look like our family. Why do we do that? That's genetics in a nutshell.

[39:55] Dr. Linda Bluestein: So if someone can't access a geneticist — I obviously evaluate people all the time for hypermobile EDS and related conditions, and I have a very different background as an anesthesiologist, and I'm much more focused on people's pain. But especially if I'm really concerned about vascular EDS, or if I find some other indicators, then I will order genetic testing from a clinical lab. I will do that for people when I feel it is appropriate. But what else do you think people can do if they can't get in to see a geneticist? Maybe they can't get in to see me or somebody like me who does do some amount of genetic testing, even though we're not a geneticist. What do you think that person should do?

[40:00] Dr. Paldeep Atwal: So there are a couple of options. Let's explore this now. First is the standard medical model where a patient goes to the doctor, the doctor decides what test to order and orders it on behalf of the patient. We're all used to that, we don't need an explanation of that.
Now there's also direct-to-consumer testing where you can go and buy a test yourself and have the results without physician involvement. We just talked about some of the pitfalls that can happen with that — not to say it's all negative, just to be clear.
[41:27] But there's a third option. There are companies out there where you can go, express an interest in a test — let's say a connective tissue gene panel — and they can either help facilitate with your own doctor or they can use one of their doctors to review the appropriateness based on some sort of questionnaire or other mechanism and say, yes, this is appropriate for you. You'll pay for the test, and then they also provide follow-up, whether with a genetic counselor, a physician, or someone with the knowledge to interpret the result.
[42:19] Because as you know very well, the results from genetic testing are challenging. I regularly have patients say, after a connective tissue gene panel comes back negative, "Well, so does this mean I don't have hypermobility?" And then I have to explain: we haven't found the gene for that subtype. We tested for all the other subtypes to make sure you don't have something that can cause vascular fragility, aneurysms, all of those things. We want to make sure you're not at risk for them. Then they understand — it's actually helping us reconfirm the diagnosis, because we looked at these other things that can be similar in presentation, that have overlapping phenotypes, but ultimately have significant, potentially life-threatening differences. That's why we did the test in the first place.
So just having that discussion and explanation of why you're doing the test is really necessary. It's not easy to navigate such a complex world of genetic testing, even for a top physician like you. It's hard. So for the patient to do it themselves, it's very challenging.

[43:42] Dr. Linda Bluestein: Yeah, it's very, very challenging. And there are specific panels, like you said, and then we can do whole exome sequencing, we can do whole genome sequencing. I've ordered whole exome sequencing before for a couple of patients — actually, one of them is a patient we have in common. And shockingly, it came back with one word: negative. This person — yeah, I was absolutely shocked. So I guess in terms of what panel to order — and when it comes back with that one word "negative" but people do have variants of uncertain significance, do they report it that way? Because if none of the variants of uncertain significance are clearly pathogenic, what are your thoughts on that?

[44:27] Dr. Paldeep Atwal: Right. The labs all have a slightly different structure in how they report things. In general, some will say "negative" when they don't find any pathogenic variants and any variants of uncertain significance, some will say "negative" only if they don't find pathogenic variants — but if they find variants of uncertain significance, they won't say negative, they'll say "intermediate" or "unclear" or something like that. And they'll qualify that with different categories of result.
[45:02] There's a whole counseling framework around genetic results, going through different categories: pathogenic variants related to their phenotype, pathogenic variants unrelated to their phenotype. So you're testing someone with EDS and then you find they have a cardiomyopathy gene variant that isn't related. Then there are unclear variants, related and unrelated to their phenotype.
[45:23] And then often in these tests, there are American College of Medical Genetics reportable variants — the BRCA genes, and others — where if you're looking at someone's exome or genome and you find a disease-causing variant in one of them, you should let the patient know, or offer to. Usually the patient has to consent to that before you do the test. And now some labs are also offering pharmacogenetic variants in addition. So now you've got all these pharmacogenetic variants, and the exome and the genome become this huge complex thing — even if it's mostly or all negatives, it's hugely complex and takes a long time to counsel through. How do you scale that discussion to ensure the patient understands what happened, and that you have enough time to do that for all patients getting genome sequencing? Because it's only increasing. As cost goes down and availability and access goes up — which are all good things — how do you ensure we have the ability to manage that? I don't really know the answer.

[46:39] Dr. Linda Bluestein: And when it comes to variants of uncertain significance, if we were to do whole genome sequencing on — pick any number of people — they're all going to have some variants of uncertain significance, right?

[46:46] Dr. Paldeep Atwal: Of course. We all have unique variants unique to us, otherwise we wouldn't be unique, right? So the question, as for the lab when they see these — some of them are easy. They're synonymous and they don't cause protein sequence changes or aren't near splice sites. So that's easy for the lab to interpret and safely call as non-disease-causing. Some of them are challenging.
[47:22] One thing that's really helped is having larger and larger databases. I remember when there was the 1000 Genomes Project, and then the UK came out with UK10K — that was a catchy phrase. Then there was the 100,000 Genomes Project. Now I don't even know how many there are in ExAC and gnomAD. We've got huge amounts of data, and it's only increasing. And more than that, we also have a more diverse database with people of more diverse ethnic backgrounds, which actually helps everyone. It tells you what's related, what's common, what's uncommon.
[48:17] All of that put together is helping our interpretation of genetic variants across all populations. You'll see very commonly that when you order a test, a year or two after you'll get an amended report from the lab saying they now feel comfortable this variant is not disease-causing — they've downgraded it from variant of uncertain significance to benign.
[48:44] One more point on variant classification. There are currently 5 levels: pathogenic, likely pathogenic, variant of uncertain significance, likely benign, benign. That may change in the future to reflect the complexities of variant interpretation — there might even be more levels. So, you know, what does this gene change mean? It's a very challenging area of genetics.

[49:15] Dr. Linda Bluestein: Yeah. And speaking of challenging and gene changes, you're probably going to anticipate this question. The TNXB gene variants that were found by the Norris Lab — are you able to share your thoughts on that?

[49:28] Dr. Paldeep Atwal: I think the fact that a knowledgeable, highly intelligent group like the Norris Lab is looking at this is great. We need more. We need 5 Norris Labs. And so I think that's fantastic. I think some of the evidence is very strong and this has a role. I would like to wait until it goes through the formal process: once you have a research finding, you put it together, you publish it, it goes through a peer review process where it's looked at by other experts in that area, it's published, and then confirmed by others as well. I'm tentatively positive, I'll say that. I'd like to wait until it goes through the process, but I'm tentatively quite positive about it, and I think it's a major step forward.
[50:38] Let's just be hypothetical for a second. If everything goes through and we can offer testing for that, I think that will be a positive step in the community. Now, obviously we're not going to suddenly find the gene change for every single patient with hypermobile EDS based on one study and one finding. There are likely many reasons that we haven't been able to find the genetic basis of hypermobile EDS. We talked about different subgroups within EDS, the complexities of genetics, variable expression, gene-gene interaction — we're not able to test for a lot of those things. So if we're not able to test for them and that's the cause, perhaps we're missing it.
[51:31] Additionally, the genetic sequencing technology we're mostly using is short-read technology, which is changing. But there are things that are difficult to capture with that technology; you require different sequencing technology — long-read — to see complex rearrangements, triplet repeat disorders, and other deletions of certain sizes that can be missed on short-read sequencing and not seen on chromosomal-level copy number variation. A deletion of around 1,000 base pairs, for example — it's very challenging to detect those things. So there are gaps, and we need to start filling those gaps. Maybe it's in one of those gaps that the answer lies. I don't know.

[52:13] Dr. Linda Bluestein: And given all of these complexities and the challenges that we face, but also with these vast panels and the dramatically reduced price of whole exome or whole genome sequencing — if someone has red flags for a rare type of EDS, or red flags for vascular fragility or something, then I would think that genetic testing would for sure be indicated in those instances. But kind of as a general rule, what do you think are good indicators for genetic testing, or good indicators that it's maybe not needed, in the context of somebody presenting with symptomatic joint hypermobility who suspects they have hypermobile EDS?

[53:04] Dr. Paldeep Atwal: So that's a very relevant question from my perspective. The threshold is fairly low given the ease with which gene panel testing is now available. So if someone presents with hypermobile EDS, we go through other features and factors related to Ehlers-Danlos syndrome and other related conditions — Marfan, other things like that. Really, if there's even a small concern at this stage, I usually order testing for them.
[53:37] I think the other problem I've seen in clinical practice is when someone doesn't have the testing. Let's say they have hypermobile EDS, you're very sure it's hypermobile EDS, and you think you don't need the testing — you're very confident they don't have anything else. That's fine. But this has happened: they'll come back later and say, "This cardiologist I saw, or the surgeon I saw — I need surgery, and they won't do the procedure because I was never tested for vascular EDS or something else. They're not comfortable doing their procedure or surgery unless that's been ruled out by a genetic test." So it creates barriers for the patient down the road, unknowingly. I'm not saying those barriers are correct, just to be clear, but they do exist.
[54:44] So with that experience and that awareness of what will happen to the patient — because patients often see me once or twice and I may or may not see them again — I want to make sure that in that interaction I've done everything possible for them. So 5 years down the line, it's still useful to them.
[55:11] And secondly, there's a big worry and anxiety component when it comes to testing, particularly for subtypes that cause vascular issues and aneurysms. Clinical reassurance by a medical geneticist is very valuable, but seeing that negative result has a personal utility for patients. And subsequently, I think that has a clinical utility.

[55:37] Dr. Linda Bluestein: Yeah, and I'm thinking of somebody that we also have in common whom you diagnosed with classical-like EDS type 2. And I know this is so incredibly rare that you might even know who this is offhand. Yeah, this is the AEBP1-related type 2 classical-like EDS, right? But that person — hypermobile EDS was what was suspected. I saw them after they had their diagnosis from you. This is somebody that you did do genetic testing on, hypermobile EDS was suspected, and it came back that they actually had this extremely rare type. So that's a great example that—

[56:12] Dr. Paldeep Atwal: It's a great example and a great example of why we do the testing. These are called subtypes of Ehlers-Danlos for a reason — subtypes of overlapping, similar-presenting conditions. Now, as good as we are clinically, we must know our limitations: sometimes it's really impossible to clinically differentiate. And so we have to rely on additional molecular and genetic testing to ensure we're doing what's best for the patient.

[56:42] Dr. Linda Bluestein: And in that classical-like type 2, if I remember correctly, there are different implications in terms of screening for vascular issues, right?

[56:48] Dr. Paldeep Atwal: Vascular issues, bone density issues, other screening issues as well — other medically relevant things that make it important to differentiate, to your point.

[57:01] Dr. Linda Bluestein: Right, right. Yeah. So that's not just an interesting thing — that's a—

[57:05] Dr. Paldeep Atwal: No. It's clinically useful. A meaningful difference that's going to benefit the patient and their health long-term.

[57:14] Dr. Linda Bluestein: Yeah, yeah, wow, definitely. And then you mentioned briefly about genomics and pharmacogenomic testing. Before we wrap up, if you could maybe tell us — and I'm sure we could have spent the whole time talking about genetics versus genomics — we touched on this a bit 4 years and 4 months ago, but if you could just tell us a little bit about where things are at with genomics, what the difference is, and why some people might be considering that, because there's also direct-to-consumer testing for that as well.

[57:46] Dr. Paldeep Atwal: Well, let me address the genetics versus genomics piece and then the pharmacogenetics or pharmacogenomics piece.
[57:59] I've heard a lot of people talk about genetics versus genomics. From my perspective as a geneticist, I don't see a difference. Some people will tell you genetics is looking at a single gene and genomics is looking at all of the genes in a holistic view of many genes interacting together. But that's not really the case with medical genetics — I think they're really interchangeable. I don't know any geneticist that looks at single genes anymore. The difference may just be reflective of changing times. When we were only able to look at one gene at a time, it was genetics. Now we're able to look at thousands of genes at the same time. Our college — the American College of Medical Genetics and Genomics — uses both terms, and I think they're just playing the field and calling it both. I'm a fellow of that college, and I call myself a medical or clinical geneticist. Maybe I should say clinical geneticist and genomicist. I don't know — what do you think?

[59:26] Dr. Linda Bluestein: I don't know — that's an excellent question.

[59:33] Dr. Paldeep Atwal: Yeah. So the second thing, on pharmacogenomics — looking at how your genetics influences your body's metabolism of certain drugs and medications. Most of these are CYP enzymes in the liver that either upregulate or downregulate your metabolism of those drugs, which — and this is also complicated — depending on whether a drug is given as a prodrug or in its active form, can cause higher or lower drug levels.
[1:00:20] It's used in a variety of different areas of medicine. Psychiatry, believe it or not, is probably the most commonly used area, which is kind of unusual — you don't think of genetics and psychiatry as being at the forefront of a certain area of genetic testing, but it is. And other areas as well, obviously even in anesthesiology — as you know — with the risk of malignant hyperthermia. There are a couple of genes, RYR1 for example, and that is on all of these panels.
[1:00:52] So the labs are now offering these pharmacogenetic variant interpretations alongside clinical variants — providing more complete coverage of testing, not just clinical variants but also pharmacogenetic variants. One of the issues is finding someone to interpret that. What we do is we have a pharmacist who's trained in pharmacogenomics and sits with the patient, makes a medication action plan based on that, and doesn't just look at the genetics in isolation. It also looks at drug-drug interactions — you can think of these as gene-drug interactions alongside drug-drug interactions — and there's also phenoconversion: being on drug A can actually influence how well or how slowly drug B is being metabolized. So that's all relevant in the soup, as you referred to earlier. It all plays a role together. You can't take them in isolation. Looking at all of them together is the right way to approach it.

[1:02:02] Dr. Linda Bluestein: Yeah, that's really fascinating. And I know that in psychiatry this was something that was done many years ago, and it does seem like it's becoming more common. Do we have the same problem with not just difficulty in interpreting the information, but possible inaccurate results if people are doing direct-to-consumer testing versus going to a clinical lab?

[1:02:29] Dr. Paldeep Atwal: I think inaccuracy exists in all aspects of testing if the right processes aren't followed. I always look for labs to have CLIA and CAP certification through all aspects of their testing, not simply the sequencing part. These regulators will regularly submit a sample to the lab and ask them what's in the sample to ensure that they're able to detect the right things — always being challenged to show that they can detect the right variants when they're there, or confirm they're absent when they're not. So yes, the short answer is the risk remains even in pharmacogenetic testing.
[1:03:24] Most pharmacogenetic testing looks at certain polymorphisms of the gene, so the testing is often a little bit simpler. But that being said, there are now groups looking at doing full sequencing of those genes to look for other variants not even on those panels that could be missed. So if you have a test that only looks for the presence of a specific variant and you don't find that variant, that doesn't mean there isn't something else that you didn't look for that's also relevant in terms of drug metabolism.

[1:03:59] Dr. Linda Bluestein: And there are other things besides drug metabolism that we can look at for genomics, right?

[1:04:05] Dr. Paldeep Atwal: Oh, absolutely. There are complex traits now. Polygenic risk scores for cardiovascular disease, diabetes risk, cholesterol, all sorts of conditions, neurodegenerative conditions as well. These are still nascent in many ways, but I think we'll see a lot more of that in the next 3 to 5 years. Maybe at our next conversation we can talk more about that.

[1:04:37] Dr. Linda Bluestein: Yeah, I can't wait to see what that conversation's going to be like. Okay. So I like to wrap up every episode with a hypermobility hack. What hack do you have for us? A quick tip or quick win for people?

[1:04:52] Dr. Paldeep Atwal: Hypermobility hack. So this has to be a physical thing?

[1:04:57] Dr. Linda Bluestein: It could be anything. So basically you've already given us quite a few — like looking for a CLIA-certified lab, for example, and looking for labs where it's not just direct-to-consumer testing but direct access for ordering. Something that people can do, a tip that they can take away right away.

[1:05:21] Dr. Paldeep Atwal: So let me give a couple. The first, very simple: avoid things that make your joints worse — impact exercises and maximal stretching. If people like to stretch, and many people like to stretch because it makes them feel better, try and limit it to around 80% of what you can do.
[1:05:41] And then the other one — I think maybe the most important — is to dedicate some time, at least for a short while, to working with a knowledgeable physical therapist trained in Ehlers-Danlos Syndrome. Ideally, that's someone who often has the condition themselves — if they have Ehlers-Danlos themselves, that's a great sign. Working with them on how certain movements are being done, reprogramming certain things, looking at how certain muscles are activating and how they're approaching certain joints and movements — it's amazing how inefficient some people are in moving, which makes them more tired and in more pain. Just reprogramming how they're doing simple movements they're doing dozens and dozens of times a day makes a huge difference.

[1:06:33] Dr. Linda Bluestein: That makes a lot of sense. Okay. So before we really wrap up, I first of all want to thank you so much for coming on the Bendy Bodies Podcast again and sharing your vast wisdom and knowledge with us. This is such an important topic, and I feel like it's so much more nuanced than most people realize. So thank you so very much.

[1:06:56] Dr. Paldeep Atwal: Thanks. It's always a pleasure. Let's plan, maybe a maximum of 2 years, if not 1 year, next time.

[1:07:04] Dr. Linda Bluestein: Yeah, yeah, definitely. And before you go, can you let us know where to find you, and also if you have any special projects or research that we should be aware of?

[1:07:17] Dr. Paldeep Atwal: Yeah, so I'm based in West Palm Beach, Florida. In terms of clinical practice, I see patients in the office in downtown West Palm Beach, and also online as well — many patients choose to see me online due to travel limitations. I have a couple of cases that we're publishing on unusual presentations and types of EDS, and some new variants in those genes that I've got some students working on.
[1:07:52] And I'm also doing more education. I'm planning a one-day conference for patients — patient-focused — where I bring in some experts, I'll talk, and we really discuss different aspects of Ehlers-Danlos. And maybe I can recruit you right now. I think, as you said, there are many patients who just don't have access to geneticists, don't have access to people like you and me, and it's something they could watch from their house — similar to the education you're doing here. They can really take away things from experts in the area that may be of use to them, or at least help them advocate for themselves. I'm hoping to do that within the next year or so.

[1:08:41] Dr. Linda Bluestein: That sounds fantastic, and I would love to be involved. I think that's a wonderful idea — like a boot camp.

[1:08:51] Dr. Paldeep Atwal: Yes, yeah, just a one-day thing.

[1:08:55] Dr. Linda Bluestein: I think so many people could benefit from that. These conditions require people from so many different disciplines, and would benefit from people with different backgrounds. I love that idea. That's really, really great.

[1:09:11] Dr. Paldeep Atwal: Yeah. Cardiologist, GI motility specialist, physical therapy, of course, physiatry — medically trained doctors who specialize in physical medicine — pain management, of course, that's huge. Neurosurgery — all of these. Actually, I'm going to have too many already.

[1:09:33] Dr. Linda Bluestein: I was just going to say, this is sounding more like a multi-day event. But that's okay — everything's an evolution, right? So you and I have been communicating for a number of years, even before we had our conversation 4 and a half or so years ago. Your practice has evolved, my practice has evolved. We're constantly trying to meet the needs of our patients as best we can and be able to offer them the kind of support that we want them to have.

[1:10:01] Dr. Paldeep Atwal: Absolutely. Yes, that's the goal.

[1:10:04] Dr. Linda Bluestein: Well, thank you so much again. It was so great to see you, and I look forward to our next conversation.

[1:10:11] Dr. Paldeep Atwal: Likewise, it was a real pleasure. Thank you, Linda.

[1:10:18] Dr. Linda Bluestein: Oh my gosh, that was such a great conversation with Dr. Atwal, and I'm so grateful to him for coming on the podcast a second time. We will definitely have him back much sooner than 4 years and 4 months from now and get a follow-up conversation about genetics. Because as I'm sure you are now very well aware, it's so much more complicated and so much more nuanced than most of us realize. I know I learned a lot from this conversation, so I hope you did as well.
[1:10:47] Thank you so much for listening to this week's episode of the Bendy Bodies with the Hypermobility MD podcast. You can help us spread the word about joint hypermobility and related disorders by leaving a review and sharing the podcast. This helps raise awareness about these very complex conditions. If you would like to meet with me one-on-one, check out the available options on the services page of my website, hypermobilitymd.com. You can also find me, Dr. Linda Bluestein, on Instagram, Facebook, TikTok, Twitter, or LinkedIn at Hypermobility MD. You can find Human Content, my producing team, @humancontentpods on TikTok and Instagram. You can find full video episodes up every week on YouTube at Bendy Bodies Podcast. To learn about the Bendy Bodies Program disclaimer and ethics policy, submission verification and licensing terms, and HIPAA release terms, or to reach out with any questions, please visit bendybodiespodcast.com. Bendy Bodies Podcast is a Human Content production. Thank you for being a part of the community, and we'll catch you next time on the Bendy Bodies Podcast.