EDS syndromes and Variants of Unknown Significance
Disclaimer – while Drs Payne and Lake have clinical experience in genetics, neither of us are fellowship trained Clinical Geneticists.
One of the toughest things with genetic testing, especially the much broader whole exome/genome sequencing, is what to do with the Variants of Unknown Significance (VUS). With the relative increase in availability of testing and limited access to expertise (some geneticists declining EDS consults completely) we have a lot of people stuck in this in between space. I have all the symptoms, but I don't have an established genetic variation that explains this … except for a VUS.
This isn’t rare either, one study of over 1.6 million people found that at least one VUS was found in 41% of folks being tested, though about ¼ of that group also had another pathogenic mutation found leaving 31.7% of the group with nothing but a VUS (1). During this same study, which spanned about 8 years, they found that only 7.3% were reclassified to benign or pathogenic, with about 4 out of 5 being reclassified as benign. Now, this was a large database study that lumped together everyone getting focused testing of a small panel of genes with those getting whole genome testing. As you might guess, the more genes we test, the more VUSs we initially find (2), which makes sense in light of there being more than 10 billion genetic variants consistent with life. However, studies that look at how many people are left with a VUS after clinically assessing the findings show a lower rate of VUSs (22-23%), meaning sometimes we can clinically determine that a diagnosis doesn’t fit and for the patient in front of us, call it likely benign. For example, an iron storage gene might be related to hemochromatosis, but screening for this disease is fairly simple and a VUS can be quickly classified as benign or pathogenic for an individual.
Determining what a VUS means is a matter of several factors. First, if we see the same genetic difference in a certain number of healthy people without a specific disease, then it can safely be assumed to be unrelated. For that you need critical mass for this to avoid mistaking variable penetrance for normal function. Variable penetrance is when a specific mutation sometimes causes disease, the classic example being some CHEK2 mutations that are maybe better understood as risk factors than classical “cancer genes” (3).
To have a good comparison group, we need a larger population who has this same mutation. Because mutations are often inherited, it makes sense that you would see similar mutations in people who are at least loosely related. Databases have more data from people who have been able to both access and accept testing (4). In the large database study mentioned above, 57.7% were white and 71.2% female. This means an American Indian male (<0.2% of the dataset)* would have a much lower likelihood of his VUS being reclassified (1). To call a VUS even likely benign, you would need at least 10 others with the same genetic finding and no disease, but this number rapidly increases if you are looking at recessive genes, incomplete penetrance, polygenetic risk scores, accounting for mosaicism, and other factors beyond the scope of this article (4).
Second, we now are more commonly doing duo and trio testing with exome/genome testing. This means taking a sample from one or both parents respectively and if the same VUS is present in a parent without the disease, then it is much less likely causing the clinical symptoms.
So what are the EDS relevant efforts to determine the meaning of VUSs? Human genetics is a gigantic field and EDS and other collagen/connective tissue diseases are only a slice of it. Since the formalization of the 2017 hEDS criteria, we now have several studies looking for candidate genes, the largest being the HEDGE study (5). The study has provided some new genes, but these have only explained a small fraction of what we see(e.g. 2.5% of the cohort found to have TXNB mutations) and many have been spun off into a newer monogenetic groups separate from hEDS (e.g. AEBP1 -> clEDS2). So we have not yet determined if there is a set of genes that can be blamed for causing hEDS/HSD, and if we can't find the answer when looking at over a thousand individuals, we aren't going to find a simple answer.
A separate and promising approach is to look from the other direction. That is, we look at the collagen itself and work backwards from there. Sometimes called the “Italian Group” a group of researchers the Lombardy region (Milan and Brescia) had reported some findings that oddly were in 100% of patients with hEDS and HSD and 0% in other cohorts (6). From what I have heard from other researchers, they have not found the same serum protein to be found in different levels that would help diagnose hEDS/HSD. Work continues, but we don't have answers at this time.
So what do we do when we have a VUS? While tools exist like REVEL, AlphaMissense and FoldX, these don’t help map a gene to what we see clinically, they help determine if a genetic change would impact the function of the relevant protein. This isn't always helpful when the VUS is in a promotor region or somewhere not included in the final protein. Ultimately, we often are left with a clinical picture and a possible causative mutation. This is where we must rely on clinical experience and judgement to determine if we are going to say someone falls in a diagnostic bucket or not.
Personally, I find it helpful to collaboratively walk through why we want a balanced approach to whether we call someone’s HSD/hEDS genetically supported or not. This is kind of like when we do procedures - we go through the chance of harm/benefit, and alternative approaches.
First there are risks of establishing a diagnosis when it doesn't fit (AKA Type 1 diagnostic error). This is a mental error for the clinician called “premature epistemic closure,” which is maybe a fancier way of saying “jumping to conclusions too early.” We might attribute all of someone's problems to this diagnosis and stop looking for answers, including something else that might explain a large piece of the symptoms. Honestly, even when people have monogenetic EDS or severe hEDS, people have other unrelated diseases come up too. Even in hEDS/HSD, genetic testing can reveal separate explanatory diseases in over 25% of people (7).
If you have your symptoms attributed to the wrong bucket (i.e. diagnosis), it might mean you are in a bucket with less problems (more benign) and key testing is missed, or with more problems and excess testing. With each test we run the risk of overdiagnosis. We might ask someone with benign thyroid nodules or an adrenal “incidentaloma” what it felt like to live with a mass that "probably isn't cancer, but we're going to just watch it." Unfortunately, many folks have had a larger surgery to remove something that didn't need to be removed, and in the case of thyroid surgery, this means a lifetime of taking thyroid replacement medication. This goes beyond the time and hassle of extra testing but the extra anxiety from having an unknown, not completely sure it is benign, mass in their body (8, 9, 10). Additionally, when given a genetic diagnosis at a younger age, a lot of people reconsider having children (16), or when they decide to, there is significant guilt, fear, and expense – as some couples do preimplantation genetic testing (15).
Along with overdiagnosis, we have overtreatment. Balancing real side effects with small improvements isn’t an easy task, and if the treatments doesn’t improve your long term health, it might actually cause harm. Harm from overtreatment is well documented in situations like antibiotics for viral infections (up to 25% of people have side effects) (11) or aggressive treatment for low risk prostate cancer (12). To take an EDS related example, if we say a VUS is related to osteogenesis imperfecta overlap (OI), someone might be started on daily injections of Forteo. This medication will make the bones look more dense, but that is not the issue in EDS/OI, so you don’t have the benefit. In younger folks where the growth plates are still open, there is a significant risk of cancer with this treatment, thus causing harm (13). Additionally, the most common side effects are nausea (8-14%), dizziness (3-9%), and whole body pain (21%) – not a great list of side effects for a community of people who often have MCAS, gastroparesis, POTS/dysautonomia, and diffuse pain from hEDS (28).
A more nebulous concern is that many times other specialists feel unprepared to know the clinical implications of a rare genetic diagnosis and will pass the responsibility to an overwhelmed PCP, something I personally know all too well from my past work. Alternatively, that specialist might recommend several new consults, all from additional physicians who themselves are not experts in this field. This compounds the medical trauma many people experience with EDS, even after diagnosis.
The other side of the coin is the risk of NOT calling the diagnosis (AKA type 2 diagnostic error). After all, the concept of a diagnosis is that we have a critical mass of similar people, and we can predict what we need to do now to make things better in the future. Sometimes this is by changing habits, adding medication, or by catching problems earlier in their course. BCRA (1 and 2) are classic genetic examples of this, but you can apply this logic to high blood pressure or clinical depression as well. So, if we were to make up a diagnosis (Lake Syndrome for example) and say that I have it, that wouldn’t be helpful. We don’t know what that means in terms of treatment or future prognosis. These labels do have meaning, and when we say that someone has a diagnosis, we are saying that someone has the signs, symptoms, and clinical findings similar enough to another defined group, so much so that we can predict what treatments will help or what problems lie ahead.
First, by inaccurately determining that someone does not fall into a diagnostic category, you may miss the screening tests that can find other problems early. Not just in vascular and bone diseases, but also obstetrical considerations that are critical for folks considering having children in the future. This is before we mention pulmonary testing, ophthalmological screening, audiology, dental, and cardiac screening that might be indicated based on phenotype (your unique signs and symptoms).
Second, some types of EDS have different treatment thresholds. Common treatments like physical therapy are not ones that require exotic diagnoses to get access via a health service (e.g. UK/EU) or insurance (e.g. USA). However, a clinician validating a diagnosis sometimes can mean access to a treatment or different precautions taken during a procedure or surgery. One of the most important times this comes up is when we are considering the types of EDS where we know that the aorta can stretch faster and sooner (aneurysm) and tear open (rupture) at a smaller size. Surgery on this part of the body is always risky, so surgeons have to make a very serious predictive decision about the best time to do the surgery. Ideally this is when the risk of the surgery is lower than the risk of rupture. Fortunately, hEDS does not have a higher risk of aortic enlargement or rupture than baseline populations (17). However, the kyphoscoliotic, periodontal, dermatosparaxis subtypes of EDS do have higher risks and sometimes classical EDS depending more on the location of the causative mutation (C vs N terminal)(19). Surgeons might move to operate sooner, at a smaller aortic diameter to reduce the risk of rupture or to change the approach (preferring open over endovascular) when there is a VUS used to clinically diagnose a high risk form of EDS (20 – table 5 esp). By not providing a diagnostic label, we may run the risk of failing to operate soon enough, but if we diagnose incorrectly, we run the risk of operating too soon (or at all) and in a way that might cause more problems.
The last reason is often the most salient – person sanity. In hEDS, and often in monogenetic EDS, people find that having a diagnosis confirmed means that the problems aren't "all in my head". This often feels incredibly validating, especially when a lot of the symptoms like lightheadedness and frequently changing joint pain are invisible to others. By the time most people get a diagnosis, there have often been profound social changes from their decreased level of function (21, 22). While the lack of validation is very hard in (h)EDS, this happens in many, if not all, chronic illnesses and even in well defined temporary situations with lots of information and support, like a cancer diagnosis (23). Even if it is only having a clinician say "what you are feeling is real but we don't know why" feels different from "what you are feeling is real and here is why we would expect you to feel that way" (14).
Many physicians, myself included, were trained to think linearly from symptoms (+/- testing) -> diagnosis -> treatment. When a diagnosis doesn’t have a clearly defined treatment or cure, it is frequently felt to be less important to the clinician. Hypermobility syndromes (and high functioning autism) are often seen in this light, an untreatable incurable illness where the label doesn’t change management. I would argue that any clinician who has worked with people through these diagnoses has seen that there is a lot of value to knowing what is going on (24, 25). While simply knowing the diagnosis might not always improve symptoms, people regularly report feeling more in control, better able to manage their own symptoms and long term care (26). Yet even the receiving of the diagnosis itself can provoke anxiety or depression and about 1/3 of people have more concerns about insurance and employment discrimination (26).
In the end, the clinical decision to diagnose someone with a genetically defined type of EDS based on a VUS is a tricky decision with a lot of uncertainty. As is true in my general approach to medicine, the best approach is to go into this in a collaborative way. I often find that online discussions in EDS communities frequently concerns about the way VUSs are handled and clinicians refusing or declining to provide a diagnosis. Sometimes continued uncertainty is the most honest answer we can provide, however, the decision to proceed with a tentative diagnosis is an option that can strike a balance, especially if there is a shared understanding of what this means it can give some direction on how to go forward.
In my next post I will be writing about some of these downstream effects of receiving a diagnosis, especially from a mental health point of view. If you have requests for additional topics, please use the form on our site. If you read this and want to work with us at Kaleidoscope Health Collective, we’d be happy to talk! The next step is completing the contact form and the process can unfold from there.
References
1) Chen E, Facio FM, Aradhya KW, et al. Rates and Classification of Variants of Uncertain Significance in Hereditary Disease Genetic Testing. JAMA Netw Open. 2023;6(10):e2339571. doi:10.1001/jamanetworkopen.2023.39571
2) Shirts BH, Pritchard CC, Walsh T. Family-Specific Variants and the Limits of Human Genetics. Trends Mol Med. 2016 Nov;22(11):925-934. doi: 10.1016/j.molmed.2016.09.007. Epub 2016 Oct 11. PMID: 27742414; PMCID: PMC6197474.
3) Burke W, Parens E, Chung WK, Berger SM, Appelbaum PS. The Challenge of Genetic Variants of Uncertain Clinical Significance : A Narrative Review. Ann Intern Med. 2022 Jul;175(7):994-1000. doi: 10.7326/M21-4109. Epub 2022 Apr 19. PMID: 35436152; PMCID: PMC10555957.
4) Rehm HL, et al; Medical Genome Initiative Steering Committee. The landscape of reported VUS in multi-gene panel and genomic testing: Time for a change. Genet Med. 2023 Dec;25(12):100947. doi: 10.1016/j.gim.2023.100947. Epub 2023 Jul 30. PMID: 37534744; PMCID: PMC10825061.
5) Scicluna K, Formosa MM, Farrugia R, Borg I. Hypermobile Ehlers-Danlos syndrome: A review and a critical appraisal of published genetic research to date. Clin Genet. 2022 Jan;101(1):20-31. doi: 10.1111/cge.14026. Epub 2021 Jul 14. PMID: 34219226.
6) Ritelli, M., Chiarelli, N., Cinquina, V., Bertini, V., Piantoni, S., Caproli, A., Della Pina, S.E.L., Franceschini, F., Zarattini, G., Gandy, W., Venturini, M., Zoppi, N. and Colombi, M. (2025), Bridging the Diagnostic Gap for Hypermobile Ehlers-Danlos Syndrome and Hypermobility Spectrum Disorders: Evidence of a Common Extracellular Matrix Fragmentation Pattern in Patient Plasma as a Potential Biomarker. Am J Med Genet, 197: e63857.
7) Forghani I, See J, McGonigle WC. Hypermobile Ehlers-Danlos Syndrome: Diagnostic Challenges and the Role of Genetic Testing. Genes (Basel). 2025 Apr 29;16(5):530. doi: 10.3390/genes16050530. PMID: 40428350; PMCID: PMC12111038.
8) Tikkinen KAO, Halme ALE, Guyatt GH, Glasziou P. The Impact of Definitions of Disease on Overdiagnosis. JAMA Intern Med. 2025;185(8):909–910. doi:10.1001/jamainternmed.2025.1727
9) Del Chiaro M, Torphy RJ, Schulick RD (University of Colorado Anschutz Medical Campus, Aurora, Colorado, USA). Pancreatic incidentalomas: Investigation and management (Review) J Intern Med 2021; 290: 969–979.
10) Ayappa S, Joffe S, Laberge AM, Nye RT, Hess M, Callahan KP. How variants of uncertain significance affect clinical decisions: A systematic review. Genet Med. 2026 Jul;28(7):102593. doi: 10.1016/j.gim.2026.102593. Epub 2026 May 12. PMID: 42117370; PMCID: PMC13192372.
11) Shehab N, Patel PR, Srinivasan A, Budnitz DS. Emergency department visits for antibiotic-associated adverse events. Clin Infect Dis. 2008 Sep 15;47(6):735-43. doi: 10.1086/591126. PMID: 18694344.
12) Donovan JL, et al. Patient-Reported Outcomes 12 Years after Localized Prostate Cancer Treatment. NEJM Evid. 2023 Apr;2(4):EVIDoa2300018. doi: 10.1056/EVIDoa2300018. Epub 2023 Mar 11.
13) Ward, L.M., Rauch, F. Anabolic Therapy for the Treatment of Osteoporosis in Childhood. Curr Osteoporos Rep 16, 269–276 (2018). https://doi.org/10.1007/s11914-018-0434-z
14) Lopes MT, Koch VH, Sarrubbi-Junior V, Gallo PR, Carneiro-Sampaio M. Difficulties in the diagnosis and treatment of rare diseases according to the perceptions of patients, relatives and health care professionals. Clinics (Sao Paulo). 2018;73:e68. doi: 10.6061/clinics/2018/e68. Epub 2018 Apr 5. PMID: 29641803; PMCID: PMC5866403.
15) Severijns Y, de Die-Smulders CEM, Gültzow T, de Vries H, van Osch LADM. Hereditary diseases and child wish: exploring motives, considerations, and the (joint) decision-making process of genetically at-risk couples. J Community Genet. 2021 Jul;12(3):325-335. doi: 10.1007/s12687-021-00510-x. Epub 2021 Feb 20. PMID: 33611773; PMCID: PMC8241960.
16) Kowal K, Domaradzki J. "To have children or not?" Between desire, responsibility, luck, and guilt: reproductive decision-making in individuals with neurofibromatosis type 1. Orphanet J Rare Dis. 2025 Oct 21;20(1):531. doi: 10.1186/s13023-025-04061-z. PMID: 41121384; PMCID: PMC12542219.
17) Lahey H, Shin H, Myers K, McBride KL. Longitudinal echocardiography in pediatric patients with hypermobile Ehlers-Danlos syndrome. Am J Med Genet A. 2024 Dec;194(12):e63844. doi: 10.1002/ajmg.a.63844. Epub 2024 Aug 16. PMID: 39148461.
18) Asher SB, Chen R, Kallish S. Mitral valve prolapse and aortic root dilation in adults with hypermobile Ehlers–Danlos syndrome and related disorders. Am J Med Genet Part A. 2018; 176A: 1838–1844.
19) Bowen JM, Sobey GJ, Burrows NP, Colombi M, Lavallee ME, Malfait F, Francomano CA. 2017. Ehlers–Danlos syndrome, classical type. Am J Med Genet Part C Semin Med Genet 175C: 27–39.
20) Morris SA, Flyer JN, Yetman AT, Quezada E, Cappella ES, Dietz HC, Milewicz DM, Ouzounian M, Rigelsky CM, Tierney S, Lacro RV; American Heart Association Council on Lifelong Congenital Heart Disease and Heart Health in the Young (Young Hearts); Council on Cardiovascular and Stroke Nursing; Council on Peripheral Vascular Disease; Council on Cardiopulmonary, Critical Care, Perioperative and Resuscitation; and Council on Cardiovascular Surgery and Anesthesia. Cardiovascular Management of Aortopathy in Children: A Scientific Statement From the American Heart Association. Circulation. 2024 Sep 10;150(11):e228-e254. doi: 10.1161/CIR.0000000000001265. Epub 2024 Aug 12. PMID: 39129620.
21) Halverson, C. M. E., Clayton, E. W., Garcia Sierra, A., & Francomano, C. (2021). Patients with Ehlers–Danlos syndrome on the diagnostic odyssey: Rethinking complexity and difficulty as a hero's journey. American Journal of Medical Genetics Part C: Seminars in Medical Genetics, 187C: 416–424.
22) Bennett SE, Walsh N, Moss T, Palmer S. Understanding the psychosocial impact of joint hypermobility syndrome and Ehlers-Danlos syndrome hypermobility type: a qualitative interview study. Disabil Rehabil. 2021 Mar;43(6):795-804. doi: 10.1080/09638288.2019.1641848. Epub 2019 Jul 18. PMID: 31318301.
23) Vaughan, D. (2021, October 28). Love lost: The effects of cancer on marriage and relationships. CURE. https://www.curetoday.com/view/love-lost-the-effects-of-cancer-on-marriage-and-relationships.
24) Estrella E, Frazier PA. Healthcare experiences among adults with hypermobile Ehlers-Danlos syndrome and hypermobility spectrum disorder in the United States. Disabil Rehabil. 2024 Feb;46(4):731-740. doi: 10.1080/09638288.2023.2176554. Epub 2023 Feb 11. PMID: 36772820.
25) Guedry SE, Langley BO, Schaefer K, Hanes DA. Patient experience of chronic illness care and complementary integrative health use: a cross-sectional study of patients with hypermobility spectrum disorders (HSD) and Ehlers-Danlos syndromes (EDS). Disabil Rehabil. 2023 Oct;45(21):3549-3559. doi: 10.1080/09638288.2022.2131003. Epub 2022 Oct 12. PMID: 36222285.
26) Wang, Y.-T., Jahani, S., Morel-Swols, D., Kapely, A., Rosen, A., & Forghani, I. (2024). Patient experiences of receiving a diagnosis of hypermobile Ehlers–Danlos syndrome. American Journal of Medical Genetics Part A, 194A:e63613. https://doi.org/10.1002/ajmg.a.63613
27) Bulbena-Cabré A, Baeza-Velasco C, Rosado-Figuerola S, Bulbena A. Updates on the psychological and psychiatric aspects of the Ehlers-Danlos syndromes and hypermobility spectrum disorders. Am J Med Genet C Semin Med Genet. 2021 Dec;187(4):482-490. doi: 10.1002/ajmg.c.31955. Epub 2021 Nov 22. PMID: 34806831.
28) Lilly. Forteo (teriparatide injection) Package insert. https://medical.lilly.com/us/products/answers/what-are-the-common-adverse-reactions-associated-with-forteo-teriparatide-injection-26358
*I want to make a note to fully accept that the use of race and ethnicity based categorization is hugely problematic, especially in genetic studies, but the very general point I’m making here still stands. There are groups of people with privilege who have access and trust in the medical system and there are groups who do not have the same prevalence of access.
