The HEDGE analysis team is currently preparing a number of papers for publication which will share their initial findings from the HEDGE (Hypermobile Ehlers-Danlos Genetic Evaluation) study. These papers are expected to be released in 2025.

Ahead of this, the team submitted abstracts describing preliminary work on the HEDGE study to the American Society for Human Genetics (ASHG) conference. 

The conference committee accepted the following three abstracts for publication and presentation at the conference:

Laukaitis C., Subramanian D.N., Janousek V., et al. Assessment of Suspected Candidate Gene Variants in Hypermobility Ehlers-Danlos Syndrome Patients from the HEDGE Study Cohort (Abstract) Presented at the Annual Meeting of The American Society of Human Genetics, November, 2024 in Denver, Colorado. 

Seth A., He W., Handsaker R., et al. Identifying Rare Variants Associated with Hypermobile Ehlers-Danlos Syndrome Using a Case-Only Cohort and Biobank Controls: Overcoming Ancestry and Technical Differences in Separately Sequenced Samples. (Abstract) Presented at the Annual Meeting of The American Society of Human Genetics, November, 2024 in Denver, Colorado. 

He., Seth A., Handsaker R., et al. Multi-Ancestry GWAS for Hypermobile Ehlers-Danlos Syndrome. (Abstract)Presented at the Annual Meeting of The American Society of Human Genetics, November, 2024 in Denver, Colorado.

The abstracts can be read HERE by clicking on the “Program” tab, then the “Browse Abstracts” tab, and searching in search bar for HEDGE. The Ehlers-Danlos Society is not permitted to publish them on its website until after the conference. 

The Ehlers-Danlos Society has produced a general Q&A webinar update on the HEDGE Study and the publications, which can be watched below.

The HEDGE researchers have provided the following summaries and FAQ’s about these abstracts:

Abstract 1. Poster at the ASHG 

Identifying Rare Variants Associated with Hypermobile Ehlers-Danlos Syndrome Using a Case-Only Cohort and Biobank Controls: Overcoming Ancestry and Technical Differences in Separately Sequenced Samples 

Summary: 

This abstract reported on our initial efforts to compare the HEDGE genetic data with a control group, to determine if any genes in the HEDGE group have an excess number of rare variants (those with frequency in the general population of less than 1%). For the controls, we used the genetic data from the large “All of Us” project, which is sponsored by the U.S. National Institutes of Health.  We eliminated people with hypermobility syndromes such hEDS and HSD and chose five controls for each HEDGE participant.   

To select control subjects with an ancestry very close to their matched HEDGE participant, we developed a novel method relying on the genetic data that we call MANCS (Multi-Ancestry Nearest Control Selection).  This allowed us to select the best five controls for each HEDGE participant from the All of Us data. 

Another challenge was the fact that the All of Us participants underwent genetic sequencing separately, so we had to be concerned about possible differences in sequencing technology and data analysis. We used several methods to adjust for possible differences, judging our success partly by the number of “synonymous variants,”  a well-known method for assessing such issues. A synonymous variant is one that does not change the resultant protein because it does not change the meaning of the genetic code specifying the sequence of amino acids. Therefore, they should not usually appear in different numbers in cases vs. controls. At the time of the abstract submission, only 8 genes remained (from over 20,000) with unexplained differences in the number of synonymous variants and the work continued to examine other possible sources of bias.  

We are quite pleased with the results of these efforts and believe they allow us to make high-quality statistical comparisons. This sort of data cleanup, though complex and time-consuming, is essential to have meaningful results.   

The next step, which we expect to complete soon, is to determine which genes have an excess burden of rare variants in our group of people with hEDS. This will provide a set of candidate genes for further study.   

What specific rare variants are being investigated in this study?

This study focuses on identifying places in the genome where most people have the same DNA sequence.  We focused on rare genetic variants that are predicted to change the amino acid sequence of a protein. These are called loss-of-function and missense variants, and they may contribute to the risk of hEDS. 

How do the researchers ensure that ancestry differences do not bias the results?

This study used an ancestry-based matching approach, where each hEDS case was matched to five ancestry-matched controls from the AllofUs biobank, described in more detail below. This matching strategy resolves ancestry differences between the cases and controls. 

What are the potential implications of this research for diagnosing hEDS?

By identifying genes that have rare genetic variants that increase or decrease the risk of hEDS, this study could provide a deeper insight into the underlying genetic basis of the condition. If we find rare variants that strongly predict the presence of hEDS, these could help to refine diagnostic criteria and potentially could be used as diagnostic genetic tests for some individuals with hEDS. Both would enable earlier and more accurate identification of hEDS in diverse patient populations. Even if we were to discover only genes with variants that are not strong enough predictors of hEDS to form the basis of a diagnostic test, such a discovery would still greatly enhance our understanding of the underlying biological causes of hEDS and give us clues as to how to think about better preventive measures or therapies. 

Why was the AllofUs biobank chosen as the control group for this study?

The AllofUs biobank was chosen because it includes over 249,000 participants with a wide range of ancestral backgrounds, allowing for precise ancestry matching to the hEDS case cohort. Ancestry-matched controls from the All of Us Biobank were carefully selected for each hEDS case to minimize potential effects of population differences and improve the accuracy of the study’s findings. The All of Us Biobank already has whole genome sequence data, making it an ideal resource for selecting matched controls for hEDS cases across various populations for a study of rare variation. 

What does the process of harmonizing transcript annotation sets involve?

Harmonizing transcript annotation sets involves aligning and standardizing the genetic data by using standard reference transcript annotations – which help accurately predict the impact of genetic variants on proteins. This step helps in mitigating artifacts that could arise from systematic transcript annotation differences, ensuring consistency in variant identification and analysis. 

How might this research affect individuals of different ancestries within the hEDS community?

This study aims to identify genetic variants associated with hEDS across various ancestries, offering valuable insights into this condition. The findings could potentially lead to more insights into the pathophysiology of hEDS and better diagnostic and treatment options, ultimately benefiting the hEDS community. The generalizability of our study is limited by the ancestral diversity of the participants. That being said, in most disease studies that identify specific genes, the genes contributing to a disease are typically consistent across different ancestries. 

How does this study’s approach differ from previous genetic studies on hEDS?

This is the first well-matched, comprehensive study of rare variants of this size for hEDS. This study also uses a novel ancestry-based matching technique and a systematic quality control pipeline to address both ancestry and technical differences between separately sequenced case and control cohorts. This comprehensive approach allows the inclusion of all cases regardless of ancestry and a well-calibrated analysis approach for rare variant testing that sets it apart from prior genetic studies that may not have addressed such biases. 

What are nonsynonymous variants, and why are they important in this context?

Nonsynonymous variants are differences in DNA sequence that result in a change in the amino acid sequence of a protein. Investigating these variants is crucial, as they can have direct and oftentimes large impacts on protein function, making them more likely to contribute to the development of rare disorders like hEDS. 

What is the significance of the Genomic Inflation Factor (λGC) in this study?

The Genomic Inflation Factor (λGC) is used to assess the presence of systematic bias in the genetic analyses. A λGC close to 1.0 (as achieved in the study’s quality control steps) indicates minimal residual inflation and suggests that the technical differences and population structure have been adequately accounted for. 

Abstract 2. Poster at the ASHG 

Assessment of Suspected Candidate Gene Variants in Hypermobility Ehlers-Danlos Syndrome Patients from the HEDGE Study Cohort 

Summary:

As part of our analysis of the HEDGE genetic data, we looked specifically for known causes of the various types of Ehlers-Danlos syndrome and other related disorders, as well as variants in genes we thought could be involved based on previously known information (a total of 90 genes).  Some of these genes were the subject of previous published reports.  Others were genes we judged to be candidates based on known connective tissue biology. 

Only six participants had variants diagnostic of another (e.g., classical EDS) that could explain their symptoms and findings. This low number shows that physicians diagnosing hEDS and HSD are doing a good job in ruling out other known conditions that could cause similar findings.  We notified each of these individuals. 

We also examined 184 candidate genes for which there was information suggesting possible involvement in causing hEDS, but we did not find any increases in variants in these genes that would lead us to believe they are involved.  At the time of submission of the abstract, we were still studying our data to better understand these findings. 

We wish to re-emphasize that HEDGE was not designed to evaluate genetic data for other conditions, such as a predisposition to cancer or other inherited diseases, so you should not rely on participation in this study to rule out any such genetic findings 

What are "candidate genes" in this study?

Candidate genes include genes known to cause rare types of EDS and other connective tissue disorders, since many HEDGE study participants have not yet had the opportunity to have clinical gene sequencing tests done. In addition, we studied genes that have been identified in previous research work as possible new causes for hEDS. One example in the first category is the gene FBN1, associated with the disease Marfan Syndrome, which has several features that overlap with hEDS (i.e. joint hypermobility, mitral valve prolapse, dental crowding). In this poster abstract, we are describing our early results for both categories of candidate genes. We focused primarily on variants found in the first category as we tried to ensure that no one participating in the HEDGE study actually had a different known connective tissue disorder. Our future work will include analysis of variants in candidate genes that are newly identified by our work in the HEDGE study as part of the ongoing genetic analysis. 

What types of genetic variants are being studied in the HEDGE project?

Genetic variants can be of a number of types—sometimes, they are single base changes where one spot in a person’s DNA sequence has a different base in place of the usual base (e.g. a C instead of an A). Sometimes, a variant takes the form of one or more bases being missing from the DNA (deletion), or there can be extra DNA (insertion -e.g. an extra few bases are inserted into the gene). Occasionally, a variant is larger: thousands of missing or extra DNA bases or more complicated changes in the chromosome structure.  

Our initial efforts are focused on more readily identified genetic variants, consisting of small deletions, insertions and single base DNA changes in the parts of the human genome that code for the proteins which form the human body. Depending on the results of this work, we plan to extend our studies to other types of genetic variants that are harder to detect and/or interpret; these include variants in parts of the genome that do not directly code for proteins, and structural variants that duplicate or remove large segments of DNA. 

What are "variants of uncertain significance" and how do they impact our understanding of hEDS?

The term ‘genetic variant’ means that the sequence (or spelling) of someone’s gene varies from the ‘usual’ sequence – called the reference genome. The reference genome is a consensus from thousands of people who have had their genomes sequenced.  Most variants that differ from the reference genome have no impact on health, but some lead to disease.  These are sometimes caused mutations but are more properly labeled as ‘pathogenic variants.’  The variants that do not cause disease are called ‘benign variants.’  Unfortunately, there is a grey area where a variant can be of ‘uncertain significance.’   

A variant of uncertain significance (VUS) is a genetic variant where there is not enough information to know whether it is benign or pathogenic and whose impact on the gene where they are found is currently unknown. Variants are labeled as VUS after an expert assessment that includes factors such as how common this variant is in people with the disease as well as in the general population, its predicted effect on the gene’s function, and other data (e.g. from experimental studies).  

If a pathogenic variant is found in a gene known to cause connective tissue in a person enrolled in HEDGE, then, if confirmed, this would change this person’s diagnosis from hEDS to the other connective tissue disease. These participants have been notified by the HEDGE PIs so that the presence of the variant can be confirmed in a clinical lab test. But, if the variant is of uncertain significance (VUS), then we aren’t sure whether it causes disease.  In this case, as would be the case if this testing were done by a physician using a commercial clinical lab, then this VUS would not be useful for changing clinical management or other courses of action such as testing other relatives. Thus, we are not routinely returning these unclear VUS results to HEDGE participants unless our analysis provides a strong indication that they may be reclassified as ‘likely pathogenic’ in the near future. It is also possible that some of these VUSs are benign and do not have any effect on the associated gene, and/or that an unrecognized variant in a different gene may be the actual cause of the hEDS symptoms in that individual. 

Why is it important to continue studying the genetic causes of hEDS?

hEDS is one of the few diseases with a strong hereditary component that currently does not have any known genetic cause. This creates problems for individuals with the disease as well as their healthcare professionals (HCPs): 

  • There is no genetic test to confirm the diagnosis in people suspected to have the condition or their wider family, creating barriers to disease recognition, management and acceptance. 
  • It contributes to under-recognition of the disease by HCPs, leading to misdiagnoses and delayed diagnoses. 
  • Our lack of understanding of the disease and its causes hinders the development of effective therapies and treatments. 

What does it mean if someone carries a gene variant but does not have hEDS?

If a person carries a ‘likely pathogenic’ variant in a gene for a connective tissue disease that causes symptoms overlapping with hEDS, then there is a good possibility they have a different diagnosis (e.g. Marfan Syndrome) that better explains their symptoms. Knowing this may improve the way this person’s medical care is managed. However, variants detected in this study will still require independent confirmation by an accredited clinical diagnostic genetics laboratory and clinical evaluation by participant’s physician, as occasionally there can be disagreement on how to interpret difficult ‘borderline’ variants that other labs may label as ‘VUSs’, and not everyone with a genetically diagnosed disorder has all the features of that disorder.  

How does this study define "likely pathogenic" variants, and how confident can we be in their role in disease?

‘Likely pathogenic’ variants, as defined by guidelines issued by the American College of Medical Genetics and Genomics and the Association for Molecular Pathology* and used by diagnostic and research genetic laboratories worldwide, have a >90% likelihood of damaging the gene it is found in, enough to cause disease. This means we can be at least 90% confident they are causing the disease associated with that gene, but there is still a small chance (up to 1 in 10) that the variant may have little or no impact i.e. it does not cause disease. Genetic counselling provided to participants with these variants will normally take this into account when they are advised of the implications and their options. 

* Richards S, Aziz N, Bale S, et al. Standards and guidelines for the interpretation of sequence variants: a joint consensus recommendation of the American College of Medical Genetics and Genomics and the Association for Molecular Pathology. Genet Med. 2015;17(5):405-424. doi:10.1038/gim.2015.30 

What is the role of expert phenotyping in the study, and why is it important?

Expert phenotyping refers to the detailed review of relevant signs, symptoms and other medical information provided by the HEDGE participants in their survey responses alongside their in-person or chart assessment. This is vitally important for interpreting any genetic data from a diagnostic or research test, as it is very difficult to deduce the relevance of any detected variants or their impact if nothing is known about the person’s health. For the HEDGE participants, we have carefully and systematically analyzed the data they provided to ensure that our variant interpretations are robust.  

Abstract 3. Poster and Presentation at the ASHG 

Multi-Ancestry GWAS for Hypermobile Ehlers-Danlos Syndrome 

Summary: 

This abstract and presentation reported our work performing a genome-wide association study (GWAS) comparing the HEDGE group to the controls.  As with the rare variant burden part of the study, we used MANCS (Multi-Ancestry Nearest Control Selection), a new method we developed to precisely choose controls for each HEDGE participant.   

GWAS examines whether common variants (with a prevalence of 1% or more in the general population) might combine to cause a disease.  For hEDS, since such common variants all occur more frequently in the general population than does hEDS, no single common variant could be the cause.  GWAS examines the possibility that multiple genes, each with a small effect, might combine to give rise to the findings (the “phenotype”).  This is a well-understood but very complex and sophisticated statistical method, with a long history of use in the study of many traits and diseases.   

Since GWAS looks only at common variants, it does not tell us anything about the impact of rare variants.  The prevailing view has generally been that hEDS is caused by one or more rare variants with a large effect size, not the combined effect of many common variants.  

The abstract reports that we found the difference in the HEDGE group from the control group was about 15% due to the impact of common variants.  This is called “heritability” – but remember it does not include the effect of rare variants.  A 15% heritability does NOT necessarily mean the condition is inherited only 15% of the time.  In fact, a condition caused exclusively by rare inherited genetic variants would have a heritability of zero, because it is not caused by common variants. 

This heritability finding could be consistent with many different genetic models.  One possible model is that 15% of cases are due to the combined effect of many common variants, each with a small effect.  With this model, the remaining 85% might be due to rare variants with a high effect, or purely due to environmental causes (infections, etc.).  A more likely model is that rare variants are the main cause, but whether the condition manifests (penetrates) is dependent on the overall milieu of common variants.   

GWAS also examines specific common variants to determine if they occur at a high frequency in the study group.  Unlike the rare gene burden analysis, GWAS does not look at the count of variants in genes, but determines if any specific variants, in and of themselves, occur frequently enough to warrant attention.  We found only one such variant meeting the strict statistical criteria, and it was present in only about 1% of the individuals with hEDS.  Some others came close but did not meet the criteria and warrant further evaluation.  Again, since these variants are common, none could be a sole cause of hEDS. 

Finally, as part of the GWAS analysis, we looked at whether the combination of common variants that give rise to hypermobility also are contributory to hEDS. In other words, we wanted to know if the 15% heritability bears any relation to hypermobility in general.  To do this, we compared our data to data from 23 and Me, where people reported Beighton scores.  We found that the common variants contributing to hypermobility in 23 and ME (the “polygenic risk score”) were also involved in the 15% contribution of common variants to hEDS.   

What does GWAS stand for?

GWAS stands for Genome-Wide Association Study. 

What is a genome-wide association study and how does this apply to HEDGE?

A genome-wide association study (GWAS) is a research method that involves testing, one-by-one, millions of places across the entire genome where we know people often have differences in their DNA (genetic variants), to identify genetic variants that alter the risk of a specific disease or affect a human trait. In the HEDGE study, GWAS was used to search for genetic variants that alter the risk of hEDS by comparing the frequencies of genetic variants in the genomes of individuals with hEDS and those of ancestry-matched controls. 

What does Matching mean in human research studies?

Matching in research studies refers to the process of selecting control participants who are similar to the cases in specific characteristics, such as genetic ancestry and sex. In this study, matching was used to select control participants from the All of Us biobank who closely resemble hEDS cases in terms of their genetic ancestry. This is an important step in ensuring that any genetic differences from differences in ancestry do not incorrectly appear to be related to hEDS, which can happen if the cases and controls are not matched correctly. 

What is Multiple Ancestry-Matched Control Selection Testing?

The Multiple Ancestry-Matched Control Selection (MANCS) method is designed to select controls that closely match hEDS cases in terms of ancestry, using genetic data rather than self-reported racial/ethnic labels. By allowing the pairing of all hEDS cases to their closely matched controls, regardless of any individual’s particular ancestry, this approach minimizes potential biases and allows us to be more inclusive, enhancing our study’s ability to detect true genetic signals associated with hEDS. This increases the reliability of the findings and ensures that they are more applicable to the diverse hEDS community. 

What is QC?

QC stands for quality control. In genetic studies, QC refers to the set of processes used to clean and standardize genetic data, ensuring that the results are accurate and not confounded by technical artifacts or errors. 

What is Logistic Regression and what is an Odds Ratio?

Logistic regression is a statistical method used to model the relationship between a yes/no outcome (in this study, the presence or absence of hEDS) and one or more possible variables that could influence the outcome (in this study, genetic variants). An Odds Ratio is a measure of the strength of the effect of a specific genetic variant on the likelihood of having the disease. An odds ratio greater than 1 indicates an increased risk of the variant, while an odds ratio less than 1 indicates a decreased risk. 

What is the significance of finding a locus on chromosome 5 associated with hEDS?

We know from studies of other diseases that genetic findings from GWAS can provide clues to the causal mechanisms of disease – what happens differently in people with hEDS to make them more susceptible to this disorder? The identification of a locus (genomic region) on chromosome 5 associated with hEDS suggests that this region of the genome may contain genes that may play a role in the development of hEDS. This finding, if confirmed in additional studies, provides a starting point for further investigations, which may offer insights into the underlying causal biological mechanisms of hEDS, which would in turn be the first step in guiding the development of possible therapies or preventive measures. 

How does the MANCS (Multi-Ancestry Nearest Control Selection) method improve the study's reliability?

The MANCS method enhances the study’s reliability by selecting ancestry-matched controls for each hEDS case. Using a multi-ancestry nearest control selection minimizes the potential false positive results that could result from ancestry differences between cases and controls, ensuring that any identified genetic variations that differ between cases and controls are more likely linked to hEDS itself rather than ancestry. Additionally, it increases the study’s statistical power by allowing the inclusion of all cases and their matched controls regardless of ancestry, improving the power to identify genetic variants associated with hEDS. 

What does it mean that hEDS is "modestly heritable" concerning common variants?

“Modestly heritable” means that part of the risk for hEDS is due to common genetic variants (those present in ~1% or more of the population), but that common genetic variants alone do not fully explain the condition. The study estimated this common variant heritability to be approximately 15%, indicating that while common genetic variation plays a role, other factors also contribute to the condition. It is also important to note that the standard error of 8% is relatively high due to the limited number of cases in the study, meaning that the actual contribution of common variants to hEDS could be higher or lower than our estimate of 15%. Increasing the number of cases in future analyses would help improve the accuracy of this heritability estimation. 

How does this study's use of polygenic scores (PGS) relate to the Beighton score and joint hypermobility?

Polygenic scores (PGS) combine the effects of many genetic variants that are known to affect a disease or a quantitative measure (trait), to estimate an individual’s genetic predisposition for that disease or trait. For this study, we used a GWAS for joint hypermobility as measured by the Beighton score to create a Beighton score PGS for each of the cases and controls. We found that hEDS cases had significantly higher Beighton score PGS values compared to controls. This tells us that there is likely a shared genetic basis between hEDS and generalized joint hypermobility. 

Why were ancestry-matched controls from the All of Us Research Program used in this study?

Ancestry-matched controls from the All of Us Biobank were carefully selected for each hEDS case to minimize potential effects of population differences and improve the accuracy of the study’s findings. The All of Us Biobank includes over 249,000 participants from diverse ancestral backgrounds, and already has whole genome sequence data, making it an ideal resource for selecting matched controls for hEDS cases across various populations. 

What are the implications of the observed polygenic contributions to hEDS?

This study provides the first evidence indicating a polygenic contribution to hEDS. This finding supports the idea that hEDS has a multifaceted genetic architecture and may share genetic risk factors with other polygenic traits, such as joint hypermobility. 

How might these findings influence future genetic research on hEDS?

This study completed the first GWAS for hEDS, uncovering suggestive genetic associations and offering initial evidence of a polygenic contribution to hEDS. These findings establish a crucial foundation for understanding the genetic basis of hEDS. Moreover, by using a stringent analytical approach that minimizes confounding effects, the study provides a more reliable identification of genetic variants associated with hEDS. Although common variants with modest effects, such as those that are discovered in GWAS, do not have diagnostic utility, variants that influence the risk of hEDS can provide clues to underlying biological mechanisms that influence the risk of hEDA. This robust methodology can guide future research efforts, helping to deepen our understanding of hEDS.  

What challenges remain in using GWAS to study rare diseases like hEDS?

The challenges include the need for larger sample sizes to detect genetic variants with the modest effect sizes that typically are seen in GWAS. Increasing the number of cases will boost the statistical power needed to identify causal variants of more modest effects. Additionally, GWAS only paints part of the picture: other genetic analyses, such as rare variant and copy number variant analyses, are important to fully investigate the genetic basis of hEDS. 

How does this study contribute to understanding the genetic architecture of hEDS?

As addressed above, this study provides valuable insights into the genetic architecture of hEDS by completing the first GWAS for hEDS, uncovering genetic variants with suggestive associations to hEDS, and offering the initial evidence of the polygenic nature of hEDS. These findings lay the groundwork for further research into the genetic basis of hEDS. 

What are the next steps for validating the findings from this GWAS?

The next steps for validating our findings include further analysis of the suggestive genetic loci identified in the hEDS GWAS and replicating these results using additional independent hEDS cohorts (which we will assemble from the AllofUs database). Furthermore, our team is integrating rare variant and copy number variant analyses to gain a more comprehensive understanding of the genetic underpinnings of hEDS. 

*** In genetics, a variant refers to a difference or alteration in the DNA sequence compared to a reference sequence. Variants can occur naturally and may be small changes, such as a single nucleotide polymorphism (SNP), or larger changes like insertions, deletions, or duplications of DNA segments. Some variants have no effect on an individual’s health, while others can influence traits or increase the risk of certain diseases. Variants can be inherited or arise spontaneously (de novo) during a person’s life. 

**** With impending completion of data standardization, we expect to finish the rare variant burden analysis before year-end, setting the scene for full publication of the results in 2025.  For many observers, the rare variant burden analysis is the heart of HEDGE.  We will share these results with each HEDGE participant, along with information specific to each participant. 

After completion of the planned study, the HEDGE team will continue to analyze the data for additional insights about regulatory and mitochondrial genes and mechanisms of disease.  We designed HEDGE to rely on probability cut-offs for significance, but we know such cutoffs can also sometimes eliminate findings that do not quite meet the strict criteria, so we may need to examine those genes as well. Additionally, some genetic mechanisms involve interaction of a handful of genes in a specific way.  We will be looking for the presence of such interactions.  Work on the HEDGE data will continue long after the initial project is completed. 

Sign up to The Ehlers-Danlos Society mailing list

Click here to sign up