Who Decides What Counts as CMT?

In October, while I was testing the Research Watch tool, it picked up a paper about a gene-editing experiment for a rare neuropathy called HMSN-P. Before I could decide whether to publish it, I had to answer a question that sounds simple: is HMSN-P a kind of CMT?

It turns out there’s no single answer. One major database files HMSN-P under CMT. Another files it next to CMT, as one of the many adjacent diseases. The expert panel that rates CMT genes has HMSN-P on its to-do list. These classifications can reflect different questions and purposes, but disagreements can also reveal inconsistencies that need resolving. No single body has the final say on CMT’s boundaries. But how would they go about doing so, and who decides? Who sits on this expert panel, and what processes do they have for deciding questions like HMSN-P’s classification as a disease?

Key concepts that are easy to confuse: A genetic testing panel is a laboratory test examining a selected set of genes. An expert panel is a group of specialists who review scientific evidence, such as what genes should be included in a genetic testing panel for a particular disease or disease group. This article refers to both.

This post is about the edges of CMT: the conditions and genes that sit on the boundary, who decides where they go, and how. If you want the story of why CMT’s type names (CMT1A, CMT2S and so on) are such a confusing mess in the first place, start with my companion post, Name Games.

CMT Names a Pattern, Not a Cause

“Charcot-Marie-Tooth disease” describes what doctors saw in the 1880s, and what we still see on the surface: muscle weakness and wasting that starts in the feet and lower legs and/or the hands and forearms. It runs in families and creeps slowly up the longest nerves. CMT was a description of this pattern long before anyone could name a gene or knew what they were in terms of chromosomes and DNA. Today more than 80 genes have been identified that can produce the CMT pattern, and many of them can also produce something that looks different: weakness without sensory loss, sensory loss without much weakness, problems starting in the hips instead of the feet, or a disease that reaches the spinal cord and brain.

So every boundary question about CMT is really a question about how far the old model can stretch to accommodate new phenomena before it becomes obsolete and is discarded or evolves into something else. Eventually a new paradigm will emerge. For now, we’re in the “stretching to accommodate new phenomena” phase. Here are the edge cases that come up most because they require the most stretching and straining to call them CMT.

The Edge Cases

HMSN-P: Weakness That Starts in the Hips

HMSN-P stands for hereditary motor and sensory neuropathy with proximal dominant involvement. It was first described in 1997 in families from Okinawa, so it is sometimes called the Okinawa type.¹ Unlike typical CMT, the weakness starts closer to the edges of the trunk or torso, in the hips and shoulders. In 2012 researchers found the cause, a single change in a gene called TFG, and autopsies surfaced the same kinds of protein clumps found in Lou Gehrig’s disease, ALS.² Families with this TFG variant have since been reported in Korea, Iran, India, Brazil and Italy.

How is it classified? Orphanet, the European rare disease database, puts HMSN-P inside its CMT group. The Mondo disease ontology, which ClinGen and many genetics tools use, puts it outside CMT, alongside it under “hereditary peripheral neuropathy.” ClinGen’s CMT expert panel lists TFG as “in scope” but hasn’t curated it yet.³ The same gene, with different changes, also causes a form of hereditary spastic paraplegia, a disease of the spinal cord,⁴ so ot differs in this respect from other diseases classified as CMT.

Motor-Only Neuropathies

Distal hereditary motor neuropathy (dHMN) looks like CMT without the sensory loss. The trouble is that the same genes cause both. For HSPB1, ClinGen’s panel notes that the same variant can cause CMT in one family member and motor-only neuropathy in another, so it rated them as a single disease. It did the same for GARS1, combining CMT2D and a motor-only form under “Charcot-Marie-Tooth disease.”

Outside ClinGen, dHMN is usually kept apart. Mondo and Orphanet both place it outside CMT. The World Health Organization’s disease codes, ICD-11, go further and file dHMN under spinal muscular atrophy, in a different section from the hereditary neuropathies altogether.⁵

SORD: One Gene, Two Names

SORD deficiency, discovered in 2020,⁶ illustrates the classification problem most acutely. ClinGen’s CMT panel rates SORD under the plain name “Charcot-Marie-Tooth disease” and calls it one of the most common recessive forms. OMIM and Mondo call the same condition “distal hereditary motor neuronopathy, autosomal recessive 8.” I’ve argued before that the “first treatable CMT disease” is neither CMT nor a disease because SORD causes a treatable enzyme deficiency that causes a disorder. You could call it any one of these things, or two or all three of them, as the CMTRF does. The databases still haven’t settled the question, so the naming varies from source to source.

This has real consequences. The completed Phase 2/3 trial of govorestat for SORD deficiency is listed on ClinicalTrials.gov as “Hereditary Neuropathy Caused by SORD Deficiency.” When I searched the registry’s condition field on October 14, a search for “Charcot-Marie-Tooth” didn’t find it; a search for “hereditary neuropathy” did.⁷

Sensory-Heavy Forms

At the other end of the spectrum, CMT2B, caused by RAB7A, mostly affects sensation, with foot ulcers and only variable weakness. It’s called CMT, but it sits right next to the hereditary sensory and autonomic neuropathies (HSAN). ClinGen’s CMT panel covers both: it rates RAB7A under CMT type 2 and SPTLC1 under HSAN type 1. SPTLC1 is a good example of a split. Some changes in it cause a sensory neuropathy; others cause a childhood form of ALS, and ClinGen treats those as separate diseases.⁸ ICD-11 gives HSAN its own code.

HNPP: Same Gene, Opposite Change

An extra copy of the PMP22 gene causes CMT1A, the most common form of CMT. A missing copy causes hereditary neuropathy with liability to pressure palsies (HNPP), where nerves are easily injured by pressure.⁹ ClinGen rates them separately, because the change in the gene works the opposite way. Mondo also keeps HNPP outside CMT. Orphanet puts it inside its CMT group, and ICD-11 indexes it with the hereditary motor and sensory neuropathies.

SMARD1 and CMT2S: One Gene, Very Different Lives

IGHMBP2 can cause SMARD1, a severe disease of infants that affects breathing,¹⁰ or it can cause CMT2S, a milder neuropathy.¹¹ Again, even a clear genetic basis for a disease is not always enough, by itself, to distinguish its unique presentation in everyone who has that genetic variation. Because of their common root in SMARD1, ClinGen combined SMARD1 and CMT2S into one disease, or rather a disease spectrum, under the broad name “hereditary peripheral neuropathy” which is broad enough to describe many kinds of CMT. Mondo keeps them apart, with SMARD1 under spinal muscular atrophy and CMT2S under CMT. One of the one-patient antisense drugs in Research Watch is for CMT2S, not SMARD1. ClinGen in this case is the lumper, and Mondo is the splitter.

Look-Alikes the CMT Panel Rates Anyway

ClinGen’s CMT panel has also rated some genes for conditions that aren’t neuropathies of the CMT kind at all. SYT2 is rated for a congenital myasthenic syndrome, a problem where nerve meets muscle that can cause foot deformities and weakness much like CMT.¹² SEPTIN9 is rated for hereditary neuralgic amyotrophy, attacks of severe shoulder and arm pain followed by weakness.¹³ The panel doesn’t say why it took these on. My guess is that they turn up when people with CMT-like symptoms are tested, so somebody has to rate them.

Who Decides

No single body owns the word “CMT.” Different groups decide different things:

• Gene names: the HUGO Gene Nomenclature Committee (HGNC). It names genes, not diseases. It’s why GARS became GARS1 and SEPT9 became SEPTIN9.
• Disease entries and numbers: OMIM, a catalogue curated at Johns Hopkins. Most lettered names, such as CMT2D, come from OMIM entries, and OMIM sometimes renames them.
• Whether a gene really causes a disease: ClinGen’s expert panels. The CMT panel became an approved ClinGen expert panel in December 2019. It’s chaired by Stephan Züchner and lists 15 members. This is the Charcot-Marie-Tooth Disease Gene Curation Expert Panel (CMT GCEP). As of October 14 it had rated 70 genes in 72 curations.¹⁴
• Where a disease sits in the family tree: disease ontologies such as Mondo (funded by the US National Human Genome Research Institute) and Orphanet (a European database) often disagree, as HMSN-P and HNPP show. Their different purposes, structures and scopes help explain some disagreements. Others involve outdated or inconsistent entries and mappings that need review. Mondo actively works to reconcile these differences.
• Medical codes: the World Health Organization’s ICD-11. It has codes for CMT1, CMT2 and intermediate CMT under “hereditary motor and sensory neuropathy,” a separate code for HSAN, and puts dHMN under spinal muscular atrophy.¹⁵
• Your own diagnosis: your testing lab and your doctor. Whether your particular variant causes a disease is judged under separate rules for interpreting variants.¹⁶ A gene’s rating says nothing about your variant.
• Research groups: consortia decide who and what to study. The Inherited Neuropathy Consortium, now funded by the CMTA, lists CMT types as the diseases it studies. It doesn’t list HNPP, dHMN or HSAN.¹⁷

How They Decide

ClinGen’s process is the most transparent, so it’s the easiest one to understand. A panel scores the published evidence that a gene causes a disease: how many unrelated families have been reported, how strong the genetic evidence is, and whether lab experiments back it up. The result is one of seven ratings, from Definitive through Strong, Moderate and Limited to Disputed, Refuted and No Known Disease Relationship.¹⁸ Of the CMT panel’s 72 curations, 52 are Definitive and one, MED25, is Disputed. (MED25 is associated with CMT2B2, Basel-Vanagaite-Smirin-Yosef syndrome, and Syndromic Intellectual Disability,)

Before scoring, the panel has to decide what “the disease” is. That’s the lumping-and-splitting question. ClinGen’s guidance is to lump two named conditions into one unless they differ in how the gene change works, in how they’re inherited, or in how they show up.¹⁹ That’s why HSPB1’s CMT and motor-only forms were lumped, and why PMP22’s duplication and deletion were split. ClinGen now also favours gene-first names such as “SORD-related neuropathy” over lettered ones.²⁰

Mondo tries to reconcile disease descriptions from OMIM, Orphanet and other sources into a shared system that computers can use. This takes more than matching names: curators must decide whether two entries describe the same condition, overlapping conditions, or different ones. A disease can also belong to more than one category. Some disagreements reflect different purposes; others reveal inconsistencies that need resolving.²¹

Is “CMT” Even the Right Name?

Many specialists think the system needs changing. In a 2016 survey of 107 CMT specialists, 65% said the classification should change.²² A new guideline on genetic testing for nerve disorders, published in September in Muscle & Nerve, uses a wider umbrella term, “inherited peripheral neuropathies.”²³ But no replacement for “CMT” has been agreed on, and recent clinical reviews still use the CMT1, CMT2, CMT4, and CMTX framework.²⁴

Here’s where I land. For understanding your own condition, the gene is what matters, and gene-first names will win in clinics and labs. In biomedical science and clinical diagnosis, eventually “CMT” will go away, and it should. That should happen gradually, as gene-first names take over, and the name shouldn’t be kept simply because charities, research consortia and drug trials are organized and marketed around it. Until then, the honest approach is to treat CMT as the outgoing family name for a group of related inherited neuropathies, with fuzzy edges, and to say clearly when a condition sits on one of those edges.

What This Means for You

This isn’t medical advice. But a few practical points follow.

• Search widely for trials. If you are following clinical trials for “CMT” on ClinicalTrials.gov, search by gene names and broader terms such as “hereditary neuropathy” or “distal hereditary motor neuropathy,” not just “Charcot-Marie-Tooth.”
• Check what was tested. A CMT gene panel may leave out genes filed under motor or sensory neuropathy, or the reverse. Ask what a panel specifically includes or excludes today, as this changes.
• A label isn’t a verdict. If your condition sits on an edge, that says something about the databases, not about how real or serious your condition is.
• The community is wider than the label. CMT groups generally welcome people with related neuropathies, and many of the same treatments and supports apply across these groups.

This last point is important to me. It’s all about the excluded lived experience of the most common rare diseases, whether their causes are known or unknown. In my view, there’s a need and opportunity to close the regrettable gap that comes from splitting up people with forms of neuropathy and neuromuscular disease in “communities” defined by and for the needs and interests of researchers and funding organizations. Both focus on diagnosis and cures, which are at once backward-facing and hopeful to a fault. Our genetics and functional losses are in the past and can’t be undone, only suffered and mitigated. A disease-free future will never arrive for most of us. The present reality — now and every day — involves coping with mobility loss, pain, and common co-morbidities, which no one owns or brings under one roof with an eye to patient-focused care.

How cmt.blog decides what counts as CMT

I needed a rule for Research Watch to follow to filter all the research articles that come out. I also needed it to decide what to exclude or put an asterisk on in the Open CMT Gene Table.

Here’s my rule: Research Watch covers CMT and the hereditary motor and sensory neuropathies by name, HNPP, SORD deficiency, distal hereditary motor neuropathy, and the genes ClinGen’s CMT panel has curated. Conditions outside that panel’s curations are described as related neuropathies, not as CMT. Each condition is named as the panel names it, so only those it files as CMT are called CMT. That keeps the feed and the gene table in step with each other, and it means that when the panel rates TFG, HMSN-P’s place here will follow.

So, as a practical example, the HMSN-P paper now appears in Research Watch, described as research on a related neuropathy, with a note that the databases disagree.

Notes

1. Takashima et al., “New Type of Hereditary Motor and Sensory Neuropathy.” ↩︎
2. Ishiura et al., “TRK-Fused Gene Is Mutated.” ↩︎
3. Orphanet, “Charcot-Marie-Tooth Disease/Hereditary Motor and Sensory Neuropathy.”; Mondo Disease Ontology.; ClinGen, “Gene-Disease Validity Curations.” ↩︎
4. Beetz et al., “Inhibition of TFG Function.” ↩︎
5. Mondo Disease Ontology.; Orphanet, “Charcot-Marie-Tooth Disease/Hereditary Motor and Sensory Neuropathy.”; World Health Organization, ICD-11. ↩︎
6. Cortese et al., “Biallelic Mutations in SORD.” ↩︎
7. ClinicalTrials.gov, “NCT05397665.” ↩︎
8. Mohassel et al., “Childhood Amyotrophic Lateral Sclerosis.” ↩︎
9. Chance et al., “DNA Deletion.” ↩︎
10. Grohmann et al., “Immunoglobulin μ-Binding Protein 2.” ↩︎
11. Cottenie et al., “Truncating and Missense Mutations in IGHMBP2.” ↩︎
12. Herrmann et al., “Synaptotagmin 2 Mutations.” ↩︎
13. Kuhlenbäumer et al., “Mutations in SEPT9.” ↩︎
14. ClinGen, “Charcot-Marie-Tooth Disease Gene Curation Expert Panel.”; ClinGen, “Gene-Disease Validity Curations.” ↩︎
15. World Health Organization, ICD-11. ↩︎
16. Richards et al., “Standards and Guidelines.” ↩︎
17. Inherited Neuropathy Consortium, “Diseases Studied.” ↩︎
18. Strande et al., “Evaluating the Clinical Validity.” ↩︎
19. Thaxton et al., “Lumping versus Splitting.” ↩︎
20. Thaxton et al., “Implementation of a Dyadic Nomenclature.” ↩︎
21. Vasilevsky et al., “Mondo: Integrating Disease Terminology across Communities.” ↩︎
22. Magy et al., “Updating the Classification.” ↩︎
23. Saporta, “Guidelines for Genetic Testing.” ↩︎
24. De Grado et al., “Charcot-Marie-Tooth Disease.” ↩︎

Sources

• Beetz, Christian, Adam Johnson, Amber L. Schuh, et al. “Inhibition of TFG Function Causes Hereditary Axon Degeneration by Impairing Endoplasmic Reticulum Structure.” Proceedings of the National Academy of Sciences 110, no. 13 (2013): 5091–5096. https://doi.org/10.1073/pnas.1217197110.
• Chance, Phillip F., Mary Kathryn Alderson, Kathleen A. Leppig, et al. “DNA Deletion Associated with Hereditary Neuropathy with Liability to Pressure Palsies.” Cell 72, no. 1 (1993): 143–151. https://doi.org/10.1016/0092-8674(93)90058-x.
• ClinGen. “Charcot-Marie-Tooth Disease Gene Curation Expert Panel.” Clinical Genome Resource. Accessed October 14, 2026. https://clinicalgenome.org/affiliation/40063/.
• ClinGen. Gene-Disease Validity Curations. n.d. https://search.clinicalgenome.org/kb/gene-validity.
• ClinicalTrials.gov. “NCT05397665.” National Library of Medicine. Accessed October 14, 2026. https://clinicaltrials.gov/study/NCT05397665.
• Cortese, Andrea, Yi Zhu, Adriana P. Rebelo, et al. “Biallelic Mutations in SORD Cause a Common and Potentially Treatable Hereditary Neuropathy with Implications for Diabetes.” Nature Genetics 52, no. 5 (2020): 473–481. https://doi.org/10.1038/s41588-020-0615-4.
• Cottenie, Ellen, Andrzej Kochanski, Albena Jordanova, et al. “Truncating and Missense Mutations in IGHMBP2 Cause Charcot-Marie Tooth Disease Type 2.” American Journal of Human Genetics 95, no. 5 (2014): 590–601. https://doi.org/10.1016/j.ajhg.2014.10.002.
• De Grado, Amedeo, Marina Serio, Paola Saveri, Chiara Pisciotta, and Davide Pareyson. “Charcot-Marie-Tooth Disease: A Review of Clinical Developments and Its Management – What’s New in 2025?” Expert Review of Neurotherapeutics 25, no. 4 (2025): 427–442. https://doi.org/10.1080/14737175.2025.2470980.
• Grohmann, Katja, Markus Schuelke, Alexander Diers, et al. “Mutations in the Gene Encoding Immunoglobulin μ-Binding Protein 2 Cause Spinal Muscular Atrophy with Respiratory Distress Type 1.” Nature Genetics 29, no. 1 (2001): 75–77. https://doi.org/10.1038/ng703.
• Herrmann, David N., Rita Horvath, Janet E. Sowden, et al. “Synaptotagmin 2 Mutations Cause an Autosomal-Dominant Form of Lambert-Eaton Myasthenic Syndrome and Nonprogressive Motor Neuropathy.” American Journal of Human Genetics 95, no. 3 (2014): 332–339. https://doi.org/10.1016/j.ajhg.2014.08.007.
• Inherited Neuropathy Consortium. “Diseases Studied.” Rare Diseases Clinical Research Network. Accessed October 14, 2026. https://inc.rarediseasesnetwork.org/diseases-studied.
• Ishiura, Hiroyuki, Wataru Sako, Mari Yoshida, et al. “The TRK-Fused Gene Is Mutated in Hereditary Motor and Sensory Neuropathy with Proximal Dominant Involvement.” American Journal of Human Genetics 91, no. 2 (2012): 320–329. https://doi.org/10.1016/j.ajhg.2012.07.014.
• Kuhlenbäumer, Gregor, Mark C Hannibal, Eva Nelis, et al. “Mutations in SEPT9 Cause Hereditary Neuralgic Amyotrophy.” Nature Genetics 37, no. 10 (2005): 1044–1046. https://doi.org/10.1038/ng1649.
• Magy, Laurent, Stéphane Mathis, Gwendal Le Masson, Cyril Goizet, Meriem Tazir, and Jean-Michel Vallat. “Updating the Classification of Inherited Neuropathies: Results of an International Survey.” Neurology 90, no. 10 (2018): e870-e876. https://doi.org/10.1212/WNL.0000000000005074.
• Mohassel, Payam, Sandra Donkervoort, Museer A. Lone, et al. “Childhood Amyotrophic Lateral Sclerosis Caused by Excess Sphingolipid Synthesis.” Nature Medicine 27, no. 7 (2021): 1197–1204. https://doi.org/10.1038/s41591-021-01346-1.
• Monarch Initiative. Mondo Disease Ontology, Release 2026-10-06. n.d. https://www.ebi.ac.uk/ols4/ontologies/mondo.
• Orphanet. “Charcot-Marie-Tooth Disease/Hereditary Motor and Sensory Neuropathy (ORPHA:166).” Orphanet. Accessed October 14, 2026. https://www.orpha.net/en/disease/detail/166.
• Richards, Sue, Nazneen Aziz, Sherri Bale, 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.” Genetics in Medicine 17, no. 5 (2015): 405–424. https://doi.org/10.1038/gim.2015.30.
• Saporta, Mario A. “Guidelines for Genetic Testing of Peripheral Nerve Disorders.” Muscle & Nerve (September 2026). https://doi.org/10.1002/mus.70394.
• Strande, Natasha T., Erin Rooney Riggs, Adam H. Buchanan, et al. “Evaluating the Clinical Validity of Gene-Disease Associations: An Evidence-Based Framework Developed by the Clinical Genome Resource.” American Journal of Human Genetics 100, no. 6 (2017): 895–906. https://doi.org/10.1016/j.ajhg.2017.04.015.
• Takashima, Hiroshi, Masanori Nakagawa, Keiichi Nakahara, et al. “A New Type of Hereditary Motor and Sensory Neuropathy Linked to Chromosome 3.” Annals of Neurology 41, no. 6 (1997): 771–780. https://doi.org/10.1002/ana.410410613.
• Thaxton, Courtney, Leslie G. Biesecker, Marina DiStefano, et al. “Implementation of a Dyadic Nomenclature for Monogenic Diseases.” American Journal of Human Genetics 111, no. 9 (2024): 1810–1818. https://doi.org/10.1016/j.ajhg.2024.07.019.
• Thaxton, Courtney, Jennifer Goldstein, Marina DiStefano, et al. “Lumping versus Splitting: How to Approach Defining a Disease to Enable Accurate Genomic Curation.” Cell Genomics 2, no. 5 (2022): 100131. https://doi.org/10.1016/j.xgen.2022.100131.
• Vasilevsky, Nicole A., Sabrina Toro, Nicolas Matentzoglu, et al. “Mondo: Integrating Disease Terminology across Communities.” Genetics 232, no. 4 (2026): iyaf215. https://doi.org/10.1093/genetics/iyaf215.
• World Health Organization. “ICD-11 for Mortality and Morbidity Statistics, 2026-01 Release.” Accessed October 14, 2026. https://icd.who.int/browse/2026-01/mms/en.