Pushing the Limits
Updated: Dec 1, 2021
Examining the Limitations of Genetic Testing in Charcot Marie Tooth Disease
Charcot Marie Tooth disease, or CMT for short, is a heterogeneous group of inherited peripheral neuropathies. CMT is rare, affecting approximately only 3 million people worldwide, but it’s the most commonly inherited peripheral neuropathy. Although CMT symptoms can be treated and well-managed, the disease itself has no known effective treatment or cure. The elusive treatment and cure are a consequence of the ever-expanding catalog of genes discovered to have CMT-causing mutations. Due to this expanding catalog, CMT is perhaps one of the fastest growing areas in human genetics.
Discoveries in CMT genetics are occurring at breakneck speed. Scientists discovered the first CMT genetic link in 1991. Scientists have continued to make new CMT genetic discoveries every year since. The number of genes discovered to have CMT-causing mutations in the last 10 years eclipses the number of genes discovered in the first 20 years. Despite these gains, a CMT genetic test result that fails to identify a known CMT genetic cause is still a common outcome, leaving many CMTers with more questions than there are answers.
It All Started When
The first CMT-associated gene discovery was published in June 1992. This discovery, a duplication of the PMP22 gene, is the cause for CMT1A (Patel 1992). We normally have two copies of this gene. CMTers who have CMT1A, however, have a third copy (in rare cases, there are four copies of this gene – a triplication). About a year prior to this discovery, scientists had narrowed down the cause of CMT1A to a duplication of the segment of chromosome 17 where the PMP22 gene lives (Raeymaekers 1991). Raeymaekers et al. (1991) found that 128 CMTers from 12 different families who had CMT1 each had a duplication of a small segment of chromosome 17 (17p11.2-p12). This duplication segregated with the disease, meaning that members of these 12 families who did not have CMT did not have this chromosome 17 segment duplication. Raeymaekers et al. (1991) concluded this discovery to be the cause for CMT1A. Patel et al. (1992), however, identified the specific gene involved. And, things were only just getting started.
In January 1993, scientists announced they had discovered the cause of a Type 1 CMT called Hereditary Neuropathy with liability to Pressure Palsies (HNPP) after finding CMTers with this phenotype each had a deletion of one-copy of the PMP22 gene (Chance 1993). The second gene discovered by scientists to have a CMT-causing mutation was the MPZ gene, with having a mutation causing CMT1. They published this finding in September 1993, calling their discovery CMT1B (Hayasaka 1993). The year closed out with scientists publishing on Christmas Eve the first X-Linked CMT discovery, after finding a CMT-causing mutation in the GJB1 gene (formerly called CONNEXIN32) (Bergoffen 1993).
CMT-associated genes were elusive for the next few years. Scientists didn’t make a new CMT-associated gene discovery until 1996 when they found CMT-causing mutations in the NTRK1 gene (Indo 1996) (this discovery would not be categorized as CMT until the 2010s). A year later, scientists discovered CMT-causing mutations in the PHYH gene (Mihalik 1997) (this discovery, too, would not be categorized as CMT until the 2010s). The last CMT-associated gene discovery made by scientists in the 1990s came in 1998 when they found CMT-causing mutations in the EGR2 gene (Warner 1998). The 90’s started with scientists wanting to discover the cause of CMT. The decade ended with scientists having discovered CMT-causing mutations in six different genes, accounting for eight unique subtypes. As if this wasn’t overwhelming enough, the turn of the century ushered in an explosion of genes discovered to have CMT-causing mutations.
The 1990’s brought six CMT-associated gene discoveries, from 1991 through 1998. New gene discovery was elusive in 1999. Scientists then announced in 2000 they had discovered CMT-causing mutations in four genes: MTMR2 (Bolino 2000), NDRG1 (Kalaydjieva 2000), NEFL (Mersiyanova 2000), and GAN (Bomont 2000) (this would not be categorized as CMT until the 2010s). Scientists discovered another five genes with CMT-causing mutations in 2001: PRX (Guilbot 2001), ELP1 (formerly called IKAP) (Anderson 2001), SPTLC1 (Dawkins 2001), KIF1B (Zhao 2001), and IGHMBP2 (Grohmann 2001). Zhao et al. (2001) found CMT-causing mutations in the KIF1B gene were responsible for causing CMT2A. However, scientists later found, in 2004, that the actual culprit was a mutation in the MFN2 gene (Züchner 2004), but the KIF1B association to CMT was not retracted until 2020 (Clinical Genome Resource 2020). The new millennium was off to one heck of a jammed packed start.
In only the first two years of the new century, of the new millennium, scientists discovered nine genes having CMT-causing mutations. Although scientists found one of those discoveries to be inaccurate, the remaining eight genes still exceed the number of CMT-associated genes discovered in the previous nine years. By the time the end of 2010 rolled around, marking the end of the first 20 years of CMT-associated gene discovery, scientists had found CMT-causing mutations in a staggering 55 different genes. Fast forward another eleven years to present day, late 2021, and scientists have discovered another 65 genes having CMT-causing mutations, for a total of 120 CMT-associated genes over the span of thirty years (Raymond, CMT-Associated Genes 2021). With all of this expansiveness and ginormity, how is it that CMT genetic testing is limited?
There are many commercial laboratories offering genetic testing services for CMT. A review of publicly available CMT genetic test gene lists from various commercial laboratories reveals there isn’t any one commercial laboratory that includes every discovered CMT-associated gene in their CMT panels (a panel is a genetic test that analyzes several genes, as opposed to a test that analyzes a single gene). Commercial laboratories each have their own criteria for deciding what to include or exclude in their panels. For this reason, there is variability from panel to panel, laboratory to laboratory. Knowing what’s included is essential, and what’s included vs. what’s excluded is the most significant variable across all CMT genetic tests.
Invitae Laboratories is arguably the most frequently used laboratory for CMT genetic testing in the US and Canada. Their Comprehensive Neuropathies Panel test 03200 analyzes 102 genes, with an option to add an additional nine for a total of 111 genes (Invitae Comprehensive Neuropathies Panel 2021). Despite this panel’s expansiveness, only 89 genes are associated with CMT. These 89 genes are associated with causing 118 individual CMT subtypes (Raymond, CMT-Associated Genes 2021). Considering there are 120 genes discovered to have CMT-causing mutations, plus an additional five chromosomal locations scientists suspect as having a CMT-causing mutation but have not yet identified the specific gene, all collectively accounting for 155 individual CMT subtypes, the reasons why a CMT genetic test might not identify a cause become clear. A much larger and different kind of genetic test often used for CMT, and one that isn’t limited to only laboratory selected genes on a panel, called Whole Exome Sequencing, has its limitations also.
Whole Exome Sequencing (WES) is a type of genetic test that attempts to look at all coding exons in all of our more than 20,000 genes. A gene has two basic parts: an intron and an exon. An intron is the part of the gene that holds noncoding-genetic-material. The exon is the part of the gene that holds all the coding-genetic-material. Coding-genetic-material is the part of a gene that codes (instructs) a certain molecular process, and each individual gene has its own molecular process that it codes. WES attempts to look at the coding part of every gene for errors in how the gene is assembled. These errors, which are akin to spelling errors, are gene mutations, and these mutations are sometimes responsible for causing diseases such as CMT.
WES analyzes our DNA and pipes all of the sequenced genetic data into a computer program. A clinician then inputs into the program symptom and condition keywords provided by the doctor who ordered the WES. The program uses an algorithm to sort the tens of thousands of pages of genetic sequence data to identify genes that scientists have associated with causing conditions that match the symptom and condition keywords the clinician entered into the algorithm. After the algorithm sorts the data, the program generates a list of genes that are responsive to the keyword inputs. Clinicians then sort and verify this list as they compile what is called the primary gene list. The primary gene list is then analyzed for any present mutations, and clinicians generate the WES interpretation report based on their findings within the primary gene list. Sometimes, this process is fully automated and sometimes there is an abundance of manual data analysis. The level of automation is at each laboratory’s discretion.
WES is not a genetic test that analyzes specific genes like what a gene panel does. WES attempts to analyze all genes. However, clinicians can manually sort WES data for specific genes, then manually add the specific genes to the dataset that is used to generate the primary gene list. WES, however, has technology limitations that might omit some genes. Another WES limitation pertaining to CMT is that scientists believe they have identified only about half of all genes that likely have CMT-causing mutations (Michael Shy 2020). If genes potentially have CMT-causing mutations but are not yet identified by scientists, WES is not going to capture these as responsive genes and clinicians tasked with compiling WES primary gene lists aren’t going to know to manually sort for these genes. Another limitation to WES is the keywords.
WES results are only as good as the keywords used to generate the final report. If keywords don’t match CMT symptom profiles, for example, or symptom keywords are too generalized and vague, WES might not return a result that accurately represents a CMTer’s true genetic profile. In short, CMT genetic testing results, whether WES or otherwise, might not be definitive. If they’re not, how can doctors definitively diagnose CMT?
When Genetic Testing isn’t Enough
CMT was first described in 1886. Doctors knew what it was, and they could tell it was inheritable from how it appeared to run in families. But, the cause was elusive for the next 100 years. Nobody could have predicted prior to the first CMT genetic discovery that underlying causes would balloon to the enormity they have in the thirty years since the first discovery. Despite the profound number of CMT-associated genes, scientists believe we are only halfway to discovering all genes with CMT-causing mutations. Today, approximately 95% of CMTers who have a demyelinating CMT are able to obtain a genetic confirmation. In sharp contrast, only about 50% of CMTers who have an axonal CMT are able to obtain a genetic confirmation, and some scientists suggest this number might be as low as 30% (Michael Shy 2020).
CMT genetics expert, Stephan Züchner, M.D., Ph.D., professor and Chair of the Dr. John T. Macdonald Foundation, Department of Human Genetics, University of Miami, Miller School of Medicine, suggests that discoveries like the CMT-causing mutations in the SORD1 gene (Cortese 2020) have the potential to cut down on the overall number of CMT-associated genes yet to be discovered given the number of CMTers who are likely to have SORD-deficiency CMT (Stephan Züchner 2021). According to Dr. Züchner, there are at least 3,300 CMTers in the US alone who likely have this newly discovered SORD-deficiency CMT. Dr. Züchner suggests that discoveries like this might quite possibly mean that scientists are much closer to discovering all CMT-associated genes than previously thought. However, because all have not been discovered, CMT genetic testing remains unfortunately diagnostically limited, and therefore cannot necessarily serve as a standalone diagnostic tool.
CMT genetic testing is but only one piece of the diagnostic picture doctors consider when diagnosing CMT. Symptoms, family history, and nerve conduction study results all play an equal role when diagnosing CMT, and together, often inform the doctor's genetic testing decisions. Richard A. Lewis, MD, Professor of Neurology, Cedars-Sinai Medical Center; Director, CMTA Center of Excellence, explains that to definitively diagnose CMT, one needs the appropriate clinical picture—numbness and weakness in the feet and usually in the hands occurring with reduced reflexes, plus a family history of the same problems that cannot be due to other causes such as diabetes. Also, the nerve conduction studies should show evidence of neuropathy—nerve conduction velocities that are very slow fit a diagnosis of CMT1, and if not very slow, CMT2. Having pes cavus (high arches) since early in life is supportive evidence of a CMT diagnosis. But other neuropathies can give a similar clinical picture and similar nerve conduction study findings. To be certain it’s CMT, Dr. Lewis explains, an abnormal genetic test would be proof. However, genetic testing doesn’t always detect a problem and that doesn’t mean that a genetic disorder such as CMT is not the cause. In addition, genetic testing results can be difficult to interpret and can be confusing to the patient. A definitive diagnosis depends on a doctor putting all this information together; and, it is important not to miss a treatable cause of neuropathy. With all of this complexity in mind, Dr. Lewis advises that it is best to get a good neurologic evaluation if you have neuropathy.
The Pursuit of…
Although CMT genetic testing is part of the diagnostic work up for CMT, the genetic testing often is a personal choice. The personal reasons for testing or not testing are as many as there are CMTers. A CMT genetic test might not be definitive insofar as the test might not identify a known CMT cause. Often, the test identifies only variants of unknown or uncertain significance (VUS) in CMT-associated genes. A VUS is a mutation in which scientists have not yet determined whether it causes something or is benign (harmless), and a VUS finding in a gene that is known to have CMT-causing mutations is a common finding. A VUS finding in a CMT-associated gene can create confusion for CMTers, the finding can be difficult for doctors to interpret, and the finding requires a more sophisticated diagnostic analysis by the ordering doctor. The frequency at which VUS findings occur in CMT genetic testing lends itself to the often lacking clear-cut and overt genetic confirmation of a CMT clinical diagnosis. This often adds to a CMTer’s frustration, and understandably so. With these pitfalls, why undergo genetic testing?
CMT genetic testing is as much of an integral part of diagnosing CMT as are symptoms, family history, and nerve conduction studies. A nerve conduction study often precedes genetic testing, and the results can inform a clinical diagnosis. Before genetic testing is performed, when nerve conduction study findings are consistent with demyelinating CMT, doctors use CMT1 as a blanket diagnosis, and when consistent with axonal CMT, CMT2 (Gondim and Thomas 2019). However, the only way doctors can identify the exact CMT subtype is through genetic testing that identifies a known CMT-causing genetic mutation (El-Abassi 2014), or, as Dr. Lewis explains, through a more sophisticated analysis of all diagnostic findings in the case of a genetic test VUS finding. The blanket diagnosis of CMT1 or CMT2 is descript as a clinical diagnosis but infers no genetic cause. Raymond (2021) identifies 27 demyelinating CMT subtypes and 112 axonal CMT subtypes. Upon genetic confirmation, a CMT1 or CMT2 clinical diagnosis can change, and is dependent upon the identified CMT-causing gene mutation. There are four categories in which demyelinating CMT is found—CMT1, CMT4, CMTX, and [Gene Name]-CMT; and there are ten categories in which axonal CMT is found—CMT2, CMTX, dHMN, dSMA, GAN, HMSN, HSAN, HSN, SMA-LEP, [Gene Name]-CMT (Raymond, CMT-Associated Genes 2021). It’s easy to see how a CMT1 or CMT2 clinical diagnosis might change after the underlying CMT-causing mutation is identified, but why does knowing the exact subtype matter?
CMT clinical trials are often subtype specific. CMT, due to its heterogeneity (many unique causes, many diverse presentations), is a collection of 155 genetically distinct diseases. Each of these 155 genetically distinct diseases is a CMT subtype. Researchers often target a specific subtype in their clinical trials. As such, clinical trials typically require that every participant have genetic confirmation of their CMT (genetic confirmation = identified subtype). Gene therapy as a potential treatment and even as a potential cure for CMT involves targeting a specific gene or gene mutation. Therefore, a potential gene therapy for CMT4C, for example, which is caused by autosomal recessive mutations in the SH3TC2 gene (Senderek 2003), would likely be of no benefit to a CMTer who has CMT1H, which is caused by autosomal dominant mutations in the FBLN5 gene (Safka Brozkova 2020). For these reasons, identifying the underlying responsible CMT-causing genetic mutation is essential. Although CMT genetic testing might come up empty, testing is critical in these circumstances. Another reason for CMT genetic testing is family planning.
CMT is inheritable. CMT is inheritable because each of the genetic mutations that cause the disease are inheritable. How CMT is inherited, that is, how it’s passed on, is dependent solely on the underlying CMT-causing genetic mutation. CMT is passed on in one of five ways. Each of these are known as an inheritance pattern, and each inheritance pattern carries its own chances of inheritance (Raymond, CMT Inheritance Patterns 2021). In order to know the chances a CMTer has for passing on their CMT to their children, especially when there is no established family history of CMT, the underlying responsible genetic mutation must be identified. Without knowing the genetic cause, understanding the chances of inheritance is guess work (albeit a highly educated and informed guess), and this can impact family planning. A carefully reviewed detailed family history, however, can reveal a likely inheritance pattern when the investigator has the specialized training needed to understand and recognize the patterns.
While CMT is inheritable, one does not have to have inherited it in order to have it. A CMTer whose CMT was not inherited from a parent has CMT that is caused by a spontaneous and random gene mutation that occurred at or shortly after conception. CMT that was not inherited from a parent is called a de novo, or new case. A CMTer who has a de novo CMT case has the same chances of passing on their CMT as though they had inherited it, and those chances are solely dependent on the underlying CMT-causing genetic mutation.
CMT can be extremely difficult for doctors to diagnose, especially when there’s no established family history. The doctor has to consider the totality of all diagnostic findings. Despite gains in CMT genetic discovery and gains in commercially available genetic testing, a genetic test might not confirm a CMT clinical diagnosis. When genetic testing does not reveal a genetic cause, having an appropriate clinical picture of CMT symptoms, together with appropriate nerve study findings, and having family members who also fit the same clinical picture, can lead to a definitive CMT clinical diagnosis after exhausting all causes of acquired neuropathy (such as diabetes) and after exhausting all causes of otherwise treatable neuropathy. A CMT clinical diagnosis is definitive insofar as serving as a reliable diagnosis for demyelinating CMT—CMT1, or axonal CMT—CMT2, but a clinical diagnosis does not infer a genetic cause, i.e., the specific subtype. Identifying the specific subtype requires genetic testing.
CMT genetic testing is required for identifying the specific CMT subtype, but the testing has inherent limits. By this, the test might not identify a CMT genetic cause. A test that does not overtly identify a CMT-causing mutation when a CMTer has the appropriate clinical picture and nerve study findings supportive of a CMT clinical diagnosis means only that the underlying cause was not yet tested for, and the result means nothing more. Genetic test results in this situation specifically do not rule out CMT. In this situation, a CMT clinical diagnosis that prompted a genetic test still stands, is still a reliable diagnosis, and is still a definitive diagnosis when all other causes of neuropathy have been exhausted, even though the exact subtype remains elusive.
About the Author
Kenneth Raymond is a CMTer who was first diagnosed with CMT1 in late 2002, at the age of 29. He was genetically confirmed to have CMT1A a year later. Kenneth has devoted his life since diagnosis to studying, researching, and learning all things CMT, with an emphasis on the genetics of CMT as they relate to everyday CMTers. As a member of the Charcot-Marie-Tooth Association’s Advisory Board, Kenneth is a CMT genetics expert and is a CMT advocate who is committed to raising CMT awareness through fact-based information rooted in the latest understandings of CMT.
Anderson, S. L., Coli, R., Daly, I. W., Kichula, E. A., Rork, M. J., Volpi, S. A., Ekstein, J., & Rubin, B. Y. 2001. "Familial dysautonomia is caused by mutations of the IKAP gene." American journal of human genetics 68 (3): 753–758. doi:https://doi.org/10.1086/318808
Bergoffen, J., Scherer, S. S., Wang, S., Scott, M. O., Bone, L. J., Paul, D. L., Chen, K., Lensch, M. W., Chance, P. F., & Fischbeck, K. H. 1993. "Connexin mutations in X-linked Charcot-Marie-Tooth disease." Science 262 (5142): 2039–2042. doi:https://doi.org/10.1126/science.826610
Bolino, A., Muglia, M., Conforti, F. L., LeGuern, E., Salih, M. A., Georgiou, D. M., Christodoulou, K., Hausmanowa-Petrusewicz, I., Mandich, P., Schenone, A., Gambardella, A., Bono, F., Quattrone, A., Devoto, M., & Monaco, A. P. 2000. "Charcot-Marie-Tooth type 4B is caused by mutations in the gene encoding myotubularin-related protein-2." Nature genetics 25 (1): 7–19. doi:https://doi.org/10.1038/75542
Bomont, P., Cavalier, L., Blondeau, F., Ben Hamida, C., Belal, S., Tazir, M., Demir, E., Topaloglu, H., Korinthenberg, R., Tüysüz, B., Landrieu, P., Hentati, F., & Koenig, M. 2000. "The gene encoding gigaxonin, a new member of the cytoskeletal BTB/kelch repeat family, is mutated in giant axonal neuropathy." Nature genetics 26 (3): 370–374. doi:https://doi.org/10.1038/81701
Chance, P. F., Alderson, M. K., Leppig, K. A., Lensch, M. W., Matsunami, N., Smith, B., Swanson, P. D., Odelberg, S. J., Disteche, C. M., & Bird, T. D. 1993. "DNA deletion associated with hereditary neuropathy with liability to pressure palsies." Cell 72 (1): 143–151. doi:https://doi.org/10.1016/0092-8674(93)90058-x
Clinical Genome Resource. 2020. Clinical Genome Resource. October. Accessed August 2021. https://search.clinicalgenome.org/kb/gene-validity/CGGV:assertion_ae6ffeae-9609-47bf-89a9-4863cf6ef05c-2020-10-05T161930.420Z
Cortese, A., Zhu, Y., Rebelo, A. P., Negri, S., Courel, S., Abreu, L., Bacon, C. J., Bai, Y., Bis-Brewer, D. M., Bugiardini, E., Buglo, E., Danzi, M. C., Feely, S., Athanasiou-Fragkouli, A., Haridy, N. A., Inherited Neuropathy Consortium, Zuchner, S. 2020. "Biallelic mutations in SORD cause a common and potentially treatable hereditary neuropathy with implications for diabetes." Nature genetics 52 (5): 473–481. doi:https://doi.org/10.1038/s41588-020-0615-4
Dawkins, J. L., Hulme, D. J., Brahmbhatt, S. B., Auer-Grumbach, M., & Nicholson, G. A. 2001. "Mutations in SPTLC1, encoding serine palmitoyltransferase, long chain base subunit-1, cause hereditary sensory neuropathy type I." Nature genetics 27 (3): 309–312. doi:https://doi.org/10.1038/85879
El-Abassi, R., England, J. D., & Carter, G. T. 2014. "Charcot-Marie-Tooth disease: an overview of genotypes, phenotypes, and clinical management strategies." PM & R : the journal of injury, function, and rehabilitation 6 (4): 342–355. Accessed November 2021. doi:https://doi.org/10.1016/j.pmrj.2013.08.611
Gondim, Francisco de Assis Aquino MD, PhD, MSc, FAAN, and Florian P MD, PhD, MA, MS Thomas. 2019. What is the role of EMG and NCS in the workup of Charcot-Marie-Tooth (CMT) disease? Edited by PharmD, PhD Francisco Talavera. February. Accessed November 2021. https://www.medscape.com/answers/1173484-176746/what-is-the-role-of-emg-and-ncs-in-the-workup-of-charcot-marie-tooth-cmt-disease
Grohmann, K., Schuelke, M., Diers, A., Hoffmann, K., Lucke, B., Adams, C., Bertini, E., Leonhardt-Horti, H., Muntoni, F., Ouvrier, R., Pfeufer, A., Rossi, R., Van Maldergem, L., Wilmshurst, J. M., Wienker, T. F., Sendtner, M., Rudnik-Schöneborn, et al. 2001. "Mutations in the gene encoding immunoglobulin mu-binding protein 2 cause spinal muscular atrophy with respiratory distress type 1." Nature genetics 29 (1): 75–77. doi:https://doi.org/10.1038/ng703
Guilbot, A., Williams, A., Ravisé, N., Verny, C., Brice, A., Sherman, D. L., Brophy, P. J., LeGuern, E., Delague, V., Bareil, C., Mégarbané, A., & Claustres, M. 2001. "A mutation in periaxin is responsible for CMT4F, an autosomal recessive form of Charcot-Marie-Tooth disease." Human molecular genetics 10 (4): 415–421. doi:https://doi.org/10.1093/hmg/10.4.415
Hayasaka, K., Himoro, M., Sato, W., Takada, G., Uyemura, K., Shimizu, N., Bird, T.D., Conneally, P.M., and Chance, P.F. 1993. "Charcot-Marie-Tooth neuropathy type 1B is associated with mutations of the myelin P0 gene." Nature Genetics 5 (1): 31-34. doi:https://doi.org/10.1038/ng0993-31
Indo, Y., Tsuruta, M., Hayashida, Y., Karim, M. A., Ohta, K., Kawano, T., Mitsubuchi, H., Tonoki, H., Awaya, Y., & Matsuda, I. 1996. "Mutations in the TRKA/NGF receptor gene in patients with congenital insensitivity to pain with anhidrosis." Nature genetics 13 (4): 485–488. doi:https://doi.org/10.1038/ng0896-485
2021. Invitae Comprehensive Neuropathies Panel. May 03. https://www.invitae.com/en/physician/tests/03200/
Kalaydjieva, L., Gresham, D., Gooding, R., Heather, L., Baas, F., de Jonge, R., Blechschmidt, K., Angelicheva, D., Chandler, D., Worsley, P., Rosenthal, A., King, R. H., & Thomas, P. K. 2000. "N-myc downstream-regulated gene 1 is mutated in hereditary motor and sensory neuropathy-Lom." American journal of human genetics 67 (1): 47–58. doi:https://doi.org/10.1086/302978
Mastaglia, F. L., Nowak, K. J., Stell, R., Phillips, B. A., Edmondston, J. E., Dorosz, S. M., Wilton, S. D., Hallmayer, J., Kakulas, B. A., & Laing, N. G. 1999. "Novel mutation in the myelin protein zero gene in a family with intermediate hereditary motor and sensory neuropathy." Journal of neurology, neurosurgery, and psychiatry 67 (2): 174–179. doi:https://doi.org/10.1136/jnnp.67.2.174 <