The centerpiece of Lévy’s reputation is a 1926 paper published in the Revue Neurologique, co-authored with Roussy, under a title that translates as Seven cases of a particular familial disease: gait disturbances, clubfeet and generalized tendon areflexia, with, incidentally, slight clumsiness of the hands. The seven patients, drawn from a single family, displayed a striking and reproducible combination of features. They walked with difficulty, a problem the authors captured in the term dystasia, and they had congenital or early-onset deformities of the feet. Most tellingly, their tendon reflexes were absent throughout, even though the clinical picture was far milder than that of the better-known hereditary ataxia of Friedreich, with which it was immediately compared.

The disorder they described became known as Roussy-Lévy syndrome, and its definition was a triumph of careful phenotyping. At a time when the tools of neurology were the hammer, the tuning fork, the ophthalmoscope and the clinical eye, Lévy and Roussy recognized that a syndrome could be defined by a constellation of signs rather than by any single finding. The combination of areflexia, pes cavus and other foot deformities, distal weakness and an essentially static or slowly progressive course distinguished their patients from those with Friedreich’s ataxia, who typically show pyramidal signs, scoliosis and cardiomyopathy, and from other hereditary neuropathies described in the same era. The 1926 paper stands as an early exercise in what would now be called genotype-phenotype correlation, conducted entirely at the bedside.

Lévy did not stop at description. In 1934, the year of her death, she and Roussy published a follow-up paper on hereditary areflexic dystasia framed as a contribution to the study of the genesis of familial diseases and their kinship with one another. That subtitle matters. The authors were explicitly wrestling with the question of how related clinical entities fit together, whether the syndrome they had described was a disease in its own right, a fruste or abortive form of Friedreich’s ataxia, or part of a broader family of hereditary degenerations. This was a conceptual debate that ran through French neurology of the period, and the 1933 paper by Darré, Mollaret and Landowski, which asked directly whether Roussy-Lévy disease was merely a mild form of Friedreich’s disease using labyrinthine and chronaxie examinations, shows how contested the boundaries remained.

The question was not settled for decades, and the eventual answer vindicated the distinctness of the syndrome in a way neither Lévy nor her contemporaries could have imagined. In 1956, Joseph Lapresle traced and re-examined four of the seven original patients, thirty years after the first publication, confirming the persistence of the phenotype and the remarkable stability of the clinical picture across the decades. Then, in 1973, Lapresle and Salisachs reported a finding that moved the debate from the clinic to the laboratory: a nerve biopsy specimen from one of the original cases of Roussy-Lévy disease showed onion bulb formations, the whorled proliferations of Schwann cell processes that are the pathological signature of a demyelinating and remyelinating hereditary neuropathy. The syndrome was, at its core, a disease of the peripheral nerves.

The molecular coda arrived at the end of the twentieth century. In 1998, Auer-Grumbach and colleagues demonstrated that the Roussy-Lévy syndrome is a phenotypic variant of Charcot-Marie-Tooth disease type 1A, associated with the characteristic duplication on chromosome 17p11.2 that underlies the commonest form of hereditary motor and sensory neuropathy. A year later, Planté-Bordeneuve and colleagues, in a paper pointedly titled The Roussy-Lévy family: from the original description to the gene, closed the loop by studying the descendants of the original kindred and confirming the genetic diagnosis. A clinical picture sketched in 1926 from seven patients and a family tree had been anchored, seventy-three years later, to a specific segment of the human genome.

That trajectory, from bedside description to chromosomal duplication, is precisely why historians of neuroscience find Lévy worth recovering. Her work illustrates how the great nosological debates of interwar neurology, over whether syndromes were separate diseases or variants of one another, were resolved not by better rhetoric but by accumulated follow-up data and, eventually, by molecular genetics. The Roussy-Lévy syndrome is now understood as part of the demyelinating Charcot-Marie-Tooth spectrum, yet its recognition as a distinct phenotype in 1926 helped clinicians for decades to counsel families, predict prognosis and distinguish patients whose disease would remain relatively static from those facing the progressive disability of Friedreich’s ataxia.

Lévy’s scientific range extended beyond hereditary neuropathy. In 1925 she published a monograph on the late manifestations of epidemic encephalitis, the prolonged and delayed forms of the disease that swept Europe in the wake of the First World War and left many patients with parkinsonian and other sequelae years after the acute infection. The work places her among the clinicians who grappled with encephalitis lethargica, one of the great unsolved puzzles of the era, and shows a career that engaged with both the degenerative and the infectious frontiers of neurology. Contemporaries remembered her within the vibrant Parisian school that included figures such as Jean Lhermitte, and the historical literature, notably Peter Koehler’s 2018 study of Lévy and the Roussy-Lévy syndrome in the Journal of the History of the Neurosciences, has begun to piece together the details of her short but productive career.

Her early death in 1934, at the age of forty-eight, cut short a career that had already produced two landmark papers on hereditary areflexic dystasia and a substantial monograph on post-encephalitic disease. It also helps explain why her name faded. Eponyms tend to attach to those who survive to promote, extend and defend their syndromes, and Lévy had no such opportunity. The new biographical account, written by Carlos Alvarado-De la Hoz, Andres Felipe Cardenas-Cruz and Alba Lucia Marentes-Cubillos, neurologists at the Hospital Universitario San Ignacio and the Pontificia Universidad Javeriana in Bogotá, forms part of a deliberate effort by the Journal of Neurology to document the pioneers whose contributions underpin modern practice, and it appeared in volume 273 of the journal as article number 601.

For a modern readership, the story carries a broader lesson about how medicine accumulates knowledge. The seven patients described in 1926 were examined with none of the instruments that would eventually explain their condition, yet the precision of the original observation was good enough that a biopsy in 1973 and a gene study in 1999 could be anchored to those very cases. Clinical description, often dismissed as the antiquarian wing of neuroscience, proved to be durable data. Gabrielle Lévy’s legacy is a reminder that the careful documentation of a single family, done well, can outlast the career and even the life of the observer, and that some of the most consequential figures in the history of the neurosciences are those whose names survive only in the hyphen of a syndrome they never lived to see explained.

Subject of Research: Historical biography of neurologist Gabrielle Lévy and the Roussy-Lévy syndrome

Article Title: Gabrielle Lévy (1886–1934)

Article References: Alvarado-De la Hoz, C., Cardenas-Cruz, A. F., & Marentes-Cubillos, A. L. (2026). Gabrielle Lévy (1886–1934). Journal of Neurology, 273(10), Article 601. https://doi.org/10.1007/s00415-026-14135-y

Image Credits: AI Generated

DOI: 10.1007/s00415-026-14135-y

Keywords: Gabrielle Lévy, Roussy-Lévy syndrome, hereditary areflexic dystasia, Charcot-Marie-Tooth disease, Friedreich's ataxia, neurology history, hereditary neuropathy, peripheral nerve, genetics, Pioneers in Neurology, Journal of Neurology, medical eponyms

The Marseille team, led by Etienne Fortanier and Shahram Attarian, combed through the records of 642 patients tested after their fiftieth birthday at their referral center. Fifty-seven of them, or 8.9 percent, met the study’s strict definition of late-onset CMT: symptoms beginning after age fifty combined with a class 5 pathogenic variant, the highest confidence category in the American College of Medical Genetics and Genomics framework for interpreting sequence variants. That threshold matters. By demanding both a late clinical onset and an unambiguously disease-causing mutation, the investigators filtered out the incidental genetic noise that often complicates interpretation of sequencing results in older populations, where variants of uncertain significance are common.

The clinical picture that emerged was strikingly mild compared with classic early-onset CMT. High arches, known as pes cavus, were present in 80.7 percent of the confirmed cases, and a family history of neuropathy appeared in 63.2 percent. Yet disability remained modest across the group, measured with validated instruments such as the Charcot–Marie–Tooth Neuropathy Score. This mildness is precisely what makes late-onset CMT so easy to miss. A patient who develops slowly progressive numbness and balance trouble at fifty-eight, without dramatic muscle wasting, is far more likely to be worked up for an acquired neuropathy than to be sent for hereditary testing, particularly when the family history is small or incomplete.

The genetic landscape revealed by the sequencing was remarkably diverse. Pathogenic variants were distributed across twenty different genes, a breadth that would have been practically impossible to capture under the old candidate-gene testing paradigm, in which laboratories screened one gene at a time based on the presumed inheritance pattern and nerve conduction profile. The single most common finding was the PMP22 duplication, the canonical cause of the demyelinating CMT1A subtype, present in 32 percent of cases. But a notable share of diagnoses involved genes only recently linked to late-onset phenotypes, including MME, which encodes the membrane metallo-endopeptidase neprilysin, LRSAM1, an E3 ubiquitin ligase involved in autophagy, MPZ, which encodes myelin protein zero, and MFN2, the mitofusin gene that governs mitochondrial fusion.

The prominence of MME and LRSAM1 in this cohort reinforces a growing international consensus that axonal neuropathies beginning in middle and old age carry a distinct genetic signature. Earlier studies from Japan, Italy, France, and the United Kingdom have pointed in the same direction, and whole-genome sequencing efforts have shown that broadening the genetic net raises diagnostic yields substantially. What the Marseille study adds is a decade of single-center consistency: the same referral service, the same testing pipeline, and the same clinical assessment framework applied uniformly across ten years, minimizing the referral and methodological heterogeneity that can distort multi-center surveys.

To understand what distinguishes genuine late-onset CMT from the many other neuropathies that afflict older adults, the researchers assembled a control group twice the size of the case series: 114 age- and sex-matched patients who had negative CMT-targeted sequencing and a confirmed alternative diagnosis. The main alternative diagnoses in this control group were inflammatory neuropathies, including chronic inflammatory demyelinating polyradiculoneuropathy, idiopathic axonal neuropathies, and toxic neuropathies. These are precisely the entities that dominate the differential diagnosis of progressive distal weakness and sensory loss after fifty, and misattributing hereditary disease to one of them carries real consequences, from unnecessary immunoglobulin therapy to missed genetic counseling for adult children who may carry the same variant.

Multivariate analysis distilled the distinction to three independent predictors. A family history of neuropathy, the presence of pes cavus, and a median motor nerve conduction velocity below 38 meters per second were each independently associated with late-onset CMT rather than an acquired alternative. The conduction threshold is technically meaningful: velocities below 38 meters per second in the median nerve conventionally demarcate demyelinating from axonal physiology, and their association with genetically confirmed CMT in this late-onset cohort indicates that even patients whose disease surfaces late often retain the electrophysiological fingerprint of their underlying molecular lesion. In practical terms, a neurologist evaluating an older patient with an unexplained neuropathy now has a compact, bedside-derivable triad that should trigger referral for comprehensive genetic testing.

The implications extend beyond diagnosis into the design of clinical trials. As gene-specific therapies for CMT edge closer to the clinic, natural history data on late-onset populations become essential, because the pace of progression, the burden of disability, and the range of causative genes all differ from the pediatric and young-adult cohorts in which most historical natural history studies were conducted. A trial designed around early-onset CMT1A assumptions may be poorly suited to a sixty-year-old with an MME-related axonal neuropathy that has progressed imperceptibly for a decade. Recognizing that nearly a tenth of tested patients over fifty harbor a genetic diagnosis means trial recruiters, outcome measure developers, and regulatory strategists must account for this older, genetically heterogeneous population.

There are also counseling dimensions that the study only begins to address. Each of the twenty implicated genes carries its own inheritance pattern, penetrance profile, and recurrence risk for offspring. An autosomal dominant MFN2 variant identified in a fifty-four-year-old has immediate implications for siblings and children, some of whom may already be experiencing subtle symptoms they have not connected. Conversely, identifying a recessive MME variant can reassure relatives and reshape reproductive planning. The Marseille findings argue for embedding genetic counselors within neuromuscular clinics that serve older adults, rather than treating hereditary neuropathy as a pediatric subspecialty concern.

The study’s limitations are those inherent to its design. It is retrospective, drawn from a single highly specialized referral center, which likely inflates the proportion of genetically confirmed cases relative to community practice, where testing rates remain low and referral thresholds high. The 8.9 percent figure should therefore be read as a prevalence among tested patients, not among all older adults with neuropathy. Still, the message travels well beyond Marseille: in the next-generation sequencing era, the age cutoffs that once guided who deserved hereditary testing have lost their justification. A family history, high arches, and slowed median nerve conduction in a patient of any age should now be enough to open the door to a comprehensive gene panel, because the genetic causes of neuropathy clearly do not respect a fiftieth birthday.

Subject of Research: Genetically confirmed late-onset Charcot–Marie–Tooth neuropathy diagnosed by next-generation sequencing in patients over 50

Article Title: Late-onset CMT neuropathy in the NGS era: a 10-year retrospective study

Article References: Fortanier, E., Bonello-Palot, N., Verschueren, A., Kouton, L., Grapperon, A.-M., Salort-Campana, E., Corazza, G., de Bovis, V. M., du Closel, L. B., Delmont, E., & Attarian, S. (2026). Late-onset CMT neuropathy in the NGS era: a 10-year retrospective study. Journal of Neurology, 273(10), Article 624. https://doi.org/10.1007/s00415-026-14154-9

Image Credits: AI Generated

DOI: 10.1007/s00415-026-14154-9

Keywords: Charcot-Marie-Tooth disease, hereditary neuropathy, next-generation sequencing, late-onset disease, PMP22 duplication, MME, LRSAM1, MPZ, MFN2, pes cavus, nerve conduction velocity, genetic diagnosis