Progressive Myoclonic Epilepsies: Unverricht-Lundborg Disease and Lafora Disease

Literature and regulatory information updated: 31 October 2026.

0. Initial Reference Table

The progressive myoclonic epilepsies (also termed progressive myoclonus epilepsies, PME) are a group of disorders defined by the combination of stimulus-sensitive myoclonus, epileptic seizures and progressive neurological deterioration [1, 2]. The following table summarises the two entities at the core of this article, Unverricht-Lundborg disease and Lafora disease, and the class to which they belong; the remaining members of the class are listed in Table 2 (section 2) and in the differential diagnosis table (section 6).

The other PMEs with their own subsection in this article — action myoclonus–renal failure syndrome, KCTD7-related PME, North Sea PME, MEAK and EPM1B — are summarised in the following table, with one row per entity.

1. Introduction

1.1 Definition of the class

The term progressive myoclonic epilepsy denotes a heterogeneous set of rare disorders, mostly autosomal recessive, that share the triad of stimulus-sensitive myoclonus, grand mal seizures and absences, and progressive neurological deterioration or regression [1, 41]. The International League Against Epilepsy (ILAE) classification of electroclinical syndromes includes the PMEs among the syndromes of adolescence and adulthood [42]. The phenotype arises mainly in late childhood, adolescence and young adulthood, and differs from other metabolic epilepsies of early childhood, whose dominant expression is epileptic encephalopathy, infantile spasms or neonatal myoclonic epilepsy [6]. In most PMEs dementia and ataxia coexist; Unverricht-Lundborg disease is the best-known exception, because cognition remains close to normal for decades [3, 6]. The distinction between "metabolic" PMEs (Lafora disease, neuronal ceroid lipofuscinoses, sialidosis type I, MERRF, Gaucher disease type 3, spinal muscular atrophy with PME due to ASAH1, ceramide synthase deficiencies, SCARB2-related action myoclonus–renal failure syndrome) and "non-metabolic" PMEs (Unverricht-Lundborg disease, dentatorubral-pallidoluysian atrophy, neuroserpinosis, KCNC1-related disorders) is used in the Saudubray classification and remains of practical value in choosing tests [6].

1.2 Historical overview

Unverricht (1891, 1895) and Lundborg (1903) described the form now named after them in families from Estonia and eastern Sweden; the disease was later found frequently in Finland, with onset at around 10 years of age, progressive disabling myoclonus, mild mental deterioration and absence of inclusion bodies in the brain [8]. In 1983, Eldridge and colleagues studied 15 North American families (27 patients) with so-called "Baltic myoclonus" and showed that faster progression than in the Baltic series was associated with phenytoin use, with marked improvement in at least eight patients after valproate was introduced and phenytoin reduced or withdrawn; the same work stressed the clinical distinction from Lafora body disease, which was invariably fatal [43]. Lafora disease was described by Lafora and Glueck in 1911; Gonzalo Rodríguez-Lafora (1886–1971), a Madrid neuropathologist trained under Ramón y Cajal, gave his name to the intraneuronal inclusions and to the disease [4]. The modern history of the class begins in 1989 with the Marseille workshop, whose consensus statement (1990) defined the PME types then known and opened the phase of genetic research [44, 45]. The same group observed that patients with so-called Ramsay Hunt syndrome could not be distinguished clinically from patients with Unverricht-Lundborg disease [8].

1.3 Milestones of molecular discovery

Lehesjoki and colleagues mapped the EPM1 locus to the distal long arm of chromosome 21 in 1991, in 12 Finnish families [8]. In 1996, Pennacchio and colleagues identified CSTB as the responsible gene, finding reduced mRNA and two point mutations (a splicing mutation and a stop codon mutation) in patients [46]. Because point mutations explained only 14% of 58 unrelated alleles, Lalioti and colleagues searched for another mechanism and in 1997 described the expansion of a dodecamer located about 70 nucleotides upstream of the CSTB transcription start site, with more than 60 copies in mutant alleles versus 2–3 in normal alleles, and intermediate alleles of 12–17 repeats that were unstable on transmission [47]. In Lafora disease, Minassian and colleagues cloned the EPM2A gene in 1998 by positional cloning at 6q24; it encodes a protein with a tyrosine phosphatase signature, laforin [48]. The second gene, NHLRC1, which encodes the E3 ubiquitin ligase malin, was identified in 2003 and accounts for about half of cases [2]. That year, Chan and colleagues mapped the EPM2B locus to a 2.2 Mb region at 6p22 by linkage analysis in four consanguineous French-Canadian families with the classic form (maximum lod score 5.2, with the same nine-marker haplotype in all) and subsequently identified 17 different NHLRC1 variants in 26 of 34 probands (eight deletions, one insertion, seven missense and one nonsense; 18 homozygous and eight compound heterozygous families) [10]. Since then, PMEs numbered up to EPM12 have been described in OMIM, the most recent associated with the SLC7A6OS gene [5].

1.4 Nomenclature

GeneReviews discourages the toponyms "Baltic myoclonus" and "Mediterranean myoclonus", because Unverricht-Lundborg disease occurs worldwide [3]. OMIM classifies the CSTB-related form as EPM1A (#254800) and the PRICKLE1-related form as EPM1B (#612437), a distinct disorder with early ataxia and limitation of upward vertical gaze [7, 39]. In Lafora disease, OMIM separated the EPM2A-related form (myoclonic epilepsy of Lafora 1, MELF1, #254780) from the NHLRC1-related form (MELF2, #620681, an entry created in January 2024, with the synonyms "EPM2B" and "Lafora disease 2"), retaining the historical designations EPM2A and EPM2B as synonyms [2, 4, 10]. The PRDM8-related form (?EPM10, #616640) is termed early-onset Lafora body disease, although Lafora bodies are found in muscle and not in sweat glands [2, 49].

1.5 Entities in this article

Unverricht-Lundborg disease is the autosomal recessive PME caused by biallelic loss of function of CSTB, mainly through homozygous expansion of the dodecamer, with onset between 6 and 15 years of age, action myoclonus and tonic-clonic seizures, ataxia that appears years later and mostly normal cognition [3]. Lafora disease is the autosomal recessive PME caused by biallelic variants in EPM2A or NHLRC1, characterised by the accumulation of insoluble polyglucosans (Lafora bodies) and by onset between 8 and 19 years of age, occipital seizures, myoclonus, early cognitive decline and death, as a rule, within the following decade [2, 50]. According to the review by Ferlazzo and colleagues, they are the two most common PMEs, followed by the neuronal ceroid lipofuscinoses, MERRF and sialidosis, and the PMEs as a whole account for less than 1% of all epilepsies [17]. Sialidosis, MERRF, the neuronal ceroid lipofuscinoses and Gaucher disease are compared here only in the differential diagnosis.

In addition to these two diseases, this article covers in dedicated subsections five rare PMEs that share their differential diagnosis and symptomatic treatment: action myoclonus–renal failure syndrome (AMRF, EPM4, SCARB2), KCTD7-related PME (EPM3, also termed CLN14), North Sea PME (EPM6, GOSR2), MEAK (EPM7, KCNC1) and EPM1B (PRICKLE1) [7, 14, 15, 31, 39]. Spinal muscular atrophy with PME (ASAH1), galactosialidosis and GM1 gangliosidosis appear here only in the differential.

2. Molecular Genetics and Pathophysiology

2.1 Genetic classification of the class

The current classification combines OMIM numbering (EPM1A to EPM12), the eponymous entities and the pathogenic mechanism (lysosomal, glycogen metabolism, complex lipids, ion channels, vesicular trafficking, repeat expansions and mitochondrial DNA) [6, 8, 51]. Table 2 brings together the types with a recognised PME phenotype in the sources consulted.

2.2 Unverricht-Lundborg disease: the CSTB gene and cystatin B

CSTB is located at 21q22.3 and encodes cystatin B (stefin B), a reversible intracellular inhibitor of papain-family cysteine proteases, mainly cathepsins B, H and L, present in the cytoplasm and nucleus and with a protective role against proteases released from lysosomes [3, 68]. In the nucleus, cystatin B interacts with histones and with cathepsin L, whose activity it may regulate [3]. In patient lymphoblastoid cell lines, reduced inhibitory activity is associated with increased activity of cathepsins B, L and S, and the protein also has an anti-apoptotic effect in cerebellar cells [45].

The mechanism is loss of function: the dodecamer expansion reduces CSTB mRNA to 5–10% of control values, and the other pathogenic variants also abolish function, some with complete absence of intracellular protein [3]. The Cstb-/- mouse reproduces the disease, with progressive ataxia, myoclonic seizures and apoptosis of cerebellar granule cells [3]. In murine models, microglial activation and brain and peripheral inflammation preceding myoclonus have been described, together with loss of cortical GABAergic terminals with reduced inhibition, and sensitisation of neurons to oxidative stress through dysregulation of the cystatin B–cathepsin B pathway [3]. Studies from 2026 added transcriptomic and proteomic data from mouse cerebellum, cortex and hippocampus (immune activation, decreased oxidative phosphorylation and lysosomal alteration) [69], oxidative stress with cell cycle arrest and apoptosis in induced pluripotent stem cells from patients [70] and, in brain organoids, altered transport of Sonic Hedgehog by extracellular vesicles with diversion of ventral progenitor specification and excitation/inhibition imbalance [71]. These are preclinical data that propose therapeutic targets, with no clinical translation as of the update date.

2.3 Spectrum of CSTB variants

The CCC-CGC-CCC-GCG dodecamer expansion is located in the promoter region and accounts for about 90% of pathogenic alleles overall and about 99% in Finland; sequence variants account for the remaining ~10% [3]. GeneReviews defines ranges of 2–3 repeats (normal), 12–17 repeats (uncertain significance, unstable alleles without a characterised phenotype) and 30 or more repeats (pathogenic, full penetrance); alleles of 4–11 and 18–29 repeats have not been observed [3]. In the notation provided by the authors of the GeneReviews chapter, the alleles are written NM_000100.4:c.-179_-190CCCCGCCCCGCG[2_3] (normal), [12_17] (uncertain) and [30_?] (full penetrance) [3]. In the Finnish national study, expansions ranged from 38 to 77 repeats [72], and in the Chinese case described in 2025 the expanded allele had at least 55 repeats [73]. There is no relevant somatic instability; contractions and expansions occur on germline transmission, with no tendency to increase over successive generations, so there is no anticipation [3].

Sequence variants occur mainly in compound heterozygosity with an expansion. Those normalised in VariantValidator (MANE transcript NM_000100.4) include c.202C>T p.(Arg68Ter), described in compound heterozygous Finnish patients, c.67-1G>C (splicing variant), c.10G>C p.(Gly4Arg) and c.116_117del p.(Lys39SerfsTer37); the latter was published with the protein notation "p.K395fs*37", which does not match the protein length and was corrected by validation [73–75]. In Italy, six of 52 families with EPM1A (11.5%) were compound heterozygous for point variants or indels: c.67-1G>C and three variants published in the authors' notation, c.168+1_18del, c.133C>T and c.168+2_169+21delinsAA [76]. In VariantValidator, the first of these three corresponds to NM_000100.4:c.168+2_168+19del (description normalised by the 3' alignment rule), whereas c.133C>T (reference base mismatch) and c.168+2_169+21delinsAA (intronic position without a corresponding exon boundary) cannot be validated on the MANE transcript and are therefore kept in their published form [74, 76].

2.4 Genotype-phenotype correlations and CSTB allelic disorders

In the Finnish study of 66 homozygous patients, the size of the longer expanded allele was independently associated with the motor threshold on transcranial magnetic stimulation and had only a modulating effect on age at onset, myoclonus severity and the cortical silent period [72]. Compound heterozygosity for an expansion and a point variant is associated with earlier onset, more severe myoclonus, drug-resistant seizures and poorer cognitive performance: five Finnish patients with c.202C>T had earlier onset than 21 patients homozygous for the expansion [75], and eight compound heterozygous Italian patients had disease onset at 7.4 ± 1.7 years, with resistant seizure clusters and wheelchair use after the age of 30 in all, although the amount of protein was similar to that in homozygotes [76]. A more severe phenotype in compound heterozygous males has been suggested, based on small numbers [3].

Biallelic complete loss-of-function variants (stop codon and frameshift) cause a distinct disorder of infantile or neonatal onset, with progressive microcephaly, severe developmental delay, dyskinesia, diffuse hypomyelination and cerebral and cerebellar atrophy [3, 77]. In 2025, five Egyptian patients homozygous for c.67-1G>C or c.10G>C, variants previously described in EPM1A, had this neonatal phenotype, and only one had seizures [77]. A Palestinian case of infantile-onset ULD with a homozygous missense variant and global delay was published in 2025 [78]. Among the modifiers studied, the ε4 allele of the APOE gene did not alter myoclonus severity or disease duration in 65 Finnish patients, but was associated with better self-reported quality of life, greater preservation of the hippocampus and amygdala and greater white matter degeneration [79].

2.5 Lafora disease: laforin, malin and glycogen

In the normal neuron there is no significant glycogen synthesis, because glycogen synthase is targeted for proteasomal degradation by the laforin-malin complex; in astrocytes, glycogen is degraded to lactate, which is exported to neurons [50]. Laforin is a dual-specificity phosphatase with a carbohydrate-binding module, which acts on glycogen and prevents its hyperphosphorylation; there are two isoforms, of which the cytoplasmic isoform A is the one relevant to the disease [2, 80, 81]. Malin, encoded by NHLRC1, a single-exon gene of 1188 base pairs, is a 395-amino-acid E3 ubiquitin ligase with a RING zinc finger and six NHL repeats; in a laforin-dependent manner it ubiquitinates the PPP1R3C and PPP1R3D subunits that target the phosphatase to glycogen, polyubiquitinates laforin and the debranching enzyme, and takes part in the clearance of aggregates by the proteasome and macroautophagy [2, 82, 83].

The current pathogenic model describes a defect in the regulation of chain length in a small fraction of glycogen: without laforin or without malin, some molecules develop excessively long chains, precipitate and aggregate into Lafora bodies, which, on reaching a density threshold (at around 14 years of age in humans), initiate and sustain progressive myoclonic epilepsy [2]. Experimental data show that the phosphatase function of laforin is dispensable for the formation of Lafora bodies (inactivation of the phosphatase domain, with the rest of the protein preserved, does not give rise to them), and that the relevant function of laforin is to direct malin to glycogen [2]. The abnormal glycogen has a high phosphate content and reduced branching, and Lafora bodies accumulate in axons and above all in dendrites (small but very numerous), in the cerebral cortex, substantia nigra, thalamus, globus pallidus and dentate nucleus [2, 50]. They are also present in muscle, liver, heart, skin and retina, without symptoms in these organs, which makes Lafora disease a generalised glycogen storage disease with neurological expression [2, 50]. A 50% reduction in glycogen synthesis abolishes the disease in murine models, which provides the rationale for therapies targeting glycogen synthase 1 (GYS1) [17, 84]. The malin-deficient mouse develops Lafora bodies at around 11 months of age, with neurodegeneration, susceptibility to kainate-induced seizures and impaired autophagy, and brain deletion of Gys1 in this model eliminated brain glycogen and prevented neurodegeneration, synaptic changes, seizure susceptibility and the autophagy defect [10]. Outside the human species, Lafora disease of the miniature wire-haired dachshund (more than 5% of animals of this breed in the United Kingdom) is caused by a biallelic dodecamer expansion in Epm2b, with mRNA reduced to less than 1/900, a parallel with the CSTB dodecamer expansion mechanism in ULD [10].

2.6 Spectrum of EPM2A and NHLRC1 variants

According to GeneReviews, each gene accounts for about 50% of cases; in the meta-analysis of 250 published cases, NHLRC1 was mutated in 56% and EPM2A in 44% [2, 85]. Saudubray attributes about 50% to EPM2A and 30–40% to NHLRC1, an older figure [50]. The proportion varies with the population studied: in the Spanish series of Gómez-Abad and colleagues (2005; 77 families), 70.1% of probands had EPM2A variants, 27.3% NHLRC1 variants and 2.6% no variant in either gene, whereas in Japan Singh and colleagues found NHLRC1 variants in five of eight families, EPM2A variants in one and none in two [10]. In EPM2A, sequencing detects 85–90% of variants and intragenic deletions or duplications account for 10–15%; in NHLRC1, sequencing detects more than 90% and deletions less than 10%, including heterozygous whole-gene deletions in one Italian and one Serbian family [2]. Biallelic intronic EPM2A variants that alter splicing cause the severe classic form [2].

GeneGuidance uses NM_001018041 as the EPM2A reference sequence; the MANE transcript confirmed in VariantValidator is NM_005670.4 [74, 80]. Epm2b P71A and D148N knock-in mice, corresponding to the human variants p.(Pro69Ala) and p.(Asp146Asn), have been created, with distinct neurological abnormalities [87].

2.7 Genotype-phenotype correlations in Lafora disease

Most variants cause complete loss of function, and the phenotypes associated with EPM2A and NHLRC1 are indistinguishable in the classic form [2]. Two earlier series, however, described a slower course in the NHLRC1-related form: Gómez-Abad and colleagues observed, in 21 patients with NHLRC1 variants compared with 70 patients from 54 families with EPM2A variants, a slightly longer disease duration and later death, with two patients reaching the fourth decade; Singh and colleagues (13 patients with NHLRC1 and 22 with EPM2A, similar age at onset, about 12 years) recorded the need for respiratory support on average 20 years and 6.5 years after onset, respectively, although cognitive decline, ataxia and spasticity appeared 2–4 years after onset in both groups [10]. OMIM summarises this difference as slower progression and later death in MELF2 [10], but the 2021 meta-analysis found no relationship between the affected gene and survival [88]. In the 2023 meta-analysis (250 cases from 70 articles, 114 variants), the NHLRC1 truncating/truncating genotype was associated with shorter survival (hazard ratio [HR] 2.88; 95% confidence interval [95% CI] 1.23–6.78) and a trend towards greater loss of autonomy (HR 2.03; 95% CI 0.75–5.56), and homozygosity for p.(Asp146Asn) was confirmed to be associated with a longer course [85]. In EPM2A, Ganesh and colleagues (22 patients from 14 families) distinguished the classic form, with onset in adolescence and associated mainly with exon 4 variants (phosphatase domain), from an atypical form with dyslexia and learning difficulties in childhood before the epilepsy, associated mainly with exon 1 variants (carbohydrate-binding domain) [4]. The PRDM8-related form, described in two families, begins at around 5 years of age and has a much more protracted course; the PRDM8 protein interacts with laforin and malin [2, 50].

2.8 Inheritance

Both diseases are autosomal recessive, and heterozygotes are asymptomatic [2, 3]. The sources consulted do not describe a second inheritance pattern for CSTB, EPM2A or NHLRC1; within the class, dominant inheritance is restricted to KCNC1, SEMA6B, DNAJC5 (CLN4) and ATN1, and maternal inheritance to MERRF [7, 15, 16, 56].

2.9 Action myoclonus–renal failure syndrome: SCARB2 and LIMP-2

The SCARB2 gene encodes lysosomal integral membrane protein 2 (LIMP-2, or LMP2), a receptor that binds β-glucocerebrosidase in the endoplasmic reticulum and delivers it to lysosomes through a mannose-6-phosphate receptor-independent pathway [14]. Loss of SCARB2 function leads to mistargeting and lysosomal depletion of the enzyme; because GBA1 variants cause neurological phenotypes in both the homozygous and heterozygous states, secondary glucocerebrosidase deficiency is considered one of the drivers of the disease, and LIMP-2 probably has other functions, possibly in cholesterol transport [14, 89]. In fibroblasts from the two Portuguese sisters described by Balreira and colleagues, β-glucosidase activity was 10% of normal, with the enzyme retained in the endoplasmic reticulum, whereas leucocyte activity was normal, which makes the leucocyte assay inadequate to exclude the disease [21, 90]. Neuropathology showed extraneuronal accumulation of autofluorescent pigmented granules in the cortex, globus pallidus and cerebellum, and kidney biopsies showed collapsing glomerulopathy, a severe variant of focal segmental glomerulosclerosis [23]. Sequencing detects 100% of known variants, and no deletions or duplications have been described [14].

GeneReviews identifies no clinically useful genotype-phenotype correlations, except late onset with p.(Gly462AspfsTer34) [14]. The review by Atasu and colleagues concluded that the variant type does not materially alter the course, but that a C-terminal location is associated with later onset, particularly with that variant [91]. No Mendelian allelic disorders are known; SCARB2 variants have been associated with Parkinson disease risk in association studies and with modification of Gaucher disease [14, 24, 91].

2.10 KCTD7-related PME

KCTD7 encodes a 289-amino-acid protein with a BTB/POZ domain, belonging to the family of potassium channel tetramerisation domain-containing proteins; it does not conduct potassium, but regulates conductance through interactions with Kv subunits, and takes part in transcriptional regulation, cytoskeletal organisation and the autophagy-lysosome pathways [7, 54]. It is expressed after birth throughout the central nervous system; loss of function depolarises the resting potential and increases excitability [7]. Sequencing detects more than 99% of variants, and intragenic deletions or duplications account for less than 1% [7]. The first variant, homozygous in a Moroccan family, was NM_153033.5:c.295C>T p.(Arg99Ter); NM_153033.5:c.550C>T p.(Arg184Cys) was described in two Mexican siblings with lipofuscinosis-type storage material in fibroblasts and lymphocytes, a finding that gave rise to the designation CLN14 [25]. The c.533C>T p.(Ala178Val) variant recurs in Italian and Indian patients, but has a conflicting classification in ClinVar [25, 92]. There is no consistent genotype-phenotype correlation; developmental delay followed by progressive regression is more likely with variants in the BTB/POZ domain [7]. Inclusions are not a constant finding: none of the four biopsied patients in the 2024 cohort had them [7]. Opsoclonus-myoclonus-ataxia syndrome and paroxysmal non-epileptic myoclonus-dystonia, described in a few patients, may represent variable expressivity rather than distinct allelic disorders [7].

2.11 North Sea PME: GOSR2

GOSR2 encodes a Golgi SNARE protein (GS27, membrin) involved in vesicle fusion in trafficking between the endoplasmic reticulum and the Golgi [31, 52]. Saudubray also classifies it among the congenital disorders of glycosylation due to defective vesicular trafficking (GOSR2-CDG), with involvement of the brain and peripheral nervous system, and the 2025 review describes glycosylation defects in several compound heterozygotes [31, 93]. The founder variant NM_004287.5:c.430G>T p.(Gly144Trp) lies in the SNARE domain, destabilises its hydrophobic core and causes partial loss of function; mutant GOSR2 retains its cis-Golgi localisation [31]. Haplotype analysis by Corbett and colleagues dated the common ancestor, probably of European origin, to about 3600 years ago [29]. By August 2024, 42 patients with 11 variants had been published: 29 homozygous for c.430G>T, and the remainder compound heterozygous or homozygous for rare variants, such as c.336+1G>A, c.491_493del p.(Lys164del) (in ClinVar, repeat notation c.485AGA[2]) and c.337-12T>A [31]. The correlation described is as follows: variants in exons 1 and 4 are associated mainly with congenital muscular dystrophy, because they affect the muscle isoform, and those in exon 5 mainly with the PME phenotype; start codon variants differ from each other, p.(Met1Leu) with isolated deafness and p.(Met1Arg) with muscular dystrophy and refractory epilepsy [31]. In the fruit fly, reduction of membrin in glial cells alone reproduces heat-triggered seizures, which suggests primary glial dysfunction and helps to explain worsening with fever [31].

2.12 MEAK: KCNC1 and the Kv3.1 channel

KCNC1 encodes the Kv3.1 subunit of voltage-gated potassium channels, expressed in fast-spiking GABAergic interneurons, whose rapid activation and rapid deactivation allow high-frequency firing [15]. The NM_001112741.2:c.959G>A p.(Arg320His) variant alters the fourth arginine of the S4 voltage sensor and causes loss of function with a dominant negative effect; all published patients with MEAK have this variant, and all carriers of this variant have MEAK [15, 35]. Other variants in the same gene cause different phenotypes: c.1262C>T p.(Ala421Val), recurrent and with severe loss of function, causes developmental and epileptic encephalopathy; gain-of-function variants, such as c.1273G>A p.(Val425Met), are associated with encephalopathy or with developmental delay and hypotonia, without myoclonus; c.1538C>T p.(Ala513Val) has been described in one case of epilepsy of infancy with migrating focal seizures [15]. GeneReviews writes p.Val245Met in a footnote and p.Val425Met in the variant table; ClinVar confirms NM_001112741.2:c.1273G>A p.(Val425Met). At high temperatures, normal Kv3.1 channels shift activation to more negative potentials, which increases channel availability and probably explains the clinical improvement with fever [36].

2.13 EPM1B: PRICKLE1

PRICKLE1 (also called RILP) encodes a protein of the non-canonical WNT planar cell polarity pathway that interacts with the transcriptional repressor REST [40]. The founder variant NM_153026.3:c.311G>A p.(Arg104Gln), homozygous in the three original consanguineous families (Israeli Arab and Jordanian), blocks the interaction with REST in vitro [39, 40]. Heterozygous variants have been described in isolated patients with myoclonic epilepsy, and disruption of prickle genes causes seizures in zebrafish, fruit flies and mice, which led to the proposal of a dosage effect; the absence of family data in these reports precludes attributing causality to them [39]. OMIM groups EPM1B with EPM1A on the basis of clinical similarity, although the mechanism is distinct [39].

2.14 Inheritance of the other PMEs

AMRF, KCTD7-related PME, North Sea PME and EPM1B are autosomal recessive, with asymptomatic heterozygotes [7, 14, 29, 39]. MEAK is autosomal dominant and almost always results from a de novo variant; one family has been described with two affected siblings and two affected children of one of them, with parents without a detectable variant in blood, consistent with germline mosaicism [15, 33].

3. Clinical Spectrum and Manifestations by System

3.1 Unverricht-Lundborg disease: presentation

The disease begins between 6 and 15 years of age in a previously healthy child [3]. In more than half of patients the first manifestation is involuntary myoclonus; in almost half, a generalised tonic-clonic seizure [3]. The myoclonus is action-activated and stimulus-sensitive (light, physical exertion, stress), predominates in the proximal limb muscles, is asynchronous, focal or multifocal, and may generalise into clusters of myoclonic seizures or into myoclonic status epilepticus with partially preserved consciousness [3]. In the North American series of Eldridge, the myoclonus was photosensitive, sometimes violent, and more intense on awakening [43]. Absences, psychomotor seizures and focal motor seizures may occur; tonic-clonic seizures are rarely absent [3]. On initial examination, which appears normal, an experienced observer finds almost imperceptible myoclonus triggered by light stimulus, threat, hand-clapping, tapping the nose, eliciting reflexes, movement or cognitive tasks [3].

3.2 Unverricht-Lundborg disease: course and natural history

During the first 5–10 years the myoclonus progresses, and about one-third of patients become severely disabled and wheelchair-dependent [3]. Seizures, infrequent at first, become more frequent over the following 3–7 years and may then cease with appropriate treatment [3]. Years after onset, ataxia, incoordination, intention tremor and dysarthria appear; cognition, particularly memory, generally remains normal, although performance on timed tests that depend on motor function is poorer [3]. Depression and emotional lability are frequent, and schooling is often interrupted [3]. The myoclonus may be of subcortical origin, which contributes to treatment resistance [3].

Follow-up of 20 genetically confirmed patients (Marseille and Messina; mean age 37.9 years; mean follow-up 25.6 years from a mean onset at 12.3 years; nine from North Africa, six Italian, two French and three of mixed European origin) showed that the myoclonus progressed only in the first 5 years; three patients became severely disabled, six were fully independent and 11 needed variable degrees of support; eight became seizure-free, and seizures became much less frequent after 10 years of disease; on EEG, background activity stabilised or improved and photoparoxysmal responses disappeared in all but two [94]. These data describe treated disease, with valproate and avoidance of phenytoin, and not untreated natural history. In the Finnish national study (66 homozygous patients; 33 men and 33 women), earlier onset and longer duration were associated with more severe action myoclonus, lower performance intelligence quotient and changes in cortical excitability [72]. Before the valproate era, many patients died 8–15 years after onset, usually before the age of 30; with current pharmacological treatment, physiotherapy and psychosocial support, life expectancy is comparable to that of controls up to the age of 40 but worse in the long term, with deaths mainly from respiratory infections [3]. Compound heterozygous patients have a more severe course, with resistant seizures, moderate to severe cognitive impairment and frequent psychiatric symptoms [75, 76].

3.3 Lafora disease: presentation

Lafora disease begins between 8 and 19 years of age (peak at 14–16 years), after apparently normal development; many patients had isolated febrile or afebrile seizures in childhood [2]. In the meta-analysis of 298 genetically confirmed cases, the mean age at onset was 13.4 years (standard deviation 3.7), and 9.1% began at 18 years or later [88]. In a Spanish series of 17 patients with NHLRC1 variants (Gómez-Abad and colleagues, 2005), onset occurred between 12 and 15 years of age, except in two patients (7 and 22 years), and presentation was a generalised tonic-clonic seizure in 50%, simple focal occipital seizures in 18.7%, focal seizures with secondary generalisation in 12.4%, absences in 6.3% and myoclonic seizures in 6.3%; one patient presented with liver failure, without neurological manifestations [10]. The main seizure types are myoclonic seizures and occipital seizures; occipital seizures manifest as transient blindness, simple or complex visual hallucinations, photomyoclonic or photoconvulsive seizures, or migraine-like headaches with scintillating scotomas [2]. Generalised tonic-clonic seizures, atypical absences, atonic seizures and focal impaired-awareness seizures also occur [2]. The myoclonus is fragmentary, symmetrical or massive, positive or negative, present at rest and exacerbated by action, light or excitement, and disappears in sleep [2]. Not all visual hallucinations are epileptic: some respond initially to antipsychotics and not to antiseizure medications [2]. In sibships, subtle myoclonus, visual hallucinations or headaches appear earlier in subsequently affected siblings than in the proband, and there is considerable variability in age at onset within and between families [2].

3.4 Lafora disease: course and natural history

The course is characterised by increasingly frequent and intractable seizures, status epilepticus of any seizure type, cognitive decline at onset or soon after the first seizures, early dysarthria and ataxia and late spasticity; the emotional disturbance and confusion of the early stages give way to dementia [2]. The myoclonus becomes almost continuous and is the main cause of early wheelchair dependence; most patients lose the ability to walk before the age of 21 and, by around 25 years, are in a vegetative state with continuous myoclonus and tube feeding; without a feeding tube, aspiration is frequent and aspiration pneumonia is a common cause of death [2]. The Apulia cohort (six families, 2010–2024) described three electroclinical stages: presenting symptoms, with seizures and then myoclonus; progressive neurodegeneration, with drug-resistant epilepsy, dementia and ataxia; and a terminal stage, with severe disability, convulsive emergencies, aspiration pneumonia and respiratory failure [95].

The individual patient data meta-analysis by Pondrelli and colleagues (73 articles, 298 cases) estimated survival of 93% (95% CI 89–96) at 5 years, 62% (95% CI 54–69) at 10 years and 57% (95% CI 49–65) at 15 years, with a median of 11 years, and probabilities of loss of autonomy of 45% at 5 years, 75% at 10 years and 83% at 15 years, with a median of 6 years; Asian origin and onset before the age of 18 were poor prognostic factors, whereas the affected gene and the initial symptoms were not related to survival [88]. These figures come from published cases, with selection bias, and mix different eras and treatments. The mild forms due to NHLRC1 p.(Asp146Asn) have later onset, a longer course and prolonged preservation of daily activities [2, 85].

3.5 Lafora disease: cognition and behaviour

In a single-centre study from Bologna (14 patients; onset at 12.6 ± 2.0 years; baseline assessment at 16.5 ± 2.6 years), the Leiter-3 intelligence quotient was 79.8 ± 22.8 and was abnormal in 10 patients (71%); executive dysfunction was the dominant deficit (abnormal Children's Color Trails Test in 12 of 14, 86%), visuomotor integration was abnormal in 71% and adaptive behaviour (Vineland-II) was on average moderately low, mainly because of communication; at 12 months (nine patients), there was a significant decline in intelligence quotient, executive functions and visuomotor integration [96]. GeneReviews describes, at onset, visual hallucinations, cognitive deficits, depressed mood and headaches and, later, agitation and predominantly frontal dementia, with impaired performance and executive function [2].

3.6 Findings by system in Lafora disease

3.7 Other members of the class: comparative clinical profile

KCTD7-related disease begins between 5 and 39 months of age, with seizures in 98–100%, multifocal myoclonus in more than 95%, regression in all and mortality of about 15% between 18 months and 21 years [7]. KCNC1-related MEAK resembles Unverricht-Lundborg disease at onset (myoclonus at 6–14 years, mean 10 years), with seizures from adolescence to the fourth decade and mild cognitive decline in more than half [15]. Action myoclonus–renal failure syndrome begins in the second or third decade with cortical tremor, and kidney involvement, which may precede the neurological picture by decades or be absent, progresses to end-stage kidney disease; death occurs 7–15 years after onset [7, 14]. GOSR2-related PME begins with ataxia and areflexia between 1 and 4 years of age, followed by myoclonus, with scoliosis and relative cognitive preservation [7, 31]. In juvenile DRPLA (onset before the age of 20), myoclonus, epilepsy and intellectual deterioration predominate [16]. In Gaucher disease type 3, one of the most common forms combines mild systemic disease with progressive myoclonic encephalopathy [54, 65]. In the classic Norrbotten form, the central neurological sign is multifocal myoclonus of cerebral origin, at rest or with movement, together with generalised tonic-clonic seizures and a progressive saccadic deficit leading to lateral gaze palsy; some authors designate as type 3a the patients with PME and dementia, who have a worse prognosis and in whom neurological deterioration may cause death in adulthood despite treatment [65]. The juvenile/adult form of galactosialidosis, described mainly in Japan, presents with generalised seizures, myoclonus, ataxia and cognitive deterioration, with bilateral cherry-red spots in most, angiokeratomas and mildly coarse facial features, and onset between 1 and 40 years [61]. In sialidosis, the spectrum ranges from the multisystem form with hydrops fetalis to slowly progressive forms with seizures and myoclonus (cherry-red spot–myoclonus syndrome), and the cherry-red spot is common to the whole spectrum [60].

3.8 Action myoclonus–renal failure syndrome: clinical features and natural history

AMRF is a continuum of two manifestations that progress independently: PME and kidney disease, apparently a steroid-resistant nephrotic syndrome [14]. In the series of Badhwar and colleagues (15 patients from 9 families in five countries, including the four original French-Canadians), fine finger tremor began at 17–26 years (mean 19.8), action myoclonus at 14–29 years (mean 21.7) and infrequent generalised seizures at 20–28 years; proteinuria was detected in all between 9 and 30 years of age and progressed to renal failure in 12 of the 15, 0–8 years after detection (mean 3.8 years) [14, 23]. Presentation was neurological in five, renal in four and simultaneous in six; nine patients died between 25 and 35 years of age, from respiratory complications, renal failure or transplant complications [21]. The tremor, which is cortical, appears during fine actions such as writing, is relieved by alcohol and is eventually masked by myoclonus; action myoclonus, the most disabling manifestation, extends to the lower limbs, trunk and bulbar musculature, with dysarthria and dysphagia, and there is negative myoclonus with postural lapses that produce a "bouncy" standing posture [14]. Severe epileptic photosensitivity and GTCS are frequent; ataxia and dysarthria appear years later; sensorimotor neuropathy, mainly demyelinating, and sensorineural deafness are less frequent, and cognitive decline is rare, described in four families [14]. In a review of published cases, the frequencies were: myoclonus 100%, ataxia 96%, tonic-clonic seizures 82%, dysarthria 68%, tremor 65%, kidney involvement 62%, peripheral neuropathy 17%, cognitive impairment 13.7% and deafness 6.8% [91].

The kidney disease begins with mild, asymptomatic proteinuria and may progress to nephrotic syndrome and end-stage kidney disease; intrafamilial variability is large, and in one family one brother underwent kidney transplantation for focal segmental glomerulosclerosis one year before neurological onset, another died of renal failure at 12 years and the sister presented with gait disturbance before proteinuria [14]. There is an extreme of the spectrum without kidney involvement: of five Italian patients with onset at 14–23 years, four died 10–15 years after onset without renal failure [21]. Three Japanese patients homozygous for p.(Gly462AspfsTer34) had onset in the fifth or sixth decade, without renal failure [14]. Dialysis and transplantation treat the renal failure but do not improve the neurological picture, which does not result from uraemic encephalopathy [14, 23]. One German family had dilated cardiomyopathy in two of three affected siblings, at 14 and 21 years [14]. Death occurs 7–15 years after onset, from neurological deterioration, aspiration, sudden unexpected death in epilepsy, end-stage kidney disease or graft rejection [14].

3.9 KCTD7-related PME: clinical features and natural history

Symptoms begin between 5 and 39 months of age; seizures are the first manifestation in 60–70%, and in about one-third onset is with developmental delay or regression, a movement disorder or ataxia [7]. In the international cohort of Yoganathan and colleagues (42 patients from 36 families), the median age at first seizure was 14 months (interquartile range 11.75–22.5), myoclonic seizures were the first seizure in 43.9%, and 38.1% had delay followed by progressive regression; six died between 1.5 and 21 years, and in the review of 59 published cases 88% had a PME phenotype and 14% died between 3 and 18 years of age [28]. According to GeneReviews, which integrates about 95 patients, epilepsy occurs in 98% (myoclonic in 95–100%, GTCS in 50%, atypical absences in 20%, atonic in 18%, tonic and focal in 15%, recurrent febrile seizures in 10%), regression in 95% and speech and language difficulties in 100%; ataxia in 59%, microcephaly in 37%, hypotonia in 41%, spasticity in 37%, tremor in 37%, dystonia and choreoathetosis in 29% and ophthalmological involvement in about 20% [7]. About three-quarters stop walking or need assistance, and all are partially or fully dependent in daily activities [7]. EEG shows a slow or disorganised background in 78%, generalised or multifocal discharges with posterior predominance and, in a few, hypsarrhythmia; photosensitivity occurs in less than one-fifth; MRI is normal in about 60% [7]. In an Indian series of six children (median onset at 11 months), all had worsening with fever, five had ataxia and three had thalamic T2 hypointensity [92].

3.10 North Sea PME: clinical features and natural history

In the series of Boissé Lomax and colleagues (12 patients), ataxia began at a mean age of 2 years and myoclonic seizures at 6.5 years, with tonic-clonic seizures, absences and falls; all developed scoliosis in adolescence, four had pes cavus and two syndactyly; all had raised CK (median 734 IU) with normal muscle biopsy; EEG showed generalised spike-wave with posterior predominance and photosensitivity in all; all became wheelchair-dependent (mean age 13 years) and four died in the third or early fourth decade [32]. In a Dutch series of 17 patients (5–46 years), severity and rate of progression varied widely: some used a wheelchair in the first decade and others walked unaided in the third; fever, intercurrent illness, heat, emotions, stress, noise and flashing light worsened the myoclonus [97]. The areflexia is due to sensory neuronopathy and anterior horn involvement [29, 97]. Cognition is usually preserved, with mild memory difficulties in the third decade [29]. One compound heterozygous patient reached 61 years of age, wheelchair-dependent since her thirties [29]. The 2025 review describes three phenotypes: myoclonic ataxia or PME (32 isolated and 4 combined among 42), congenital muscular dystrophy (1 isolated and 4 combined) and deafness (4 isolated and 1 combined); in the mixed phenotype, CK ranged from 833 to 5582 U/L, with refractory seizures from 2–3 years of age, and in the classic phenotype there was hypothyroidism in three patients and delayed puberty in two [31].

3.11 MEAK: clinical features and natural history

In the series of 20 patients of Oliver and colleagues, symptoms began between 3 and 15 years of age (median 9.5), with progressive myoclonus and rare tonic-clonic seizures; the early ataxia was rapidly overshadowed by the myoclonus; 10 were in a wheelchair by late adolescence, half had mild cognitive decline and there were no early deaths [36]. EEG showed generalised spike and polyspike-wave, with photosensitivity in most; polygraphic recordings confirmed cortical myoclonus with co-activation of agonist and antagonist muscles, and MRI showed symmetrical, apparently progressive cerebellar atrophy, with a prominent corpus callosum [36]. Six patients improved transiently during febrile episodes [36]. In the series pooled by GeneReviews, myoclonus occurred in 30/30, epilepsy in 31/32, ataxia in 29/30, prior delay in 3/30 and cognitive decline in 16/28; seizures appear from adolescence to the fourth decade and are usually controlled, and about half need a walking aid or wheelchair by late adolescence or early adulthood; severe dementia has not been observed, and the disease does not appear to shorten life [15]. The family with vertical transmission had a slightly milder phenotype [33].

3.12 EPM1B: clinical features

In the Israeli Arab family of Berkovic and colleagues (eight affected), seizures began at a mean age of 7.3 years (5–10), myoclonic, tonic-clonic or both, aggravated by sunlight; four had delayed walking and difficulty running before the seizures, three became wheelchair users and there was no progressive dementia [39]. In another family, ataxia began at around 4 years of age, with dysarthria, upward vertical gaze palsy, pyramidal signs and axonal sensory neuropathy; in the Jordanian family, gait ataxia began at 15 months, with action tremor at 4 years and atonic seizures at 8–10 years, which later evolved into myoclonic seizures [39]. The picture approaches that of ULD, with earlier onset and a somewhat more severe course [39].

4. Clinical Recognition and Warning Signs

4.1 When to suspect

A PME should be considered in any previously healthy child, adolescent or young adult with myoclonus and seizures that worsen over time, particularly if the neurological examination, cognition or background EEG ceases to be normal [1, 6]. In adults, Saudubray lists as clues to a metabolic disorder epilepsy that does not fit a classic syndrome, a mixture of myoclonus and focal seizures, association with cerebellar or pyramidal signs, worsening with classic antiseizure medications, unexplained status epilepticus and slowing of background activity with photoparoxysmal responses at low stimulation frequencies [98]. The Oxford Desk Reference recommends EEG in a child with ataxia when Unverricht-Lundborg PME or lipofuscinosis is suspected [99].

4.2 The juvenile myoclonic epilepsy pitfall

Juvenile myoclonic epilepsy (JME) is the main initial misdiagnosis in both diseases: according to the Oxford Desk Reference, JME accounts for 5–10% of epileptic seizures, begins in adolescence and presents with myoclonus, tonic-clonic seizures in 90% and absences in 30% [100]. Most patients with Lafora disease receive a brief diagnosis of JME [2], and the Chinese ULD case went undiagnosed for ten years [73]. The finding that separates the entities is the background EEG, normal in JME and already slow in Lafora disease when myoclonus appears; in ULD, the EEG is always abnormal, even before symptoms, with a labile, sometimes slow, background and marked photosensitivity; and the neurological examination is normal in JME [2, 3].

4.3 Specific warning signs

In Unverricht-Lundborg disease, the pointers are stimulus-sensitive action myoclonus that makes running, sport or climbing stairs difficult, intense photosensitivity, ataxia that sets in years later and a normal MRI [3]. In Lafora disease, the warning signs are occipital seizures with transient blindness or visual hallucinations, declining school performance and changes in behaviour or mood at onset, rapid progression of seizures and myoclonus, and a slow EEG with posterior-predominant discharges [2, 6]. Visual hallucinations, withdrawal and cognitive decline may raise suspicion of schizophrenia, a less likely hypothesis once convulsions and an epileptiform EEG appear [2].

In the other PMEs, the warning signs are specific to each entity: fine hand tremor in an adolescent or young adult with proteinuria, nephrotic syndrome or a kidney transplant (AMRF), bearing in mind that ciclosporin-induced tremor may be wrongly blamed before molecular diagnosis [14]; multifocal myoclonus and regression in the first three years of life, with febrile worsening (KCTD7-related PME) [7, 92]; ataxia beginning in the first years of life with areflexia, scoliosis and raised CK, particularly in families from North Sea countries (GOSR2) [32]; and a picture similar to ULD with a negative CSTB expansion, a negative family history and improvement of myoclonus during fever (MEAK) [15, 36]. The kidney disease of AMRF may precede the neurological disease by decades, so the absence of neurological signs does not exclude the diagnosis in a young person with familial focal segmental glomerulosclerosis [14].

4.4 Emergency situations

In both diseases, status epilepticus (myoclonic, convulsive or non-convulsive) and seizure clusters may require intensive care [2, 101]. In advanced Lafora disease, aspiration and respiratory failure are frequent causes of hospital admission and death [2, 95]. Abrupt withdrawal of long-term clonazepam may precipitate myoclonic status epilepticus in ULD [17].

4.5 Newborn screening

The sources consulted describe no newborn screening programmes for any of the PMEs in this article, and the CSTB expansion is not detected by the sequencing platforms used in genomic studies [3].

5. Diagnosis: Criteria, Tests and Algorithms

5.1 Criteria

No published consensus clinical criteria exist for Lafora disease [2], and the GeneReviews chapter on Unverricht-Lundborg disease defines suggestive findings rather than formal criteria [3]. In both, the diagnosis is established by molecular confirmation; skin biopsy retains a role in Lafora disease when genetic testing is inconclusive [2, 3].

5.2 Suggestive findings

In Unverricht-Lundborg disease, GeneReviews describes as suggestive, in a previously healthy child aged 6 to 15 years, involuntary action-activated myoclonus and/or generalised tonic-clonic seizures, photosensitive generalised spike-wave and polyspike-wave paroxysms, an EEG that is always abnormal (even before the manifestations), with labile and sometimes slow background activity, gradual worsening of myoclonus and ataxia, and a normal MRI [3]. In Lafora disease, in a previously healthy adolescent, the suggestive findings are occipital seizures with transient blindness or visual hallucinations, fragmentary, symmetrical or generalised myoclonus, generalised seizures, progressive neurological degeneration, an EEG with background slowing, loss of the alpha rhythm and of sleep features, irregular posterior-predominant spike-wave and photosensitivity, PAS (periodic acid–Schiff)-positive inclusions on skin biopsy, a normal MRI at onset and a family history consistent with recessive inheritance, the absence of which does not exclude the diagnosis [2]. Table 3 summarises the comparison proposed by Orsini and colleagues.

Sources for Table 3: [2, 3, 45].

5.3 Ancillary investigations

EEG is the central non-molecular investigation. A baseline recording should be obtained before starting antiseizure medications, because this is the phase in which it is most characteristic [3]. In Lafora disease, EEG deterioration often precedes clinical deterioration, and the 2025 scientific symposium stressed that EEG is the first investigation to worsen even in minimally symptomatic patients [84]. The sleep EEG may progressively lose spindles and K-complexes in ULD, in parallel with motor and cognitive worsening, according to a case report with quantitative EEG [102]. Giant somatosensory evoked potentials occur in several PMEs, including late infantile lipofuscinosis [6], and in Lafora disease visual and somatosensory potentials are of high voltage at onset [2]. MRI is normal at the onset of both diseases and excludes structural lesions [2, 3]. FDG-PET showed frontoparietal hypometabolism in the Chinese ULD case [73] and progressive parietal and precuneus hypometabolism in two patients with Lafora disease followed for six months, even with electroclinical stability [103]. Standardised scales (UMRS, Unified Myoclonus Rating Scale, for myoclonus; SARA, ICARS or BARS for ataxia) serve to quantify and monitor the disease, not to diagnose it [3, 68].

5.4 Biomarkers under investigation

Neurofilament light chain (NfL) was increased in 31 patients with Lafora disease (26 families; mean age 21 years; Ulm, Bologna and Dallas): mean serum NfL 13.95 pg/mL (95% CI 11.20–16.69) versus 7.72 pg/mL (95% CI 6.79–8.65) in controls, and cerebrospinal fluid (CSF) NfL 576.9 versus 306.8 pg/mL; the area under the curve was 0.80 in serum and 0.88 in CSF, and serum NfL correlated with disease stage (r = 0.56) and with the clinical performance scale (r = −0.49) [104]. It is an exploratory biomarker for trials and not a validated diagnostic test.

5.5 Molecular testing in Unverricht-Lundborg disease

The first-line test is targeted analysis of the dodecamer expansion, requested explicitly, because multigene sequencing panels, exome and genome do not detect this expansion [3, 68]. Conventional PCR detects alleles of 2–3 repeats, but apparent homozygosity for a normal allele does not exclude an expansion; expansion-specific long-range PCR or Southern blot is required [3]. Exact sizing of the repeats has no demonstrated clinical utility [3]. If only one expanded allele is found, or none in a patient with a typical phenotype, CSTB is sequenced to look for point variants or indels, which account for about 10% of alleles [3]. A negative panel or exome does not exclude the disease and should not be used to rule it out [68]. The proportion of alleles detected by each method (90% and 10%) is not equivalent to the clinical sensitivity of a single test.

5.6 Molecular testing in Lafora disease

GeneReviews proposes a multigene PME panel when the phenotype suggests the disease, and exome (or genome) sequencing when the picture is atypical, because most EPM2A and NHLRC1 variants lie in the coding region [2]. Targeted deletion and duplication analysis is needed for the 10–15% of structural EPM2A variants and for NHLRC1 deletions, less than 10% [2]. Biallelic variants of uncertain significance, or one pathogenic variant with one of uncertain significance, neither establish nor exclude the diagnosis [2]. Apparent homozygosity should prompt consideration of an undetected deletion of one or more exons (artefactual homozygosity) or uniparental isodisomy; if the variant is found in only one parent, with confirmed parentage, a de novo event or parental post-zygotic mosaicism is possible [2]. The report should state the gene, the versioned transcript (NM_005670.4 for EPM2A; NM_198586.3 for NHLRC1; NM_000100.4 for CSTB), the HGVS nomenclature, the zygosity and the ACMG/AMP classification (American College of Medical Genetics and Genomics/Association for Molecular Pathology), in which "pathogenic" and "likely pathogenic" have the same diagnostic value [2, 74].

5.7 Skin biopsy

Lafora bodies, starch-like, poorly branched and insoluble polyglucosans, are found in eccrine duct cells or apocrine myoepithelial cells and stain with PAS [2]. Because the apical PAS-positive granules of normal axillary apocrine glands can mimic Lafora bodies, the biopsy should be taken outside the axilla and the genital region, where Lafora bodies in the eccrine ducts are unequivocal [2]. There is a risk of false negatives, particularly early in the disease, and of false positives through confusion with normal polysaccharides of the apocrine glands [2]. Andrade and colleagues (2003) showed that patients with variants in either gene have Lafora bodies in both apocrine myoepithelial cells and eccrine duct cells, and, by molecular analysis, reclassified as Unverricht-Lundborg disease a patient previously diagnosed with an atypical form of Lafora disease on the basis of an axillary biopsy with PAS-positive material lining the glandular lumen, without bodies in the myoepithelial cells or the eccrine glands (a case published in 2000 in Arquivos de Neuro-Psiquiatria); the authors therefore preferred skin biopsy outside the axilla [10]. The biopsy identifies the disease without distinguishing the gene, and is now needed only when molecular testing is uninformative [82]. In Unverricht-Lundborg disease, skin biopsy shows no Lafora bodies, which distinguishes it from Lafora disease [2].

5.8 Practical algorithm

The following algorithm integrates the recommendations of GeneReviews and the reference textbooks.

In an Italian study cited by Zimmern and Minassian, 47 of 165 patients with PME had no diagnosis, and next-generation sequencing in 38 of them yielded 16 new diagnoses, raising the overall yield to 82% [51].

5.9 Prenatal and preimplantation diagnosis

Once the two pathogenic variants have been identified in an affected relative, prenatal diagnosis and preimplantation genetic testing are possible in both diseases [2, 3]. In Unverricht-Lundborg disease, the laboratory must use a method capable of detecting the expansion [3].

5.10 Diagnosis of the other PMEs

None of these entities has consensus diagnostic criteria, and the diagnosis rests on identifying biallelic variants (AMRF, KCTD7, GOSR2, PRICKLE1) or a heterozygous variant (MEAK) in a patient with a suggestive picture [7, 14, 15]. PME, ataxia, epilepsy, nephrotic syndrome or lysosomal disease panels and exome sequencing detect most variants; single-gene analysis is rarely useful, except when suspicion is very specific [7, 14, 15]. A variant of uncertain significance, or a pathogenic variant with another of uncertain significance, neither confirms nor excludes the diagnosis [7, 14]. In AMRF, leucocyte glucocerebrosidase activity may be normal or only low (7.5 nmol/h/mg protein, against a normal value of ≥8.7, in the patient of Quraishi and colleagues), outside the diagnostic range for Gaucher disease, and cannot be used to exclude the disease [24, 90]. In KCTD7-related PME, skin biopsy may show lipofuscinosis-type inclusions, but their absence does not exclude the diagnosis [7]. In MEAK, the recurrent variant p.(Arg320His) accounted for 13% of the 84 undiagnosed PMEs studied by exome sequencing and 7% of a second cohort of 28 [35].

6. Differential Diagnosis

6.1 Comparative table

6.2 Main differentials

The most frequent and most important distinction in practice is between the two PMEs of this article and JME, made on the background EEG and the neurological examination [2, 3]. Between ULD and Lafora disease, age at onset, occipital seizures, rapidity of progression, early cognitive decline and Lafora bodies point to the latter [2, 3]. KCNC1-related EPM7 is the closest mimic of ULD, with an almost overlapping presentation; because the recurrent variant is dominant and almost always de novo, the family history is negative and the CSTB expansion test is normal [3, 15]. MERRF shares action myoclonus and onset in childhood or adolescence with ULD, but is distinguished by myopathy with ragged-red fibres, deafness and basal ganglia abnormalities [62]; valproate is discouraged in mitochondrial forms [3, 17]. In MERRF, the m.8344A>G variant in MT-TK accounts for 80–90% of cases, ragged-red fibres appear in 92% of muscle biopsies and the estimated prevalence in Europe is 0.9/100,000; valproate inhibits carnitine uptake and, when it is indispensable, should be combined with L-carnitine [17, 63]. Cardiomyopathy (hypertrophic, dilated or histiocytoid) occurs in about 40% and there may be ventricular pre-excitation [64]. The risk to the offspring of a woman with m.8344A>G depends on the mutant load in maternal blood: in an analysis of 47 women, less than 5% of children were affected with a load below 35%, 15% with 40–59%, 20% with 60–79% and 80% with 80–100% [107]. SGCE-related myoclonus-dystonia, autosomal dominant with reduced penetrance, is distinguished by the associated dystonia and marked improvement with alcohol [106]. Sialidosis type I, juvenile/adult galactosialidosis and Gaucher disease type 3 require ophthalmological examination and targeted enzyme studies; ERG helps to rule out the lipofuscinoses [2, 6, 61]. In the oligosaccharidoses, urinary oligosaccharide testing precedes enzyme assay in leucocytes; thin-layer chromatography may miss some cases, and tandem mass spectrometry reduces this risk [60].

Among the other PMEs, age at onset separates KCTD7-related PME (before 4 years, with regression) from North Sea PME (ataxia in the first years of life, myoclonus in childhood, scoliosis and raised CK) and from the second-decade forms (ULD, MEAK, EPM1B) and the young adult form (AMRF) [7, 14, 15]. MEAK is described as almost identical to ULD and GOSR2-related PME in the early phase; it is distinguished by de novo dominant inheritance, absence of scoliosis and of raised CK, and febrile improvement, whereas in North Sea PME fever causes worsening [15, 36, 97]. In AMRF, proteinuria, demyelinating neuropathy and deafness are the extraneurological clues, and the myoclonus, which occurs in the kinetic phase of movement, causes myoclonic dysmetria often mistaken for cerebellar ataxia, to the point that one patient was included in a spinocerebellar ataxia cohort [14]. In an adult with tremor, AMRF is distinguished from familial cortical tremor by progression and proteinuria [14]. In the young child, the differential of KCTD7-related PME includes subacute sclerosing panencephalitis, paraneoplastic or parainfectious opsoclonus-myoclonus syndrome, the neuronal ceroid lipofuscinoses (retinopathy), SMA-PME (proximal weakness and denervation), and the DHDDS-, NUS1-, IRF2BPL- and TBC1D24-related disorders [7].

7. Treatments and Risk Management Strategies

7.1 General principles and nature of the evidence

No disease-modifying therapy is approved for Unverricht-Lundborg disease or Lafora disease; treatment is symptomatic, pharmacological and rehabilitative, with psychosocial support [2, 3]. There are no clinical practice guidelines with formal grading of the evidence: the GeneReviews chapter on Lafora disease states explicitly that its recommendations are based on the author's experience and follow those for the other PMEs [2], and the therapeutic references for the class are expert reviews (Michelucci and colleagues, 2016; Ferlazzo and colleagues, 2017; Orsini and colleagues, 2019), built mainly on small series and case reports [17, 45, 101]. The only randomised trials conducted in a genetically confirmed PME were the piracetam crossover trial and the two phase III brivaracetam trials, both in ULD [108, 109]. Antiseizure medications have a clear effect on generalised seizures, sometimes for many months, but do not alter the cognitive and behavioural progression of Lafora disease [2]. Regimens of three to four drugs are common, and pharmacological overload and neurotoxicity must be avoided, particularly in Lafora disease, in which resistant myoclonus leads to a risk of overmedication [2, 17].

7.2 Therapeutic hierarchy

According to Ferlazzo and colleagues, valproate is the drug of choice in PMEs, except in mitochondrial forms; levetiracetam or, alternatively, clonazepam are the first adjuncts; zonisamide and perampanel are alternatives; piracetam is reserved for disabling myoclonus that does not respond to the above; phenobarbital and primidone are used for resistant myoclonus or seizures; and lamotrigine should be used with caution because of its unpredictable effect on myoclonus [17]. The ULD GeneReviews places valproate first, clonazepam as an adjunct, high-dose piracetam for myoclonus, levetiracetam, brivaracetam and perampanel as effective in myoclonus and generalised seizures, and topiramate and zonisamide as adjuncts [3]. The Lafora disease GeneReviews organises treatment into three lines: first line with valproate and benzodiazepines (clonazepam, clobazam, diazepam); second line with levetiracetam, zonisamide, topiramate and perampanel, with limited evidence; third line with primidone, phenobarbital, piracetam and ethosuximide; the most commonly used combinations bring together valproate, perampanel, topiramate, zonisamide or levetiracetam and a benzodiazepine [2]. Divergences between sources are summarised in section 7.12.

7.3 Drugs: doses and sources

The reference dose is that of the expert reviews and trials in PMEs; the official label is cited alongside for comparison. None of these drugs has an approved indication for ULD or Lafora disease; use is off-label, except for piracetam in cortical myoclonus in countries where that indication is authorised.

The evidence supporting these doses comes from small series [17]. In a Finnish prospective open-label study from the pre-molecular era (26 patients, most probably with ULD), withdrawal of all previous antiseizure medications, including phenytoin and carbamazepine, and introduction of valproate with clonazepam (plus phenobarbital if tonic-clonic seizures persisted) markedly improved myoclonus and seizures, with benefit maintained in 73% at six years; withdrawal of the aggravating drugs may have contributed to the effect [17]. For piracetam, in addition to the ULD trial, there is a placebo-controlled crossover trial in 21 patients with cortical myoclonus of various causes (2.4–16.8 g/day), an open-label series of 12 patients with PME with doses up to 45 g/day and benefit maintained in some for up to seven years, and a series of 11 patients with advanced PME (ULD, Lafora, MERRF, sialidosis) with significant improvement of myoclonus, mainly in the first 12 months, and no effect on tonic-clonic seizures; the cost (about €300 per month for 30 g/day in Italy) and the number of doses limit its use [17]. Adjunctive zonisamide produced lasting improvement of myoclonus and seizures in five of seven patients (six with ULD and one with Lafora disease), reduced myoclonic seizures by more than 50% in about half of 30 patients with PME (five withdrawals because of anorexia, drowsiness or asthenia) and reduced myoclonus severity on the UMRS in 12 patients with ULD in a 6–8-week open-label trial [17]. Topiramate improved myoclonus and seizures in five of eight patients with PME (five of the eight with Lafora disease and three without an aetiological diagnosis), with withdrawal in two because of cognitive decline and vomiting [17]. In the brivaracetam trials, the variable disease duration (2–50 years) and the previous or concomitant use of piracetam or levetiracetam, which act on the same target (SV2A), may have reduced the response; brivaracetam suppresses photosensitivity [17]. Non-pharmacological measures complete the treatment: avoiding bright lights and striped patterns and wearing dark glasses when photosensitivity persists; alcohol has an antimyoclonic effect, and the authors allow small amounts on social occasions [17]. In MERRF, benefits have been described with coenzyme Q10 (50–100 mg three times daily) and L-carnitine (1000 mg three times daily) [17]. In neuronopathic Gaucher disease, enzyme replacement does not cross the blood-brain barrier and has a limited or no effect on the neurological manifestations, and combination with miglustat showed no benefit on seizures in an open-label trial [17]; piracetam may help to control the myoclonus [65].

7.4 Agents to avoid

Phenytoin should be avoided as maintenance treatment in all PMEs, because it worsens the neurological symptoms of ULD and may accelerate cerebellar degeneration; the 1983 North American experience associated it with motor and intellectual deterioration, marked ataxia and death [3, 43]. Sodium channel blockers (carbamazepine, oxcarbazepine, phenytoin and probably lacosamide), GABAergic drugs (tiagabine and vigabatrin), gabapentin and pregabalin may worsen myoclonus and myoclonic seizures [3, 17]. Lamotrigine is discouraged by the review of Michelucci and colleagues, has anecdotal reports of worsening myoclonus in Lafora disease and worsened or failed to improve myoclonus in five patients with ULD [2, 17, 101]. In the laforin-deficient mouse, lithium activated glycogen synthase and increased Lafora bodies, a preclinical warning against the direct translation of treatments acting on glycogen [122].

In the other PMEs, the avoidance recommendations are the same: in AMRF, GeneReviews advises against phenytoin, which may accelerate cerebellar degeneration, sodium channel blockers (carbamazepine, oxcarbazepine), tiagabine, vigabatrin, gabapentin and pregabalin [14]; in KCTD7-related PME, the list also includes eslicarbazepine and lacosamide [7]. In the patient of Quraishi and colleagues, clobazam was associated with worsening of myoclonus [24]. In North Sea PME, fever, heat (including hot baths) and intercurrent infections trigger worsening of myoclonus and seizures, which justifies treating fever and preventing infections [31, 97]. In AMRF, renally eliminated and nephrotoxic drugs require adjustment to renal function; miglustat is not recommended with a creatinine clearance below 30 mL/min/1.73 m² [123, 124].

7.5 Unverricht-Lundborg disease: approach

Valproate is the drug of first choice, because it reduces myoclonus and the frequency of generalised seizures; its introduction, together with withdrawal of phenytoin, changed the prognosis of the disease [3, 43]. Clonazepam is the classic adjunct for myoclonus, and levetiracetam is an alternative or complement, with good tolerability [3, 17]. High-dose piracetam has the best-quality evidence: in the randomised crossover trial of Koskiniemi and colleagues, 24 g/day significantly improved the global myoclonus score (p = 0.005), functional disability and global assessments, with a linear dose-effect relationship and good tolerability [108]. Perampanel improved myoclonus in 10 of 12 patients and suppressed seizures in the six who still had them, with psychological and behavioural effects in half [115]; in the Italian series of 49 patients with PME, those with ULD or a similar phenotype improved most (p = 0.011) [116]. Zonisamide and topiramate are possible adjuncts [3, 17]. N-acetylcysteine has had variable results, and vagus nerve stimulation was described as reducing seizures and cerebellar dysfunction in one report [3]. Compound heterozygous patients have seizures resistant to the regimens that usually control homozygotes [76]. Extreme stimuli (light, noise, stress) should be avoided [3].

7.6 Lafora disease: symptomatic approach

Valproate is the traditional treatment and suppresses tonic-clonic seizures and myoclonus for some time; benzodiazepines are adjuncts for myoclonus, without clear evidence in Lafora disease and with possible tolerance [2]. In the open-label study of perampanel in 10 patients (disease duration 2 to 27 years; mean dose 6.7 mg/day), four had a seizure reduction greater than 74% and seven a major improvement of myoclonus, three dropped out because of lack of efficacy or adverse effects, and there was no improvement in disability or cognition [117]. In the Italian series, 12 of the 49 patients had Lafora disease [116]. Primidone, phenobarbital, piracetam and ethosuximide are third-line options, and vagus nerve stimulation has only two anecdotal reports [2]. With progression, generalised seizures become more frequent and myoclonus more disabling, requiring intensive home support or institutional care, depending on resources and family preferences [2].

7.7 Status epilepticus and seizure clusters

In Lafora disease, GeneReviews recommends intravenous benzodiazepines or a phenytoin loading dose in severe worsening with recurrent seizures or status epilepticus, without continuing phenytoin once the episode has been controlled [2]. The review of Michelucci and colleagues proposes intravenous benzodiazepines, valproate or levetiracetam in motor status epilepticus in PMEs [101], and Ferlazzo and colleagues advise considering levetiracetam in myoclonic status epilepticus and reserving phenytoin for selected cases in the late stages [17]. Seizure clusters and status epilepticus may require admission to intensive care [2]. The Apulia cohort described frequent convulsive emergencies in the terminal stage and fewer hospital admissions with home care by a specialised team [95].

7.8 Rehabilitation, nutrition and supportive care

In ULD, physiotherapy (balance, gait training, strengthening) maintains mobility, occupational therapy optimises daily activities with adapted devices (weighted cutlery, dressing hooks), and home adaptations (grab bars, raised toilets, ramps) prevent falls [3]. Speech and language therapy treats dysarthria, with alternative means of communication when necessary; feeding programmes and videofluoroscopic swallowing studies guide food consistency; gastrostomy is an option in advanced cases; nutritional assessment aims to avoid obesity, which hampers mobility [3]. Psychotherapy and peer support are part of treatment, given the frequency of depression [3]. In Lafora disease, physiotherapy with stretching reduces contractures and falls and prevents scoliosis and hip dislocation; hypertonia and dystonia can be treated with baclofen, tizanidine, botulinum toxin or antiparkinsonian drugs; oromotor assessment at each visit and swallowing studies guide the thickening or cooling of food and the decision on a nasogastric tube or gastrostomy, which reduces the risk of aspiration pneumonia in advanced disease [2]. Augmentative and alternative communication, social services, care coordination, palliative care and home nursing complete the approach [2].

7.9 Diet

Between 2000 and 2002 the NINDS (National Institute of Neurological Disorders and Stroke) conducted a six-month ketogenic diet study in 15 patients with advanced Lafora disease, with no results posted on ClinicalTrials.gov [125]. In KCTD7-related disease, the ketogenic diet reduced seizures in a few cases [7].

7.10 Lafora disease: targeted and investigational therapies

Reducing glycogen synthesis is the rational target, because a 50% reduction in glycogen synthase abolishes the disease in murine models [17]. Metformin received orphan designation EU/3/16/1803 from the European Medicines Agency (EMA) for the treatment of Lafora disease on 12 December 2016; an orphan designation recognises the rarity criterion and plausibility, but is not a marketing authorisation [18]. In a retrospective series from three Italian centres, 12 patients (six with EPM2A and six with NHLRC1) in intermediate or late stages received adjunctive metformin for a mean of 18 months (6–36), with a mean maintenance dose of 1167 mg/day (500–2000 mg); gastrointestinal effects limited the dose in four, there were no serious adverse effects, and three had a clinical response, temporary in two [126]. In a Spanish study, early metformin was more effective than late metformin in Epm2a-/- and Epm2b-/- mice, and in an observational cohort of 18 patients the eight treated had slower deterioration of daily activities than the 10 untreated [127]. These are data from uncontrolled series and an observational study; metformin is used off-label, and the clinical case in section 10 is an example of such use [86].

ION283 is an antisense oligonucleotide targeting GYS1 (glycogen synthase 1) mRNA, developed with Ionis, which halted several aspects of progression in laforin-deficient and malin-deficient mice, with electrophysiological improvement [84]. On 9 December 2020 the EMA granted orphan designation EU/3/20/2377 to a 2'-O-(2-methoxyethyl) antisense oligonucleotide targeting GYS1 pre-mRNA for Lafora disease [128]. The open-label phase 1/2 study NCT06609889 (sponsor-investigator B. Minassian, UT Southwestern, Dallas; collaborator Elpida Therapeutics) administers 15 mg intrathecally in repeated doses to 10 patients aged 10 to 18 years, with a primary safety endpoint at two years; it began on 3 December 2024 and has an estimated primary completion date of October 2027 [19]. According to a February 2025 statement from the patient association Chelsea's Hope, the first patient was treated in December 2024 and the second in January 2025, with no safety issues at the one-month assessment, and the safety committee authorised continuation [129]. No results had been published as of the update date.

VAL-1221 is an antibody-enzyme fusion protein that degrades glycogen. In a 12-month compassionate use programme in Bologna, five patients aged 17 to 24 years with intermediate or advanced disease received 20 mg/kg intravenously every 2 weeks; four completed (one stopped at eight months after status epilepticus), tolerability was good (five mild infusion reactions: rash in one and hypotension in four), but the disease progressed in all, the drug was not detected in CSF and CSF metabolomics did not change; the authors concluded that the intravenous route is safe but ineffective and that direct administration into the central nervous system is needed [130]. The expanded access protocol NCT05930223 (Parasail; up to 10 patients aged 12 to 28 years; 20 mg/kg intravenously every two weeks) remains listed as available, last updated in 2023 [131]. A related antibody-amylase fusion protein, VAL-0417, is being tested in a canine model [84].

Empagliflozin, a sodium-glucose cotransporter 2 (SGLT2) inhibitor, was given to two patients with confirmed Lafora disease, 10 mg/day for six months, in a pilot study published in October 2026: it was safe, and seizures, myoclonus, EEG and MRI remained stable, but FDG-PET showed progressive and spreading hypometabolism, so no conclusions about efficacy can be drawn [103, 132]. In the epm2a-/- zebrafish, dapagliflozin, canagliflozin and empagliflozin reduced hyperexcitability and the locomotor deficit [133]. In gene therapy, intravenous administration of rAAV2/9P31 vectors carrying EPM2A or EPM2B reversed neuropathological changes in presymptomatic mice [134], but overexpression of EPM2A, unlike that of EPM2B, formed Lafora bodies, a finding with safety implications for gene replacement [135]. Overexpression of endogenous amylase by SINEUP has been tested in cells and patient fibroblasts [136]. The Ionis-sponsored natural history study (33 patients; Los Angeles, Dallas, Bologna and Madrid; 2019–2022) and the Lafora disease-specific performance scales prepared the endpoints for these trials [137].

7.11 Unverricht-Lundborg disease and other PMEs: research

Clinical development in ULD includes the two negative phase III brivaracetam trials [109], a ropinirole trial (Turku, 16 patients, status unknown since 2008) and an intravenous immunoglobulin trial (Mantua, one patient, status unknown since 2017) [138, 139]. The registry and natural history study of EPM1 and complete CSTB loss-of-function disease at Boston Children's Hospital, with the Epilepsy Foundation, began in October 2024 and plans to include 200 patients, with the UMRS as endpoint [140]. In EPM7, the Kv3 channel modulator AUT00201 was tested in a single-dose crossover phase 1 trial in six patients [37]. In SMA-PME, the first intravenous administration of an AAV9 vector carrying ASAH1 (2.2 × 10¹⁴ vector genomes/kg, under prednisolone and sirolimus) in a 15-year-old girl with advanced disease caused complement activation, capillary leak syndrome, refractory shock and death on the eighth day [141].

7.12 Divergences between sources

7.13 Regulatory status

The European register of orphan designations also includes designation EU/3/05/315, from 2005, for the "treatment of progressive myoclonic epilepsies", now withdrawn [144]. FDA orphan designations were not consulted.

7.14 Trials and therapeutic studies

None of these studies allows indirect comparison between drugs: they differ in populations, disease stage, scales and duration, and only the piracetam trial and the brivaracetam trials were blinded and placebo-controlled [108, 109]. The negative brivaracetam result, with completion rates of 95.3%, showed that rigorous trials are feasible in a PME, and the authors acknowledged that the action myoclonus score, highly variable within the same patient, may not be the ideal instrument [109]. The open-label perampanel and metformin series are susceptible to selection bias and regression to the mean, and the responses described provide no information on modification of the disease course.

7.15 Developments in the last 12 months (November 2025 to October 2026)

7.16 Pregnancy, contraception and breastfeeding

In typical Lafora disease, pregnancy is exceptional, but it may occur in the mild forms (for example, NHLRC1 p.(Asp146Asn)), and the universal use of antiseizure medications requires counselling on fetal exposure [2]. In ULD, with preserved cognition and survival into adulthood, pregnancy is a frequent issue. In a meta-analysis cited by the Oxford Desk Reference, the overall risk of major malformations in women with epilepsy was 7%, 16.8% with polytherapy and 10.7% with valproate; in the registries, the rate with valproate was 11% (28/263) in the Finnish registry, 5% (29/619) in the Swedish registry and 6% (44/715) in the UK registry, and with lamotrigine 2% (35/1558) in the international registry [146]. Valproate increases the risk of neural tube defects about tenfold, has a dose-dependent effect and is associated with neurodevelopmental disorders, including attention deficit hyperactivity disorder (ADHD) and autism spectrum disorder; levetiracetam has not shown a significant excess of major malformations in the available studies [146, 147]. In the European Union, valproate should not be used for epilepsy during pregnancy unless there is no effective alternative, nor in women of childbearing potential outside the pregnancy prevention programme; since 2024, treated men should be informed of the potential risk of neurodevelopmental disorders in children conceived during the following three months and discuss effective contraception [142, 143]. The US label contraindicates valproate for migraine prophylaxis in pregnant women and reserves it, in epilepsy, for when other drugs have failed [110]. The Oxford Desk Reference recommends preconception folic acid 5 mg/day, not stopping medication without speaking to the neurologist and monitoring lamotrigine levels and possibly levetiracetam levels [146]. This tension is particularly relevant in ULD, in which valproate is the drug of first choice [3]. The sources consulted describe no specific breastfeeding data in PMEs.

7.17 Perioperative care and transition

The sources consulted describe no specific perioperative recommendations for ULD or Lafora disease; uninterrupted continuation of antiseizure medication (clonazepam and valproate should not be stopped abruptly) and assessment of aspiration risk in patients with dysphagia apply [2, 17]. GeneReviews recommends planning the transition to adult care from the age of 10 years in Lafora disease, with realistic plans for adult life, and continuously assessing the need for palliative care [2]. In ULD, transition involves educational and vocational planning, given the frequency of interrupted schooling [3].

7.18 Other PMEs: treatment by entity

None of the other PMEs has a clinical practice guideline or an approved treatment; treatment of myoclonus and seizures follows the general principles for PMEs, with the doses in section 7.3, and the GeneReviews chapters are based on the authors' experience [7, 14, 15]. The AMRF GeneReviews indicates valproate as first choice, because it reduces myoclonus and the frequency of generalised seizures, clonazepam as an adjunct, high-dose piracetam and N-acetylcysteine, the latter with variable results, and cites vagus nerve stimulation, which in a ULD series reduced seizures and improved cerebellar function [14]. The kidney disease follows the IPNA (International Pediatric Nephrology Association) recommendations for steroid-resistant nephrotic syndrome, with renin-angiotensin-aldosterone system blockade and, sometimes, immunosuppression (corticosteroids, calcineurin inhibitors, mycophenolate mofetil, rituximab), bearing in mind that genetic forms rarely benefit from prolonged immunosuppression; dialysis and transplantation prolong survival without improving the neurological picture [14].

Miglustat, a glucosylceramide synthase inhibitor that partially crosses the blood-brain barrier, is the only mechanism-targeted therapy with clinical reports in AMRF, as substrate reduction therapy, and its use is off-label [14, 24]. One of the Portuguese sisters received 600 mg/day for two years from the age of 24, after one year of ineffective enzyme replacement, with improvement of myoclonus and dysphagia, and died of pneumonia at 26 years [24, 105]. The patient of Quraishi and colleagues started on 100 mg twice daily and reached 200 mg three times daily in 6 months, the upper limit of the doses tolerated in the Gaucher disease trials and the maximum label dose (200 mg three times daily, Niemann-Pick type C in adults and adolescents ≥12 years), with arrest of progression for three years [24]. Venglustat, with better central nervous system penetration, has not been used in AMRF, and the manufacturer declined compassionate use in that case [14, 24]. As there is no published dose for AMRF beyond these two reports, the following table presents the dose from the reports as reference and the label doses for Niemann-Pick disease type C, the closest neurological indication, as a declared extrapolation.

In KCTD7-related PME, clonazepam, valproate, levetiracetam and lamotrigine were effective in some patients, as monotherapy or in combination, without the possibility of comparing drugs; one patient benefited from callosotomy, the role of vagus nerve stimulation and deep brain stimulation is unknown, and the ketogenic diet reduced seizures in a few cases [7]. In an Indian child, epilepsia partialis continua responded to corticosteroids [92]. Physiotherapy, management of spasticity and dystonia (baclofen, tizanidine, botulinum toxin), gastrostomy feeding and influenza vaccination are part of supportive care [7].

In North Sea PME, treatment combines 3–5 antiseizure medications, mainly levetiracetam, clonazepam and valproate, and myoclonus and seizures persist despite this; the ketogenic diet was used in four patients and improved quality of life in one, who continued it; deep brain stimulation has been applied, with no pooled results in the review; severe scoliosis sometimes requires surgery [31]. The authors of the review suggest evaluating ganaxolone, with no clinical data in the disease [31].

In MEAK, GeneReviews suggests levetiracetam, valproate, clonazepam and/or primidone for the myoclonus, and seizures are usually controlled [15]. In a 44-year-old man with refractory myoclonus and epilepsy, bilateral deep brain stimulation of the region of the subthalamic nucleus and substantia nigra pars reticulata reduced the UMRS myoclonus score by about 70% and abolished GTCS for more than 24 months, with no effect on ataxia and dysarthria [38]. AUT00201, a positive modulator of Kv3 channels, was tested as a single 100 mg dose in a placebo-controlled crossover trial in six adults, completed in 2024: adverse effects were similar with drug and placebo, with no serious events, and the exploratory endpoints of myoclonus and intracortical inhibition showed no apparent difference [37]. Fluoxetine, which inhibits Kv3.1 currents, improved seizures and motor function in one patient with a gain-of-function variant, and an n-of-1 trial with 2.5–5 mg/day has been completed; it does not apply to MEAK, which is caused by loss of function [15, 145].

The sources consulted describe no gene therapy in clinical trials for AMRF, KCTD7-related PME, North Sea PME, MEAK or EPM1B; for North Sea PME, the 2025 review identifies as prerequisites for future gene replacement or editing the identification of the glial cells involved, the degree of functional loss of the SNARE complexes and the level of GOSR2 restoration needed for clinical benefit, to be studied in induced pluripotent stem cell and organoid models [31].

In EPM1B, the sources consulted describe a response of seizures and tremor to medication in the Jordanian family, without specifying the drugs [39].

8. Reproductive Options and Family Counselling

8.1 Recurrence risks

Both diseases are autosomal recessive [2, 3]. The parents of an affected individual are obligate heterozygotes, but should be tested to confirm that each carries a variant and to calculate the recurrence risk reliably; de novo variants are rare but possible in recessive disorders [3]. When both parents are heterozygous, each sibling of an affected individual has, at conception, a 25% chance of being affected, a 50% chance of being an asymptomatic carrier and a 25% chance of not being a carrier [2, 3]. An affected sibling tends to have a picture similar to that of the proband, but intrafamilial variability exists, particularly in compound heterozygotes with ULD and in age at onset of Lafora disease [2, 3]. The proband's aunts and uncles have a 50% chance of being carriers [2, 3].

8.2 Particular situations

In Lafora disease, if the variant is found in only one parent (with confirmed biological parentage), a de novo event in the proband or parental post-zygotic mosaicism should be considered; apparent homozygosity may correspond to a deletion of one or more exons not detected by sequencing or to uniparental isodisomy [2]. In ULD, transmission of expanded alleles shows some germline instability, with contractions and expansions, without anticipation; alleles of 12–17 repeats, of uncertain significance, are unstable on transmission [3, 47].

8.3 Offspring of affected individuals and partner testing

The children of an individual with ULD are obligate carriers, unless the partner is a carrier or affected; because the carrier frequency in the general population is low, the risk of affected children is very low, except in populations with a higher frequency, such as the Finnish and North African populations [3]. Testing the partner of a carrier begins with targeted testing for the expansion and, if negative, CSTB sequencing [3]. In typical Lafora disease, affected individuals rarely have children; EPM2A and NHLRC1 testing should be offered to the partners of known carriers, particularly in consanguineous couples [2].

8.4 Cascade testing, prenatal and preimplantation diagnosis

Carrier testing of relatives requires prior identification of the family's variants [2, 3]. Once both variants have been identified in an affected relative, prenatal diagnosis and preimplantation genetic testing are possible; views on their use vary among professionals and families, and the decision is personal [2, 3]. The ideal time to determine genetic risk and discuss these options is before pregnancy, and counselling should be offered to young adults who are affected, carriers or at risk of being carriers [2, 3].

8.5 Predictive testing in minor siblings

In Lafora disease, subtle symptoms appear earlier in subsequently affected siblings than in the proband, and the EEG worsens before the clinical picture [2, 84]. The sources consulted do not establish specific recommendations on predictive testing of asymptomatic minor siblings; the existence of a trial enrolling patients from the age of 10 years and the preclinical evidence that early treatment is more effective (metformin in mice) [19, 127] are factors to weigh in the discussion with the family, according to the general principles of testing in minors.

8.6 Other PMEs

In AMRF, KCTD7-related PME, North Sea PME and EPM1B, the risk to siblings is 25% when both parents are heterozygous, and the parents should be tested; if only one is a carrier, a de novo variant, parental mosaicism, an undetected deletion or uniparental isodisomy are considered [7, 14]. In AMRF, the order and range of manifestations vary between siblings, and a sibling may have only kidney disease or only neurological disease, which makes it necessary to test siblings with proteinuria, renal failure or tremor [14]. Patients with KCTD7-related PME have no known offspring [7]. In MEAK, each child of an affected individual has a 50% chance of inheriting the variant; when neither parent has it in blood, the risk to siblings is slightly higher than that of the general population because of germline mosaicism, and a parent with somatic and germline mosaicism may have minimal symptoms [15]. Once the variants have been identified, carrier testing (in the recessive forms), prenatal diagnosis and preimplantation genetic testing are possible [7, 14, 15]. In women with AMRF planning a pregnancy, GeneReviews recommends folic acid 1 mg/day, increased to 4 mg/day three months before conception and during pregnancy, continuation of antiseizure medications with doses divided into four administrations or prolonged-release formulations, and monitoring of levels [14].

9. Special Considerations: Population Diversity and Epidemiology

9.1 Unverricht-Lundborg disease

ULD is the PME with the highest incidence and occurs worldwide [3]. Finland has the highest known prevalence: 2/100,000 inhabitants and an estimated incidence of 1:20,000 births (5/100,000), GeneReviews figures based on personal communication and validated by Orphanet [3, 11]. Orsini and colleagues report about 4/100,000 affected individuals in Finland, without indicating the primary source of that value [45]. Prevalence is also higher in Tunisia, Algeria and Morocco, with no published figures, in the Mediterranean region, on Réunion Island and in areas of high consanguinity [3, 45]. In Finland, about 99% of alleles are the dodecamer expansion [3]. In a haplotype study, 29 of 47 North African patients (61.7%) shared the same haplotype of the expanded allele, establishing a founder effect in that population, whereas the haplotypes of 48 patients from Western Europe were heterogeneous [8]. In the French and Italian long-term follow-up series, nine of 20 patients were of North African origin [94]. In the USA, Eldridge and colleagues identified 15 families in 1983 [43]. The first genetically confirmed Chinese patient was published in 2025, in a population of more than 1.4 billion people [73]. Lafora disease has not been described in Finland [2].

9.2 Lafora disease

GeneReviews estimates the worldwide frequency at 4/1,000,000, based on all published variants, with no exact prevalence data [2]. Orphanet records a validated point prevalence of 0.1/100,000 in France (expert estimate) and at least 300 cases or families described worldwide [12], and the 2021 meta-analysis brought together 298 genetically confirmed cases [88]. The disease is relatively rare in the non-consanguineous populations of the USA, Canada, China and Japan, and relatively common in the Mediterranean basin (Spain, France, Italy), in regions of Central Asia, India, Pakistan, North Africa and the Middle East, in isolates in the southern USA and Quebec and in other regions of high consanguinity [2]. In Italy and Japan NHLRC1 variants predominate; in Spain and France, EPM2A variants; in India and Arab populations, the distribution is similar [2]. Series data confirm this gradient: 70.1% of probands with EPM2A in 77 Spanish families and NHLRC1 in five of eight Japanese families [10]. In the French-Canadian population, the NHLRC1 founder variant p.(Cys26Ser) prompted a targeted DNA test for screening in that population [10]. The Apulia cohort suggests a higher regional prevalence, yet to be clarified [95]. The meta-analysis identified Asian origin as a poor prognostic factor [88].

9.3 Portugal and Brazil

The NHLRC1 variant c.205C>G p.(Pro69Ala), the most common in this gene, is present in all patients of Portuguese origin described and recurs in Italians, French and Spanish patients, suggesting a founder effect [2]. Within the class, the SLC7A6OS variant c.191A>G (EPM12) was initially described in one family from Turkey and another from Portugal, and later in three patients of Puerto Rican ancestry; haplotype analysis indicates a common ancestor about 1100 years ago, consistent with migrations between the eastern Mediterranean, the Iberian Peninsula and Puerto Rico [58]. The sources consulted include no Brazilian series of ULD or Lafora disease, which does not mean that no published data exist; OMIM cites a report of two cases published in 2000 in Arquivos de Neuro-Psiquiatria, one of which, diagnosed by axillary biopsy as an atypical form of Lafora disease, later proved, by molecular analysis, to be Unverricht-Lundborg disease [10].

The first pathogenic SCARB2 variant associated with human disease was described in a Portuguese family: two sisters, daughters of consanguineous parents, with PME without intellectual disability and nephrotic syndrome with C1q accumulation in the glomerular capillaries and mesangium; the younger was homozygous for NM_005506.4:c.533G>A p.(Trp178Ter) [90, 105]. The first had onset at 15 years, became fully dependent at 20 and died at 23; the second had onset at 17, developed nephrotic syndrome at 21 and died at 26 [21, 24]. The sources consulted include no Brazilian cases of the other PMEs in this article.

9.4 Founder populations in the class

In addition to those already mentioned, the sources describe the "Spanish" EPM2A variant p.(Arg241Ter), the NHLRC1 deletion c.468_469del in a tribal isolate in Oman, the NHLRC1 variant p.(Cys26Ser) in French-Canadian isolates [2], the GOSR2 variant p.(Gly144Trp) in patients from North Sea countries [31, 51] and the CERS1 form in an Algerian family [53].

In the other PMEs, AMRF was first described in French-Canadian families, in which most probands are homozygous for SCARB2 NM_005506.4:c.862C>T p.(Gln288Ter); the 2004 series extended it to the USA, Australia, Cuba and Germany, and there are late-onset forms in Japan with a distinct variant [14, 23]. The 12 patients with the GOSR2 founder variant c.430G>T in the 2013 series came from countries bordering the North Sea, as far as the coast of northern Norway, with a common ancestor about 3600 years ago [29, 32]. KCTD7-related PME has the largest number of published cases in consanguineous families from Turkey, Morocco, Pakistan and India, in addition to cases in Mexico and Italy [25, 92]. PRICKLE1 p.(Arg104Gln)-related EPM1B is a founder disorder in Israeli Arab and Jordanian families [40]. MEAK, because it results from a recurrent de novo variant, occurs without geographical clustering; Muona and colleagues estimated a rate of 1.75 × 10⁻⁷ per person for p.(Arg320His) [15].

9.5 Reference networks and patient associations

The relevant European Reference Network is EpiCARE (https://epi-care.eu), which includes centres with published PME series, such as Bologna, Kuopio, Helsinki, Lyon and Genoa [20, 79, 104]. Chelsea's Hope Lafora Children Research Fund (https://chelseashope.org) is the patient association listed by GeneReviews for Lafora disease, and the Lafora Epilepsy Cure Initiative takes part in research [84, 148]. The ULD GeneReviews lists the American Epilepsy Society and the Epilepsy Foundation, and the Oxford Desk Reference lists Epilepsy Action (United Kingdom) [1, 3].

For AMRF, GeneReviews also lists ERKNet (the European Reference Network for rare kidney diseases) and associations such as NephCure Kidney International and the PodoNet registry; for KCNC1-related disorders, the KCNC1 Foundation and the international KCNC1 registry at the Children's Hospital of Philadelphia [14, 15].

10. Illustrative Clinical Case

10.1 EPM2A-related Lafora disease

In 2025, Naderian and colleagues described a 15-year-old Iranian boy with normal development who had completed secondary schooling [86]. At 14 years he had tonic-clonic and myoclonic seizures, with normal EEG and MRI (Figure 1 of the article, brain MRI: https://pmc.ncbi.nlm.nih.gov/articles/PMC12121156/figure/Fig1/), and valproate initially controlled the seizures [86]. Eight months later, tonic-clonic seizures with myoclonus, hand tremor, developmental changes and lower limb paralysis, described by the authors as bilateral, appeared [86]. The EEG showed symmetrical 6 Hz theta background activity, with abundant bursts of generalised spike-wave and polyspike-wave that increased on successive recordings (Figure 2, serial EEG: https://pmc.ncbi.nlm.nih.gov/articles/PMC12121156/figure/Fig2/) [86]. Sequencing identified the homozygous EPM2A variant c.794A>G p.(His265Arg) in exon 4 (NM_005670.4, confirmed by Sanger sequencing and normalised in VariantValidator), and the relatives tested were asymptomatic heterozygotes [74, 86]. A 19-year-old brother and a 22-year-old cousin had each had a single tonic-clonic seizure and were not tested [86]. Three months later, with worsening and resistance to levetiracetam and valproate, phenobarbital (120 mg/day), metformin and zonisamide (100 mg/day, with increments of 100 mg every two weeks) were added, with partial seizure control and subsequent readmissions for worsening [86]. The case illustrates the initially normal MRI, EEG deterioration, drug resistance and off-label use of metformin; the choice of phenobarbital corresponds to the GeneReviews third line and is not routine early management [2].

10.2 Unverricht-Lundborg disease due to compound heterozygosity

In 2025, Miao and colleagues described the first genetically confirmed Chinese case of ULD: a 21-year-old woman, born to non-consanguineous parents, with normal development [73]. The first tonic-clonic seizure occurred at 8 years, triggered by bright light, followed by recurrent nocturnal seizures; oxcarbazepine was ineffective, an example of the drugs that the sources advise against [3, 73]. At 12 years, myoclonus triggered by flashing light and by menstruation appeared, which decreased with valproate; at 18 years, the response to valproate and levetiracetam became insufficient (one to two nocturnal tonic-clonic seizures per month and two to three myoclonic seizures per week) [73]. At 19 years, examination showed only truncal ataxia; MRI was normal, FDG-PET showed bilateral frontoparietal hypometabolism and video-EEG showed multifocal discharges with occipital predominance and normal posterior background activity (Figure 1a; https://pmc.ncbi.nlm.nih.gov/articles/PMC12121152/figure/Fig1/) [73]. Targeted testing found a dodecamer expansion (at least 55 repeats) on one allele and exome sequencing the c.116_117del variant on the other, with familial segregation (Figure 1b); the published protein notation ("p.K395fs*37") corresponds, in VariantValidator, to NM_000100.4:c.116_117del p.(Lys39SerfsTer37) [73, 74]. Perampanel at 2 mg/day markedly improved myoclonus and gait and suppressed the seizures; for 24 months the patient remained seizure-free, and recurrent myoclonus responded to an increase up to 6 mg/day, without psychiatric effects, with levetiracetam 1000 mg/day in combination [73]. Diagnosis took ten years, because of confusion with juvenile myoclonic epilepsy, and the case shows that exome sequencing alone would not have detected the expanded allele [3, 73].

10.3 Action myoclonus–renal failure syndrome treated with miglustat

In 2021, Quraishi and colleagues described an American lawyer and dancer, of French-Canadian ancestry on her mother's side, who at 20 years began to have imbalance and falls due to jerks of the lower limbs [24]. At 21 years, upper limb myoclonus and the first of several bilateral tonic-clonic seizures appeared; at 22 her handwriting deteriorated, and she needed a cane at 23 and a walker at 25, when dysphagia appeared [24]. At 25 years she had bilateral action myoclonus, titubation, wide-based gait, ocular dysmetria and dysarthria, with normal mental state, strength, sensation and reflexes; video-EEG showed sleep-activated generalised polyspikes and numerous jerks without an EEG correlate, and MRI showed only mild diffuse atrophy [24]. Clonazepam had no effect, levodopa, levetiracetam and perampanel had mild and transient effects, valproate a mild but more sustained effect, and clobazam worsened the myoclonus [24]. Exome sequencing identified two SCARB2 alleles inherited from different parents: the nonsense variant NM_005506.4:c.862C>T p.(Gln288Ter), of paternal origin, and the two-nucleotide insertion with the published designation c.431_432insAG, of maternal origin, predicted to produce a truncated protein; leucocyte glucocerebrosidase was 7.5 nmol/h/mg (normal ≥8.7) and creatinine 0.9 mg/dL [24]. The pedigree and Sanger sequencing are shown in Figure 1 of the article (https://pmc.ncbi.nlm.nih.gov/articles/PMC8320328/figure/F1/), and the clinical course before and after miglustat in Figure 2 (https://pmc.ncbi.nlm.nih.gov/articles/PMC8320328/figure/F2/) [24]. With miglustat started at 100 mg twice daily and increased over six months to 200 mg three times daily, the myoclonus and range of movement improved, the dysphagia resolved, the patient was again able to use her mobile phone and the television remote control, remained seizure-free, the EEG at 18 months showed no epileptiform discharges and progression stopped for three years, with no definite adverse effects [24]. Renal function worsened slowly without an increase in proteinuria; biopsy showed no focal segmental glomerulosclerosis, and a pelviureteric junction obstruction, probably congenital, was corrected surgically [24]. The authors acknowledge the subjectivity of assessing myoclonus and dysphagia in a single case; the use of miglustat in AMRF is not established practice and does not appear in guidelines [14, 24].

10.4 MEAK treated with deep brain stimulation

In 2023, Sobstyl and colleagues described a 44-year-old Polish man with genetically confirmed MEAK [38]. From 4–5 years of age he had intention tremor, followed by myoclonus, upper limb ataxia and increasing difficulty walking; the first bilateral tonic-clonic seizure occurred at 22 years, and both the myoclonus and the epilepsy became drug-refractory [38]. He underwent staged bilateral implantation of electrodes in the region of the subthalamic nucleus and substantia nigra pars reticulata; with follow-up of more than 24 months, the UMRS myoclonus score fell by almost 70% and the bilateral tonic-clonic seizures disappeared, but the ataxia and dysarthria did not improve [38]. It is an isolated, uncontrolled report, and GeneReviews cites it as a response in a single patient; deep brain stimulation is not an established treatment for MEAK [15, 38]. The case also illustrates the diagnostic delay possible when tremor precedes myoclonus and seizures, and the usefulness of genetic testing for counselling [38].

11. Prognosis and Long-Term Follow-Up

11.1 Prognosis

In treated ULD, progression of myoclonus is concentrated in the first 5–10 years, about one-third of patients become wheelchair-dependent, seizures become rare or disappear in many, and life expectancy is comparable to that of controls up to the age of 40 but lower in the long term, with deaths mainly from respiratory infections [3, 94]. Before valproate and with the use of phenytoin, many patients died 8–15 years after onset, usually before the age of 30 [3, 43]. Compound heterozygosity, early onset and longer duration are associated with a worse course [72, 76]. In Lafora disease, most patients die within 10 years of onset, from status epilepticus or complications of neurological deterioration; median survival in the meta-analysis was 11 years, with 62% surviving at 10 years and 57% at 15, and loss of autonomy had a median of 6 years [2, 88]. The NHLRC1 truncating/truncating genotype is associated with shorter survival, and p.(Asp146Asn) homozygosity with a longer course [85].

In the other PMEs, prognosis differs greatly: in AMRF, death occurs 7–15 years after onset, with no long-term survival data, and in the 2004 series nine patients died between 25 and 35 years of age [14, 21]; in KCTD7-related PME, published mortality is about 15%, between 18 months and 21 years, and the oldest surviving patient described was 25 years old [7]; in North Sea PME, all patients in the 2013 series became wheelchair users and four died in the third or early fourth decade, but some patients walk in the third decade and one patient is 61 years old [29, 32, 97]; in MEAK, about half need a wheelchair by late adolescence, with no apparent shortening of life [15, 36].

11.2 Surveillance

Serum NfL, sleep EEG and quantitative FDG-PET are investigational markers, not yet part of routine surveillance [102–104].

12. Summary and Final Considerations

The progressive myoclonic epilepsies form a class of rare genetic disorders, mostly autosomal recessive, united by the triad of stimulus-sensitive myoclonus, seizures and neurological deterioration, and Unverricht-Lundborg disease and Lafora disease are the two most common forms. Unverricht-Lundborg disease, caused mainly by homozygous expansion of a dodecamer in the CSTB promoter, begins between 6 and 15 years of age, usually preserves cognition and, with appropriate treatment, tends to stabilise after the first 5–10 years, with survival similar to that of controls up to the age of 40. Lafora disease, caused by biallelic variants in EPM2A or NHLRC1, results from the accumulation of insoluble polyglucosans in neurons, begins in adolescence with occipital seizures and myoclonus and progresses to dementia and death with a median of 11 years. The common first diagnostic step is the background EEG, which distinguishes both from juvenile myoclonic epilepsy; confirmation requires, in ULD, a test targeted at the expansion, which panels, exome and genome sequencing do not detect, and, in Lafora disease, sequencing with deletion analysis, supplemented by skin biopsy outside the axilla when necessary. Treatment remains symptomatic, with valproate as first choice and clonazepam, levetiracetam, piracetam and perampanel as adjuncts, and with avoidance of maintenance phenytoin, sodium channel blockers and GABAergic drugs; the restrictions on valproate in pregnancy require reproductive planning in women with ULD. In Lafora disease, metformin has European orphan designation and only observational evidence, intravenous VAL-1221 was ineffective, and the antisense oligonucleotide ION283, targeting GYS1, has been in a phase 1/2 trial since December 2024. Genetic counselling, rehabilitation, monitoring of swallowing and the airway, and follow-up in EpiCARE network centres complete the approach, in a context in which the data from Portugal are limited to the NHLRC1 founder variant p.(Pro69Ala) and the description of the SLC7A6OS variant in a Portuguese family.

The other PMEs in this article share symptomatic treatment but differ in course and clinical clues: action myoclonus–renal failure syndrome (SCARB2) combines young-adult PME and independently progressing kidney disease, and is the only one with a mechanism-targeted therapy described in patients, miglustat, supported by only two reports; KCTD7-related PME begins before the age of 4 with regression and refractoriness; North Sea PME (GOSR2) begins with ataxia, with scoliosis, areflexia and febrile worsening; MEAK (KCNC1), caused by a recurrent de novo variant, mimics ULD, improves with fever, does not shorten life and has had a phase 1 trial with a Kv3 modulator and a report of effective deep brain stimulation; EPM1B (PRICKLE1) is an early-ataxia form in founder families from the Middle East.

13. References

1. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 308–311 (Seizures with developmental delay/intellectual disability).
2. Minassian B. Progressive myoclonus epilepsy, Lafora type. 2007 Dec 28 [updated 2025 Jan 23]. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews®. Seattle (WA): University of Washington, Seattle; 1993–2026. Available from: https://www.ncbi.nlm.nih.gov/books/NBK1389/
3. Lehesjoki AE, Kälviäinen R. Progressive myoclonic epilepsy type 1. 2004 Jun 24 [updated 2020 Jul 2]. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews®. Seattle (WA): University of Washington, Seattle; 1993–2026. Available from: https://www.ncbi.nlm.nih.gov/books/NBK1142/
4. Online Mendelian Inheritance in Man (OMIM). #254780 Myoclonic epilepsy of Lafora 1; MELF1 [Internet]. Baltimore (MD): Johns Hopkins University. Available from: https://omim.org/entry/254780
5. Online Mendelian Inheritance in Man (OMIM). #619191 Epilepsy, progressive myoclonic, 12; EPM12 [Internet]. Baltimore (MD): Johns Hopkins University. Available from: https://omim.org/entry/619191
6. Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 1, Clinical approach to inborn errors of metabolism in paediatrics, section 1.5.2.2 and Tables 1.17–1.18.
7. Yoganathan S, Jain P, Costain G, Whitney R. KCTD7-related progressive myoclonic epilepsy. 2025 Nov 20. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews®. Seattle (WA): University of Washington, Seattle; 1993–2026. Available from: https://www.ncbi.nlm.nih.gov/books/NBK619245/
8. Online Mendelian Inheritance in Man (OMIM). #254800 Myoclonic epilepsy of Unverricht and Lundborg; ULD [Internet]. Baltimore (MD): Johns Hopkins University. Available from: https://omim.org/entry/254800
9. Online Mendelian Inheritance in Man (OMIM). *601145 Cystatin B; CSTB [Internet]. Baltimore (MD): Johns Hopkins University. Available from: https://omim.org/entry/601145
10. Online Mendelian Inheritance in Man (OMIM). #620681 Myoclonic epilepsy of Lafora 2; MELF2 [Internet]. Baltimore (MD): Johns Hopkins University; created 2024-01-17, edited 2025-01-03. Available from: https://omim.org/entry/620681
11. Orphanet. Progressive myoclonic epilepsy type 1 (ORPHA:308). Available from: https://www.orpha.net/en/disease/detail/308
12. Orphanet. Lafora disease (ORPHA:501). Available from: https://www.orpha.net/en/disease/detail/501
13. Orphanet. PRDM8-related progressive myoclonus epilepsy (ORPHA:324290). Available from: https://www.orpha.net/en/disease/detail/324290
14. Xiao C, Ahn H, Kibrom S, Toro C. SCARB2-related action myoclonus – renal failure syndrome. 2015 Dec 17 [updated 2023 Feb 9]. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews®. Seattle (WA): University of Washington, Seattle; 1993–2026. Available from: https://www.ncbi.nlm.nih.gov/books/NBK333437/
15. Ginn N, Goldberg EM. KCNC1-related disorders. 2025 Dec 18. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews®. Seattle (WA): University of Washington, Seattle; 1993–2026. Available from: https://www.ncbi.nlm.nih.gov/books/NBK619809/
16. Prades S, Melo de Gusmao C, Grimaldi S, Shiloh-Malawsky Y, Felton T, Houlden H. DRPLA. 1999 Aug 6 [updated 2023 Sep 21]. In: Adam MP, Feldman J, Mirzaa GM, et al., editors. GeneReviews®. Seattle (WA): University of Washington, Seattle; 1993–2026. Available from: https://www.ncbi.nlm.nih.gov/books/NBK1491/
17. Ferlazzo E, Kasteleijn-Nolst Trenité D, de Haan GJ, Nitschke F, Ahonen S, Gasparini S, et al. Update on Pharmacological Treatment of Progressive Myoclonus Epilepsies. Curr Pharm Des. 2017;23(37):5662-5666. doi:10.2174/1381612823666170809114654
18. European Medicines Agency. EU/3/16/1803: Orphan designation for the treatment of progressive myoclonic epilepsy type 2 (Lafora disease) — metformin [Internet]. Amsterdam: EMA; designated 2016-12-12, updated 2022-11-11. Available from: https://www.ema.europa.eu/en/medicines/human/orphan-designations/eu-3-16-1803
19. ClinicalTrials.gov. A Phase 1/2 Open Label Study to Evaluate the Safety and Efficacy of Intrathecally Administered ION283 in Patients With Lafora Disease (NCT06609889) [Internet]. Bethesda (MD): National Library of Medicine; updated 2025-12-18. Available from: https://clinicaltrials.gov/study/NCT06609889
20. EpiCARE — European Reference Network for rare and complex epilepsies [Internet]. Available from: https://epi-care.eu
21. Online Mendelian Inheritance in Man (OMIM). #254900 Epilepsy, progressive myoclonic, 4, with or without renal failure; EPM4 [Internet]. Baltimore (MD): Johns Hopkins University. Available from: https://omim.org/entry/254900
22. Orphanet. Action myoclonus-renal failure syndrome (ORPHA:163696). Available from: https://www.orpha.net/en/disease/detail/163696
23. Badhwar A, Berkovic SF, Dowling JP, Gonzales M, Narayanan S, Brodtmann A, et al. Action myoclonus-renal failure syndrome: characterization of a unique cerebro-renal disorder. Brain. 2004;127(Pt 10):2173-82. doi:10.1093/brain/awh263
24. Quraishi IH, Szekely AM, Shirali AC, Mistry PK, Hirsch LJ. Miglustat therapy for SCARB2-associated action myoclonus-renal failure syndrome. Neurol Genet. 2021;7(5):e614. doi:10.1212/NXG.0000000000000614
25. Online Mendelian Inheritance in Man (OMIM). #611726 Epilepsy, progressive myoclonic, 3, with or without intracellular inclusions; EPM3 [Internet]. Baltimore (MD): Johns Hopkins University. Available from: https://omim.org/entry/611726
26. Orphanet. Progressive myoclonic epilepsy type 3 (ORPHA:263516). Available from: https://www.orpha.net/en/disease/detail/263516
27. Orphanet. CLN14 disease (ORPHA:699708). Available from: https://www.orpha.net/en/disease/detail/699708
28. Yoganathan S, Whitney R, Thomas M, Danda S, Chettali AM, Prasad AN, et al. KCTD7-related progressive myoclonic epilepsy: report of 42 cases and review of literature. Epilepsia. 2024;65(3):709-24. doi:10.1111/epi.17880
29. Online Mendelian Inheritance in Man (OMIM). #614018 Epilepsy, progressive myoclonic, 6; EPM6 [Internet]. Baltimore (MD): Johns Hopkins University. Available from: https://omim.org/entry/614018
30. Orphanet. Progressive myoclonic epilepsy type 6 (ORPHA:280620). Available from: https://www.orpha.net/en/disease/detail/280620
31. Polet SS, Siegal EZ, Fuchs SA, Tijssen MAJ, de Koning TJ. The Genotypic and Phenotypic Spectrum of GOSR2 Mutations: Clinical and Pathophysiological Insights. J Inherit Metab Dis. 2025;48(6):e70115. doi:10.1002/jimd.70115
32. Boissé Lomax L, Bayly MA, Hjalgrim H, Møller RS, Vlaar AM, Aaberg KM, et al. 'North Sea' progressive myoclonus epilepsy: phenotype of subjects with GOSR2 mutation. Brain. 2013;136(Pt 4):1146-54. doi:10.1093/brain/awt021
33. Online Mendelian Inheritance in Man (OMIM). #616187 Epilepsy, progressive myoclonic 7; EPM7 [Internet]. Baltimore (MD): Johns Hopkins University. Available from: https://omim.org/entry/616187
34. Orphanet. Progressive myoclonic epilepsy type 7 (ORPHA:435438). Available from: https://www.orpha.net/en/disease/detail/435438
35. Muona M, Berkovic SF, Dibbens LM, Oliver KL, Maljevic S, Bayly MA, et al. A recurrent de novo mutation in KCNC1 causes progressive myoclonus epilepsy. Nat Genet. 2015;47(1):39-46. doi:10.1038/ng.3144
36. Oliver KL, Franceschetti S, Milligan CJ, Muona M, Mandelstam SA, Canafoglia L, et al. Myoclonus epilepsy and ataxia due to KCNC1 mutation: analysis of 20 cases and K+ channel properties. Ann Neurol. 2017;81(5):677-89. doi:10.1002/ana.24929
37. ClinicalTrials.gov. Safety, Blood Levels and Effects of AUT00201 in Patients With MEAK (NCT05873062) [Internet]. Available from: https://clinicaltrials.gov/study/NCT05873062
38. Sobstyl M, Kożuch N, Iwaniuk-Gugała M, Stapińska-Syniec A, Konopko M, Jezierski P, et al. Deep brain stimulation in a patient with progressive myoclonic epilepsy and ataxia due to potassium channel mutation (MEAK). A case report and review of the literature. Epilepsy Behav Rep. 2023;24:100627. doi:10.1016/j.ebr.2023.100627
39. Online Mendelian Inheritance in Man (OMIM). #612437 Epilepsy, progressive myoclonic, 1B; EPM1B [Internet]. Baltimore (MD): Johns Hopkins University. Available from: https://omim.org/entry/612437
40. Bassuk AG, Wallace RH, Buhr A, Buller AR, Afawi Z, Shimojo M, et al. A homozygous mutation in human PRICKLE1 causes an autosomal-recessive progressive myoclonus epilepsy-ataxia syndrome. Am J Hum Genet. 2008;83(5):572-81. doi:10.1016/j.ajhg.2008.10.003
41. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 128–131 (Developmental regression).
42. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 416–419 (Epilepsy in adults).
43. Eldridge R, Iivanainen M, Stern R, Koerber T, Wilder BJ. "Baltic" myoclonus epilepsy: hereditary disorder of childhood made worse by phenytoin. Lancet. 1983;2(8354):838-42. doi:10.1016/s0140-6736(83)90749-3
44. Genton P, Striano P, Minassian BA. The history of progressive myoclonus epilepsies. Epileptic Disord. 2016;18(S2):3-10. doi:10.1684/epd.2016.0834
45. Orsini A, Valetto A, Bertini V, Esposito M, Carli N, Minassian BA, et al. The best evidence for progressive myoclonic epilepsy: A pathway to precision therapy. Seizure. 2019;71:247-257. doi:10.1016/j.seizure.2019.08.012
46. Pennacchio LA, Lehesjoki AE, Stone NE, Willour VL, Virtaneva K, Miao J, et al. Mutations in the gene encoding cystatin B in progressive myoclonus epilepsy (EPM1). Science. 1996;271(5256):1731-4. doi:10.1126/science.271.5256.1731
47. Lalioti MD, Scott HS, Buresi C, Rossier C, Bottani A, Morris MA, et al. Dodecamer repeat expansion in cystatin B gene in progressive myoclonus epilepsy. Nature. 1997;386(6627):847-51. doi:10.1038/386847a0
48. Minassian BA, Lee JR, Herbrick JA, Huizenga J, Soder S, Mungall AJ, et al. Mutations in a gene encoding a novel protein tyrosine phosphatase cause progressive myoclonus epilepsy. Nat Genet. 1998;20(2):171-4. doi:10.1038/2470
49. Online Mendelian Inheritance in Man (OMIM). #616640 Epilepsy, progressive myoclonic, 10; EPM10 [Internet]. Baltimore (MD): Johns Hopkins University. Available from: https://omim.org/entry/616640
50. Walter JH, Labrune P, Laforêt P. The glycogen storage diseases and related disorders. In: Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 5, section 5.3 Brain glycogenoses. doi:10.1007/978-3-662-63123-2_5
51. Zimmern V, Minassian B. Progressive Myoclonus Epilepsy: A Scoping Review of Diagnostic, Phenotypic and Therapeutic Advances. Genes (Basel). 2024;15(2). doi:10.3390/genes15020171
52. Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 30, section 30.6.1 Disorders of SV exocytosis.
53. Online Mendelian Inheritance in Man (OMIM). #616230 Epilepsy, progressive myoclonic, 8; EPM8 [Internet]. Baltimore (MD): Johns Hopkins University. Available from: https://omim.org/entry/616230
54. Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 40, sections 40.1.3, 40.2.1, 40.2.8 and 40.5.
55. Online Mendelian Inheritance in Man (OMIM). #616540 Epilepsy, progressive myoclonic, 9; EPM9 [Internet]. Baltimore (MD): Johns Hopkins University. Available from: https://omim.org/entry/616540
56. Online Mendelian Inheritance in Man (OMIM). #618876 Epilepsy, progressive myoclonic, 11; EPM11 [Internet]. Baltimore (MD): Johns Hopkins University. Available from: https://omim.org/entry/618876
57. Altıntaş M, Yıldırım M, Bektaş Ö, Teber S. Progressive Myoclonus Epilepsy and Beyond: A Systematic Review of SEMA6B-related Disorders. Neuropediatrics. 2025;56(2):83-93. doi:10.1055/a-2442-5741
58. Grinton BE, Ellis CA, Parnes M, Stripe LL, Munro JE, Mazzola L, et al. SLC7A6OS Founder Mutation: A Rare Cause of Progressive Myoclonus Epilepsy Dated to 1100 Years Ago. Mov Disord. 2026;. doi:10.1002/mds.70516
59. Online Mendelian Inheritance in Man (OMIM). #159950 Spinal muscular atrophy with progressive myoclonic epilepsy; SMAPME [Internet]. Baltimore (MD): Johns Hopkins University. Available from: https://omim.org/entry/159950
60. Jones S, Wijburg FA. Glycosaminoglycans and oligosaccharides disorders: glycosaminoglycans synthesis defects, mucopolysaccharidoses, oligosaccharidoses and sialic acid disorders. In: Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 41, section 41.3 Oligosaccharidoses and mucolipidoses (41.3.1–41.3.5). doi:10.1007/978-3-662-63123-2_41
61. Nyhan WL, Hoffmann GF, Al-Aqeel AI, Barshop BA. Atlas of Inherited Metabolic Diseases. 4th ed. Boca Raton: CRC Press; 2020. p. 733–740 (Galactosialidosis).
62. Nyhan WL, Hoffmann GF, Al-Aqeel AI, Barshop BA. Atlas of Inherited Metabolic Diseases. 4th ed. Boca Raton: CRC Press; 2020. p. 393–398 (MERRF).
63. Bezerra Filho COA, Damo PS, Ribeiro EM. MERRF (myoclonic epilepsy with ragged red fibers). In: Ribeiro EM, coord. Síndromes Genéticas: Reconhecer para Tratar. Fortaleza: EdUnichristus; 2021. p. 134–136.
64. Blank AC, Breur JMPJ, Fuchs SA, Koop K, Baas AF. Mitochondrial cardiomyopathies. In: Baars HF, van der Smagt JJ, Doevendans PAFM, van Tintelen JP, editors. Clinical Cardiogenetics. 3rd ed. Cham: Springer; 2020. p. 179–183 (11.3 Clinical presentation). doi:10.1007/978-3-030-45457-9_11
65. Nyhan WL, Hoffmann GF, Al-Aqeel AI, Barshop BA. Atlas of Inherited Metabolic Diseases. 4th ed. Boca Raton: CRC Press; 2020. p. 678–687 (Gaucher disease).
66. Jones KL, Jones MC, del Campo M. Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 852–853 (Berardinelli-Seip congenital lipodystrophy syndrome).
67. Nyhan WL, Hoffmann GF, Al-Aqeel AI, Barshop BA. Atlas of Inherited Metabolic Diseases. 4th ed. Boca Raton: CRC Press; 2020. p. 787–792 (Disorders of creatine synthesis or transport).
68. GeneGuidance. CSTB [database]; version of 2026-09-26.
69. Malyutina A, Lund C, Tegelberg S, Hakala P, Nyman TA, Lehesjoki AE, et al. Transcriptomic and proteomic insights into progressive myoclonus epilepsy type 1. Dis Model Mech. 2026;19(4). doi:10.1242/dmm.052681
70. Singh S, Plotnikova L, Tiukuvaara T, Kälviäinen R, Hämäläinen RH. Oxidative Stress Drives Cell Cycle Stalling, Apoptosis and Metabolic Suppression in Cystatin B Deficient EPM1 Patient iPSCs. Cell Prolif. 2026;59(10):e70232. doi:10.1111/cpr.70232
71. Forero A, Pravata V, Pipicelli F, Frenna E, Soloperto A, Ianni M, et al. Defective EV-mediated transport of SHH alters neural fate specification in EPM1 epilepsy. Sci Adv. 2026;12(36):eadu3955. doi:10.1126/sciadv.adu3955
72. Hyppönen J, Äikiä M, Joensuu T, Julkunen P, Danner N, Koskenkorva P, et al. Refining the phenotype of Unverricht-Lundborg disease (EPM1): a population-wide Finnish study. Neurology. 2015;84(15):1529-36. doi:10.1212/WNL.0000000000001466
73. Miao P, Ding Y, Cen Z, Chen Y, Luo W, Zhang B, et al. A novel c.116 - 117 del variant in Unverricht-Lundborg disease: first ULD report in large Chinese population and review of the pathogenetic variants in CSTB gene. Acta Epileptol. 2025;7(1):32. doi:10.1186/s42494-025-00216-4
74. Freeman PJ, Hart RK, Gretton LJ, Brookes AJ, Dalgleish R. VariantValidator: accurate validation, mapping, and formatting of sequence variation descriptions. Hum Mutat. 2018;39(1):61-68. doi:10.1002/humu.23348
75. Koskenkorva P, Hyppönen J, Aikiä M, Mervaala E, Kiviranta T, Eriksson K, et al. Severer phenotype in Unverricht-Lundborg disease (EPM1) patients compound heterozygous for the dodecamer repeat expansion and the c.202C>T mutation in the CSTB gene. Neurodegener Dis. 2011;8(6):515-22. doi:10.1159/000323470
76. Canafoglia L, Gennaro E, Capovilla G, Gobbi G, Boni A, Beccaria F, et al. Electroclinical presentation and genotype-phenotype relationships in patients with Unverricht-Lundborg disease carrying compound heterozygous CSTB point and indel mutations. Epilepsia. 2012;53(12):2120-7. doi:10.1111/j.1528-1167.2012.03718.x
77. Abdel-Hamid MS, Abdel-Ghafar SF, Sayed ISM, Zaki MS, Abdel-Salam GMH. Delineating the Clinical and Brain Imaging Characteristics of the Neonatal Form of CSTB -Related Neurodevelopmental Disorders. Clin Genet. 2025;108(2):184-193. doi:10.1111/cge.14720
78. Zaben B, Tuqan AR, Hammad O, Jweiles R, Awad S, Barabrah AM, et al. Infantile-Onset Unverricht-Lundborg Disease: A Unique Case With Severe Neurological Impairment and Positive Family History. J Investig Med High Impact Case Rep. 2025;13:23247096251407837. doi:10.1177/23247096251407837
79. Gunnar J, Liu Y, Eronen H, Joensuu T, Äikiä M, Hyppönen J, et al. Exploring the role of apolipoprotein ε4 in progressive myoclonic epilepsy type 1. Epileptic Disord. 2026;28(1):55-67. doi:10.1002/epd2.70112
80. GeneGuidance. EPM2A [database]; version of 2026-09-26.
81. Online Mendelian Inheritance in Man (OMIM). *607566 EPM2A glucan phosphatase, laforin; EPM2A [Internet]. Baltimore (MD): Johns Hopkins University. Available from: https://omim.org/entry/607566
82. GeneGuidance. NHLRC1 [database]; version of 2026-09-26.
83. Online Mendelian Inheritance in Man (OMIM). *608072 NHL repeat-containing protein 1; NHLRC1 [Internet]. Baltimore (MD): Johns Hopkins University. Available from: https://omim.org/entry/608072
84. Williams MI, Donohue KJ, Sanz P, Messahel S, Serratosa JM, Duran J, et al. The 9th annual Lafora science symposium: a rare epilepsy community makes progress towards clinical readiness. Epilepsy Behav. 2025;171:110654. doi:10.1016/j.yebeh.2025.110654
85. Pondrelli F, Minardi R, Muccioli L, Zenesini C, Vignatelli L, Licchetta L, et al. Prognostic value of pathogenic variants in Lafora Disease: systematic review and meta-analysis of patient-level data. Orphanet J Rare Dis. 2023;18(1):263. doi:10.1186/s13023-023-02880-6
86. Naderian R, Vafaeian F, Hoseini SM, Sanami S. Case report of Lafora disease: a rare genetic disorder manifesting as progressive myoclonic epilepsy. BMC Neurol. 2025;25(1):230. doi:10.1186/s12883-025-04253-x
87. Iglesias-Cabeza N, Zafra-Puerta L, Worby CA, Fernández-Burgos D, Sciaccaluga M, Imperatore V, et al. Epm2b(P71A) and Epm2b(D148N) knock-in mouse models of Lafora disease exhibit distinct and pronounced neurological alterations. Prog Neurobiol. 2026;261:102912. doi:10.1016/j.pneurobio.2026.102912
88. Pondrelli F, Muccioli L, Licchetta L, Mostacci B, Zenesini C, Tinuper P, et al. Natural history of Lafora disease: a prognostic systematic review and individual participant data meta-analysis. Orphanet J Rare Dis. 2021;16(1):362. doi:10.1186/s13023-021-01989-w
89. Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 44, Disorders of cellular trafficking, tables of neurological and renal manifestations.
90. Balreira A, Gaspar P, Caiola D, Chaves J, Beirão I, Lima JL, et al. A nonsense mutation in the LIMP-2 gene associated with progressive myoclonic epilepsy and nephrotic syndrome. Hum Mol Genet. 2008;17(14):2238-43. doi:10.1093/hmg/ddn124
91. Atasu B, Acarlı ANO, Bilgic B, Baykan B, Demir E, Ozluk Y, et al. Genotype-phenotype correlations of SCARB2 associated clinical presentation: a case report and in-depth literature review. BMC Neurol. 2022;22(1):122. doi:10.1186/s12883-022-02628-y
92. Sangeeth TA, Viswanathan LG, Padmanabha H, Janardhanan A, Asranna A, Arunachal G, et al. KCTD7-related progressive myoclonic epilepsy: clinical and genetic characterization of six Indian patients and review of literature. Seizure. 2026;135:28-33. doi:10.1016/j.seizure.2026.01.005
93. Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 43, section 43.1 Congenital disorders of protein N-glycosylation, table of vesicular trafficking defects.
94. Magaudda A, Ferlazzo E, Nguyen VH, Genton P. Unverricht-Lundborg disease, a condition with self-limited progression: long-term follow-up of 20 patients. Epilepsia. 2006;47(5):860-6. doi:10.1111/j.1528-1167.2006.00553.x
95. d'Orsi G, Di Claudio MT, Liantonio A, Imbrici P, Altomare CD, Palumbo O, et al. Clinical course and management challenges in Lafora disease: a narrative analysis in an Apulian cohort. Orphanet J Rare Dis. 2025;20(1):447. doi:10.1186/s13023-025-03976-x
96. Mazzone S, Muccioli L, Tappatà M, Pasini E, Bifolchi G, Pietrafusa N, et al. Neuropsychological functioning and progression in Lafora disease. Epilepsia. 2026;67(10):5477-5488. doi:10.1002/epi.70376
97. Polet SS, Anderson DG, Koens LH, van Egmond ME, Drost G, Brusse E, et al. A detailed description of the phenotypic spectrum of North Sea Progressive Myoclonus Epilepsy in a large cohort of seventeen patients. Parkinsonism Relat Disord. 2020;72:44-48. doi:10.1016/j.parkreldis.2020.02.005
98. Saudubray JM, Baumgartner MR, García-Cazorla Á, Walter J, editors. Inborn Metabolic Diseases: Diagnosis and Treatment. 7th ed. Berlin: Springer; 2022. Chapter 2, Inborn errors of metabolism in adults, section 2.3.6 Epilepsy, Table 2.8.
99. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 74–77 (Ataxic child).
100. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 410–415 (Epilepsy in infants and children).
101. Michelucci R, Pasini E, Riguzzi P, Andermann E, Kälviäinen R, Genton P. Myoclonus and seizures in progressive myoclonus epilepsies: pharmacology and therapeutic trials. Epileptic Disord. 2016;18(S2):145-153. doi:10.1684/epd.2016.0861
102. Pophal E, Taha M, Doll E, Nordli DR Jr, Minassian B, Nordli DR 3rd. Sleep architecture and qEEG patterns in PME type 1 diagnosis. Epileptic Disord. 2025;27(5):1032-1037. doi:10.1002/epd2.70079
103. d'Orsi G, Liantonio A, Imbrici P, Lolli G, Dinoi G, Conte E, et al. Empagliflozin in Lafora disease: A first-in-human 6-month pilot study with electroclinical and FDG-PET findings. Epileptic Disord. 2026;. doi:10.1002/epd2.70442
104. Muccioli L, Ganceviciute B, Becker F, Minardi R, Tappatà M, Bachhuber F, et al. Neurofilament Light Chain as a Biomarker of Disease Progression in Lafora Disease. Neurol Genet. 2025;11(6):e200319. doi:10.1212/NXG.0000000000200319
105. Chaves J, Beirão I, Balreira A, Gaspar P, Caiola D, Sá-Miranda MC, et al. Progressive myoclonus epilepsy with nephropathy C1q due to SCARB2/LIMP-2 deficiency: clinical report of two siblings. Seizure. 2011;20(9):738-40. doi:10.1016/j.seizure.2011.06.018
106. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 138–141 (Dystonia).
107. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 490–495 (Mitochondrial DNA diseases).
108. Koskiniemi M, Van Vleymen B, Hakamies L, Lamusuo S, Taalas J. Piracetam relieves symptoms in progressive myoclonus epilepsy: a multicentre, randomised, double blind, crossover study comparing the efficacy and safety of three dosages of oral piracetam with placebo. J Neurol Neurosurg Psychiatry. 1998;64(3):344-8. doi:10.1136/jnnp.64.3.344
109. Kälviäinen R, Genton P, Andermann E, Andermann F, Magaudda A, Frucht SJ, et al. Brivaracetam in Unverricht-Lundborg disease (EPM1): Results from two randomized, double-blind, placebo-controlled studies. Epilepsia. 2016;57(2):210-21. doi:10.1111/epi.13275
110. DEPAKOTE (divalproex sodium) — prescribing information [Internet]. NDA 018723; label effective 2026-03-31. Available from: https://dailymed.nlm.nih.gov/
111. KLONOPIN (clonazepam) tablets — prescribing information [Internet]. NDA 017533; label effective 2025-12-01. Available from: https://dailymed.nlm.nih.gov/
112. UCB Inc. KEPPRA (levetiracetam) tablets and oral solution — prescribing information [Internet]. NDA 021035; label effective 2025. Available from: https://dailymed.nlm.nih.gov/
113. UCB Pharma Ltd. Nootropil 1200 mg film-coated tablets — Summary of Product Characteristics [Internet]. electronic medicines compendium (eMC). Available from: https://www.medicines.org.uk/emc/product/101131/smpc
114. Generis Farmacêutica. Piracetam Generis 1200 mg comprimidos — Resumo das Características do Medicamento [Internet]. Lisboa: INFARMED; approved 2021-02-26. Available from: https://extranet.infarmed.pt/INFOMED-fo/ (med_id 38133)
115. Crespel A, Gelisse P, Tang NP, Genton P. Perampanel in 12 patients with Unverricht-Lundborg disease. Epilepsia. 2017;58(4):543-547. doi:10.1111/epi.13662
116. Canafoglia L, Barbella G, Ferlazzo E, Striano P, Magaudda A, d'Orsi G, et al. An Italian multicentre study of perampanel in progressive myoclonus epilepsies. Epilepsy Res. 2019;156:106191. doi:10.1016/j.eplepsyres.2019.106191
117. Goldsmith D, Minassian BA. Efficacy and tolerability of perampanel in ten patients with Lafora disease. Epilepsy Behav. 2016;62:132-5. doi:10.1016/j.yebeh.2016.06.041
118. FYCOMPA (perampanel) — prescribing information [Internet]. NDA 202834/208277; label effective 2022-12-28. Available from: https://dailymed.nlm.nih.gov/
119. Advanz Pharma (US) Corp. ZONEGRAN (zonisamide) capsules — prescribing information [Internet]. Revised 2023-11. Available from: https://dailymed.nlm.nih.gov/
120. Janssen Pharmaceuticals. TOPAMAX (topiramate) tablets and sprinkle capsules — prescribing information [Internet]. Revised 2026-03. Available from: https://dailymed.nlm.nih.gov/
121. BRIVIACT (brivaracetam) — prescribing information [Internet]. NDA 205836/205837/205838; label effective 2026-08-20. Available from: https://dailymed.nlm.nih.gov/
122. Wu J, Lynn TC, Nitschke S, Mitra S, Minassian BA. Lithium exacerbates Lafora body formation in the Epm2a(-/-) Lafora disease mouse model. Neurosci Lett. 2025;856-858:138250. doi:10.1016/j.neulet.2025.138250
123. European Medicines Agency. Zavesca (miglustat): EPAR – product information (summary of product characteristics) [Internet]. Amsterdam: EMA. Available from: https://www.ema.europa.eu/en/documents/product-information/zavesca-epar-product-information_en.pdf
124. Actelion Pharmaceuticals US. ZAVESCA (miglustat) capsules: prescribing information [Internet]. Revised 04/2023. DailyMed. Available from: https://dailymed.nlm.nih.gov/dailymed/search.cfm?query=zavesca
125. ClinicalTrials.gov. A Trial of Ketogenic Diet in Lafora Disease (NCT00007124) [Internet]. Available from: https://clinicaltrials.gov/study/NCT00007124
126. Bisulli F, Muccioli L, d'Orsi G, Canafoglia L, Freri E, Licchetta L, et al. Treatment with metformin in twelve patients with Lafora disease. Orphanet J Rare Dis. 2019;14(1):149. doi:10.1186/s13023-019-1132-3
127. Burgos DF, Machío-Castello M, Iglesias-Cabeza N, Giráldez BG, González-Fernández J, Sánchez-Martín G, et al. Early Treatment with Metformin Improves Neurological Outcomes in Lafora Disease. Neurotherapeutics. 2023;20(1):230-244. doi:10.1007/s13311-022-01304-w
128. European Medicines Agency. EU/3/20/2377: Orphan designation for the treatment of progressive myoclonic epilepsy type 2 (Lafora disease) — 2'-O-(2-methoxyethyl) modified antisense oligonucleotide targeting glycogen synthase 1 pre-mRNA [Internet]. Amsterdam: EMA; designated 2020-12-09. Available from: https://www.ema.europa.eu/en/medicines/human/orphan-designations/eu-3-20-2377
129. Chelsea's Hope Lafora Children Research Fund. First patients receive ION283 dosing for Lafora disease [press release]. Sacramento (CA); 20 Feb 2025. Available from: https://chelseashope.org/wp-content/uploads/2025/02/Safety-Study-Press-Release-February-2025.pdf
130. Muccioli L, Tappatà M, Farolfi A, Singh PK, Pasini E, Mazzone S, et al. New therapeutic strategies for Lafora disease: Evaluation of the safety, efficacy, pharmacokinetics and metabolomic profile of intravenous VAL-1221 treatment. Neurotherapeutics. 2026;23(1):e00808. doi:10.1016/j.neurot.2025.e00808
131. ClinicalTrials.gov. Intravenous VAL-1221 Lafora Expanded Access Protocol (LEAP) (NCT05930223) [Internet]. Available from: https://clinicaltrials.gov/study/NCT05930223
132. d'Orsi G, Liantonio A, Imbrici P, Gambacorta N, Dinoi G, Altomare CD, et al. Empagliflozin Repurposing for Lafora Disease: A Pilot Clinical Trial and Preclinical Investigation of Novel Therapeutic Targets. Methods Protoc. 2025;8(3). doi:10.3390/mps8030048
133. Della Vecchia S, Imbrici P, Liantonio A, Naef V, Damiani D, Licitra R, et al. Dapagliflozin ameliorates Lafora disease phenotype in a zebrafish model. Biomed Pharmacother. 2025;183:117800. doi:10.1016/j.biopha.2024.117800
134. Zafra-Puerta L, Iglesias-Cabeza N, Sciaccaluga M, Bellingacci L, Canonichesi J, Sánchez-Martín G, et al. Advances in gene therapy for Lafora disease: Intravenous recombinant adeno-associated virus-mediated delivery of EPM2A and EPM2B genes. Clin Transl Med. 2025;15(11):e70514. doi:10.1002/ctm2.70514
135. Alao EO, Sheibani M, Wu J, Marriam U, Evans D, Kasiri S, et al. Lafora disease gene therapy: EPM2A but not EPM2B overexpression results in Lafora body formation. Neurotherapeutics. 2026;23(2):e00866. doi:10.1016/j.neurot.2026.e00866
136. Allegri L, Baldan F, Mio C, Imperatore V, Costa C, Prontera P, et al. SINEUP-Mediated Overexpression of Endogenous α-Amylase as a Therapeutic Approach in Lafora Disease. Genes (Basel). 2026;17(3). doi:10.3390/genes17030321
137. ClinicalTrials.gov. Prospective, Longitudinal, Observational Study of the Natural History and Functional Status of Patients With Lafora Disease (NCT03876522) [Internet]. Available from: https://clinicaltrials.gov/study/NCT03876522
138. ClinicalTrials.gov. Effect of Ropinirole Hydrochloride in Progressive Myoclonic Epilepsy of Unverricht-Lundborg Type (NCT00639119) [Internet]. Available from: https://clinicaltrials.gov/study/NCT00639119
139. ClinicalTrials.gov. Intravenous Immunoglobulin for Unverricht-Lundborg Disease (NCT03351569) [Internet]. Available from: https://clinicaltrials.gov/study/NCT03351569
140. ClinicalTrials.gov. Registry and Natural History Study for Progressive Myoclonus Epilepsy Type 1 (EPM1) (NCT06593951) [Internet]. Available from: https://clinicaltrials.gov/study/NCT06593951
141. Boespflug-Tanguy O, Valent A, Rambaud J, Léger PL, Plu I, Seilhean D, et al. Death following high-dose AAV9 gene therapy in a patient with advanced SMA-PME. Mol Ther. 2026;34(8):4442-4448. doi:10.1016/j.ymthe.2026.05.016
142. European Medicines Agency. Valproate and related substances — referral (Article 31): new measures to avoid valproate exposure in pregnancy endorsed (CMDh, 21 March 2018) [Internet]. Available from: https://www.ema.europa.eu/en/medicines/human/referrals/valproate-related-substances-0
143. European Medicines Agency. Potential risk of neurodevelopmental disorders in children born to men treated with valproate medicines: PRAC recommends precautionary measures [Internet]. 12 Jan 2024. Available from: https://www.ema.europa.eu/en/news/potential-risk-neurodevelopmental-disorders-children-born-men-treated-valproate-medicines-prac-recommends-precautionary-measures
144. European Medicines Agency. Medicines output: orphan designations — JSON report [Internet]. Amsterdam: EMA. Available from: https://www.ema.europa.eu/en/documents/report/medicines-output-orphan_designations-json-report_en.json
145. ClinicalTrials.gov. A Single Patient Trial of Fluoxetine in KCNC1-related Disorder (NCT06341127) [Internet]. Available from: https://clinicaltrials.gov/study/NCT06341127
146. Firth HV, Hurst JA. Oxford Desk Reference: Clinical Genetics and Genomics. 2nd ed. Oxford: Oxford University Press; 2017. p. 732–735 (Fetal anticonvulsant syndrome).
147. Jones KL, Jones MC, del Campo M. Smith's Recognizable Patterns of Human Malformation. 8th ed. Philadelphia: Elsevier; 2022. p. 794–795 (Fetal valproate syndrome).
148. Chelsea's Hope Lafora Children Research Fund [Internet]. Available from: https://chelseashope.org