Skip to main content
NCBI home page
NIHPA Author Manuscripts logoLink to NIHPA Author Manuscripts
. Author manuscript; available in PMC: 2026 Nov 2.
Published in final edited form as: Genet Med. 2023 Nov 10;26(2):101027. doi: 10.1016/j.gim.2023.101027

Neurological manifestations in PMM2-congenital disorders of glycosylation (PMM2-CDG): Insights into clinico-radiological characteristics, recommendations for follow-up, and future directions

Karthik Muthusamy 1,*, Judit M Perez-Ortiz 2,3, Anna N Ligezka 4, Ruqaiah Altassan 5,6, Christin Johnsen 4,7, Matthew J Schultz 8, Marc C Patterson 2,3,4, Eva Morava 4,8,9
PMCID: PMC13723945  NIHMSID: NIHMS2214708  PMID: 37955240

Abstract

Purpose:

In the absence of prospective data on neurological symptoms, disease outcome, or guidelines for system specific management in phosphomannomutase 2-congenital disorders of glycosylation (PMM2-CDG), we aimed to collect and review natural history data.

Methods:

Fifty-one molecularly confirmed individuals with PMM2-CDG enrolled in the Frontiers of Congenital Disorders of Glycosylation natural history study were reviewed. In addition, we prospectively reviewed a smaller cohort of these individuals with PMM2-CDG on off-label acetazolamide treatment.

Results:

Mean age at diagnosis was 28.04 months. Developmental delay is a constant phenotype. Neurological manifestation included ataxia (90.2%), myopathy (82.4%), seizures (56.9%), neuropathy (52.9%), microcephaly (19.1%), extrapyramidal symptoms (27.5%), stroke-like episodes (SLE) (15.7%), and spasticity (13.7%). Progressive cerebellar atrophy is the characteristic neuroimaging finding. Additionally, supratentorial white matter changes were noted in adult age. No correlation was observed between the seizure severity and SLE risk, although all patients with SLE have had seizures in the past. “Off-label” acetazolamide therapy in a smaller sub-cohort resulted in improvement in speech fluency but did not show statistically significant improvement in objective ataxia scores.

Conclusion:

Clinical and radiological findings suggest both neurodevelopmental and neurodegenerative pathophysiology. Seizures may manifest at any age and are responsive to levetiracetam monotherapy in most cases. Febrile seizure is the most common trigger for SLEs. Acetazolamide is well tolerated.

Keywords: Acetazolamide, Congenital disorders of glycosylation, Phosphomannomutase 2 enzyme, PMM2-CDG, Stroke-like episode (SLE)

Introduction

Congenital disorders of glycosylation (CDG) are a rare group of inherited metabolic disorders that affect the complex process of synthesis and remodeling of glycoproteins and glycolipids. Attachment of the glycan chain is important for most proteins to reach their destination in the cellular compartment and function appropriately. There are approximately 160 known CDG to date.1,2 The most common type of glycosylation defect involves the biochemical pathway of N-linked glycosylation.3 CDG are further classified based on either the enzymatic defect impacting the glycan assembly and transfer in the endoplasmic reticulum (CDG-I) or their maturation in the Golgi apparatus (CDG-II).4 Given the ubiquity of glycosylation in human physiology,5 these disorders usually affect multiple organ systems with variable spectrum of clinical manifestations and severity.6

Phosphomannomutase 2 (PMM2) enzyme is involved in the early stage of the N-linked glycosylation pathway, and it is essential for conversion of mannose-6-phosphate to mannose-1-phosphate.7 Biallelic pathogenic variants in PMM2 gene (HGNC ID: 9115) lead to PMM2-CDG (MIM: #212065), which is the most common N-linked glycosylation disorder, with more than 1000 patients reported so far.8 Affected individuals can manifest with either an isolated neurological phenotype or with a combination of neurological and visceral phenotypes. Neurological involvement is the core clinical feature in PMM2-CDG with both central and peripheral nervous system being involved and is often the initial clinical presentation. Systemic involvement includes endocrine, hematological, immunological, cardiac, gastrointestinal, skeletal, and renal systems, in addition to the vision and hearing changes, which all could cause secondary neurological deficits thus adding up to the neurological phenotype. Neurological manifestations range from developmental delay, hypotonia, epilepsy, cerebellar ataxia, myopathy, and demyelinating neuropathy to movement disorders and stroke-like episodes (SLEs).8,9 There is no curative treatment for the neurological manifestation of PMM2-CDG besides supportive therapy including seizure management, occupational, physical, and speech therapy. Moreover, the pathophysiology of seizures and SLEs are poorly understood at this time, and there are no standardized treatment protocols that would guide the physicians when encountering these well-recognized neurologic emergencies. An open-label pilot study with oral acetazolamide in individuals with PMM2-CDG has shown positive effect on the motor symptoms, including the cerebellar ataxia and the speech.10

Understanding the neurological phenotype is essential because many additional treatments are on the horizon, and its ability to modify the neurological features is yet to be understood. Despite several reports in the literature on different neurologic symptoms in PMM2-CDG, no systematic prospective studies are available focusing on neurological presentations, imaging, and follow-up. As part of the Frontiers of Congenital Disorders of Glycosylation Consortium (FCDGC) natural history study, we reviewed the data of a large cohort of individuals with PMM2-CDG for their neurological involvement. Additionally, we prospectively evaluated a smaller portion of this cohort for their response to off-label acetazolamide use. The study aimed to systematically explore the neurologic phenotype, seeking to identify common and uncommon manifestations, predict severity, and derive recommendations for clinical practice and future directions from a neurology perspective.

Materials and Methods

Subjects

Patients with a biochemically and molecularly confirmed diagnosis of PMM2-CDG were prospectively enrolled to the Frontiers of Congenital Disorders of Glycosylation Consortium (FCDGC) natural history study (IRB: 19-005187). Data were systematically collected according to IRB protocol. Retrospective data collection was also part of the enrollment process because the patients have been previously followed up by standard of care. All patients provided signed informed consent.

Data extraction

Demographic information, head circumference (at birth when available and at the time of evaluation), developmental milestones, neurological history, and examination preferably by a single investigator, carbohydrate-deficient transferrin (CDT), PMM2 gene variants, electrophysiology, and neuroimaging data were collected. For epilepsy and SLE, age at onset, precipitating events, clinical details, treatment details, response to treatment, electroencephalogram (EEG), and neuroimaging findings during the episodes were collected and analyzed. Microcephaly was defined per Center for Disease Control definition of more than 2 standard deviations (SDs) below the average or less than the 3rd percentile for age. Diagnosis of muscle involvement/myopathy was based on standard neurologic physical examination by a neurologist. Seizure control was assessed per International League against Epilepsy definition for drug-resistant epilepsy.11 Electrophysiological studies included EEG and electromyography. Neuroimaging included brain computed tomography (CT) and magnetic resonance imaging (MRI). If the patients had more than 1 neuroimaging evaluation, each were analyzed separately for temporal changes. Magnetic resonance spectroscopy (MRS) and spine imaging were analyzed whenever available.

Nijmegen Progression CDG Rating Scale

The Nijmegen Progression CDG Rating Scale (NPCRS) is a clinical tool developed as a means of quantifying clinical severity of individuals with CDG in a global and comprehensive way and is validated across all age groups.12 NPCRS scores are used to scale patients into a mild (0-14), moderate (15-25), and severe (>26) category.13 The assessment was performed in every patient in the study by the same investigator (E.M.).

Genotype-phenotype severity analysis

Correlations between PMM2 genotype and phenotype severity in the cohort was based on previously described pathogenic variant effects predicted on PMM2 enzyme function.14

Laboratory values

CDT glycoform analysis was performed in serum by standard clinical testing by mass spectrometry in all patients and expressed as isoform ratios.15

Off-label acetazolamide therapy

Based on the previously seen tolerability and benefit of acetazolamide treatment in a pilot trial in PMM2-CDG,10 6 patients were started on off-label acetazolamide. Patients were started at a standard dose of 5 to 10 mg/kg/day divided twice daily. Acetazolamide dose was uptitrated to a maximum of 18 mg/kg/day as tolerated. Safety clinical labs were monitored, including comprehensive metabolic panel (Na, K, Cl, pCO2, Base excess, aspartate aminotransferase, alanine aminotransferase, gamma-glutamyl transferase, indirect and direct bilirubin, total protein, albumin, and alkaline phosphatase), kidney function tests (blood urea nitrogen, creatinine, urinalysis, urine calcium/creatinine ratio, and urine protein/creatinine ratio), in addition to clinical exam, growth (height, weight, and head circumference), and vital signs (blood pressure, respiratory rate, and heart rate). NPCRS and the modified version of the International Cooperative Ataxia Rating Scale (ICARS) (miniICARS) were scored every 6 months. Based on no significant change in several sections of ICARS score on acetazolamide treatment in the previous pilot study in PMM2-CDG, mini-ICARS scores were established. The mini ICARS is an abbreviated ataxia scoring system, modified from the ICARS10 but excluding oculomotor disorders, with addition of the Archimedes spiral. Relative to ICARS, the mini ICARS tests 4 of 7 posture and gait disturbances (walking, standing, and body sway eyes open or eyes closed) and 4 of 10 kinetic functions (knee-tibia test, finger-finger test, pronation-supination alternating movements, and speech fluency and clarity).

Statistical analysis

All statistical analyses were performed using GraphPad Prism version 8.3. For all statistical tests, P value < .05 was considered significant. Genotype-phenotypic severity correlations were calculated as odds ratio for each neurological deficit using Fisher’s exact test. Correlations between genotype versus CDG severity were graded according to NPCRS, and 2-way analysis of variance was used for analysis. Pre- versus post-treatment scores in off-label acetazolamide data was analyzed using paired two-tailed t test.

Results

Fifty-one (34 males and 17 females) individuals with PMM2-CDG were included in the study. Duration of prospective follow-up of these patients in natural history study ranged from 4 months to 4.8 years, and retrospective data were available for all patients from their birth. Mean age at establishment of the diagnosis was 28.04 months. Delay in diagnosis was significant for most individuals and ranged from 1 month to 32 years. Evaluations in the past done in view of overlapping clinical features before suspecting PMM2-CDG included testing for spinal muscular atrophy (n = 3), congenital myopathy (n = 2), carnitine uptake defect (n = 1), and Prader-Willi syndrome and chromosomal anomalies (n = 4). Disproportionate delay in the motor domain relative to other developmental domains, such as socio-cognitive and language, associated hypotonia during early infancy, and dysmorphism were noted as reasons for suspecting an alternate diagnosis. Mean age at assessment in our clinic was 9.45 years (0.75 to 34 years). All patients (n = 51) had NPCRS scores obtained at the time of assessment. Severity score based on NPCRS was mild in 7.8% of patients, moderate in 62.7%, and severe in 29.4% (Figure 1A). Microcephaly at birth was seen in 1 of 27 (3.7%), and microcephaly at assessment was noted in 9 of 47 (19.1%). The patient with microcephaly at birth at 2nd percentile was noted to have normal head circumference percentile at the age of 16 years (34.3 percentile). All 9 patients with microcephaly at our initial evaluation were normocephalic at birth. All patients had global developmental delay, and only 4 of them were able to walk unassisted. Developmental status at the time of our assessment were reported as static in 2 patients (at 25 and 28 years of age), 1 had regression of attained milestones after head injury at age 14, and the rest continued to have slow developmental gains. Strabismus was noted in all but 1 patient, 16 patients had retinopathy, 1 had optic nerve hypoplasia, and 1 had macular hypoplasia. Among patients with retinopathy, progression of retinal findings was documented in 3 patients at ages 9, 12, and 26, respectively. Nine patients (17.6%) had sensorineural hearing loss without any alternative explanation and required hearing aids. The most common neurological manifestations (Figure 1B) were ataxia (90.2%), followed by myopathy (82.4%), seizures (56.9%), neuropathy (52.9%), extrapyramidal symptoms, such as dystonia, chorea, athetosis, and tics (27.5%), SLEs (15.7%), and spasticity (13.7%).

Figure 1. PMM2-CDG disease severity, neurological involvement, and genotype-phenotype correlation.

Figure 1

Open in a new tab

A. Categorization of severity based on NPCRS. B. Spectrum of neurological manifestations. C. Patients with 1 copy of non-functioning enzyme variants (R141H, T237R, T237M, and D188G) exhibit lower odds of the spasticity phenotype relative to those with 2 copies of PMM2 pathogenic variants with functioning enzyme (OR 0.09, CI 0.02-0.42, P = .0024). This effect is not seen in other neurological deficits, or (D) in CDG severity scores (NPCRS). E. Spectrum of this cohort’s pathogenic variants in PMM2 (NM_000303.3) listed in proteinpaint (https://proteinpaint.stjude.org/), digitally reproduced for color manipulation. Numbers within the circle denote the number of occurrence (when more than 1) of the specific variant. Missense variants are coded in blue, splice site variants in light purple, and intronic variants in pink. EPS, extrapyramidal symptoms; SLE, stroke-like episodes; SNHL, sensorineural hearing loss.

More than half (56.9%) of patients in the cohort experienced seizures. Mean age at onset of first seizure was 35 months and ranged from 3 months to 24 years. Among patients with seizures, 24.1% had isolated afebrile seizures, 27.5% had isolated febrile seizures, and 48.3% had both febrile and afebrile seizures (Figure 2A). Status epilepticus requiring rescue medications and hospitalization was seen in 38% of patient with seizures. The most common semiology of seizures was generalized seizures in 72.4%, both generalized and focal seizures in 24.1%, and isolated focal seizures in 3.4%. Isolated generalized tonic clonic seizures were seen in 66.6% patients with generalized seizures, atonic seizures in 1 patient, and the rest had multiple seizure types (generalized tonic clonic seizures, atypical absence, tonic, and myoclonic seizures). Among patients with seizures, 13 were on monotherapy with levetiracetam and 1 with oxcarbazepine. The rest of the patients were on polypharmacy, with varying combinations of levetiracetam, carbamazepine, lacosamide, phenytoin, lamotrigine, and clonazepam. Of the patients with seizures, 79.3% had good seizure control, 13.8% had initial drug-resistant epilepsy that gradually became controlled over the course of follow-up, and 6.9% of them continued to have drug refractory epilepsy at the time of follow-up.

Figure 2. Seizure burden and stroke-like episodes.

Figure 2

Open in a new tab

A. Types of seizures. B. Twenty-nine of 51 patients carried seizure diagnosis, with a mean age of 34 months at first seizure. Eight of these patients later developed stroke-like episodes (SLE). There was no difference between age of onset of seizures in patients that went on to develop SLE versus those that did not (21 vs 27 months, t test, P = .4301). All patients with SLE invariably had a seizure diagnosis before SLE event. An average of 38 months elapsed between first seizure and first stroke-like episode (38 ± 25.04). Only 1 patient, with severe infantile form of PMM2-CDG, had their first SLE within days of first seizure. In all the other cases, the mean latency from first seizure to first SLE were longer (43 months). C. Transferrin glycosylation in seizure patients did not correlate with stroke-like episodes (asialo/di-oligo OR 1.863, CI 0.3796-7.7630, P = .6981; mono/di-oligo OR 0.600, CI 0.1032-3.5240, P = .6232). Features of seizure severity did not correlate with stroke-like episodes, including status epilepticus (OR 2.000, CI 0.4563-8.8845, P = .4327) and poor seizure control (OR 3.600, CI 0.6427-18.5, P = .3045).

A total of 62 EEGs were available from 32 patients, of which 47 were normal EEG. Among the abnormal EEGs, 9 had hemispheric/focal slowing, 4 with diffuse slowing, 1 with focal epileptiform activity, and 1 with generalized epileptiform activity. Twelve nerve conduction studies were available from 10 patients; 3 of these patients, who had 1 study each, had normal nerve conductions, the rest of the 7 patients had abnormal studies showing demyelinating sensory motor polyneuropathy.

Eight patients (15.7%) in the cohort had a total of 13 SLEs, with 2 of them having more than 1 SLE. Details on SLEs, including the relation to seizures, precipitating factors, and details on evaluation, including imaging and EEG during the episodes, were analyzed and tabulated (Table 1, Figure 3). Deficits lasted for 1 day to 1 year (less than 1 week in 11, 6 weeks in 1, and 1 year in 1). Nine neuroimaging were performed during SLE (both CT and MRI), of which 1 MRI was abnormal with scattered hemispheric swelling and associated restriction diffusion. Four EEGs were done during SLE, of which 2 were abnormal with contralateral hemispheric slowing (Figure 3). All patients with SLE invariably carried seizure diagnosis before the SLE event. There was no difference between age of onset of seizures in patients who went on to develop SLE versus those that did not (34 vs 38 months, with means of 21 vs 27 months when outlier excluded, t test, P = .4301). An average of 38 months elapsed between first seizure and first SLE (38 ± 25.04) (Figure 2B). Because all patients with SLE had history of seizures, anticonvulsants were continued during management of SLE, along with other symptomatic and supportive treatments, such as hydration and antipyretics. Additionally, rescue treatment for seizures were administered when they were noted to have active seizures during the hospital stay or persistent EEG abnormality. Although patients with poor seizure control tend to have SLE, the results were not statistically significant, and in patients with seizures, we found no correlations between transferrin glycoform ratios and likelihood of SLE. (Figure 2C).

Table 1.

Characteristics of stroke-like episodes (SLE) in the cohort

Patients with SLE Patient 1 Patient 2 Patient 3 Patient 4 Patient 5 Patient 6 Patient 7 Patient 8
No of SLE 1 2 1 1 1 5 1 1
History of seizures + + + + + + + +
Age at first seizure (months) 24 16 18 7 24 49 6 58
Age at first SLE (months) 60 54 44 81 50 123 6 88
Time from first seizure to stroke (months) 36 38 26 74 26 74 0 30
Precipitating factors Seizures Infection, seizures Seizures Fall Fever, mastoiditis Fever, seizures Seizures Seizures
Laterality of involvement Left Both episodes were right Left Right Right Right - 2, left 3 Right Left
Seizure control Initially refractory, now under control Yes Refractory seizures Yes Yes Initially refractory, now under control Yes Yes
EEG during SLE Normal EEG Left hemispheric dysfunction (Figure 5C) Right temporal slowing - - Normal - -
Imaging during SLE Normal CT Left hemispheric swelling with restricted diffusion (Figure 5) No new findings in CT and MRI Normal CT Normal MRI Normal MRI - MRI normal
Time to recovery 6 weeks 1 year, 1 week 1 day 1 week 2 days 1 day to 1 week 1 day 2 days
Open in a new tab

CT, computed tomography; EEG, electroencephalogram; MRI, magnetic resonance imaging; SLE, stroke-like episodes.

Figure 3. Neuroimaging findings in PMM2-CDG.

Figure 3

Open in a new tab

MRI Brain of a child during a stroke-like episode with right hemiparesis with T2 FLAIR axial (A and B) and DWI images (D and E) showing cortical swelling along with multiple foci of restricted diffusion in the left cerebral hemisphere. Corresponding EEG (C) revealing diffuse left hemispheric slowing. MRI Brain T2W axial image (F) shows midbrain atrophy with relatively preserved superior cerebellar peduncle giving a molar tooth-like appearance. Single-voxel MR spectroscopy (G) in cerebellum revealing reduced NAA:creatine and choline:creatine ratios. NAA, N-acetylaspartate.

Ninety-six brain MRIs were available from 43 patients. Mean age at MRI Brain was 4.1 years of age (1 month to 29 years). Eighty-nine MRIs revealed cerebellar and brain stem atrophy; 7 were normal (obtained between 2 months to 4.75 years of age). Three patients’ early MRI revealed delayed myelination at 4, 7, and 8 months of age, respectively, and all had progression of myelination in subsequent scans. Serial MRIs were available for 28 patients, and the interscan interval ranged from 1 month to 23.5 years. Findings in serial MRI done before 20 years of age revealed improvement in supratentorial myelination in those with initially delayed myelination and a progressive atrophy of cerebellum and brain stem (Figure 4). Findings in serial MRI done after 20 years of age revealed mild diffuse cerebral atrophy and subcortical white matter changes suggestive of axonal loss in addition to cerebellar and brain stem atrophy (Figure 5). In some images, cerebellum had markedly atrophied with no further noticeable progression in serial images, presumably related to extensive loss of Purkinje cells and a complete Bergman gliosis. A total of 6 MRS were available, and 4 of them revealed low N-acetylaspartate (NAA) levels (Figure 3). Seven magnetic resonance imaging spines were available and were normal, except in 1 that showed short pedicles and platyspondyly. Thirty-two CTs scans were available from 15 patients, and all of them revealed cerebellar and brain stem atrophy.

Figure 4. Serial MRI Brain findings in the same patient in early childhood.

Figure 4

Open in a new tab

Upper panel (A-C) done at age 3 months, middle panel (D-F) and lower panel (G-I) of the same patient done at 4 years and 7 years of age, respectively, shows progressive volume loss with hyperintensity involving cerebellar vermis more than the cerebellar hemispheres in T2W axial images (A, D, and G) and T1 sagittal images (D, F, and I). Also note the initial delay in myelination of optic radiations in supratentorial white matter, and deep cerebellar white matter (arrow marks in [B] and [A]) that improved later (arrow marks in [E] and [D]).

Figure 5. MRI Brain findings in late adolescence/adult age group.

Figure 5

Open in a new tab

MRI Brain of the same patient at 18 (A-D) and 25 (E-H) years of age showing interval increase in the extent of T2W hyperintensity of supratentorial periventricular white matter with axonal loss/gliosis (arrow marks in E and F). T2 FLAIR coronal images (C and G) show severe vermian atrophy and hyperintense atrophic cerebellar hemispheres, along with supratentorial white matter changes. T1 sagittal images (D and H) reveals small pontine protuberance in (D) with interval progression with further flattening in (H).

CDT

CDT assay was available for 44 patients, of which 36 had abnormal mono-oligo/di-oligo ratios and 34 had abnormal asialo/di-oligo ratios.

We found no correlations between transferrin glycoform ratios and any section of NPCRS, including section III containing neurologic symptoms, total NPCRS or SLE (Figure 2C).

Genetic studies

Genetic confirmation was available for all patients. All the variants listed (Supplemental Table 1) are either published as confirmed pathogenic or likely pathogenic, or when it was a variant of uncertain significance, we measured the enzyme activity to confirm its pathogenicity. All but 1 had compound heterozygous pathogenic/likely pathogenic variants in PMM2 gene, whereas 1 had homozygous pathogenic variant related to uniparental disomy. There was no correlation between genotype and phenotypic severity. Of 10 neurological features evaluated, only spasticity correlated negatively with null PMM2 variants. (Figure 1C and D). Distribution of these gene variants are plotted along the PMM2 protein using proteinpaint software (https://pecan.stjude.cloud/proteinpaint) (Figure 1E).

Treatment

Six patients in our cohort (not participating in NCT04679389) pursued off-label acetazolamide treatment. The starting dose ranged from 5 to 10 mg/kg/day and individually increased as tolerated with the highest tolerated dose being 18mg/kg/day (Figure 6A). Safety monitoring labs were reassuring. Stable NPCRS scores were observed in all 6 participants (Figure 6B). Parents endorsed subjective improvement in tasks requiring coordination and language fluency, but objective ataxia scores did not change significantly based on the available data points (Figure 6C and D).

Figure 6. Off-label acetazolamide in 6 patients with PMM2-CDG in this cohort.

Figure 6

Open in a new tab

A. Dose by weight over time. B. Patients taking off-label acetazolamide had stable NPCRS score over the course of 2 to 2.5 years. C. Four of these had available ataxia severity scores (ICARS) at baseline and with treatment. D. There is a negative mean difference in ICARS on treatment vs after treatment (−5.500), but this difference is not statistically significant (P = .0758, paired two-tailed t test). ICARS, International Cooperative Ataxia Rating Scale; NPCRS, The Nijmegen Progression CDG Rating Scale.

Discussion

The neurological phenotypic spectrum in PMM2-CDG has not been prospectively studied. In the absence of data on neurological outcome, and no guidelines for disease specific neurological symptom management in individuals with PMM2-CDG, we collected and reviewed natural history data and analyzed the neurological and the neurodevelopmental aspects of the disease.

Developmental delay is a prominent presentation in individuals with PMM2-CDG, involving mainly the motor function and cognition.8,16 In our cohort, all patients had global developmental delay. Cerebellar ataxia has been reported in 95% to 100%.8,17,18 In our cohort, 90% of patients had cerebellar ataxia. Myopathy is an underreported disease manifestation12,13 because 82% of our patient had clinical myopathy with variable severity. More than half of our patients exhibited peripheral neuropathy, which is comparable to previous reports, and nerve conduction studies were consistent with a demyelinating sensory motor neuropathy.8,18 Movement disorders have been rarely described; however, we found that around one fourth of our cohort had mixed movement disorder suggestive of extrapyramidal involvement.8,18 Movement disorders did not require any symptomatic treatment because they were mild to moderate and were not considered to be the major cause of their motor disability. Sensorineural hearing loss was seen in 17.6% of patients, which is comparable to the literature range of 8% to 33%.8,19–21 Stagnation of head circumference with resultant microcephaly was noted in our cohort. No significant statistical difference was observed in genotype or CDT values in individuals who walked unassisted versus others.

Various seizure semiologies that have been documented in individuals with PMM2-CDG are generalized tonic clonic, partial, myoclonic, and absence seizures.8 Overall reported seizure rate was 23%18 in the literature. Interestingly, in comparison to previous literature, more than half of our patients had seizures. Generalized seizures were the most common seizure semiology; however, significant proportion of our patients also had focal seizures as well. Although most patients responded well to monotherapy with levetiracetam, 7% had drug-resistant epilepsy. Based on our cohort, it is also possible that the first seizure can occur even during adulthood. Our findings on seizures emphasize that seizures are common and possibly one of the defining phenotypic features in PMM2-CDG. Interrogation about seizures should be an integral component during follow-up because early identification and aggressive seizure control could result in improving the quality of life. Monotherapy with levetiracetam appears to be the most common and safely used medication in the presence of liver involvement as a co-morbidity in PMM2-CDG.

SLEs are one of the acute neurologic emergencies encountered in PMM2-CDG. Recurrent SLE could lead to cumulative disability over time. Sudden onset neurological deficits can occur because of varying etiologies that include thrombotic or hemorrhagic strokes, subclinical seizures, or post-ictal weakness. SLE is a usually a diagnosis of exclusion once these are ruled out by appropriate investigations. Pathophysiology of SLE has not been fully elucidated, and currently there are no standardized guidelines for management. Several postulates have been put forward that includes hypoglycosylation-driven channelopathy, transient mitochondrial dysfunction, active epileptic inhibitory phenomena, and altered function of hypoglycosylated mechanoreceptors in the setting of trauma.22–25 Incidence of SLE in our cohort was 15.6% in comparison to a higher incidence reaching 52% across various studies.21,26 Age at SLE ranged from 6 months to 10.25 years in our cohort and has been reported as late as 24 years of age in a previous study.27 SLE can occur at any age and appears independent of clinical/biochemical severity, seizure control, and genotype as evidenced in our study. Precipitating factors for SLE in this cohort were febrile illness, seizures, and trauma as described in the previous literature.22,23,27 An interesting observation in our cohort is that all individuals with SLE had prior history of seizures, and the lag time from history of first seizure to SLE ranged from 0 to 74 months. In addition to contralateral hemispheric slowing in EEG in 1 of our patients during SLE, most of them exhibited clinical seizures around the time of SLE. This emphasizes the possibility of an epileptic phenomena as the underlying cause for SLE. In fact, a previous study had depicted similar observation and response to antiepileptic medications.22,27 Although patients with drug resistant epilepsy tend to have SLE in our cohort, the data were not statistically significant. We hypothesize that SLE are due to an epileptic phenomenon and might represent focal nonconvulsive status epilepticus with prolonged post-ictal Todd’s paralysis due to overshoot of neuronal hyperpolarization after seizures, or an active epileptic inhibitory phenomenon.

Brain imaging is often required during SLE to rule out ischemic and hemorrhagic strokes because these patients are predisposed for both thrombosis and coagulopathies.28 Strokes due to thrombosis and SLE have been described to coexist in the same patient.26 Neuroimaging findings are usually normal during SLE, although there are rare reports of cytotoxic edema with restricted diffusion in the involved hemisphere.8 One of our patients had similar findings of signal changes and restricted diffusion and had prolonged time to recover. This indicates that abnormal neuroimaging findings during a SLE implies possibly prolonged seizures and energy failure resulting in cytotoxic edema and thus lead to poor recovery. Lactate peak in MRS during a SLE has been documented and might point to transient mitochondrial dysfunction that is presumably secondary.25

Classic neuroimaging findings in PMM2-CDG are the cerebellar signal changes along with volume loss predominantly involving the vermis relative to the cerebellar hemispheres and pontine atrophy. This has been described as shrunken, bright cerebellum in the literature.29 Neuroimaging changes involving cerebellum have been documented even in early infancy, emphasizing the onset of disease before birth.30 Milder cases may initially have a normal MRI and the atrophy eventually develops with disease progression.30 Delay in myelination was observed in images performed during early infancy in our cohort that eventually improved with age. This finding has not been classically described in the literature in PMM2-CDG as in other CDGs.31,32 Supratentorial white matter appears typically normal during childhood in PMM2-CDG. However, imaging during adulthood in our cohort demonstrated subcortical and periventricular white matter changes in cerebrum suggesting progressive axonal loss, which has rarely been described in literature.33 Our study is among the few to describe the evolution of cerebellum and brain stem atrophy in serial imaging and findings during adulthood. We also describe a molar tooth-like appearance in axial images that appears because of midbrain atrophy with relatively preserved superior cerebellar peduncle. There are several cerebellar volumetric studies that have been observed to correlate with the clinical progression of cerebellar ataxia.17,34,35 MRS demonstrates reduced NAA consistent with axonal loss and can reveal elevated lactate during SLE.25,30 Brain autopsy in a child with PMM2-CDG had revealed reduced thickness of the molecular layer of cerebellum and associated cerebellar vermis and hemispheric atrophy, and hypoplasia of cerebellum.36 Because the initial clinical features are often nonspecific, awareness of the imaging findings could shorten diagnostic odyssey by raising suspicion for PMM2-CDG earlier on in the disease course.

Molecular mechanisms underlying neurologic impairment have not been fully elucidated. Hypoglycosylation of several receptor proteins intrinsic to neurodevelopment, such as matrix metalloproteinase 2 or N-cadherins, is considered central to neuropathology. Although some observations have linked variants affecting the homodimer folding and/or stability domains with milder disease severity,8,14 a consistent genotype-phenotype correlation has not been established. We were unable to replicate these observations in our cohort. Moreover, we did not observe any correlation between blood CDT results and the neurological severity, including seizures and SLE. Twenty percent (8/44) of patients had normal CDT results. CDT ratios have been shown to improve spontaneously over time in natural history studies along with various other biochemical parameters, making it less reliable for inclusion as disease’s monitoring biomarker in treatment trials.37,38

Currently, there is no curative treatment available for individuals with PMM2-CDG. A clinical trial with acetazolamide had shown effectiveness in motor cerebellar syndrome and possible improvement in epilepsy and SLE.10 Few patients in our cohort were on “off-label” use of acetazolamide. Given the smaller number, we did not observe any statistically significant improvement in neurological manifestation in the short monitoring period. A double-blind, randomized acetazolamide clinical trial in individuals with PMM2-CDG is underway (https://clinicaltrials.gov/ct2/show/NCT04925960) to assess efficacy and follow-up over a longer period. Furthermore, preclinical studies on epalrestat, an aldose reductase inhibitor, showed increase in the PMM2 enzyme activity in vitro, and an off-label use of epalrestat in 1 patient showed improvement in the ICARS and the transferrin glycosylation.39,40 Several other measures that had shown efficacy in preclinical studies are morpholino oligonucleotides and pharmacologic chaperones.41,42

Developmental disability and other neurological morbidities in PMM2-CDG are due to multi-neuraxial involvement that includes cerebellum, brain stem connections, corticospinal tract, extrapyramidal system, cerebral cortex, peripheral nerve, muscle, vision, and hearing systems. Preferential involvement of cerebellar vermis and brain stem, as well as focal neurological manifestations, such as focal seizures and SLE in PMM2-CDG, are intriguing, given the generalized hypoglycosylation and multisystemic involvement. Contrary to the stabilization or even improvement in clinical and biochemical manifestations in most organ systems,37 our observation is that the neurological manifestations and the related disability tend to slowly progress even in adulthood with new onset seizures and may associate with progression of imaging changes, development of white matter changes suggestive of axonal loss, and MRS revealing reduced NAA over time. We propose an algorithm for diagnosis and management for individuals with PMM2-CDG from the neurologic perspective (Figure 7).

Figure 7. Management algorithm for neurological manifestations in PMM2-CDG.

Figure 7

Open in a new tab

BARS, Brief Ataxia Rating Scale; EEG, electroencephalogram; MRI, magnetic resonance imaging; ICARS, International Cooperative Ataxia Rating Scale; NPCRS, The Nijmegen Progression CDG Rating Scale.

We propose that neurological involvement in PMM2-CDG is both developmental and degenerative in varying proportions. Hence, the ideal time for intervention would be early infancy or even in utero. Seizures are common, and levetiracetam monotherapy is efficient in most patients. Epileptic phenomena often precipitated by intercurrent infections is the cause for SLEs, and an aggressive control of seizures and hyperthermia could help prevent a SLE based on our clinical observation, although experimental data are lacking.

Future directions would include exploring the pathophysiology of the neurological involvement with ex vivo cell models and brain organoids for altered neurogenesis and synaptic architecture that underlies the developmental defects, as well as the factors that underlie progressive neurodegeneration in specific regions of the brain. Incorporation of various advanced structural neuroimaging techniques like volumetric studies, diffusion tensor imaging, and functional imaging such as positron emission tomography, multimodal electrophysiologic studies such as evoked potentials, central motor conduction time, glycoproteomic studies, and exploration of metabolic parameters such as glucose levels and transcriptomics/proteomics in serum and cerebrospinal fluid over time and during acute events like seizures and SLEs, would help us to explore the less well-understood pathophysiology neurologic manifestation in PMM2-CDG. This would eventually help us discover biomarkers with therapeutic implications and promote translational research in this rare group of disorder.

Supplementary Material

Spreadsheet suppl. file

The online version of this article (https://doi.org/10.1016/j.gim.2023.101027) contains supplemental material, which is available to authorized users.

Acknowledgments

The authors sincerely thank all the patients and their families for participating in the study.

Funding

This work was funded by the grant titled Frontiers in Congenital Disorders of Glycosylation (1U54NS115198-01) from the National Institute of Neurological Diseases and Stroke (NINDS) and the National Center for Advancing Translational Sciences (NCATS), and the Rare Disorders Clinical Research Network (RDCRN), at the National Institute of Health.

Conflict of Interest

Mayo Clinic and Eva Morava have a financial interest related to this research. This research has been reviewed by the Mayo Clinic Conflict of Interest Review Board and is being conducted in compliance with Mayo Clinic Conflict of Interest policies. This work was funded by the grant titled Frontiers in Congenital Disorders of Glycosylation (1U54NS115198-01) from the National Institute of Neurological Diseases and Stroke (NINDS) and the National Center for Advancing Translational Sciences (NCATS), and the Rare Disorders Clinical Research Network (RDCRN), at the National Institute of Health. All other authors declare no conflicts of interest.

Abbreviations

CDG

congenital disorders of glycosylation

CDT

carbohydrate-deficient transferrin

CT

computed tomography

EEG

electroencephalogram

FCDGC

Frontiers of Congenital Disorders of Glycosylation Consortium

ICARS

International Cooperative Ataxia Rating Scale

MRI

magnetic resonance imaging

MRS

magnetic resonance spectroscopy

NCPRS

Nijmegen CDG Patient Rating Scale

PMM2

phosphomannomutase 2

SLE

stroke-like episodes

Footnotes

Ethics Declaration

This study was approved by the Institutional Review Board of Mayo Clinic under protocols #18-007276 and #19-005187. All procedures followed were in accordance with the ethical standards of the responsible committee on human experimentation (institutional and national) and with the Helsinki Declaration of 1975, as revised in 2000 (5). Informed consent was obtained from all patients for being included in the study.

Data Availability

Data are available from corresponding author upon request.

References

  • 1.Wilson MP, Matthijs G. The evolving genetic landscape of congenital disorders of glycosylation. Biochim Biophys Acta Gen Subj. 2021;1865(11):129976. 10.1016/j.bbagen.2021.129976 [DOI] [PubMed] [Google Scholar]
  • 2.Boyer SW, Johnsen C, Morava E. Nutrition interventions in congenital disorders of glycosylation. Trends Mol Med. 2022;28(6):463–481. 10.1016/j.molmed.2022.04.003 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 3.Péanne R, de Lonlay P, Foulquier F, et al. Congenital disorders of glycosylation (CDG): quo vadis? Eur J Med Genet. 2018;61(11):643–663. 10.1016/j.ejmg.2017.10.012 [DOI] [PubMed] [Google Scholar]
  • 4.Freeze HH, Eklund EA, Ng BG, Patterson MC. Neurology of inherited glycosylation disorders. Lancet Neurol. 2012;11(5):453–466. 10.1016/S1474-4422(12)70040-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 5.Fisher P, Spencer H, Thomas-Oates J, Wood AJ, Ungar D. Modeling glycan processing reveals Golgi-enzyme homeostasis upon trafficking defects and cellular differentiation. Cell Rep. 2019;27(4):1231–1243.e6. 10.1016/j.celrep.2019.03.107 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 6.Verheijen J, Tahata S, Kozicz T, Witters P, Morava E. Therapeutic approaches in Congenital Disorders of Glycosylation (CDG) involving N-linked glycosylation: an update. Genet Med. 2020;22(2):268–279. 10.1038/s41436-019-0647-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 7.Matthijs G, Schollen E, Heykants L, Grunewald S. Phosphomannomutase deficiency: the molecular basis of the classical Jaeken syndrome (CDGS type Ia). Mol Genet Metab. 1999;68(2):220–226. 10.1006/mgme.1999.2914 [DOI] [PubMed] [Google Scholar]
  • 8.Altassan R, Péanne R, Jaeken J, et al. International clinical guidelines for the management of phosphomannomutase 2-congenital disorders of glycosylation: diagnosis, treatment and follow up. J Inherit Metab Dis. 2019;42(1):5–28. 10.1002/jimd.12024 [DOI] [PubMed] [Google Scholar]
  • 9.Lam C, Krasnewich DM. PMM2-CDG. August 15, 2005; updated May 20, 2021. In: Adam MP, Mirzaa GM, Pagon RA, et al. , eds. GeneReviews® [internet]. 1993–2023. https://www.ncbi.nlm.nih.gov/books/NBK1110/
  • 10.Martínez-Monseny AF, Bolasell M, Callejón-Póo L, et al. AZATAX: acetazolamide safety and efficacy in cerebellar syndrome in PMM2 congenital disorder of glycosylation (PMM2-CDG). Ann Neurol. 2019;85(5):740–751. 10.1002/ana.25457 [DOI] [PubMed] [Google Scholar]
  • 11.Kwan P, Arzimanoglou A, Berg AT, et al. Definition of drug resistant epilepsy: consensus proposal by the ad hoc Task Force of the ILAE Commission on Therapeutic Strategies. Epilepsia. 2010;51(6):1069–1077. 10.1111/j.1528-1167.2009.02397.x [DOI] [PubMed] [Google Scholar]
  • 12.Ligezka AN, Mohamed A, Pascoal C, et al. Patient-reported outcomes and quality of life in PMM2-CDG. Mol Genet Metab. 2022;136(2):145–151. 10.1016/j.ymgme.2022.04.002 [DOI] [PubMed] [Google Scholar]
  • 13.Achouitar S, Mohamed M, Gardeitchik T, et al. Nijmegen paediatric CDG rating scale: a novel tool to assess disease progression. J Inherit Metab Dis. 2011;34(4):923–927. 10.1007/s10545-011-9325-5 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 14.Vaes L, Rymen D, Cassiman D, et al. Genotype-phenotype correlations in PMM2-CDG. Genes (Basel). 2021;12(11):1658. 10.3390/genes12111658 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 15.Lacey JM, Bergen HR, Magera MJ, Naylor S, O’Brien JF. Rapid determination of transferrin isoforms by immunoaffinity liquid chromatography and electrospray mass spectrometry. Clin Chem. 2001;47(3):513–518. 10.1093/clinchem/47.3.513 [DOI] [PubMed] [Google Scholar]
  • 16.Pearl PL, Krasnewich D. Neurologic course of congenital disorders of glycosylation. J Child Neurol. 2001;16(6):409–413. 10.1177/088307380101600604 [DOI] [PubMed] [Google Scholar]
  • 17.Serrano M, de Diego V, Muchart J, et al. Phosphomannomutase deficiency (PMM2-CDG): ataxia and cerebellar assessment. Orphanet J Rare Dis. 2015;10:138. 10.1186/s13023-015-0358-y [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 18.Pettinato F, Mostile G, Battini R, et al. Clinical and radiological correlates of activities of daily living in cerebellar atrophy caused by PMM2 mutations (PMM2-CDG). Cerebellum. 2021;20(4):596–605. 10.1007/s12311-021-01242-x [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 19.Barone R, Carrozzi M, Parini R, et al. A nationwide survey of PMM2-CDG in Italy: high frequency of a mild neurological variant associated with the L32R mutation. J Neurol. 2015;262(1):154–164. 10.1007/s00415-014-7549-7 [DOI] [PubMed] [Google Scholar]
  • 20.Kasapkara ÇS, Barış Z, Kılıç M, et al. PMM2-CDG and sensorineural hearing loss. J Inherit Metab Dis. 2017;40(5):629–630. 10.1007/s10545-017-0073-z [DOI] [PubMed] [Google Scholar]
  • 21.Kjaergaard S, Schwartz M, Skovby F. Congenital disorder of glycosylation type Ia (CDG-Ia): phenotypic spectrum of the R141H/F119L genotype. Arch Dis Child. 2001;85(3):236–239. 10.1136/adc.85.3.236 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 22.Dinopoulos A, Mohamed I, Jones B, Rao S, Franz D, deGrauw T. Radiologic and neurophysiologic aspects of stroke-like episodes in children with congenital disorder of glycosylation type Ia. Pediatrics. 2007;119(3):e768–e772. 10.1542/peds.2006-0763 [DOI] [PubMed] [Google Scholar]
  • 23.Izquierdo-Serra M, Martínez-Monseny AF, López L, et al. Stroke-like episodes and cerebellar syndrome in phosphomannomutase deficiency (PMM2-CDG): evidence for hypoglycosylation-driven channelopathy. Int J Mol Sci. 2018;19(2):619. 10.3390/ijms19020619 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 24.Li JV, Ng CA, Cheng D, et al. Modified N-linked glycosylation status predicts trafficking defective human Piezo1 channel mutations. Commun Biol. 2021;4(1):1038. 10.1038/s42003-021-02528-w [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 25.van Baalen A, Stephani U, Rohr A. Increased brain lactate during stroke-like episode in a patient with congenital disorder of glycosylation type Ia. Brain Dev. 2009;31(2):183. 10.1016/j.braindev.2008.08.014 [DOI] [PubMed] [Google Scholar]
  • 26.Schiff M, Roda C, Monin ML, et al. Clinical, laboratory and molecular findings and long-term follow-up data in 96 French patients with PMM2-CDG (phosphomannomutase 2-congenital disorder of glycosylation) and review of the literature. J Med Genet. 2017;54(12):843–851. 10.1136/jmedgenet-2017-104903 [DOI] [PubMed] [Google Scholar]
  • 27.Serrano M. Stroke-like episodes in PMM2-CDG: when the lack of other evidence is the only evidence. Front Pediatr. 2021;9:717864. 10.3389/fped.2021.717864 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 28.Stefanits H, Konstantopoulou V, Kuess M, Milenkovic I, Matula C. Initial diagnosis of the congenital disorder of glycosylation PMM2-CDG (CDG1a) in a 4-year-old girl after neurosurgical intervention for cerebral hemorrhage. J Neurosurg Pediatr. 2014;14(5):546–549. 10.3171/2014.7.PEDS14102 [DOI] [PubMed] [Google Scholar]
  • 29.Feraco P, Mirabelli-Badenier M, Severino M, et al. The shrunken, bright cerebellum: a characteristic MRI finding in congenital disorders of glycosylation type 1a. AJNR Am J Neuroradiol. 2012;33(11):2062–2067. 10.3174/ajnr.A3151 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 30.Chiapparini L, Moscatelli M. Neuroimaging of pediatric cerebellum in inherited neurodegenerative diseases. Appl Sci Basel. 2021;11(18):8522. 10.3390/app11188522 [DOI] [Google Scholar]
  • 31.Imtiaz F, Al-Mostafa A, Al-Hassnan ZN. Further delineation of the phenotype of congenital disorder of glycosylation DPAGT1-CDG (CDG-Ij) identified by homozygosity mapping. JIMD Rep. 2012;2:107–111. 10.1007/8904_2011_57 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 32.Schorling DC, Rost S, Lefeber DJ, et al. Early and lethal neurodegeneration with myasthenic and myopathic features: a new ALG14-CDG. Neurology. 2017;89(7):657–664. 10.1212/WNL.0000000000004234 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 33.Monin ML, Mignot C, De Lonlay P, et al. 29 French adult patients with PMM2-congenital disorder of glycosylation: outcome of the classical pediatric phenotype and depiction of a late-onset phenotype. Orphanet J Rare Dis. 2014;9:207. 10.1186/s13023-014-0207-4 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 34.Serrano NL, De Diego V, Cuadras D, et al. A quantitative assessment of the evolution of cerebellar syndrome in children with phosphomannomutase-deficiency (PMM2-CDG). Orphanet J Rare Dis. 2017;12(1):155. 10.1186/s13023-017-0707-0 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 35.de Diego V, Martínez-Monseny AF, Muchart J, et al. Longitudinal volumetric and 2D assessment of cerebellar atrophy in a large cohort of children with phosphomannomutase deficiency (PMM2-CDG). J Inherit Metab Dis. 2017;40(5):709–713. 10.1007/s10545-017-0028-4 [DOI] [PubMed] [Google Scholar]
  • 36.Wurm D, Hänsgen A, Kim YJ, Lindinger A, Baghai A, Gortner L. Early fatal course in siblings with CDG-Ia (caused by two novel mutations in the PMM2 gene): clinical, molecular and autopsy findings. Eur J Pediatr. 2007;166(4):377–378. 10.1007/s00431-006-0240-y [DOI] [PubMed] [Google Scholar]
  • 37.Witters P, Edmondson AC, Lam C, et al. Spontaneous improvement of carbohydrate-deficient transferrin in PMM2-CDG without mannose observed in CDG natural history study. Orphanet J Rare Dis. 2021;16(1):102. 10.1186/s13023-021-01751-2 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 38.Witters P, Honzik T, Bauchart E, et al. Long-term follow-up in PMM2-CDG: are we ready to start treatment trials? Genet Med. 2019;21(5):1181–1188. 10.1038/s41436-018-0301-4 [DOI] [PubMed] [Google Scholar]
  • 39.Iyer S, Sam FS, DiPrimio N, et al. Repurposing the aldose reductase inhibitor and diabetic neuropathy drug epalrestat for the congenital disorder of glycosylation PMM2-CDG. Dis Model Mech. 2019;12(11): dmm040584. 10.1242/dmm.040584 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 40.Ligezka AN, Radenkovic S, Saraswat M, et al. Sorbitol is a severity biomarker for PMM2-CDG with therapeutic implications. Ann Neurol. 2021;90(6):887–900. 10.1002/ana.26245 [DOI] [PMC free article] [PubMed] [Google Scholar]
  • 41.Gámez A, Serrano M, Gallego D, Vilas A, Pérez B. New and potential strategies for the treatment of PMM2-CDG. Biochim Biophys Acta Gen Subj. 2020;1864(11):129686. 10.1016/j.bbagen.2020.129686 [DOI] [PubMed] [Google Scholar]
  • 42.Park JH, Marquardt T. Treatment options in congenital disorders of glycosylation. Front Genet. 2021;12:735348. 10.3389/fgene.2021.735348 [DOI] [PMC free article] [PubMed] [Google Scholar]

Associated Data

This section collects any data citations, data availability statements, or supplementary materials included in this article.

Supplementary Materials

Spreadsheet suppl. file
NIHMS2214708-supplement-Spreadsheet_suppl__file.xlsx (11.6KB, xlsx)

Data Availability Statement

Data are available from corresponding author upon request.

RESOURCES