The study was approved by the Ethics Committee of the Hospital Josep Trueta (Girona, Spain) and Hospital Sant Joan de Déu (Barcelona, Spain), following the Helsinki II declaration. Written informed consent to participate in this study was provided by the participants’ legal guardians. Written informed consent was also obtained from all relatives included in the study.

The study enrolled 26 pediatric patients previously diagnosed with any type of LMNA muscular dystrophy (2014–2020) and carrying a definite pathogenic rare variant in LMNA. We retrospectively collected all available data from each patient’s first clinical contact up to their enrollment in our study. Clinical evaluation of index cases included a complete physical examination by a pediatric neurologist, neuromuscular specialist, and pediatric cardiologist. Non-pediatric relatives enrolled in our study also were clinically assessed. Saliva or peripheral blood samples were obtained from each patient as well as all available family members. All individuals were clinically assessed at Hospital Sant Joan de Déu (Barcelona, Catalonia, Spain). The complete pedigree of each family was obtained, including history of neuromuscular and cardiac diseases, syncope, and unexplained deaths.

Genomic DNA was analyzed using next-generation sequencing (NGS). A total of 105 genes involved in neuromuscular diseases and risk of malignant cardiac arrhythmias were analyzed (ACTA1, AGRN, ANO5, B3GALNT2, B4GAT1, BAG3, BIN1, CAPN3, CAV3, CCND3, CFL2, CHAT, CHKB, CHRNA1, CHRNB1, CHRND, CHRNE, CNTN1, COL6A1, COL6A2, COL6A3, COLQ, DAG1, DES, DMD, DNAJB6, DNM2, DOK7, DPAGT1, DPM1, DPM2, DPM3, DSC2, DSG2, DSP, DYSF, EMD, FHL1, FKRP, FKTN, FLNC, FOS, FXN, GAA, GFPT1, GMPPB, GOSR2, HRAS, ISPD, ITGA7, KBTBD13, KCNQ1, KCNH2, LAMA2, LAMP2, LARGE, LDB3, LMNA, MOK2, MTM1, MUSK, MYF6, MYOT, MYH7, MYBPC3, NEB, NESPRIN2, NUP88, PCNA, PLEC, PKCA, PKP2, POMGNT1, POMGNT2, POMK, POMT1, POMT2, RAPSN, RYR1, RYR2, SCN4A, SCN5A, SEPN1, SGCA, SGCB, SGCD, SGCG, SLC25A4, SREBP, TAZ, TCAP, TMEM5, TMEM43, TMPO, TNNT1, TNNI3, TNNT2, TNPO3, TPM2, TPM3, TRIM32, and TTN). All gene isoforms described in Ensembl 75 (www.ensembl.org) that have been linked with either a RefSeq code (www.ncbi.nlm.nih.gov/refseq) or CCDS (www.ncbi.nlm.nih.gov/CCDS) were included. Sequence data coordinates were based on UCSC human genome version hg19 (NCBI GRCh37 built). Biotinylated cRNA probe solution was used as a capture probe (Agilent Technologies, Santa Clara, CA, United States). Probes were designed using eArray (Agilent Technologies).

Non-common genetic variants [minor allele frequency (MAF) < 1%] identified throughout NGS analysis were confirmed using Sanger sequencing. Exons and exon–intron boundaries of each gene were amplified (Verities PCR, Applied Biosystems, Austin, TX, United States), and the resulting PCR products were purified (Exosap-IT, Affymetrix Inc., USB Products, Cleveland, OH, United States) and directly sequenced in both directions (Big Dye Terminator v3.1 and 3130XL Genetic Analyzer, both from Applied Biosystems). The Posterior SeqScape Software v2.5 (Life Technologies, Carlsbad, CA, United States) was used to compare results with the reference sequence from hg19. The identified rare variants were contrasted with the Human Gene Mutation Database (www.hgmd.cf.ac.uk/ac/index.php) and Genome Aggregation Database (gnomAD) (www.gnomad.broadinstitute.org). To detect copy number variation (CNV), we looked for significant differences between expected and obtained normalized coverage for a given sample in a region of interest. Several samples were analyzed to corroborate similar levels of coverage between samples. All CNVs were compared with the CNV Control database (www.gwas.biosciencedbc.jp/cgi-bin/cnvdb/cnv_top.cgi), Database of Genomic Variants (www.dgv.tcag.ca/dgv/app/home), DECIPHER (www.decipher.sanger.ac.uk), and gnomAD (www.gnomad.broadinstitute.org). Rare variants that were potentially deleterious and confirmed in the index case were analyzed using the Sanger method in the relatives.

Each rare variant was classified following current recommendations of the American College of Medical Genetics and Genomics (ACMG) (Richards et al., 2015). A vast majority of pathogenic (P) variants are extremely rare (<0.01%). All available data concerning each rare genetic variant was updated until submission time (June, 2022). Variants classified as Variant of Unknown Significance (VUS) in alternative genes were further sub-classified. Variants identified showed no reported MAF or low MAF. Certain association with any neuromuscular disease were considered as VUS with highly suspicious Likely Pathogenic role (VUS-LP); thus, they were included to clarify their potential role in clinical practice. To avoid bias, five investigators independently investigated genetic data concerning each analyzed variant in our study. Finally, all investigators discussed data included in each item of the ACMG and consensus as well as final classification of all rare variants.

Our study included 26 pediatric patients (mean age 8.2 years at enrollment; IQR, 4–12.5 years; 53.8% males) of 25 families, with a total of 76 individuals (26 index cases and 50 relatives). Two of the index cases enrolled were female monozygotic twins (index cases 23 and 24). Families were originally from Spain (12), United Kingdom (3), United States (3), Australia (2), Canada (1), France (1), Greece (1), Russia (1), and Argentina (1). No consanguinity occurred in any of families. No potential common ancestor was identified after the family interviews. All index cases accomplished with clinical criteria for LMNA-related muscular disease: 11 Patients (42.3%) were L-CMD, 13 (50%) were EDMD, and 2 (7.7%) presented as LGMD1B. Early-onset skeletal muscle impairment before 2 years of age was detected in 23 of the 26 cases (88.4%) (Figure 1). Clinical assessment was performed in all patients included in our cohort, confirming previous diagnosis for each LMNA-related muscular disease.

Our NGS analysis showed an average call rate of 99.25% achieved at 30x coverage. The median coverage per sample was 892 (749–1286). An average of four failed whole exons occurred in each sample, and all these exons were amplified using Sanger sequencing. All rare variants (MAF < 1%) were confirmed also using Sanger sequencing, discarding false positive signal. No CNV were identified in any of genes analyzed, including LMNA. Previous karyotype (performed at time of diagnosis, out of our centre) also discarded any large chromosomic alteration in all patients included in our cohort.

All patients included in our study had a previous genetic analysis of the LMNA gene. This previous analysis identified rare variants in this gene as potential cause of the disease. Our targeted-gene panel analysis confirmed all previous rare variants in the LMNA gene (Figure 2). No additional rare or common variant classified as pathogenic (P) or likely pathogenic (LP) were identified in LMNA. As above mentioned, previous genetic analysis identified rare variants in the LMNA gene, which were classified according to available data at the moment of genetic analysis was performed. Rare variants in LMNA were reclassified following ACMG guidelines and accordingly to current data available (June, 2022). All LMNA variants remain classified as P or LP, without any modification in comparison to previous genetic report.

A total of 19 LMNA rare variants were identified (17 exonic and 2 intronic). Of exonic rare variants, 4 were delins and 13 missense. All variants were identified in heterozygous state (Figure 2). Eleven rare variants (57.89%) were classified as LP and 7 (36.84%) as definitively P. The most frequent rare variant identified in LMNA was p.Arg249Trp (5 patients, 19.23%), as previously reported (Quijano-Roy et al., 2008; Pasqualin et al., 2014; Heller et al., 2017; Ben Yaou et al., 2021; Fan et al., 2021; Jedrzejowska et al., 2021). These 5 patients were diagnosed with L-CMD (4 patients -index case 1, 2, 6, and 26-) and EDMD (1 patient -index case 14-). Other two rare variants were identified two times in different patients each one (p.Asn39Ser, patients 3 and 7; p.Arg453Trp, patients 20 and 21). Both rare variants were also previously reported (Pasqualin et al., 2014; Ben Yaou et al., 2021; Bennett et al., 2021; Fan et al., 2021; Jedrzejowska et al., 2021). A total of 6 cases showed DCM (patient 3, 5, 9, 11, 13, and 16), five of them diagnosed with EDMD and only one with L-CMD (patient 11). Among the patients with ICD (patients 1, 3, 5, 13, and 16), four were diagnosed with EDMD and one with L-CMD (patient 1). Index case number 2, diagnosed with L-CMD and carrying a pacemaker (PM), showed prolonged asystole episodes. Unfortunately, two cases died after inclusion in our study (patient 3 due to rapidly progressive heart failure despite optimal treatment and ICD carrier and 22due to severe respiratory infection) (Tables 1–3) (Figure 1). Finally, focused on intronic LMNA rare variant (c.810 + 1G>C) in patient 17, the clinical diagnostic was LGMD1B. The patient was not a carrier of any other rare variant in either the LMNA gene or any other gene. In addition, none of relatives showed any clinical symptom (Table 1).

Segregation of genetic variants in families showed that in 24 cases (92.3%), the rare variant in LMNA was de novo (including intronic variant in case number 17), as widely published (Ben Yaou et al., 2021; Jedrzejowska et al., 2021). Only in two families (index cases 11 and 18), one of parents carried the same deleterious variant in LMNA. In both these cases, the parents showed minor muscular impairment (Figure 3).

Eleven rare variants were identified in 8 genes (AGRN, CHRND, COL6A3, DYSF, FLNC, GAA, LAMP2, and NEB) encoding structural proteins. The AGRN gene encodes the protein Agrin, associated with Congenital myasthenic syndrome following an autosomal recessive (AR) pattern of inheritance. The CHRND gene encodes the Cholinergic receptor nicotinic delta subunit, mainly related to Myasthenic syndrome following autosomal dominant (AD) or AR pattern of inheritance. The COL6A3 encodes Collagen type VI alpha 3 chain protein and deleterious variants are mainly associated with muscular dystrophy following both AD and AR patterns of inheritance. The DYSF gene encodes the protein Dysferlin, mainly associated with muscular dystrophy following an AR pattern of inheritance. The FLNC gene encodes the protein Filamin C, mainly associated wit cardiomyopathies following an AD pattern of inheritance. The GAA gene encodes Alpha Glucosidase and deleterious variants in this gene are related to Pompe disease following an AR pattern of inheritance. The LAMP2

Gene encodes the Lysosomal associated membrane protein 2, related to Danon disease following a X-linked pattern of inheritance. Finally, the NEB gene encodes the protein Nebulin, and pathogenic variants in this gene cause mainly myopathy following an AR pattern of inheritance (Table 2). At this point, it is important to remark that none of these proteins have a close relation to the LMNA gene according to consulted protein databases.

These rare variants were identified in 8 families −1, 2, 5, 7, 11, 18, 22, and 23/24- (30.76%). No other rare or common alterations (single variants, delins, or CNV), classified as P, LP, or VUS-LP, were identified in any of additional genes analyzed. All the new 11 rare variants were exonic and in heterozygous state. One variant was non-sense (patient 2), 3 not-in-frame deletions (patients 2, 11, and 18), 2 not-in-frame delins (patients 23/24 -twins-), and 5 missense (patients 1, 5, 7, 18, and 22). All rare variants were classified as LP or VUS-LP following ACMG recommendations. None rare variant was repeated in more than one patient except in twins (patients 23/24) (Tables 1–3) (Figure 1).

Segregation of genetic variants in families showed that none of these rare variants was de novo. Curiously, only two variants in GAA (p.Arg594His, CM121193-patient 1-and p.Asp645Asn, CM980804-patient 18-) were previously identified and associated with Glycogen Storage Disease (GSD) (Huie et al., 1998; McCready et al., 2007; Liu et al., 2014). None of parents (except families 11 and 18, due to LMNA) were previously diagnosed or showed any symptom associated with the genes identified in this study (Figure 3).

Previous clinical diagnosis and follow-up showed 9 patients with more severe phenotypes (6 EDMD -patients 3, 5, 9, 13, 16 and 22- and 3 L-CMD -patients 1, 2 and 11-). All nine patients carried a rare missense deleterious variant in LMNA except two (patient 9-EDMD-and 11-L-CMD-) whose carried deletions (p.Glu31ArgfsTer65 and c.1488_1488+9del, respectively) (Tables 1–3).

Six patients diagnosed with EDMD (patients 3, 5, 9, 13, 16, and 22) showed different phenotypes: DCM (patient 9) or DCM and carried an ICD (patients 3, 5, 13, and 16). Curiously, patient 22 did not show any risk factor but died suddenly. Patient 3 also died suddenly despite carrying an ICD. Only patient 9 carried a rare variant in LMNA (c.91delG) while all other patients carried a missense variant in the same gene. Concerning additional rare variants in other genes, only patient 5 carried a rare missense variant (c.1952G>A) in GAA, inherited from their unaffected mother (Figure 1).

Three patients diagnosed with L-CMD (patients 1, 2, and 11) also showed different phenotypes: DCM (patient 11) or carried an ICD (patient 1) or a PM (patient 2) but without DCM. Curiously, patients 1 and 2 (no family relationship) carried the same missense rare variant in LMNA (p.Arg249Trp) and patient 11 carried an intronic rare variant (c.1488_1488+9del). Additionally, all patients carried at least one rare variant: patient 1 in GAA (p.Arg594His) inherited from their mother and without any symptom to date, patient 2 carried two rare variants, one in AGRN and other in FLNC (p.Arg740Ter and p.Cys644TrpfsTer27, respectively), both rare variants inherited from their healthy mother, and patient 11 carried a rare variant in NEB (p.Met5792_Arp5794delinsIle), inherited from their mother who showed a mild neuromuscular affectation but their brother carried the same variant in NEB without any symptom diagnosed to date (Tables 1–3) (Figures 1, 2).

All LMNA variants were deleterious, which was the main cause of the clinically diagnosed disease. However, different onset as well as disease progression seems to be modified by other variants, in concordance to previously suggested but not exhaustively analyzed to date. The variant p.Arg249Trp was identified in 5 patients (4 L-CMD and 1 EMD2). This variant was previously reported in several patients diagnosed with L-CMD (Ben Yaou et al., 2021). The variant was de novo in all reported cases and, in addition to early onset of muscular involvement, patients also showed cardiac involvement and malignant arrhythmias (Quijano-Roy et al., 2008; Komaki et al., 2011; Pasqualin et al., 2014; Ben Yaou et al., 2021; Fan et al., 2021; Jedrzejowska et al., 2021). In view of age-dependent penetrance for heart involvement due to deleterious variants in LMNA, a regular cardiological supervision should have been offered (Jedrzejowska et al., 2021), particularly in p.Arg249Trp carriers due to clinical severity (Ben Yaou et al., 2021).

Patients 1 and 2, diagnosed with L-CMD showed most severe phenotype and carried additional potentially deleterious rare variants (patient 1 in GAA and patient 2 in AGRN and FLNC). In patient 1, the GAA_p.Arg594His (CM121193) was previously reported and associated with GSD following an autosomal recessive pattern of inheritance (Liu et al., 2014). It was inherited from their asymptomatic mother and none showed any symptom of GSD or Pompe disease due to the heterozygous form. In patient 2, both variants were novel, inherited from their mother and classified as deleterious. The variants in FLNC are mainly associated with Hypertrophic cardiomyopathy (not observed in patient 2 or the mother) and variants in AGRN are mainly associated with congenital myasthenic syndrome following an autosomal recessive pattern of inheritance. Therefore, neither mother or sister showed any symptom of muscular weakness as both are carriers of the same heterozygotic variant. Therefore, the presence of any deleterious rare variant may explain the most severe muscular weakness and malignant arrhythmias observed in patients 1 and 2, in comparison to patients 6 and 26 (both showing same diagnosis and carrying the same LMNA variant but without aggressive phenotype). However, further molecular studies should be performed to unravel the pathophysiological mechanism involved in these potential phenotype modifications. Curiously, patient 14, diagnosed with EDMD and carrying the same variant p.Arg249Trp, showed a different phenotype possibly due to an unidentified alteration, reinforcing the targeted genetic analysis not only limited to the LMNA gene in patients diagnosed with LMNA-related muscular diseases. In addition, in p.Arg249, other deleterious variants were identified in patient 8, who was diagnosed with EDMD. This variant (p.Arg249Gln) was also previously reported in 3 infants showing slow muscular degeneration and slight arrhythmias (Bonne et al., 2000; Fan et al., 2021; Jedrzejowska et al., 2021), similar to this patient.

Patient 5, who showed a severe phenotype of EDMD, with DCM and implanted ICD due to malignant arrhythmias, carried two rare variants (LMNA_p.Arg453Pro and GAA_p.Gly651Asp). Both variants were novel. The GAA variant was inherited from the asymptomatic mother and no symptoms of GSD/Pompe disease in any carrier were observed due to widely-accepted recessive pattern of inheritance in this gene. Curiously, in the same aminoacid p.Arg453, another deleterious rare variant was previously reported as de novo in several cases (p.Arg453Trp) (Bonne et al., 2000; Fan et al., 2021; Jedrzejowska et al., 2021). In these reported cases, the diagnosis was EDMD concomitant with arrhythmias and slow muscular degeneration. Only in one case, the diagnosis was LGMD1B. In our cohort, two cases carried this rare variant LMNA_p.Arg453Trp, patient 20 diagnosed with EDMD and patient 21 diagnosed with LGMD1B (both showing slow muscular degeneration, without any arrhythmia or cardiac alteration).

Patient 7, diagnosed with EDMD but without any cardiac alteration, carried the deleterious LMNA_p.Asn39Ser variant. She carried an additional rare variant in the LAMP2 gene, also identified in her mother. Rare variants in this gene are associated with Danon disease, following an X-linked pattern of inheritance. However, the mother did not show any symptom/phenotype related to Danon disease. The rare variant in LMNA was previously published in 5 patients, of which 3 were diagnosed with L-CMD and slow muscular weakness progression, and 2 with EDMD and no cardiac affectation (Pasqualin et al., 2014; Fan et al., 2021). In our cohort, patient 3 also carried the same variant but with a clinical diagnosis of EDMD. Despite no additional deleterious variant identified in any of all analyzed genes, DCM and malignant arrhythmias were documented. Unfortunately, this patient died due to rapidly progressive heart failure despite optimal treatment and being an ICD carrier. It suggests the targeted genetic analysis looking for other new genes not currently associated with any muscular diseases. Curiously, in the same aminoacid LMNA_p.Asn39, twins included in our cohort (patients 23/24) carried the deleterious variant (p.Asn39Lys) responsible of L-CMD diagnosed. This variant was recently reported in one case of L-CMD with slow muscular weakness progression and no cardiac affectation (Fan et al., 2021) and in two cases showing hypotonia, waddling gait and normal heart (Jedrzejowska et al., 2021), in concordance to our twins.

Patient 11, diagnosed with concomitant L-CMD and DCM, carried a de novo and novel deletion LMNA_c.1488_1488+9del. This variant was inherited from her mother who showed a minor muscular impairment. No history of any muscular disease was documented in previous generations. Patient 11 also carried an additional deletion in NEB, a gene associated with myopathies following an autosomal recessive pattern of inheritance. Both their mother and brother carried this NEB variant in heterozygosis form and, as expected, showing no symptom to date. In our cohort, three more de novo deletions in LMNA were identified in patients 4, 9, and 25, where two were diagnosed with L-CMD (patients 4 and 25) and showed slow muscular weakness progression with no cardiac affectation, while patient 9 was diagnosed with EDMD and DCM. In concordance, both deleterious variants were previously reported in L-CMD patients showing phenotypes similar to our patients (Fan et al., 2021).

Patient 18, diagnosed with EDMD, showed slow muscular weakness progression and no cardiac affectation. The patient carried LMNA_p.Gln36His, inherited from the father who showed a minor muscular impairment. No history of any muscular disease was documented in previous generations. This variant was never reported, to the best of our knowledge. This patient also carried the deleterious variant GAA_p.Asp645Asn, previously identified and associated with GSD (CM980804) (McCready et al., 2007). This GAA variant was inherited during heterozygosis from the healthy mother, not showing any symptom of GSD or Pompe disease. In addition, this patient also carried a deleterious indel in COL6A3. This variant has not been reported so far, and the gene is associated with dystonia and muscular dystrophy. The variant was inherited from the father who showed a minor muscular impairment. As mentioned above, no history of any muscular disease was documented in previous generations.

Patient 22, diagnosed with EDMD, showed slow muscular weakness progression and no cardiac affectation. The LMNA_p.Glu31Lys variant was de novo. Despite no aggressive phenotype, the patient died at 10 years old due to severe respiratory infection. This variant has been recently reported in one patient diagnosed with L-CMD, slow muscular weakness progression, and no cardiac affectation (Fan et al., 2021). This patient also carried a deleterious variant in DYSF (p.Arg1849Lys), the gene associated with an autosomal recessive muscular dystrophy. This variant was inherited from the asymptomatic mother, as also observed in the brother who also carried the same DYSF variant. No history of any muscular disease was documented in previous generations. Finally, patients 13 and 16, both diagnosed with EDMD, DCM and with an ICD, carried only one deleterious variant in LMNA (p.Leu38Phe and p.Arg455Pro, respectively). Both rare variants are novel and de novo, after segregation of both variants in relatives.

In conclusion, laminopathies associated with muscular disorders are a group of heterogeneous conditions with different onset and development. We suggest that a targeted genetic diagnosis including LMNA as well as other genes related to muscular diseases may help to unravel additional potential rare variants that could be associated with more severe phenotypes. However, translation into clinical practice should be performed with caution due to further studies in large cohorts are necessary to clarify role additional variants.