Genetic profile of Charcot-Marie-Tooth disease in the Saudi population: A retrospective study highlighting the role of consanguinity

Discussion

We retrospectively reviewed all CMT cases admitted to NGHA hospitals, in Riyadh, Jeddah, and Alahsa between 2016 and 2022 to reveal the underlying genetic causes of the disease. In our cohort, only 23 out of 43 (53%) of the CMT cases underwent genetic testing. Relevant genetic variants to CMT were identified in 19 out of 23 (82%) of tested cases, underscoring its clinical utility. However, the proportion of patients in our cohort who received a genetic diagnosis, 19 out of 43 (44%), is lower than the 76% to 63% rates reported in previous studies.(6,22)

Prior studies highlight PMP22, GJB1, MFN2, and MPZ as major causative genes for CMT in other populations.¹ However, in our cohort, the predominant genes were MTMR2, PMP22, and FGD4, with no cases linked to MPZ and only one to GJB1, excluded due to incomplete medical records. Although the number of cases is limited, this lack of similarity with other studies in causative genes suggests that the genetic architecture of CMT in the Saudi population differs from that observed in other populations, highlighting both the genetic heterogeneity of CMT and notable inter-population variations.

In contrast to worldwide data where the most common mode of CMT inheritance is autosomal dominant, our findings highlight the high prevalence of autosomal recessive CMT, with no clear predominance of a single CMT subtype. This pattern aligns with observations from other regions with high consanguinity rates, where autosomal-recessive forms of CMT—encompassing multiple genetic subtypes—account for a larger proportion of cases, and correlates with the high consanguinity rates reported in the Saudi population.^8,15 Moreover, despite the fact of scarce specific epidemiological data in Saudi Arabia, the genetic implication of consanguineous marriages is evident by the tendency towards a more severe phenotype in patients with autosomal recessive CMT among our cohort. Autosomal recessive inheritance is common across several inherited neurological disorders in regions with high consanguinity, such as Saudi Arabia and other Middle Eastern countries. This pattern is seen in autosomal recessive spinal muscular atrophy,²³ and certain metabolic neuropathies,²⁴ which occur at higher relative frequencies compared to Western cohorts. In contrast, global data show a predominance of autosomal dominant inheritance for CMT.¹ The high incidence of positive family history in 53% of cases further stresses the importance of genetic diagnosis for providing informed counseling to affected families. In addition, we noticed a stratification effect through intra-tribe marriages as 2 out of 4 patients with MTMR2 variants have the c.1736_1745delinsCC p.(Tyr579SerfsTer22), and 2 out of 3 patients with FGD4 variants having the c.1740C>A p.(Tyr580Ter) are related to the same tribe.²⁵ Given these insights, we strongly recommend integrating genetic testing into the standard diagnostic workflow for CMT to enhance the identification of causative genes, particularly in regions with high rates of consanguinity and intra-tribe marriages. By revealing these genes, healthcare providers can offer more targeted genetic counseling and risk assessment to families, helping them make informed decisions about family planning and managing the potential risks associated with consanguineous unions. Furthermore, as advances in precision medicine and gene therapy continue to evolve, the identification of CMT genes relevant to the local population enables personalized treatment strategies.^10,26–28 These advances hold the potential not only to improve the quality of life for individuals affected by CMT but also to reduce the incidence of the disease in future generations through early interventions.

In this study, we report 5 novel variants in clinically diagnosed CMT patients. The NM_016156.6: c.357+1del in the MTMR2 gene variant was identified as a homozygous variant in a patient with severe distal muscle wasting and atrophy in both upper as well as lower limbs, flexion contractures of the proximal interphalangeal joints, foot drop with a high-steppage gait, hypotonia, and absent reflexes, while superficial touch and deep pain sensations remained intact. The patient’s symptoms began at the age of 11 years. The variant is a null variant affecting the splice donor site at a conserved nucleotide. It was absent from control chromosomes within the GnomAD database,¹⁸ and had not been reported in ClinVar.¹⁷ In silico prediction suggests a potential impact on splicing, which may contribute to the disease phenotype. Given the available evidence, this variant is classified as likely pathogenic. The NM_001370298.3:c.1471C>T p.(Arg491Trp) in the FGD4 gene was identified as a homozygous variant using WES in a patient with a classical CMT1 phenotype developed at 6 years old, accompanied by kyphoscoliosis, bilateral hip dysplasia, and delayed motor development. Positive family history was reported in this case; however, segregation analysis was not performed due to the lack of genetic material from other family members. The variant is a missense variant that alters arginine, a positively charged amino acid, with tryptophan, a hydrophobic amino acid at codon 491. It is observed in the GnomAD database,¹⁸ at a frequency of 3.7×10-6 with no homozygous event. Multiple in silico tools predicted a pathogenic effect of this variant. As of September 2, 2024, there was only one submission in ClinVar (Variation ID: 2122304) for a variant affecting the same codon with different amino acid change p.(Arg491Leu). Given the available evidence, this variant is classified as a VUS. The NM_181882.3:c.586del p.(Arg196GlufsTer117) in the PRX gene was detected as a homozygous variant using WES in a patient with severe pes cavus, stiff movements, scoliosis, and MRI findings revealing spina bifida at the S1 vertebral level. The patient’s symptoms began at the age of 9 years. The variant is a frameshift deletion, leading to the alteration of codon 196 of the PRX gene. This variant was not present in control chromosomes within the GnomAD database,¹⁸ and had not been reported in ClinVar.¹⁷ However, a nonsense mutation at the same position (Arg196) has previously been identified in a patient with CMT and has been classified as pathogenic in ClinVar.^29,30 Given the available evidence, the variant is classified as pathogenic. The NM_002972.4:c.5583+1G>A and c.4631dup p.(Arg1545ProfsTer10) in the SBF1 gene were detected as compound heterozygous, each inherited from a different parent. The variants were detected using WES in a patient with microcephaly, and significant gross motor developmental delay, with MRI findings indicating multifocal abnormalities in the brainstem and conus medullaris. The patient’s symptoms began in the first year of life. Neither variant had been reported in ClinVar. The c.5583+1G>A is an intronic variant affecting a splice donor site. The variant is observed in the GnomAD database at a frequency of 1.85×10^-6 with no homozygous event. Multiple in silico tools predicted a negative effect on splicing and a pathogenic impact. The c.4631dup is an insertion of one nucleotide resulting in a frameshift mutation. This variant is not present in control chromosomes within the GnomAD database. Given the available evidence, the variant is classified as likely pathogenic. The NM_014365.3:c.421A>C p.(Lys141Gln) in HSPB8 is detected as a heterozygous variant using WES in a patient with classical autosomal dominant CMT2 phenotype. The patient’s symptoms began at the age of 12 years. This is a missense variant that replaces lysine, a positively charged amino acid, with glutamine a neutral amino acid. This variant is not present in control chromosomes within the GnomAD database. Multiple in silico tools predicted a damaging effect on the protein. Variants that disrupt the p.Lys141 amino acid residue in HSPB8 have been observed in affected individuals,^31,32 and alternative variants affecting the same codon are reported in ClinVar as pathogenic (421A>G p.K141E (VCV000002618.30), c.423G>T p.(K141N) (VCV000002619.5), and 423G>C p.(K141N) (VCV000002617.3). Given available evidence, the variant is classified as likely pathogenic.

Globally, reported prevalence for CMT varies widely, with the highest estimates in Northern Europe (~31/100,000) and the lowest in Asia (~11/100,000).² There is a lack of epidemiological studies on CMT in Saudi Arabia and other Arabian Gulf countries. This gap limits our ability to directly compare national or regional prevalence with other parts of the world. This study is constrained by its relatively small sample size, encompassing patients from only three regions of the country between 2016 and 2022, coinciding with the adoption of electronic medical records. Furthermore, 52% of CMT cases were diagnosed based solely on clinical symptoms without genetic confirmation, indicating a significant gap in our understanding of the genetic underpinnings of CMT in nearly half of the cases. In addition, most of the included cases (13 out of the 17) are from the Riyadh hospital, which serves as a national referral center, and has a higher rate of genetic testing. As a result, geographical prevalence and regional subtype distributions cannot be inferred. Future research should aim to include a larger sample size across more hospitals and capture the region of origin to better elucidate the genetic landscape and the utility of genetic testing in CMT, as well as to determine regional subtype distribution of the disease. Additionally, extending genetic testing to all clinically diagnosed cases could uncover further causative genes and clarify the sub-classification of CMT, thereby improving diagnostic accuracy and therapeutic strategies.