All participants provided written informed consent under protocols approved by the local research ethics committee (2019/ETH12538). The clinical phenotypes of three affected individuals investigated in the present study are presented in Table 1. All three individuals are from the same family with an extensive history of autosomal dominant ataxia. The family is of Italian-Australian ethnicity and individual III-3 reports she has several members of her extended family in Malta who are also likely to be affected by ataxia, but who were not available for clinical or genetic assessment (Fig. 1).
The three affected individuals presented with slowly progressive and longstanding ataxia. The age of onset could not be recalled by the individuals, however individuals IV-1 and III-3 reported that they have had the symptoms at least since their early teenage years. They all exhibited signs of gait ataxia including broad-based unsteady gait with a negative Romberg’s test. Signs of mild limb ataxia were observed in two individuals (III-3 and IV-1), including ballistic dysmetria, dysdiadochokinesia and heel-shin ataxia. Slight dysarthria was observed only in one individual (III-3). One individual exhibited horizontal nystagmus (IV-5), whereas another individual displayed both bidirectional gaze-evoked horizontal and vertical nystagmus (III-3). All three individuals showed saccadic intrusions into smooth pursuit eye movement. They all had normal tone, power, deep tendon reflexes (DTRs) and down-going plantar responses with no evidence of sensory neuropathy on upper and lower limb examination. Symptoms related to autonomic disturbances were not reported. Importantly, the affected individuals showed no evidence of past or current EK-related skin lesions. Of note, one individual (IV-5) was formally reviewed by a dermatologist who confirmed the lack of EK, although this patient had mild eczema of the hands and tinea corporis on the right buttock confirmed on skin biopsy. There were no signs that indicated possible cognitive decline in any of the individuals. On brain magnetic resonance imaging (MRI), all three individuals showed cerebellar and pontine atrophy, with the older affected member (III-3) showing severe cerebellar atrophy (Fig. 2). Another individual (IV-1) also showed pontine T2 signal hyperintensity which is a sign typically associated with multisystem atrophy.
Genetic studies:
The proband (IV-5) tested negative for common repeat expansions causing ataxia (SCA1, 2, 3, 6, 7, 12, 17 and Friedreich's ataxia). Following this, the proband underwent next generation sequencing-based custom neuromuscular gene panel (Neuro v6) testing at PathWest Laboratory Medicine. A heterozygous missense variant in the ELOVL4 gene, NM_022726.4:c.698C > T (p.Thr233Met) was identified. This result was confirmed by Sanger sequencing. This rare variant is not found in the GnomAD population database and has a high CADD score (v1.6) of 27.5m. Importantly, this variant was previously reported in SCA34 cases exhibiting typical EK lesions (6, 8, 9). The variant was originally classified as a variant of uncertain significance (VUS) according to ACMG criteria (PM2, PP3, PP5). The two additional affected individuals subsequently underwent the Sanger sequencing, confirming that they carried the variant (Fig. 1), consistent with segregation of the ELOVL4 variant with ataxia in the family. The original VUS classification of this ELOVL4 variant was thus upgraded to likely pathogenic (ACMG criteria: PM2, PP3, PP5 and PP1 (moderate)).
Literature review:
A systematic literature review was performed by searching databases, including PubMed, ScienceDirect, Web of Science and SCOPUS, using the terms “SCA34” or “ELOVL4 ataxia”. Publications with English abstracts describing cases with an identified ELOVL4 variant were included.
In total, we identified 59 reported cases of SCA34 (Table 2, Table 3). The estimated mean of age of onset was 30.94 (± 13.71) years (Mean ± SD). Most cases had cerebellar ataxic features including gait ataxia (88.1%), limb ataxia (76.3%), dysarthria (62.7%), and nystagmus (57.6%). On brain MRI, cerebellar atrophy was reported in all cases (100%) and the hot cross bun sign, usually associated with multiple system atrophy cerebellar type, was seen in 30.3% of cases.
Table 2
Clinical findings according to variant in reported cases of ATX-ELOVL4
| ELOVL4 variant | Number of cases | Age of onset | Cerebellar signs |
| Gait ataxia | Limb ataxia | Dysarthria | Nystagmus |
Present study | c.698C > T (p.T233M) | 3 | 13.3 | 3 | 2 | 1 | 2 |
Beaudin et al. (2020) | c.504G > C (p.L168F) | 9 | 47 | 9 | 9 | 9 | 7 |
Bourassa et al. (2015) | c.539A > C (p.G180P) | 1 | 25 | 1 | 1 | 1 | 1 |
Bourque et al. (2018) | c.698C > T (p.T233M) | 1 | 15 | 1 | 0 | 0 | 0 |
Cadieux-Dion et al. (2014) | c.504G > C (p.L168F) | 19 | 51 | 12 | 9 | 6 | 7 |
Haeri et al. (2021) | c.539A > C (p.Q180P) | 1 | 13 | 1 | 1 | 1 | 1 |
Moreno-Escobar and Tripathi (2022) | c.736T > G (p.W246G) | 1 | 30 | 1 | 1 | 1 | 1 |
Mukherjee et al. (2021) | c.539A > C (p.Q180P) | 1 | 22 | 1 | 1 | 1 | 1 |
Ozaki et al. (2019) | c.698C > T (p.T233M) | 2 | 47 | 2 | 2 | 2 | 2 |
Ozaki et al. (2015) | c.736T > G (p.W246G) | 9 | 33.9 | 9 | 9 | 9 | 7 |
Wang et al. (2021) | c.698C > T (p.T233M) | 2 | 31.5 | 2 | 2 | Not reported | Not reported |
Xiao et al. (2019) | c.512T > C (p.I171T) | 10 | 42.6 | 10 | 8 | 6 | 5 |
Total | | 59 | | 52 | 45 | 37 | 34 |
Mean (± SD) | | | 30.94 (± 13.71) | | | | |
Percentage | | | | 88.1% | 76.3% | 62.7% | 57.6% |
Table 3
Summary of clinical findings in reported cases of ATX-ELOVL4
Extracerebellar signs | Brain MRI findings |
Pyramidal signs | Autonomic disturbances | EK-related skin lesions | Cognitive impairment | Ophthalmologic findings | n | Cerebellar atrophy | Hot cross bun sign |
0 | 0 | 0 | 0 | Not reported | 3 | 3 | 0 |
2 | 2 | 0* | 5 | 1 | 6 | 6 | 2 |
0 | Not reported | 1 | Not reported | Not reported | 1 | 1 | 0 |
1 | Not reported | 1 | Not reported | Not reported | 1 | 1 | 0 |
7 | Not reported | 14 | 0 | Not reported | 7 | 7 | 0 |
0 | 0 | 0** | 0 | 0 | 1 | 1 | 1 |
1 | 1 | 0 | Not reported | Not reported | 1 | 1 | 1 |
1 | Not reported | 1 | 1 | Not reported | 1 | 1 | 0 |
2 | Not reported | 1 | 1 | Not reported | 2 | 2 | 2 |
8 | 4 | 0 | Not reported | 0 | 6 | 6 | 4 |
Not reported | Not reported | 2 | Not reported | Not reported | 1 | 1 | 0 |
6 | 0 | 0 | Not reported | 4 | 3 | 3 | 0 |
28 | 7 | 20 | 7 | 5 | 33 | 33 | 10 |
47.5% | 11.9% | 33.9% | 11.9% | 8.5% | | 100.0% | 30.3% |
*One individual from this study exhibited nummular dermatitis |
**This case had congenital ichthyosis |
Importantly, EK-related skin lesions were only seen in a minority of cases (33.9%), which is in contrary to the reported prototypic manifestation of SCA34/ATX-ELOVL4 (OMIM #133190). Whilst the majority of cases with EK skin lesions had EKV (3), several cases had lesions more consistent with progressive symmetric erythrokeratoderma (PSEK) rather than EKV (9, 10). Of note, one study reported a 16-year-old boy with PSEK skin lesions with no neurological or cerebellar clinical signs who was found to have c.698C > T ELOVL4 variant (10). This case highlights the potential clinical phenotypic variability for ELOVL4-associated disease, though it is plausible that this individual develops ataxia later in his life. Similarly, the aforementioned large French-Canadian family study also reported three variant carriers who displayed EKV but no ataxia which may be attributed to their younger ages (3). Additionally, some cases demonstrated skin lesions unrelated to EK, such as ichthyosis (11) and nummular dermatitis (4) and thus were not included in the calculation of prevalence of EK-related skin lesions (Table 2, Table 3). Interestingly, there were total of five reported SCA34 cases with the c.698C > T variant in ELOVL4 and four of them exhibited EK skin lesions (6, 8, 9) in contrast to the present study where the three individuals did not exhibit EK-related skin lesions despite having the exact same nucleotide variant. This finding supports inter-familial phenotypic variability and points against a clear genotype-phenotype correlation. It may also suggest that additional factors, such as environmental and genetic modifiers, may affect the dermatological manifestation of ELOVL4-related disease.
In terms of other extra-cerebellar manifestations, pyramidal tract signs, such as DTRs abnormalities or abnormal plantar reflexes, were found in 47.5% of the cases. Autonomic disturbances, cognitive impairment and ophthalmologic pathologies were found in 11.9%, 11.9% and 8.5% of reported cases respectively, however these values may not be accurate considering the relatively small number of reports on these phenotypic manifestations. However, it should be noted that one study reported the presence of retinitis pigmentosa in four out of eight affected members in one family (5) and another study described cognitive impairment and psychiatric symptoms in five affected members of the studied family who underwent cognitive evaluation (4).