The majority of CLN5 pathogenic and likely pathogenic variants in ClinVar (https://www.ncbi.nlm.nih.gov/clinvar/?term=CLN5%5Bgene%5D, accessed Dec.12018) are null variants including 31 frameshift, 15 nonsense, 3 splicing and 3 intragenic deletion variants. In addition, there are 11 pathogenic missense variants. The frameshift variant detected in our patient had not been reported in HGMD or ClinVar. Large deletion CNVs including CLN5 and of variable size and boundary including monosomy 13q had been reported in over 20 individuals in Clinvar, but none was detected in CLN5 patients. Our patient is the first of such situation that a large deletion involving CLN5 is compounded with a pathogenic frameshift variant in CLN5 gene.
Children with CLN5 disease often have normal early development before regression initiates around the age of four. Our patient exhibited early developmental delay involving motor function and language usage. It is conceivable that the early developmental delay is due to the large de novo deletion since overlapping deletions in postnatal patients reported in Decipher (for example 299003, 261869, 1583 and 250039) often exhibited global developmental delay and hypotonia. This large deletion involved EDNRB gene which is associated with Waardenburg syndrome type 4A (WS4A) and susceptibility to Hirschsprung disease. Both conditions exhibit incomplete penetrance and the penetrance is dose dependent. Biallelic mutations lead to high penetrance. Our patient with one copy deletion did not exhibit features of WS4A or Hirschsprung disease is consistent with the reduce penetrance of these conditions.
The clinical features of Finnish patients with CLN5 mutations have been well described in 1980’s [13, 14]. Visual deterioration, decreased attention span, gait ataxia, tremors, seizure, motor difficulty and cognitive regression were reported at early stages of CLN5 in many clinical cases [9, 10, 15,16,17]. Holmberg et al. reported the clumsiness and concentration disturbances between the age of 2 to 6, followed by mental decline and visual deterioration in eight Finnish vLINCL patients [18, 19]. Epileptic seizures, myoclonia, and ataxia started between the age of 7 to 11, which are the major reasons for severe disability. Simonati et al. characterized the phenotypic profile of 15 children with CLN5 disease . The onset of the disease for these patients ranged from 2 years old to 7 years and 6 months old (median 5y). The most common symptoms included impaired learning and cognition, hyperactivity, aggression, intolerance and motor stereotypes. Seizures and vision impairment occurred relatively later, usually 3 to 4 years after onset of the disease.
We reviewed all previously published CLN5 cases and plotted the age of onset distributions for key features (Additional file 2: Figure S1). Most patients have vision problems around 5–10 years of age. In some patients, visual failure is the first presenting feature. Vision decline would continue and possibly lead to blindness 1 to 5 years after the onset of the symptom [9, 21]. The visual impairment of our patient happened at age 6 (Additional file 2: Figure S1 a). The penetrance of visual impairment reached 80% based on the published cases. This feature has the highest penetrance among other CLN5-related features.
Motor clumsiness mainly occurs around 4 years to 8 years old (Additional file 2: Figure S1 b). Motor impairment or regression, fine motor and gross motor problems, progressive motor retardation, and total loss of motor functions would follow after the initial onset of the symptom [10, 15, 17, 22]. Loss of the ability to walk independently occurs around age 8 to age 13 (Additional file 2: Figure S1 c). In some patients, it may take more than 10 years for motor skill to deteriorate from unsteady gait to inability to walk . Our patient gradually lost motor skills from 4 years old to 7 years old and became bedridden 6 years after the symptom onset. The penetrance of motor clumsinessis 71%. The penetrance of total loss of walking capability is as low as 19% based on the current published cases however the penetrance is age dependent.
Intellectual disability occurs mainly at age 3 to 7 which affects preschool and school age children (Additional file 2: Figure S1 d). The penetrance of intellectual disability is 79%. The onset of language impairment ranged from 2 to 11 years of age (Additional file 2: Figure S1 e). The penetrance of language regression is 60%. Our patient presented language impairment at the age of 4 and showed below the average cognitive competence at the age of 5.
The onset of epilepsy among CLN5 patients mainly occurs around 4–10 years of age and epilepsy penetrance is as high as 66% (Additional file 2: Figure S1 f). Motor impairment usually occurs several years after the onset of seizure [15, 18] and ataxia occurs at 4 to 11 years of age in most patients (Additional file 2: Figure S1 g) with a penetrance of 26% based on reported cases. Both epilepsy and dystaxia appeared when our patient was 4 years old. The initial symptoms of our patient started at age of 4 including decline of language capacity, difficulty in walking and convulsion. It is rarely reported that these three features presented at the initial stage of disease onset.
In this study, we reported the first case of CLN5 patient with a large de novo heterozygous deletion which unmasked a novel pathogenic frameshift variant. Figure S1 presented that most of the CLN5-related features in our patient has an earlier onset in comparison to the average age of symptoms onset, it may well be attributable to the presence of a large deletion accompanying with a loss of function variant. The simultaneous detection of both large and small pathogenic variants in CLN5 demonstrated the advantage of exome sequencing. The molecular findings and the phenotypic information of this patient can help to delineate genotype-phenotype correlations for CLN5.