Introduction

Gillespie syndrome (OMIM #206700) is a rare disorder characterized by the typical association of cerebellar ataxia, bilateral aniridia and intellectual disability. Specifically, a particular type of partial aniridia, without lens opacities or risk for glaucoma development, is distinctive of the syndrome. It was first described by the ophthalmologist Gillespie in 1965, but it took over 50 years to understand its genetic basis, when Gerber and colleagues identified in 2016 either dominant negative pathogenic variants or recessive variants of ITPR1 gene (OMIM*147265) as the cause of the condition [1, 2].

The gene encodes one of the three members of the IP3-receptors family that form Ca2 release channels, localized in endoplasmic reticulum [2].

We now know that ITPR1 pathogenic variants are responsible for a spectrum of clinical presentation, depending on the type and location of the variant and on the zygosity [2,3,4,5,6]. Heterozygous ITPR1 loss-of-function variants, mostly deletions, cause spinocerebellar ataxia 15 (SCA15, OMIM #606658), one of the most frequent form of adult-onset ataxia not related to repeat expansions [6, 7], while heterozygous missense variants are responsible for the childhood-onset form of the disorder, SCA29 (OMIM #117360) [4, 8]. The key element distinguishing Gillespie syndrome from these two SCAs is the presence of bilateral partial aniridia, deriving from specific dominant ITPR1 variants causing an alteration of the functional ITPR1 domains controlling the formation and/or maintenance of the iris sphincter muscle, or by biallelic variants completely compromising the gene function [5].

Clinically, the two pediatric forms of ITPR1-related disorder are both characterized by infantile hypotonia, global developmental delay (DD) and possible intellectual disability (ID). Ataxia is usually non-progressive, often despite a radiological intensification of cerebellar atrophy is detected. Diagnosis of Gillespie syndrome is driven by the evidence of dilated pupils (fixed midriasis).

Gillespie syndrome is an extremely rare disorder. There is no current therapy for the syndrome, besides motor, cognitive, linguistic and behavioral rehabilitation according to the functioning and chronological age of the subject. Since its first description in 1965, less than 100 patients have been reported and only 37 with molecular confirmation (29 with dominant variants, 8 with the recessive form).

Here, we present two additional cases, both presenting a single pathogenic variant affecting the Gly2554 residue. We describe the clinical evolution of the patients, one of whom is now 17 years old, and discuss the updated phenotypic spectrum of the disorder.

Methods

Genetic analyses were performed by next generation sequencing (NGS) on an Illumina MiSeq apparatus. For patient 1, NGS-targeted resequencing analysis of the entire ITPR1 gene was performed using a probe-based customized gene panel (Illumina Nextera Rapid Capture Custom Kit—Illumina Inc., San Diego, CA, USA) including > 200 genes associated with ataxia and spastic paraplegia. For patient 2, ITPR1 exons 57–60 encoding the mutational-hotspot transmembrane domain were sequenced using a gene-specific amplicon-based NGS protocol (Illumina Nextera XT—Illumina Inc., San Diego, CA, USA).

The general developmental/intelligence quotient (GQ/IQ) level was obtained through standardized scales: Griffiths Mental Development Scales (GMDS) or Wechsler Intelligence Scale for Children (WISC).

A comprehensive literature review was performed revising all papers reporting Gillespie patients present in PubMed database, both single case reports and patient cohorts, ranging from the firsts reports (1965) and up to date. Inclusion criteria were the presence of the 3 main clinical signs defining Gillespie Syndrome (1. iris hypoplasia 2. cerebellar atrophy 3. ataxia) in a patient with confirmed ITPR1 pathogenic variant/s. Exclusion criteria were the absence of a genetic confirmation in a patient with Gillespie phenotype or the presence of ITPR1 variant/s in a patient with a non-Gillespie phenotype.

All variants reported in literature have been rewritten using the NM_001378452.1 transcript in order to have homogeneous molecular records.

Case Reports

Case 1 (Patient 6)

The first new patient is a boy, now aged 9. He is the second child of healthy nonconsanguineous parents. He was born by vaginal delivery after an uneventful pregnancy. Growth parameters at birth were normal (W 3120 g, L 49 cm, HC 33 cm) and he experienced no perinatal distress (Apgar 10/10). Multiple angiomas (face, nape of neck, left buttock) were found, so that abdominal, medullary and brain US were performed, without particular findings.

At 3 weeks of age, the child was taken to the ER for episodes of projectile vomiting: abdominal US excluded a pyloric stenosis, but bilateral midriasis and hypotonia were found, so that the child underwent further exams. An ophthalmic checkup confirmed pupillary midriasis, with mild bilateral iris hypoplasia, hypopigmented fundus, and the presence of epipupillary remnants of Wackendorf's membrane. Brain MRI was normal, as well as a cardiological examination with echocardiogram.

Because of the presence of delay in early motor development, he started physiotherapy and was later brought to our attention at the age of 11 months for further consultation.

He was in good general health, except for a slight underweight (L 71 cm = -1.31 SD, W 7.4 kg = -2.7 SD, HC 45 cm = -0.86 SD). He presented flat angiomas at forehead, both upper eyelids, upper lip, and left buttock. Pupils were mydriatic, with torpid light reflex; visual pursuit was discontinuous. Diffuse muscle hypotonia was present and the child was able to control the head but not to sit autonomously. Brain MRI was repeated and confirmed to be normal; visual evoked potentials (VEP) and electroretinogram (ERG) showed the presence of scotopic-dominant retinal alterations, with photopic responses and visual responses in the normal range; brainstem auditory evoked potentials (BAEP) and electroencephalogram (EEG) were normal.

Despite the absence of cerebellar atrophy, Gillespie syndrome was suspected and confirmed by ITPR1 genetic analysis, which identified a heterozygous de novo pathogenic variant [ITPR1(NM_001378452.1):c.7660G > A p.(Gly2554Arg)], affecting residue Gly2554 that represents a mutational hotspot for Gillespie syndrome [5].

The child was followed with annual clinical controls. He showed a constant improvement of psychomotor development: as for motor skills, he became able to sit independently by the age of 2 and to walk with help at the age of 4, while language development progression was initially slow but reached normal levels during school-age.

Growing up, he developed an ataxic syndrome, characterized by oculomotor dyspraxia, dysarthria, dysmetria, tremor, truncal titubation, and gait instability. SARA score at 8 years old was 23 (Gait 5, Stance 4, Sitting 2, Speech disturbance 4, Finger chase 2, Nose-finger test 2, Fast alternating hands movements 2, Heel-shin slide 2).

A third brain MRI performed at the age of 4 years and 2 months documented an involutional enlargement of the cerebellar-vermian cerebrospinal fluid spaces, with cerebellar vermis and hemispheres atrophy, hyperintense signal alterations of cerebellar gyrus and superior cerebellar peduncles in FLAIR images and shaded hyperintense signal alterations in T2/FLAIR images of occipital subcortical white matter (Fig. 1). A subsequent MRI control at 8 years appeared unchanged.

Fig. 1
figure 1

Brain MRI of Case 1 (Patient 6). Panel 1—Brain MRI at the age of 4 months: Sagittal TSE-T1w (1a) and coronal TSE-T2w (1b) showed no significant morphological abnormality, with asymptomatic small pineal cyst; axial T2w-FLAIR (1c) revealed a mild delay in deep white matter myelination. Panel 2—Follow-up MRI at the age of 4 years: Sagittal TSE-T1w and coronal TSE-T2w showed marked cerebellar and vermis atrophy (2a, arrow) with folia prominence of the superior aspect of cerebellar hemispheres (2b, star). Residual altered signal intensity on T2w was still visible within deep white matter of occipital lobes (2c, dotted)

Full size image

BAEP, PEV, ERG were also repeated twice, resulting normal. EEG registration at 5 and 8 years showed the presence of multifocal epileptic anomalies in sleep, but the child never reported seizures.

Last cognitive evaluation at the age of 8 confirmed cognitive skills within normal limits in all the considered domains, except for Processing Speed score, influenced by severe deficits in visual-perceptual analysis and visuomotor integration. The overall IQ score at Wechsler Intelligence Scale for Children (WISC-IV) was 70 (subscales: Verbal Comprehension 74, Perceptual Reasoning 82, Working Memory 94, Processing Speed 59). The evaluation also evidenced difficulties in school learning processes and some sub-clinical behavioral problems in attention, and emotional and social responses.

Case 2 (Patient 19)

The second new case is a girl, now aged 17 years. She is the first child of healthy nonconsanguineous parents. She was brought to our attention at the age of 2 for the occurrence of developmental delay and a peculiar iris abnormality, initially diagnosed as a partial coloboma with a suspect of glaucoma but later characterized as bilateral iris hypoplasia.

Neurological examination was suggestive for an ataxic syndrome, with ocular motility anomaly, trunk titubation and instability with difficulties in reaching and maintaining the stance position.

The first brain MRI, performed at the age of 2, revealed global cerebellar atrophy with predominant involvement of cerebellar vermis, and a slight T2-hyperintensity of cerebellar cortex. The patient underwent genetic tests, including array-CGH and sequencing of FOXC1, PITX2, and PAX6 genes, all resulting normal. There was a suspicion of Gillespie syndrome, but the genetic bases of the disease were not yet known at that time.

The girl was followed over the years. The administration of Griffiths Scales for Mental Development (GSMD) at 2 years revealed a moderate global developmental delay (DQ 44) with significant weaknesses in motor skills. At the age of 5 she became able to stand independently and walk few steps with help or with a walking frame. Clinical presentation has been stable during follow-up until the age of 17 years, without any worsening, and is currently characterized by an overt ataxic syndrome with oculomotor and orobuccal dyspraxia, dysarthria, truncal titubation in sitting and standing position, dysmetria, intention tremor exaggerated from anxiety or stress, and ataxic gait with severe instability in half turnings. Mild global hypotonia is present, but with vivid deep tendon reflexes and a slight bilateral Babinski sign. At the age of 16, she had an overall SARA score of 25 (Gait 7, Stance 4, Sitting 1, Speech disturbance 4, Finger chase 2, Nose-finger test 3, Fast alternating hand movement 2, Heel-shin slide 2). Consistently with iris hypoplasia, she shows bilateral mydriasis and a very slow fotomotor reflex. The global developmental delay evolved into moderate intellectual disability (no standardized intelligence tests available).

Despite a stable clinical presentation, an evolution of cerebellar atrophy was observed at the age of 3, but later examinations (at 4 and 5 years of age) showed no further progression.

Ophthalmologic and orthoptist controls documented an overall improvement with normal intraocular pressure. VEP and ERG showed a mild alteration of retino-cortical transmission and cortical activation; EEGs never disclosed significant anomalies.

ITPR testing revealed a heterozygous de novo variant: ITPR1(NM_001378452.1):c.7661G > T p.(Gly2554Val). The variant was never reported or annotated before (absent from GnomAD with an 88.42% coverage) but involves the same mutational hotspot residue (Gly2554) as in Patient 1. Including de novo criteria, it is classifiable as pathogenic according to ACMG guidelines (criteria PM2, PM5, PP2, PS2).

Results

In order to give a comprehensive and updated description of Gillespie syndrome, we reported the complete characterization of all the cases present in literature, expanded by the report of 2 further cases. The main features are summarized in Table 1 [2, 5, 9,10,11,12,13,14,15,16,17]. A recent article collecting ITPR variants causing all the related phenotypes has been recently published, but data from patients carrying the Gly2554Val variant are merged, while data about the Lys2611del variant regard only one patient and could be included in the results count [17]. For this reason, Table 1 also contains data from the Tolonen patients but those regarding the 6 cases with variants in the Gly2554 hotspot are not used to calculate the prevalence of the clinical issues and the total count was performed on 33 patients. Molecular data of our patients have also been reported in a recent study on expansion-negative SCAs [18].

Table 1 Molecular and clinical features of the so-far described patients with genetically confirmed Gillespie syndrome
Full size table

One of the patients has been followed from infancy to adolescence, allowing a long follow-up of the disease course. All patients shared the 3 main signs of the condition (mydriasis, ataxia, hypotonia) dominating the clinical picture.

Data concerning development indicate that motor skills progressively improve over time and patients may reach some competences (e.g. sitting, walk with help) in late childhood. Developmental stages are available in 23 of the 33 patients. Motor skill acquisition is significantly delayed in all patients. Age range for sitting position is 7–40 months (mean 21 months, median 18 months), with patients with biallelic mutation showing a significantly more severe impairment (P34 and P35 after 3 years of age, P36, P37, and P38 only able to control the head). Walking appears to be possible only with assistance in most cases (the information is missing in some cases); 11 patients are not able to walk even if assisted, but 4 of them (P11, P22, P28, P38) are still very young. For the other 12, age range for walking was 2y6mo-16y (mean and median 7.5y). Patients with the recurrent Gly2554 aminoacidic change seems to have worse motor performances, as none of them was able to walk by the age of 3; globally, patients with recessive Gillespie show the worse performances.

Interestingly, although a delay in developmental milestones is described in most cases, ID is not invariably present. In the present cohort, data about presence/absence of ID are available in 24 of the 32 patients (data refer to development in 6 patients, only “learning difficulties” are reported in 2, and no information is available in one case). Four out of the 23 patients (17%) have normal intelligence (P1, P3, P6, P23), 17/23 (74%) have mild-to-moderate disability, and only 2 (9%) have severe impairment. For normal-to-mild/moderate disability, no genotype–phenotype correlation seems to emerge, as, for example, patients with the recurrent c.7660G > A variant may have normal intelligence as well as moderate ID. By contrast, patients with biallelic variants demonstrate a worse cognitive profile, as they are the only ones to score in the severe ID range and none of them has normal IQ. For two patients without ID (P1 and P6), similar weaknesses in visuo-motor integration and processing speed abilities have been reported, while information about specific neuropsychological domains is generally not provided in the other clinical reports. A deeper cognitive evaluation considering the degree of impairment not only in general developmental/intelligence quotient but also in specific neuropsychological domains (e.g. visuo-perceptual analysis, visuo-motor integration, language-learning processes, executive functioning) may help to describe cognitive features of the conditions more sensitive to genotypic variability.

Unfortunately, neurological data about the patients reported in literature are poor, in most cases limited to the annotation of hypotonia and ataxia. Therefore, it is not possible to evaluate a possible genotype–phenotype correlation or the presence/absence of evolution of the neurological signs through age. In our cases, although staggering and titubation persist over the years (as observed by the high SARA score), the children have progressively developed compensation mechanisms that allowed a slow but constant progression of motor skills.

Brain MRI confirms in all patients cerebellar atrophy, reported as predominantly affecting the vermis and sometimes associated with signal hyperintensity of diverse cerebellar structures (cortex of hemispheres, centrum semiovale, cerebellar peduncles and gyri) and, in 4 cases, with supratentorial nonspecific white matter alterations (P6, P29, P30, P22, P37). For 8 patients, a clearly evaluable follow-up MRI is available confirming cerebellar atrophy progression in all cases but one. Brain MRI performed in patient 28 did not show cerebellar atrophy, but the boy underwent a single exam at the age of 12 months. It should be noted that also other patients (P22, P23, P35) showed no cerebellar alteration at first MRI performed between 3 and 18 months.

Neurophysiological tests have been performed in some cases. VEP and ERG were altered in only one of the 4 patients who underwent examination (P21). EEG was performed in our 2 cases, showing epileptic anomalies in one of them (P6), who, however, never experienced seizures.

Gillespie syndrome is not usually associated with serious malformations and/or general health issues. General examination revealed normal growth in the majority of patients. Height below the standard curves for age is reported in 3 patients (P11, P25, P26) and poor weight in 3 (P2, P25, P37). Among the 12 patients with available data, 5 exhibited relative (P6, P11, P27, P37) or absolute (P24) macrocephaly. Facial dysmorphisms were reported in some cases (P1, P6, P24, P26, P27, P32, P33, P34), albeit without a specific recognizable pattern. In the cohort of reported patients, the most recurrent medical problems besides the neurological and ophthalmological ones were scoliosis/kyphosis (5 patients) and heart malformations (4 patients). Overall, children with biallelic variants exhibit worse general conditions, with more recurrent presence of medical problems and/or malformative defects.

Discussion

ITPR1-associated ataxias are among the most common causes of genetic ataxia, both in children (SCA29) and adults (SCA15) [6]. Despite sharing similar molecular bases, Gillespie syndrome is extremely rare, as it requires specific dominant or biallelic alterations of ITPR1 to manifest. The first report of a chromosomal abnormality presenting a de novo translocation t(X;11) (p22.32p12) detected in a patient with Gillespie syndrome was published in 1998 [19]. Since then, knowledge on the disease progressively increased for both molecular and clinical aspects.

We describe here two new cases along with a thorough description of all genetically-confirmed cases so far reported, thus offering the most updated overview of the condition. Although some of the results emerging from this complete literature review are important confirmation of what already known from small case-studies or single-case reports, reanalysis of all the 39 cases allows to underline new features and correlation and to better delineate the natural history of the disease.

The first important confirmation is that Gillespie syndrome, like its allelic disorder SCA29, may be included in the group of the so-called Non Progressive Cerebellar Ataxias (NPCA) [20]. In particular, data about development indicate not only that the disease is non-progressive but also that the motor skills may gradually improve over time with patients reaching some relevant milestones (e.g., sitting, walk with help) in late childhood, as described in other early-onset SCAs [21]. As regards life expectancy, the number of middle-aged patients is still very limited to establish whether it is affected in this disorder.

A recent study of a wide cohort of ITPR1 patients, published by Tolonen et al. [17], demonstrated that Gillespie-related variants are sparse within the whole ITPR1 gene, with truncating variant being located in the first three gene domains (suppressor, IP3 binding, and regulatory/coupling) and the less deleterious variants (missense and one in-frame deletion) at the end of the third domain (regulatory/coupling) or in the last one (channel domain). Despite this, no clear genotype–phenotype correlation is found among patients carrying single ITPR1 variants with a Gillespie Syndrome phenotype. In previous studies, it has been hypothesized that the severity of the disease could be modulated by the position of the mutations, the presence of the wild-type protein, and the ability of the mutant protein to be incorporated into ITPR channels [14]. Comparison of all known cases shows clear clinical variability among patients sharing variants at the same amino acid residue, thus implying that molecular information cannot predict evolution of the disease in a given patient.

Otherwise, cognitive functions, neurological presentation, and the general clinical features appear to be more severe in patients with the recessive form of the disease. Although this could be partly expected because of the likely lack of the ITPR1 gene product, it was not mentioned in previous reports.

Conclusions

In conclusion, the study delineates the clinical phenotype of Gillespie syndrome by collecting and revising all existing literature data and reporting two additional patients, giving an overview not only on the neurological features of the condition, but also on the developmental and cognitive aspects and on the general health issues possibly regarding such patients.

This complete overview could serve as a disease guide for the clinicians who have in charge Gillespie patients and need to read a compendium on the condition, but also as a starting point to deepen the knowledge about as-yet-unknown aspects of the disease (e.g., life expectancy and possible genotype phenotype correlation).

Since its first descriptions and even before the identification of ITPR1 as the causal gene, Gillespie syndrome has been considered a non-degenerative ataxia [1, 2]. Data about development and the collection of middle-aged patients lend support to this hypothesis, making this concept a key point both for the clinicians and therapists providing care to Gillespie patients and for the families, particularly those with the younger patients.