A Case of HDR Syndrome and Ichthyosis: Dual Diagnosis by Whole-Genome Sequencing of Novel Mutations in GATA3 and STS Genes

Abstract

Context:

Atypical presentations of complex multisystem disorders may elude diagnosis based on clinical findings only. Appropriate diagnostic tests may not be available or available tests may not provide appropriate coverage of relevant genomic regions for patients with complex phenotypes. Clinical whole-exome/-genome sequencing is often considered for complex patients lacking a definitive diagnosis.

Case Description:

A boy who is now 7 years old presented as a newborn with congenital ichthyosis. At 6 weeks of age, he presented with failure to thrive and hypoparathyroidism. At 4 years of age, he was diagnosed with sensorineural hearing loss. Whole-genome sequencing identified novel mutations in GATA3, which causes HDR syndrome (hypoparathyroidism and deafness), and STS, which causes X -linked congenital ichthyosis.

Conclusion:

Whole-genome sequencing led to a definitive clinical diagnosis in a case where no other clinical test was available for GATA3, and no sequencing panel would have included both genes because they have disparate phenotypes. This case demonstrates the power of whole-genome (or exome) sequencing for patients with complex clinical presentations involving endocrine abnormalities.

Clinical whole-exome and whole-genome sequencing (WES/WGS) has emerged as a powerful diagnostic tool in clinical medicine, particularly in patients with unidentified diagnosis despite extensive workup using available diagnostic tests. Prior publications have demonstrated a diagnostic yield for WES/WGS in the range of 25–30%, particularly for disorders with neurological symptoms (1–3). Patients with endocrine phenotypes have been less well studied.

The syndrome of hypoparathyroidism, deafness, and renal dysfunction, or HDR syndrome (Online Mendelian Inheritance in Man [OMIM] no. 146255), is caused by haploinsufficiency of the GATA3 gene on chromosome 10p (4, 5). GATA3 is a transcription factor important in embryonic development, and it is expressed in the parathyroid glands, inner ear, kidney, thymus, and central nervous system. Variable presentations of patients with HDR syndrome are not uncommon, and the absence of renal involvement has been reported in up to 20% of reported cases (5–8). There can be marked variability in the age of presentation of the various components of the syndrome as well (5, 8).

X-Linked congenital ichthyosis (OMIM no. 308100) is a dermatological disorder caused by a hemizygous loss of function mutations or deletions of the steroid sulfatase gene (STS). The disorder involves excess accumulation of cholesterol sulfate in the surface epithelial cells of the epidermis that fail to exfoliate, resulting in characteristic scale formation in early infancy (9). The STS enzyme is responsible for the hydrolysis of sulfated steroids, and the accumulation of cholesterol sulfate in the epidermis leads to corneocyte retention and thickening of the stratum corneum, resulting in a scaly appearance of the skin (9).

We present a case of a child born with congenital ichthyosis who presented with hypoparathyroidism at 6 weeks of age and sensorineural deafness at 4 years of age. Prior clinical testing (detailed below) failed to identify a diagnosis. However, WGS identified novel mutations in the GATA3 and STS genes that caused the HDR syndrome and X-linked ichthyosis, respectively.

Patient and Methods

Testing for the patient and both parents was performed as part of a hospital-funded program to provide WGS for clinical diagnostic purposes to undiagnosed patients with a wide range of phenotypes. Patients were selected for enrollment by a multidisciplinary committee of representatives from divisions/departments caring for patients who might benefit from this testing. The committee established enrollment criteria that included both biological parents also being available for testing. Informed consent was obtained from both parents after meeting with a geneticist and genetic counselor, at which time the benefits, risk, and limitations of the study were discussed in detail. Institutional Review Board approval was not required because obtaining WGS on the patient was clinically indicated, and testing was performed in a Clinical Laboratory Improvement Amendment (CLIA)-certified laboratory.

WGS was performed by Illumina in a CLIA facility. Ingenuity Variant Analysis software (2012 beta release; QIAGEN) was used to assess WGS data. Filtering for variants with allele frequency < 0.01 and prediction of being deleterious (nonsense, frameshift, consensus splice mutation, or missense with in silico prediction of pathogenicity) yielded 709 variants. Further refinement by matching with phenotypic terms “hypoparathyroidism” and “ichthyosis” yielded a list of 31 variants. Comparison to parental data revealed the de novo status of a GATA3 variant and X-linked inheritance of the STS variant. After review by an American Board of Medical Genetics-certified clinical molecular geneticist, Sanger sequencing was performed on the GATA3 and STS variants in a CLIA-certified lab.

Case Report

The male infant was born at 37 weeks gestation to a 36-year-old G6P3–4 mother. The pregnancy was complicated by maternal nephrolithiasis and pregnancy-induced hypertension. The mother went into labor at 37 weeks, and the infant was delivered by repeat cesarean section. The birth weight was (∼15% for gestational age), and birth length was 18.25 in (∼10% for gestational age).

Soon after birth, the infant developed dry, scaly skin, and he was diagnosed with congenital ichthyosis by a pediatric dermatologist. He was treated effectively with Aquaphor (Beiersdorf Inc), although the condition persisted in a milder form in childhood. He developed problems with recurrent vomiting after discharge from the newborn nursery. Formula intolerance was suspected, and he was changed to a soy formula and prescribed ranitidine as an outpatient.

At 6 weeks of age, he was admitted to the hospital for failure to thrive and a normocytic anemia. He was found during this admission to have hypocalcemia (7.7 mg/dL [8–10.5]), hyperphosphatemia (9.7 mg/dL [3.5–6.6]), and an inappropriately low PTH level (9.5 pg/mL [10–65]). His hypoparathyroidism was initially treated with low phosphate formulas, but calcium supplements were added at 17 months, and calcitriol was added at 2 years of age. Due to the association of hypoparathyroidism with microdeletion of chromosome 22q11.2, multiplex ligation and probe amplification for this region was performed to assess for deletions/duplications, and the results were negative. The anemia was felt to be secondary to poor nutrition and exaggerated physiological nadir. The anemia resolved by 5 months of age.

The patient also failed a newborn hearing screen, and subsequent hearing evaluations were abnormal but were attributed to middle ear fluid. Sequential audiology testing ultimately diagnosed mild to moderate bilateral sensorineural hearing loss (mostly in the high frequency range). Hearing aids were prescribed at 4 years of age. He was followed by neurology for gross and fine motor delays that normalized by age 2 years. He had persistent expressive speech delay and mild cognitive delays. Magnetic resonance imaging of the head and spine were normal. He had transient hypertonia in infancy, which may have been secondary to hypocalcemia.

He was also noted to have short stature (1st percentile), whereas the midparental height was in the 75th percentile. His IGF-1, IGF binding protein-3, and thyroid function studies were normal for age. He demonstrated limited improvement of linear growth by 6 years of age, with height attainment to the 5th percentile.

The mother reported a history of kidney stones presenting with flank pain and hematuria during her pregnancy with the patient, but she also developed kidney stones with her previous pregnancies. Her symptoms resolved after delivery. The etiology of the maternal nephrolithiasis was never fully evaluated, and the mother's medical history was unavailable. The biological father reported no health problems. The parents were unrelated. There were three healthy older male siblings. The maternal grandfather had a history of very dry skin.

Additional diagnostic tests included a 46,XY G-banded karyotype and negative chromosomal microarray (Agilent 244k array performed as a clinical test). Clinical sequencing of the calcium sensing receptor gene (CASR) was negative. A renal ultrasound and cardiac echocardiogram were also normal. Electrolytes, blood urea nitrogen, and creatinine were all normal.

WGS followed by analysis using commercial software (Ingenuity Variant Analysis) revealed a de novo novel GATA3 c.192C>A;p. Tyr64Term nonsense mutation in exon 1. Sanger sequencing confirmed the variant (Supplemental Figure 1). A novel maternally inherited hemizygous c.1679A>G/p.Gln560Arg missense mutation in the STS gene was also found. Sanger sequencing confirmed the sequence variant (Supplemental Figure 2).

The GATA3 and STS sequence variants are novel by comparison to ClinVar, dbSNP, ExAC, and HGMD databases (accessed July 27, 2015). Mean coverage in ExAC was 60X for GATA3 and 58X for STS, albeit with lower coverage of 10–20X for exon 1 of GATA3 where the p.Tyr64Ter mutation is located. ExAC shows one individual with a synonymous change resulting in p.Tyr64Tyr, but no other changes reported at that position. The STS variant is also novel, but another source reported a mutation at the same position in a patient with ichthyosis, c.1679A>C, p.Gln560Pro (10).

Discussion

This clinically complex case illustrates the value of WGS or WES in rendering a definitive diagnosis in a patient with an underlying endocrine abnormality in the context of a complex multisystem disorder of unclear etiology. This case highlights some of the challenges in arriving at a clinical diagnosis and in selecting the appropriate clinical test.

Regarding the challenge of ascribing causality to a genetic variant, genetic disorders, even those caused by a single gene, often demonstrate variable symptoms among patients with mutations in the gene, even for the same exact mutation. In this case, our patient has two of the phenotypes related to the identified GATA3 mutation (hypoparathyroidism and deafness), but lacks another distinctive phenotype within the HDR spectrum (renal disease). In our patient, the symptoms of ichthyosis were mild, but in other people with the same mutation, symptoms might be more severe. Variability of symptoms is likely due to the effect of an individual's genetic background (genetic modifiers) and possibly to the environment and other variables. It is important to note that phenotypic variability has been reported in X-linked ichthyosis, even for patients with the same STS mutation. A recently published series of 22 cases with an STS deletion included phenotypic description of skin findings in 10 cases, none of which had exactly the same skin findings (10). This supports the notion that other genetic or environmental factors could modify the phenotype.

Also, interpreting genetic test results for disorders caused by polymorphisms (variants that are relatively common in the population) is more challenging than for variants in genes associated with single-gene syndromes. Variants in the former category have small effects that may or may not result in any symptoms, and the effects are only appreciated when looking at a population of individuals who carry such variants and comparing them to a control population without the variants (also known as a genetic association study). Variants in single-gene syndromes have large effect sizes, implying that the presence of the variant will essentially always result in symptoms. In general, conditions caused by common variants cannot be interpreted in the context of a WES test, and the test is ideally directed at conditions caused by single variants of large effects. In our case, the STS variant may not have a large effect, and therefore the symptoms are subtle.

Selecting the appropriate clinical genetic test for a patient with a complex phenotype can be a challenge for multiple reasons. First, an appropriate test may not be available. In this case, GATA3 sequencing was not available as a clinical test at the time of enrollment in our clinical sequencing pilot program.

Second, it may be difficult to recognize the full clinical picture because symptoms evolve over time, or there is variability in the symptoms in a particular individual. Our patient presented with isolated hypoparathyroidism early in infancy, and therefore the diagnosis of HDR was not readily apparent. Formal diagnosis of hearing loss was not made until 4 years of age. Also, our patient does not have renal disease, but approximately 20% of HDR cases without renal disease have been reported (5–8). In addition, the renal component of the HDR syndrome may not appear until later in life. Third, the genetic mechanism of disease (eg, sequence change vs genomic deletion) may not be assessed by each clinical test. In this case, the patient had been tested by chromosomal microarray that would have detected the cause of most cases of X-linked ichthyosis. Deletions or mutations in the STS gene cause X-linked ichthyosis in males. Over 90% of cases are due to deletions of the STS gene (detectable by microarray), although point mutations (not detectable by microarray) have been increasingly recognized (11).

Finally, the presence of two distinct Mendelian disorders created the appearance of a complicated phenotype that made it difficult to identify a unifying diagnosis to explain all of our patient's symptoms. Clinical WES has identified mutations in two distinct genes in 4.6% of patients in one large study, underscoring the unique role of WGS in evaluating complex medical disorders (2). The complicated phenotype not only made it difficult to make a clinical diagnosis, but also made it difficult to select a single test because genes with such disparate phenotypes would not be included on the same gene sequencing panel.

This case also allows us to discuss the relative value of WES or WGS testing. In current clinical practice, clinical laboratories cannot interpret most sequence variants outside the coding exons and immediate flanking intronic sequence. Therefore, WES is the current standard for clinical testing (as opposed to WGS). There are two important ways to assess the value of WES. First, the diagnostic yield is in the range of 25–30% and is about two to three times higher than for chromosomal microarray. Second, WES could offset the cost of other testing. Our patient had five genetic diagnostic tests before WGS, including: karyotype, fragile X testing, 22q11 deletion testing, chromosomal microarray analysis, and CASR gene sequencing. All five prior tests were negative. Based on our hospital's current rates, the cost of WES is only 9% more than the other five tests combined. Although this does not represent an absolute offset in costs, WES (equivalent to clinical WGS as performed on our patient) yielded two positive results, whereas all other testing had been negative, and the cost was comparable.

Conclusion

WGS identified two novel mutations in GATA3 and STS genes in a patient with complex multisystem disease, leading to the definitive genetic diagnosis of two different syndromes. The mutation in GATA3 caused hypoparathyroidism and hearing loss (HDR syndrome) and the STS mutation caused X-linked ichthyosis. These two genes would never occur on the same clinical sequencing panel, underscoring the ability of a WES or WGS approach to detect mutations for more than one Mendelian syndrome in patients with complex phenotypes.