Novel synonymous and missense variants in FGFR1 causing Hartsfield syndrome

Hartsfield syndrome is a rare clinical entity characterized by holoprosencephaly and ectrodactyly with the variable feature of cleft lip/palate. In addition to these symptoms patients with Hartsfield syndrome can show developmental delay of variable severity, isolated hypogonadotropic hypogonadism, central diabetes insipidus, vertebral anomalies, eye anomalies, and cardiac malformations. Pathogenic variants in FGFR1 have been described to cause phenotypically different FGFR1‐related disorders such as Hartsfield syndrome, hypogonadotropic hypogonadism with or without anosmia, Jackson–Weiss syndrome, osteoglophonic dysplasia, Pfeiffer syndrome, and trigonocephaly Type 1. Here, we report three patients with Hartsfield syndrome from two unrelated families. Exome sequencing revealed two siblings harboring a novel de novo heterozygous synonymous variant c.1029G>A, p.Ala343Ala causing a cryptic splice donor site in exon 8 of FGFR1 likely due to gonadal mosaicism in one parent. The third case was a sporadic patient with a novel de novo heterozygous missense variant c.1868A>G, p.(Asp623Gly).


| PATIENTS' MEDICAL REPORTS
We describe two siblings born to healthy, non-consanguineous Swiss parents. Their healthy sister has no reported medical issues and the family history is otherwise unremarkable. Patient 1 (II:1) was born at term after an uneventful pregnancy with a birth weight of 3,195 g (−1.3 SD). He was referred to the genetic department at the age of 18 months with a craniosynostosis of the sagittal suture, ectro-/poly-/ syndactyly of both feet, syndactyly of both hands (Figure 1a), cryptorchidism, a single maxillary incisor, as well as global developmental delay.
His length at the time of examination was 79 cm (− 1.5 SD), his weight 9.6 kg (−1.4 SD), and his head circumference 44 cm (−3.4 SD). During the following years he developed an ocular apraxia and ataxia, mild intellectual disability and was diagnosed with a hypogonadotropic hypogonadism. A cranial MRI at the age of 7 years showed a lobar holoprosencephaly. An echocardiogram and abdominal ultrasound were normal. His younger sister, Patient 2 (II:2), was born at term after an uneventful pregnancy with a birth weight of 2,760 g (−1.6 SD). At the age of 3 weeks, she was referred to the genetic department with poly-/ syndactyly of the right foot ( Figure 1b)  Genomics (Richards et al., 2015), the FGFR1 variant c.1029G>A, p.

| Patient 3
Initial genetic testing revealed a normal karyotype of 46XY as well as normal chromosomal microarray analysis (1.4M Roche NimbleGen oligonucleotide CGH array). Subsequent Sanger sequencing revealed the heterozygous FGFR1 missense variant c.1868A>G, p.(Asp623Gly), which was subsequently confirmed to be de novo with parental segregation analysis by Sanger sequencing (Figure 2c). The variant was predicted to be damaging to the protein function by the common bioinformatic algorithms used for pathogenicity assessment (Ioannidis et al., 2016). Additionally, the variant was not annotated in ExAC, gnomAD, EVS, or dbSNP databases, as well as 200 ethnically matched control chromosomes. However, a substitution within the same amino acid position leading to a aspartic acid to tyrosine p. (Asp623Tyr) instead of glycine has been described in another case of Hartsfield syndrome (Simonis et al., 2013).
According to the recommendations of the American College of Medical Genetics and Genomics (Richards et al., 2015), the FGFR1

| DISCUSSION
Here we report a novel de novo synonymous pathogenic variant in FGFR1, which was found in two siblings (Patients 1 and 2), and a novel  Loss-of-function variants in FGFR1 are associated with a wide phenotypic spectrum and identical variants may present with a variable phenotype attributed to incomplete penetrance and variable expressivity (Simonis et al., 2013). Interestingly, Patient 3 p.-(Asp623Gly) showed a more severe phenotype than a previously reported case p.(Asp623Tyr; Simonis et al., 2013), suggesting differing pathogenic mechanism dependent on the substituted amino acid at that position.
The WES data from DNA extracted from leukocytes of both parents of Patients 1 and 2 showed a normal sequence at this position with a coverage of approximately 300-fold, each. Thus, we assume gonadal mosaicism of this variant in one of the parents or alternatively somatic mosaicism including germ cells. Gonadal mosaicism has previously been reported in a case affected by Hartsfield syndrome (Dhamija et al., 2014;Shi et al., 2016). Herein, we report a second family with gonadal mosaicism in Hartsfield syndrome and emphasize that such genetic constellations should be taken into consideration when counseling clinically unaffected parents regarding their potential recurrence risk.