The girl was born at a gestational age of 40 weeks and her birth weight was 3530 grams. She had feeding difficulties and at three months of age she was hospitalized after subacute developing of paleness and petechiae. A pancytopenia was found with critical low levels of thrombocytes as well as moderate anemia and lymphopenia. She recovered spontaneously, but the cause of the pancytopenia remained unclear. At the age of five it was obvious that she was both slower and weaker than other children of her age, thus a neurological investigation was initiated. She ran with clumsiness and had difficulties standing on heels and toes, and she had weakened tendon reflexes. Nerve conduction studies and electromyography suggested demyelinating neuropathy and hereditary polyneuropathy was suspected. At eight years of age the investigation was reactivated since nystagmus and lively tendon reflexes in patella had developed. Her cognitive function and general development were normal. MRI showed (figure 1) abnormal signaling in white matter and disseminated cysts in white and grey matter and a pronounced atrophy of the cerebellum. No medical reason was found that could explain these findings. For the following seven years she was followed at her local hospital, and was considered clinically stable. But at the age of 13 she developed an intra cerebral hemorrhage and an extensive investigation was therefore performed at the university hospital. A new MRI scan of the brain and the medulla were conducted. The brain was assessed as unchanged (except for the hemorrhage). However, the MRI of the medulla revealed long gone dural ectasias involving sella tursica resulting in a mechanical pressure on the hypophysis. EEG was normal and no signs of pathology were observed in the collected cerebrospinal fluid (CSF) samples except for slight increased levels of neurofilament light chain. In addition, generalized joint laxity was observed and this generated the hypothesis that she could suffer from a soft tissue disease or a mitochondrial disease. Muscle biopsy and genetic testing regarding Charcot-Marie-Tooth disease type 1A (CMT1A), Friedrichs ataxia (trinucleotide repeat in FXN), Loeys-Dietz syndrome (TGFBR1, TGFBR2) and Ehler Danlos syndrome type IV (COL3A1) were all normal as well as the karyotype of lymphocytes (46,XX). She did fully recover from the stroke associated symptoms, but she had problems walking without support (walked 10 meter without support and used wheelchair otherwise), a finding not attributed to her cerebral bleeding. Her gait disturbance was caused by the paresis of central origin in her legs. If it could be explained by the pathology in the brain or the spinal cord or both could not be determined. Although the cerebellum was seemed atrophied on MRI the only obvious clinical features from that area was nystagmus. No explanation was found for her widespread symptoms. During the remainder of her youth she continued with regular follow-ups at her local hospital. At 18 years of age she was referred to the department of neurology at the university hospital. No further focal neurological deficits were developed during the following years but she is considered mentally fatigued. At the age of 27 whole exome sequencing (WES) was performed.
Sequencing results: Patient was heterozygous for a de novo variant in SAMD9L; NM_152703.4:c.2686T>G, p.(Phe896Val), no signs of mosaicism in the WES or Sanger analyses.. The variant has previously not been reported to our knowledge. It is a missense mutation and generates an amino acid change, phenylalanine to valine, in a highly conserved region. In silico tools (SIFT, Mutation Taster, Align GVGD and PolyPhen2) predict the variant as deleterious. There was one more variant detected in SAMD9L, inherited from the patients healthy father, NM_152703.4:c.1565C>T, p.(Ala522Val) with an allele frequency of 1,7% in SweGen database and 1,4% in Exome Aggregation Consortium (ExAC) database, We could not exclude cis with c.2686T>G by Integrative Genomics Viewer (IGV).The variant was interpreted as not disease causing due to its allele frequency and inheritance from an healthy individual. Analysis of WES data of chromosome 7q showed no signs of acquired UPD(7q).
Whole exome sequencing and data filtering
Extraction of DNA was performed from whole blood using QIA Symphony DSP DNA Mini Kit (Qiagen, Hilden, Germany), according to the manufacturer's instructions and was subjected to WES. DNA library preparation and exome capturing was preformed using the Agilent Sureselect Clinical Research Exome v2. Sequencing was undertaken with paired-end 150 base pair reads using Illumina NextSeq500.
Sequence reads were aligned to the reference genome (hg19) using CLC biomedical workbench and variant calling was also performed with CLC biomedical workbench. Filtering of called variants was performed in several steps using Bench Lab NGS. We analyzed known disease-causing genes within the OMIM database. Further filtering was done to predict pathogenic variants (using SIFT, Mutation Taster, Align GVGD and PolyPhen2) and the level of conservation. How common the variants are in the population (using ExAC, Exome Aggregation Consortium project, 1,000 Genome; NHLBI Exome Sequencing Project (ESP)) were investigated to reduce the number of variants.
Confirmation of variants and analysis of parental samples was undertaken by Sanger sequencing according to standard procedures of the manufacturer (Thermo Fisher Scientific) Genomic DNA from the patient and the parents were analyzed.