Genetic diagnosis and clinical evaluation of severe fetal akinesia syndrome

Abstract Objective In this retrospective study, we describe the clinical course, ultrasound findings and genetic investigations of fetuses affected by fetal akinesia. Materials and Methods We enrolled 22 eukaryotic fetuses of 18 families, diagnosed with fetal akinesia between 2008 and 2016 at the Department of Obstetrics and Feto‐Maternal Medicine at the Medical University of Vienna. Routine genetic evaluation included karyotyping and chromosomal microarray analysis. Retrospectively, exome sequencing was performed in the index case of 11 families, if stored DNA was available. Confirmation analyses and genetic diagnosis of siblings were performed by using Sanger sequencing. Results Whole exome sequencing identified pathogenic variants of CNTN1, RYR1, NEB, GLDN, HRAS and TNNT3 in six cases of 11 families. In three of these families, the variants were confirmed in the respective sibling. Conclusions The present study demonstrates a high diagnostic yield of exome sequencing in fetuses affected by akinesia syndrome, especially if family history is positive. Still, in a large part the underlying genetic cause remained unknown, whereas precise clinical evaluation in combination with exome sequencing shows to be the best tool to find the disease causing variants.


| INTRODUCTION
The term fetal akinesia represents a category of disorders within a broad spectrum of diseases leading to reduced or absent fetal movements. This clinical finding is often recognized as a sequence of deformational changes related to reduced fetal movements called fetal akinesia deformations sequence (FADS). Many different alternative nomenclatures have been described in the literature including multiple congenital contractures (MCC), and most commonly arthrogryposis multiplex congenita (AMC).
For the definition of AMC or FADS at least two or more joint contractures in different body areas have to be present. 1,2 The phenotype of FADS includes features like intrauterine growth restriction (IUGR), craniofacial anomalies, limb contractures, pulmonary hypoplasia, short umbilical cord, together with pregnancy complications such as polyhydramnios and abnormal intrauterine positioning. This phenotype can occur isolated or associated with additional organ system anomalies. [3][4][5] Regarding prenatal diagnosis ultrasound findings of arthrogryposis, defined as multiple congenital contractures, club feet and clenched hands may be signs of decreased fetal movements. Moreover, nuchal edema, fetal hydrops and abnormal facial profile are observed in affected fetuses (FADS). 6 Later in pregnancy maternal perception of decreased intrauterine movements and developing polyhydramnios are important indications. 7 Previous investigations showed that prenatal MRI can visualize fetal akinesia and helps to identify associated findings, in particularly associated CNS abnormalities and lung hypolasia. 8 Concerning the etiology of this heterogeneous presenting phenotype a variety of genetic, non-genetic, parental and environmental causes are known. Environmental causes include circulating maternal antibodies to neurotransmitters, myelin and muscle proteins, restricted intrauterine space, drug use, maternal illness and ischemia. 9 Regarding genetic etiologies generally FADS can result from defects at any point along the motor system. 10 Although many well-defined genetic syndromes, are associated with decreased or absent fetal movements, the genetic diagnostic rate of these cases remains low. 3,6,9,10 This is due to the rarity and the genetic heterogeneity of these phenotypes and compounded by the difficulty of assigning unspecific prenatal clinical findings to a defined syndrome 6 . In the last years, through the availability of next generation sequencing methods, continuously genes associated with neuromuscular disease have been identified. 11 In this study, we comprehensively investigated 22 cases of prenatally diagnosed FADS and identified diverse genetic disorders responsible for this severe phenotype.

| METHODS
Ethical approval was obtained from the Ethics Committee of the Medical University of Vienna (1496/2015).   (Table S1). Additionally, variants in genes associated with congenital myopathy, muscle dystrophy and disease-causing genes, associated with any kind of muscle involvement were screened for pathological variants. Furthermore, (especially if negative) after the first two steps of filtering, all data were analyzed for known pathogenic variants according to ClinVar and HGMD Databases. All other variants were classified according to ACMG guidelines. 14 3 | RESULTS What does this study add? Abbreviations: CMA, chromosomal microarray analysis; EFW, expected fetal weight; GA, gestational age; IUFD, intrauterine fetal death; IUGR, intrauterine growth restriction; LGA, large for gestational age; SGA, small for gestational age; ToP, termination of pregnancy; WoP, weeks of pregnancy.

| Clinical findings
depending on the time of referral to our department. The mean gesta- For the disease causing variants in these fetuses detected by exome sequencing the mode of inheritance was autosomal dominant in two families (33%) and autosomal recessive in four families (67%).
In case 9 exome sequencing revealed two heterozygous variants (c.3986A>C/C.13328A>G) of NEB. Both missense variants revealed different results following two different prediction tools, so that these variants were classified as variants of unknown significance (VOUS).
Summarized, exome sequencing helped to establish a diagnosis in six of 11 families, respectively in nine of 15 cases, revealing a diagnostic yield of 55% (respectively 60%).

| DISCUSSION
In this retrospective study, we described the clinical course, ultrasound findings and genetic investigations of 22 affected fetuses.
Generally, structural malformation and ultrasound findings described in our cohort are comparable with previous published papers. 7,17 In all cases reduced or missing fetal movements and as a consequence arthrogryposis were recorded. Polyhydramnios, the most common associated finding in these cases, occurred more frequently in this cohort compared to others. This is probably explained by the rather small rate of termination of pregnancy in this cohort. Associated IUGR has been reported before, but in varying frequencies from 20% to 40%. 7,17 In our collective, only 9% of fetuses showed IUGR. However, we could also show an association of fetal akinesia and large for gestational age in 27% of fetuses, which may be explained by the high rate of hydropic fetuses (37%) in our cohort. In the remaining cases, the poor outcome with the high perinatal mortality, represents the severe phenotype and the precise clinical diagnosis performed by ultrasounds and fetal MRI.
Copy number variations (CNV) detected by Chromosomal microarray analyses in fetal akinesia seemed to be less promising compared to exome sequencing, according to our results. This is supported by the fact that the majority of literature in this field is focusing on molecular variants, rather than copy number variations. The small number of investigations using aCGH in the diagnostics of fetal akinesia syndrome focused on CNV in certain known disease associated genes like NEB. 19 The case of paternal UPD 14 (case 20), primary presented with polyhydramnios, decreased fetal movements and skin edema, so that fetal akinesia was suspected.  Concerning the application of exome sequencing in a prenatal setting the diagnostic yields ranges from 10% to 57% by investigating fetuses with different sonographic abnormalities. [26][27][28][29][30][31] In this wide range of diagnostic yield, the high diagnostic rate in our investigation is most likely explained by the indication and the previous probability of an underlying genetic cause within the cohort. In particular, the rate of family history, rate of consanguinity and the severe disease entity of the fetuses influenced the diagnostic yield.
In general, due to the heterogeneous clinical presentation, the variety of possible candidate genes, the use of WES is an efficient method for genetic diagnosis, within this indication. Because of the fast and constant identification of new associated genes, clearly demonstrated by Kiefer and Hall listing 320 genes associated with arthrogryposis 2016 32 and 402 in the gene ontology article of 2019, 33 WES is preferable to gene panels.
The major limitation of this study is the retrospective design, which does not allow to screen for environmental factors like for example maternal antibodies. Stored DNA was available more often in cases with positive family history. Thus, positive family history and consanguinity was overrepresented in this cohort, possibly leading to a higher diagnostic yield of WES in this study.
In conclusion, the present study demonstrates a high diagnostic yield of WES in FADS, especially in eukaryote cases with positive family history of FADS and if consanguinity is known. Still, in a large part the underlying genetic cause remained unknown, whereas precise clinical evaluation in combination with WES shows to be the most efficient tool to find the disease causing variants.