PIEZO1 mutation: a rare aetiology for fetal ascites

Abstract

We present a case of isolated fetal ascites diagnosed at 20 weeks’ gestation. No aetiology was identified on extensive prenatal workup, including prenatal microarray. The patient terminated the pregnancy at 23 weeks’ gestation. Exome sequencing was performed on the products of conception, which ended up giving insight into a possible cause for the ascites. Two heterozygous missense variants of uncertain significance were identified in the PIEZO1 gene. The paternal variant has been linked to dehydrated hereditary stomatocytosis. The father of the baby suffers from haemolytic anaemia, splenomegaly and has had jaundice throughout his life. His brother and father have similar conditions. We suspect that at least one of the gene variants identified in our exome sequencing may be responsible for the illness that runs in this family, including the fetus with isolated ascites.

Background

Fetal ascites may be isolated or occur as part of generalised hydrops. It is diagnosed on ultrasound, when fluid is seen in the intra-abdominal cavity. The fetus must also be evaluated for pericardial effusion, pleural effusion, subcutaneous oedema, cystic hygroma, polyhydramnios and other malformations which could direct further workup.

Many cases of fetal ascites are described in the literature, with a large variation in severity and outcomes. One case series examined 79 cases of non-immune fetal ascites. On initial evaluation, 26% had hydrops fetalis, 42% had other anomalies and 32% were found to have isolated ascites. Out of the 25 cases which were found to be isolated, 68% were found after birth to have another anomaly, infection or metabolic disorder as the cause of ascites and 32% were confirmed to be truly idiopathic.¹ They reported a 48% fetal loss rate in cases of isolated fetal ascites and a neonatal mortality rate of 25% in cases of idiopathic ascites. Another retrospective chart review, which examined 12 cases of isolated fetal ascites over a 10-year period, reported a relatively good prognosis with an 80% survival rate.²

Abnormal lymphatic drainage and fetal anaemia resulting in decreased plasma oncotic pressure are two known mechanisms by which fetal ascites may develop. The PIEZO1 gene codes for an ion channel in the erythrocyte membrane that plays a role in erythrocyte function and the development of lymphatic structures. Mutations in PIEZO1 have been linked to disorders such as dehydrated hereditary stomatocytosis (DHS), generalised lymphatic dysplasia and non-immune hydrops of unknown aetiology.

Case presentation

A healthy 32-year-old primigravida at 20 weeks’ gestation was found to have fetal ascites and a 3 mm pericardial effusion on routine anatomic survey (figure 1). She previously had a normal first trimester ultrasound at 12 weeks and 4 days gestation. Paternal medical history includes a haemolytic disorder which was diagnosed at birth. He has chronic jaundice and has had splenomegaly since childhood. He underwent an extensive haematologic workup, but the aetiology of the haemolytic anaemia remains unclear. Paternal family history is significant for his father and brother having a similar disorder. His father was more severely affected and had a splenectomy. He also suffered from cirrhosis, heart failure and diabetes due to haemochromatosis, and passed away due to complications related to iron overload. His brother suffers from chronic jaundice. Maternal medical history includes dyslipidaemia and migraines.

Figure 1.

Ultrasound at 20 weeks 0 days demonstrating fetal ascites and pericardial effusion.

The fetus in this case was ruled out for many potential causes of non-immune hydrops, including congenital infection, haemoglobinopathy, cardiovascular abnormalities, fetal malformations and genetic or metabolic disorders. Prenatal microarray was negative. On follow-up imaging at 22 weeks, the pericardial effusion resolved, but the ascites persisted. The patient and father of the baby underwent extensive counselling regarding the unclear aetiology and prognosis. They elected to terminate the pregnancy at 23 weeks’ gestation. Whole exome sequencing was performed on products of conception, which ended up giving insight into a possible cause for the ascites.

Investigations

Maternal testing included type and screen, haemoglobin electrophoresis, thyroid stimulating hormone, G6PD screening, and testing for coxsackie A2, A4, A7, A9, A10 and B 1–6 virus types, herpes simplex virus, rubella, HIV and syphilis. The patient is Rh positive with a negative antibody screen. Amniocentesis revealed a male fetus with a normal microarray. PCR was done on amniotic fluid for toxoplasmosis, parvovirus and cytomegalovirus, all of which were negative.

The patient underwent termination of pregnancy via dilatation and evacuation. The products of conception and placenta were submitted for pathology. Placental pathology was unremarkable, with no gross lesions, infarcts or haemorrhage identified. The fetal evaluation was very limited due to the procedure. However, the upper and lower extremities exhibited moderate oedema.

Whole exome sequencing was performed using fetal tissue, and it revealed two heterozygous PIEZO1 gene variants which were reported as variants of unknown significance. Parental blood testing indicated that each parent transmitted one of those variants to the fetus. The variant PIEZO1: NM_001142864: exon42:c.C6008A: p.A2003D (chr16:88786633) was inherited from the father, and PIEZO1: NM_001142864:exon46:c.C6751T: p.P2251S (chr16:88783216) was inherited from the mother (figure 2).

Figure 2.

Exome sequencing results.

Outcome and follow-up

The working diagnosis at the time of termination was fetal ascites of unknown aetiology. Later on, the finding of two PIEZO1 gene variants of unknown significance raised the question of whether one or both variants may have been pathogenic, manifesting as isolated fetal ascites.

The fetus inherited the PIEZO1: NM_001142864: exon 42;c.C6008A: p.A2003D(chr 16:88786633) from the father. This missense variant has previously been linked to isolated DHS when in a heterozygous state.³ Multiple lines of computational evidence support a deleterious effect of this gene variant. The fetus inherited the PIEZO1: NM_001142864: exon 46;c.C6751T: p.P2251S(chr 16:88783216) from the mother. This variant is present at an allele frequency corresponding to the prevalence of DHS.

We suspect that the haemolytic anaemia and jaundice seen in the father of the baby are a manifestation of DHS related to his PIEZO1 gene variant. This diagnosis fits when considered in the context of the paternal family history. Further genetic testing in living relatives is needed to determine the exact cause of the father’s haemolytic anaemia. Recommendations were made for him to pursue workup and treatment at Yale University in New Haven, Connecticut, USA. Their institution serves as a reference laboratory for the diagnosis of rare anaemias.

Our patient conceived again few months later. The couple declined invasive prenatal testing in the first trimester for familial variants after being counselled on the limitations of testing for variants of unknown significance. They elected to have early sonographic evaluation in the second trimester and reconsider testing if there was any evidence of ascites or hydrops. Anatomic evaluation throughout the pregnancy was reassuring, with no evidence of ascites or hydrops, and the patient recently delivered a healthy male infant at term. The patient plans to have the baby tested for the two PIEZO1 variants.

Discussion

The PIEZO1 gene codes for a component of a mechanical ion channel which is found in the plasma membrane of red blood cells. It is also expressed in fetal lymphatic vasculature, liver and bone marrow.

A recent study explored the role of PIEZO1 mechanical ion channels in the regulation of hydration and volume of mature erythrocytes. The mechanosensitive ion channels translate mechanical signals into calcium ion influx. In mature erythrocytes obtained from patients with DHS, gain-of-function PIEZO1 mutations resulted in slower inactivation of the ion channel. This leads to excessive calcium influx and secondary activation of the Gardos channel in red cells, thereby causing potassium leakage, water loss and erythrocyte dehydration.⁴ This study also found that the PIEZO1 gene is expressed in erythroid progenitor cells, which had not been described in previous literature. Cells with the PIEZO-1 mutation exhibited a delay in erythroid differentiation. The authors suggest that this finding may contribute to the pathogenesis of the DHS, specifically as it pertains to the iron overload often seen in these patients.⁴

Mutations in PIEZO1 have been reported with autosomal dominant DHS, autosomal recessive generalised lymphatic dysplasia (a rare form of primary lymphedema) and non-immune hydrops of unknown aetiology.⁵ DHS is a rare autosomal dominant disorder in which the malfunctioning ion channel on erythrocyte cell membranes allows for leakage of cations, leading to dehydration of the red blood cells. Mutations in a variety of genes, including PIEZO1, have been implicated in the development of these diseases. DHS can be difficult to recognise due to a wide variation in clinical presentation, and patients are often not diagnosed until adulthood. Patients may have chronic anaemia, recurrent jaundice, gallstone disease, thrombosis after splenectomy or iron overload with hemosiderosis.^{6 7} Peripheral smears in affected patients show stomatocytes and spherocytes. The only biological test for DHS is ektacytometry, which is not widely available.⁸ Perinatally, the presentation of DHS has been described as a pleiotropic syndrome of perinatal oedema, anaemia and pseudohyperkalaemia due to leakage of potassium from cells.⁹,¹⁰ Affected fetuses may have non-immune hydrops fetalis (NIHF).¹¹ Perinatal and fetal ascites can be isolated or occur in concert with other findings of DHS. The pathophysiology of the ascites remains unclear, but generally resolves within the first few months of life. Liver disease associated with DHS has been suggested as an aetiology of the ascites in infants.⁷ As in generalised lymphatic dysplasia, the ascites may be caused by abnormalities in lymphatic vasculature.

Multiple lines of computational evidence support a deleterious effect of the paternal variant which the fetus in our case inherited.³ The father’s personal and family history suggest that he has a pathogenic mutation which is inherited in an autosomal dominant fashion. If the maternal variant is benign, the risk of recurrence would be 50%; however, the phenotype could vary widely. It is also possible that the maternal variant is pathogenic, and the two variants together may be responsible for the presentation of the fetus in our case. Testing of other family members may help to clarify the significance of these variants.

Whole exome sequencing sequences the protein coding regions of the genome. Prenatal phenotyping is an essential component for whole exome sequencing interpretation and significantly improves the diagnostic yield.¹² According to a recent study, in cohorts of all prenatal anomalies, whole exome sequencing has demonstrated a clear or likely diagnostic yield in 9%–47% of cases when microarray and/or karyotype are normal.¹²

There are currently no clinical guidelines regarding an investigation for mutations in PIEZO1 in patients with abnormal ultrasound findings, and routine genetic screening does not look for these mutations. Several recent case reports demonstrate the importance of exome sequencing when cytogenetics do not identify the cause of fetal hydrops.

A recent case report published in the European Journal of Medical Genetics describes a patient with a history of two consecutive pregnancies affected by NIHF of unknown aetiology. During her second pregnancy, exome sequencing was performed on amniocytes during the second trimester and revealed compound heterozygosity for two PIEZO1 variants. These variants were later found to be present in the first fetus and in the heterozygous parents.¹³ The first pregnancy ended in a fetal demise, and the second pregnancy was a live birth. The child was ultimately diagnosed with PIEZO1-related generalised lymphatic dysplasia. The couple was informed about the variants, their implication in the fetal phenotypes, and recurrence risks for future pregnancies.¹³ Another recent case report, which was published in the American Journal of Medical Genetics, describes a case of recurrent fetal hydrops. A patient with a history of IUFD at 36 and 32 weeks gestation due to hydrops fetalis presented at 16 weeks with NIHF. Prenatal whole exome sequencing revealed compound heterozygous mutations in PIEZO1 which were inherited from the mother and father, resulting in a form of severe generalised lymphatic dysplasia.¹⁴ Whole exome sequencing had not been performed until the patient’s third affected pregnancy. Sequencing also confirmed that the same two variants were present in her prior pregnancy which demised at 32 weeks’ gestation.¹⁴

A case report published in the Taiwanese Journal of Obstetrics and Gynecology details a case in which a heterozygous PIEZO1 mutation was identified in a patient and her two children. The three of them have haemolytic anaemia, splenomegaly, hepatomegaly and cholelithiasis.¹⁵ The workup started during her fourth pregnancy, when her fetus was found to have marked fetal ascites at 20 weeks’ gestation. There was no fetal anaemia and the karyotype was normal. The ascites was drained at 22 weeks, and the remainder of the pregnancy was uncomplicated. She delivered at term. The child appeared healthy at birth, but later experienced abdominal pain and jaundice. During her fifth pregnancy, she was found to have marked fetal ascites and large bilateral pleural effusions. Once again, there was no fetal anaemia and the karyotype was normal. She underwent multiple fetal treatments including the placement of peritoneoamniotic shunts, a thoracic–amniotic shunt, and therapeutic amniocentesis. With the shunts in place, they achieved complete resorption of the effusions. The child was born at 32 weeks after preterm premature rupture of membranes. The neonate had mild respiratory distress. The drains which were placed in utero were removed on the first day of life without complication or need for further interventions. Ektacytometry revealed similar findings in the mother and both children, which were consistent with DHS.¹⁵

Society of maternal fetal medicine guidelines for workup of NIHF, which were published in 2015, recommend obstetric ultrasound with dopplers, karyotype, chromosomal microarray, fetal echocardiography and viral PCR in the first line workup of NIHF. Since those guidelines were published in 2015, clinical genetic sequencing technologies have made rapid advancements. A recent systematic review put together a list of 131 genes associated with NIHF in order to promote the use of exome sequencing studies and to support the development of gene panels for NIHF.¹⁶ Exome sequencing can identify these single gene variants, which may be of great clinical value because it can prevent delays in initiating available treatment and can guide counselling on likely pregnancy outcomes and long term health and developmental outcomes for the child. Knowing a molecular diagnosis may also provide psychological benefit to the parents and allow for fetal genetic screening or preimplantation genetic diagnosis in future pregnancies.¹⁶

As we have demonstrated in our case, and the case reports described in recent literature, more research efforts should be directed towards elucidating the specific PIEZO1 gene variants responsible for disease phenotypes, and an effort made to pursue an understanding of how these diseases manifest during the antenatal period. This will help to appropriately counsel patients on their fetus’ diagnosis, prognosis and treatment options. It may also allow affected individuals to pursue treatment for haemolytic anaemia and avoid the long term detrimental effects of haemolytic anaemia on their health.

Learning points.

• In cases of fetal ascites of unknown aetiology, exome sequencing to identify rare causes is a necessary next step, especially if there is a family history of genetic disorders.

• Mutations in PIEZO1 have been linked to disorders including dehydrated hereditary stomatocytosis with or without pseudohyperkalaemia, generalised lymphatic dysplasia and non-immune hydrops of unknown aetiology.

• There are no current guidelines regarding the treatment of these conditions in the antenatal period; however, several case reports in the literature report favourable fetal outcomes.