Knowledge of the genetic status of the fetus in an on-going pregnancy gives couples the power to make an informed decision about their unborn child. When a fetus is known to have a particular genetic abnormality, a decision may be made either to choose termination or to continue with the pregnancy and take steps to provide appropriate care for the newborn child. Prenatal testing falls into two categories: screening and diagnosis. Prenatal screening is offered to all pregnant women as part of routine prenatal care to determine if the fetus is at substantial risk of having a particular disorder such as Down Syndrome or sickle cell anaemia. In cases deemed to be at high risk, prenatal diagnosis is offered to provide a definitive diagnosis and determine whether the fetus has inherited a disorder.
Prenatal genetic diagnosis is often used where there is a family history of a sex-linked disease. Most sex-linked diseases are recessive X-linked diseases caused by a particular mutation on the X chromosome. The disease is normally manifested only in males, who carry a single X chromosome, whilst in females the normal allele on the second X chromosome compensates for the diseased allele. The most common X-linked recessive diseases include haemophilia (a blood clotting disorder) and Duchenne muscular dystrophy (a progressive muscle wasting disease), although numerous others can result in severe conditions. Whilst each disease is individually relatively rare, it has been estimated that in combination they occur in around 5 in 10,000 live births [1
In the UK, sex-linked diseases are usually diagnosed through referral to a clinical geneticist when there is a known family history of a particular disease. Although fetal sex can often be determined using an ultrasound scan of the fetus in the second or third trimester, a definitive prenatal diagnosis can only be made through invasive testing in which a sample of fetal cells is physically removed from the uterus for genetic analysis, using either chorionic villus sampling (CVS) at 11–14
weeks gestation, or amniocentesis from 15
weeks gestation. Both these invasive techniques carry a small but significant risk of miscarriage (1-2%) [2
] and although currently the gold standard for prenatal diagnosis, many women are reluctant to undergo invasive testing. However, there are substantial advantages to earlier diagnosis. Where future management might involve a decision to terminate the pregnancy, early termination carries fewer risks (being medically induced or involving surgical vacuum aspiration), whilst late termination (at more than 14
weeks) may require the induction of labour, potentially causing significantly greater physical, emotional and psychological complications.
Due to the risk of miscarriage with these traditional prenatal diagnostic methods, enormous interest has arisen in the field of non-invasive prenatal diagnosis (NIPD). In 1997, Lo et al.
] demonstrated the presence of fetal DNA in the maternal blood, opening the possibility that a simple blood test could provide a non-invasive method for prenatal diagnosis. Fragments of cell-free fetal DNA (cffDNA) originating from the placenta are detectable in the maternal blood stream [3
] from 5
weeks gestation until birth [4
], when they are rapidly cleared from the circulation and are undetectable within 2
]. It has been proposed that cffDNA could be used for non-invasive prenatal diagnosis. However, cffDNA only comprises around 3% to 6% [4
], although up to 10% has also been reported [6
], of the total cell-free DNA in the maternal circulation during pregnancy, the rest being maternal in origin. Therefore, distinguishing, or ideally isolating, fetally derived cell-free DNA in an overwhelming background of maternal DNA is a significant technical challenge due to the high level of molecular similarity between it and the maternally derived cell-free DNA. As a result, a number of different protocols have been developed to extract the cell-free DNA from a blood sample and analyze it for fetal specific sequences, usually with real time quantitative polymerase chain reaction (qPCR).
To date, the most advanced application of cffDNA for prenatal diagnosis is fetal sex determination for pregnancies at high risk of an X-linked disease (or certain masculinising endocrine disorders), in order to reduce the need for invasive testing. This is achieved through selective amplification and detection of Y chromosome sequences not otherwise present in the mother, most commonly the sex determining region Y (SRY), but sometimes using the testis specific protein Y linked 1 (also known as DYS14). The fetus is presumed to be female if no Y chromosome DNA can be detected. This technology is already being translated into a clinical setting and is used routinely in some clinics in the UK and elsewhere, and has been shown to reduce the need for invasive testing by 45% [7
]. In addition, several companies currently offer commercial mail-order fetal sexing using cffDNA from a home finger prick sample.
Since 1997, a number of large studies examining the accuracy of prenatal fetal sex determination using cffDNA have been published [8
] as well as many smaller ones (reviewed by Avent & Chitty [10
]). Devaney et al.
] have recently published a systematic review and meta-analysis documenting the overall test performance of non-invasive fetal sex determination using cffDNA including data from 57 studies and approximately 6,500 pregnancies. This review was limited however to English language publications and only searched for journals listed in PubMed. Here we report our review and meta-analysis of the wider literature in order to further evaluate the use of cffDNA in the maternal circulation for non-invasive prenatal determination of fetal sex.