In the news

Examine how reliably it was possible to predict fetal sex by non-invasive prenatal testing using cell-free fetal DNA, and identify factors that affect test performance. 

A systematic review of the published literature was carried out followed by meta-analysis of results from a total of 57 studies comprising a total of 3524 pregnancies with male fetuses and 3017 pregnancies with female fetuses. 

Overall, the ability of the test to detect the presence of Y-chromosome sequences (ie. a male fetus) had a sensitivity of 95.4% and a specificity of 98.6%, though figures varied considerably between studies. The most important factors in affecting test performance were the method used for the DNA processing (the use of PCR or RTQ-PCR techniques), and the gestational age of the fetus; earlier than 7 weeks test sensitivity was low, but rose to 94.8% (with specificity 98.9%) from 7-12 weeks.

The best performance for testing was achieved using maternal blood and RTQ-PCR after the 20^th week of pregnancy; testing using maternal urine or maternal blood before the 7^th week of pregnancy was found to be unreliable, calling into question some commercial claims of high accuracy at 5-7 weeks. 

The only other method of determining fetal sex is by ultrasound, which is unreliable before 11 weeks and not always successful after this point. Non-invasive testing therefore offers sex determination at least a month before current methods. This is very valuable for pregnancies at risk of a sex-linked genetic disease, so that fetuses not at risk by virtue of their sex can be spared the invasive testing currently required for diagnosis.

The authors note the need for a large clinical trial to reliably validate test performance, and call for research to develop methods for confirming the presence of cell-free fetal DNA, which would ensure that negative test results for Y-chromosome sequences genuinely show that the fetus is female, excluding the possibility that the cell-free fetal DNA was not detected.