The identification of novel biomarkers for the small for gestational age human fetus

Babies who grow poorly in the womb (called small for gestational age [SGA]) are at increased risk of a number of adverse outcomes, including stillbirth. Screening pregnant women in an attempt to detect SGA babies has the potential to reduce the number of these adverse outcomes, in particular stillbirth, which accounts for the death of about 4000 babies per annum in the UK. Current practice in the UK in screening the general population for SGA babies is confined to measuring the woman?s bump (technically called ?symphysis-fundal height?) with a tape measure. However, this approach is known not to be particularly good at picking up small babies. The approach to SGA is very crude in comparison to the approach to screening for Down?s syndrome babies in pregnancy. Down?s syndrome screening utilises a combination of scan and blood markers and now detects 90% of cases with only a small proportion of women requiring further tests. There are also a number of scan and blood markers for SGA babies, but none of these is good enough on its own to predict them. We believe that it will be possible to identify novel markers which may be more effective than existing blood tests. In fact, relatively few studies have systematically looked for better markers and most of the existing blood tests which could help predict SGA were discovered by chance. We have access to scan information, blood samples and placental (afterbirth) samples from a large scale study of unselected women in their first pregnancy ( 2800 women and on target to recruit 5000). The overarching aim of the present application is to see if we can find better markers in the mother?s blood of poorly grown babies. Specifically, we plan to determine the key differences in the genes expressed in the placenta of SGA babies and matched controls. We then aim to see if the proteins encoded by these genes are found at different concentrations in the placenta and in the mother?s blood. Finally, we aim to determine whether information about the levels of these proteins improves the prediction of SGA babies. The end result will be a predictive tool (algorithm) to detect small babies. We could then design trials of the tool to see if it improved outcome. The ultimate aim is to make care of pregnant women more effective and to reduce the number of babies lost to stillbirth.

Small for gestational age infants (SGA) are at increased risk of a number of adverse outcomes. Screening for SGA has the potential to reduce the number of adverse outcomes, in particular population rates of stillbirth, which account for the death of about 4000 babies per annum in the UK. The 2008 NICE Guideline on Antenatal Care identified improving detection of SGA infants as one of 5 major research priorities. The current exemplar of pregnancy screening is Down?s syndrome, which utilises a combination of ultrasonic and biochemical markers and now detects 90% of cases for a less than 5% false positive rate. We have previously reported associations between a range of ultrasonic and biochemical measurements and the risk of delivery of a SGA infant, but none of these has the discrimination to provide clinically useful prediction. We have hypothesised that this may be achieved by combined ultrasonic and biochemical assessment of the fetus and placenta. The current project will use the Pregnancy Outcome Prediction study, a prospective cohort study of unselected nulliparous women. The study combines serial ultrasonic fetal biometry and utero-placental Doppler and maternal blood sampling at 12, 20, 28 and 36 weeks; and systematic sampling of the placenta following delivery. The study has already recruited 2800 women, has placental samples stored from 1900 women and continues to recruit approximately 100 women per month. The design of the study allows optimal phenotyping of cases of SGA (using both Doppler and growth velocity) and comparison with controls matched in relation to key maternal and obstetric characteristics. Preliminary analysis of the ultrasonic data indicated that an algorithm based on scan data alone only picked up about half of SGA infants for a 10% screen positive rate. The aims of the present study are (1) to compare expression gene array of placental samples from infants with different phenotypes of SGA at term and to compare them with matched controls, (2) to validate variation in placental levels of candidate biomarkers using qRT-PCR, Western blot and immunohistochemistry in a separate series of cases and matched controls, (3) to use a case cohort approach to compare maternal serum levels of candidate biomarkers and assess their screening performance in combination with serial ultrasound. The ultimate aim is to generate an algorithm for the identification of pathologically SGA infants that might ultimately be tested in a randomised controlled trial of screening the general population.