Pre-eclampsia screening in Denmark (PRESIDE): national validation study

OBJECTIVES

To investigate the predictive performance of the Fetal Medicine Foundation (FMF) first-trimester screening algorithm for pre-eclampsia in a Danish population and compare screening performance with that of the current Danish strategy, which is based on maternal risk factors.

METHODS

This was a prospective study of women with a singleton pregnancy attending for their first-trimester ultrasound scan and screening for aneuploidies at six Danish university hospitals between May 2019 and December 2020. Prenatal data on maternal characteristics and medical history were recorded, and measurements of mean arterial pressure (MAP), uterine artery pulsatility index (UtA-PI), serum pregnancy-associated plasma protein-A (PAPP-A) and serum placental growth factor (PlGF) were collected without performing a risk assessment for pre-eclampsia. Information on acetylsalicylic acid use was recorded. After delivery, pregnancy outcome, including gestational age at delivery and pre-eclampsia diagnosis, was recorded. Pre-eclampsia risk assessment for each woman was calculated blinded to outcome using the FMF screening algorithm following adjustment to the Danish population. Detection rates (DRs) of the FMF algorithm were calculated for a fixed screen-positive rate (SPR) of 10% and for the SPR achieved in the current Danish screening.

RESULTS

A total of 8783 pregnant women were included, with a median age of 30.8 (interquartile range (IQR), 28.1-33.9) years. The majority were white (95%), naturally conceiving (90%), non-smokers (97%) and had no family history of pre-eclampsia (96%). The median body mass index was 23.4 (IQR, 21.2-26.6) kg/m2 . A complete risk assessment including maternal characteristics, MAP, UtA-PI, PlGF and PAPP-A was available for 8156 women (92.9%). In these women, UtA-PI was measured bilaterally with a median value of 1.58 (IQR, 1.27-1.94) and the median resting MAP of 80.5 (IQR, 76.1-85.4) mmHg in two consecutive measurements. Among these, 303 (3.7%) developed pre-eclampsia, including 55 (0.7%) cases of pre-eclampsia with delivery < 37 weeks of gestation and 16 (0.2%) cases of pre-eclampsia with delivery < 34 weeks. At a SPR of 10%, combined screening using the FMF algorithm based on maternal characteristics, MAP, UtA-PI, PlGF and PAPP-A had a DR of 77.4% (95% CI, 57.6-97.2%) for pre-eclampsia with delivery < 34 weeks, 66.8% (95% CI, 54.4-79.1%) for pre-eclampsia with delivery < 37 weeks and 44.1% (95% CI, 38.5-49.7%) for pre-eclampsia with delivery at any gestational age. The current Danish screening strategy using maternal risk factors detected 25.0% of women with pre-eclampsia with delivery < 34 weeks and 19.6% of women with pre-eclampsia with delivery < 37 weeks at a SPR of 3.4%. When applying the FMF algorithm including maternal characteristics, MAP, UtA-PI and PlGF at the fixed SPR of 3.4%, the DRs were 60.5% (95% CI, 36.9-84.1%) for PE with delivery < 34 weeks and 45.2% (95% CI, 32.0-58.5%) for PE with delivery < 37 weeks.

CONCLUSION

In this large Danish multicenter study, the FMF algorithm based on maternal characteristics, MAP, UtA-PI, PlGF and PAPP-A predicted 77.4% of cases with pre-eclampsia with delivery < 34 weeks and 66.8% of cases with pre-eclampsia with delivery < 37 weeks of gestation at a SPR of 10%, suggesting that the performance of the algorithm in a Danish cohort matches that in other populations. © 2023 The Authors. Ultrasound in Obstetrics & Gynecology published by John Wiley & Sons Ltd on behalf of International Society of Ultrasound in Obstetrics and Gynecology.

National data on the early clinical use of non-invasive prenatal testing in public and private healthcare in Denmark 2013-2017

INTRODUCTION

In Denmark, non-invasive prenatal testing (NIPT) has been used since 2013. We aimed to evaluate the early clinical use of NIPT in Danish public and private healthcare settings before NIPT became an integrated part of the national guidelines on prenatal screening and diagnosis in 2017.

MATERIAL AND METHODS

NIPT data were collected between March 2013 and June 2017 from national public registries and private providers. Results from follow-up samples (chorionic villi, amniotic fluid, postnatal blood or fetal tissue) were included from The Danish Cytogenetics Central Registry and indications and outcome from The Danish Fetal Medicine Database.

RESULTS

A total of 3936 NIPT results were included in the study from public hospitals (n = 3463, 88.0%) and private clinics (n = 473, 12.0%). The total number of prenatal tests was 19 713 during the study period: 20% were NIPT analyses (n = 3936) and 80% invasive procedures (n = 15 777). Twenty-five percent of NIPTs in the private clinics were performed before gestational week 11⁺⁰ , whereas NIPT in public settings was used only after combined first trimester screening (P < .001). Regardless of indication, the national public sensitivity was 96.9% (95% CI 82.0%-99.8%) for trisomy 21, 100% (95% CI 46.3%-100%) for trisomy 18, 100% (95% CI 5.5%-100%) for trisomy 13, and 87.0% (95% CI 74.5%-92.4%) for any fetal chromosomal aberration. Forty-seven true-positive NIPT results included cases of common aneuplodies (trisomy 21, n = 31; trisomy 18, n = 5; and trisomy 13, n = 1), sex chromosomal aberrations (n = 7) and atypical chromosomal aberrations (n = 3). One false-negative NIPT result occurred (trisomy 21). Of 47 cases, 21 (45%) cases with a true-positive NIPT result resulted in live births by choice; 11 of these children had Down and 4 had Edwards syndrome.

CONCLUSIONS

The total number of NIPT analyses was low compared with the number of invasive procedures in the implementation period. In contrast to the generally high termination rate after a positive result following invasive testing in Denmark, a high proportion of true-positive NIPT results from the public setting resulted in live births. NIPT may be an important risk-free alternative to invasive testing for a minority of women in the public setting who wish to use prenatal genetic testing for information only and not for reproductive decision-making.

Non-Invasive Prenatal Testing (NIPT) in pregnancies with trisomy 21, 18 and 13 performed in a public setting - factors of importance for correct interpretation of results

OBJECTIVES

We have established an open source platform for non-invasive prenatal testing (NIPT) based on massively parallel whole-genome sequencing in a public setting. The objective of this study was to investigate factors of importance for correct interpretation of NIPT results to ensure a high sensitivity and specificity.

STUDY DESIGN

This investigation is a retrospective case-control study performed in a public NIPT center. The study included 108 aneuploid cases and 165 euploid controls. MPS was performed on circulating cell-free DNA in maternal blood. The pipeline included automated library preparation and sequencing on a HiSeq1500 (Illumina). The software programmes WISECONDOR and SeqFF were used for data analysis of aneuploidy status and fetal fraction of cell-free DNA, respectively. Lower limit of fetal fraction for aneuploidy testing was 0.02.

RESULTS

We identified four false negative aneuploidy cases of which two were explained by a vanishing twin. The number of no-call cases due to low fetal fraction was 8 out of 273 (2.9%). The sensitivity and specificity, when no-calls and vanished twins were excluded, were 100% and 99.5% for T21, 91% and 99.2% for T18, and 100% and 99.6% for T13. By multiple regression analysis we found a significant association between fetal fraction and gestational age, maternal BMI and ART treatment.

CONCLUSION

With a non-commercial open source NIPT set-up having the same high test-performance as reported by large private laboratories, we show that fetal fraction, a vanishing twin, BMI, gestational age and ART treatment are important factors in the interpretation of NIPT results.

Maternal exposure to UV filters: associations with maternal thyroid hormones, IGF-I/IGFBP3 and birth outcomes

BACKGROUND

Several chemical UV filters/absorbers ('UV filters' hereafter) have endocrine-disrupting properties in vitro and in vivo. Exposure to these chemicals, especially during prenatal development, is of concern.

OBJECTIVES

To examine maternal exposure to UV filters, associations with maternal thyroid hormone, with growth factor concentrations as well as to birth outcomes.

METHODS

Prospective study of 183 pregnant women with 2nd trimester serum and urine samples available. Maternal concentrations of the chemical UV filters benzophenone-1 (BP-1) and benzophenone-3 (BP-3) in urine and 4-hydroxy-benzophenone (4-HBP) in serum were measured by liquid chromatography-tandem mass spectrometry (LC-MS/MS). The relationships between 2nd trimester maternal concentrations of the three chemical UV filters and maternal serum concentrations of thyroid hormones and growth factors, as well as birth outcomes (weight, height, and head and abdominal circumferences) were examined.

RESULTS

Positive associations between maternal serum concentrations of 4-HBP and triiodothyronine (T₃), thyroxine (T₄), insulin-like growth factor I (IGF-I) and its binding protein IGFBP3 were observed in mothers carrying male fetuses. Male infants of mothers in the middle 4-HBP exposure group had statistically significantly lower weight and shorter head and abdominal circumferences at birth compared to the low exposure group.

CONCLUSIONS

Widespread exposure of pregnant women to chemical UV filters and the possible impact on maternal thyroid hormones and growth factors, and on fetal growth, calls for further studies on possible long-term consequences of the exposure to UV filters on fetal development and children's health.

Presence of benzophenones commonly used as UV filters and absorbers in paired maternal and fetal samples

BACKGROUND

Previous studies have demonstrated widespread exposure of humans to certain benzophenones commonly used as UV filters or UV absorbers; some of which have been demonstrated to have endocrine disrupting abilities.

OBJECTIVES

To examine whether benzophenones present in pregnant women pass through the placental barrier to amniotic fluid and further to the fetal blood circulation.

METHODS

A prospective study of 200 pregnant women with simultaneously collected paired samples of amniotic fluid and maternal serum and urine. In addition, unique samples of human fetal blood (n=4) obtained during cordocentesis: and cord blood (n=23) obtained at delivery, both with paired maternal samples of serum and urine collected simultaneously, were used. All biological samples were analyzed by TurboFlow-liquid chromatography - tandem mass spectrometry for seven different benzophenones.

RESULTS

Benzophenone-1 (BP-1), benzophenone-3 (BP-3), 4-methyl-benzophenone (4-MBP), and 4-hydroxy-benzophenone (4-HBP) were all detectable in amniotic fluid and cord blood samples and except 4-HBP also in fetal blood; albeit at a low frequency. BP-1 and BP-3 were measured at ~10-times lower concentrations in fetal and cord blood compared to maternal serum and 1000-times lower concentration compared to maternal urine levels. Therefore BP-1 and BP-3 were only detectable in the fetal circulation in cases of high maternal exposure indicating some protection by the placental barrier. 4-MBP seems to pass into fetal and cord blood more freely with a median 1:3 ratio between cord blood and maternal serum levels. Only for BP-3, which the women seemed to be most exposed to, did the measured concentrations in maternal urine and serum correlate to concentrations measured in amniotic fluid. Thus, for BP-3, but not for the other tested benzophenones, maternal urinary levels seem to be a valid proxy for fetal exposure.

CONCLUSIONS

Detectable levels of several of the investigated benzophenones in human amniotic fluid as well as in fetal and cord blood calls for further investigations of the toxicokinetic and potential endocrine disrupting properties of these compounds in order for better assessment of the risk to the developing fetus.

Effects of forage neutral detergent fibre and time after feeding on medial and ventral rumen pH and volatile fatty acids concentration in heifers fed highly digestible grass/clover silages

The major microbial fermentation of forages and production of volatile fatty acids (VFA) takes place in the medial part of the rumen, whereas the absorption of VFA occurs through the rumen epithelium, for example the ventral sac. The objective was to study effects of forage neutral detergent fibre (NDF) content and time after feeding on the medial to ventral VFA and pH gradient as well as rumen motility in the rumen of heifers fed grass/clover silages. Four silages were harvested at different growth stages with NDF contents of 31–45% of DM and in vitro organic matter digestibilities of 75–82% and fed to four rumen-fistulated Jersey heifers at 90% of ad libitum level in a Latin square design, with half the ration fed at 0800 hours and 1530 hours. Rumen fluid was sampled hourly from 0730 hours to 1530 hours in the medial and ventral rumen, and analysed for pH and concentrations of VFA, L-lactic acid, and ammonia to assess ruminal chemical gradient. Reticular contractions were continuously recorded by a pressure transducer. Time relative to feeding affected rumen parameters as pH was generally lower and VFA content greater in medial compared with ventral rumen fluid. Greater NDF content of the silage caused lower VFA concentration and higher pH in the rumen mat, and therefore the gradient diminished at greater NDF content in the silages; an effect probably caused by reduced organic matter digestibility rather than digesta NDF properties. This study therefore suggests that VFA production decreased with greater NDF content of forages, whereas intra-ruminal equilibration increased.

The outcome of antenatal ultrasound diagnosed anomalies of the kidney and urinary tract in a large Danish birth cohort

OBJECTIVE

Antenatal ultrasound diagnosed anomalies of the kidney and urinary tract (AUDAKUT) are reported in 0.3%-5% on prenatal ultrasound (US) and 0.3%-4.5% on postnatal US. The anterior-posterior diameter of the renal pelvis (APD) is an essential measurement. Series with low threshold values of APD prenatally and postnatally will include healthy infants. It is important to avoid follow-up of such infants.

INTERVENTIONS

In 2006, new Danish guidelines for AUDAKUT were introduced.

AIM OF STUDY

Investigations of incidences and type of AUDAKUT based on Danish guidelines, including long-term follow-up.

DESIGN

Cohort study.

SETTING

Copenhagen University Hospital Hvidovre and Copenhagen University Hospital Rigshospitalet, Denmark.

PATIENTS

Consecutive cases with AUDAKUT in the second and third trimesters, which were either terminated before 22 completed weeks of gestation or born in the 8-year period January 2006-December 2013. Patients were followed until June 2014.

RESULTS

50 193 live born children and 24 terminated fetuses (0.05%) were included. The prevalence of AUDAKUT was only 0.39% prenatally, 0.29% at first postnatal US and 0.22% at the end of follow-up, including terminated cases. The greater the prenatal and postnatal APD, the higher risk of febrile urinary tract infection (fUTI) and surgical intervention, and lower probability of resolution. 25% of the identified patients had fUTI and/or surgery.

CONCLUSIONS

We recommend threshold values of APD at least 10 mm in the third trimester and in general at least 12 mm at first postnatal US for intensive follow-up. In this largest to date unselected birth cohort of AUDAKUT, the incidences of clinically significant AUDAKUT were in the lowest range of those previously published.

Autocorrelation and cross-correlation between hCGβ and PAPP-A in repeated sampling during first trimester of pregnancy

BACKGROUND

Theoretically, repeated sampling of free β-human chorionic gonadotropin (hCGβ) and pregnancy associated plasma protein-A (PAPP-A) in the first trimester of pregnancy might improve performance of risk assessment of trisomy 21 (T21). To assess the performance of a screening test involving repeated measures of biochemical markers, correlations between markers must be estimated. The aims of this study were to calculate the autocorrelation and cross-correlation between hCGβ and PAPP-A in the first trimester of pregnancy and to investigate the possible impact of gestational age at the first sample and time between sampling on the correlation.

METHODS

A prospective study was conducted including 3891 unaffected singleton pregnancies. Two measurements of hCGβ and PAPP-A were obtained during the first trimester in each pregnancy. Correlations between the four parameters, hCGβ first, hCGβ second, PAPP-A first and PAPP-A second, were estimated and presented in terms of Pearson's r coefficients. Furthermore, the correlation between paired samples as a function of time between samples was investigated.

RESULTS

The study demonstrated high correlation between first and second samples of hCGβ and PAPP-A with a correlation coefficient of 0.80 and 0.79, respectively. By contrast, the correlations between hCGβ and PAPP-A were low. In addition, the study demonstrated that the correlation between paired samples of hCGβ and PAPP-A decreases with earlier gestational age at the first sample and with increasing time between samples.

CONCLUSIONS

We have developed a parameter set in terms of correlations between biochemical markers, which can be incorporated into a T21 screening algorithm based on repeated measures within the first trimester.

Recording rumination time by a rumination monitoring system in Jersey heifers fed grass/clover silage and hay at three feeding levels

This study was conducted with the objective to validate a rumination monitoring system (RMS). The RMS was developed by SCR Engineers Ltd. (Netanya, Israel) to record individual rumination times in commercial dairy herds. Four heifers were fed 4 forage types at 3 feeding levels in 4 periods in a Latin square design. The forage types were spring growth of grass/clover harvested early (May 9) and late (May 25) conserved as silage and hay, creating 4 forage types: early harvest silage (ES), early harvest hay (EH), late harvest silage (LS), and late harvest hay (LH). Feeding levels were ad libitum, 90% of ad libitum, and 60% of ad libitum. Data were collected for 24 h at ad libitum feeding level, for 96 h at 90% of ad libitum, and for 24 h at 60% of ad libitum. Rumination time was recorded by the RMS in minutes per 2-h interval (RT(RMS2_i)) and per 24-h interval (RT(RMS24)). As a reference method, rumination time was recorded by chewing halters identifying rumination time by jaw movements (JM) and was accumulated into minutes per 2-h interval (RT(JM2_i)) and per 24-h interval (RT(JM24)) over the same time intervals as RT(RMS2_i) and RT(RMS24). Differences between RT(RMS2_i) and RT(JM2_i) were observed for EH, LS, and LH, with differences of 3.4 (P < 0.001), 1.1 (P = 0.03), and 3.3 (P < 0.001) min per 2-h interval. In addition, feeding restrictively at 90% and 60% of ad libitum resulted in greater rumination time recorded by the RMS compared to JM, with differences of 2.6 min per 2-h interval (P < 0.001) for 90% of ad libitum and 1.6 min per 2-h interval (P = 0.01) for 60% of ad libitum. Further, the differences between rumination times recorded by the RMS compared to JM were different among the 4 heifers. Despite these effects, rumination time recorded by the RMS correlated well with rumination time recorded by JM with correlation between RT(RMS2_i) and RT(JM2_i) of r = 0.91 (P < 0.001) and between RT(RMS24) and RT(JM24) of r = 0.79 (P < 0.001).

Temporal effects of maternal and pregnancy characteristics on serum pregnancy-associated plasma protein-A and free β-human chorionic gonadotropin at 7-14 weeks' gestation

OBJECTIVE

The aim of this study was to investigate gestational age-dependent effects of racial origin, smoking status and mode of conception on maternal serum levels of free β-human chorionic gonadotropin (β-hCG) and pregnancy-associated plasma protein-A (PAPP-A) at 7-14 weeks' gestation.

METHODS

This was an analysis of data from prospective first-trimester combined screening for aneuploidies in singleton pregnancies, with β-hCG and PAPP-A measured at 7 + 1 to 14 + 3 weeks' gestation. We included 27,908 pregnancies from three centers in the U.K. and 125,461 pregnancies from 22 centers in Denmark, all with known normal fetal karyotype or resulting in the birth of a phenotypically normal neonate. Multiple regression modelling of log10 -transformed marker concentrations was used to produce log10 multiple of the median (MoM) values for free β-hCG and PAPP-A and to examine pregnancy characteristics that have significant effects on marker concentrations.

RESULTS

Serum free β-hCG and PAPP-A concentrations were significantly affected by gestational age, maternal weight, racial origin, parity, smoking and mode of conception. There were significant gestational age-dependent effects attributed to Afro-Caribbean race, smoking and conception through in-vitro fertilization (IVF) on PAPP-A and free β-hCG levels. In women of Afro-Caribbean race there was a weekly increase in PAPP-A of 5.3% and in free β-hCG of 1.8%. In smokers there was a weekly decrease in PAPP-A of 2.4% and in free β-hCG of 1.6%. In cases of IVF conceptions there was a weekly increase in PAPP-A of 4.5% and in free β-hCG of 4.6%.

CONCLUSIONS

Serum free β-hCG and PAPP-A concentrations at 7-14 weeks' gestation are affected by several pregnancy characteristics. The effects of Afro-Caribbean race, smoking and IVF conception change with gestational age.

Prospective study evaluating performance of first-trimester combined screening for trisomy 21 using repeat sampling of maternal serum markers PAPP-A and free β-hCG

OBJECTIVE

To prospectively evaluate the performance of first-trimester combined screening for trisomy 21 using the biochemical markers pregnancy-associated plasma protein-A (PAPP-A) and free beta-human chorionic gonadotropin (free β-hCG) obtained before and at the time of the nuchal translucency (NT) scan.

METHODS

Three fetal medicine departments in Denmark participated in the study. Screening for trisomy 21 was set up as a two-step approach with blood sampling performed before the NT scan (early sample) and again at the time of the NT scan (late sample). PAPP-A and free β-hCG were measured on both the early and late samples. Age-standardized detection and false-positive rates for different screening protocols were calculated.

RESULTS

We collected two blood samples in 27 pregnancies affected by trisomy 21 and in 3891 control pregnancies. The early samples were taken between gestational ages 8 + 0 and 13 + 6 weeks, and the late samples between 11 + 3 and 14 + 6 weeks. The median interval between the samples was 17 (range, 1-40) days. We found a significantly better estimated screening performance when using early sampling vs late sampling (P < 0.05). With a risk cut-off of 1 in 100, at the time of the risk assessment the estimated detection and false-positive rates when using the early sample were 91% (95% CI, 81-98%) and 1.6% (95% CI, 1.3-2.0%), respectively. For fixed false-positive rates the highest detection rates were achieved using both blood samples. When comparing early sampling vs double sampling there was no significant difference in screening performance.

CONCLUSION

In combined first-trimester screening for trisomy 21, use of early sampling with measurement of PAPP-A and free β-hCG before the time of the NT scan can optimize screening performance. Using maternal serum markers obtained both before and at the time of the NT scan has the potential to further improve performance, but larger studies are needed to confirm this potential.

Glucocorticoids and histone deacetylase inhibitors cooperate to block the invasiveness of basal-like breast cancer cells through novel mechanisms

Aggressive cancers often express E-cadherin in cytoplasmic vesicles rather than on the plasma membrane and this may contribute to the invasive phenotype of these tumors. Therapeutic strategies are not currently available that restore the anti-invasive function of E-cadherin in cancers. MDA-MB-231 cells are a frequently used model of invasive triple-negative breast cancer, and these cells express low levels of E-cadherin that is mislocalized to cytoplasmic vesicles. MDA-MB-231 cell lines stably expressing wild-type E-cadherin or E-cadherin fused to glutathione S-transferase or green fluorescent protein were used as experimental systems to probe the mechanisms responsible for cytoplasmic E-cadherin localization in invasive cancers. Although E-cadherin expression partly reduced cell invasion in vitro, E-cadherin was largely localized to the cytoplasm and did not block the invasiveness of the corresponding orthotopic xenograft tumors. Further studies indicated that the glucocorticoid dexamethasone and the highly potent class I histone deacetylase (HDAC) inhibitor largazole cooperated to induce E-cadherin localization to the plasma membrane in triple-negative breast cancers, and to suppress cellular invasion in vitro. Dexamethasone blocked the production of the cleaved form of the CDCP1 (that is, CUB domain-containing protein 1) protein (cCDCP1) previously implicated in the pro-invasive activities of CDCP1 by upregulating the serine protease inhibitor plasminogen activator inhibitor-1. E-cadherin preferentially associated with cCDCP1 compared with the full-length form. In contrast, largazole did not influence CDCP1 cleavage, but increased the association of E-cadherin with γ-catenin. This effect on E-cadherin/γ-catenin complexes was shared with the nonisoform selective HDAC inhibitors trichostatin A (TSA) and vorinostat (suberoylanilide hydroxamic acid, SAHA), although largazole upregulated endogenous E-cadherin levels more strongly than TSA. These results demonstrate that glucocorticoids and HDAC inhibitors, both of which are currently in clinical use, cooperate to suppress the invasiveness of breast cancer cells through novel, complementary mechanisms that converge on E-cadherin.