Cell-free DNA screening has superior aneuploidy detection compared with traditional tests.1 Its use as a primary or secondary screening is currently recommended for both the general obstetric population and high-risk pregnancies from 10 weeks of gestation. Despite the high sensitivity and specificity in screening for Down syndrome,2,3 the positive predictive value can be low, particularly among young women and for the detection of other fetal aneuploidies with low prevalence in the population.1 The higher the proportion of fetal DNA in the maternal circulation (fetal fraction), the more accurate the test is,4 and low fetal fraction is the main cause of noninformative results.5
Studies have consistently found that fetal fraction increases with gestational age and decreases with rising maternal weight.3,5–11 It is hypothesized that this reduction in fetal fraction is a consequence of a dilution effect and proportionally higher release of maternal cell-free DNA in obese women.7
It has been suggested that delaying cell-free DNA testing may render better test success rates among obese women.6 Livergood et al6 reported a reduction in the predicted failure rate from 14.9% to 10.4% between 8 and 16 weeks of gestation in obese women. However, the gestational age at which failure rates in obese women were not significantly increased compared with women with normal body mass index (BMI, calculated as weight (kg)/[height (m)]2) was 21 weeks.6 Although the increase in fetal fraction was found not to be constant across different gestational ages,12 changes in fetal fraction levels throughout gestation in different BMI groups have not been previously reported.
This study aims to analyze the influence of maternal BMI on the incremental increase in fetal fraction as gestation advances and its effects on test failure rates.
MATERIALS AND METHODS
This was an observational cross-sectional study including consecutive singleton pregnancies undergoing cell-free DNA screening for fetal chromosomal abnormalities from 10 weeks of gestation, between May 2013 and January 2018, at two fetal medicine clinics in Melbourne and Sydney, Australia. Minors (younger than 18 years of age), multiple pregnancies, and cases in which cell-free DNA testing was not performed (as a result of a missed miscarriage, cotwin demise, or blood collection problems) were excluded.
Blood samples were collected onsite from consenting participants following pretest counseling provided by genetic counselors or fetal medicine specialists. Women who opted to use cell-free DNA testing as a primary screening test before 12 weeks of gestation undertook ultrasonography before blood sampling to confirm fetal number, viability, and gestational age. Subsequently, these patients were advised to return for fetal structural assessment between 12 and 13 6/7 weeks of gestation.
Demographic information, including maternal age, ethnicity, and method of conception, was recorded. Weight and height were measured on the day of blood sampling, and BMI was calculated.
During the study period, screening for trisomy 21, 18, and 13 and, when requested, sex chromosome conditions was performed by cell-free DNA analysis on maternal plasma using two different platforms. Platform A utilized digital analysis of selected regions for chromosome analysis and single nucleotide polymorphism analysis for estimating the fetal fraction in the sample,13 and platform B used next-generation sequencing and massive parallel sequencing for aneuploidy screening and fetal fraction measurement.14 No risk assessment was issued when fetal fraction was below 4% (platform A) or 2% (platform B) or when a technical issue did not allow accurate analysis of the sample (“technical failure”). Both platforms were used in both clinics. Platform A was the platform utilized until November 2016. Platform B has been used since then. The demographic characteristics of the study population, in each clinic, were similar during the study period, and no significant changes in technologies for fetal fraction estimation occurred during the study period.
Test results and fetal fraction were recorded. All women who received a noninformative report in the first attempt (“no-call”) were offered repeat cell-free DNA testing (“redraw”) at no additional cost and, if they were under 14 weeks of gestation, the option to have blood collected for first-trimester combined screening in case the cell-free DNA is noninformative again (“test failure”). Analysis of maternal serum biochemistry in the blood sample for combined screening included assessment of free β human chorionic gonadotropin and pregnancy-associated plasma protein A. As with all patients attending the practice, ultrasound assessment of fetal anatomy and nuchal translucency was also recommended.
The study complied with the National Health and Medical Research Council 2014 Ethical Considerations in Quality Assurance and Evaluation Activities15 and was approved by Monash Health Human Ethics Committee (NMH HREC Reference Number: LNR/17/MonH/101, Monash Health Ref: RES-17-0000-134-L).
Baseline characteristics of the population were expressed in absolute values and percentages for categorical variables and medians and interquartile ranges for continuous variables. The main outcome measure was the increase in fetal fraction throughout gestation and secondary outcomes were no-call and test failure rates.
Body mass index was categorized as normal (BMI less than 25.0), overweight (BMI 25.0–29.9), obesity class I (BMI 30.0–34.9), and obesity classes II and III (BMI 35.0 or greater). As a result of low numbers, separate categories for low BMI (less than 18.5) and class III obesity (greater than 40) were unable to be assessed. Thirty-one cases (0.2%) in which BMI was not available were excluded from analysis. Fetal fraction changes throughout gestation for each platform were compared among the four groups with the use of scatterplots and lines of best fit derived from linear regression for all patients in whom cell-free DNA testing was performed between 10 and 20 weeks of gestation. Goodness-of-fit was assessed by R2 values, and hypothesis testing was performed using logarithmically transformed fetal fraction values.
Median fetal fraction values between the groups were compared with Kruskal-Wallis test and Jonckheere-Terpstra test for ordered alternatives. Proportions of categorical variables (no-call results, test failure rates, and informative redraw) were compared between the groups using χ2 or Fisher exact test. Odds ratios (ORs) with 95% CIs and adjusted OR, allowing for ethnicity and mode of conception, were calculated for each platform and BMI group using univariable and multivariable logistic regression analysis. Predicted probabilities of no-call results and test failure were also derived from logistic regression analysis. All tests were two-tailed and a significance level of 5% was adopted.
Statistical analysis was performed using SPSS 25.0 and the results of this study were reported according to the STrengthening the Reporting of OBservational studies in Epidemiology (STROBE) statement.16
A total of 14,223 singleton pregnancies had cell-free DNA testing performed and were included in the analysis, including 8,583 (60.3%) patients screened with platform A and 5,640 (39.7%) with platform B. Classification according to BMI in normal, overweight, obesity class I, and obesity classes II and III comprised, respectively, 64.1% (9,118), 23.2% (3,304), 8.2% (1,171), and 4.5% (630) of the patients. Baseline characteristics of the study population are shown in Table 1.
Only 9 cases of 14,112 (0.06%) in which results were informative had to be excluded from analysis because fetal fraction was not available, five of them in platform A and four in platform B.
Median fetal fraction values decreased with increasing BMI across both platforms and were significantly lower in platform B than in platform A within the same BMI group (P<.001; Table 2). There was an average increase in fetal fraction of 0.36 units/wk between 10 and 20 weeks of gestation in the normal BMI group. In the group with BMIs of 35 or greater, this change was less pronounced (0.10 units/wk) and the slope was not statistically different from zero (Fig. 1; Table 3). The rate of increase in fetal fraction in the whole study population was significantly higher after 20 weeks of gestation than before this gestational age (1.0 vs 0.3 units/wk, P<.001; Fig. 2).
There were 316 cases (2.2%) of no-call results, of which 214 occurred with platform A (1.5% of total, 2.5% of platform A) and 102 with platform B (0.7% of total, 1.8% of platform B). Of the patients with no-call results, 98.4% (311/316) had repeat testing, and test failure occurred in 120 patients (0.8% of total, 39% of the patients who had a repeat test) with 81 failures after testing with platform A (0.6% of total, 0.9% of platform A) and 39 failures when platform B was used (0.3% of total, 0.7% of platform B). In the highest BMI group, the no-call rates were 16.6% (platform A) and 7.8% (platform B) and the failure rates were 6.4% (platform A) and 2.7% (platform B). After adjustment for independent predictors in the multivariable logistic regression analysis (ethnicity and mode of conception), women with BMIs of 35 or greater were much more likely to receive a no-call (adjusted OR 22.0 [A] and 8.0 [B]) or failed result (adjusted OR 25.0 [A] and 5.8 [B]) as compared with women with normal BMIs. To correct for possible overfitting of the models, unadjusted ORs were calculated with univariable logistic regression analysis with values that were similar to those obtained after adjustment (Table 2).
The risk of test failure after adjustment for gestational age relatively increased by 13.6%/unit increase in BMI. No-call and test failure results were less common with platform B in all BMI groups. The rates of informative redraw were similar across different BMI groups and between the two platforms (Table 2). Between 10 and 20 weeks of gestation, there was no significant decrease in the predicted probability of no-call results in the group with BMIs of 35 or greater, and no overlap of CIs between this group and lower BMI categories was found (Fig. 3).
Across increasing BMI groups, the average increase in fetal fraction with gestation is progressively lower, being minimal in women with BMIs of 35 or greater. The average increase in fetal fraction per week in the normal BMI group was 0.36 units; in women with BMIs of 35 or greater, the average increase was 0.10 units/wk.
Among high BMI groups, the median fetal fraction is significantly lower. This finding is consistent with previously published studies.6–9,12 Kinnings et al12 reported a 1.17% decrease in fetal fraction for every 5-kg/m2 increase in maternal BMI.
No-call and failure rates of cell-free DNA testing were higher among women with obesity classes II and III. Livergood et al6 demonstrated an 8.5-fold increase in risk of a no-call result among women with obesity class III. They reported a decrease in the predicted no-call rates between 8 and 16 weeks of gestation from 14.9% to 10.4%, but they also found that significant differences in no-call rates between women with high and normal BMIs no longer exist at 21 weeks of gestation, at which time reproductive choices may be limited.6 Our results suggest that there is no significant reduction in the predicted risk of test failure among women with BMIs of 35 from 10 to 20 weeks of gestation. Consequently, delaying the test to a later gestation is unlikely to reduce no-call and failure rates in this population.
Comparing the platforms, no-call and failure rates were significantly higher among obese women using single nucleotide polymorphism analysis compared with next-generation sequencing, suggesting that sequencing methods may be preferable in high BMI groups. Although the second platform reports lower fetal fraction values than the first, the threshold for issuing a no-call report is lower, leading to a reduced failure rate. How these differences affect the performance in detecting fetal aneuploidies needs to be addressed in further research. Because the technologies used are not the same, the fetal fraction values may not be comparable, and a given test may be able to accurately discriminate affected from unaffected cases even in lower fetal fraction ranges.17
The rate of informative redraw in women with a prior no-call result was similar across the BMI groups, which is supported by a previous large study.12 Wang et al,9 however, observed a lower informative redraw rate in higher weight groups (18.2% for weight greater than 140 kg vs 71.4% for weight less than 90 kg), thought to be related to a slower fetal fraction rise and indicating that delaying the redraw several weeks may be worthwhile in obese women.
Although cell-free DNA testing has the benefit of higher accuracy compared with combined screening, the possibility of a no-call result is its main downside, especially among women with increased BMI, because it impairs timely pregnancy management and may increase maternal anxiety levels.
An ultrasonogram at 11–14 weeks of gestation is recommended in all cases,18 and first-trimester combined screening can provide additional information in patients in whom cell-free DNA testing fails, because serum biochemistry is independent of fetal fraction and nuchal translucency measurement can be obtained in a significant proportion of women with increased BMI.19 Hence, it may be helpful to store blood for maternal serum biochemistry at 10 and 13 6/7 weeks of gestation,6 especially in high BMI groups, allowing the calculation of aneuploidy risk in the event of a failed cell-free DNA test and additionally to screen for other complications such as preeclampsia in this particularly high-risk population who could benefit from preventive measures.20–22
The main limitation of this study is its cross-sectional design. Consequently, we did not analyze the performance of the test in detecting fetal aneuploidies between different BMI groups. As a result of lower fetal fraction, poorer performance of the test in women with higher BMIs would be expected.4
Maternal factors that were not available in our study may be related to lower fetal fraction. Chronic hypertension is more prevalent in obese women and related to lower fetal fraction.23,24 Because this information was not available in a significant number of cases, no adjustments could be made for this confounder.
Despite the large sample size, the absolute numbers of no-call and failed results in specific BMI groups were limited, leading to wide CIs and possible overfitting of the regression models. For this reason, we reported crude ORs, which, reassuringly, were similar to the adjusted ones. The number of women having the test after 26 weeks of gestation was small. This is inherent to the fact that cell-free fetal DNA testing is usually performed as a primary screening test before 14 weeks of gestation to allow timely decision-making, and we therefore restricted the regression models to cases before 26 weeks of gestation.
In increased BMI groups, fetal fraction is lower and the rates of no-call and failed results on cell-free DNA testing are higher. The rate of increase in fetal fraction is minimal in women with BMIs above 35, and therefore delaying the test is unlikely to reduce the risk of test failure. These high rates of test failure should be taken into account during pretest counseling of pregnant women.
1. Gil MM, Quezada MS, Revello R, Akolekar R, Nicolaides KH. Analysis of cell-free DNA in maternal blood in screening for fetal aneuploidies: updated meta-analysis. Ultrasound Obstet Gynecol 2015;45:249–66.
2. Screening for fetal aneuploidy. Practice Bulletin No. 163. American College of Obstetricians and Gynecologists. Obstet Gynecol 2016;127:e123–37.
3. Lee TJ, Rolnik DL, Menezes MA, McLennan AC, da Silva Costa F. Cell-free fetal DNA testing in singleton IVF conceptions. Hum Reprod 2018;33:572–8.
4. Wright D, Wright A, Nicolaides KH. A unified approach to risk assessment for fetal aneuploidies. Ultrasound Obstet Gynecol 2015;45:48–54.
5. Revello R, Sarno L, Ispas A, Akolekar R, Nicolaides KH. Screening for trisomies by cell-free DNA testing of maternal blood: consequences of a failed result. Ultrasound Obstet Gynecol 2016;47:698–704.
6. Livergood MC, LeChien KA, Trudell AS. Obesity and cell-free DNA ‘no calls’: is there an optimal gestational age at time of sampling? Am J Obstet Gynecol 2017;216:413.e1–9.
7. Ashoor G, Poon L, Syngelaki A, Mosimann B, Nicolaides KH. Fetal fraction in maternal plasma cell-free DNA at 11-13 weeks' gestation: effect of maternal and fetal factors. Fetal Diagn Ther 2012;31:237–43.
8. Ashoor G, Syngelaki A, Poon LC, Rezende JC, Nicolaides KH. Fetal fraction in maternal plasma cell-free DNA at 11-13 weeks' gestation: relation to maternal and fetal characteristics. Ultrasound Obstet Gynecol 2013;41:26–32.
9. Wang E, Batey A, Struble C, Musci T, Song K, Oliphant A. Gestational age and maternal weight effects on fetal cell-free DNA in maternal plasma. Prenat Diagn 2013;33:662–6.
10. Canick JA, Palomaki GE, Kloza EM, Lambert-Messerlian GM, Haddow JE. The impact of maternal plasma DNA fetal fraction on next generation sequencing tests for common fetal aneuploidies. Prenat Diagn 2013;33:667–74.
11. Scott FP, Menezes M, Palma-Dias R, Nisbet D, Schluter P, da Silva Costa F, et al. Factors affecting cell-free DNA fetal fraction and the consequences for test accuracy. J Matern Fetal Neonatal Med 2018;31:1865–72.
12. Kinnings SL, Geis JA, Almasri E, Wang H, Guan X, McCullough RM, et al. Factors affecting levels of circulating cell-free fetal DNA in maternal plasma and their implications for noninvasive prenatal testing. Prenat Diagn 2015;35:816–22.
13. Sparks AB, Wang ET, Struble CA, Barrett W, Stokowski R, McBride C, et al. Selective analysis of cell-free DNA in maternal blood for evaluation of fetal trisomy. Prenat Diagn 2012;32:3–9.
14. Bianchi DW, Platt LD, Goldberg JD, Abuhamad AZ, Sehnert AJ, Rava RP, et al. Genome-wide fetal aneuploidy detection by maternal plasma DNA sequencing. Obstet Gynecol 2012;119:890–901.
15. NHRMC. Ethical considerations in quality assurance and evaluation activities. Available at: https://www.nhmrc.gov.au/guidelines-publications/e111
. Retrieved September 22, 2017.
16. STROBE statement—checklist of items that should be included in reports of observational studies (STROBE initiative). Int J Public Health 2008;53:3–4.
17. Artieri CG, Haverty C, Evans EA, Goldberg JD, Haque IS, Yaron Y, et al. Noninvasive prenatal screening at low fetal fraction: comparing whole-genome sequencing and single-nucleotide polymorphism methods. Prenat Diagn 2017;37:482–90.
18. Salomon LJ, Alfirevic Z, Audibert F, Kagan KO, Paladini D, Yeo G, et al. ISUOG updated consensus statement on the impact of cfDNA aneuploidy testing on screening policies and prenatal ultrasound practice. Ultrasound Obstet Gynecol 2017;49:815–6.
19. Thornburg LL, Mulconry M, Post A, Carpenter A, Grace D, Pressman EK. Fetal nuchal translucency thickness evaluation in the overweight and obese gravida. Ultrasound Obstet Gynecol 2009;33:665–9.
20. Poon LC, Kametas NA, Maiz N, Akolekar R, Nicolaides KH. First-trimester prediction of hypertensive disorders in pregnancy. Hypertension 2009;53:812–8.
21. Rolnik DL, Wright D, Poon LC, O'Gorman N, Syngelaki A, de Paco Matallana C, et al. Aspirin versus placebo in pregnancies at high risk for preterm preeclampsia. N Engl J Med 2017;377:613–22.
22. O'Gorman N, Wright D, Syngelaki A, Akolekar R, Wright A, Poon LC, et al. Competing risks model in screening for preeclampsia by maternal factors and biomarkers at 11-13 weeks gestation. Am J Obstet Gynecol 2016;214:103.e1–12.
23. Rolnik DL, da Silva Costa F, Lee TJ, Schmid M, McLennan AC. Association between fetal fraction on cell-free DNA testing and first trimester markers for pre-eclampsia. Ultrasound Obstet Gynecol 2018 Jan 10 [Epub ahead of print].
24. Zhou Y, Zhu Z, Gao Y, Yuan Y, Guo Y, Zhou L, et al. Effects of maternal and fetal characteristics on cell-free fetal DNA fraction in maternal plasma. Reprod Sci 2015;22:1429–35.