Sánchez-Durán et al. BMC Pregnancy and Childbirth (2019) 19:274 https://doi.org/10.1186/s12884-019-2434-0 RESEARCH ARTICLE Open Access Clinical application of a contingent screening strategy for trisomies with cell-free DNA: a pilot study María Ángeles Sánchez-Durán1,2, Andrea Bernabeu García1,2, Inés Calero1,2, Jordi Ramis Fossas2,3, Tamara Illescas1,2, María Teresa Avilés1,2, Nerea Maiz1,2* and Elena Carreras1,2 Abstract Background: Different strategies have been designed for clinical implementation of cell-free DNA (cfDNA) testing. We aimed to evaluate the performance of a contingent strategy based on conventional screening and offering cfDNA to the intermediate-risk group, for the screening for trisomies 21, 18 and 13. Secondary objectives were to assess the uptake of cfDNA in women with intermediate-risk, to evaluate the performance of cfDNA testing, and the preferences of pregnant women with intermediate risk. Methods: Prospective observational pilot study between February 2016 and March 2017. Singleton pregnancies with a known outcome were included in the study. At the conventional screening (first trimester combined test or second trimester quadruple test) women were classified in high (risk ≥1:250) or low risk (< 1:250). For the study, a contingent strategy was applied: following the conventional screening women were classified into three groups: high risk (risk ≥1:10 or nuchal translucency ≥3 mm), intermediate-risk (risk 1:11 to 1:1500) and low risk (< 1:1500), and a cfDNA test was offered to those at the intermediate risk. Results: For the analysis, 2639 women were included, 2422 (91.8%) had a first trimester combined test and 217 (8.2%) a second trimester quadruple test. There were 5 cases of trisomy 21, 4 of trisomy 18 and none of trisomy 13. For the contingent strategy, the detection rate and false positive rates were 88.9% (8/9) and 1.3% (35/2630), respectively. For the conventional strategy, the detection rate and false positive rates were 66.7% (6/9) and 5.3% (140/2630), respectively. The cfDNA test had a detection rate for trisomy 21 of 100% (3 out of 3), and a false positive rate of 0.2% (1/466). In a survey, 81.8% (374/457) of women in the intermediate-risk group would choose cfDNA testing as the second line test, mainly due to the lack of risk for the fetus. Conclusion: A contingent screening strategy for trisomies 21, 18 and 13, based on conventional screening, and offering a cfDNA test to women with a risk between 1:11 to 1:1500, reduced the false positive rate and increased the detection rate for these trisomies. Moreover, this strategy is well accepted by women. Keywords: Cell-free DNA, Contingent screening, Fetal trisomy, Prenatal screening * Correspondence: nmaiz@vhebron.net 1 Maternal-Fetal Medicine Unit, Maternal-Fetal Medicine Department, Hospital Universitari Vall d’Hebron, Ps. Vall d’Hebron 119-129, 08035 Barcelona, Spain 2 Universitat Autònoma de Barcelona, Barcelona, Spain Full list of author information is available at the end of the article © The Author(s). 2019 Open Access This article is distributed under the terms of the Creative Commons Attribution 4.0 International License (http://creativecommons.org/licenses/by/4.0/), which permits unrestricted use, distribution, and reproduction in any medium, provided you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons license, and indicate if changes were made. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Sánchez-Durán et al. BMC Pregnancy and Childbirth (2019) 19:274 Page 2 of 7 Background these data, according to prenatal diagnosis protocol for First trimester combined testing has become the stand- fetal congenital anomalies [5]. A second trimester quad- ard screening strategy for Down syndrome, with a detec- ruple test, including measurement of serum β-hCG, tion rate of 90% for trisomy 21, 95% for trisomies 18 and alpha-fetoprotein, unconjugated-oestriol, and inhibin-A, 13, and a false positive rate of 5% [1]. More recently, the was offered to those women attending the screening study of cell-free DNA (cfDNA) in maternal blood has antenatal clinic after 13 + 6 weeks. shown higher detection rates [2, 3], with an important reduction in false positive rates, minimizing the number of invasive procedures and their complications. The Conventional screening universal use of cfDNA testing is controversial due to In the routine clinical practice, following first trimester concerns about cost. An alternative is a contingent combined test or second trimester quadruple test strategy in two steps; firstly a conventional screening is women were classified into two groups: high-risk, if the performed, and secondly, a cfDNA test is offered only to risk for trisomies 21, 18 or 13 was ≥1:250 or the nuchal those patients with intermediate-risk. This strategy translucency ≥3 mm, and low risk, if the risk was would increase the detection rate and reduce the false lower than 1:250. An invasive test was offered to positive rate [4]. women at high risk. In the low-risk group, no add- The aim of the study was to assess the performance of itional tests were offered and a second trimester scan the contingent screening for trisomies 21, 18 and 13. was scheduled (Fig. 1). Secondary objectives were to assess the uptake of cfDNA in women with intermediate-risk, to evaluate the performance of cfDNA testing, and the preferences of Contingent screening (study) pregnant women with intermediate risk. For the study, women were classified in three groups following the conventional screening: High-risk (risk ≥1: Methods 10 or nuchal translucency ≥3 mm), intermediate-risk Study design and population (risk between 1:11 and 1:1500) and low-risk (risk < 1: This was a prospective observational study performed 1500). An invasive test was offered to women at high between February 2016 and March 2017 at Vall d’Heb- risk, no further test to women at low risk, and a cfDNA ron University Hospital, Barcelona, Spain. This was a test was offered to those with intermediate-risk. Never- pilot study, previous to the implementation of a contin- theless, if the risk was between 1:11 and 1:250 addition- gent strategy based on conventional screening (first tri- ally an invasive test was offered, according to the mester combined test and second-trimester quadruple national screening protocol (Fig. 2). test) and offering cfDNA to the intermediate-risk group A chorionic villus sampling or amniocentesis was at the public health system in our region, carried out depending on gestational age. Prenatal diag- The study population was women referred consecu- nostic results of chromosomes 13, 18, 21, X and Y were tively to the hospital for aneuploidies screening. Eligible released within 48 h of sample collection using QF-PCR patients were at least 18 years of age, with a singleton and results of all chromosomal abnormalities were re- pregnancy. Exclusion criteria were vanishing twin preg- leased with a time frame of 2–3 weeks by karyotype and/ nancy and unknown karyotype or unknown neonatal or array analysis. phenotype. This study was approved by the institutional Regarding cfDNA, peripheral maternal blood (20 mL) research ethics committee (CEIC-VHIR) and informed was collected into a standard ethylenediaminetetraacetic consent was obtained from all women included in this acid blood collection tube during the screening visit and study. sent via courier to the USA for cfDNA testing (Har- mony® prenatal test, Ariosa Diagnostics, Inc., San Jose, Clinical protocol CA, USA). As required in the regional prenatal screening protocol, Genetic counseling was carried out by a trained obstet- all pregnant women were offered to screen for fetal tri- rician and screening results of the most common triso- somies at the first prenatal care visit. The first-trimester mies (21, 18 and 13) were expected to be obtained screening protocol included measurement of serum con- within 7–10 days. Women with a screen-positive result centration of β-human chorionic gonadotropin (β-hCG) (high-risk result from cfDNA testing) were advised to and pregnancy-associated plasma protein-A (PAPP-A) at have an invasive procedure in order to confirm the 8 to 12 weeks, and an ultrasound scan at 11+ 0 to 13+ 6 diagnosis, whereas women with a screen-negative result weeks, where fetal crown-rump length [CRL] and nuchal (low-risk result from cfDNA testing) continued conven- translucency were measured. Subsequently, the risk for tional obstetric control during pregnancy as low-risk trisomies 21, 18 and 13 was calculated combining all patients. Sánchez-Durán et al. BMC Pregnancy and Childbirth (2019) 19:274 Page 3 of 7 Fig. 1 Conventional screening strategy After cfDNA results delivery, women were asked to fill visit. The risks for trisomies 21, 18 and 13 and the pro- out an anonymous survey concerning the available cedure performed after screening (invasive test, cfDNA contingent second-line options (Additional file 1). or routine obstetric control) were collected in the data- All pregnant women were followed up until the end of base. We recorded the invasive test results, the cfDNA pregnancy and all the newborns were studied for a results (risk for trisomy 21, 18 or 13, fetal fraction in phenotype by a pediatrician and a karyotype was re- cfDNA from maternal plasma, test with no results, repe- quested if necessary. titions and time for result) and perinatal results of pregnancy. Study variables The primary outcome of this study was the perform- Demographic maternal characteristics, including mater- ance (detection rate and false positive rate) of the nal age, weight, and race, were collected in the study contingent strategy based on conventional screening and electronic database by obstetricians during the screening using cfDNA in the screening for trisomy 21, 18 and 13. Fig. 2 Contingent screening strategy Sánchez-Durán et al. BMC Pregnancy and Childbirth (2019) 19:274 Page 4 of 7 Secondary outcomes were the detection rate and false Caucasian, 131 (5.0%) North African, 73 (2.8%) Sub-Sa- positive rates of the cfDNA test, as well as the prefer- haran, 71 (2.7%) East Asian, 51 (1.9%) South Asian and ences of pregnant women. 6 (0.2%) mixed race. Statistical analysis Conventional screening For the descriptive analysis, categorical data are shown According to the conventional test (first trimester com- as frequency and percentage, and continuous variables bined test or second trimester quadruple test), 146 of as the median and interquartile range (IQR). Pearson 2639 (5.5%) women had a high risk (screen positive rate) correlation was used to determine the association and 2493 (94.5%) had a low risk (Fig. 1). The detection between fetal fraction and body mass index (BMI), and rate for all trisomies (trisomy 21, 18 and 13) was 66.7% between the fetal fraction and the gestational age. Detec- (6/9) and the false positive rate 5.3% (140/2630). The tion rate and false positive rate of both conventional and specific detection rates for trisomy 21 and 18 were 60% contingent screening strategies were calculated. Statis- (3/5), and 75% (3/4), respectively. tical significance level was set at p < 0.05. The statistical software package SPSS 23.0 (IBM SPSS Statistics for Contingent screening Windows, Version 23.0. Armonk, NY: IBM Corp) was Following the conventional screening 33 (1.3%) women used for the analysis of the data. had a high risk (14 with a risk of 1:10 or more, and 19 fetuses with nuchal translucency of 3 mm or more), 496 Results (18.8%) patients had an intermediate risk (between 1:11 Study population and 1:1500) and 2110 (79.9%) women had a low risk Two thousand nine hundred and eighty-six patients (< 1:1500). were screened in this period, 2706 women were eligible From the 496 women with intermediate-risk 27 (5.4%) for the study, from these 67 (2.5%) were excluded, 2639 declined cfDNA testing, 7 had an invasive test (5 of women were included for analysis, 2422 (91.8%) who them had a high risk), 2 had a cfDNA in a private center, had a first trimester combined test and 217 (8.2%) a sec- one miscarried, one moved to another city, one refused ond trimester quadruple test (Fig. 3). In this population, due to difficulties for blood extraction and the other 15 there were 5 cases of trisomy 21, 4 of trisomy 18 and no refused to give a reason. In the 469 women who had a cases of trisomy 13 (Table 1). cfDNA test, 3 had a high risk for trisomy 21, one had a Median maternal age was 32.1 years (IQR, 28.1 to high risk for trisomy 13 and 463 had a low risk, and in 36.0) and median weight 63 kg (IQR, 55.2 to 72.0). Two two cases no result was obtained after repeating the test thousand three hundred and seven women (87.4%) were two and three times, respectively. The reason for the no Fig. 3 Enrollment Sánchez-Durán et al. BMC Pregnancy and Childbirth (2019) 19:274 Page 5 of 7 Table 1 Description of the 9 cases of aneuploidies Case Aneuploidy Screening Trimester Risk 1 in Conventional screening cfDNA Contingent screening 1 Trisomy 21 1 3 High risk – High risk 2 Trisomy 21 1 7 High risk – High risk 3 Trisomy 21 1 54 High risk High-risk High risk 4 Trisomy 21 1 596 Low risk High-risk High risk 5 Trisomy 21 2 741 Low risk High-risk High risk 6 Trisomy 18 1 1 High risk – High risk 7 Trisomy 18 1 1 High risk – High risk 8 Trisomy 18 1 1 High risk – High risk 9 Trisomy 18 1 13,735 Low risk – Low risk cfDNA cell-free DNA result in both cases was a low fetal fraction. The risk for invasive test was performed and the result was normal. Down syndrome at the conventional screening in these The overall no-result rate was 0.43% (2/469). cfDNA two cases was 1:213 and 1:908, respectively. Both women results were issued after a mean of 6.7 (SD2.3) days, 90 who had no result and those who refused cfDNA testing and 87% of results were available within 8 and 7 days, were reclassified according to the result at the conven- respectively. tional screening. Following the cfDNA testing in the The median fetal fraction was 11.9% (IQR, 9.1 to 14.7). intermediate-risk group, 43 (1.6%) women had a high There was a significant correlation between fetal fraction risk (screen positive rate), and 2596 (98.4%) had a low and BMI, r = -0.364 (95% CI: − 0.356 to − 0.220; p < risk (Fig. 2). 0 .001), and there was no significant correlation between In addition to the cfDNA test, an invasive test was fetal fraction and gestational age (r = 0.086, 95% CI: − offered to the group of 113 women with a risk between 0.013 to 0.354; p = 0.068). 1:11 and 1:250, 32 (28.3%) of them declined it and 81 A total of 457 (97.4%) patients at intermediate-risk (71.7%) underwent an invasive test. filled out the anonymous survey, the remaining patients For the contingent strategy, the detection rate for all of this group did not participate due to a language bar- three trisomies was 88.9% (8/9) and the false positive rier. The results showed that 374 (81.8%) women would rate was 1.3% (35/2630). The specific detection rates for have preferred cfDNA testing as the second line contin- trisomy 21 and 18 were 100% (5/5) and 75% (3/4), gent test, 80 (17.5%) would have preferred an invasive respectively. procedure, and 3 (0.7%) women not doing anything. The reasons for these answers are given in Table 2. Regarding Performance of cfDNA test the time for results, 239 women (52.3%) considered that In the 469 women with intermediate-risk who accepted it was short, 184 (40.3%) that it was appropriate, 30 cfDNA testing, there were 3 cases of trisomy 21 and (6.6%) thought that it was too long, and 4 (0.9%) did not none of trisomy 18 or 13. In one case the result in- respond to this question. formed of high risk for trisomy 13, and the karyotype was normal. Median gestational age at cfDNA testing Discussion was 13.2 weeks (interquartile range, 12.5–14.1). Main findings of the study The detection rate for trisomy 21 was 100% (3 out of This study confirms that a contingent screening strategy, 3), and the overall false positive rate was 0.2% (1/466). based on the first trimester combined test or second The test was repeated in 15 (3.2%) women, in 8 cases trimester quadruple test and offering a cfDNA test to due to a sample mismanagement (4 cases where the the intermediate-risk group, can improve the perform- samples were not labeled, 4 cases where the sample was ance of the screening for trisomies 21, 18 and 13 by both lost during transport), 2 cases had a low fetal fraction increasing the detection rate and reducing the false posi- and in 5 cases the samples did not meet quality criteria. tive rate. CfDNA was well accepted by this group of In the second analysis, 12 cases showed a low risk for women, with an uptake of 94.6%, including women trisomies, and three samples again had no results, two of classified as ‘high risk’ by the conventional screening them were repeated, one had a low risk, and the other test. The main reason for accepting this test was the lack sample had no results for the third time. The latter was of risk for the fetus. reported as a normal newborn at the end of follow-up. Focusing on the cfDNA test, the performance was ex- Regarding the sample which was not repeated, an cellent in this small sample, with a detection rate of Sánchez-Durán et al. BMC Pregnancy and Childbirth (2019) 19:274 Page 6 of 7 Table 2 Results of the survey Q1. Following the conventional screening, which procedure would you have preferred as the first option? cfDNA (n = 374) Invasive testing (n = 80) No further testing (n = 3) Q2.Why? Diagnostic accuracy 16 (4.3%) 76 (95.0%) 0 No risk for fetus 349 (93.3%) 1 (1.3%) 0 Earlier diagnosis 2 (0.5%) 1 (1.3%) 0 Othera 7 (1.9%) 2 (2.5%) 3 (100%) cfDNA cell-free DNA a Other: cfDNA: 5/7 cases preferred to avoid a painful test; Invasive testing: did not give a specific reason; No further testing: 2/3 cases committed to the pregnancy. The rest of the cases did not specify a reason Women’s opinion about available prenatal testing options 100% for trisomy 21, and an overall false positive rate of women of this group opted to avoid invasive testing. We 0.2%. The no-result rate of the test was 0.43% after re- believe that this is explained by the lack of confidence of peating the test up to three times. In a survey after the the women in the test due to the novelty. screening process, more than 80% of women in the intermediate-risk group would choose cfDNA testing as the first option, and more than 90% of women consid- Strengths and limitations ered that the time to deliver the results was appropriate. The main limitation of this study is the sample size. The small number of affected fetuses in this cohort does not Compare to the literature guarantee an accurate assessment and, therefore, a study Our findings are similar to those reported by other with a larger sample size would ensure these results. groups where a contingent screening, by means of a This study evaluated the feasibility of a contingent cfDNA test in the intermediate-risk group increases the screening in real conditions, where 8% of women missed detection rate and reduces the false positive rate [6, 7]. the first-trimester screening and had a second-trimester However, the performance and cost-effectiveness are dif- screening. ficult to compare among studies where different cut-offs The uptake of our study is not comparable to other have been used [8]. The cut-offs chosen in our study studies since this was a pilot study carried out under a were 1:11 and 1:1500. When we have addressed the issue regional clinical protocol, and women could not be of implementing a contingent screening, we have set deprived of the options included in this protocol, hence two goals: reducing the invasive test rate and increasing high-risk patients were offered an invasive procedure the detection rate. In order to reduce the invasive rate, despite participating in the study and having a cfDNA we had to offer a cfDNA test to the high-risk group (≥1: test. 250). We observed that most chromosomal abnormal- ities were in the group of risk of 1:10 or more, or in those with a nuchal translucency higher than 3 mm. As Clinical implications a consequence, we opted for this cut-off at the high-risk The confirmation of the good performance of the con- site. On the opposite side, we aimed to increase the de- tingent strategy, the high uptake and good acceptance of tection rate and chose 1:1500 based on our own data the test by the patients will permit us to introduce the and adjusting to our budget. contingent screening in the clinical practice modifying The no-result rate in our population was 3.2% follow- the regional protocol. Invasive testing will remain as ing the first sample collection, and it was reduced to the first choice in cases of increased nuchal translu- 0.43% after repeating the sample collection up to three cency, fetal structural abnormalities or risk higher times. Other studies have reported no-results rates for than 1:10, and as a diagnostic test following a positive trisomies up to 5.9% [2]. cfDNA result. It is important to highlight that the performance of the contingent strategy depends on the performance of the conventional screening (first trimester combined test Conclusions or second trimester quadruple test), it is, therefore, im- A contingent screening strategy for trisomies 21, 18 and portant to maintain the standards of quality for these 13, based on conventional screening, and offering a screening tests. cfDNA test to women with a risk between 1:11 to 1: Previous studies have shown a reduction in the rate of 1500, reduced the false positive rate and increased the invasive testing from 65 to 37% in the high-risk group detection rate for these trisomies. Moreover, this strategy (from the combined test) [6]. In our study, only 28% of was well accepted by women. Sánchez-Durán et al. BMC Pregnancy and Childbirth (2019) 19:274 Page 7 of 7 Additional file fetal congenital anomalies]. Primera edición. Barcelona: Direcció General de Salut Pública; 2008. 6. Gil MM, Revello R, Poon LC, Akolekar R, Nicolaides KH. Clinical Additional file 1: Anonymous survey concerning available prenatal implementation of routine screening for fetal trisomies in the UK NHS: cell- testing options. (DOCX 12 kb) free DNA test contingent on results from first-trimester combined test. Ultrasound Obstet Gynecol. 2016;47:45–52. Abbreviations 7. Miltoft CB, Rode L, Ekelund CK, Sundberg K, Kjaergaard S, Zingenberg H, et BMI: Body mass index; CfDNA: Cell-free DNA; CI: Confidence interval; al. Contingent first-trimester screening for aneuploidies with cell-free DNA CRL: Crown-rump length; IQR: Interquartile range; PAPP-A: Pregnancy- in a Danish clinical setting. Ultrasound Obstet Gynecol. 2018;51:470–9. associated plasma protein-A; β-hCG: β-human chorionic gonadotropin 8. García-Pérez L, Linertová R, Álvarez-de-la-Rosa M, Bayón JC, Imaz-Iglesia I, Ferrer-Rodríguez J, et al. Cost-effectiveness of cell-free DNA in maternal blood testing for prenatal detection of trisomy 21, 18 and 13: a systematic Acknowledgments Not applicable. review. Eur J Health Econ. 2018;19:979–91. Authors’ contributions Publisher’s Note MAS, ABG: Conceived and designed the study, analyzed and interpreted the Springer Nature remains neutral with regard to jurisdictional claims in data and wrote the first draft of the work. IC, MTA, JRF, TI: Contributed to the published maps and institutional affiliations. acquisition of data and critically reviewed the manuscript. NM: Contributed to the analysis and interpretation of data and critically reviewed the manuscript. EC: has critically reviewed the manuscript. All authors read and approved the final manuscript. Funding The study was financially supported by Ariosa Diagnostics, Inc. San Jose, CA, USA. This organization had no role in the study design, data collection, data analysis, data interpretation or writing of the report. Ariosa reviewed the final manuscript prior to its disclosure to ensure that no Ariosa Confidential Information is contained therein. Availability of data and materials The datasets used and analyzed during the current study are available from the corresponding author on reasonable request. Ethics approval and consent to participate Ethical approval for this study was provided by the Ethical Committee CEIC Vall d’Hebron Research Institute (CEIC-VHIR), PR(AG)276/2014. Written informed consent was obtained from all women included in this study. Consent for publication Not applicable. Competing interests The authors declare that they have no competing interests. Author details 1 Maternal-Fetal Medicine Unit, Maternal-Fetal Medicine Department, Hospital Universitari Vall d’Hebron, Ps. Vall d’Hebron 119-129, 08035 Barcelona, Spain. 2 Universitat Autònoma de Barcelona, Barcelona, Spain. 3Department of Laboratory, Hospital Universitari Vall d’Hebron, Barcelona, Spain. Received: 29 January 2019 Accepted: 26 July 2019 References 1. Kagan KO, Wright D, Valencia C, Maiz N, Nicolaides KH. Screening for trisomies 21, 18 and 13 by maternal age, fetal nuchal translucency, fetal heart rate, free beta-hCG and pregnancy-associated plasma protein-A. Hum Reprod. 2008;23:1968–75. 2. Gil MM, Accurti V, Santacruz B, Plana MN, Nicolaides KH. Analysis of cell-free DNA in maternal blood in screening for aneuploidies: updated meta- analysis. Ultrasound Obstet Gynecol. 2017;50:302–14. 3. Taylor-Phillips S, Freeman K, Geppert J, Agbebiyi A, Uthman OA, Madan J, et al. Accuracy of non-invasive prenatal testing using cell-free DNA for detection of down, Edwards and Patau syndromes: a systematic review and meta-analysis. BMJ Open. 2016;6:e010002. 4. American College of Obstetricians and Gynecologists Committee on Genetics. Committee opinion no. 545: noninvasive prenatal testing for fetal aneuploidy. Obstet Gynecol. 2012;120:1532–4. 5. Generalitat de Catalunya. Departamento de Salud. Protocolo de diagnóstico prenatal de anomalías congénitas fetales. .[Prenatal diagnosis protocol for