Clinical and Economic Evaluation after Adopting Contingent Cell-Free DNA Screening for Fetal Trisomies in South Spain

1 Clinical and economic evaluation after adopting contingent cell free DNA 1 screening for fetal trisomies in South Spain. 2 3 Authors: José A. Sainz (1,2), María R. Torres (3), Ignacio Peral (4), Reyes 4 Granell (3), Manuel Vargas (1), Pilar Carrasco (5), José A. Garcia-Mejido 5 (1,2), Belén Santacruz (6,7), María M Gil (6,7) 6 7 8 Affiliation: 9 (1) Obstetrics and Gynecology Department, Hospital Universitario Valme, Seville, Spain. 10 (2) Obstetrics and Gynecology Department, Universidad de Sevilla, Spain. 11 (3) Obstetrics and Gynecology Department, Hospital Juan Ramn Jimnez, Huelva, 12 Spain 13 (4) Clinical Analysis Department. Hospital Universitario Valme, Seville, Spain. 14 (5) Genetics Unit, Clinical Analysis Department, Hospital Juan Ramn Jimnez, Huelva, 15 Spain. 16 (6) Obstetrics and Gynecology Department, Hospital Universitario de Torrejón, Torrejón 17 de Ardoz, Madrid, Spain. 18 (7) School of Health Sciences, Universidad Francisco de Vitoria (UFV), Pozuelo de 19 Alarcón, Madrid, Spain. 20 21 Corresponding author: José Antonio Sainz., Obstetrics and Gynecology Department, 22 Hospital Universitario Valme, Seville, Spain. 23 Phone number: 0034 955015385 E-mail: [email protected] 24 Short Title: cf-DNA contingent screening 25 26 Key Words: cell-free DNA; contingent screening; combined test; trisomies; 27 aneuploidies; Down syndrome; first trimester screening; non-invasive prenatal testing. 28 2 Abstract. 29 Introduction: Contingent cell-free (cf) DNA screening on the basis of the first-trimester 30 combined test (FCT) results has emerged as a cost-effective strategy for screening of 31 trisomy 21 (T21). 32 Objectives: To assess performance, patients’ uptake and cost of contingent cfDNA 33 screening and to compare it to that of the stablished FCT. 34 Methods: This is a prospective cohort study including all singleton pregnancies attending 35 to their FCT for screening of T21 at two university hospitals in South Spain. When the 36 FCT risk was 1:50, there were major fetal malformations or the nuchal translucency 37 was ≥3.5mm, women were recommended invasive testing (IT); if the risk was between 38 1:50 and 1:270, women were recommended cfDNA testing and for risks bellow 1:270, 39 no further testing was recommended. Detection rate (DR), false positive rate (FPR), 40 patients’ uptake and associated costs were evaluated. 41 Results: We analyzed 10,541 women, including 46 T21 cases. DR of our contingent 42 strategy was 89.1% (41/46) at 1.4% (146/10,541) FPR. Uptake of cfDNA testing was 43 91.2% (340/373) and overall IT rate was 2.0%. The total cost of our strategy was 44 €1,467,235.7, similar to €1,446,525.7 had cfDNA testing not been available. 45 Conclusions: Contingent cfDNA screening shows high DR, low IT rate and high uptake 46 at a similar cost than traditional screening. 47 48 Key Words. cell-free DNA; contingent screening; Combined test; fetal trisomy; first 49 trimester screening; prenatal diagnosis 50 51 3 INTRODUCTION 52 53 In singleton pregnancies, first-trimester combined test (FCT) for screening of trisomies 54 21, 18 and 13 using a combination of maternal age, fetal nuchal translucency (NT) 55 thickness and serum free β-human chorionic gonadotropin (β-hCG) and pregnancy- 56 associated plasma protein A (PAPP-A) has a detection rate of about 90% for trisomy 21 57 and about 95% for trisomies 18 and 13, at an overall false positive rate (FPR) of 5% [1]. 58 This method of screening is the first-line screening for aneuploidies in many countries in 59 Europe, including Spain. Recently, analysis of cell-free (cf) DNA of maternal blood has 60 been incorporated in clinical practice, providing effective screening for the major 61 trisomies as early as 10 weeks’ gestation [2]. A recent meta-analysis of clinical validation 62 and implementation studies reported that the detection rate (DR) of cfDNA testing for 63 trisomies 21, 18 and 13 are 99.7%, 97.9% and 99.0% respectively, at a combined false 64 positive rate (FPR) of 0.12 [3]. Therefore, since cfDNA testing is highly effective in 65 screening for trisomies and it only involves the simple taking of a maternal blood sample, 66 it could be argued that universal cfDNA screening should be introduced in routine clinical 67 practice. However, such approach is still limited by the higher cost of the test in 68 comparison to the traditional FCT. Over the last few years, several studies have been 69 published assessing the economic impact of different strategies for implementing cfDNA 70 testing in health care systems [4-6]. An alternative strategy to universal screening is to 71 offer cfDNA testing contingent on the results of another method of screening used as 72 first-line screening, preferably the FCT. By this approach, only women at high- and / or 73 intermediate-risk would be offered cfDNA testing and, therefore, it would still be possible 74 to retain the main advantage of the test in terms of early results and high performance, 75 but the cost of such screening program would be considerably lower [7-9]. This strategy 76 would also allow to retain the advantages of the first-trimester combined assessment 77 such as pregnancy dating, early detection of major defects and prediction and potential 78 prevention of a series of pregnancy complications [10]. 79 80 In Spain, like in many countries in Europe, screening for trisomy 21 is carried out by the 81 FCT, both in private and public settings. However, unlike other European countries, 82 screening for trisomy 21 is not yet part of the Spanish National Screening Programs and 83 therefore, there is no regulation to coordinate and monitor it. Recently, the Spanish 84 Society of Gynecology and Obstetrics (Sociedad Española de Ginecología y Obstetricia, 85 SEGO) has updated its guidelines to incorporate cfDNA testing as a screening option, 86 and current recommendation is universal screening by the FCT followed by contingent 87 cfDNA testing for risks between 1 in 50 and 1 in 250 [11]. The aim of this recommendation 88 is to reduce the rate of invasive procedures without modifying DR or increasing the cost. 89 Some Spanish public hospitals have already reported their experience with this approach 90 [9] but, to the best of our knowledge, no economic evaluation after implementation has 91 yet been published. 92 93 The objectives of our study are first, to analyse the influence of implementation of cfDNA 94 contingent screening in the global performance of screening, second, to assess patients’ 95 acceptability of the cfDNA test and third, to evaluate the difference in costs after 96 implementing this new screening strategy. 97 98 99 MATERIALS AND METHODS 100 101 The data from this study derived from prospective screening for trisomy 21 at 11 to 13 102 weeks’ gestation by contingent cfDNA testing on the basis of the results from the FCT. 103 All women with singleton pregnancies attending to their first-trimester hospital visit at one 104 of two university hospitals in South Spain (Hospital Universitario de Valme in Seville and 105 Hospital Juan Ramón Jiménez in Huelva) from March 2016 to March 2018 were 106 4 included. Ethics approval was obtained from the Local Research Ethics Committee 107 (0109-N-16). 108 109 Clinical implementation of cfDNA contingent screening for trisomy 21 110 In the two participating hospitals, the FCT is routinely performed at 11-13 weeks’ 111 gestation. During the first-trimester scan, we confirm number of fetuses, check viability, 112 diagnose major fetal defects, measure crown-rump length (CRL) for pregnancy dating 113 [12] and fetal NT. This measurements are combined with maternal age and maternal 114 serum concentrations of free β-hCG and PAPP-A measured at 9–12 weeks’ to calculate 115 the patient-specific risk for trisomy 21 [1]. If the risk is more than 1 in 270, the mother is 116 explained that her risk for trisomy 21 is low and she is booked for another scan at 19-21 117 weeks’ gestation to examine fetal anatomy. If the risk is between 1 in 50 and 1 in 270, 118 she is classified as high risk and given the options of invasive testing (chorionic villus 119 sampling or amniocentesis) or cfDNA testing. Finally, if the risk is more than 1 in 50 or if 120 there are any major fetal malformation or the fetal NT is ≥ 3.5 mm, the mother is 121 explained that not only the risk of trisomy 21 is increased but also that of other 122 chromosomal and subchromosomal abnormalities and therefore, she is advised to have 123 an invasive test with array analysis (figure 1). 124 125 Women opting for cfDNA testing provided written informed consent and maternal blood 126 (20 mL) was collected into Roche Cell-Free DNA Collection Tubes (Roche, Pleasanton, 127 CA). The tubes were shipped without any processing to the cfDNA laboratory in Madrid, 128 Spain. Targeted cfDNA testing for fetal trisomy was performed using the Harmony® 129 prenatal test. In brief, Harmony® uses digital analysis of selected regions (DANSR) 130 assays targeting sequences on chromosomes 13, 18, and 21 for chromosome 131 quantitation and single nucleotide polymorphisms on chromosomes 1 to 12 for fetal 132 fraction measurement. Products of the DANSR assays are quantified using a custom 133 microarray. The FORTE (Fetal fraction Optimized Risk of Trisomy Evaluation) algorithm 134 is used to include fetal fraction in data analysis and provide patient-specific risk 135 assessments for trisomy [13]. A risk of ≥ 1% is considered to be high probability. In the 136 study sites, women receiving a low-risk result are reassured that trisomies are unlikely 137 and they are booked for anomaly scan at 19-21 weeks. However, women receiving a 138 high-risk result are advised to consider invasive testing for prenatal diagnosis. For the 139 cases where the cfDNA test does not provide results, women are offered a second draw 140 and for those cases without results from second analysis, they are advised to have 141 invasive testing (figure 1). 142 143 Performance of screening 144 DR and FPR with their confidence intervals (CI) were calculated for both, cfDNA 145 contingent screening and traditional FCT. Different cut-offs were explored to estimate 146 performance of the contingent strategy when the group offered cfDNA testing is 147 increased. 148 149 Pregnancy outcome 150 Pregnancy outcome was ascertained by two methods: first, prenatal or postnatal 151 karyotyping and second, neonatal examination by a qualified physician within the first 152 three days of the newborn’s life. Cases raising any suspicion were followed up at least 6 153 months after birth. Cases lost to follow up, including those ending up in miscarriage or 154 stillbirth without karyotyping, were excluded. 155 156 Economic assessment 157 We performed short-term economic analysis including all procedures carried out until 158 delivery. For this analysis we took into account only direct costs, including tests and 159 procedures performed during pregnancy and delivery, as established by the public health 160 5 system of Andalucía, Spain [14], except for the case of the cfDNA test which was 161 externalised to a private laboratory. 162 163 Statistical analysis 164 Descriptive data were expressed as median and interquartile range (IQR) and in 165 proportions (absolute and relative frequencies). Comparisons between treatment groups 166 were performed by Mann-Whitney U-test or two-tailed χ2-test as appropriate. 167 168 The statistical software package SPSS 20.0 (SPSS Inc.,Chicago, IL) was used for data 169 analyses. 170 171 RESULTS 172 173 Study population 174 During the study period a total of 10,794 women attended their first-trimester hospital 175 visit at one of the two participating hospitals. 10,677 (98.5%) of those accepted FCT for 176 screening of trisomy 21, 117 (1.1%) declined screening and 136 were lost to follow up. 177 Maternal and pregnancy characteristics as well as results from FCT and cfDNA test when 178 performed are shown in table 1. In the study population there were 68 (0.62%) 179 chromosomal abnormalities, including 46 cases of trisomy 21 (0.43%). 180 181 Performance of contingent screening 182 Following our strategy, we detected 41 (89.1%; 95% CI: 77.0 to 95.3) of the 46 trisomy 183 21 cases and 58 (85.3%; 95% CI: 75.0 to 91.8) of the 68 cases of other aneuploidies at 184 1.4% (146/10,541; 95% CI: 1.2 to 1.6) invasive testing rate in the first trimester of 185 pregnancy. After including the results from the 19-21 weeks’ anomaly scan, we increased 186 the detection rate for trisomy 21 to 91.3% (42/46; 95% CI: 79.7 to 96.6) and for the other 187 aneuploides diagnosed before or after birth, to 94.1% (64/68; 95% CI: 85.8 to 97.7). 188 Had cfDNA testing not being available, FCT alone would have also detected 89.1% 189 (41/46; 95% CI: 77.0 to 95.3) of the trisomy 21 cases but at 4.3% (457/10,541; 95% CI: 190 4.0 to 4.7) FPR. 191 192 Performance of cfDNA testing alone 193 In total, we carried out 340 cfDNA tests. We did not get a result after the first draw in 17 194 (5.0%) cases but after repeating the test in all 17 cases, only two (0.6%) cases were left 195 without a result. The cfDNA test detected all 15 cases of trisomy 21 with no false 196 positives. 197 198 Women’s preferences on clinical management (table 2) 199 The screening protocol of our hospitals is shown in figure 1. In 148 (1.4%) cases we 200 detected any major fetal malformation, fetal NT was ≥ 3.5 mm or the FCT risk was ≥ 1 in 201 50. In this first group, there were 43 aneuploidies, including 26 cases of trisomy 21. 145 202 (98.0%) women chose to have an invasive test but 3 (2.0%) opted against it and had 203 cfDNA testing instead. Among the women having invasive testing, there was one 204 miscarriage at 16 weeks’ gestation. In 373 (3.5%) cases, the FCT risk was between 1 in 205 50 and 1 in 270 without major fetal malformations or increased NT. In this second group, 206 there were 15 cases of trisomy 21. 340 (91.2%) women in this group chose to have 207 cfDNA testing, 30 (8.0%) chose to have invasive testing and 3 (0.8%) women decided 208 not to have any further testing. In this group, there were 2 miscarriages at 13 and 20 209 weeks’ respectively. In 10,156 (95.1%) cases, the FCT risk was less than 1 in 270 without 210 major fetal malformations or increased NT. In this third group, there were 10 211 aneuploidies, including five cases of trisomy 21 (one case of spontaneous miscarriage). 212 In total, we performed 210 (2.0%) invasive procedures, 145 in the first group, 30 in the 213 second group and 35 in the third one. 214 6 Of the 17 cases that did not receive a result after the first attempt, all women decided to 215 repeat the test. The two women that did not get a result from the second draw decided 216 not to do more studies and follow the usual pregnancy care. 217 218 Cost analysis 219 220 Cost of screening of trisomy 21 by our cfDNA contingent strategy was estimated in 221 €1,467,235.7 and cost / effectiveness was estimated in €22,925.5 (table 3). If cfDNA 222 testing had not been available, the estimated cost and cost/effectiveness would have 223 been similar (€1,446,525.7 and €22,601.9 respectively). Therefore, implementation of 224 cfDNA testing contingently after the FCT only resulted in a marginal 1.4% increase in the 225 total cost of the program. 226 227 Performance of different strategies of contingent cfDNA testing 228 We finally evaluated performance and associated costs of contingent screening at 229 different cut-offs from FCT and results are reported in table 4. 230 231 DISCUSSION 232 233 Main findings of the study 234 In this study we found that first, within our public health system, a strategy in which cfDNA 235 testing is implemented contingently after the FCT is accepted by 91.2% of the women; 236 second, our contingent strategy allows to reduce the invasive testing rate from 4.2% to 237 1.4% for the same DR of about 90%; and third, this strategy can be implemented at a 238 similar cost than traditional screening. 239 240 Comparison with previous studies 241 Our results are consistent with those from previous studies, which showed that 242 contingent screening of aneuploidies by FCT and cfDNA test is feasible and well 243 accepted by patients [7-9,15,16]. The first study reporting on the performance of this 244 contingent strategy used a cut-off of 1 in 2500 from the FCT to offer cfDNA testing [15]. 245 The authors reported that, although this cut-off could potentially increase the DR up to 246 97% for trisomy 21 and up to 95% for trisomies 18 and 13, it would require that about 247 24% of the screened population had cfDNA testing [17]. Similarly, had we offered cfDNA 248 testing to women with a risk of 1 in 2,500 or more, we would have detected 97.8% of the 249 cases of trisomy 21 and 98.5% of the other aneuploidies by performing the test in about 250 26% of our population. In contrast, the SEGO proposal aims to ensure a DR of about 251 90% but only about 4% of the women to require cfDNA testing, as shown in our study. 252 The main advantage of this strategy is the secondary reduction of invasive tests at a 253 similar cost. During the study period, our invasive testing rate was 2.0% (210/10,541), 254 which is considerably lower than our previously reported rate of 4.8% in 2005 to 2010 (p 255 < 0.0001) [18]. Another Spanish study conducted in a public hospital in Madrid region, 256 reported only 75% uptake of cfDNA testing within the high-risk group, defined as a risk 257 of ≥ 1 in 250 at the time of screening [9]. However, this uptake increased from 8% in the 258 very high-risk group (risks ≥ 1 in 10) to 100% in the less high-risk group (risks between 259 1 in 150 and 1 in 250), and the uptake of cfDNA testing in the women whose risk was 260 between 1 in 50 and 1 in 250, was about 90% like in the present study [9]. 261 262 Strengths and weaknesses of the study 263 The main strength of our analysis is the use of real clinical data, collected in fully funded 264 public hospitals. Thus, these results reflect women’s behaviour in real life regarding 265 uptake of trisomy 21 screening, cfDNA testing and invasive testing regardless of 266 economic status and therefore, lead to real inputs for our model. However, although it 267 was not the aim of our study, the small number of affected pregnancies included did not 268 allow us to accurately assess the performance of neither the FCT nor the cfDNA test. 269 7 Another limitation is that we only assessed short-term costs for economic evaluation, 270 acknowledging that indirect costs, although difficult to quantify, are also of great 271 importance. Additionally, we have not taken into account the costs related to personnel 272 involved but we believe that both, indirect costs and personnel costs, would be higher in 273 the strategy without cfDNA testing; first, because sick leave is more likely to happen after 274 invasive testing than after cfDNA testing, second, because the cost of one or even two 275 Fetal Medicine specialists performing an invasive procedure is higher than that of a nurse 276 drawing blood for cfDNA analysis. 277 278 Interpretation 279 Essentially, there are two options for clinical implementation of cfDNA testing in 280 screening of the major trisomies: first, universal screening and second, contingent 281 screening based on the results of first-line screening by another method. Universal 282 screening would definitely lead to the best performance. However, the high marginal cost 283 associated leaves this strategy out for most public health systems. Therefore, introducing 284 cfDNA testing in a contingent fashion seems to be a reasonable alternative. Following 285 cfDNA testing, the most accurate method for screening of trisomy 21 is the FCT and the 286 results from our study prove that, only having a good-quality first-trimester scan and FCT, 287 we can ensure high performance of any contingent screening proposal and keep the 288 costs as previously determined. When the cost of the test decreases, current cut-offs 289 may be replaced by lower ones and the proportion of women opting for cfDNA testing 290 may be expanded; however continuous audit and monitoring of performance and costs 291 is necessary to keep them stable. 292 293 Conclusions 294 First, clinical implementation of contingent cfDNA screening following a high-risk result 295 from the FCT as recommended by the Spanish Society of Obstetrics and Gynecology is 296 feasible and shows similar DR and costs and lower invasive testing rate than traditional 297 screening. Second, patients’ uptake of such strategy is high. Third, expanding the group 298 of patients eligible for cfDNA testing would increase the DR but at the expense of an 299 increase in the total cost of the program. 300 301 Acknowledgement 302 We thank Prof. A. Fernández-Palacín, Department of Preventive Medicine and Public 303 Health, University of Seville, Spain, for helping in the statistical evaluation 304 Statement of Ethics 305 Ethical approval was given by the Biomedical Ethics Committee of the Junta of Andalucía 306 (01-N-16), Spain. Approval date: January 31, 2016. 307 308 Disclosure Statement 309 The authors declare that they have no conflicts of interest. 310 311 Funding Sources 312 None 313 314 Contribution to authorship 315 JAS, RT, IP, RG, MV, PC, JAGM participated in protocol and study development, data 316 collection, data analysis and manuscript writing and approval. BS and MMG participated 317 in data analysis and manuscript writing and approval. 318 319 8 REFERENCES 320 321 1. Wright D, Syngelaki A, Bradbury I, Akolekar R, Nicolaides KH. First-trimester 322 screening for trisomies 21, 18 and 13 by ultrasound and biochemical testing. Fetal 323 Diagn Ther 2014; 35: 118-126. 324 2. Gil MM, Quezada MS, Bregant B, Ferraro M, Nicolaides KH. Implementation of 325 maternal blood cell-free DNA testing in early screening for aneuploidies. Ultrasound 326 Obstet Gynecol 2013; 42: 34-40. 327 3. Gil MM, Accurti V, Santacruz B, Plana MN, Nicolaides KH. Analysis of cell-free DNA 328 in maternal blood in screening for aneuploidies: updated meta-analysis. Ultrasound 329 Obstet Gynecol 2017; 50: 302-314. 330 4. Nshimyumukiza L, Menon S, Hina H, Rousseau F, Reinharz D. Cell-free DNA 331 noninvasive prenatal screening for aneuploidy versus conventional screening: A 332 systematic review of economic evaluations. Clin Genet 2018; 94:3-21. 333 5. Bianchi DW, Chiu RWK. Sequencing of Circulating Cell-free DNA during Pregnancy. 334 N Engl J Med 2018; 379: 464-473. 335 6. Kostenko E, Chantraine F, Vandeweyer K, Schmid M, Lefevre A, Hertz D , Zelle L, 336 Bartha JL, Di Renzo GC. Clinical and Economic Impact of Adopting Noninvasive 337 Prenatal Testing as a Primary Screening Method for Fetal Aneuploidies in the 338 General Pregnancy Population. Fetal Diagn Ther 2018:1-11. 339 7. Chitty LS, Wright D, Hill M, Verhoef TI, Daley R, Lewis C, Mason S, McKay F, 340 Jenkins L, Howarth A, Cameron L, McEwan A, Fisher J, Kroese M, Morris S. Uptake, 341 outcomes, and costs of implementing non-invasive prenatal testing for Down's 342 syndrome into NHS maternity care: prospective cohort study in eight diverse 343 maternity units. BMJ 2016; 354: i3426. 344 8. Gil MM, Revello R, Poon LC, et al. Clinical implementation of routine screening for 345 fetal trisomies in the UK NHS: cell-free DNA test contingent on results from first- 346 trimester combined test. Ultrasound Obstet Gynecol 2016; 47: 45–52. 347 9. Gil MM, Brik M, Casanova C, Martin-Alonso R, Verdejo M, Ramírez E, Santacruz B. 348 Screening for trisomies 21 and 18 in a Spanish public hospital: from the combined 349 test to the cell-free DNA test. J Matern Fetal Neonatal Med 2017; 30: 2476-2482. 350 10. Nicolaides KH. Turning the pyramid of prenatal care. Fetal diagnosis and therapy 351 2011; 29: 183 – 196. 352 11. Sociedad Española de Ginecología y Obstetricia 2017. Guía de asistencia práctica: 353 Cribado y diagnóstico precoz de anomalías genéticas. [Internet]. [accesed 2019 Apr 354 1]. Available from: https://sego.es/ 355 12. Robinson HP, Fleming JE. A critical evaluation of sonar “crown-rump length” 356 measurements. Br J Obstet Gynaecol 1975; 82: 702–710. 357 13. Juneau K, Bogard PE, Huang S, Mohseni M, Wang ET, Ryvkin P, Kingsley C, 358 Struble CA, Oliphant A, Zahn JM. Microarray-based cell-free DNA analysis improves 359 noninvasive prenatal testing. Fetal Diagn Ther 2014; 36: 282-286. 360 14. Corrección de errores de la Orden de 8 de mayo de 2018, por la que se modifica la 361 Orden de 14 de octubre de 2005, por la que se fijan los precios públicos de los 362 servicios sanitarios prestados por centros dependientes del Sistema Sanitario 363 Público de Andalucía. (BOJA núm. 92, de 15.5.2018). [Internet]. [accesed 2019 Jan 364 14]. Available from: 365 http://www.juntadeandalucia.es/servicioandaluzdesalud/ordenpreciospublicos/defa 366 ult.asp. 367 15. Gil MM, Giunta G, Macalli EA, Poon LC, Nicolaides KH. UK NHS pilot study on cell- 368 free DNA testing in screening for fetal trisomies: factors affecting uptake. Ultrasound 369 Obstet Gynecol 2015; 45: 67-73. 370 16. Miltoft CB, Rode L, Tabor A. Positive view and increased likely uptake of follow-up 371 testing with analysis of cell-free fetal DNA as alternative to invasive testing among 372 Danish pregnant women. Acta Obstet Gynecol Scand 2018; 97: 577-586. 373 9 17. Nicolaides KH, Syngelaki A, Poon LC, Gil MM, Wright D. First-trimester contingent 374 screening for trisomies 21, 18 and 13 by biomarkers and maternal blood cell-free 375 DNA testing. Fetal Diagn Ther 2014; 35: 185-192. 376 18. Sainz JA, Peral I, Borrero C, Almeida C, Moro A, Turmo E. Stepwise sequential 377 screening for Down's syndrome (combined test associated with modified genetic 378 sonography) in pregnant women with low risk for chromosomal disorders. J Perinat 379 Med 2012; 40: 647-651. 380