LAUSR

STUDY QUESTION Does ART-based conception influence fetal fraction (FF) estimation and cell-free fetal DNA (cffDNA) testing performance? SUMMARY ANSWER Mode of conception (ART versus natural) does not impact FF estimation or cffDNA test informativity rates. WHAT IS KNOWN ALREADY Pregnancies achieved via ART are increasing, and cffDNA testing is displacing traditional prenatal screening methods due to its high sensitivity and specificity and noninvasive nature. However, conflicting data exist on cffDNA testing performance and FF in ART pregnancies compared with natural pregnancies. STUDY DESIGN, SIZE, DURATION We performed a case-control study that included 21 558 consecutive pregnancies (spontaneous, n = 15 707; ART, n = 5851). ART-conceived pregnancies were stratified into two groups according to oocyte origin. Samples were collected from April 2015 to September 2019. PARTICIPANTS/MATERIALS, SETTING, METHODS The study included women from different centers worldwide. Blood samples were drawn from the 10th week of gestation onward. Massive parallel whole-genome sequencing was used to analyze cffDNA content in blood plasma. Two different types of technologies (single-end and paired-end) were applied because of analysis technology changes made by the sequencing provider over time. FF was determined using different methods depending on the type of technology used. Cases with an FF <2% or with failure in any quality control metrics were classified as noninformative. An analysis of covariance model was selected to identify which qualitative (sequencing methodology, mode of conception, type (i.e. multiplicity) of gestation and age (women >35 or <35 years old)) and quantitative (gestational age, BMI) variables were predictors of FF value. Multinomial logistic regression was used to evaluate whether the mode of conception impacted cffDNA testing performance. MAIN RESULTS AND THE ROLE OF CHANCE A univariate t-test demonstrated no significant differences in FF values between ART (median FF = 9.2%) and spontaneous pregnancies (median FF = 9.2%). Also, a multivariate analysis showed that the mode of conception, did not strongly impact the percentage of FF. ART-treated women showed a lower incidence of high-risk cffDNA results compared to women who conceived naturally, specifically for trisomy (T)21 (0.7% versus 1.3%, P = 0.001) and T18 (0.1% versus 0.3%, P = 0.001). A multivariate model stratified by type of aneuploidy suggested that these differences were conditioned by oocyte origin, especially for the T21 risk classification (P < 0.0001). False-positive rates (FPRs) were significantly higher in the ART population, mainly for T13 (P = 0.001) and sexual chromosome aneuploidies (SCAs; P < 0.001). A multivariate model suggested that the differences observed in SCAs were caused by sequencing modality rather than by mode of conception. Likewise, ART-treated women who used their own oocytes had a higher probability of a false positive for T13 (P = 0.004). LIMITATIONS, REASONS FOR CAUTION Our study lacks follow-up data for low- and high-risk cases of both ART-conceived and naturally conceived pregnancies. Therefore, the results comparing FPR in both populations should be interpreted carefully. Also, collecting information about different ART modalities and regarding preimplantation genetic testing for aneuploidy treatments would help draw definite explanations for the trends observed in this study. WIDER IMPLICATIONS OF THE FINDINGS This is the first study that demonstrates, with a large sample size, that FF is not influenced by mode of conception, demystifying the notion that patients undergoing ART have a higher probability of noninformative cffDNA testing results. Multivariate models stratified by oocyte origin and type of aneuploidy demonstrated that ART-conceived pregnancies do not have a higher probability of classification as a high-risk pregnancy in prenatal testing. This information is especially valuable to clinicians and genetic counselors when informing patients about the risks and limitations of cffDNA testing in ART pregnancies. STUDY FUNDING/COMPETING INTEREST(S) This study was financially supported by Igenomix Lab S.L.U. All authors declare no conflict of interest. TRIAL REGISTRATION NUMBER N/A.