How big is the time window for cell-free fetal DNA testing after pregnancy loss and which factors are associated with a successful result?

Abstract

Study question: How fast does cell-free fetal DNA (cffDNA) decline after early pregnancy loss and which factors affect the decline?

Summary answer: After pregnancy loss cffDNA declines gradually with detectable levels persisting up to 3 days post-tissue passage correlating with β-hCG decline.

What is known already: Postpartum clearance of cffDNA occurs within hours, but little is known about its decline following early pregnancy loss. Initial results from the Copenhagen Pregnancy Loss (COPL) study showed slower clearance in relation to pregnancy loss with detectable levels found up to 24 h after tissue passage.

Study design, size, duration: This prospective cohort study included 1463 women from the COPL cohort, enrolled between 12 November 2020 and 19 December 2022. Participants were divided into three groups based on sampling time: the standard group (samples collected before tissue passage), the delayed sample group (samples collected after tissue passage at maximum 24 h), and the repeated sample group (samples collected at multiple time points after medical and surgical treatment).

Participants/materials, setting, methods: Eligible women were 18 years old with a confirmed intrauterine pregnancy loss before 22 gestational weeks. Exclusion criteria included ectopic, molar, or unknown-location pregnancies and inability to consent. For the repeated sample group, additional exclusions were vaginal bleeding at diagnosis, anembryonic pregnancies, and opting for expectant management. Blood samples were analyzed for β-hCG and cffDNA, with fetal fraction measured using the sequencing-based fetal fraction (SeqFF) method. In the repeated sampling group, analyses were performed on Days 2 and 3 for surgically treated and Days 7 and 14 for medically treated.

Main results and the role of chance: After pregnancy tissue passage, both cffDNA and β-hCG levels declined consistently over time with a corresponding increase in no-call rates. The decline in SeqFF following pregnancy loss occurred more gradually than the immediate clearance previously reported after delivery, remaining measurable for up to 3 days after pregnancy loss, though 30% of samples became inconclusive at this timepoint. The fetal fraction of cffDNA was highest when tissue remained in utero and declined significantly beyond 12-h post-passage (median 4.7% <6 h vs 2.8% >12 h). The no call rate was 9.1% when the tissue was still in situ but increased to 27.1% in the 12- to 24-h post-passage group. Higher β-hCG levels correlated with increased odds of a conclusive cffDNA test (OR 1.30, 95% CI: 1.20-1.43, P < 0.001). β-hCG is accounting for 20% of the variability in fetal fraction measurements. Variability in the decline of cffDNA between groups reflects the differences in the timing of sampling and treatment approaches.

Limitations, reasons for caution: The small size of the repeated sample group may limit the generalizability of the findings. Reliance on the SeqFF method introduces variability, as it may not perform consistently. Adjustment for key confounders including BMI, gestational age, type of pregnancy loss, and timing of sampling was performed, but residual confounding cannot be ruled out.

Wider implications of the findings: As a non-invasive method, cffDNA testing offers critical genetic insights for couples facing time-sensitive reproductive decisions or those with recurrent pregnancy loss. The correlation between β-hCG and fetal fraction suggests β-hCG screening could optimize sequencing decisions. These findings support refining cffDNA diagnostics to enhance pregnancy loss evaluation and reproductive care.

Study funding/competing interest(s): The study was funded by the Ole Kirks Foundation, the BioInnovation Institute Foundation (BII21SG1020554, NNF20SA0066125, and NNF15OC0016662), the Novo Nordisk Foundation (NNF22OC0077221 and NNF23OC0087269), and the A.P. Møller Foundation. The authors have declared their COI.

Trial registration number: N/A.