Noninvasive Prenatal Testing Test Versus Chorionic Villus Sampling, Where Are We Now?

Introduction

The discovery of cell-free fetal DNA (cff-DNA), also known as noninvasive prenatal testing (NIPT), by Lo et al. in 1997,¹ has significantly changed the screening for common chromosomal abnormalities. Its widespread acceptance globally underscores its impact. NIPT analyzes fetal DNA originating from the apoptosis of the outer cytotrophoblast, the fetal hematopoietic system, and lysis of other fetal cells. The DNA fraction increases from 10% to 15% between 10 and 20 gestational weeks. Rapid cff-DNA clearance occurs in maternal circulation, spanning from 4 minutes to 2 weeks postpartum.

NIPT has recently garnered recognition as a valuable prenatal screening tool from several professional societies.^2–5 Nevertheless, these societies emphasize the significance of advocating for invasive procedures, such as chorionic villus sampling (CVS) and amniocentesis, particularly in cases of high risk of NIPT results, structural abnormalities, or in instances where guidance from a genetic counselor is involved.

While CVS remains regarded as one of the gold standards for making a prenatal diagnosis, concerns arise due to the presence of placenta mosaicism and localized sampling, casting shadows on the test’s accuracy. Increasing case reports detailing the inconsistencies between CVS and amniocentesis^6,7 prompt questions about the test’s implementation in clinical practice. Given that the testing targets of NIPT and CVS originate from DNA fragments of placental origin, considerations arise regarding the potential future replacement of CVS by NIPT and whether CVS remains the preferred diagnostic method in high-risk NIPT cases, considering its historically high accuracy rate.

Performance comparison between noninvasive prenatal testing and chorionic villus sampling

With advancements in massively parallel shotgun sequencing, chromosome selective sequencing, and single nucleotide polymorphisms analysis, NIPT has been clinically implemented to screen for chromosomal abnormalities, copy-number variants (CNVs) and microdeletions.⁸

Numerous systematic reviews have established its accuracy in detecting common chromosomal abnormalities, such as trisomy 21, 18, and 13. The reported sensitivity of T21 is as high as 99%, while the sensitivity for T18 and T13 is reported at 98% and 99% at a combined false-positive rate of 0.13%.⁹ Microarray-based cff-DNA testing demonstrates sensitivity and specificity for T21, T18, and T13 close to 100%.¹⁰ Even in twin pregnancies, NIPT exhibits a pooled sensitivity and specificity for trisomy 21 of 99% and 100%, respectively.¹¹

The performance of NIPT on sex chromosomal aneuploidies (SCAs) varies. The detection of 45, X achieves a sensitivity of 93.9% and a specificity of 99.6%. In contrast, the sensitivities and specificities for other SCAs, such as 47, XXX, 47, XXY, and 47, XYY, range around 99.5% and 99.9%, 100% and 100%, and 91.7% and 100%, respectively.¹² In addition to its high detection of common chromosomal abnormalities, NIPT shows promise in detecting copy number variations and microdeletions.¹³ Neofytou et al.¹⁴ and Koumbariet et al.¹⁵ identified all microdeletions correctly without any false-negative events, exhibiting 100% sensitivity and 100% specificity. As for inherited monogenic diseases, genome-wide cfDNA haplotyping demonstrates high consistency. Xu et al. first reported an accuracy of 99.98% for Duchenne muscular dystrophy with the haplotype-based approach for NIPT, consistent with invasive procedures.¹⁶

The no-call rate is considered a drawback of NIPT, which ranges from 1.9% to 6.3%. However, a new methodology using rolling circle replication has reduced the no-call rate to 0.07% with an overall sensitivity of 98% and overall specificity of 99% for trisomy 21, 18, and 13.¹⁷ A no-call result may indicate a higher chance of atypical chromosomal abnormalities, adverse pregnancy outcomes, and maternal neoplasia such as lymphoma.

Positive predictive values (PPVs) for different types of aneuploidies vary due to their population-based nature. For example, PPVs for T21 could be 84%–99.2%, whereas the PPVs for T18 ranged from 45% to 82%, and PPVs for T13 are less than 50%.¹⁸ Inconsistently, higher PPVs reaching 100% were also reported in T18¹⁹ and T13²⁰ due to selected populations, sequence-read depth, algorism, and so on. Apart from placenta mosaicism, potential causes of false-positive NIPT results include vanishing twin syndrome and maternal malignancies.²¹

Cell-based NIPT (cb-NIPT), a promising new method, has become attractive because of its advantage in distinguishing fetal cells from maternal cells. This new technology is clinically implemented to detect aneuploidies, unbalanced translocation, sub-chromosomal deletions and duplications, copy number variations (CVS), and monogenic disorders. Hatt et al. reported that cb-NIPT detected all genetic aberrations (32/32) found in CVS with no false-positive results, including trisomy 13, 18, and 21 (23/23), pathogenic CNVs (6/6), and sex chromosome aberrations (3/3).²² In addition to isolating the trophoblast, several studies have reported isolating fetal nucleated blood cells (FNBCs) from maternal circulation. Testing FNBC improves cb-NIPT’s diagnostic capacity and avoids confined placental mosaicism (CPM).

Despite being recognized as a diagnostic procedure, CVS carries a risk of false-positive detection due to confined placenta mosaicism, with a reported incidence of 1%–2%. Malvestiti et al.’s study described 1317 (2.18%) mosaic cases among 60,347 CVS, and only 13% of these cases were confirmed as true fetal mosaicism.²³ Approximately 3%–4% of CVS cases needed further amniotic investigation, with respective mosaicism rates of 1.6%–2.0% in trisomy 21, 3.2%–4.0% in trisomy 18, and 8.3%–22.0% in trisomy 13.^24,25 Following a high-risk NIPT, Reilly et al. exhibited the CVS mosaicism rates for T21, T18, T13, and 45, X as 2%, 4%, 22%, and 59%, respectively, with genuine fetal mosaicism rates confirmed by amniocentesis being 44%, 14%, 4%, and 26% accordingly.²⁶ Van Opstal et al. demonstrated that NIPT, compared with CVS, is more sensitive for detecting CPM involving the cytotrophoblast restricted to a part of the placenta.²⁷ In this study, four cases of CPM were detected by NIPT with normal CVS results.

In a cross-sectional questionnaire study on patients’ choices and opinions on CVS and NIPT after undergoing in vitro fertilization and preimplantation genetic testing, 89.2% of respondents affected by hereditary disorders opted for NIPT instead of CVS.²⁸ Although the whole-genome cf-NIPT and cb-NIPT are still under further exploration, they have demonstrated vast and promising potential to replace CVS.

In clinical scenarios where diagnostic tests should be offered after receiving high-risk NIPT results, the question remains: should CVS or amniocentesis be offered? Mardt et al. recommended proceeding with CVS and examining all cell lines. Amniocentesis was recommended in cases where any of the cell lines demonstrated mosaicism.²⁹ In contrast, Okoror et al. proposed using amniocentesis over CVS, especially when no structural anomalies were noted.³⁰ Scott et al. also supported the importance of late first-trimester ultrasound (LFTU) in favoring amniocentesis rather than CVS in the context of high-risk NIPT.³¹

Conclusion

Since both CVS and NIPT receive the material from the placenta, can NIPT truly serve as a substitute for CVS? Advancements in NIPT technology present the potential to challenge the role of CVS in detecting fetal chromosomal abnormalities. Its efficacy in diagnosing T21 has already been acknowledged and continues to evolve, especially with the advent of cell-based technology. This not only enhances screening capabilities but also offers the potential for diagnosing specific chromosomal abnormalities in fetuses during the early stages of pregnancy.

Conflicts of Interest

None.