- Top of page
- Subjects And Methods
Chromosome-selective sequencing of loci from chromosomes 21 and 18 in maternal plasma cell-free DNA (cfDNA) has been successfully applied in non-invasive prenatal testing (NIPT) for fetal trisomies 21 and 18[1-5]. In pregnancies with a trisomic fetus, the extra cfDNA molecules derived from the extra fetal chromosome can be detected, given the higher proportion of molecules relative to a reference disomic chromosome.
Sparks et al. used a training set of euploid and trisomic pregnancies to perform selective sequencing of cfDNA and develop a novel algorithm for the estimation of individualized trisomy risk. They subsequently applied this approach for assessment of a blinded validation set and correctly discriminated the 36 cases of trisomy 21 and eight cases of trisomy 18 from the 123 euploid cases. Ashoor et al. performed a nested case–control study of cfDNA in plasma obtained at 11–13 weeks' gestation before chorionic villus sampling (CVS) from 300 euploid pregnancies, 50 pregnancies with trisomy 21 and 50 pregnancies with trisomy 18. Chromosome-selective sequencing correctly detected all cases of trisomy 21 and 49 (98.0%) of the cases of trisomy 18, with a false-positive rate (FPR) of 0%. Norton et al. performed chromosome-selective sequencing on chromosomes 21 and 18 in a multicenter cohort of high-risk pregnancies at 10–39 (mean, 17) weeks' gestation. They correctly detected all 81 cases of trisomy 21 with an FPR of 0.03% (1/2888 normal cases) and detected 37 (97.4%) of the 38 cases of trisomy 18 with an FPR of 0.07% (2/2888). More recently, the chromosome-selective sequencing approach was applied for NIPT in pregnant women undergoing routine screening for aneuploidies at 11–13 weeks' gestation. All eight cases of trisomy 21, and two cases of trisomy 18, had a trisomy risk score of > 99%, whereas the risk score for trisomy 21 was < 1% in all 1939 euploid pregnancies (FPR, 0%) and the risk score for trisomy 18 was < 1% in 1937 (FPR, 0.1%).
The objective of this study was to assess the performance of selective sequencing of cfDNA in maternal plasma for the prenatal detection of fetal trisomy 13.
- Top of page
- Subjects And Methods
This two-phase case–control study shows that chromosome-selective sequencing of cfDNA in maternal plasma can detect the majority of trisomy 13 pregnancies with an FPR of 0.05%. The first phase of the study was used to optimize the trisomy 13 algorithm, which was then applied via a blinded analysis to a second set of cases. In eight of the 10 cases of trisomy 13, the estimated risk for this aneuploidy was > 99%, whereas in 99.9% of the euploid cases the risk score for trisomy 13 was ≤ 0.01%.
The detection rate of trisomy 13 by chromosome-selective sequencing of maternal plasma cfDNA is lower than that for the detection of trisomies 21 and 18. Previous studies utilizing DANSR and FORTE, both in high-risk and routinely screened populations, reported that the detection rates of trisomy 21 and 18 were 100 and 97–98%, respectively, with FPRs of 0.1% or less[2-5].
Three previous studies, utilizing non-selective massively parallel shotgun sequencing, have examined NIPT for trisomy 13 from analysis of cfDNA in maternal plasma. Chen et al. examined 25 trisomy 13 pregnancies and reported that with the use of a previously standardized z-score method developed for trisomy 21, they detected nine (36.0%) of the cases of trisomy 13 at an FPR of 7.6%. After adjustments, including removal of repeat-masking and correction for the GC content bias in the sequencing data, the detection rate increased to 100% and the FPR decreased to 1.1%. Palomaki et al. used a similar approach and detected 11 of 12 (91.7%) cases of trisomy 13 at an FPR of 0.9%. Bianchi et al. examined 16 cases of trisomy 13 and with a z-score cut-off of 2.5 the detection rate and FPR were 81.2% (13/16) and 0%, respectively, whereas at a z-score cut-off of 4.0 the detection rate was 68.8% (11/16) at an FPR of 0%[10, 11].
One possible explanation for the observed lower detection rate of trisomy 13 with NIPT, compared with trisomy 21, is that in pregnancies with trisomy 13 fetuses there is placenta-confined mosaicism with a high proportion of cells being disomic, whereas in trisomy 21 all placental cells are invariably trisomic. As the primary source of cfDNA in the maternal circulation is thought to be the placenta, a mosaic placenta that is predominantly disomic would lead to a false-negative result when using NIPT for the detection of trisomy 13. Likewise, a mosaic placenta that is predominantly trisomic in which the fetus is euploid could lead to a false-positive result.
At 11–13 weeks' gestation, the relative prevalences of trisomies 18 and 13 to trisomy 21 are 1:3 and 1:7, respectively[14-16]. However, in trisomies 18 and 13 the rate of spontaneous abortion or fetal death between 12 and 40 weeks' gestation is about 80%, therefore the relative prevalence of these trisomies to trisomy 21 at birth are 1:12 and 1:28, respectively.
Effective first-trimester screening for trisomy 21 is provided by a combination of maternal age, fetal nuchal translucency (NT) thickness and maternal serum free β-hCG and PAPP-A, with a detection rate of more than 90% for an FPR of 5%. A beneficial consequence of screening for trisomy 21 is the early diagnosis of about 75% of cases of trisomies 18 and 13, which are the second and third most common chromosomal abnormalities. All three trisomies are associated with increased maternal age, increased fetal NT and decreased maternal serum PAPP-A, but in trisomy 21 serum free β-hCG is increased, while in trisomies 18 and 13 it is decreased[6, 7, 17-19]. When specific algorithms for trisomies 18 and 13 in addition to the one for trisomy 21 are also used, about 90% of fetuses with trisomy 21 and 95% of those with trisomies 13 and 18 can be detected for an overall increase in FPR of only 0.1%. This increasing FPR has generally been considered acceptable despite the low prevalence of trisomies 13 and 18.
Trisomy 13 is often associated with abnormalities that can be readily identified by ultrasonography, not only in the second but also in the first trimester[20, 21]. A study of 181 fetuses with trisomy 13 reported that at the 11–13-week scan 92 (50.8%) had holoprosencephaly, exomphalos and/or megacystis and in 129 (71.3%) the heart rate was above the 95th percentile. Another study of holoprosencephaly, exomphalos and megacystis detected at 11–13 weeks reported that these defects are associated with a high incidence of chromosomal abnormalities, mainly trisomies 18 and 13, found in about 65, 55 and 30% of cases, respectively.
The performance of NIPT in screening for both trisomies 21 and 18 is far superior to that of currently available screening methods, with a substantial increase in detection rate and decrease in FPR[5, 6]. A study of NIPT with a chromosome-selective sequencing approach in pregnant women undergoing routine screening for aneuploidies at 11–13 weeks' gestation reported that the estimated trisomy risk score was > 99% in all cases of trisomy 21 and trisomy 18 and < 1% in 99.9% of the euploid cases.
This study has shown that NIPT is also useful in screening for trisomy 13. Although the total number of cases of trisomy 13 examined is too small for accurate assessment of the detection rate, the FPR was only 0.05%. Consequently, if the NIPT result is positive for trisomy 13 there is a 1600-fold (80/0.05) increase in risk for this trisomy, therefore such patients should be offered the option of invasive diagnostic testing. If the NIPT result is negative for trisomy 13 there is a 5-fold (99.95/20) decrease in the a-priori risk. In NIPT there is a delay of 1–2 weeks between sampling and obtaining results. The blood sample could be collected at 9–10 weeks' gestation so that the results would be available by 12 weeks, which is the best time to carry out the first-trimester ultrasound examination. If the 12-week scan demonstrates holoprosencephaly, exomphalos or megacystis, where the risk for aneuploidies is very high, the 5-fold reduction in risk following a negative NIPT test is unlikely to reassure the parents, and they should still be offered invasive testing. Since the prevalence of these defects is less than 0.1%, the effect on the overall proportion of pregnancies requiring an invasive test would be minimal.
The sensitivity and specificity of NIPT for trisomies 21, 18 and 13 is not 100%, therefore NIPT should not be considered a diagnostic test to replace invasive testing in high-risk pregnancies. It is a new high-performance screening test that identifies a high-risk group requiring further investigation by invasive testing. Similarly, the introduction of NIPT in universal screening for trisomies 21, 18 and 13 will complement rather than replace the 11–13-week scan. Not only is the latter useful in screening for aneuploidies but it is also a diagnostic test for many major fetal defects, some of which, such as holoprosencephaly, require further investigation by invasive testing; additionally, in combination with biochemical and other biophysical markers, the scan can provide effective early screening for pregnancy complications, including pre-eclampsia and preterm birth[22-25].
Trisomy 13 is associated with a high rate of fetal death, and the prevalence in live births is about 30 times lower than that of trisomy 21. Additionally, unlike individuals with trisomy 21, who can survive for more than 60 years, individuals with trisomy 13 rarely survive beyond the first few months following birth. Chromosome-selective sequencing of cfDNA can detect the majority of cases of trisomy 13 with an FPR of less than 0.1%, but the detection rate is lower than that reported for trisomies 21 and 18.