SNP-based non-invasive prenatal testing detects sex chromosome aneuploidies with high accuracy

Abstract

Objective: This study aimed to develop a single-nucleotide polymorphism-based and informatics-based non-invasive prenatal test that detects sex chromosome aneuploidies early in pregnancy.

Methods: Sixteen aneuploid samples, including thirteen 45,X, two 47,XXY, and one 47,XYY, along with 185 euploid controls, were analyzed. Cell-free DNA was isolated from maternal plasma, amplified in a single multiplex polymerase chain reaction assay that targeted 19,488 polymorphic loci covering chromosomes 13, 18, 21, X, and Y, and sequenced. Sequencing results were analyzed using a Bayesian-based maximum likelihood statistical method to determine copy number of interrogated chromosomes, calculating sample-specific accuracies.

Results: Of the samples that passed a stringent quality control metric (93%), the algorithm correctly identified copy number at all five chromosomes in all but one of the 187 samples, for 934/935 correct calls as early as 9.4 weeks of gestation. We detected 45,X with 91.7% sensitivity (CI: 61.5-99.8%) and 100% specificity (CI: 97.9-100%), and 47,XXY and 47,XYY. The average calculated accuracy was 99.78%.

Conclusion: This method non-invasively detected 45,X, 47,XXY, and 47,XYY fetuses from cell-free DNA isolated from maternal plasma with high calculated accuracies and thus offers a non-invasive method with the potential to function as a routine screen allowing for early prenatal detection of rarely diagnosed yet commonly occurring sex aneuploidies.

Figure 1

Histogram of samples stratified by fetal fraction (Figure 1A) and gestational age (Figure 1B).

Figure 1

Histogram of samples stratified by fetal fraction (Figure 1A) and gestational age (Figure 1B).

Figure 2

Graphical representation of sequencing data obtained from one euploid (Figure 2A), one Monosomy X (Figure 2B), and one 47,XXY (Klinefelter syndrome, Figure 2C) sample. All interrogated SNPs are assumed to be dimorphic and are designated as A and B for simplicity. Briefly, for each graph the number of A allele reads as a fraction of the total reads is plotted (y-axis) against the position of each of the several thousand interrogated SNPs on the chromosomes-of-interest (x-axis). The x-axis represents the linear position of each SNP along the chromosome, and each spot corresponds to a single SNP. As plasma cfDNA is a mixture of fetal and maternal cfDNA, the vertical position of each spot represents the sum of the contribution of both fetal and maternal allele reads, and is a function of the fetal fraction. To more readily visualize the maternal and fetal contributions, the spots are colored according to maternal genotype: SNPs for which the mother is homozygous for the A allele (AA) are colored red, SNPs for which the mother is homozygous for the B allele (BB) are colored blue, and SNPs for which the mother is heterozygous (AB) are colored green. Since the majority of plasma cfDNA is maternal in origin, the spots mainly distribute according to maternal genotype. The contribution of fetal allele reads results in segregation into distinct clusters. Because the loci targeted on the Y chromosome are homologous to loci on the X chromosome, but differ by one nucleotide, probes hybridize to both chromosomes. However, the targeted alleles have chromosomally distinct, non-dimorphic identities, so are not color-coded; all alleles from the X chromosome are assigned as A alleles, and all alleles from the Y chromosome are assigned as B alleles. Interrogated chromosomes are indicated above the plot. Fetal and maternal genotypes at individual SNPs are indicated to the right of the plot, with the maternal contribution to the fetal genotype color-coded. A. Euploidy, 38% fetal cfDNA fraction. The presence of three green clusters in the center of the plot (around 0.31, 0.50, and 0.69), as well as the presence of two red (around 1 and 0.81) and two blue (around 0 and 0.19) clusters, indicate the presence of two fetal chromosomes. Thus, this fetus has two copies of chromosomes 13, 18, 21, and X. For the Y chromosome, the A alleles are tightly associated with the upper limit of the plot, indicating the absence of a Y chromosome. Together, this indicates a 46,XX fetal chromosomal complement. B. Monosomy X, 18% fetal cfDNA fraction. The center trio of green clusters (centered around 0.41, 0.50, and 0.59) condense towards the center of the plot, and the red and blue peripheral clusters (centered around 0.91 and 0.09, respectively) regress towards the limits of the plot, due to the decreased contribution of fetal alleles to the overall number of A allele reads. Although copy number is difficult to identify by eye using this method at lower fetal fractions, the NATUS algorithm is able to make highly accurate copy number calls as low as 4% fetal fraction. Here, the pattern indicates two copies of chromosomes 13, 18, and 21. However, for the X chromosome, the center trio of clusters is replaced with a duo of clusters and the peripheral red and blue clusters (around 0.91 and 0.09, respectively) are absent, indicating the presence of a single fetal X chromosome. Additionally, the Y chromosome cluster remains tightly associated with the upper limit of the plot, indicating the lack of a Y chromosome. Together, this indicates a 45,X fetal chromosomal complement. C. Klinefelter syndrome, 30% fetal cfDNA fraction. For the autosomal chromosomes the typical “two-chromosome” cluster distribution is apparent. However, this pattern also appears for the X chromosome, indicating the presence of two copies. In the absence of reads from the Y chromosome, the fetal chromosomal complement would be identified as 46,XX. However, the presence of B alleles from the Y chromosome shifts the cluster downward, indicating the presence of a single Y chromosome (based on the distribution of the reads). Together, this indicates a fetal chromosomal complement of 47,XXY. The method is similarly able to call 47,XYY using SNPs from the homologous non-recombining regions of the X and Y chromosomes.

Figure 2

Graphical representation of sequencing data obtained from one euploid (Figure 2A), one Monosomy X (Figure 2B), and one 47,XXY (Klinefelter syndrome, Figure 2C) sample. All interrogated SNPs are assumed to be dimorphic and are designated as A and B for simplicity. Briefly, for each graph the number of A allele reads as a fraction of the total reads is plotted (y-axis) against the position of each of the several thousand interrogated SNPs on the chromosomes-of-interest (x-axis). The x-axis represents the linear position of each SNP along the chromosome, and each spot corresponds to a single SNP. As plasma cfDNA is a mixture of fetal and maternal cfDNA, the vertical position of each spot represents the sum of the contribution of both fetal and maternal allele reads, and is a function of the fetal fraction. To more readily visualize the maternal and fetal contributions, the spots are colored according to maternal genotype: SNPs for which the mother is homozygous for the A allele (AA) are colored red, SNPs for which the mother is homozygous for the B allele (BB) are colored blue, and SNPs for which the mother is heterozygous (AB) are colored green. Since the majority of plasma cfDNA is maternal in origin, the spots mainly distribute according to maternal genotype. The contribution of fetal allele reads results in segregation into distinct clusters. Because the loci targeted on the Y chromosome are homologous to loci on the X chromosome, but differ by one nucleotide, probes hybridize to both chromosomes. However, the targeted alleles have chromosomally distinct, non-dimorphic identities, so are not color-coded; all alleles from the X chromosome are assigned as A alleles, and all alleles from the Y chromosome are assigned as B alleles. Interrogated chromosomes are indicated above the plot. Fetal and maternal genotypes at individual SNPs are indicated to the right of the plot, with the maternal contribution to the fetal genotype color-coded. A. Euploidy, 38% fetal cfDNA fraction. The presence of three green clusters in the center of the plot (around 0.31, 0.50, and 0.69), as well as the presence of two red (around 1 and 0.81) and two blue (around 0 and 0.19) clusters, indicate the presence of two fetal chromosomes. Thus, this fetus has two copies of chromosomes 13, 18, 21, and X. For the Y chromosome, the A alleles are tightly associated with the upper limit of the plot, indicating the absence of a Y chromosome. Together, this indicates a 46,XX fetal chromosomal complement. B. Monosomy X, 18% fetal cfDNA fraction. The center trio of green clusters (centered around 0.41, 0.50, and 0.59) condense towards the center of the plot, and the red and blue peripheral clusters (centered around 0.91 and 0.09, respectively) regress towards the limits of the plot, due to the decreased contribution of fetal alleles to the overall number of A allele reads. Although copy number is difficult to identify by eye using this method at lower fetal fractions, the NATUS algorithm is able to make highly accurate copy number calls as low as 4% fetal fraction. Here, the pattern indicates two copies of chromosomes 13, 18, and 21. However, for the X chromosome, the center trio of clusters is replaced with a duo of clusters and the peripheral red and blue clusters (around 0.91 and 0.09, respectively) are absent, indicating the presence of a single fetal X chromosome. Additionally, the Y chromosome cluster remains tightly associated with the upper limit of the plot, indicating the lack of a Y chromosome. Together, this indicates a 45,X fetal chromosomal complement. C. Klinefelter syndrome, 30% fetal cfDNA fraction. For the autosomal chromosomes the typical “two-chromosome” cluster distribution is apparent. However, this pattern also appears for the X chromosome, indicating the presence of two copies. In the absence of reads from the Y chromosome, the fetal chromosomal complement would be identified as 46,XX. However, the presence of B alleles from the Y chromosome shifts the cluster downward, indicating the presence of a single Y chromosome (based on the distribution of the reads). Together, this indicates a fetal chromosomal complement of 47,XXY. The method is similarly able to call 47,XYY using SNPs from the homologous non-recombining regions of the X and Y chromosomes.

Figure 2

Graphical representation of sequencing data obtained from one euploid (Figure 2A), one Monosomy X (Figure 2B), and one 47,XXY (Klinefelter syndrome, Figure 2C) sample. All interrogated SNPs are assumed to be dimorphic and are designated as A and B for simplicity. Briefly, for each graph the number of A allele reads as a fraction of the total reads is plotted (y-axis) against the position of each of the several thousand interrogated SNPs on the chromosomes-of-interest (x-axis). The x-axis represents the linear position of each SNP along the chromosome, and each spot corresponds to a single SNP. As plasma cfDNA is a mixture of fetal and maternal cfDNA, the vertical position of each spot represents the sum of the contribution of both fetal and maternal allele reads, and is a function of the fetal fraction. To more readily visualize the maternal and fetal contributions, the spots are colored according to maternal genotype: SNPs for which the mother is homozygous for the A allele (AA) are colored red, SNPs for which the mother is homozygous for the B allele (BB) are colored blue, and SNPs for which the mother is heterozygous (AB) are colored green. Since the majority of plasma cfDNA is maternal in origin, the spots mainly distribute according to maternal genotype. The contribution of fetal allele reads results in segregation into distinct clusters. Because the loci targeted on the Y chromosome are homologous to loci on the X chromosome, but differ by one nucleotide, probes hybridize to both chromosomes. However, the targeted alleles have chromosomally distinct, non-dimorphic identities, so are not color-coded; all alleles from the X chromosome are assigned as A alleles, and all alleles from the Y chromosome are assigned as B alleles. Interrogated chromosomes are indicated above the plot. Fetal and maternal genotypes at individual SNPs are indicated to the right of the plot, with the maternal contribution to the fetal genotype color-coded. A. Euploidy, 38% fetal cfDNA fraction. The presence of three green clusters in the center of the plot (around 0.31, 0.50, and 0.69), as well as the presence of two red (around 1 and 0.81) and two blue (around 0 and 0.19) clusters, indicate the presence of two fetal chromosomes. Thus, this fetus has two copies of chromosomes 13, 18, 21, and X. For the Y chromosome, the A alleles are tightly associated with the upper limit of the plot, indicating the absence of a Y chromosome. Together, this indicates a 46,XX fetal chromosomal complement. B. Monosomy X, 18% fetal cfDNA fraction. The center trio of green clusters (centered around 0.41, 0.50, and 0.59) condense towards the center of the plot, and the red and blue peripheral clusters (centered around 0.91 and 0.09, respectively) regress towards the limits of the plot, due to the decreased contribution of fetal alleles to the overall number of A allele reads. Although copy number is difficult to identify by eye using this method at lower fetal fractions, the NATUS algorithm is able to make highly accurate copy number calls as low as 4% fetal fraction. Here, the pattern indicates two copies of chromosomes 13, 18, and 21. However, for the X chromosome, the center trio of clusters is replaced with a duo of clusters and the peripheral red and blue clusters (around 0.91 and 0.09, respectively) are absent, indicating the presence of a single fetal X chromosome. Additionally, the Y chromosome cluster remains tightly associated with the upper limit of the plot, indicating the lack of a Y chromosome. Together, this indicates a 45,X fetal chromosomal complement. C. Klinefelter syndrome, 30% fetal cfDNA fraction. For the autosomal chromosomes the typical “two-chromosome” cluster distribution is apparent. However, this pattern also appears for the X chromosome, indicating the presence of two copies. In the absence of reads from the Y chromosome, the fetal chromosomal complement would be identified as 46,XX. However, the presence of B alleles from the Y chromosome shifts the cluster downward, indicating the presence of a single Y chromosome (based on the distribution of the reads). Together, this indicates a fetal chromosomal complement of 47,XXY. The method is similarly able to call 47,XYY using SNPs from the homologous non-recombining regions of the X and Y chromosomes.