Single-cell sequencing of circulating trophoblasts for non-invasive fetal pathogenic copy number variant screening: first large-scale clinical validation study results
Prenatal Genetics
-
Primary Categories:
- Prenatal Genetics
-
Secondary Categories:
- Prenatal Genetics
Introduction:
Pathogenic microdeletions/duplications (pCNVs) account for a significant perinatal morbidity/mortality and have a 1/100-200 prevalence in the general prenatal population. Cell-free DNA noninvasive testing (cfNIPT) shows significant limitations for comprehensive profiling of pCNV<7-10Mb. We have recently published a proof-of-concept study demonstrating the feasibility of a novel cell-based NIPT (cbNIPT) for circulating extravillous trophoblasts (cEVTs) isolation and single cell sequencing for fetal CNV detection down to ~800kb. Our current effort is aimed to demonstrate cbNIPT clinical performance in a large blinded prospective study comparing screening by genomic analysis of isolated fetal cells to invasive diagnostic testing results (ClinicalTrial.gov NCT05671744).
Methods:
20mL of venous blood was collected from pregnant women at 11+0 to 22+6 weeks before the clinically indicated invasive diagnostic procedure and processed with a proprietary laboratory workflow. Concordance between chromosomal microarray (CMA) and/or karyotype and cbNIPT was conducted on genomic imbalances reported to patients based on the clinical centers’ standard of care.
Results:
1390 pregnant participants were enrolled and 995 patient samples ultimately met criteria for inclusion in the final analysis with 778 having CMA as diagnostic comparator while the remainder karyotype only. At the subject level, cbNIPT cumulative sensitivity and specificity for common autosomal trisomies (CATs) was 98.3% (119/121) and 99.2% (8/874), for sex chromosome aneuploidies (SCAs) was 77.8% (4/18) and 99.3% (7/977) and pCNV≥1Mb 100.0% (28/28) and 96.8% (31/967), respectively.
Secondary analysis for pCNV<1Mb showed sensitivity and specificity of 100.0% (2/2) and 99.7% (2/666) for the size range 600Kb-1Mb. Cumulatively, test performances for pCNV≥600Kb are 100.0% (30/30) and 96.6% (33/965).
The lowest level of detection achieved was down to 300Kb. For pCNV≥300Kb cumulative performances are 94.4% (34/36) and 96.1% (33/959), respectively.
Of the 37 false positives, the vast majority are given by cases where a ‘private’ large segmental imbalance >12Mb was detected in only 1 out of multiple cEVT analyzed (n=24; 65%) and in cases where a pCNV was consistently seen in multiple cEVTs but not reported by clinical centers (n=6; 16%).
Conclusion:
Analysis of circulating trophoblasts from maternal blood with the current workflow and bioinformatic pipeline allows the potential for detecting genomewide pCNVs at an early gestational age for any pregnant women. This would represent ‘specific’ and ‘intentional’ (not incidental) comprehensive screening for a wide range of true pathogenic CNVs at high resolution, with high sensitivity. Screen negative results would drastically lower the ‘residual risk’ in women screened compared to conventional prenatal screening tools, including cell-free DNA testing, for true pathogenic CNVs.
Pathogenic microdeletions/duplications (pCNVs) account for a significant perinatal morbidity/mortality and have a 1/100-200 prevalence in the general prenatal population. Cell-free DNA noninvasive testing (cfNIPT) shows significant limitations for comprehensive profiling of pCNV<7-10Mb. We have recently published a proof-of-concept study demonstrating the feasibility of a novel cell-based NIPT (cbNIPT) for circulating extravillous trophoblasts (cEVTs) isolation and single cell sequencing for fetal CNV detection down to ~800kb. Our current effort is aimed to demonstrate cbNIPT clinical performance in a large blinded prospective study comparing screening by genomic analysis of isolated fetal cells to invasive diagnostic testing results (ClinicalTrial.gov NCT05671744).
Methods:
20mL of venous blood was collected from pregnant women at 11+0 to 22+6 weeks before the clinically indicated invasive diagnostic procedure and processed with a proprietary laboratory workflow. Concordance between chromosomal microarray (CMA) and/or karyotype and cbNIPT was conducted on genomic imbalances reported to patients based on the clinical centers’ standard of care.
Results:
1390 pregnant participants were enrolled and 995 patient samples ultimately met criteria for inclusion in the final analysis with 778 having CMA as diagnostic comparator while the remainder karyotype only. At the subject level, cbNIPT cumulative sensitivity and specificity for common autosomal trisomies (CATs) was 98.3% (119/121) and 99.2% (8/874), for sex chromosome aneuploidies (SCAs) was 77.8% (4/18) and 99.3% (7/977) and pCNV≥1Mb 100.0% (28/28) and 96.8% (31/967), respectively.
Secondary analysis for pCNV<1Mb showed sensitivity and specificity of 100.0% (2/2) and 99.7% (2/666) for the size range 600Kb-1Mb. Cumulatively, test performances for pCNV≥600Kb are 100.0% (30/30) and 96.6% (33/965).
The lowest level of detection achieved was down to 300Kb. For pCNV≥300Kb cumulative performances are 94.4% (34/36) and 96.1% (33/959), respectively.
Of the 37 false positives, the vast majority are given by cases where a ‘private’ large segmental imbalance >12Mb was detected in only 1 out of multiple cEVT analyzed (n=24; 65%) and in cases where a pCNV was consistently seen in multiple cEVTs but not reported by clinical centers (n=6; 16%).
Conclusion:
Analysis of circulating trophoblasts from maternal blood with the current workflow and bioinformatic pipeline allows the potential for detecting genomewide pCNVs at an early gestational age for any pregnant women. This would represent ‘specific’ and ‘intentional’ (not incidental) comprehensive screening for a wide range of true pathogenic CNVs at high resolution, with high sensitivity. Screen negative results would drastically lower the ‘residual risk’ in women screened compared to conventional prenatal screening tools, including cell-free DNA testing, for true pathogenic CNVs.