The clinical utility of genome-wide NIPT: a look into new opportunities Francesco Fiorentino Scientific and Lab Director Eurofins GENOMA Group - Molecular Genetics Laboratories Rome - Milan – Italy [email protected]
The clinical utility of genome-wide NIPT:
a look into new opportunities
Francesco Fiorentino Scientific and Lab Director
Eurofins GENOMA Group - Molecular Genetics
Laboratories
Rome - Milan – Italy
Disclaimer
The opinions expressed during this
presentation are those of the speaker and
may not represent the opinions of Illumina.
Any uses of Illumina’s products described
in this presentation may be uses that have
not been cleared or approved by the FDA or
any other applicable regulatory body
Conventional cfDNA screening
cfDNA-based NIPT has rapidly become integrated into clinical
practice.
High sensitivity and specificity resulting from multiple large-scale
clinical trials.
Conventional cfDNA-based NIPT approaches focus only on detection
of common trisomies and sex-chromosome aneuploidies.
This leaves a gap of ~17% of clinically relevant
chromosomal/subchromosomal abnormalities that would go
undetected (Wellesley et al., 2012).
Prevalence of chromosomal abnormalities
Wellesley et al. (2012) Eur J Hum Genet;20:521–6
Genome-wide cfDNA screening
Genome-wide cfDNA analysis would greatly expand the range of
chromosomal rearrangements detectable by NIPT
It extends screening to include also rare trisomies and structural
chromosome anomalies throughout the fetal genome
Potential to improve overall pregnancy management, providing a
significantly higher sensitivity compared to standard screening
Massively parallel sequencing (MPS) used for NIPT of common fetal
aneuploidies can also be used for detecting other fetal unbalanced
chromosomal rearrangements (sequencing data already contains
genome-wide informations)
Clinical implementation of
Genome-wide cfDNA screening
Limited clinical data are currently available.
Lack of prospective clinical results
It makes difficult to accurately determine the test performance
parameters.
The clinical utility of genome-wide cfDNA screening is controversial,
because this could lead to a decrease of the specificity, potentially
increasing invasive testing.
Risk of overdiagnosis with a higher number of false positives because
of chromosomal rearrangements which are confined to the placenta.
Aim of the study
In this study, we expanded conventional cfDNA-based non-invasive
prenatal testing(NIPT) to cover the entire genome.
We aimed to compare the performance of the two test in a large
general population of pregnant women, in order to assess the
clinical utility of the genome-wide screening.
Genome-wide cfDNA analysis was offered to pregnant women
undergoing NIPT for common fetal aneuploidy
Massively parallel sequencing (MPS) using NextSeq 550
High Output v1.2 kit (36 bp single-end reads)
Rapid Run module (2 days TAT)
Sequencing depth: ~30 millions unique reads per sample
Methods
Bioinformatic Analysis
Conventional
cfDNA screening Genome-wide
cfDNA screening
Sequencing data from a single run underwent to two different bioinformatic
analyses
Bayindir et al. (2015) Eur J Hum Genet;23:1286–93.
Genome partition into 50-kb bins; GC-normalized 50-kb bin
count; bin aggregation per 5Mb windows, sliding by 50 kb;
Z scores calculation per 5 Mb bins.
Chromosomally abnormal results of cfDNA testing were confirmed by a
metaphase and/or array-CGH-based karyotyping (amniocentesis or
CVS or POC)
Chromosomally normal results were confirmed by newborn physical
examination. A newborn with a normal physical examination was
considered to be euploid.
Follow-up information was obtained by telephone and recorded in an
internal database.
Karyotyping or clinical follow-up results were used as the gold
standard to calculate sensitivity and specificity of NIPT in this
population
Follow-up
Patients enrolled in the study
Genome-wide
cfDNA screening
was offered
Patients not accepting
genome-wide
cfDNA screening
Patients accepting
genome-wide
cfDNA screening
Demographics and pregnancy characteristics
Characteristics n
No. of eligible patients 12.114
Maternal age-yr
- Mean ±SD 35.3±4.1
- Min-max 20-58
Gestational age at sample collection -wk
- Mean ±SD 12.3±2.1
- Min-max 10-29
Indications for NIPT
- Parental Anxiety 3804 (31.4%)
- Advanced maternal age 4446 (36.7%)
- Positive prenatal screen 1199 (9.9%)
- Fetal ultrasound abnormality 472 (3.9%)
- Prior pregnancy with fetal aneuploidy 157 (1.3%)
- More than one indication 2035 (16.8%)
Results
Clinically relevant chromosomal abnormalities
classes detected by genome-wide cfDNA analysis
Types of chromosome anomalies detected by
conventional vs genome-wide cfDNA testing
10.7%
del5p15.33p13.2
Chr. 5 Chr. 5
Array-CGH
cfDNA Amniotic
Fluid
34 Mb
Detection of a deletion 5p
Genome-wide cfDNA
screening
31 y.o patient
13^ weeks gestation
Indication for testing:
Parental anxiety
Del18p11.32-p11.31
Chr. 18
Array-CGH
3.8 Mb
Del18q21.32-q23
21.3 Mb
Identification of 2 deletions (18p / 18q) Chr. 18
Amniotic
Fluid
Genome-wide
cfDNA screening
32 y.o patient
20^ weeks gestation
Indication for testing:
Fetal ultrasound
abnormality
32 y.o patient
11^ weeks gestation
Carrier of a reciprocal translocation
46,XX,t(7;9)(p15;q22)
Chr. 7
Chr. 9
Patients unaware to be carrier of translocation
Chr. 13 Chr. 20
33 y.o patient
12^ weeks gestation
Carrier of a reciprocal translocation
46,XX,t(13;20)(q33;q13.3)
Indication for testing:
Parental anxiety
Array-CGH
CVS
Array-CGH
CVS
Dup7p22.3p21.2 15 Mb Dup9p24.3-q31.1 107.5 Mb
Fetal karyotype with an unbalanced translocation
Genome-wide
cfDNA screening
Genome-wide
cfDNA screening
Del13q33.1q34 11.4 Mb Dup20q13.33 1.9 Mb
Array-CGH
(CVS)
Chr. 13 Chr. 20
Array-CGH
(CVS)
Fetal karyotype with an unbalanced translocation
Genome-wide
cfDNA screening
Genome-wide
cfDNA screening
Chr.
18
Detection of a duplication 18p
Genome-wide
cfDNA screening
Detection of a duplication 11p Chr. 11
Array-CGH
cfDNA Amniotic
Fluid
Dup11p15.4p15.1
16 Mb
Dup11p15.4p15.1
16 Mb
Genome-wide
cfDNA screening
Array-CGH
cfDNA Amniotic
Fluid
delXp22.33p11.1
58.5 Mb
delXp22.33p11.1
58.5 Mb
Detection of a deletion Xp
Genome-wide
cfDNA screening
Array-CGH
cfDNA
del22q11.21
DiGeorge Syndrome
4 Mb
Chr.22
Amniotic
Fluid
Identification of 22q11.2 deletion
Genome-wide
cfDNA screening
Chr. 9
46,XX[80]/47,XX,+9[20]
Detection of Trisomy 9 mosaic (20%)
Traditional Karyotyping
Genome-wide
cfDNA screening
42 y.o patient
14^ weeks gestation
Indication for testing:
AMA
Chr.22
46,XY[42]/47,XY,+22[8] Traditional Karyotyping
Detection of Trisomy 22 mosaic (16%)
Genome-wide
cfDNA screening
40 y.o patient
10^ weeks gestation
Indication for testing:
AMA
Chr. 7
46,XX[48]/47,XX,+7[2]
Traditional Karyotyping
Detection of Trisomy 7 mosaic (4%)
Genome-wide
cfDNA screening
^ Clinically relevant chromosomal abnormalities, not detected by conventional cfDNA screening,
potentially resulting in the birth of babies with chromosomal anomalies, have been considered as
false negative.
§ A P-value of less than 0.05 was considered to indicate statistical significance (**).
Performance of conventional cfDNA screening vs.
genome-wide analysis
Conventional
cfDNA
screening
Genome-
wide cfDNA
screening
P-value§
No. of pregnancies assessed 12.114 12.114
Clinical relevant chromosomal abnormalities detected - no. 166 196
Pregnancies confirmed as chromosomally abnormal - no. 151 169
False Positive 15 27
False Negative 12* 0
True Positive 151 169
True Negative 11.936 11.918
Sensitivity 92.64% 100.00% <0.001
Specificity 99.87% 99.77% 0.064
Positive Predictive Value (PPV) 90.96% 86.22% 0.161
Negative Predictive Value (NPV) 99.90% 100.00% <0.001
Performance of Genome-wide cfDNA screening
Rare trisomies must be interpreted with caution
Risk of overdiagnosis, with a higher number of false positives,
because mosaicism confined to the placenta (CPM).
Most of the rare trisomies identified are not viable and, therefore,
may not require an invasive testing follow-up, thus limiting this risk.
Such additional data has important clinical implications and may be
helpful in improving pregnancy management.
Placental mosaicism carries a small but significant risk for:
intrauterine growth restriction (IUGR),
small-for-gestational-age infants
unfavorable pregnancy outcome (e.g., T16)
mosaic fetal aneuploidy
fetal uniparental disomy (UPD)
Rare Trisomies
Rare Trisomies are associated with an
increased risk of feto-placental disease
Pertile et al. Sci Transl Med. 2017 Aug 30;9(405)
Pertile et al. Sci Transl Med. 2017 Aug 30;9(405)
Rare Trisomies are associated with an
increased risk of feto-placental disease
Genome-wide cfDNA testing resulted accurate in the detection of
common and rare aneuploidies, as well as segmental chromosome
abnormalities;
The approach improved the detection rate of genetic aberrations as
compared with conventional cfDNA screening (100% vs 92.64%,
p<0.001);
12 (7.4%) clinically relevant fetal chromosome anomalies would
have been missed if only conventional NIPT had been performed
No statistically significant increase in false positive results: the
specificity of genome-wide screening resulted similar to that obtained
with standard cfDNA testing (99.87% vs 99.77%, p=0.064).
Summary of the Study
40 y.o. patient
NIPT (5 chromosomes screening)
Trisomy 21 not detected Maternal CNV detected
causing the false positive result
Positive Result High risk for Trisomy 21
Test performed by a different lab
Case Study: Copy Number variation
13^ weeks gestation
Genome-wide NIPT
Lower the risk of false positive
results determined by maternal CNVs
Genome-wide
cfDNA screening Array-CGH
Maternal blood
Chudova et al., 2016 NEJM
Lower the risk of false positive
results determined by maternal CNVs
Strom et al., 2017 NEJM
Lower the risk of false positive
results determined by maternal CNVs
Conclusion This study provided preliminary data towards the clinical utility of
genome-wide cfDNA analysis.
This level of screening provided a significant higher sensitivity
compared to standard screening while maintaining a high specificity
Genome-wide cfDNA screening has the potential to improve overall
pregnancy management.
Additional clinical data are required before this approach can be
evaluated for routine integration into NIPT programs.
The clinical utility of genome-wide NIPS
Fiorentino et al. (2017) Prenatal Diagnosis 37, 593–601
Acknowledgements
NIPT Laboratory Team
Sara BONO
Francesca PIZZUTI
Arianna POLVERARI
Sara DUCA
Monica FAIETA
Sara AMATUCCI
Ilaria PAGANO
Donatella VIOLA
Vanessa DE PAOLIS
Fabiana ROMANO
Elena D’ANGELOSANTE
Giulia PREDEBON
Clarissa LOCCI
Elena CORSETTI
Scientific Coordinator
Francesca SPINELLA
NIPT Genetic Counselors
Marina BALDI
Mariateresa SESSA
Laura GIGANTE
Laura DIANO
Thank you for your attention
Francesco Fiorentino
http://www.laboratoriogenoma.eu/
Rome Milan
Major limitation related with the estimation of sensitivity of segmental
imbalances
The study design did not included infant follow-up by chromosome
microarray analysis (CMA), so we were unable to determine whether
negative genome-wide screening results were actual true-negative
We cannot rule out that very small rearrangements may have
remained unnoticed in the newborns.
Similarly, low level mosaic aneuploidies are not necessarily apparent
at newborn physical examination
Limitations
Results of samples tested
Total
No. of patients analyzed 12.114
Samples with a call - no. (%) 12.078 (99.7)
Total cancellations - no. (%) 182 (1.5)
- Samples with low FF - no. (%) 145 (1.2)
- Samples with assay failure - no. (%) 36 (0.3)
Samples with a conclusive result - no. (%) 11.932 (98.5)
Chromosomally abnormal results
- Genome-wide cfDNA screening - no. (%) 196 (1.6)
- Conventional cfDNA screening - no. (%) 166 (1.4)
Pregnancies confirmed as chromosomally abnormal - no. (%)
- Genome-wide cfDNA screening - no. (%) 169 (1.4)
- Conventional cfDNA screening - no. (%) 151 (1.3)
Del5q35.3
Array-CGH
(Amniotic Fluid) Array-CGH
(Amniotic Fluid)
69.4 Mb Dup3q21.3-q29
2.2 Mb
Fetal karyotype with an unbalanced translocation
Genome-wide cfDNA
screening
Genome-wide cfDNA
screening
Array-CGH
cfDNA Amniotic
Fluid
del5p15.33p13.3
30.7 Mb
del5p15.33p13.3
30.7 Mb
Detection of a deletion 5p
Genome-wide
cfDNA screening