Los días 8 y 9 de junio de 2017 organizamos en la Fundación Ramón Areces con el Ciberer y la Fundación Jiménez Díaz un simposio internacional sobre 'Diagnóstico prenatal no invasivo y diagnóstico genético reproductivo'. Coordinado por la doctora Ana Bustamante, del servicio de Genética del Hospital Universitario Fundación Jiménez Díaz, tuvo como objetivo mostrar los últimos avances en el campo de la genética reproductiva a nivel preimplantacional, prenatal, e, incluso, preconcepcional.

1. Sarah Mason
2nd February 2016
Implementing Non-Invasive Prenatal Diagnosis (NIPD) of
monogenic diseases in a National Health Service Laboratory
Fiona McKay
Principal Clinical Scientist, NE Thames Regional Genetics Service
8th June 2017

2. Invasive vs non-invasive
CVS:
 normally between 11-14 weeks gestation
 Small increased risk of miscarriage
 Small risk of infection
 Repeat sampling if insufficient material obtained
Amniocentesis
 normally 15-20 weeks gestation
 small increased risk of miscarriage
 Small risk of infection
 cffDNA
 no increased risk of miscarriage
 easier sample collection
 earlier diagnoses (from 7 weeks)
AMNIOCENTESIS
CVS

3. Cell-free fetal DNA (cffDNA)
 Originates from placenta (trophoblast)
 represents whole fetal genome
 Detectable from 4 weeks
 % increases with gestation
 10-20% of total cfDNA
 Cleared from circulation within 30mins of delivery
 Shorter than maternal cfDNA
 maternal cfDNA 166bp
 cffDNA 143bp
Cell free DNA

4. Cell-free fetal DNA - problems
 Relative abundance of maternal cfDNA
 Emanates from the placenta - risk of mosaicism?
 Limitations for use in multiple pregnancies
 When used in early pregnancy, issues associated with ‘vanishing twin’

5. Non-invasive prenatal … definitions
• NIPD: DIAGNOSIS - Does NOT require invasive test for confirmation
• Single gene disorders
• Sex determination
• RHD
• NIPT or NIPS: TESTING or SCREENING
• Aneuploidy
Requires invasive testing for confirmation

6. Background
In the UK there are ~20,000 molecular prenatal tests per annum
85% because of a high risk for Down’s syndrome
Over 3000 prenatal diagnoses are for a risk of a monogenic disorder, 70% after invasive
testing.
2015-2016:
NIPT for aneuploidy included for the first
time 1138 reports (NHS only)
2006-2007:
NIPD for fetal sex determination was
included in our national audit for the
first time: 60 tests (1 lab)
In 2015-16: 447 tests (5 labs)
2013-2014: NIPD for monogenic disease (TD)
included in Top Prenatal Reports Category
ACGS Audit 2013-2014

7. NIPD in use in the UK from 2003
Current practice
Fetal sex determination - genes on the Y-chromosome
RHD in Rhesus negative mothers at high risk of HDN (not at GOSH)
 Women at risk of carrying a fetus with single gene disorder
 Scan abnormalities
 Known genetic history in family
Paternally inherited dominant disorders
De-novo gene changes (e.g. Achondroplasia)
Paternal mutation exclusion for recessive conditions where parents are
carriers of different mutations

8. Non-invasive prenatal diagnosis (NIPD):
Detection of cell-free fetal DNA (cffDNA) in maternal plasma to offer a safe
alternative to invasive testing.
Our ISO15189 accredited NHS Regional Genetics Laboratory offers a clinical NIPD
service using targeted NGS panels (Illumina MiSeq) to detect the presence of a
paternal or de novo allele:
 FGFR3-related skeletal dysplasia including achondroplasia and
thanatophoric dysplasia
 FGFR2-related craniosynostosis syndromes e.g. Apert & Crouzon
 10 common cystic fibrosis mutations
30% of all our molecular prenatal diagnosis since 2015 have been non-invasive
7883

9. NIPD for FGFR3-related skeletal dysplasias
Achondroplasia
 Most common viable skeletal dysplasia
 Most present de novo, in late pregnancy
 98% cases due to single mutation in FGFR3
Thanatophoric dysplasia (TD)
 Most common lethal skeletal dysplasia
 At least 13 mutations in FGFR3
N
ACH
TD

10. Next generation sequencing (MiSeq Illumina)
 30 FGFR3 mutations
 (+Severe Achondroplasia with Developmental Delay and Acanthosis
Nigricans (SADDAN), hypochondroplasia, craniosynostosis)
 6 amplicon panel
 Sequence 1000’s of reads
 Can determine number of reads of wild type allele and number of reads of
mutant allele

11. NIPD for FGFR3-related skeletal dysplasias
 Validation
 40 achondroplasia – 39/40 correct (rare mutation not on panel)
 38 TD - all correct
 Several different mutations detected
Next generation sequencing - MiSeq
NIPD FGFR3 gene panel approved by UKGTN in 2013
cfDNA cfDNA mat gDNA
Type2wt 244325 248812 135147
c.1948A>C 139 215 115
c.1948A>G 16768 16937 25
c.1949A>T 204 23 109
c.1949A>C 284 160 168
c.1950G>C 27 33 7
c.1950G>T 107 39 38

12. Benefits of NIPD for FGFR3-related skeletal dysplasias
Achondroplasia
 Confirmation of sonographic diagnosis without risk of preterm labour
 Safe exclusion of Down’s syndrome and other more severe dysplasias
 Allows for accurate parental counselling and preparation for delivery
Thanatophoric dysplasia
 May be detected early. NIPD allows risk free early definitive diagnosis
 Excludes differential diagnoses which may carry high recurrence risk
 Allows for a surgical termination if requested
 Provides accurate diagnosis in twins without putting the normal twin at risk
For both, NIPD allows safe exclusion of a recurrence early in future pregnancies

13. Apert syndrome
 Autosomal Dominant, 2 common mutations in the FGFR2 gene
 NIPD aids sonographic diagnosis from 12 weeks
 Allows for exclusion of recurrence in subsequent pregnancies from 9 weeks
 Definitive, safe diagnosis in twins
 12 diagnostic NIPD NGS tests performed
MiSeq NGS results showing large number of wildtype counts in
maternal plasma and genomic DNA, but only mutant counts in the
two plasma samples analysed
NIPD for Apert syndrome approved by UKGTN in 2014

14. NIPD for Cystic Fibrosis – paternal exclusion
 Autosomal recessive - paternal mutation exclusion
 5 CFTR amplicons for MiSeq (Illumina)
 Only applicable to couples
 1) who are known carriers of different CF mutations AND
 2) the paternal mutation is one of the 10 mutations listed below
 10 mutations including neonatal screening common mutations (bold)
 c.1521_1523delCTT p.(Phe508del)
 c.489+1G>T
 c.1624G>T p.(Gly542*)
 c.1652G>A p.(Gly551Asp)
 c.3846G>A p.(Trp1282*)
 c.1657C>T p.(Arg553*)
 c.1519_1521delATC p.(Ile507del)
 c.1679G>C p.(Arg560Thr)
 c.1646G>A p.(Ser549Asn)
 c.1647T>G p.(Ser549Arg)
 13 diagnostic NIPD NGS tests performed
NIPD for Cystic Fibrosis approved by UKGTN in 2014
FALSE NEGATIVE NIPD CASE
• CF NIPD referral for pat mutation exclusion
• Paternal mutation c.3909C>G – detected in
paternal gDNA using assay
• Not detected in cfDNA in 2 separate
samplings, presence of cfDNA confirmed
• Baby affected with CF
• Mutation withdrawn from panel

15. Aim: To increase availability of NIPD
Bespoke NIPD:
Target Population
Couples with a pregnancy at risk of a genetic condition where a de novo or
paternal mutation is known.
Relative Haplotype Dosage Analysis:
Target Population
Couples with a pregnancy at high risk of inheriting a
recessive genetic condition irrespective of genotype.

16. Aim: To increase availability of NIPD
Inclusion criteria:
1-3 base pair mutation
Control gDNA
De Novo / Paternal Dominant
For recessive conditions, parents must
carry different mutations
Target Population
Couples with a pregnancy at risk of a genetic condition where the mutation is
known and invasive prenatal diagnosis would otherwise be the only option
Exclusion criteria:
Pseudogene
Polymorphisms

17. Bespoke NIPD
Assay development for each couple:
Ideally, pre-pregnancy
gDNA from affected proband or carrier parent
Primer design, SNP check & order
NGS confirmation of mutation detection
Report within 8 weeks to offer NIPD
cffDNA testing in pregnancy:
Minimum 9 weeks gestation
May require 2 samples one week apart at early gestation

18. FRAS1 exon 66
hg19_dna range=chr4:79437011-79437070
ctccctccctggcagAGGCAGGGTTCCTGGATGATGTGGTCTATGATAGCACTGCCCTGGGGCCTGGCTACGATCGCCC
CTTCCAGTTTGACCCCAGCGTGCGAGAGCCGAAGACCATCCAGCTCTACAAACACCTGAACCTGAAGAGCTGCGTGTGG
ACCTTTGATGCTTATTATGACATGACTGAGCTGATTGACGTCTGTGGGGGCTCTGTAACCGCTGACTTCCAGgtaggtg
ccccggggcttgtctgaggactctgc
Bespoke Amplicon Based Design
C primer
Re familial mutation
G exon
Sample index - 6bp
Illumina sequence adaptor (MiSeq)
60bp

19. Bioinformatic Analysis
Automated scripting to count wild type and
mutant sequences
Monitoring sample indexes
for potential contamination
Poor quality data is filtered outID6 ID7 ID8
cffDNA (1) cffDNA (2) Mat gDNA
FRAS1 wt 319077 141751 440538
FRAS1 c.10261C>T 34498 12055 292
ZFX 36743 44017 350940
ZXY 6 0 2

20. Bioinformatic Analysis
Confirmation of Fetal DNA
ZFX/Y and HLA assays are run in parallel to confirm presence of cffDNA in the
same plasma extract undergoing mutation test:
• Calculate fetal fraction
• Important in mutation negative cases
• Uninformative (HLA/ZFY) mutation negative cases: can’t confirm fetal DNA present
ID01 ID08 ID02
cfDNA cfDNA gDNA
HLABexon3-VAR01 0 1 2
HLABexon3-VAR02 3509 2734 2805
HLABexon3-VAR03 1 9 0
HLABexon3-VAR04 9 26 11
HLABexon3-VAR05 4108 3582 2876
HLABexon3-VAR06 3 0 0
HLABexon3-VAR07 336 656 0
HLABexon3-VAR08 0 0 0
ZFX 27046 48258 55986
ZFY 764 1456 6
Apert wt 97397 163414 97801
Apert c.755C>G 15 15 13
Apert c.758C>G 16 26 8
Apert
c.755_756delCGinsTT 0 2 0
FETAL
FRACTION ID01
VARIANT Counts
Maternal allele 1: VAR#2 3509
Maternal allele 2: VAR#5 4108
TOTAL maternal
counts: 7617
VARIANT Counts
Fetal HLA variant: VAR#7 336
Fetal Fraction: 8.45

21. Bespoke NIPD tests
Confirmed inheritance of paternal FGFR2 mutation (Crouzon syndrome)
Pre-pregnancy work-up
Bespoke NIPD test in subsequent pregnancy – fetus predicted to be AFFECTED

22. Bespoke NIPD tests
Exclusion of recurrence of de novo SMARCB1 mutation
Pre-pregnancy work-up
Bespoke NIPD test in subsequent pregnancy – fetus predicted to be UNAFFECTED

23. Audit of bespoke service - delivery
Phenotype Gene Mutation
Tuberous sclerosis TSC2 c.133_134delCT
c.2713C>T
c.4351dupC
c.4645T>A
Neurofibromatosis
type 1
NF1 c.4823T>C
c.4330A>G
c.5170C>T
Rhabdoid Tumour
Predisposition Syndrome
SMARCB1 c.157C>T
c.118C>T
Epileptic encephalopathy,
early infantile
SCN8A
KCNQ2
c.669G>T
c.431G>A
Skeletal Dysplasia
(AR)
FGFR3
TCF12
SLC26A2
c.779C>G
c.1916del
c.296T>C
Fraser syndrome (AR) FRAS1 c.10261C>T
c.10261C>T
c.10261C>T
Late infantile neuronal
ceroid lipofuscinosis
TPP1 c.509-1G>C
Noonan syndrome PTPN11 c.178G>A
Porencephaly COL4A1 c.324+1G>A
Marfan syndrome FBN1 c.8268G>A
Polycystic kidney disease
(AR)
PKHD1 c.1486C>T
Phenotype Gene Mutation
Intellectual disability ARID1B
KDM5B
BCAP31
CASK
c.6129del
c.3343C>T
c.2359T>C
c.678+1G>A
c.37G>T
Congenital Nephrotic
Syndrome
WT1 c.1301G>A
Cornelia De Lange syndrome NIPBL c.4160dupA
Osteogenesis imperfecta COL1A1
COL1A2
c.543+4A>T
c.543+4A>T
c.1875+1G>A
c.1801G>A
c.1801G>A
c.1801G>A
c.1801G>A
Renal tubular dysgenesis
(AR)
ACE c.96del
Epidermolytic Palmoplantar
Keratoderma
KRT9 c.482A>G
c.482A>G
Hereditary spastic paraplegia SPAST c.1361A>G
Familial dilated
cardiomyopathy
ACTC1 c.664G>A
Alexander disease GFAP c.235C>A
42 bespoke tests offered:
9/42 mutation positive

24. Audit of bespoke service - requests
Designs Requested Gene Mutation
Zellweger PEX6 c.1314_1321del
Intellectual disability KATA6A c.4213G>T
Smith-Lemli-Optiz (AR) DHCR7 c.964-1G>C
TPM1-related cardiomyopathy TPM1 c.742A>G
Neurofibromatosis type 2 NF2 c.523_524del
Alpers syndrome POLG c.2542G>A
Laron syndrome GHR c.922G>A
Spondyloepiphyseal dysplasia
congenita
COL2A1 c.2104G>A
Osteogenesis imperfecta COL1A2 c.965G>A
Osteogenesis imperfecta COL1A1 c.544-1G>A / c.941G>A
Ornithine Transcarbamylase
Deficiency
OTC c.521C>T
Congenital glaucoma FOXC1 c.392C>A
Neurofibromatosis type 1 NF1 c.6007-2A>G
NIPD not possible

25. NIPD for recurrence
NIPD to exclude recurrence of an apparently de novo mutation
 All pregnancies tested on this basis detected no mutation
 Is there any point in testing low risk pregnancies?
Benefits
 Safe non-invasive test for low risk pregnancies
 Welcomed by patients, many of whom prefer to accept the low risk of
recurrence rather than undergo invasive testing
 In some mosaic cases, NGS analysis has indicated recurrence risk may be
higher
 For recessive conditions, exclusion of the paternal allele reduces the rate of
invasive testing as only pregnancies where the paternal allele is inherited
require invasive test

26. Number of Invasive and Non-Invasive Prenatal Diagnosis tests carried out per year
Prenatal service delivery – where are we now?
Single gene tests: TOTAL: 383 Fails: 6 (1.5%) Inconclusive: 4 (1.0%)
0
10
20
30
40
50
60
70
80
90
2008-9 2009-10 2010-11 2011-12 2012-13 2013-14 2014-15 2015-16 2016-17
SD Invasive
SD NIPD
0
10
20
30
40
50
60
70
80
90
2008-9 2009-10 2010-11 2011-12 2012-13 2013-14 2014-15 2015-16 2016-17
SD Invasive
SD NIPD
Bespoke
0
10
20
30
40
50
60
70
80
90
2008-9 2009-10 2010-11 2011-12 2012-13 2013-14 2014-15 2015-16 2016-17
SD Invasive
SD NIPD
Bespoke
CF Exclusion
CF RHDO

27. How is bespoke NIPD being offered?
Not all NHS trusts are funding the same level of NIPD
 cost for bespoke work-up (as assay likely only to be used by that family)
 additional cost for NIPD
Three funding scenarios:
 Trust pays for work-up and NIPD
 Parents pay for work-up & trust pays for NIPD
(Some trusts won’t pay for recurrence testing)
 Parents pay for work-up & NIPD
MOST COMMON

28. Challenge
fetal DNA present amongst high background of maternal DNA
detect slight increase in mutant reads
NIPD for autosomal recessive conditions
cffDNA fraction directly affects
expected extent of allelic imbalance
Concentration of cffDNA varies with
gestation & between pregnancies
Relative mutation dosage-RMD
Measuring fetal fraction:
Y-chr markers
size differential
heterozygous SNPs

29. NIPD by relative haplotype dosage analysis
Maternal
Paternal Proband
Haplotype analysis
Lo et al 2010 Sci Trans Med; New et al 2014 J Clin End
Metab
Link SNPs flanking CYP21A2 to mutant allele
(Hap 1 = mutant allele)
Agilent Sureselect Custom Enrichment Assay
~6500 SNPs (7Mb flanking CYP21A2)

30. Principles of linkage analysis
SNPs flanking CYP21A2 are genotyped in
maternal and paternal DNA
Next generation sequencing of
SNPs in cell-free DNA
Assess all informative SNPs
along chromosome to
determine inheritance of
paternal haplotype
SNP informative if;
Homozygous in Mother
Heterozygous in Father
Hap 1 / Hap 2
Maternal
Paternal
Parental
genotypes
Paternal Hap 1
transmitted if ‘C’
allele NOT
detected
Or
A A A C
Fetal genotypes
A A A C
Maternal
plasma
Paternal Hap 2
transmitted if ‘C’
allele IS detected
Paternal haplotype transmission

31. Linkage analysis
Advantages
• Useful when there is a pseudogene/high homology with another gene
• Useful for diagnostic service as can have single NIPD assay for all families with
mutations in the same gene, regardless of mutation
Challenges
• Sufficient DNA for library preparation
• robust assay with variable cfDNA
• PCR – duplicates
• Labour intensive, time consuming assay

32. Adapted from Lo et al 2010, Sci. Transl. Med. 2, 61ra91
Relative Haplotype Dosage Analysis (RHDO)
 Involves analysis of multiple SNPs which flank the disease gene / locus of interest
 If the phase of the mutation carried by both parents is known then RHDO can assess
whether the maternal mutant or wild-type allele has been transmitted to the fetus
 Provides definite diagnosis for recessive conditions
NIPD for Cystic Fibrosis by RHDO – in service from 3rd
October 2016
NIPD for Spinal Muscular Atrophy (SMA) and Congenital
Adrenal Hyperplasia (CAH) by RHDO – under development

33. 170394 2017-05-13
Fetal fraction
Fetal fraction Qualified SNPs Disqualified SNPs
Type 1 SNPs 0.11 20 0
Type 3AD SNPs 0.103 18 0
Type 3BC SNPs 0.0815 2 51
Maternal haplotype
Homozygous.for.HapI Heterozygous Homozygous.for.HapII Conclusion
Type_4a 0 3 0 Discordant to Proband
Type_4a_rev 0 3 0 Discordant to Proband
Type_4b 0 0 6 Discordant to Proband
Type_4b_rev 0 0 6 Discordant to Proband
Paternal haplotype
Qualified.SNPs Disqualified.SNPs Conclusion
Type 3AD SNPs 18 0Discordant to Proband
Type 3BC SNPs 2 51NA
Proband Affected
Maternal haplotype Discordant to Proband
Paternal haplotype Discordant to Proband
cffDNA result Predict UNAFFECTED
Relative Haplotype Dosage Analysis (RHDO)

34. RHDO – Cystic Fibrosis Service
Case Gestation Fetal Fraction Outcome
1 10+3 8% Paternal high risk
Maternal high risk
2 11 4.8% Paternal high risk
Maternal low risk
3 9+1 15.4% Paternal low risk
Maternal low risk
4 12+2 14.6% Paternal low risk
Maternal high risk
5 9+4 11% Paternal high risk
Maternal high risk
6 9+0 6.4% Paternal high risk
Maternal low risk
7 9+0 11.6% Paternal low risk
Maternal low risk

35. Case study G129789
• Maternal mutation c.1521_1523delCTT p.(Phe508del)
• Paternal mutation c.489+1G>T
• Affected son
• 2015 – NIPD paternal mutation exclusion
• no mutation detected, invasive not required
• 2017 – NIPD RHDO
• paternal high risk, maternal low risk alleles detected
• Invasive not required (would have been required if only paternal mutation
exclusion had been available)
2015 2017

36. Conclusions
Technical challenges
• Delivering an accredited NIPD service for paternal exclusion is relatively
straightforward.
• Expanding the service to make NIPD available for rare conditions time consuming
and costly.
• High NGS sensitivity reveals mosaicism. Essential to test maternal samples in
parallel by NGS. Particular importance for panel based NGS to detect common
mutations.
Patient Perspective
• Uptake is high; patients have welcomed the availability of our bespoke NIPD
service
• Funding / Commissioning in our NHS service is an issue for this costly test. Equity
of access.
• High rate of testing for information only.

37. Acknowledgments
NETRGL
Lyn Chitty
Lucy Jenkins
Sarah Mason
Sandra Moore
Natalie Chandler
Helena Ahlfors
Funding
NETRGL
RAPID NIHR PGfAR
GOSH CC Charity
GOSH BRC
This work described here was partially funded by the National Institute for Health Research (NIHR) under its Programme
Grants for Applied Research Programme (RP-PG-0707-10107 – “RAPID”) and the GOSH BRC. The views expressed are
those of the author and not necessarily those of the NHS, the NIHR or the Department of Health.
Research Team
Samantha Edwards
Melissa Hill
Suzanne Drury
Jane Hayward