Non-invasive Testing - Part 2

Question 1

Outline the principle of how digital PCR works.

Answer 1

• The principle of this is that if you dilute the template DNA such that you have just less than 1 molecule per well and then do thousands of PCRs you actually get some PCRs which are positive and some which are negative.
• By counting the positives from 1 molecule you get an absolute quantification of starting DNA without the need for standards and therefore it is more accurate than real time PCR.
• In practice you segregate individual molecules into separate wells so that there is no interference between those molecules in the PCR.
• This number of replicates is possible with nano fluidics.
• By using this technique you can check if the foetus has inherited maternally inherited point mutations by using relative mutation dosage (RMD).

Question 2

Describe the principle and use of Relative Mutation Dosage (RMD).

Answer 2

• Relative Mutation Dosage can be used to actually detect whether the foetus has inherited the mutation from the mother against all the excess maternal cfDNA in the test sample.
• Used in pregnancies where both parents carry the same mutation.
• The mother is heterozygous for the location of interest.
• The principe of this is that we are looking at a situation where the mother is heterozygous for a particular mutation. If the foetus is homozygous then there will be a greater proportion of mutant signal to wildtype signal compared to the heterozygote result where there would be an equal proportion and a wildtype result in the foetus where there would be an excess of the wildtype gene.
• Using digital PCR or NGS.
• Allelic ratio between mutant and wildtype alleles determined by counting maternal and foetal contributions.
• The fraction of foetal DNA concentration directly affects the expected extent of allelic imbalance. Therefore it is crucial in this calculation to know what proportion of the cfDNA is from the foetus.
• The proportion of cffDNA varies between samples.

Question 3

How do you determine the fractional cffDNA concentration?

Answer 3

• Requires a foetal allele not present in the maternal tissue.
• calculate the percentage of foetal marker out of total DNA.
• For male foetus we can use SRY or DYS14.
• For female foetus there is no universal foetal marker available.
• 24 biallelic indel polymorphisms were tested for suitability using digital PCR assays.
• Amplicons over 143bp were redesigned to make them shorter or excluded from further analysis.
• 13 currently identified as being informative.
• Another informative marker would be HLA typing using MiSeq NGS.

Question 4

Outline Illumina NGS.

Answer 4

• The DNA is fragmented and adaptors are ligated onto the ends.
• DNA is then attached to a solid matrix and PCR takes place at the various sites so you get multiple DNA molecules amplified from a single DNA molecule.
• Primers are attached and bases are added - each base with a different colour - you get parallel sequencing of millions of these DNA fragments.
• NGS and digital PCR have a similar sensitivity for NIPT but SNGS is more cost effective as samples can be multiplexed.
• More mutations are covered in a single NGS assay.

Question 5

Describe NIPT for trisomy testing.

Answer 5

• Trisomy 21 occurs in about 1 in 700 births an currently in the UK primary testing is commonly done by risk stratification involving a combination of USS, NT testing and serum biomarkers followed by invasive testing if they are at high risk (1% miscarriage risk). Therefore non invasive technique is desireable.
• Screening = phenotypic features not genetic pathology.
• Sensitivity and specificity suboptimal.
• Combined use, strict gestational window.
• Problem with developing trisomy 21 test using NIPT mainly the low proportion of cffDNA.
• There were efforts to lower % of maternal DNA with formaldehyde to stabilise cells before centrifugation - initial findings not replicated by other groups.
  1) . One approach is to look at heterozygous SNPs on chromosome 13, 18, 21 - in trisomy should be a ratio of 1:2 or 1:1:1 instead of 1:1. Need method to measure only copy number from foetal cells. Reliant on informativeness of polymorphisms. Similar concept to current method for rapid trisomy testing with CVS, amnio using QF-PCR.
  2) . Could measure the copy number of chromosome 13, 18 and 21 by accurately measuring foetal specific sequences on those chromosomes where there is a difference in methylation between the foetal genes and maternal genes which occur on those chromosomes.
  3) . Could measure the number of sequences obtained from chromosomes 13, 18 and 21 by very accurate measurement of foetal and maternal DNA and compare them to the number of sequences obtained from other chromosomes to determine if there is a chance that there are any extra copies of any of these sequences in the foetal material.
• This testing has been developed and is available in the USA and China.

Question 6

Outline the use of heterozygous SNPs for foetal copy number measurement.

Answer 6

• It is possible to use a combination of SNPs and genes that are differentially methylated between the trophoblast and maternal tissues, or actually are expressed only in the placenta
• Use differential methylation of genes between trophoblast and maternal tissue in chromosome 13, 18, 21 (epigenetic allelic ratio approach).
  e. g. maspin (SERPINB5) on chromosome 18 = hypomethylated in placental tissues, hypermethylated in maternal blood cells - use polymorphisms. Extensive searches for similar chromosome 21 genes.
• Quantitative analysis of polymorphic alleles in free foetal RNA for genes only expressed in placenta (RNA-SNP approach).
  e.g. PLAC4 on chromosome 21 - 90% sensitivity and 96% specificity.
  CffRNA = stable.
• Needs panel of SNPs - not informative in all cases.

Question 7

Outline the epigenetic-genetic chromosome dosage approach to trisomy testing in NIPT.

Answer 7

• You count the number of sequences left after digesting away maternal DNA.
• Chromosome 21 - promoter of holocarboxylase synthetase (HLCS) gene hypermethylated in placenta, hypomethylated in blood.
• Digest away maternal hypomethylated DNA with BstU1.
• Count number of sequences relative to control (ZFY on Y in male foetuses or could also be a paternal SNP allele).
• Unlike epigenetic allele ratio method does not require the methylation specific marker and SNP to be present together on a short stretch of DNA.
• Can also compare the HLCS gene agains the RASSF1a marker on Chr3.
• A problem with DNA methylation approaches is that there is often an overlap in ratio ranges and they do not seem to be very accurate. May be as a result of variation in methylation.

Question 8

How can digital PCR be used for trisomy detection from cffDNA?

Answer 8

• Has been shown that simply counting the numbers of copies of chr21 in the cffDNA sample can demonstrate over representation of chr21 amplicon in trisomy 21 cases.
• However, often only work if there is a very large proportion of cffDNA in their sample. The problem fund was that higher numbers of PCR analyses or foetal enrichment required when foetal DNA concentrations were lower.
• Working close to the limits of the tech for this.
• Actually a very small difference between levels of DNA in trisomy and non trisomy samples.

Question 9

What is nucleic acid size selection (NASS)?

Answer 9

• NASS is a method that has been proposed for increasing the relative proportion of foetal DNA which is actually tested from the cell free DNA fraction.
• Relies on the fact that the average size of the cell free DNA from the foetus appears to be smaller than that from the mother.
• Relative mutation dosage works in the same way as for trisomy detection.
• Similar problems re low foetal DNA concentration.
• Nucleic acid size selection.
• Foetal DNAs shorter than maternal - can enrich by selecting for shorter DNAs.
• Physical.
• Digital nucleic acid size selection NASS.
• Digital duplex digital PCR targeting overlapping amplicons - only wells showing the presence of short DNA molecules counted.
• Has been used for improvement of the relative mutation dosage approach.

Question 10

How can we overcome the problem of low concentration of cffDNA for NIPT for trisomy 21 for example?

Answer 10

1) . Enrich foetal DNA.
2) . Increase PCRs if using digital PCR.
3) . Could increase statistical power by only sequencing the relevant chromsomes but sequencing them more times (increase sampling of the same chromosome).

Can use target enrichment strategies.

Question 11

What might be the benefit of using NIPT for trisomy testing?

Answer 11

• False positive rate for current screening programme is about 5% and this means that 5% of women have inappropriate invasive testing.
• With 2-plex test could rule out trisomy 21 in 98% of these cases = reduced false positive.
• Conceivable that could use it as a first tier screening test but would be expensive due to the number of women being tested.

Question 12

Where is the cffDNA actually derived from? What problems can this cause? What are some other things that may confound the use of cffDNA?

Answer 12

• ccfDNA is actually derived from the placenta.
• You can therefore get potential problems including confined placental mosaicism for problems that are not present in the foetus.
• Other potential problems - demised co-twin, maternal malignancy, maternal mosaicism for aneuploidy etc.

Question 13

What does the future hold for NIPT?

Answer 13

• NIPT to detect large deletions and duplications in the chromosomes.
• ?Sequencing the whole foetal genome/exome.