18.02.10 Low level mutation detection

Question 1

What is low level mutation detection?

Answer 1

Detection of a variant population of DNA in a sample where the wild-type (wt) DNA greatly exceeds the variant DNA contribution.

Need high selectivity and enrichment for successful detection and identification.

Question 2

What are the applications of low level mutation detection?

Answer 2

Somatic muts in tumour samples which are a heterogeneous mix of wt & tumour DNA (tumour itself is a heterogeneous mix of cells reducing mut level further.

Muts in fetus using NIPD, in which the cell-free fetal DNA load is only 3-6% of the entire DNA population (i.e. maternal DNA).

Heteroplasmic muts in mtDNA genomes, where mutant mtDNA is a proportion of wt mtDNA.

Question 3

Give some examples of why low level mutation detection is useful in oncology?

Answer 3

1. Early cancer detection in tissue biopsy/plasma/serum/blood-including cell free circulating tumour DNA (ctDNA), e.g. identification of muts in the Kras gene in pancreatic adenocarcinoma, lung cancer and colorectal cancer.
2. Assessment of residual disease after surgery or radiochemotherapy.
3. Disease staging/risk stratification and molecular profiling for prognosis or tailoring therapy
4. Monitoring of therapy outcome and cancer remission/relapse e.g. Minimal Residual Disease (MRD) monitoring & identification of emerging resistance 10-3 to 10-6 mutant to wt DNA.

Question 4

What is enrichment and why is it necessary for low level mutations?

Answer 4

Enrichment = the process that increases mutant allele concentration relative to wt alleles.

The use of enrichment methods is often necessary to increase the mutant concentration to a level at which accurate analysis is feasible.

Enrichment methods typically segregate by their ability to enrich either known or unknown muts.

Design of mut enrichment assays for the detection of known muts is much easier than for unknown muts as sequence data can be used and specific nucleotides can be targeted; as a result more methodologies exist for enrichment of known muts than unknown muts.

Question 5

What broad categories of enrichment categories exist for low level mutations?

Answer 5

Techniques with moderate-high selectivity

Techniques with very high selectivity.

Question 6

Many allele-specific amplification methodologies exist for known mutations. How do these techniques amplify the variant?

Answer 6

Moderate-high specificity
Preferentially destroying/blocking the wt allele
Preferentially amplifying the variant allele

Question 7

Give some examples of enrichment methods which preferentially destroy/block the wt allele.

Answer 7

1. Restriction endonuclease-mediated selective PCR (REMS-PCR)
2. Artificial introduction of a restriction site (AIRS) RFLP
3. Peptide nucleic acid-mediate PCR and Locked nucleic acid-mediate PCR clamping or wild type blocking.

Question 8

How does Restriction Endonuclease-Mediated Selective PCR (REMS-PCR) work?

Answer 8

Restriction Endonuclease-Mediated Selective PCR (REMS-PCR) – when a pathogenic mut also alters the sequence of a restriction enzyme (RE) site, inclusion of that RE during PCR will cause digestion of the wt amplicons (with the intact RE site) and amplification products only for the mutated allele (mutated RE site > not recognised by RE > PCR product).

Question 9

How does Artificial Introduction of a Restriction Site (AIRS) RFLP (Restriction Fragment Length Polymorphism) work?

Answer 9

If the mut does not alter a RE site, AIRS uses a modified primer that selectively binds to the wt allele and introduces a RE site into the wt PCR product during PCR. Exposure to the RE > digestion of wt product producing fragments of a smaller size to those of mutated PCR product > fragments separated by gel electrophoresis. Enrich alleles present at 10-3 level.

Question 10

Give some examples of enrichment methods which preferentially amplify the variant allele.

Answer 10

AS-PCR (Allele-Specific PCR; key technique),

ARMS (Amplification Refractory Mut System),

PASA (PCR Amplification of Specific Alleles),

PAMSA (PCR Amp of Multiple Specific Alleles),

Question 11

How do enrichment methods that preferentially amplify the variant allele broadly work?

Answer 11

Uses a primer with a sequence at the 3’ end that matches that of the mutant allele but not the wt allele e.g. the site of a point mut (base matching at 3’ terminus is required for primer extension and hence PCR product). Was used for detection of point muts in CF & β-thal genes. These methods can enrich alleles present at 10-1 to 10-3 level.

Question 12

What enrichment techniques are available for known mutations with have high selectivity?

Answer 12

RFLP-PCR based methods - uses thermostable RE to differentiate between wt and mutant, such as RSM-OCR and APRIL-ATM method

Digital PCR

Question 13

What are the principles of dPCR?

Answer 13

Within a sample individual nucleic acid molecules are partitioned into separate regions (Fig1). Each partition contains either a negative or positive reaction, i.e. “0” or “1” (i.e. digital output) (see figure). These regions can be generated using, micro well plates, capillaries, emulsion PCR (separation using droplet techniques based on oil-water emulsions (e.g. RainDance) and nanofluidic chips (e.g. Fluidigm). This separation allows absolute quantification and a more reliable collection and sensitive measurement of nucleic acid amounts.

Question 14

What are the applications of dPCR?

Answer 14

dPCR can be used to detect both known & unknown muts (known=allele specific fluorescent probe detection using real time-PCR, unknown=use NGS to sequence products) to a level of 10-3.

The limit of detection is defined by the no of regions that can be analysed and the intrinsic rate of the polymerase used for the application (=PCR errors). 36,960 chamber plate is available from Fluidigm; they have also developed a 200k chamber plate.

Question 15

What is random mutation capture PCR?

Answer 15

Random Mutation Capture (RMC)-PCR a combination of RSM and digital PCR; DNA digestion step prior to PCR amplification in order to remove wt DNA (with non-mutated RE sites). This ensures that only molecules known to contain a mut are isolated. The product then goes through a digital PCR step (is diluted to isolate single mols & qPCR is performed to amplify the mutant seqs). Allows the detection of variants present at 10-8

Question 16

What are some of the applications of RMC-PCR?

Answer 16

Note: The differences in DNA methylation between maternal and fetal DNA can be exploited using sodium bisulphite converted DNA in combination with dPCR for the quantification of cffDNA in maternal plasma.

E.g. (Ref 14): Use of dPCR and Raindance to detect IDH1 mRNA in glioma patient cerebrospinal fluid samples. These muts could be used to track tumours and also inform treatment decisions if drugs are developed to treat such muts.

Question 17

What methodologies exist for the enrichment of unknown mutations?

Answer 17

COLD-PCR

NGS

Question 18

Describe COLD-PCR

Answer 18

COLD-PCR (Co-amplification at lower denaturation temperature PCR; key technique): A single-step method that can be used for the enrichment of both known and unknown minority alleles during PCR, irrespective of mut type and position.

Can combine COLD-PCR with downstream method e.g. sequencing, MALDI-TOF, Pyrosequencing, real-time TaqMan PCR to improve detection sensitivity by up to 100 fold.

For a DNA sequence close to a critical denaturation temperature (Tc), the percent denaturation becomes sensitive to the exact DNA sequence and any mismatches along the dsDNA sequence.

In practice, the method has been shown to be unpredictable for enrichment.

Question 19

What are the advantages of COLD-PCR?

Answer 19

Advantages include: Single-step method capable of enriching both known and unknown minority alleles irrespective of mutation type and position. Does not require any extra reagents or specialized machinery. Therefore the cost is not increased. Better than conventional PCR for the detection of mutations in a mixed sample. Does not significantly increase experiment run time compared to conventional PCR.

Question 20

What are the disadvantages of COLD-PCR?

Answer 20

Disadvantages include: vulnerable to polymerase induced error, needs sequences less than 200bp, all variants may not have a Tc that differs significantly from the wt, success depends on the sequence and position of the variant as this affects the dynamics of amplification.

Question 21

What the the types of COLD-PCR?

Answer 21

Full
Fast
ICE

Question 22

Describe Full COLD-PCR

Answer 22

Enrichment of all possible muts.

Induces the formation of heteroduplexes at positions of muts.

By using a lower denaturation temperature during PCR, double-stranded DNA (dsDNA) containing mismatches (heteroduplexes) denature first.

True homoduplexes have a higher melting temperature (Tm) & denature less than heteroduplexes at the Tc; thus their amplification is relatively suppressed

Question 23

Describe fast COLD-PCR.

Answer 23

Fast – Enrichment of known Tm-reducing muts (vs. wt).

Selectively denature only variant sequences by denaturing at the Tc and then only amplify these

Question 24

Describe ICE.

Answer 24

ICE (Improved and Complete Enrichment):

Improvement of mut enrichment in Tm-neutral and Tm-raising muts:

A ss oligonucleotide (60-90nt) complimentary to the sense strand of the wt sequence is added.

The oligo contains a 3’ non-extendible phosphate group and so when anneals to wt, PCR amplification is inhibited; the variant sequence is amplified and enriched.

Question 25

What are the challenges of detecting low level variants by NGS?

Answer 25

population skewing due to differential amplification in heterogeneous mixtures

polymerase mistakes resulting from base misincorporations and rearrangements due to template switching

In order to detect low level muts by NGS, much higher read depth (deep sequencing) is needed

Seq error is unevenly distributed throughout the genome; may be influenced by the seq context, position on the read, and molecule structure. Results in sequencing error ‘hot spots’ where the error rate can be 10x greater than the genome average. ~1% of bases are incorrectly identified, depending on the specific platform and sequence context.

To detect rare single nucleotide variants at 0.1% requires robust processing methods, data pipeline, 10,000’s high quality parallel reads (see: Cushing et al 2013). Beyond this level (<0.1%) it begins to become impractical (unit cost, PCR resampling, data volume) to use NGS read depth alone to detected minority populations.

Question 26

What approaches are there to improve sequencing accuracy at the biochemical and data processing level?

Answer 26

Exogenous tagging (spike-ins)
Duplex sequencing
Tagged amplicon sequencing.

Question 27

How is exogeneous tagging used to improve sequencing accuracy?

Answer 27

Before amplification enabling identification of all amplicons derived from a particular starting molecule so any variation in the seq or copy number of identically tagged seq reads can be discounted as technical error. E.g. Kinde et al, reported a 20x reduction in error freq with a tagging method that is based on labeling ssDNA fragments with a primer containing a 14-bp degenerate seq.

Question 28

How is duplex sequencing used to improve sequencing accuracy?

Answer 28

Schmitt et al, report a technique called Duplex sequencing that was used to detect mtDNA muts .

Both strands of the DNA duplex are tagged and sequenced. True muts are found at the same position in both strands. In contrast, PCR or sequencing errors result in muts in only one strand and can be discounted as technical error (error rate of less 1 in a billion).

Question 29

How is tagged amplicon sequencing used to improve sequencing accuracy?

Answer 29

ctDNA: Forshew et al, demonstrates the feasibility of detecting muts at 2% allele frequency, in multiple loci, directly from blood plasma in 88 patient samples. Using Tagged Amplicon sequencing or TAm-seq with fludigm access array and Illumina sequencing. Tailed locus-specific primers were used for PCR amplification. Each sample is pre-amplified in a multiplex PCR reaction to enrich for all targeted loci.