Skip to main content

The utility and limitation of single nucleotide polymorphism analysis on whole genome amplified mesenchymal tumour DNA in formalin fixed tumour samples.

Liang CW, Lee YS, Marino-Enriquez A, Tsui K, Huang SH

Publication: Pathology (2012) • Vol 44 • Page 33-41PMID: 22157688View on PubMed

Purpose

The purpose of this paper was to determine the effects of whole genome amplification (WGA) and using formalin-fixed paraffin-embedded (FFPE) specimens rather than frozen specimens for genetic analysis using single nucleotide polymorphism (SNP) arrays.

Conclusion

In general, the use of unamplified DNA or DNA subjected to WGA from frozen specimens followed by unamplified DNA from FFPE specimens performed superiorly to whole genome amplified DNA from FFPE specimens with respect to probe detection, matched calls between DNA types, determination of loss of heterozygosity (LOH), and correct copy number determination. When smaller regions were considered for LOH, the false positive and negative rates using amplified DNA from frozen specimens or unamplified DNA from FFPE specimens increased as region size decreased. Loss of detection and miscalls increased when the GC content was higher than 55%, particularly in frozen specimens, but high GC content increased accuracy of copy number determination. FFPE specimens also had increased loss of detection of probes with GC content below 35%. The authors report the two WGA kits performed similarly for SNP determination except when the GC content was below 35%, in which case, REPLI-g was preferable to GenomePlex, but that GenomePlex led to more accurate determination of copy number.

Studies (1)

Purpose

The purpose of this study was to determine the effects of WGA and using FFPE rather than frozen gastrointestinal stromal tumor specimens for SNP and LOH analysis using Affymetrix NSP250k arrays. DNA was extracted with QIAamp with an increased proteinase K digestion time.

Summary of Findings

After WGA of DNA from frozen specimens, 6% of calls did not match unamplified specimens. Using amplified DNA from FFPE specimens, 71% of probes were detected, but 18% of these did not match the call from the unamplified frozen specimen. Determining loss of heterozygosity (LOH) of 10-30 Mb regions was generally good using DNA from frozen specimens (amplified or not) or unamplified DNA from FFPE specimens, but use of amplified DNA from FFPE specimens resulted in up to 33% false positives. When smaller regions were considered, the false positive and negative rates using amplified DNA from frozen specimens or unamplified DNA from FFPE specimens increased as region size decreased, and the rates were highest when amplified DNA from FFPE specimens was used, followed by unamplified DNA from FFPE specimens. Copy number was correctly assigned 68% of the time using amplified DNA from frozen specimens, 59% of the time using unamplified DNA from FFPE specimens and 48% of the time using amplified DNA from FFPE specimens. Loss of detection and miscalls increased when the GC content was higher than 55%, particularly in frozen specimens, but high GC content increased accuracy of copy number determination. FFPE specimens also had increased loss of detection of probes with GC content below 35%. The authors report the two WGA kits performed similarly for SNP determination except when the GC content was below 35%, in which case, REPLI-g was preferable to GenomePlex, but that GenomePlex led to more accurate determination of copy number.

  • Biospecimens

    TissueStomach
    TissueEsophagus
    TissueSmall Bowel
    TissueColorectal
  • Preservation Types

    FormalinFrozen
  • Diagnoses

    Neoplastic
  • Platforms

    DNAArray CGH
    DNADNA microarray
    DNAWhole genome amplification
    DNAElectrophoresis
    DNASNP assay
  • Pre-analytical Factors

    ClassificationPre-Analytical FactorValue(s)
    Biospecimen PreservationType of fixation/preservation
    Formalin (buffered)
    Frozen
    Whole genome amplification SpecificNucleic acid amplification
    Genomeplex (Sigma)
    REPL-g (Qiagen)
    Whole genome amplification SpecificTemplate/input amount
    60 ng
    10 ng
    100 ng
    150 ng
    30 ng