Skip to main content

Copy-number analysis of topoisomerase and thymidylate synthase genes in frozen and FFPE DNAs of colorectal cancers.

Yu J, Miller R, Zhang W, Sharma M, Holtschlag V, Watson MA, McLeod HL

Publication: Pharmacogenomics (2008) • Vol 9 • Page 1459-66PMID: 18855534View on PubMed

Purpose

The purpose of this paper was to determine the effects of using formalin-fixed paraffin-embedded (FFPE) instead of snap-frozen colorectal tumor specimens on copy number determination by real-time PCR.

Conclusion

Despite using the same quantity of input DNA, the cycle thresholds (CTs) for topoisomerase I (TOP1), Thymidylate synthase (TYMS), long interspersed repeat element-1 (LINE-1) and beta actin were more than 10 fold lower for FFPE colorectal tumor specimens than frozen colorectal tumor specimens, which prevented copy number determination in 12 of the 46 FFPE specimens. TOP1 and TYMS copy numbers were generally highly correlated between FFPE and frozen colorectal tumor specimens (r2=0.86 and r2=0.75, respectively, when normalized to beta actin), and only one copy number difference was identified (1.59 versus 1.33). The TOP1 mRNA and protein levels were affected by TOP1 copy number, but TYMS mRNA and protein levels were not affected by copy number.

Studies (1)

Purpose

The purpose of this study was to determine the effects of using FFPE instead of snap-frozen OCT-embedded colorectal tumor specimens on copy number determination by real-time PCR. Copy numbers were compared to RNA and protein levels. The effects of determining copy number from FFPE glioma specimens using real-time PCR instead of fluorescent in situ hybridization (FISH) were also investigated. After overnight proteinase K digestion, DNA was extracted from 50 um sections of snap-frozen and OCT embedded specimens using QIAmp and from FFPE specimens using phenol:chloroform:isoamyl alcohol.

Summary of Findings

Reliable copy number data was obtained from all 52 frozen and 34 of 46 FFPE colorectal cancer specimens, with exclusion of 12 FFPE specimens due to CT vales over 35. Further, despite using the same quantity of input DNA, the CTs for TOP1, TYMS, LINE-1 and beta-actin were more than 10 fold lower for FFPE colorectal tumor specimens than frozen colorectal tumor specimens. The CT target to reference ratios were dependent on fixation method, so copy number had to be normalized to a similarly fixed specimen with no copy number changes. TOP1 and TYMS copy numbers were generally highly correlated between FFPE and frozen colorectal tumor specimens (r2=0.86 and r2=0.75, respectively, when normalized to beta actin), and only one copy number difference was identified (1.59 versus 1.33). The TOP1 mRNA and protein levels were affected by TOP1 copy number (p<0.001 and p<0.01 respectively), but TYMS mRNA and protein levels were not affected by copy number. Using real-time PCR to determine epidermal growth factor receptor (EGFR) copy numbers from 18 FFPE gliomas (normalized to FFPE colorectal cancer), the copy number was slightly underestimated compared to copy numbers determined previously by FISH. The authors postulate this was most likely due to dilution of the cancer cells with normal cells.

  • Biospecimens

    TissueColorectal
  • Preservation Types

    OCTFormalin
  • Diagnoses

    NeoplasticCarcinoma
    NeoplasticOther
  • Platforms

    DNAReal-time qPCR
    RNAReal-time qRT-PCR
    ProteinImmunohistochemistry
  • Pre-analytical Factors

    ClassificationPre-Analytical FactorValue(s)
    Biospecimen PreservationType of fixation/preservation
    Formalin (buffered)
    OCT
    Real-time qPCR SpecificTargeted nucleic acid
    Beta actin
    EGFR
    LINE1
    TOPI
    TYMS
    Real-time qPCR SpecificTechnology platform
    FISH
    Real-time qRT-PCR SpecificTargeted nucleic acid
    TOP1
    TYMS
    Immunohistochemistry SpecificTargeted peptide/protein
    TOPI
    TYMS