NIH, National Cancer Institute, Division of Cancer Treatment and Diagnosis (DCTD) NIH - National Institutes of Health National Cancer Institute DCTD - Division of Cancer Treatment and Diagnosis

Comparison of the miRNA expression profiles in fresh frozen and formalin-fixed paraffin-embedded tonsillar tumors.

Author(s): Vojtechova Z, Zavadil J, Klozar J, Grega M, Tachezy R

Publication: PLoS One, 2017, Vol. 12, Page e0179645

PubMed ID: 28644855 PubMed Review Paper? No

Purpose of Paper

This paper compared the microRNA (miRNA, miR) expression profiles of matched formalin-fixed paraffin-embedded (FFPE) and frozen tonsil specimens using TaqMan Low Density Arrays (TLDA) and investigated the effects of data-handling on differentially expressed gene lists in both specimen types. miRNA expression differences between tumor and non-tumor specimens and between tumor specimens based on known human papillomavirus (HPV) status were further investigated in a second cohort by real-time PCR. 

Conclusion of Paper

As expected, the RNA integrity numbers (RIN) were higher for frozen specimens than matched FFPE specimens (>7.0 versus 1.9-2.6). Principal component analysis showed that specimens clustered by preservation method and tumor status.  The correlation in expression between matched FFPE and frozen specimens was modest to strong (R=0.4642-0.7361) with no apparent effect of tumor percentage on the level of correlation. More miRNAs were found to be differentially expressed between tumor and non-tumor frozen specimens than between tumor and non-tumor FFPE specimens and the overlap between lists was only 27-38%, regardless of normalization method employed.

When miRNA expression was investigated in 64 additional specimens, only 7 of 11 miRNAs displayed the expected differences in expression based on tumor or HPV status found in the initial cohort. Interestingly an opposite effect of tumor or HPV status was found in the larger cohort for four of the remaining five miRNAs.

Studies

  1. Study Purpose

    This study compared the miRNA expression profile in matched frozen and FFPE specimens from 10 tonsilar tumors and 5 non-malignant tonsil specimens using TLDA arrays and two different normalization methods. miRNA expression differences between tumor and non-tumor specimens and between tumor specimens based on known HPV status were further investigated in a second cohort by real-time PCR. Five of the tumor specimens were previously shown to be HPV positive while the other five were known to be HPV negative. Details of specimen processing were not included. Frozen sections were cut on a cryostat and the percentage of tumor cells were determined (>50% for all specimens). RNA was extracted from frozen specimens using the miRVana kit and from four macrodissected 10 µm sections of FFPE specimens using Ambion RecoverAll Total Nucleic Acid Isolation kit for FFPE. RNA yield was determined by spectrophotometer and chip electrophoresis. miRNA expression was investigated using the TaqMan Array Human MicroRNA A+B Cards Set v3.0. Differences in expression of 11 selected miRNAs were confirmed by individual TaqMan real-time PCR in the initial cohort and a larger cohort of 64 FFPE tonsil specimens for which no details of handling were provided.

    Summary of Findings:

    As expected, the RIN for frozen specimens were >7 but only 1.9-2.6 for the matched FFPE specimens. Principal component analysis showed that specimens clustered by preservation method and tumor status. The correlation in expression between matched FFPE and frozen specimens was modest to strong (R=0.4642-0.7361), but consideration of only the miRNAs on card A slightly increased the strength of the correlations (R=0.5194-0.8302). There was no apparent effect of tumor percentage on the level of correlation.

    Using a threshold of >1.33-fold change and P<0.05 with the 50th percentile shift normalization method, 83 miRNAs were differentially expressed between frozen tumor and non-tumor specimens and 58 were differentially expressed between tumor and non-tumor FFPE specimens. However, only 27-38% overlap was found between the lists.  When global normalization was also applied, 72 miRNAs were differentially expressed between frozen tumor and non-tumor specimens and 57 were differentially expressed between tumor and non-tumor FFPE specimens with 28-37% overlap. The overlap between lists generated by the two different normalization methods was 58-67% for frozen specimens and 62-63% for FFPE specimens.  

    In the same 15 specimens, real-time PCR confirmed the TLDA findings of higher expression of three miRNAs in tonsillar tumors compared to non-tumor specimens and higher expression of two miRNAs and lower expression of six miRNAs in HPV-positive than HPV-negative tumors. Real-time PCR analysis of 64 additional FFPE specimens confirmed increased expression of miR-155 and miR-126 in tonsillar tumor specimens compared to non-tumor tissue, but contrary to the results in the initial cohort, expression of miR-205 was found to be non-significantly higher in non-tumor specimens in the larger cohort. Similarly, only three of six miRNAs found to be expressed at higher levels in HPV-negative tumors in the initial study were also found to have higher expression in HPV-negative tumors in the larger cohort by real-time PCR (miR-16 and miR-34a significantly and miR-193b not significantly) with miR-31, miR-221, and miR-21 having non-significantly lower expression in HPV-negative tumors in the larger cohort. The initial study and the confirmatory study both found higher expression of miR-106b and miR-9 in HPV-positive tumors.

    Biospecimens
    Preservative Types
    • Frozen
    • Formalin
    Diagnoses:
    • Neoplastic - Not specified
    • Other diagnoses
    • Not specified
    Platform:
    AnalyteTechnology Platform
    RNA Real-time qRT-PCR
    RNA Low density array
    Pre-analytical Factors:
    ClassificationPre-analytical FactorValue(s)
    Biospecimen Preservation Type of fixation/preservation Formalin (buffered)
    Frozen
    Preaquisition Diagnosis/ patient condition HPV-positive tonsillar tumor
    HPV-negative tonsillar tumor
    Non-tumor
    Real-time qRT-PCR Specific Targeted nucleic acid miR-106b
    miR-9
    miR-16
    miR-34a
    miR-193b
    miR-31
    miR-221
    miR-21
    miR-155
    miR-126
    miR-205
    Low density array Specific Technology platform Real-time PCR
    Low density array Specific Data handling 50'th percentile shift normalization only
    50'th percentile shift normalization with global normalization

You Recently Viewed  

News and Announcements

  • April 24, 2024: Biobanking for Precision Medicine Seminar

  • Most Popular SOPs in March 2024

  • New SOPs Available

  • More...