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

miRNA analysis of formalin-fixed squamous cell carcinomas of the tongue is affected by age of the samples.

Author(s): Rentoft M, Fahlén J, Coates PJ, Laurell G, Sjöström B, Rydén P, Nylander K

Publication: Int J Oncol, 2011, Vol. 38, Page 61-9

PubMed ID: 21109926 PubMed Review Paper? No

Purpose of Paper

This paper assessed whether storage of formalin-fixed, paraffin-embedded (FFPE) tissue blocks for up to 50 years adversely affected microRNA (miRNA, miR) analysis by microarray and qRT-PCR, and aimed to identify analytical methods to adjust for storage-induced changes.

Conclusion of Paper

There was no correlation between total RNA yield and the duration of FFPE block storage, although yields were lower in benign than cancerous tongue specimens. Unsupervised hierarchical clustering and principal component analysis (PCA) of miRNA microarray data revealed clustering, and the two largest sources of variation were identified as FFPE block storage duration (56%) and specimen type (carcinoma or benign; 13%). Effects on miRNA levels determined by microarray were observable after one year of FFPE block storage.  Expression of only 16 of the 808 miRNAs evaluated by the array were significantly different between tumor and benign tongue specimens (P<0.01). Further analysis of b-coefficient values indicated miRNA-specific effects of FFPE block storage. A new normalization method reported by the authors increased the number of differentially expressed miRNA between tumor and benign specimens from 16 to 54 miRNAs. qRT-PCR confirmed differences in 3 (miR-424, miR-21, and miR-203) of the 54 miRNAs that were differentially expressed between FFPE tumor and benign specimens  with case-matched tumor and normal adjacent specimens that were fresh frozen. Susceptibility to block storage-induced effects was significantly associated with the number of consecutive adenines and guanines in the miRNA sequence (P=0.004), but not with not the total number of each nucleotide.

Studies

  1. Study Purpose

    This study assessed whether storage of formalin-fixed, paraffin-embedded (FFPE) tissue blocks for up to 50 years adversely affected miRNA analysis by microarray and qRT-PCR, and aimed to identify methods to adjust the data collected to compensate for storage-induced changes. Twenty-one squamous cell carcinoma and eight benign specimens were formalin-fixed and paraffin-embedded, and stored as blocks under unspecified conditions for <1 y to > 50 y.  Case-matched tongue squamous cell carcinoma and normal adjacent specimens from an additional five patients were fresh frozen.  Additional details on formalin fixation and paraffin processing, and freezing were not provided. RNA was extracted from ten 5 µm-thick FFPE sections with the High Pure miRNA isolation kit, and from frozen specimens with the trizol method. Isolated miRNA was stored at -80°C until analysis.  RNA concentration and quality was determined by Nanodrop spectrophotometry. Expression profiling of miRNA was performed with the miRCURY LNA array v 11.0. Normalization of microarray data included background correction, dye-normalization, and array normalization across arrays by the median absolute deviation (MAD) method.  A new and additional normalization method based on linear regression was also applied. Differences observed with microarray were confirmed by qRT-PCR.

    Summary of Findings:

    There was no correlation between total RNA yield and the duration of FFPE block storage, although yields were lower in benign  than cancerous tongue specimens. Unsupervised hierarchical clustering of miRNA microarray data generated three main clusters that were based on differences in FFPE block storage duration. Principal component analysis (PCA) confirmed that clustering was modestly correlated to differences in FFPE block storage duration (r2=0.56).   The largest source of variation was attributable to the duration of FFPE block storage (56%), followed by specimen type (carcinoma or benign; 13%). Effects on miRNA levels were observable after one year of FFPE block storage.  Of the 808 miRNAs evaluated by the array, 16 displayed differential expression between tumor and benign tongue specimens. Of these 10 miRNAs were expressed at higher levels and 6 at lower levels in squamous cell carcinoma compared to benign specimens (P<0.01).  When b-coefficient values were calculated, miRNAs were divided into two groups, suggesting that the rate of effect with respect to block storage duration is miRNA specific.  A new normalization method reported by the authors increased the number of differentially expressed miRNA between tumor and benign specimens from 16 to 54.  qRT-PCR confirmed differences in 3 (miR-424, miR-21, and miR-203) of the 54 miRNAs that were differentially expressed between FFPE tumor and benign specimens  with case-matched tumor and normal adjacent specimens that were fresh frozen. Levels of miR-21 and miR-424 were 4.3 and 4.0-fold higher in tumor specimens than normal adjacent specimens, respectively, while miR-203 was 2.6-fold lower.  Susceptibility to block storage-induced effects was significantly associated with the number of consecutive adenines and guanines  in the miRNA sequence (P=0.004), but was not associated with not the total number of each nucleotide.

    Biospecimens
    Preservative Types
    • Frozen
    • Formalin
    Diagnoses:
    • Neoplastic - Carcinoma
    • Neoplastic - Normal Adjacent
    • Neoplastic - Benign
    Platform:
    AnalyteTechnology Platform
    RNA DNA microarray
    RNA Real-time qRT-PCR
    RNA Spectrophotometry
    Pre-analytical Factors:
    ClassificationPre-analytical FactorValue(s)
    Storage Storage duration 0.8 y
    1.3-1.7 y
    2.1-2.8 y
    3.0-3.4 y
    4.0-4.7 y
    5.7 y
    6.5-6.7 y
    7.4-7.9 y
    8.0-8.9 y
    9.8 y
    10.8 y
    11.2 y
    >50 y
    Real-time qRT-PCR Specific Targeted nucleic acid miR-424
    miR-21
    miR-203
    Preaquisition Diagnosis/ patient condition Squamous cell carcinoma
    Benign

You Recently Viewed  

comments powered by Disqus

News and Announcements

  • New Article Published from NCI's Biospecimen Pre-analytical Variables Program

  • SOPs from the Biospecimen Pre-analytical Variables (BPV) Program are now available!

  • New SOPs added to the BRD!

  • More...