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

Residual tissue repositories as a resource for population-based cancer proteomic studies.

Author(s): Piehowski PD, Petyuk VA, Sontag RL, Gritsenko MA, Weitz KK, Fillmore TL, Moon J, Makhlouf H, Chuaqui RF, Boja ES, Rodriguez H, Lee JSH, Smith RD, Carrick DM, Liu T, Rodland KD

Publication: Clin Proteomics, 2018, Vol. 15, Page 26

PubMed ID: 30087585 PubMed Review Paper? No

Purpose of Paper

This paper evaluated the quantity and quality of isolated peptides and proteomic and phosphoproteomic profiles of archived formalin-fixed paraffin-embedded (FFPE) ovarian adenocarcinoma specimens that were stored as FFPE blocks for different durations (7-32 y) at different registries of the Surveillance, Epidemiology, and End Results (SEER) Program.  Proteomic and phosphoproteomic profiles of the SEER FFPE specimens were also compared to frozen OCT-embedded ovarian serous carcinoma specimens from The Cancer Genome Atlas (TCGA) Program that had been stored for 3-10 y (temperature not specified).

Conclusion of Paper

While peptide yields, protein abundance, and expression proteome profiles and phosphoproteome profiles did not differ significantly between SEER registry sites or due to the duration of FFPE block storage (which ranged between 7 and 32 y), peptide yields did exhibit considerable variation among archived FFPE SEER specimens. Archived FFPE SEER high grade serous adenocarcinoma specimens analyzed by tandem mass tag (TMT)-10 LC/MS-MS and archived TCGA frozen OCT-embedded high grade ovarian serous carcinoma specimens analyzed by isobaric tags for relative and absolute quantitation (iTRAQ)-4 differed substantially from one another despite the correction factors applied; the number of peptides and phosphopeptides were approximately 50% and 76% lower in FFPE specimens, respectively, relative to frozen OCT-embedded specimens. While DNA-binding proteins were a consistently affected gene ontology (GO) category when FFPE and frozen OCT specimens were compared, differences in protein extraction protocols made it difficult for the authors to determine whether differences were introduced during formalin fixation or protein extraction.  To minimize potential differences associated with analysis by two different assays, the authors applied correction factors that were identified in an unpublished study in which case-matched breast tumor specimens were results obtained with each assay were compared.

Studies

  1. Study Purpose

    This paper evaluated the quantity and quality of isolated peptides and proteomic and phosphoproteomic profiles of archived formalin-fixed paraffin-embedded (FFPE) ovarian adenocarcinoma specimens that were stored as FFPE blocks for different durations (7-32 y) at different registries of the Surveillance, Epidemiology, and End Results (SEER) Program.  Spectrum identification rates, protein abundance, proteome and phosphoproteome profiles, and results of gene set enrichment analysis (GSEA) were compared among the same archived FFPE SEER specimens and frozen OCT-embedded ovarian serous carcinoma specimens from The Cancer Genome Atlas (TCGA) Program that had been stored for 3-10 y (temperature not specified).

     In total, 60 high grade serous ovarian adenocarcinoma FFPE specimens that were archived at three different SEER registry sites and represented a range of storage durations (13 specimens were stored for 3-12 y, 31 specimens were stored for 13-22 y, 15 specimens were stored for 23-32 y, and age was not recorded for 1 specimen) were used for comparison. Formalin fixation times and conditions, and the conditions of FFPE block storage were not recorded. FFPE sections flanking those used for proteomic analysis were used to confirm that each sample had ≥50% tumor nuclei and ≤50% necrotic cells. Five 10 µm-thick FFPE sections were placed in a tube and deparaffinized with xylene, treated with a phosphatase inhibitor cocktail, incubated with a protease inhibitor, followed by trypsin before extraction using an optimized FFPE protocol and centrifugation to remove cellular debris.  Protein concentration was quantified by the bicinchoninic acid (BCA) assay. A portion of each sample was pooled and used as a tandem mass tag (TMT) reference of normalization before isobaric labeling of individual samples and the pooled sample. Eighteen of the 60 samples (>400 µg protein yield) underwent phosphopeptide enrichment using magnetic Fe3+ NTA agarose beads before being TMT labeled and fractionated by capillary liquid chromatography (LC). Peptide samples were further separated by ultra-performance LC (UPLC) and analyzed using a Q-Exactive PI mass spectrometer. Phosphopeptides were further separated by UPLC and analyzed using a Q-Exactive HF mass spectrometer. Spectra were matched against the RefSeq human protein sequence database. Potentially significant differences in TMT reported intensities due to SEER registry site and FFPE block storage duration were evaluated using linear modeling. Protein abundance (estimated from spectral counts) from FFPE specimens were compared to those from frozen OCT-embedded high grade ovarian serous carcinoma specimens stored by The Cancer Genome Atlas (TCGA) for 3-12 y, although no additional details were provided.

    Summary of Findings:

    Proteomic analysis of 60 high grade serous ovarian adenocarcinoma FFPE specimens yielded 64,682 peptides corresponding to 8,582 proteins and 8,073 phosphopeptides representing 3,089 phosphoproteins. While peptide yields exhibited substantial variation among the archived SEER FFPE specimens (135-560 µg), peptide yields did not display significant differences due to registry site or the duration of FFPE block storage regardless of whether comparisons were based on total protein yield (from five 10 µm-thick sections) or yield was normalized to the estimated tumor volume. Further evaluation by principle component analysis (PCA) and ANOVA confirmed that the expression proteome, protein abundance, and the phosphoproteome were not substantially or significantly affected by SEER registry site or the duration of FFPE block storage.

    When two similar tumor types (serous ovarian adenocarcinoma and serous ovarian carcinoma) were used to compare FFPE and frozen OCT-embedded specimens that were analyzed using similar but different assays (TMT-10 and iTRAQ-4, respectively), FFPE specimens exhibited substantially lower numbers of identified peptides (~50% lower in FFPE SEER specimens than OCT-embedded TCGA specimens) and phosphopeptides (~76% lower in FFPE than OCT-embedded frozen specimens), and reduced identification rates.  These results were observed even after correction factors were applied that were identified in a separate study of case-matched breast tumor specimens analyzed by the two platforms.  When FFPE specimens were compared to frozen OCT-embedded specimens, DNA-binding proteins were a consistently affected GO category, although the authors were unable to determine if effects were introduced during formalin fixation or extraction because different protein extraction protocols were applied.

    Biospecimens
    Preservative Types
    • OCT
    • Formalin
    Diagnoses:
    • Neoplastic - Carcinoma
    Platform:
    AnalyteTechnology Platform
    Peptide LC-MS or LC-MS/MS
    Pre-analytical Factors:
    ClassificationPre-analytical FactorValue(s)
    Storage Storage duration 7-12 y
    13-22 y
    23-32 y
    Biospecimen Preservation Type of fixation/preservation Formalin (buffered)
    OCT
    Biospecimen Acquisition Locale of biospecimen collection 4 Different SEER Program Registries

You Recently Viewed  

News and Announcements

  • April 24, 2024: Biobanking for Precision Medicine Seminar

  • Most Popular SOPs in March 2024

  • New SOPs Available

  • More...