Clinical biomarker studies are often hindered by the scarcity or suboptimal quality of biological specimens. EdgeSeq, a transcriptomics analysis platform, combines quantitative nuclease protection assay technology with next-generation sequencing, using small amounts of starting material and delivering reproducible gene expression profiles from challenging material, such as formalin-fixed, paraffin-embedded (FFPE) tissue. To evaluate EdgeSeq for analysis of archives of stained FFPE tissue, EdgeSeq was performed on unstained, hematoxylin and eosin (H&E)-stained, and immunohistochemistry-stained slides from patients with small-cell and non-small-cell lung cancer. Pairwise comparisons of gene expression profiles from stained and unstained slides showed higher Pearson correlation coefficients with H&E staining (0.86 to 0.97) than with immunohistochemistry staining (0.21 to 0.56). A 25-gene interferon-γ signature score from unstained slides showed a Pearson correlation coefficient of 0.92 with H&E-stained slides and a significant Spearman correlation (P = 0.0025) with immune scores. To test gene expression profiling in small samples, FFPE sample equivalents were examined from 5.0 to 0.08 mm2 of a section (5 μm thick); sample equivalents ≥0.31 mm2 showed alignment rates >69% and pairwise Pearson correlation coefficients ≥0.87. EdgeSeq can, thus, be used to profile small and H&E-stained FFPE tumor specimens to obtain biomarker data from limited tissue in oncology clinical trials and enable research into tumor microenvironment and immune cell engagement with tumors at the locoregional level.
Copyright © 2019 American Society for Investigative Pathology and the Association for Molecular Pathology. Published by Elsevier Inc. All rights reserved.