Cancer Diagnosis Program | Biorepositories and Biospecimen Research Branch

Search BRD Papers

Identification and validation of differentially expressed transcripts by RNA-sequencing of formalin-fixed, paraffin-embedded (FFPE) lung tissue from patients with Idiopathic Pulmonary Fibrosis.

Author(s): Vukmirovic M, Herazo-Maya JD, Blackmon J, Skodric-Trifunovic V, Jovanovic D, Pavlovic S, Stojsic J, Zeljkovic V, Yan X, Homer R, Stefanovic B, Kaminski N

Publication: BMC Pulm Med, 2017, Vol. 17, Page 15

PubMed ID: 28081703 PubMed Review Paper? No

Purpose of Paper

Conclusion of Paper

Studies

1. Study Purpose

   This study investigated the effects of specimen storage on NGS results of FFPE specimens and compared lists of differentially expressed genes between patients with IPF and healthy controls obtained using NGS of FFPE specimens, with those obtained by microarray analysis of unmatched frozen specimens and NanoString analysis of unmatched FFPE and frozen specimens. FFPE Lung biopsies of seven IPF and six normal tissues were stored for a median of six years before sectioning. For each FFPE specimen, two RNA isolations were performed on five 10 µm sections that were deparaffinized in xylene twice using the MasterPure kit.  RNA yield was determined by spectrophotometry and integrity by bioanalyzer. rRNA was removed using the TiboZer rRNA removal kit and libraries were prepared using NEBNext Ultra Directional RNA Library Prep kit. Libraries were sequenced using the Illumina HiSeq 2000. Levels of 10 genes with significant differences and 15 without significant differences were further evaluated by NanoString.  NGS data was compared with Agilent microarray data from 19 IPF and 16 pathologically-normal age and gender matched frozen specimens. An additional seven healthy control and eight IPF frozen specimens and five healthy control and seven IPF FFPE specimens were analyzed by NanoString.

   Summary of Findings:

   Although all FFPE specimens generated pure RNA with an optical density (OD) ratio of 260 to 280 of greater than 1.9, the RNA was highly degraded, as indicated by low RNA integrity numbers (2.1-2.6). The RNA degradation was unaffected by disease state or storage duration.  Overall, an average of 116 million reads were produced, of which, ~62 million were mapped, but one FFPE specimen generated only 15.9 million reads and had a mapping rate of only 20.94%. The mapping rate was lower for specimens stored for more than 7 years (40-50 million reads) than when all specimens were considered (62 million). Of the 23,615 annotated genes, sufficient coverage was produced for 15,149 genes of which 4,131 were differentially expressed in IPF versus controls (1,920 increased and 2,211 decreased in IPF). Based on expression patterns, the specimens segregated by diagnosis. Of the 1,920 genes found to be increased in FFPE IPF cases compared to controls by NGS, 760 were also found to be increased in frozen IPF cases relative to controls by microarray. Similarly, of the 2,211 genes found to be decreased in FFPE IPF cases compared to controls by NGS, 1,413 were also found to be decreased in frozen IPF cases relative to controls by microarray. Importantly, a hypergeometric test showed that the overlap in differentially expressed genes was highly significant (P<10^-182) and not due to random association.  A total of 92 genes found to be differentially expressed by NGS in FFPE specimens were found to be differentially expressed in the opposite direction by microarray analysis of frozen specimens. For these genes, the hypergeometric test had a probability of approximately 1, indicating these changes were due to random association.  There were 11,039 genes that were not differentially expressed in IPF in either the FFPE specimens analyzed by NGS or the frozen specimens analyzed by microarray. Pathway analysis showed enrichment of genes in pathways known to be affected by IPF. Further, 14 of the 21 genes in the MMP7 had comparable changes in IPF in the frozen specimen analyzed by microarray and the FFPE specimen analyzed by NGS.

   Expression by NanoString was very strongly correlated with microarray expression for frozen specimens (r=0.92) and with RNA-seq for FFPE specimens (r=0.90).  All 15 genes with concordant differences between IPF and control were also found by NanoString. Of the 10 genes with discordant differences between IPF and control in frozen specimens analyzed by NGS and FFPE specimens analyzed by microarray, only four were discordant between frozen and FFPE specimens using NGS.

   Biospecimens

   • Tissue - Lung

   Preservative Types

   • Formalin
   • Frozen

   Diagnoses:

   • Other diagnoses
   • Normal

   Platform:

   Analyte	Technology Platform
   RNA	Automated electrophoresis/Bioanalyzer
   RNA	DNA microarray
   RNA	Next generation sequencing
   RNA	Spectrophotometry

   Pre-analytical Factors:

   Classification	Pre-analytical Factor	Value(s)
   Biospecimen Preservation	Type of fixation/preservation	Formalin (buffered) Frozen
   Storage	Storage duration	>7 years <7 years
   Preaquisition	Diagnosis/ patient condition	IPF Healthy
   DNA microarray Specific	Technology platform	NanoString Agilent microrray
   Next generation sequencing Specific	Technology platform	Agilent microarray NanoString

   View more