Cancer Diagnosis Program | Biorepositories and Biospecimen Research Branch

Search BRD Papers

Gene Expression Profiling of FFPE Samples: A Titration Test.

Author(s): Manjunath HS, Al Khulaifi M, Sidahmed H, Ammar A, Vadakekolathu J, Rutella S, Al-Mohannadi MJ, Elawad M, Mifsud W, Charles A, Maccalli C, Tomei S

Publication: Technol Cancer Res Treat, 2022, Vol. 21, Page 15330338221129710

PubMed ID: 36415121 PubMed Review Paper? No

Purpose of Paper

Conclusion of Paper

Studies

1. Study Purpose

   The purpose of this paper was to compare the yield and quality of RNA extracted from non-matched pediatric formalin-fixed paraffin embedded (FFPE) gut biopsies using two different kits.  The impact of RNA integrity and input amount on expression profiling using NanoString was investigated by comparing results obtained using two different input amounts and from five FFPE gut specimens with different DV200 values. The study included 11 FFPE gut tissues surgically resected from patients with colorectal dysplasia or neoplasia and a history of inflammatory bowel disease (IBD) and 35 FFPE biopsies from pediatric patients with either IBD, inflamed, or normal tissues. Details regarding FFPE processing and block storage were not included. RNA was extracted from two 10 µm sections using either the AllPrep DNA/RNA FFPE Kit (all 11 adult specimens and 20 pediatric biopsies) or the RNAstorm FFPE RNA Extraction Kit (15 pediatric biopsies).  RNA yield was quantified by NanoDrop and RNA integrity was evaluated using the Agilent RNA 6000 Pico Kit on a Bioanalyzer instrument.  Gene expression in five specimens from adult patients with different DV200 values was assessed using the PanCancer IO360 Panel on a nCounter Digital Analyzer at 555 fields of view. Gene expression for each specimen was assessed using both the recommended input and twice the recommended input.

   Summary of Findings:

   Extraction with the AllPrep Kit yielded an average of 5.41 µg (range: 0.38-23.85 μg) of RNA from the adult FFPE specimens and 392.02 ng (range: 39.0-2039.0 ng) from the pediatric FFPE specimens, with A260/A280 ratios of 2.01 and 2.11, respectively, and A260/A230 ratios of 0.97 and 0.15, respectively. Use of the RNAstorm Kit for extraction from FFPE pediatric specimens led to slightly higher yields (999.0 ng, range of 33.0-2534.0 ng) and A260/A280 and A260/A230 ratios of 1.8 and 0.89, respectively.  The RNAstorm Kit yielded significantly more RNA and higher A260/A230 values from non-matched pediatric biopsies than the AllPrep Kit (P=0.0266 and P<0.0001, respectively). Because the quantity of RNA obtained from pediatric biopsies was insufficient for analysis by NanoString, NanoString analysis was only used to assess two different RNA input amounts in adult FFPE specimens that underwent extraction with the AllPrep Kit. The binding density in the NanoString assay was only proportionate to the amount of RNA input for one of the five specimens (DV200=50-70%). The specimen with a DV200 <30% was flagged during quality control for having a normalization factor >10-fold different from the average regardless of input amount, after normalization of data to the geometric mean of the housekeeping genes. The authors stated that this likely indicates an insufficient number of gene counts, which was confirmed by < 100 genes detected even when the RNA input was doubled.  As expected, the number of genes detected above background and the expression correlation for housekeeping genes increased for all five specimens when twice as much RNA was used as input.  Clustering based on the first four principal components was observed for sample pairs that differed in in RNA input amounts, but clustering was not based on DV200 value or RNA input amount. While housekeeping gene expression was very strongly correlated between the two different RNA input amounts for specimens with a DV200 >30% (R²>0.9), the correlation was strong for specimens with a DV200 <30% (R²>0.739). While levels of highly expressed genes (top 50%) were very strongly correlated between samples from the same specimen that had different RNA input amounts among the four specimens with a DV200 value >30% (R²>0.9), levels of lowly expressed genes (bottom 50%) were only strongly correlated (R²>0.8 for all but one specimen with a DV200 >70%). Upon further analysis, the authors determined that the specimen with lower correlation values and a DV200 >70% had an “abnormal smear” in the gel image which may reflect contamination and have interfered with the DV200 calculation.

   Biospecimens

   • Tissue - Colorectal

   Preservative Types

   • Formalin

   Diagnoses:

   • Irritable Bowel Syndrome
   • Normal
   • Neoplastic - Carcinoma

   Platform:

   Analyte	Technology Platform
   RNA	Automated electrophoresis/Bioanalyzer
   RNA	DNA microarray
   RNA	Spectrophotometry

   Pre-analytical Factors:

   Classification	Pre-analytical Factor	Value(s)
   Analyte Extraction and Purification	Analyte isolation method	AllPrep DNA/RNA FFPE kit RNAstorm FFPE RNA extraction kit
   DNA microarray Specific	Template/input amount	Recommended input Twice the recommended input
   DNA microarray Specific	Quality metrics	DV200 <30% DV200=30-50% DV200=50-70% DV200>70%

   View more