Cancer Diagnosis Program | Biorepositories and Biospecimen Research Branch

Search BRD Papers

RNA Extraction Method Impacts Quality Metrics and Sequencing Results in Formalin-Fixed, Paraffin-Embedded Tissue Samples.

Author(s): Decruyenaere P, Verniers K, Poma-Soto F, Van Dorpe J, Offner F, Vandesompele J

Publication: Lab Invest, 2023, Vol. 103, Page 100027

PubMed ID: 37039153 PubMed Review Paper? No

Purpose of Paper

Conclusion of Paper

Studies

1. Study Purpose

   This study compared RNA yields, purity and fragment size as well as total RNA and exome NGS metrics among RNA extracted from sections of FFPE DLBCL tumors using three different extraction kits.  FFPE blocks from 16 patients with DLBCL were retrieved from the archive. Details of specimen processing, including procurement and fixation, were not provided.  Additionally, FFPE blocks were created from microsatellite-stable (MSS) and microsatellite-instable (MSI) cell lines.  RNA was extracted from one to two 10 µm thick sections using the miRNeasy FFPE Kit, Ionic FFPE to Pure RNA Kit and the iCatcher FFPE Tissue RNA Kit. After RNA extraction, DNA was removed by DNase treatment. RNA was quantified by Qubit and spectrophotometer and the ratio of absorbance at 260 to 280 nm (A260/280) and 260 to 230 nm (A260/A230) were used to assess RNA purity.  RNA integrity was evaluated using an Agilent fragment analyzer.  Sequencing libraries were prepared using the TruSeq RNA Exome library preparation Kit and the SMARTer Stranded Total RNA-Seq Pico v3 library preparation Kit; samples were paired-end sequenced on a NovaSeq 6000 machine.

   Summary of Findings:

   Although RNA yields were higher when measured by spectrophotometer than Qubit before DNase treatment, the difference between quantification methods was eliminated by DNase treatment. RNA concentrations were very strongly correlated between the two quantification methods (R=0.91-0.99), regardless of DNase treatment. Tissue volume and RNA yields were strongly correlated when measured by both spectrophotometer and Qubit when  extraction was with the miRNeasy FFPE Kit (r=0.84; P<0.001 and r=0.85; P<0.001, respectively) or the iCatcher FFPE Tissue RNA Kit (r=0.81; P<0.001 and r=0.81; P<0.001, respectively) When extraction was with the Ionic FFPE Kit, there was only a modest correlation between tissue volume and RNA yield when quantified by spectrophotometer (r=0.66; P=0 .0054) and no correlation when RNA yield was quantified by Qubit. The DV200 values did not decrease following DNase treatment and were not correlated with tissue input volume or RNA yield. The DV200 value was higher when extraction was with the iCatcher FFPE Tissue RNA Kit than the miRNeasy FFPE Kit (66.12% versus 41.66%, P=0.0015) or the Ionic FFPE Pure RNA Kit (66.12% versus 39.53.79%, P=0.0041).  The ratios of absorbance at 260 nm to 280 nm and 230 nm were lower when extraction was with the Ionic FFPE to Pure RNA Kit than the other two kits, indicating a higher level of contamination when the Ionic FFPE to Pure RNA Kit was used.

   The total RNA sequencing percentage of uniquely mapped reads and the number of detectable genes with >1 or >10 counts were significantly lower and the percentage of duplicate reads was higher when extraction was with miRNeasy FFPE Kit compared to the Ionic FFPE to Pure RNA Kit or the iCatcher FFPE Tissue RNA Kit (P<0.05 all); but, no differences in these metrics were found between extraction methods when evaluated by exome sequencing. As expected, fragment size of mapped reads (for both library preparation methods), library yields (for both library preparation methods), and uniquely mapped reads in total RNA sequencing were correlated with DV200 values (r=0.68-0.90, P<0.001; r=0.51-0.85, P<0.05; and r=0.58-0.90, P<0.01, respectively), but the DV200 value and fraction of uniquely mapped reads in exome sequencing were only correlated when extraction was with miRNeasy FFPE Kit (r=0.70, P<0.001) or the Ionic FFPE to Pure RNA Kit (r=0.68, P<0.001).

   In principle component analysis, specimens made from microsatellite-stable (MSS) and microsatellite-instable (MSI) cell lines clustered separately, although there was some sub-clustering based on RNA extraction method. There was considerable overlap in the genes that were identified to be differentially expressed in the MSS versus MSI cell lines among the three extraction methods (42.4-55.6%).  Specimens from immunohistochemistry-positive patients had higher normalized expression counts of CD10, BCL6, MUM1, and BCL2 than specimens from immunohistochemistry-negative patients, regardless of extraction method. However, B-cell receptor analysis found significantly more and longer IgH CDR3 total RNA sequences and longer kappa sequences after extraction with the Ionic FFPE to Pure RNA Kit or iCatcher FFPE Tissue RNA Kit than the miRNeasy FFPE Kit (P<0.05, for all), but no differences in IgH exome sequence counts or length occurred among the extraction methods evaluated.

   Biospecimens

   • Tissue - Lymph Node

   Preservative Types

   • Formalin

   Diagnoses:

   • Neoplastic - Lymphoma

   Platform:

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

   Pre-analytical Factors:

   Classification	Pre-analytical Factor	Value(s)
   Analyte Extraction and Purification	Analyte isolation method	miRNeasy FFPE kit Ionic FFPE to Pure RNA kit iCatcher FFPE Tissue RNA kit
   Spectrophotometry Specific	Technology platform	NanoDrop Qubit
   Next generation sequencing Specific	Technology platform	TruSeq RNA Exome library preparation kit SMARTer Stranded Total RNA-Seq Pico v3 library preparation
   Analyte Extraction and Purification	Nucleic acid digestion	DNase treated Before DNase treatment

   View more