Enhanced stability of microRNA expression facilitates classification of FFPE tumour samples exhibiting near total mRNA degradation.
Author(s): Hall JS, Taylor J, Valentine HR, Irlam JJ, Eustace A, Hoskin PJ, Miller CJ, West CM
Publication: Br J Cancer, 2012, Vol. 107, Page 684-94
PubMed ID: 22805332 PubMed Review Paper? No
Purpose of Paper
This paper investigated whether qRT-PCR array technical success for mRNA and miRNA profiling was affected by the age of the formalin-fixed, paraffin-embedded (FFPE) tissue blocks used. RNA quality metrics were also evaluated for their ability to predict successful profiling.
Conclusion of Paper
FFPE block age was not correlated with RNA yield or RIN. Interestingly, RNA yield, spectrophotometry-generated ratios (OD 260/230, OD 260/290) and RNA integrity numbers (RIN) were not correlated with exon array technical performance, as measured by percent detection above background (%DABG). However, %DABG was weakly correlated to cDNA yield post-amplification (R=0.19 for bladder specimens; R=0.39 for cervix specimens)(P<0.05 for all) and inversely correlated with FFPE block age, with a higher %DABG observed among FFPE blocks collected more recently for both cervix (R= -0.69, P<0.01) and bladder (R= -0.30, P<0.01) specimens. The majority of cervical specimens stored less than 15 years were correctly classified as adenocarcinoma or squamous cell carcinoma (SCC) based on gene expression signature, but blocks stored for more than 15 y displayed ratios that were at or below background resulting in non-classification. The authors conclude that biological signal has been lost in FFPE blocks stored for prolonged periods of time. The abundance of the p63 transcript quantified by the Exon array declined with FFPE block age for cervical specimens (R= -0.63), a trend that was mirrored when p63 protein expression was examined in the same 152 specimens by immunohistochemistry. FFPE block age similarly affected the abundance of pre-miRNAs (~110 nt) by Exon array, while qRT-PCR and Affymetrix miRN v2.0 analyses of mature miR-205 revealed it was unaffected by FFPE block age, exhibiting a predicted elevation in SCC specimens compared to AC specimens regardless of block storage duration.
Studies
-
Study Purpose
This paper investigated whether qRT-PCR array technical success for mRNA and miRNA profiling was affected by the age of the formalin-fixed, paraffin-embedded (FFPE) tissue blocks used. RNA quality metrics were also evaluated for their ability to predict successful profiling. Pretreatment FFPE biopsies of bladder transitional cell and non-muscle invasive carcinomas collected from 141 patients at 11 different hospitals using individual standard operating procedures (SOPs) between November 2000 and April 2006 were analyzed. Details of fixation including formalin type and block storage were not provided, but all bladder specimens contained ≥10% tumor content. Cervix carcinoma biopsy specimens were collected from 160 patients at a single hospital between 1987 and 2002 using a single set of SOPs. Cervical specimens were fixed in 4% neutral buffered formalin and stored at room temperature, but additional processing details were not provided, but all cervical specimens contained ≥30% tumor content.
RNA was extracted from all specimens with the RecoverAll Total Nucleic Acid Isolation Kit and then DNase treated. RNA yield and integrity were determined using a bioanalyzer, and 260/230 and 260/280 absorbance ratios determined by spectrophotometry.
For Exon array hybridization, 100 ng of total RNA was amplified using the NuGen WT-Ovation FFPE v2 kit, then strand transfer (ST) cDNA was generated and 3.8-4 µg was hybridized to the Human Exon 1.0 ST array, with the exception of two cervix carcinoma specimens with low RNA yields (50 ng was used). Microarray data was Robust Multichip Average (RMA) normalized, and technical performance was measured by the percent detection above background (%DABG). Sample variation among specimen cohorts was evaluated by principal component analysis. For miRNA array hybridization, 100 ng of total RNA was tailed and ligated prior to hybridization to Affymetriz miRNA v2.0 arrays.
Expression of miR-205 was verified by qRT-PCR analysis and normalization to housekeeping genes hsa-miR-26b and hsa-miR-16.
Summary of Findings:
FFPE block age was not correlated with RNA yield or RIN. Interestingly, RNA yield, spectrophotometry-generated ratios (OD 260/230, OD 260/290) and RNA integrity numbers (RIN) were not correlated with exon array technical performance, as measured by percent detection above background (%DABG). However, %DABG was weakly correlated to cDNA yield post-amplification (R=0.19 for bladder specimens; R=0.39 for cervix specimens)(P<0.05 for all) and inversely correlated with FFPE block age, with a higher %DABG observed among FFPE blocks collected more recently for both cervix (R= -0.69, P<0.01) and bladder (R= -0.30, P<0.01) specimens. A previously reported gene signature that can discriminate AC from SCC cervical specimens was evaluated in two subsets of cervical specimens that ranged in FFPE block age (subset 1: 114 specimens, stored for 8-18 y; subset 2: 48 specimens, stored for 16-23 y). The majority of cervical specimens stored less than 15 years were correctly classified as adenocarcinoma or squamous cell carcinoma (SCC) based on gene expression signature, but blocks stored for more than 15 y displayed ratios that were at or below background resulting in non-classification. The authors conclude that biological signal has been lost in FFPE blocks stored for prolonged periods of time. The abundance of the p63 transcript quantified by the Exon array declined with FFPE block age for cervical specimens (R= -0.63), a trend that was mirrored when p63 protein expression was examined in the same 152 specimens by immunohistochemistry. FFPE block age also affected the abundance of pre-miRNAs (~ 110 nt) by Exon array. However, qRT-PCR and Affymetrix miRN v2.0 analyses of mature miR-205 revealed it was unaffected by FFPE block age, exhibiting a predicted elevation in SCC specimens compared to AC specimens regardless of block storage duration.
Biospecimens
Preservative Types
- Formalin
Diagnoses:
- Neoplastic - Carcinoma
Platform:
Analyte Technology Platform RNA Real-time qRT-PCR RNA Low density array Protein Immunohistochemistry RNA DNA microarray Pre-analytical Factors:
Classification Pre-analytical Factor Value(s) Storage Storage duration 6-8 y
8-23 y
<15 y
>15 y
Immunohistochemistry Specific Targeted peptide/protein p63
Real-time qRT-PCR Specific Targeted nucleic acid miR-205