Separate quality-control measures are necessary for estimation of RNA and methylated DNA from formalin-fixed, paraffin-embedded specimens by quantitative PCR.
Korlimarla A, Prabhu JS, Anupama CE, Remacle J, Wahi K, Sridhar TS
Purpose
The purpose of this paper was to determine the effects of formalin fixation duration, block storage, and normalization method on the analysis of RNA in formalin-fixed paraffin-embedded (FFPE) breast tumor specimens and to determine if specimens with well-preserved RNA were suitable for DNA methylation analysis.
Conclusion
The mean cycle-threshold (CT) values of the three normalization genes were not affected by fixation duration, and the authors report that block storage did not affect the classification of RNA as well-preserved based on mean CT values of the reference genes beta-actin (ACTB), ribosomal protein large P0 (RPLP0) and Pumilio RNA-Binding Family Member 1 (PUM1). By excluding specimens with an average reference gene CT of >1 SD above the mean (poorly-preserved specimens), the concordance between estrogen receptor (ESR1) transcript levels and estrogen receptor (ER) protein levels increased, and the sensitivity of ESR1 transcript levels for determining ER positivity increased but the specificity did not increase. While not all specimens with well-preserved RNA were suitable for methylation analysis, specimens with well-preserved RNA had lower average ALU-C4 CT values than those with poorly preserved RNA (19.9 versus 21.6), and concordance of DNA and RNA status occurred in 108 of 149 (72%) specimens.
Studies (2)
Purpose
The purpose of this study was to determine the effects of formalin fixation duration, block storage, and normalization method on the analysis of RNA in 250 FFPE breast tumor specimens and the concordance with immunohistochemistry (IHC) results. Breast tumor specimens were fixed for 16-96 h and stored as FFPE blocks for <45 months at an unspecified temperature. >50% tumor content was verified for all specimens by hematoxylin and eosin (H&E) staining. RNA was extracted from 2 20 um sections with TriReagent after deparaffinization by heating in TE with SDS and proteinase K digestion. Poorly preserved RNA was defined as specimens with an average CT value of the three normalization genes that was >1 standard deviation above the mean
Summary of Findings
The mean CT values of the three normalization genes were 29.9 in the 147 specimens fixed for <48 h and 30.3 in the 103 specimens fixed for >48 h, but the difference was not significant. The authors report similar insignificant differences in CT values when a cut-off of 24 h fixation time was used, but data was not presented. The authors report that having well-preserved RNA was not related to block storage duration.. The M-values (variation in comparison to the other control genes) were lower for ACTB and RPLP0 than PUM1 (m=1.019, m=1.019 and m=1.51, respectively), but all three genes were used to determine if the RNA was well-preserved. When the abundance of ESR1 transcript was plotted against ER protein Allred score, the concordance was k=0.54. The sensitivity of ESR1 transcript levels for determining ER positivity was 77%, and the specificity was 80%. Using only well preserved RNA, the concordance between ESR1 transcripts levels and protein levels increased to k=0.71, and the sensitivity of ESR1 transcript levels for determining ER positivity increased to 92%, but the specificity did not increase.
Biospecimens
TissueBreastPreservation Types
FormalinDiagnoses
NeoplasticCarcinomaPlatforms
RNAReal-time qRT-PCRProteinImmunohistochemistryPre-analytical Factors
Classification Pre-Analytical Factor Value(s) Biospecimen Preservation Time in fixative 16-48 h48-96 hStorage Storage duration 3-45 monthsReal-time qRT-PCR Specific Targeted nucleic acid ACTBESR1PUM1RPLP0Real-time qRT-PCR Specific Technology platform IHCImmunohistochemistry Specific Targeted peptide/protein ER
Purpose
The purpose of this study was to determine if there was an association between specimens having well-preserved RNA and suitability for DNA methylation analysis. 149 breast tumor specimens were fixed for 16-96 h and stored as FFPE blocks for <45 months. >50% tumor content was verified for all specimens by H&E staining. DNA was extracted from 2 20 um sections with TriReagent after deparaffinization by heating in TE with SDS and proteinase K digestion. DNA underwent bisulfate modification as necessary for methylation studies. Poorly preserved RNA was defined as specimens with an average CT value of the three normalization genes > 1 standard deviation above the mean.
Summary of Findings
The average ALU-C4 CT of specimens using real-time PCR was lower in specimens with well preserved RNA than those with poorly preserved RNA (19.9 versus 21.6). Further, 98 of 149 specimens had well-preserved RNA and DNA suitable for methylation analysis while 10 had poorly preserved RNA and DNA unsuitable for PCR. The remaining 41 specimens had either poorly preserved RNA or poorly preserved DNA indicating exclusion must be determined individually not by the specimen.
Biospecimens
TissueBreastPreservation Types
FormalinDiagnoses
NeoplasticCarcinomaPlatforms
DNAReal-time qPCRRNAReal-time qRT-PCRPre-analytical Factors
Classification Pre-Analytical Factor Value(s) Real-time qRT-PCR Specific Targeted nucleic acid ACTBPUM1RPLP0Real-time qPCR Specific Targeted nucleic acid ALU-C4 sequence