Evaluating the repair of DNA derived from formalin-fixed paraffin-embedded tissues prior to genomic profiling by SNP-CGH analysis.
Author(s): Hosein AN, Song S, McCart Reed AE, Jayanthan J, Reid LE, Kutasovic JR, Cummings MC, Waddell N, Lakhani SR, Chenevix-Trench G, Simpson PT
Publication: Lab Invest, 2013, Vol. 93, Page 701-10
PubMed ID: 23568031 PubMed Review Paper? No
Purpose of Paper
This paper investigated the effects of DNA extraction method and restoration on single nucleotide polymorphism (SNP) call rates and copy number determination using SNP arrays and investigated if quality control PCR values could predict assay performance.
Conclusion of Paper
Quality control cycle threshold values were a predictor of array performance but they did not accurately predict array success for every specimen. Although specimens extracted with the DNeasy Blood and Tissue Kit and the High Pure DNA Template Preparation kit performed comparably in quality control PCR reactions, the DNeasy Blood and Tissue Kit resulted in better or comparable array performance. In the limited number of specimens investigated, sodium thiocyanate treatment before extraction did not affect the results; however, restoration of the extracted DNA improved SNP call rates and prediction of copy number changes and decreased noise, in a reproducible manner. Importantly, restored DNA retained a high level of concordance with the original data and allowed for detection of copy number alterations missed in unrestored data. FFPE specimens showed greater magnitude of copy number changes reflecting the tumor enrichment from macrodissection.
Studies
-
Study Purpose
This study examined the effect of restoring DNA isolated by two different methods on single nucleotide polymorphism (SNP) call rates and copy number determination using SNP arrays. DNA was extracted using both the DNeasy Blood and Tissue Kit and the High Pure DNA Template Preparation Kit from four matched tumor and two lymph specimens from a single patient stored 25 years. Additionally, DNA was extracted from three matched FFPE specimens from a patient with breast cancer stored less than two years using the High Pure DNA Template Preparation Kit and from a frozen specimen from the same patient using the DNeasy Blood and Tissue Kit. For the FFPE specimens, 1 mm cores of tumor rich areas were de-waxed prior to extraction and the digestion was extended to be three days. All of the specimens stored for 25 years and two of the FFPE specimens stored less than two years were treated with 1M sodium thiocyanate overnight at 37 ˚C before extraction. DNA from both lymph specimens and four of the seven tumor specimens was restored using the Infinium HD FFPE DNA Restore protocol.
Summary of Findings:
Restoration of DNA extracted with the High Pure DNA Template Preparation kit resulted in higher SNP call rates (from 0.699 to 0.964 and from 0.796 to 0.839) and lower noise (from 0.26270 to 0.01090 and from 0.14394 to 0.02481), and while restoration of DNA isolated with the DNeasy Blood and Tissue kit had only marginal effects of SNP call rates (from 0.831 to 0.841) the decrease in noise post-restoration was still apparent (from 0.04993 to 0.0134). Importantly, concordance in SNP genotype between matched restored and unrestored specimens was 84.5–86.11% and 97.97% when DNA was extracted using the High Pure DNA Template Preparation kit and 99.71-99.75% when DNA was extracted with the DNeasy Blood and Tissue kit. The copy number prediction plots obtained with restored FFPE specimens more closely resembled those obtained with frozen specimens than when the DNA was not restored but as expected, the FFPE specimens showed greater magnitude of copy number changes reflecting the tumor enrichment from macrodissection (70% tumor cells in the FFPE specimen versus 50% tumor cells in the frozen specimen). The restoration process was reproducible as SNP detection rates between replicate restored specimens were within 0.038 of each other and copy number plots were also highly reproducible. Further, an amplification of human epidermal growth factor receptor 2 (HER2 or ERBB2) observed in the restored but not the unrestored DNA from an FFPE specimen was verified by immunohistochemistry (IHC) and chromogenic in situ hybridization (CISH) and amplification of epidermal growth factor receptor (EGFR) observed in restored and unrestored DNA from a specimen was confirmed by IHC.
Biospecimens
Preservative Types
- Frozen
- Formalin
Diagnoses:
- Neoplastic - Normal Adjacent
- Neoplastic - Carcinoma
Platform:
Analyte Technology Platform DNA Array CGH Protein Immunohistochemistry DNA In situ hybridization DNA DNA microarray Pre-analytical Factors:
Classification Pre-analytical Factor Value(s) Analyte Extraction and Purification Analyte isolation method DNeasy Blood and Tissue Kit
High Pure DNA Template Preparation Kit
Biospecimen Preservation Type of fixation/preservation Formalin (buffered)
Frozen
Array CGH Specific Technology platform CISH
IHC
Immunohistochemistry Specific Targeted peptide/protein EGFR
HER2
Array CGH Specific Template modification Restored
Unrestored
-
Study Purpose
This study compared SNP call rates and copy number determination among specimens extracted with the DNeasy Blood and Tissue Kit and the High Pure DNA Template Preparation Kit and investigated if quality control PCR was an indicator of array success. Specimens included two matched FFPE breast tumor and blood specimens stored < 2 years, five matched tumor and normal lymph specimens stored 22-25 years, and autopsy specimens of normal kidney and lymph, lung and bone metastasis, and breast tumor from 18 patients with metastatic breast cancer stored 11-53 years. One millimeter cores of tumor-rich areas were dewaxed and DNA was extracted using the DNeasy Blood and Tissue Kit or the High Pure DNA Template Preparation Kit modified by extension of the digestion period to three days. Some specimens were treated with 1M sodium thiocyanate overnight at 37 ˚C before extraction.
Summary of Findings:
Although specimens extracted with the DNeasy Blood and Tissue Kit and the High Pure DNA Template Preparation Kit performed comparably in quality control PCR reactions, the DNeasy Blood and Tissue Kit had higher SNP detection rates and lower variance for 3 of 10 specimens and comparable SNP detections rates and lower variance for the remaining 7 specimens. Pretreatment of specimens with sodium thiocyanate before extraction using the High Pure DNA Template Preparation did not affect the results, but no comparisons between treated and untreated specimens were made with older specimens or specimens extracted using the DNeasy Blood and Tissue kit.
Quality control cycle threshold values for specimens with less than 5 cycles difference from the internal control were correlated with a higher rate of detection of SNPs (R2 = 0.2722; P = 0.0324) and reduced noise (R2 = 0.06942; P = 0.0728); however, there were some specimens for which this value did not accurately predict array success or failure.
Biospecimens
Preservative Types
- Frozen
- Formalin
Diagnoses:
- Neoplastic - Carcinoma
- Autopsy
Platform:
Analyte Technology Platform DNA Array CGH DNA DNA microarray DNA Real-time qPCR Pre-analytical Factors:
Classification Pre-analytical Factor Value(s) Array CGH Specific Template modification Sodium thiocyanate treated
Untreated
Analyte Extraction and Purification Analyte isolation method DNeasy Blood and Tissue Kit
High Pure DNA Template Preparation Kit