Format

Send to

Choose Destination
Sci Rep. 2017 Jun 22;7(1):4055. doi: 10.1038/s41598-017-04419-9.

Systematic comparison of two whole-genome amplification methods for targeted next-generation sequencing using frozen and FFPE normal and cancer tissues.

Author information

1
Thoracic Oncology Laboratory, Department of Surgery, University of California San Francisco, San Francisco, CA, USA.
2
Thoracic Oncology Laboratory, Department of Surgery, University of California San Francisco, San Francisco, CA, USA. David.Jablons@ucsf.edu.
3
Comprehensive Cancer Center, University of California San Francisco, San Francisco, CA, USA. David.Jablons@ucsf.edu.
4
Thoracic Oncology Laboratory, Department of Surgery, University of California San Francisco, San Francisco, CA, USA. Il-Jin.Kim@ucsf.edu.
5
Comprehensive Cancer Center, University of California San Francisco, San Francisco, CA, USA. Il-Jin.Kim@ucsf.edu.

Abstract

Sequencing key cancer-driver genes using formalin-fixed, paraffin-embedded (FFPE) cancer tissues is becoming the standard for identifying the best treatment regimen. However, about 25% of all samples are rejected for genetic analyses for reasons that include too little tissue to extract enough high quality DNA. One way to overcome this is to do whole-genome amplification (WGA) in clinical samples, but only limited studies have tested different WGA methods in FFPE cancer specimens using targeted next-generation sequencing (NGS). We therefore tested the two most commonly used WGA methods, multiple displacement amplification (MDA-Qiagen REPLI-g kit) and the hybrid or modified PCR-based method (Sigma/Rubicon Genomics Inc. GenomePlex kit) in FFPE normal and tumor tissue specimens. For the normalized copy number analysis, the FFPE process caused none or very minimal bias. Variations in copy number were minimal in samples amplified using the GenomePlex kit, but they were statistically significantly higher in samples amplified using the REPLI-g kit. The pattern was similar for variant allele frequencies across the samples, which was minimal for the GenomePlex kit but highly variable for the REPLI-g kit. These findings suggest that each WGA method should be tested thoroughly before using it for clinical cancer samples.

PMID:
28642587
PMCID:
PMC5481435
DOI:
10.1038/s41598-017-04419-9
[Indexed for MEDLINE]
Free PMC Article

Supplemental Content

Full text links

Icon for Nature Publishing Group Icon for PubMed Central
Loading ...
Support Center