Development of NIST standard reference material 2373: Genomic DNA standards for HER2 measurements

摘要

Abstract NIST standard reference material (SRM) 2373 was developed to improve the measurements of the HER2 gene amplification in DNA samples. SRM 2373 consists of genomic DNA extracted from five breast cancer cell lines with different amounts of amplification of the HER2 gene. The five components are derived from the human cell lines SK-BR-3, MDA-MB-231, MDA-MB-361, MDA-MB-453, and BT-474. The certified values are the ratios of the HER2 gene copy numbers to the copy numbers of selected reference genes DCK , EIF5B , RPS27A , and PMM1 . The ratios were measured using quantitative polymerase chain reaction and digital PCR, methods that gave similar ratios. The five components of SRM 2373 have certified HER2 amplification ratios that range from 1.3 to 17.7. The stability and homogeneity of the reference materials were shown by repeated measurements over a period of several years. SRM 2373 is a well characterized genomic DNA reference material that can be used to improve the confidence of the measurements of HER2 gene copy number. Keywords Cancer ; Reference material ; Calibrator ; Nucleic acid ; HER2 ; Digital PCR prs.rt("abs_end"); 1. Introduction The human epidermal growth factor receptor 2 ( HER2 , official symbol ERBB2 ) is a proto-oncogene protein, a 185-kDa transmembrane glycoprotein with tyrosine kinase activity [1] . Amplifications of the HER2 gene, resulting in protein overexpression, are present in approximately 20% of breast cancers, and have been associated with poor patient prognosis [2] and [3] . The most common methods for HER2 measurements in clinical laboratories are the detection of protein overexpression by immunohistochemistry (IHC) and the evaluation of gene amplification by in situ hybridization (ISH) methods. A number of studies have reported problems with the accuracy and the concordance of the results obtained from different laboratories using IHC and fluorescence ISH (FISH) methods [4] and [5] . The American Society of Clinical Oncology and the College of American Pathologists published guidelines to improve the performance of HER2 testing by IHC and ISH methods in 2007, and an update in 2013 [6] and [7] . However, standardization of both IHC and ISH methods across laboratories remains a major challenge. Approximately 20% of HER2 testing performed may be inaccurate [8] . Recently, genomic analytical methods have been developed that enable DNA copy number variations (CNV) to be measured with high sensitivity and specificity. As few as 50 cells extracted from archival formalin-fixed paraffin-embedded (FFPE) tissues may be quantified using quantitative PCR (qPCR) [9] and [10] . The results from qPCR of HER2 measurements have been positively correlated with the results from IHC and FISH analysis [10] , [11] , [12] and [13] . Koudelakova et al. compared qPCR to IHC and FISH data in breast cancer samples, and found that high sensitivity and specificity of the new method was achieved and the results obtained with the qPCR method and FISH/IHC agreed [14] . Digital PCR (dPCR) was used to measure HER2 copy number in FFPE breast cancer tissue and these results agreed with the results from FISH and IHC analysis [15] . Garcia-Murillas used dPCR to measure the gene copy ratio of HER2 to reference genes in the microdissected DNA from HER2 amplified and HER 2 non-amplified cancers. They too, obtained high sensitivity and specificity and good agreement with the traditional detection methods [16] . Many cancer cells, including these breast cancer cell lines, have highly abnormal karyotypes, with multiple chromosome copies and major structural changes [17] . The selection of the reference genes is important in cancer cells because of the frequent gene mutations and gains or losses of DNA that have occurred. Chromosomal alterations in 15 breast cancer cell lines detected frequent gains at 1q, 8q, 20q, 7, 11q, 13, 17q, 9q and 16p and frequent losses at 8p, 11q14-qter, 18q and Xq [18] . Spectral karyotyping (SKY) using fluorescent staining for each chromosome showed a large number of complex alterations in the chromosome complement of breast cancer cell lines [18] ( http://old-www.path.cam.ac.uk/～pawefish/index.html ). A study of HER2 amplified tumors showed increased gains at 1q, 8q, 20q and losses at 18q, 13q, and 3p [19] . Comparative Genomic Hybridization (CGH) of 89 breast cancer tumors detected frequent gains at 1q, 8q, 11q, and 16p and losses at 4q, 5q, 6q, 8p, and 14q [20] . CGH was used to examine the chromosome complement of 51 breast cancer cell lines and 145 primary breast cancer tumors showed similar genetic changes in the cell lines and tumor samples with some differences: losses in 5q and losses in chromosome 18 [21] . These studies showed that the chromosomal locations of the reference genes need to be carefully considered and the assays have to be tested to ensure that the reference genes have not been specifically amplified or deleted. Suitable reference materials are needed for the new g