Abstract

Abstract Background Copy number (CN) and somatic mutation studies of cancer are powerful tools for discovering reliable biomarkers that can predict clinical outcomes. We have shown that the Molecular Inversion Probe (MIP) assay used in the OncoScan™ platform works well with thousands of degraded FFPE samples.1 In collaboration with leading cancer researchers, the coverage of the existing platform was further enhanced by supplementing the probe panel with new probes to cover an additional 200 cancer-relevant genes. The OncoScan™ FFPE Express 2.0 platform is able to deliver three types of high-quality data (SNP, CN, and somatic mutation calls) from just 75 ng non-amplified genomic DNA. Experimental design New MIP probes were designed to target the 200 cancer relevant genes and provide: (a) greater marker density; (b) high MAF (minor allele frequency) for SNP analysis; (c) good quantitative performance for reproducibility and dynamic range. Our platform employs a measurement comparing adjacent markers across the genome: this median of absolute pairwise distribution, or MAPD, is a reliable metric for assessing data quality.2 Results The median resolution increased from 1500 to 500 bp/probe in tumor suppressor genes and from 3500 to 1500 bp/probe in oncogenes. The performance of additional probes were assessed in the following categories: (1) For genotyping - the average call rate for 48 HapMap samples with 335K passing SNPs is 99.82%; for normal FFPE, 99.78%. The average accuracy for 48 HapMap is 99.38%. (2) For copy number - probe reproducibility was measured at CN = 0, 1 and 2. The average reproducibility of copy number changes is >95% as measured at the probe level in the SNPs that reside in the 200 genes. Furthermore, >97% of the newly designed probes passed this criteria. (3) For somatic mutation - the new assay was tested in cell lines with known mutations, including EGFR_pE746_A750del, KRAS_pG12D/V, PIK3CA_pE545K and PIK3CA_pH1047. Concordance of 100% was observed from six repeats of each mutation; testing was performed in two different laboratories. The false positive rate is 0.63% when assessing performance with normal samples. Conclusion We have developed a powerful new version of the OncoScan™ platform that works well on both frozen and archival FFPE samples. All three types of mutation analysis (SNP, CN, and somatic mutations) can be interrogated by the same assay, offering an unprecedented opportunity for both retrospective studies where only FFPE samples are available and for ongoing clinical research where only small amounts of biopsy sample are accessible

Full Text
Published version (Free)

Talk to us

Join us for a 30 min session where you can share your feedback and ask us any queries you have

Schedule a call