The efficacy of targeted next-generation sequencing for detection of clinically actionable mutations in cancer.

Abstract

10598 Background: The emergence of next-generation sequencing (NGS) technologies has significantly accelerated the identification of cancer-causing mutations and the development of personalized cancer care. However, the clinical application of these technologies to detect cancer gene mutations has been extremely limited due to the long turn around time, the high cost, and large amount of input DNA required by existing NGS-based tests. Methods: We have assessed the performance of a novel NGS technology that merges multiplex PCR with ion semiconductor sequencing (AmpliSeq, Life Technologies, Inc.) in our clinical diagnostic laboratory. The test interrogates 739 common mutations in 46 cancer genes including many clinically actionable mutations concurrently. First, we studied 12 tumor samples including 4 archived FFPE, 4 blood/bone marrow, and 4 cell line samples with known mutations to evaluate the sensitivity and specificity of the test. We then studied 34 de-identified, archived FFPE tumor samples of unknown genotype to further evaluate the efficacy of the test. Results: We successfully identified all known mutations previously detected by Pyposequencing or Sanger sequencing technologies. Multiple serial dilution studies showed that the test could detect mutations at frequencies as low as 5% with 99% confidence. For the samples of unknown genotype, we detected 23 COSMIC mutations in 16 samples including HRAS, BRAF, MET, TP53 mutations in lung cancer, KRAS, PIK3CA, TP53, APC, BRAF, ERBB2 mutations in colon cancer, TP53 and KRAS mutations in breast cancer, and KIT and PDGFRA mutations in GIST. Analysis of the variant call data showed that a minimum of 100X coverage is required in order to detect mutations at 10% frequency or above; a minimum 300K final library reads are necessary in order to minimize/eliminate amplicon dropout. Conclusions: The targeted NGS test can effectively detect cancer gene mutation with input DNA as low as a few nanograms, turn around time can be as short as two days, and can significantly lower cost compared to traditional Sanger sequencing. Our experience demonstrates that this technology holds great potential for clinical use, including diagnostic and therapeutic applications.