A validated algorithm for sensitive and cost-effective mutation detection in clinical cancer specimens.

e12009 Background: Patients with EGFR-mutant lung cancer experience tumor shrinkage and prolonged survival when treated with EGFR inhibitors. Furthermore, drugs targeting additional mutational activated proto-oncogenes are currently being evaluated clinically. Thus, validated approaches for sensitive and accurate mutation detection in a clinical setting are urgently needed but have, so far, been lacking. Methods: Here, we performed a careful and systematic validation of two sequencing approaches (namely, Sanger sequencing and pyrosequencing) against the gold standard, massively parallel sequencing. Results: Prospective mutational annotation of clinical lung tumor samples as a function of tumor content revealed previously recognized limitations of Sanger sequencing but also indicated that pyrosequencing is a reliable and sensitive approach to mutation detection in the setting of limited tumor content. Importantly, pyrosequencing is highly cost effective and dramatically extends the fraction of tumors that ...