Large-scale tumor profiling studies have generated massive amounts of data that have been instrumental for the detection of recurrent driver mutations in many tumor types. These driver mutations as well as the concurrent passenger mutations are now being used for a more accurate diagnosis of the tumor and prognosis for the patient. Moreover, therapeutic inhibitors toward specific mutations are already on the market and many clinical trials are ongoing to approve novel therapeutic drugs. The broad-range identification of these somatic mutations is key to this tailored personalized medicine approach, which preferentially has to be performed by a multigene multihotspot method such as massive parallel sequencing, also called next generation sequencing (NGS). The implementation of NGS in molecular diagnostics of tumor profiling however, requires a firm validation to minimize the occurrence of false positives and false negatives, thereby yielding highly accurate and robust clinical data.Here, we describe the different performance characteristics as well as quality metrics that should be analyzed for the robust diagnostic validation of tumor profiling in order to meet the requirements of international standards specific for medical laboratories, such as the ISO15189:2012 standard. These metrics include assays that assess the precision, limit of detection, accuracy, sensitivity, specificity, and robustness of the entire workflow from DNA enrichment up to the final report.
Keywords: Diagnostic screening; NGS; Solid tumor; Target enrichment; Validation; Variant classification.