Clinical Implementation of a High-Throughput Automated Comprehensive Genomic Profiling Test

Abstract

The adoption of comprehensive genomic profiling in oncology has rapidly increased the demand for standardized tumor sample processing in diagnostic laboratories. Automation of DNA and RNA library preparation workflows offers the possibility to scale-up and standardize sample processing. We report on the clinical implementation of the automated TruSight Oncology 500 High-Throughput library preparation workflow from formalin-fixed, paraffin-embedded tumor samples using the Biomek i7 hybrid Workstation. Using the same input amount, the automated workflow was validated against manual library preparation. Quality control metrics (total and mapped reads, median insert size, and median exon coverage) and the detection of tumor mutational burden, a complex biomarker, were concordant between the manual and automated workflows. The automated workflow was implemented on a total of 2997 pan-cancer clinical samples to detect genomic variants and complex biomarkers. Workflow automation resulted in a 4-fold reduction in hands-on time and a 1.7-fold reduction in total runtime compared with manual library preparation (6 hours vs. 23 hours; 24 hours vs. 42.5 hours, respectively) for a 48 DNA + 48 RNA sample batch. The automated workflow required one technician versus three technicians to manually prepare the same number of libraries. This study shows that implementation of the automated TruSight Oncology 500 High-Throughput workflow significantly reduced hands-on time and processing time per sample compared with manual library preparation.