Non-Small Cell Lung Cancer Testing on Reference Specimens: An Italian Multicenter Experience.

Abstract

Biomarker testing is mandatory for the clinical management of patients with advanced non-small cell lung cancer (NSCLC). Myriads of technical platforms are now available for biomarker analysis with differences in terms of multiplexing capability, analytical sensitivity, and turnaround time (TAT). We evaluated the technical performance of the diagnostic workflows of 24 representative Italian institutions performing molecular tests on a series of artificial reference specimens built to mimic routine diagnostic samples. Sample sets of eight slides from cell blocks of artificial reference specimens harboring exon19 EGFR (epidermal growth factor receptor) p.E746_AT50del, exon2 KRAS (Kirsten rat sarcoma viral oncogene homologue) p.G12C, ROS1 (c-ros oncogene1)-unknown gene fusion, and MET (MET proto-oncogene, receptor tyrosine kinase) Δ exon14 skipping were distributed to each participating institution. Two independent cell block specimens were validated by the University of Naples FedericoII before shipment. Methodological and molecular data from reference specimens were annotated. Overall, a median DNA concentration of 3.3ng/µL (range 0.1-10.0ng/µL) and 13.4ng/µL (range 2.0-45.8ng/µL) were obtained with automated and manual technical procedures, respectively. RNA concentrations of 5.7ng/µL (range 0.2-11.9ng/µL) and 9.3ng/µL (range 0.5-18.0ng/µL) were also detected. KRAS exon2 p.G12C, EGFR exon19 p.E736_A750del hotspot mutations, and ROS1 aberrant transcripts were identified in all tested cases, whereas 15 out of 16 (93.7%) centers detected MET exon14 skipping mutation. Optimized technical workflows are crucial in the decision-making strategy of patients with NSCLC. Artificial reference specimens enable optimization of diagnostic workflows for predictive molecular analysis in routine clinical practice.