Targeted RNA sequencing for upfront analysis of actionable driver alterations in non-small cell lung cancer

Objectives: Targeted RNA-based Next-Generation Sequencing (tRNA-seq) is increasingly being used in molecular diagnostics for gene fusion detection in non-small cell lung cancer (NSCLC). However, few data support its clinical application for the detection of single nucleotide variants (SNVs) and small insertions/deletions. In this study, we evaluated the performance of tRNA-seq using Archer FusionPlex for simultaneous detection of actionable gene fusions, splice variants, SNVs and indels in formalin-fixed, paraffin-embedded NSCLC tissue.

Materials and methods: A total of 126 NSCLC samples, including 20 validation samples and 106 diagnostic cases, were analyzed by targeted DNA-based Next-Generation Sequencing (tDNA-seq) followed by tRNA-seq.

Results: All 28 SNVs and indels in the validation set, and 34 out of 35 mutations in the diagnostic set were identified by tRNA-seq. The only mutation undetected by tRNA-seq, ERBB2 p.(Ser310Tyr), was not included in the current Archer panel design. tRNA-seq revealed one additional BRAF p.(Val600Glu) mutation not found by tDNA-seq. SNVs and indels were correctly called by the vendor supplied software, except for ERBB2 duplication p.(Tyr772_A775dup) which was only detected by an additional in-house developed bio-informatics pipeline. Variant allelic frequency (VAF) values were generally higher at the expression level compared to the genomic level (range 6-96% for tRNA-seq versus 6-61% for tDNA-seq) and low VAF mutations in DNA (6-8% VAF) were all confirmed by tRNA-seq. Finally, tRNA-seq additionally identified a driver fusion or splice variant in 10 diagnostic NSCLC samples including one MET exon 14 skipping variant not detected by tDNA-seq.

Conclusion: Our results demonstrate that tRNA-seq can be implemented in a diagnostic setting as an efficient strategy for simultaneous detection of actionable gene fusions, splice variants, SNVs and indels in NSCLC provided that adequate RNA-seq analysis tools are available, especially for the detection of indels. This approach allows upfront identification of currently recommended targetable molecular alterations in NSCLC samples.