Comparative performance of amplicon-based DNA and RNA NGS co-detection versus hybrid-capture DNA-NGS for SNV/INDEL profiling in Chinese NSCLC driver genes

Background: Although amplicon-based DNA+RNA NGS co-detection has demonstrated clear advantages over hybrid-capture DNA-NGS for identifying gene fusions in non-small cell lung cancer (NSCLC), the clinical relevance of single-nucleotide variants (SNVs) and small insertions/deletions (INDELs) in NSCLC remains equally critical for precision oncology. However, whether amplicon-based DNA+RNA NGS co-detection provides comparable or improved performance for detecting NSCLC driver-gene SNVs/INDELs relative to hybrid-capture DNA-NGS has not been systematically evaluated. Methods: We performed a comparative analysis of SNV and INDEL detection across 19 clinically relevant NSCLC driver genes, including AKT1, ALK, ARAF, BRAF, EGFR, ERBB2, FGFR1/2, HRAS, IDH1/2, KIT, KRAS, MAP2K1, MET, MTOR, NRAS, NTRK, PDGFRA, PIK3CA, and RET. Detection rates generated by a 35-gene amplicon-based DNA+RNA NGS co-detection assay (n = 2,242) were compared with those obtained using hybrid-capture DNA-NGS (n = 2,504). Results: Amplicon-based DNA+RNA NGS co-detection demonstrated significantly higher detection rates for EGFR (50.68% vs 45.73%, P < 0.05), KRAS (12.25% vs 10.78%, P < 0.05), and KIT (0.40% vs 0.08%, P < 0.05) compared with hybrid-capture DNA-NGS. The increase in EGFR detection was primarily driven by improved identification of EGFR L858R (23.39% vs 20.01%, P < 0.05). For KRAS, differences were mainly contributed by KRAS G12C (4.12% vs 3.75%) and G12D (2.43% vs 1.80%), though these did not reach statistical significance. In contrast, hybrid-capture DNA-NGS showed a significantly higher detection rate for PIK3CA (6.11% vs 4.51%, P < 0.05). For all remaining genes, including AKT1(0.32% vs 0.28%), ALK(0.11% vs 0.12%), ARAF(0.12% vs 0.12%), BRAF(2.89% vs 2.6%), ERBB2(2.98% vs 3%), FGFR2(0.08% vs 0.08%), FGFR3(0.21% vs 0.16%), HRAS(0.08% vs 0.16%), IDH1(0.28% vs 0.28%), IDH2(0.12% vs 0.12%), MAP2K1(0.31% vs 0.48%), MET(0.52% vs 0.52%), MTOR(0.26% vs 0.32%), NRAS(0.41% vs 0.52%), NTRK(0.01% vs 0.04%), PDGFRA(0.12% vs 0.16%), RET(0.1% vs 0.12%), the mutation detection rates showed no significant difference between the two platforms. Conclusion: This large-cohort comparison demonstrates that amplicon-based DNA+RNA NGS co-detection performs comparably to hybrid-capture DNA-NGS for most NSCLC driver-gene SNV/INDEL alterations, while offering significantly higher sensitivity for critical mutations in EGFR, KRAS, and KIT. These findings indicate that co-detection workflows integrating both DNA and RNA may improve mutation detection efficiency for key actionable NSCLC biomarkers, supporting their broader adoption in routine clinical genomic profiling.