Amplicon-based DNA and RNA unified NGS for enhanced fusion variant detection in suboptimal real-world NSCLC FFPE specimens

BACKGROUND: Timely identification of actionable fusion and exon-skipping events is essential for selecting targeted therapies in non-small cell lung cancer (NSCLC), yet routine RNA-based assays frequently fail in formalin-fixed paraffin-embedded (FFPE) specimens, leading to missed treatment opportunities. Amplicon-based DNA and RNA unified next-generation sequencing (D+R NGS) may enable guideline-recommended profiling from limited or degraded tissue, but its performance in real-world specimens is unclear. METHODS: The D+R NGS assay uses 20 ng of co-extracted RNA and DNA from FFPE tissue. Library preparation is performed in a single-tube amplicon-based workflow, followed by simultaneous sequencing and automated bioinformatics analysis. Its performance was evaluated in 759 NSCLC FFPE samples. Concordance was assessed against droplet digital polymerase chain reaction (ddPCR) in 46 matched fresh surgical specimens. Detection rates were evaluated in 172 surgical, 168 biopsy, and 142 cytology samples, all processed within one year. Clinical applicability was examined in 231 archived specimens stored for 1-8 years with known reverse transcription polymerase chain reaction (RT-PCR) or DNA-based next-generation sequencing (DNA-based NGS) fusion/skipping results. RESULTS: The D+R NGS assay demonstrated exceptional analytical validity, showing 100% concordance with ddPCR in 46 fresh NSCLC samples for 28 mutations and 18 fusions/skipping events, with strong correlations in variant allele frequencies (R = 0.997, p < 0.001) and fusion read counts (R = 0.799, p < 0.001). It achieved high success rates across diverse sample types: 99.4% in surgical, 98.6% in cytology, and 95.2% in biopsy specimens, with robust performance even in samples with tumor areas <3 mm². In challenging low-input RNA scenarios (as low as 5-10 ng), the assay reliably detected fusions, and it identified actionable alterations in 5 of 11 samples that had failed routine quality control (QC) due to insufficient RNA. When applied to 231 archival samples stored for 1-8 years, the assay confirmed 139 known fusions/skipping events (99.3%) and detected 18.2% additional fusions in DNA-NGS-negative cases, including receptor tyrosine kinase (ROS1) fusions and MET proto-oncogene, receptor tyrosine kinase (MET) exon 14 skipping, while maintaining a > 95% success rate despite a decline in read counts over time. CONCLUSION: The amplicon-based D+R NGS method demonstrates high compatibility with diverse NSCLC FFPE samples, overcoming RNA degradation challenges in small-biopsy and long-term archived specimens.