Comparative analysis of genomic profiles between tissue-based and plasma-based next-generation sequencing in patients with non-small cell lung cancer.

• High concordance was found between tissue-NGS and plasma-NGS in cases with ctDNA. • Plasma samples without detectable ctDNA caused most of the inconsistent results. • Plasma-NGS showed high sensitivity in detecting SNV and indel mutations. • Plasma-NGS was significantly less capable of detecting CNVs and gene fusions. • Tissue-NGS remains the preferred method for molecular testing in NSCLC patients. Genotype-guided personalized therapy has become an essential part of routine clinical care in non-small cell lung cancer (NSCLC) patients. However, small tissue specimens often yield inadequate molecular testing material. Plasma ctDNA-based liquid biopsy is an increasingly common non-invasive alternative to tissue biopsy. This study examined the similarities and differences in the molecular profiling of tissue and plasma samples to provide insight into sample selection in clinical practice. Sequencing data from 190 NSCLC patients who underwent concurrent tissue-based next-generation sequencing (tissue-NGS) and plasma-based NGS (plasma-NGS) using a 168-gene panel were analyzed. Tissue-NGS identified genomic alterations in 97.4% (185/190) of the enrolled patients and plasma-NGS identified genomic alterations in 72.1% (137/190) of the enrolled patients. Considering all NSCLC guideline-recommended biomarkers in the entire cohort of 190 cases, 81 patients had positive concordant mutations detected in both tissue and plasma samples, while 69 patients had no predefined alterations detected in either tissue or plasma samples. Additional mutations were found in the tissues of 34 patients and the plasma of six patients. The overall concordance rate between tissue and plasma samples was 78.9% (150/190). The tissue-NGS and plasma-NGS sensitivities were 95.0% and 71.9%, respectively. In the 137 patients with detectable ctDNA in plasma samples, the concordance rate between tissue and plasma samples reached 91.2%, and the sensitivity of plasma-NGS reached 93.5%. Our findings indicate that plasma-NGS is less capable of detecting genetic alterations than tissue-NGS, especially for copy number variations and gene fusions. Tissue-NGS remains the preferred method for evaluating the molecular profile of NSCLC patients when tumor tissue is available. We suggest that the concurrent use of liquid biopsy and tissue biopsy is the optimal approach in clinical practice; alternatively, plasma can be used as substitute material when tissue is unavailable. [ABSTRACT FROM AUTHOR]