EGFR L858R x EGFR T790M x EGFR-Ex19del x EGFR exon 20

Background: The feasibility and value of pericardial effusion as a liquid biopsy sample for actionable alteration detection in patients with non-small cell lung cancer (NSCLC) has not been adequately investigated. Here, we aim to reveal genomic alterations between pericardial effusion and paired tumor tissue, plasma (plasma cfDNA), and pleural effusion supernatant (PE-cfDNA) based on second-generation sequencing technology. Material and methods: A total of 26 advanced NSCLC patients were retrospectively studied. The following samples were collected and sequenced using two targeted next-generation sequencing panels: pericardial effusion (n = 26), matched tumor tissue (n = 6), plasma (n = 16), and pleural effusion supernatant (n = 5). Results: A total of 10 actionable alterations were identified in pericardial effusion of the NSCLC patients, including MET amplification, EGFR L858R, EGFR T790M, EGFR exon 19 deletion, EGFR L861Q, KRAS G12C, EML4-ALK (exon 18: exon 20) fusion, EML4-ALK (exon 20: exon 20) fusion, EML4-ALK (exon 6: exon 20) fusion, and ERBB2 exon 20 insertion. All these actionable alterations harbored multiple drug-sensitive targets as well as several drug-resistant targets, such as EGFR T790M. Compared to plasma cfDNA of 16 patients, paired pericardial effusion had higher number of actionable alterations (p = 0.08) as well as higher percentage of the population with actionable alterations (p = 0.16). Moreover, 8 out of 10 actionable alterations with single nucleotide variations (SNVs) or insertions/deletions (indels) had a higher variant allele frequency (VAF) in pericardial effusion than plasma cfDNA. In addition, we identified two actionable alterations in paired pericardial effusion, which were absence in PE-cfDNA. Clearly, 2 out of 3 actionable alterations with SNVs/indels in pericardial effusion had a higher VAF than those in PE-cfDNA. Our finding suggested the importance of pericardial effusion in the optimal selection of patients for targeted therapy. Conclusion: Among liquid biopsy specimens from the advanced NSCLC patients, pericardial effusion may be a better candidate for genomic profiling than plasma cfDNA, while it could serve as a supplement to PE-cfDNA in detecting actionable alterations. Therefore, pericardial effusion might provide a new alternative for selection of patients for better treatment management.