Discordance Found Between Liquid and Tissue Biopsies in NGS in Patients With Non–Small Cell Lung Cancer at Disease Progression

Guideline recommendations for lung cancer include panel-based next-generation sequencing (NGS) as initial diagnostic testing. Tissue-based NGS is able to detect targetable mutations to help guide therapy, but it has a number of drawbacks. It is expensive, sample acquisition can be associated with adverse events, sample errors may occur due to tumor heterogeneity, and turnaround time can be prolonged. Blood-based liquid biopsy can overcome some of these limitations and has become a cost-effective tool in patients with non–small cell lung cancer at initial evaluation, for monitoring therapy response, and in progression prediction. Despite these benefits, liquid biopsy uptake has been limited for a variety of reasons, including uncertainty regarding concordance between it and tissue-based NGS when evaluating disease progression or in the setting of intertumoral heterogeneity after targeted therapy.

To evaluate if there is concordance between liquid biopsy and tissue-based NGS in the real-world setting at the time of diagnosis and at progression, Tran and colleagues used a large, single-center database of patients with histologically confirmed lung adenocarcinoma. Patients had both liquid biopsy and tissue-based NGS within 24 weeks of one another without having a change in therapy. In-house UCM-Onco Plus (University of Chicago) was used for tissue testing. This testing reported 155 genes for clinical use and 1057 genes for research purposes, and it had a >99% sensitivity. Later in the study period, a fusion panel to detect NTRK1, NTRK2, and NTRK3 fusions was also performed. Guardant360 CDx was used for liquid biopsy, which reported variants in 55 genes. Variants were defined as clinically actionable if they guided systemic therapy decisions or assisted in progression evaluation. Individual tests were considered clinically informative if they had ≥1 clinically actionable result. If both NGS liquid and tissue biopsy individual test results led to the same clinical action, this was termed clinically informative concordance.

The study comprised 126 patients who had 138 paired tests: 77 patients were treatment naïve and 61 patients experienced progression. The study population was predominately White, female, and never or former smokers, and patients had a median Eastern Cooperative Oncology Group Performance Score of 1. The most common therapies before progression were osimertinib, erlotinib, afatinib, and chemotherapy. In the treatment-naïve group, the concordance rate was 83.1%, and in the progression group the concordance rate was 41.0% among all paired tests. In paired tests that had ≥1 clinically informative finding, the concordance rate for the treatment-naïve group was 79.4%, and for the progression group the concordance rate was 40.0%. Evaluation of a population of patients who were osimertinib refractory found common clinically informative genetic events missed by tissue biopsy, including EGFR T790M and CCDC6-RET, and those missed by liquid biopsy included GOPC-ROS1 fusion, RB1 loss, and K1F5B-RET fusion.

Source: Tran MC, Strohbehn GW, Karrison TG, et al. Brief report: discordance between liquid and tissue biopsy-based next-generation sequencing in lung adenocarcinoma at disease progression. Clin Lung Cancer. 2023;24(3):e117-e121.