Overview
Liver metastasis is a serious complication for patients with advanced non-small cell lung cancer (NSCLC), often leading to worse outcomes. Predicting which patients are at higher risk for liver metastasis has been challenging, and the biological factors driving liver metastasis are not well understood. In this study, researchers used the US-based deidentified Flatiron Health-Caris Life Sciences NSCLC Clinical-Molecular Database (CMDB), with clinical data from the Flatiron Health Research Database linked to whole exome sequencing, whole transcriptome sequencing, and immunohistochemistry data derived from Caris tumor profiling tests. The team analyzed data from over 1,100 patients with advanced NSCLC adenocarcinoma who received first-line chemoimmunotherapy, focusing on gene expression patterns in lung tumor biopsies collected at diagnosis.
By applying advanced statistical methods, the researchers identified a set of 259 genes for which expression was strongly associated with the risk of developing liver metastasis. This gene signature allowed them to score all patients based on risk and demonstrate clear differences in the incidence of developing liver metastasis over time in the high- vs low-risk groups. This signature was specific to liver metastasis and not significantly associated with brain metastases.
Why this matters
Using the CMDB, this research provides a new tool—a gene expression signature—that can help identify which patients with advanced NSCLC are most at risk for developing liver metastasis. By improving risk prediction, clinicians may be able to tailor surveillance and treatment strategies more effectively, potentially improving patient outcomes. Additionally, understanding the biological processes linked to liver metastasis could open new avenues for research and therapy development. Further validation in other patient groups will be important to confirm these findings and support their use in clinical practice.