Lung adenocarcinoma has a strong propensity to metastasize to the brain. The brain metastases are difficult to treat and can cause significant morbidity and mortality. Identifying patients with increased risk of developing brain metastasis can assist medical decision-making, facilitating a closer surveillance or justifying a preventive treatment. We analyzed 27 lung adenocarcinoma patients who received a primary lung tumor resection and developed metastases within 5 years after the surgery. Among these patients, 16 developed brain metastases and 11 developed non-brain metastases only. We performed targeted DNA sequencing, RNA sequencing and immunohistochemistry to characterize the difference between the primary tumors. We also compared our findings to the published data of brain-tropic and non-brain-tropic lung adenocarcinoma cell lines. The results demonstrated that the targeted tumor DNA sequencing did not reveal a significant difference between the groups, but the RNA sequencing identified 390 differentially expressed genes. A gene expression signature including CDKN2A could identify 100% of brain-metastasizing tumors with a 91% specificity. However, when compared to the differentially expressed genes between brain-tropic and non-brain-tropic lung cancer cell lines, a different set of genes was shared between the patient data and the cell line data, which include many genes implicated in the cancer-glia/neuron interaction. Our findings indicate that it is possible to identify lung adenocarcinoma patients at the highest risk for brain metastasis by analyzing the primary tumor. Further investigation is required to elucidate the mechanism behind these associations and to identify potential treatment targets.
Keywords: CDKN2A; brain metastasis; lung adenocarcinoma; omics data analysis; p16.