Background
Previous studies found that lung adenocarcinomas (LUAD) with EGFR-positive and ALK-positive were less responsive to immunotherapy, which may be associated with a suppressive tumor immune microenvironment (TIME). Given the discordance in the TIME between primary lung cancer and brain metastasis, it is urgent to explore the TIME in patients with EGFR/ALK-positive LUAD with brain metastases (BMs).
Methods
The transcriptome feature of formalin-fixed and paraffin-embedded samples of BMs and paired primary LUAD from 70 patients with LUAD BMs was illustrated by RNA-sequencing. Six of them were available for paired sample analysis. Then, after excluding 3 co-occurring patients, we divided 67 BMs patients into 41 EGFR/ALK-positive and 26 EGFR/ALK-negative patients. The differences in immune profiling between the two groups were analyzed from three dimensions: TIME, T-cell receptor repertoire, and immunohistochemistry. Finally, the survival data of 55 patients were collected.
Results
Compared with primary LUAD, BMs present an immunosuppressed TIME, manifested as: inhibition of immune-related pathways; low expression of immune checkpoint; decreased infiltration of CD8+T cells and cytotoxic lymphocyte; increased proportion of suppressive M2 macrophages. In different subgroups based on EGFR/ALK gene variation status, both EGFR-positive and ALK-positive tumors present a relatively immunosuppressive microenvironment, but the heterogeneity of tumor microenvironment may undergo different mechanisms. EGFR-positive BMs showed decreased CD8+T cells and increased regulatory T cells (Treg) cells, while ALK-positive BMs showed decreased CD8+T cells and increased M2 macrophages. Moreover, in the TCGA-LUAD cohort, EGFR-positive tumors showed reduced CD8+T cell infiltrations (p<0.001) and borderline significantly higher Tregs than EGFR/ALK-negative (p=0.072). In parallel, ALK-positive tumors had higher median M2 macrophages infiltrations than EGFR/ALK-negative (p=0.175), although there was no statistical significance. Collectively, there was a similar immunosuppressive milieu between EGFR/ALK-positive primary LUAD and BMs. Moreover, survival analysis uncovered higher CD8A expression, cytotoxic lymphocyte infiltration, and immune scores were significantly associated with better prognosis in both EGFR/ALK-positive and EGFR/ALK-negative groups.
Conclusion
This study found that LUAD-derived BMs exhibited an immunosuppressive TIME and revealed that EGFR-positive and ALK-positive BMs exhibited different immunosuppressive characteristics. Meanwhile, EGFR-negative BMs showed a potential benefit to immunotherapy. These findings boost molecular and clinical understanding of LUAD BMs.