Immune-Biomarkers in Advanced Non-Small Cell Lung Cancer

Background: Developing personalized immunotherapy-based approaches, through the identification of predictive immune-related biomarkers, is still a main topic for translational lung cancer research. In the present study we investigated whether baseline tissue “immune-related gene signatures”, as well as plasma chemo-cytokines profiles, could predict sensitivity/resistance to immune-checkpoint inhibitors in pre-treated patients with advanced non-small cell lung cancer (NSCLC). Methods: From September 2015 to September 2018, 150 patients with previously treated advanced NSCLC who received either anti-PD1 or anti-PD-L1 inhibitors in the second-third line setting were included within this translational study. All the patients underwent CT-scan every 6 cycles and responses were evaluated by Response Evaluation Criteria in Solid Tumors (RECIST). Gene expression analysis was performed on 86 FFPE tissue samples by using the nCounter® PanCancer IO360™ Panel applied on NanoString platform, in order to analyze 770 genes involved in key immuno-oncology pathways. Peripheral blood samples were obtained from 57 patients at baseline and 37 patients at the 6th week under IO-therapy. A panel of cytokines and chemokines (IL-6, IL-8, CXCL10, CX3CL1, CCL2, VEGF, and IFN- gamma) were quantified in plasma by Cytokine Bead Array and their association with OS and TTP was assessed by Adaptive Index Modeling multivariable analysis. NLR and PLR were also assessed for potential association with clinical outcomes. Results: The gene expression levels of IL12RB2, ESR1 and CCL8 (p-values = 0.000308, 0.00162 and 0.0398, respectively) resulted significantly higher in responders versus non-responders patients. In patients with 0S > 18 months, a significant upregulation of the ESR1 (p-value = 0.00813) and IL12RB2 (p-value = 2.48e-08) genes, as well as a higher score for lymphoid compartment (p-value = 0.0117), cytokine and chemokine signaling (p-value = 0.0268), costimulatory signaling (p-value = 0.0323), and cytotoxicity (p-value = 0.0412), was observed. As regards peripheral blood samples analysis, an Immune-suppressive blood index score (ISBIS) was identified clustering patients into 3 groups with progressively worsening TTP and OS. The score was composed by higher IL-8 and CCL-2 levels, higher NLR, and lower IFN-gamma level. The differences among both PFS and OS Kaplan Meyers curves across the different subgroups were statistically significant (p < 0.0001). Conclusion: These data suggest that the systemic balance between neutrophil-related inflammation and lymphocyte anti-tumor immunity may condition response to immunotherapy in lung cancer, with clinical benefit primarily occurring in patients with such a preexisting, intra-tumoral T-cell adaptive immune response. Correlative tissue immune gene signatures as well as circulating cyto-chemokine emerged as potential indicators of a T cell– inflamed phenotype necessary for the clinical activity of PD-1/PD-L1 inhibitors.

N.A