CD34 as a potential prognostic indicator for camrelizumab response in advanced non-small-cell lung cancer: insights from digital spatial profiling

Background: Given that only a small subset of patients with advanced non-small-cell lung cancer (aNSCLC) benefit from immune checkpoint inhibitors (ICIs), the effectiveness of ICIs is often compromised by the complex interplay within the tumor microenvironment (TME). Objectives: To identify predictive biomarkers associated with ICI resistance at a multi-omics spatial level. Design: A total of eight aNSCLC patients who received first-line anti-programmed cell death protein-1 (PD-1) monoclonal antibody camrelizumab at Shandong Cancer Hospital and Institute between 2021 and 2022 were included in the discovery cohort. An additional validation cohort of 45 samples from camrelizumab-treated aNSCLC patients was also enrolled. Methods: NanoString GeoMx (R) digital spatial profiling was conducted at the transcriptomic and proteomic level within pan-cytokeratin (panCK+), CD45+, and CD68+ compartments. For validation, multiplex immunofluorescence (mIF) staining was performed. Results: Distinct spatial expression patterns and levels of immune infiltration were observed between tumor and leukocyte compartments. Higher CD34 expression in the macrophage compartment correlated with poorer prognosis and response to camrelizumab (p < 0.05). mIF validation confirmed the association of elevated CD34 expression level with reduced progression-free survival (PFS; hazard ratio (HR) = 5.011, 95% confidence interval: 1.057-23.752, p = 0.042), outperforming traditional tumor markers in predictive accuracy. Conclusion: Our findings identify CD34 as a novel spatial biomarker for anti-PD-1 therapy efficacy, potentially guiding the selection of aNSCLC patients who are more likely to benefit from ICI treatment.