Decoding Immune Dysregulation in Newly Diagnosed Cancer through integrated Single-Cell RNA-Seq, Spectral Immune Phenotyping and Machine Learning

Early cancer detection remains a major clinical challenge. Circulating immune biomarkers provide a promising, non-invasive diagnostic opportunity, yet their potential remains insufficiently defined. Here, we present an integrated multi-omics analysis of peripheral blood mononuclear cells (PBMCs) from treatment-naïve cancer patients, combining immune phenotyping (flow cytometry, FC), multiplex cytokine profiling, and single-cell RNA sequencing (sc-RNA-seq). Compared with healthy controls, patients exhibited widespread immune dysregulation, including expansion of FOXP3+ regulatory T cells, depletion of CD16+CD11b+ monocytes and CD56dim NK cells, and elevated plasma IL-6/IL-4 levels. Sc-RNA-seq identified novel cancer-specific immune signatures, notably consistent upregulation of THBS1 and CH25H, indicative of systemic imprinting by tumor-derived cues. Deep learning models integrating single cell multi-omics data (sc-FC + sc-RNA-Seq) achieved performance comparable to clinical models, enabling cancer-type stratification and mechanistic insight. These findings establish a framework for immune-based, multi-omics diagnostics in early cancer detection and disease monitoring.