Dynamic changes in cytokine levels as potential prognostic biomarkers in cancer immunotherapy

Background Immune checkpoint inhibitors (ICIs) have reshaped cancer therapy, yet fewer than 20% of patients achieve durable benefit and many develop immune-related adverse events, highlighting the need for accessible biomarkers to support treatment decisions. Circulating cytokines obtained through liquid biopsy offer a minimally invasive means to monitor systemic immune activity and represent exploratory, hypothesis-generating indicators of immune modulation during ICI therapy. Methods We retrospectively analyzed clinical data and paired plasma samples from patients with advanced or metastatic solid tumors treated within early-phase ICI-based trials at IRCCS Humanitas Research Hospital (2019–2023). Blood was collected at baseline and at first radiologic assessment. Thirteen cytokines (IL-1β, IL-2, IL-4, IL-6, IL-10, IL-12p70, IL-17A, TNF-α, IFN-γ, IL-8, CXCL10, CCL2, TGF-β1) were quantified using a multiplex bead-based assay, and longitudinal changes were classified as positive, negative, or null. Exploratory associations with disease control rate (DCR), overall survival (OS), and treatment duration were evaluated. Immunohistochemistry for CD3, CD68, and CD66b was performed to explore descriptive correlations between cytokine modulation and tumor immune contexture. Results Sixty-six patients were included. Positive IL-1β variation was associated with higher DCR (p=0.022). OS showed exploratory associations with positive changes in IL-1β, IL-12p70, and IL-4. Concordant cytokine modulation across these cytokines was frequent (all p<0.001). Increased IFN-γ correlated with longer treatment duration. Patients with at least one positive cytokine change displayed higher CD3+ T-cell density and lower CD66b+ neutrophil infiltration. Conclusions These hypothesis-generating findings suggest that paired longitudinal cytokine profiling may complement tissue-based biomarkers and inform future personalized immunotherapy strategies.