Immunological Drug-Drug Interactions in Immune Checkpoint Inhibitor Therapy: Mechanisms, Clinical Evidence, and...

Immune checkpoint inhibitors (ICIs) have transformed cancer therapy, producing durable responses across multiple malignancies. However, treatment outcomes may be influenced by immunological drug-drug interactions (DDIs) arising from commonly prescribed concomitant medications. Unlike classical pharmacokinetic or pharmacodynamic DDIs, these interactions operate through systemic mechanisms that modulate anti-tumour immunity, including alterations to the gut microbiome, immune signalling pathways, and the tumour microenvironment. This review proposes a conceptual framework for "immunological DDIs" (iDDIs), extending beyond metabolic interactions toward a system-level understanding of immune regulation.

We synthesise current evidence on commonly used medication classes-organised by their primary immunological mechanisms: (1) gut microbiome-mediated effects, (2) systemic immunosuppression, and (3) tumour microenvironment modulation-and their impact on ICI efficacy and safety. Meta-analyses suggest that certain medications, particularly antibiotics and proton pump inhibitors, are associated with poorer clinical outcomes, although confounding by indication and disease severity remain important limitations. Artificial intelligence (AI) is an emerging approach to detect and characterise complex DDIs using large-scale clinical and real-world data. Natural language processing, machine learning models, and large language models show potential for extracting medication exposure, predicting adverse events, and supporting clinical decision-making. Most AI applications remain at an early stage, with limited external validation and uncertain clinical utility. Future research should integrate mechanistic biology, prospective clinical studies, and explainable AI approaches to improve identification of iDDIs and optimise the safe and effective use of ICIs in oncology.