Network Pharmacology and Systems Medicine

Abstract

Network pharmacology models drug action as perturbation of disease-associated molecular networks rather than single-target inhibition, enabling rational design of multi-target therapies and drug combinations for complex diseases, yet adoption of network-based approaches into drug discovery and clinical practice varies across methodological categories. We evaluated 218 network pharmacology programmes across centres in France and Estonia between 2018 and 2022, spanning five method categories: network proximity-based drug repurposing, multi-target drug design optimisation, drug combination synergy prediction, disease module identification and targeting, and patient stratification through network-based biomarkers. A Network Pharmacology Systems Medicine Index (NPSMI) was constructed from five sub-scores -- target-disease network proximity accuracy, experimental validation rate, clinical translation evidence, multi-scale mechanistic coherence, and computational scalability to genome-wide networks -- with weights from regression against sustained adoption. NPSMI correlated with adoption at r = +0.84 and discriminated adopted from non-adopted methods with an AUC of 0.884. Network proximity repurposing scored highest (mean NPSMI 0.826), while patient stratification trailed at 0.602. Only 35.3 percent exceeded the 0.75 threshold. Network proximity accuracy carried the largest weight (beta = +0.278), followed by experimental validation (beta = +0.230).