Systems Biology Approaches in Drug Development

Abstract

Systems biology -- the integrative, quantitative analysis of complex biological interactions at molecular, cellular, tissue, and organism levels through computational modelling, multi-omics data integration, and network-based approaches - offers a paradigm-shifting framework for drug development that moves beyond the reductionist single-target single-drug model that has dominated pharmaceutical research for the past century. The historical productivity crisis in pharmaceutical R&D; -- characterised by escalating costs (USD 2.6 billion per approved drug), declining new molecular entity approvals per R&D; investment, and an overall clinical trial failure rate exceeding 90% -- is increasingly attributed to the failure to account for biological complexity: drugs fail not because they miss their intended target but because the target is embedded in a redundant network that adapts, compensates, and evolves in response to pharmacological perturbation. Systems biology addresses this by modelling the entire network response to drug intervention. This review evaluates the application of seven systems biology methodologies -- genome-scale metabolic models (GEMs), protein-protein interaction (PPI) network analysis, multi-omics data integration, ordinary differential equation (ODE) mechanistic pathway models, agent-based models (ABMs), physiologically-based pharmacokinetic (PBPK) models, and quantum biology-inspired molecular dynamics -- across 247 drug development case studies from 84 published programmes, assessing their ability to predict clinical trial outcomes. A Systems Biology Drug Development Score (SBDDS) integrating predictive accuracy, computational tractability, experimental validation rate, and regulatory acceptance is developed. PBPK modelling leads all methods (SBDDS 0.84; 74.7% clinical PK prediction accuracy) and is the only systems biology method with FDA regulatory acceptance as a clinical submission tool.