Phenotypic response surfaces in pharmacology: Mechanistic and translational perspectives on optimizing drug...

Combination drug therapies are central to the treatment of diseases with multifactorial etiology, including cancer, infectious diseases, and autoimmune disorders. Despite their widespread use, optimizing drug combinations remains challenging because of complex pharmacokinetic and pharmacodynamic interactions and substantial inter-patient variability. Phenotypic Response Surfaces (PRS) have emerged as a computational framework for mapping multidimensional drug-dose response landscapes while accounting for patient-specific biological and clinical factors. However, current PRS-oriented approaches vary considerably in how molecular, biochemical, and clinical determinants are incorporated into response models. In this review, we critically evaluate the mechanistic foundations of PRS-based methods for optimizing drug combinations. Particular emphasis is placed on how structural biology, biomarker-defined target dependence, and pharmacokinetic/pharmacogenetic models can be linked to response-surface frameworks to improve mechanistic interpretability and patient-specific dose optimization. We also examine recent advances in artificial intelligence (AI) platforms that support phenotypic optimization across multiple therapeutic areas. Although these approaches show promise, current PRS applications remain limited by challenges in integrating multi-omics data, parameterizing drug exposure and target engagement, validating model transportability across patient populations, and translating predictive models into clinical practice. Overall, PRS represents a promising framework for combination-therapy optimization, but its clinical utility will depend on more rigorous integration of molecular mechanism, exposure-response relationships, and biomarker-constrained modeling.