Cancer Drug Discovery in the Digital Age: A Computational Perspective

The overarching goal of multi‐scale computational integration seeks to enhance early clinical outcome prediction in order to increase the efficiency and productivity of the therapeutic development pipeline. Rather than incremental improvements offered by tweaking existing isolated methods like QSAR or molecular docking, fundamentally reinventing the modeling paradigm to appropriately represent biological mechanisms across interconnected scales promises a transformational advance. Challenges certainly exist in managing model complexity while retaining biological traceability. However, the expected dividends from even partially attaining mechanistically articulated dynamical mappings between salient molecular events through physiological responses and ultimately clinical data warrant sustained collaborative pursuit by the quantitative pharmacology modeling community in coming years. Ultimately, the grand aim centers on inverting the dominant workflow overly reliant on brute empirical force. Instead, it remains eminently feasible to judiciously supplement but largely replace serial experimentation driving late stage optimization with rationally designed in silico trials simulating and troubleshooting various what‐if intervention scenarios.