Kinetic Fingerprints as Mechanistic and Clinical Roadmaps Across KIT Activation States

In cancer therapy, traditional approaches often overlook the dynamic nature of drug-target interactions. We introduce kinetic fingerprints as a mechanistically informative tool to guide kinase inhibitor design and predict clinical performance. Profiling 172 compounds across multiple KIT conformations, including the oncogenic D816V mutation, show that prolonged residence time determines therapeutic success, while mutations accelerating dissociation rates (koff) drive resistance, positioning koff as a robust predictor of clinical failure. Beyond efficacy and resistance, kinetic signatures map molecular behavior: fast-associating scaffolds engage readily populated KIT states, slow binders overcome conformational barriers like juxtamembrane repositioning, and extended residence times highlight ligands stabilizing regulatory elements (G-loop and regulatory spine). Kinetic profiling further unveils mechanisms invisible to conventional methods, such as drug-induced kinase degradation, and exposes selectivity dimensions beyond affinity: avapritinib exhibits durable KIT D816V engagement yet transient off-target binding. Our findings redefine the evaluation of KIT inhibitors, establishing a framework for rational, kinetics-guided drug discovery in KIT-driven cancers. Table of Contents O_FIG O_LINKSMALLFIG WIDTH=200 HEIGHT=166 SRC="FIGDIR/small/726488v1_ufig1.gif" ALT="Figure 1"> View larger version (36K): org.highwire.dtl.DTLVardef@17f522borg.highwire.dtl.DTLVardef@1ebabc6org.highwire.dtl.DTLVardef@15a5e61org.highwire.dtl.DTLVardef@1ea5e0e_HPS_FORMAT_FIGEXP M_FIG Kinetic profiling of 172 compounds across KIT conformationsincluding D816Vreveals kinetic fingerprints that predict efficacy, selectivity, resistance, and inhibitors ability to stabilize key regulatory elements. Far from a secondary metric, binding kinetics provide mechanistic insights beyond affinity, offering a powerful framework for rational drug design in KIT-driven cancers. C_FIG