Despite unprecedented access to digital health data and AI modelling tools, most early-phase clinical trials in drug development stick to outdated, rigid dosing regimens.

The tension lies between the desire to gather reliable efficacy and safety data and the risk of missing 'the right dose for the right patient' due to insufficient adaptation during trials.

This not only leads to higher rates of failures in later trial phases but can also compromise patient safety and mask promising therapeutic effects, leaving clinical and economic value on the table.

Traditional population-based dosing, basic model-informed drug development platforms, and limited use of simple dose escalation schemes.

These do not fully utilize available data or adapt in real time, often leading to suboptimal results.