Dose escalation represents a fundamental methodology in early clinical development, providing a structured approach to exploring the safety and tolerability of increasing doses of an investigational product in humans. Beginning with doses predicted to be safe based on preclinical studies, investigators carefully advance to higher doses only after demonstrating adequate safety at each level. This methodical progression allows identification of the maximum tolerated dose while minimizing participant exposure to potentially harmful dose levels.

The starting dose for human studies is typically derived from preclinical toxicology studies, often calculated as a fraction of the dose causing serious toxicity in the most sensitive animal species, adjusted for body surface area. From this starting point, doses are escalated according to a predetermined scheme, with common approaches including the traditional 3+3 design, modified Fibonacci sequences, and more sophisticated model-based methods such as the continual reassessment method. Each approach balances the competing goals of minimizing participant exposure to subtherapeutic or toxic doses while efficiently identifying the optimal dose range.

Safety monitoring during dose escalation requires particular vigilance. Dose-limiting toxicities are predefined adverse events of specified severity that trigger protocol-defined responses, which may include cohort expansion to gather additional safety data, dose reduction, or cessation of further escalation. Safety review committees evaluate accumulating data before each dose increase, typically requiring resolution of any acute toxicities in the current cohort before advancing. This careful, stepwise approach allows early identification of safety signals while establishing the foundation for dose selection in subsequent development.