Dose finding for continuous and ordinal outcomes with a monotone objective function: a unified approach

Area of Science:

• Clinical Trials
• Biostatistics
• Pharmacology

Background:

• Phase I trials commonly aim to identify a specific target dose.
• Dose-finding objectives vary, including target toxicity rates, weighted toxicity grades, or continuous response means.
• Existing designs may not universally address these diverse objectives.

Purpose of the Study:

• To present a unified dose-finding design applicable to various phase I trial objectives.
• To establish a method for identifying target doses defined by a prespecified monotone function.
• To offer a flexible approach for dose escalation and de-escalation strategies.

Main Methods:

• The proposed design calculates the normalized difference between the current and target dose at each step.
• Dose adjustments (repeat, increase, or decrease) are determined by this normalized difference.
• The design accommodates trials targeting specific toxicity rates, toxicity grade sums, or continuous response means.

Main Results:

• The design provides a systematic approach to dose finding across different trial goals.
• It allows for dose repetition when the current dose is near the target.
• The method enables adaptive dose adjustments based on the calculated difference.

Conclusions:

• The described dose-finding design offers a versatile framework for phase I clinical trials.
• It effectively identifies target doses defined by monotone functions of dose.
• This approach enhances flexibility in dose selection for various clinical objectives.