Choice of starting dose for molecularly targeted agents evaluated in first-in-human phase I cancer clinical trials

Area of Science:

• Oncology
• Clinical Pharmacology
• Translational Medicine

Background:

• Conventional methods for cytotoxic agents rely on the lethal dose to 10% of mice (LD10) to set safe starting doses in Phase I trials.
• No established consensus exists for selecting preclinical models and parameters to define starting doses for molecularly targeted agents (MTAs).

Purpose of the Study:

• To review current practices in determining starting doses for first-in-human Phase I trials of molecularly targeted agents.
• To assess the safety and variability of preclinical models and toxicologic parameters used for dose selection.

Main Methods:

• A systematic review of 81 Phase I trials involving 60 different molecularly targeted agents administered as monotherapy.
• Analysis of preclinical data sources (rodent vs. non-rodent models) and toxicologic parameters used to establish the initial dose.
• Definition of maximum-tolerated dose (MTD) and maximum-administered dose (MAD).

Main Results:

• Fifty-seven trials specified animal models, with rodent and non-rodent data used almost equally (51% vs. 49%).
• A broad spectrum of toxicologic parameters was employed for starting dose selection.
• The starting dose exceeded the human MTD in 3.7% of trials, with nonhematologic toxicity being dose-limiting in these cases.
• The median ratio of MTD/MAD to starting dose was 12, suggesting potential for higher starting doses.
• Doubling the starting dose appeared safe, but tripling was not.

Conclusions:

• The process of deriving starting doses for Phase I trials of molecularly targeted agents is generally safe.
• Current practices are diverse, utilizing a variety of preclinical toxicologic parameters and models.
• Standardization of preclinical models and parameters could potentially optimize dose selection for molecularly targeted agents in early-phase clinical trials.