PKBOIN-12: A Bayesian Optimal Interval Phase I/II Design Incorporating Pharmacokinetics Outcomes to Find the Optimal...

Area of Science:

• Pharmacology and Clinical Trial Design
• Oncology Drug Development
• Biostatistics in Medicine

Background:

• Early-phase clinical trials are shifting focus from maximum tolerated dose (MTD) to optimal biological dose (OBD) for improved risk-benefit balance.
• Pharmacokinetic (PK) data, reflecting drug exposure, correlate with toxicity and efficacy but are underutilized in current OBD selection.
• PK data availability within days offers a timely advantage over slower toxicity and efficacy outcomes.

Purpose of the Study:

• To introduce PKBOIN-12, a model-assisted design integrating PK information into dose-finding and OBD selection.
• To extend the design to TITE-PKBOIN-12, addressing challenges with late-onset toxicity and efficacy.
• To enhance the efficiency and accuracy of identifying the optimal biological dose in early-phase trials.

Main Methods:

• Development of the PKBOIN-12 design, enhancing the existing BOIN12 framework by incorporating PK parameters.
• Extension to TITE-PKBOIN-12 to accommodate time-to-event data for toxicity and efficacy.
• Extensive simulation studies and sensitivity analyses to evaluate design performance.

Main Results:

• PKBOIN-12 demonstrates superior performance over BOIN12 in identifying the OBD and allocating more patients to it.
• The new design reduces the selection of inefficacious doses by excluding those with inadequate drug exposure.
• Both PKBOIN-12 and TITE-PKBOIN-12 show robustness across various simulated scenarios.

Conclusions:

• PKBOIN-12 offers a more effective and efficient method for optimal biological dose selection in early-phase clinical trials.
• Integrating PK data into trial design optimizes the risk-benefit assessment by leveraging timely drug exposure information.
• The proposed designs provide a robust framework for advancing precision medicine through improved dose-finding strategies.