The BOIN clinical trial design21 Jan 2026 15:04
The BOIN (Bayesian Optimal Interval) design is a model-assisted, adaptive statistical method for early-phase oncology trials, used primarily to find the Maximum Tolerated Dose (MTD) by comparing the observed dose-limiting toxicity (DLT) rate with pre-set escalation and de-escalation boundaries, offering a simple, efficient, and statistically sound alternative to older methods like the 3+3 design. It's known for its transparency, operational simplicity, and strong performance in guiding dose escalation/de-escalation to safely identify the optimal dose level for further study.
Primarily used in Phase I oncology trials to find the MTD or the Recommended Phase II Dose (RP2D).
Key Advantages
• Simplicity & Efficiency: Easier for clinicians to use than complex model-based designs, while being more adaptive and faster than traditional methods.
• Improved Safety: Better overdose control, reducing patient exposure to overly toxic doses.
• Flexibility: Can handle various scenarios, including drug combinations, late-onset toxicity, and different endpoints.
• Strong Performance: High probability of correctly identifying the true MTD.
There is a series of complex (as in involving strange mathematical symbols) decision criteria. More details here:
An overview of the BOIN design and its current extensions for novel early-phase oncology trials ( https://www.sciencedirect.com/science/article/pii/S2451865422000606 )