MATS

MATS implements a multi-arm two-stage sequential adaptive design to evaluate two selected doses from a dose-escalation trial for proof-of-concept (PoC) and dose optimization across oncology indications.

Key Features:

• Two-dose evaluation: Evaluates two selected doses (higher and lower) originating from a dose-escalation trial.
• Sequential, adaptive two-stage design: Employs a first stage of broad assessment and a second stage that proceeds conditional on first-stage results.
• Stage 1 — higher-dose screening: Screens the higher of the two doses for anti-tumor activity across multiple oncology indications.
• Stage 2 — randomized comparison: Conducts a randomized comparison between the higher and lower doses within an indication if the higher dose is promising.
• Bayesian hierarchical model: Governs decision-making and enables borrowing of information across doses, indications, and stages.
• Simulation validation: Performance characteristics and operating characteristics have been evaluated through simulation studies.

Scientific Applications:

• Proof-of-concept (PoC): Establishes anti-tumor activity signals for the higher dose across indications to identify indications for further study.
• Dose optimization: Uses randomized comparisons to identify the dose with superior therapeutic benefit while considering safety, informing dose-finding beyond maximum tolerated dose (MTD) or recommended phase 2 dose (RP2D).
• Early-phase oncology trials: Supports multi-indication early-phase development by integrating evidence across indications and doses to improve decision-making.

Methodology:

Decision rules are implemented via a Bayesian hierarchical model with information borrowing across doses, indications, and stages within a sequential multi-arm two-stage adaptive design (stage 1 higher-dose screening; stage 2 randomized comparison), and the design has been evaluated using simulation studies.