Unlearn.ai

Unlearn.ai is a pioneering platform that leverages artificial intelligence and digital twins to revolutionize clinical trial design, planning, and analysis. By creating AI-generated forecasts of clinical trial participants' expected control outcomes, Unlearn aims to streamline clinical development, accelerate decision-making, and ultimately bring new treatments to patients faster. The platform offers a unified, upstream workspace that replaces fragmented workflows, allowing teams to achieve alignment more quickly and test assumptions earlier in the development process.

Overview of Unlearn's Solutions

Unlearn provides a comprehensive suite of solutions designed to enhance various stages of clinical development. Their core offerings, TrialPioneer and Trial Analyses with Digital Twins, are built upon a foundation of advanced AI and robust data integration. These tools are engineered to provide clarity and confidence throughout the entire clinical development lifecycle, from initial trial design to late-stage analysis.

TrialPioneer serves as a single workspace for upstream trial design. It integrates critical components necessary for iterative decision-making, assumption testing, and maintaining clear, defensible rationale across review cycles. Within TrialPioneer, users can access:

• Scout: A tool for continuously searching, structuring, and summarizing relevant literature and regulatory precedent from sources like PubMed, ClinicalTrials.gov, and drugs@FDA. This eliminates scattered searches and accelerates alignment on precedent.
• Hindsight: Enables exploration of harmonized clinical trial and real-world datasets to validate clinical and statistical assumptions. It helps assess feasibility, population characteristics, endpoint behavior, and benchmarks using transparent historical data.
• SimLab: Allows users to build and compare various trial-design scenarios to evaluate endpoints, inclusion/exclusion criteria, sample size, and constraints. Each scenario is explainable, reproducible, and linked to underlying evidence, supporting informed design trade-offs before protocol finalization.

Trial Analyses with Digital Twins strengthens trial analyses by utilizing AI-generated forecasts of clinical trial participants’ expected control outcomes. These digital twins act as external comparators in early-stage and open-label studies, significantly reducing variability and improving the ability to detect treatment effects. For randomized trials, this approach supports smaller sample sizes or increased statistical power. This methodology has been qualified by the EMA and aligns with current FDA guidance, facilitating clearer go/no-go decisions earlier in development and enabling more efficient late-stage trials, leading to measurable reductions in trial size, cost, and time to readout.

Key Highlights:

• Utilizes AI and digital twins to optimize clinical trial design and analysis.
• Offers a unified workspace, TrialPioneer, for upstream trial planning.
• Includes tools like Scout, Hindsight, and SimLab for literature review, data exploration, and scenario building.
• Digital twins reduce variability and improve treatment effect detection in early-stage trials.
• Supports smaller sample sizes and increased power in randomized trials.
• Methodology is qualified by EMA and aligned with FDA guidance.
• Aims to reduce trial size, cost, and time to readout.