Digital Twins

Digital Twins, developed by Unlearn.ai, are AI-generated forecasts of individual clinical trial participants' control outcomes. By predicting clinical outcomes at every future time point with high precision, they serve as the core powering technology for more rigorous, sensitive, and efficient clinical analyses. This technology is designed for pharmaceutical companies, biotechs, and clinical researchers seeking to improve the statistical performance of their trials across all stages of drug development.

Unlearn's digital twin methodology is built on disease-specific machine learning models trained on extensive historical clinical data. Using only a participant's baseline data, these models generate a digital twin that forecasts that individual's clinical trajectory across every future time point — enabling a more precise and credible comparison between treatment and control groups.

Clinical Trial Applications

• Randomized Controlled Trials: Increase statistical power while maintaining existing sample sizes, or reduce control arm size while preserving power. This approach improves sensitivity across both primary and secondary endpoints, delivering a clearer signal of efficacy.
• Early-Stage Trials and Rare Diseases: Generate participant-level synthetic control arms to enable credible treatment comparisons in settings where randomization is infeasible. Strengthen high-stakes go/no-go decisions through improved statistical sensitivity.
• Interim and Retrospective Analyses: Improve sensitivity in interim looks and subgroup analyses to detect signals that traditional methods may miss. Re-evaluate historical trial data using regulatory-aligned methods to support learning across development programs.

Regulatory Acceptance

• Unlearn's PROCOVA methodology has been officially qualified by the European Medicines Agency (EMA) via the Committee for Medicinal Products for Human Use (CHMP) for use in Phase 2 and Phase 3 trials with continuous outcomes, enabling increases in power and/or decreases in sample size.
• The U.S. FDA has provided positive feedback on PROCOVA, supporting its use in covariate-adjusted analyses across clinical development.
• The FDA recommends that sponsors adjust for covariates most strongly associated with the outcome of interest, and supports the use of previous studies to select prognostic covariates or form prognostic indices.
• Regulatory guidance confirms that sample size and power calculations in covariate-adjusted trials can be based on either adjusted or unadjusted methods, providing sponsors with flexibility in trial design.

Demonstrated Impact Across Therapeutic Areas

• Reducing sample sizes while maintaining power, or boosting power without adding participants: Collaborative research with AbbVie, Johnson & Johnson, and Roche has demonstrated the methodology's effectiveness in Alzheimer's disease trials. Unlearn has also published findings in ALS, Parkinson's disease, and early Alzheimer's disease across major conferences including AAIC, AD/PD, MDS, and the International Symposium on ALS/MND.
• Using digital twins as an external comparator in early-stage studies: Partnerships with ProJenX and QurAlis have demonstrated the use of digital twins to augment early-stage ALS clinical trials, improving confidence in efficacy findings when traditional control arms are limited.
• Reducing variability to improve decision-making in early-stage trials: Research presented at AD/PD Boston 2025 with remynd demonstrated the use of digital twins in a Phase 2a Alzheimer's disease trial to predict disease progression and support early signals of efficacy.

Technology Foundation

• Disease-specific machine learning models are trained on extensive historical clinical datasets.
• Digital twins are generated for each individual trial participant using only their baseline data.
• The models forecast clinical outcomes at every future time point with unparalleled precision.
• The technology underpins Unlearn's broader suite of products, including TrialPioneer and Digital Twin Generators.

Unlearn.ai's Digital Twins technology is backed by collaborative research and successful implementation with global leaders in drug development, including top pharmaceutical and biotech organizations. The methodology is aligned with regulatory guidance from both U.S. and European authorities, making it a credible and compliant choice for sponsors seeking to modernize their clinical trial design and analysis strategies.