Unlearn develops AI models that generate an individual digital twin for every clinical trial participant to forecast placebo outcomes before randomization, enabling highly powered "TwinRCTs" with smaller control groups. The company’s technology is qualified by the European Medicines Agency for use as the primary analysis in phase 2 and 3 trials with continuous outcomes and aligns with current FDA guidance. Unlearn partners with pharmaceutical sponsors, biotech companies, and academic institutions to accelerate enrollment and reduce trial timelines. Founded in 2017 and based in San Francisco, the company says its methods help more patients receive experimental treatment and bring therapies to market faster. Unlearn has raised over $130M to date and plans to invest the new funding in people, data, engineering capabilities, and longer-term R&D while expanding into additional therapeutic areas. Unlearn uses machine learning and generative AI to create a neural-network-based “digital twin” for each patient in a clinical trial, which can populate the control arm and reduce placebo enrollments. The company primarily targets Phase III trials and positions its platform to enable smaller, faster studies that lower costs for drug developers. Unlearn secured approval from the European Medicines Agency in September to employ its AI approach and has signed multi‑million dollar deals with pharmaceutical partners including Merck. The startup was founded in 2017 by Charles K. Fisher, Aaron Smith, and Jon Walsh and is based in San Francisco. To commercialize globally, Unlearn is prioritizing regulatory engagement—particularly with the FDA—and partnership projects with drugmakers to set precedents for the technology’s use. Financially, the company has attracted several high‑profile investors and recently raised additional funding to expand partnerships and accelerate regulatory approval. Unlearn builds TwinRCTs™, a machine-learning platform that creates digital twins of patients to reduce control-arm sizes and run smaller, faster randomized trials. TwinRCTs™ compute a digital twin for every patient using models trained on historical control data, then adjust for each patient’s prognostic score to estimate treatment effects with greater precision. Unlearn’s PROCOVA™ method has received a draft qualification opinion from the European Medicines Agency for use in primary analyses of phase II and III studies. The company partners with pharma sponsors, biotech firms, and academic institutions and publishes its work in scientific journals and conference abstracts. It has a multi-year collaboration with Merck KGaA to incorporate prognostic information from digital twins into randomized controlled trials. Unlearn says it is expanding its footprint in clinical trials and continuing discussions with global regulatory authorities. Unlearn builds a machine-learning platform that creates Digital Twins to populate Intelligent Control Arms for clinical studies. Its flagship product, Digital Twins for Alzheimer’s Disease, uses a proprietary AI model to generate simulated patient data that the company says is statistically indistinguishable from actual Alzheimer’s patient data. Study sponsors can incorporate cohorts of Digital Twins matched to trial patients either prospectively to enable smaller, more efficient trials or retrospectively to add power to ongoing trials. Unlearn Digital Twins for Alzheimer’s Disease are already being used as part of an ongoing pivotal trial. The company is led by founder and CEO Charles K. Fisher, Ph.D., and is based in San Francisco, CA. On the financing side, Unlearn announced a Series A extension that increases its Series A to $15M.

Total raised

$118M

Funding rounds

4

Latest round

Series C

Latest activity

Feb 2024