Unlearn develops AI models that generate an individual digital twin for every trial participant to forecast outcomes under placebo and enable TwinRCTs with smaller control groups. Its technology aims to reduce timelines, speed enrollment, and allow more patients to receive experimental treatment. The company says its methods are qualified by the European Medicines Agency for use as the primary analysis in phase 2 and 3 trials with continuous outcomes and align with current FDA guidance. Unlearn partners with pharmaceutical sponsors, biotech companies, and academic institutions and currently powers trials for leading global pharmaceutical companies. Founded in 2017 and based in San Francisco, Unlearn plans to invest in people, data, engineering capabilities, longer-term R&D, and expand into additional therapeutic areas. The company has raised over $130 million to date. Unlearn.AI is a San Francisco-based biotech AI startup that builds generative-AI "digital twin" profiles of patients for clinical trials. Its platform computes a twin for each trial participant and populates control arms to potentially reduce placebo enrollments and lower trial costs. The company says its neural networks enable smaller, quicker clinical trials and won European Medicines Agency approval for this approach in September. Unlearn has struck several multi-million dollar deals with pharmaceutical companies, including Merck. In March 2023 it raised $15 million at a $265 million valuation to expand partnerships and accelerate regulatory approval. OpenAI CTO Mira Murati joined Unlearn's board as part of the latest developments. Unlearn.AI builds disease-specific digital twins by training machine-learning models on aggregated historical clinical trial datasets to generate longitudinal virtual patient records. Its product produces virtual medical records that mimic demographic information, common test results and biomarkers to support analysis in clinical trials. The company works with pharma, biotech and academic researchers and has at least one named partner, Merck KGaA, which is using Unlearn to incorporate prognostic information from digital twins into randomized controlled trials. Unlearn says its approach can enable smaller control groups and more precise estimates of treatment effects while supporting regulatory decision-making. The company was founded in 2017 by Charles Fisher, Aaron Smith and Jon Walsh and currently has a 40-person headcount with plans to double staff. With the new capital, Unlearn intends to expand operations into additional disease areas. Unlearn builds a machine-learning platform that creates Digital Twins to populate Intelligent Control Arms in clinical studies. Its flagship product, Digital Twins for Alzheimer’s Disease, generates AI models statistically indistinguishable from real patient data. The company says sponsors can add cohorts of Digital Twins prospectively to enable smaller trials or retrospectively to add power to ongoing studies. Unlearn’s AD Digital Twins are already being used as part of an ongoing pivotal trial. The firm works with biopharmaceutical and medical device companies as well as regulators to meet scientific and regulatory standards. CB Insights ranked Unlearn among its Digital Health 150 companies in 2020. Unlearn.AI has developed the DiGenesis machine learning platform that builds disease-specific "digital twin" profiles from clinical-trial datasets to act as controls. The platform matches patients by demographics, lab tests and biomarkers to reduce reliance on physical matched controls and lower placebo enrollment. Unlearn is initially applying DiGenesis to neurological diseases, notably Alzheimer’s disease and multiple sclerosis. Founder and CEO Charles K. Fisher, a former principal scientist at Pfizer, leads the company and has been developing the platform since 2017. The company does not disclose its commercial partners or details of active live trials, though it says partners are already running studies using Digital Twins. Recent funding is intended to move the company closer to commercial rollouts and to advance regulatory engagement.

Total raised

$127M

Funding rounds

5

Latest round

Series C

Latest activity

Feb 2024