Unlearn.AI Closes On $12M To Improve Clinical Trials

It is often difficult to find enough patients for clinical trials.

So San Francisco-based Unlearn developed DiGenesis, a machine-learning platform that generates control patient data for clinical studies. Now the company has secured a $12 million Series A financing.

Subscribe to the Crunchbase Daily

The Unlearn platform processes historical clinical trial datasets from thousands of patients to build the disease-specific machine-learning models used to create what is called “Digital Twins” and their corresponding virtual medical records.

Digital Twins are records that include demographic information, lab tests and biomarkers that look identical to actual patient records in a clinical trial. By generating this kind of data, Unlearn aims to reduce the number of patients required to run a trial while maintaining rigorous standards of evidence and study randomization and blinding, said Charles Fisher, Ph.D., founder and CEO of Unlearn.AI.

“Patients are limited, which limits the number of trials that we can run,” he said. “If there is a database of 1,000 patients, but you need 1,000 patients for a trial, you can only run one trial. By reducing the number of patients needed for a trial, you can run more than one trial using the same patient population.”

Currently, Unlearn works with clinical trials involving Alzheimer’s and Multiple Sclerosis. Fisher said Alzheimer’s clinical studies can take an average of eight years, with the trial coordinators spending one year following the patient and seven years finding patients for the study.

“If we could reduce that number, we could do trials faster,” he added.

Growth plans

The $12 million Series A financing was led by new investor 8VC with participation from all the company’s existing investors including DCVC, DCVC Bio and Mubadala Capital Ventures.

In addition, 8VC Principal Francisco Gimenez, Ph.D., has joined the Unlearn board of directors.

To date, Unlearn has raised $17 million, and the company’s last fundraise was about a year ago, Fisher said.

Unlearn plans to use the new funds to grow its headcount from about 15 currently to double that over the next year, he said. The company also plans to expand its business to determine other appropriate areas for Digital Twins in future clinical studies.

Revenue is not a major key performance indicator for Unlearn, but making clinical trials faster and needing fewer patients is.

“If you are a pharmaceutical company developing a drug that is going to sell $10 billion per year, and you can get the drug two to three years faster, you can then be making tens of billions of dollars in revenue,” Fisher said. “That requires us to achieve a certain track record and validation. It’s not about revenue for us, but building that track record.”

Illustration: Dom Guzman