Berg applies machine-learning platform to PhII pancreatic cancer trial


Berg is taking its tech-enabled approach to drug development into the clinic. The biotech, which is known for making brash statements about its ability to slash preclinical timelines, has incorporated its machine-learning technology into a Phase II pancreatic cancer trial in an attempt to identify the patients who are most likely to respond to the treatment.

In its early years, Berg, which was cofounded in 2006 by real estate billionaire Carl Berg, applied a combination of genomics, systems biology, computational modeling and artificial intelligence to the discovery of a pipeline of products. Now, with two candidates in the clinic in multiple indications, the biotech is aiming to use similar capabilities to improve its odds of success in human trials. A Phase II study combining Berg’s lead candidate, BPM 31510, with gemcitabine is acting as an early testing ground for the concept.

Berg is typically confident about the capabilities of its technology. "With use of Berg’s Interrogative Biology platform, we will be applying our precision medicine approach where output from this trial will allow us to match patients to this given combination based on their biological profile,” Berg CEO and cofounder Niven Narain said in a statement.

This will entail analyzing blood and tissue samples from patients to create what Berg has dubbed “Molecular Maps,” a resource the company thinks will enable it to predict who is likely to respond positively to the treatment. Framingham, MA-based Berg is planning to enroll 25 patients with metastatic pancreatic adenocarcinoma.

Mayo Clinic is among the small list of centers signed up to enroll patients in the study, which is scheduled to wrap up early in 2019. By then, Berg is hoping to have generated data to show its treatment can have an effect on overall response rates. The expectation at Berg is that BPM 31510 will achieve this objective by reactivating cell death pathways in mitochondria.

- read the release