Excelsior Sciences raises $95 million to speed small molecule drug development using AI

By Kamal Choudhury

Dec 3 (Reuters) - Excelsior Sciences said on Wednesday it has raised $95 million to develop technology that uses machines and artificial ​intelligence to accelerate the development and testing of small molecules, ‌aiming to cut years off drug discovery timelines.

The company announced a $70 million Series A round ‌co-led by Deerfield Management, Khosla Ventures and Sofinnova Partners, along with a $25 million grant from New York's Empire State Development.

It did not disclose the valuation at which the funds were raised.

Small molecules make up the majority of medicines ⁠approved in the United ‌States. While treatments such as antibodies and cell therapies are growing, Excelsior CEO Michael Foley said around 60% of ‍new U.S. FDA-approved drugs are small molecules.

However, discovering and making small molecules is slow and costly, often taking more than a decade and billions of dollars as ​each molecule requires a custom synthesis process.

Excelsior's "smart bloccs" approach, which one ‌of its lead investors described as a new modular language, can help AI "better predict how to create, optimize new therapies".

Jim Flynn, managing partner at Deerfield, said a typical cycle of four months or more, often involving work split across the United States, Asia and multiple contractors, could be reduced ⁠to about two weeks in a single ​automated facility.

The same process can then be ​scaled up for manufacturing, potentially saving another year to 18 months before clinical trials begin, Flynn added.

The New York-based firm, ‍spun out of ⁠investment firm Deerfield, expects to demonstrate its full platform working within 12 months and apply it to at least one drug discovery ⁠program.

The funding round also received backing from Eli Lilly, Cornucopian Capital, Illinois Ventures and academic ‌institutions such as MIT and Princeton University.

(Reporting by Kamal Choudhury ‌in Bengaluru; Editing by Vijay Kishore)