How to Improve Drug Development? Fail Fast!

In this morning’s PSA panel on “Development Dilemmas and Opportunities,” Michael Clayman, MD, VP of Lilly Research Laboratories at Eli Lilly & Co., presented a unique option for optimizing clinical pipeline success. Perversely, it depends on failing fast. Clayman heads Lilly’s Chorus division, an organization that is trying to create a new model for drug development built on not reducing attrition but increasing the chances post-clinical proof of concept that a drug will make it to market. We took an in-depth look at Chorus in May in IN VIVO.

Clayman estimates that 90% of drugs in development will fail anyway, so why devote the time, the resources—the dollars—driving a product forward if it’s not going to make to market? The goal of his group: cut costs, and dramatically narrow the time to a decision point—typically proof of concept in man, what Clayman jokingly refered to today as “pull out your checkbook”—down to as little as twelve months.

It’s a goal Clayman claims Chorus is well on its way to achieving. To date, the company has shown that it can shave 12 to 18 months off the time it takes a drug to reach proof of concept and reduce the R&D dollar spend from $30 million to $3 million.

But, outside these metrics, there aren’t obvious ways to measure the group’s success. It’s not as if the company can use drug approvals as a measure, since the goal of Chorus isn’t to get drugs on the market, but to de-risk them as much as possible. Indeed, it’s an organizational tool to manage Lilly’s vast portfolio of drug products so that the bias is on the ultimate winners. And while nearly 80% of Lilly molecules might be pushed forward according to this program, to date the strategy has been applied to just 10.

According to Clayman, one critical component of the strategy is that Chorus is compound agnostic. No one on the 24-person team has a driving loyalty to a molecule that might sway him or her to push one project forward over another. The group also operates as an autonomous division within Lilly so that it is not hide-bound by the operational infrastructure of the larger organization. “Once a molecule is transferred to us, it’s no longer worked on by Lilly scientists. We outsource the experimentation,” he says.

That level of outsourcing is likely to be troubling to most other major pharmas. It seems unlikely that many outfits would be willing to adopt such a strategy unless there were significant proof that it improves R&D productivity. Until such time, expect the refrain to remain simply Lilly’s chorus.