Pharma’s R&D Fix: Share Risk, Decide Early, Sell Services

Panelists mull ways around failure-prone Phase II bounty

High failure rates in trials plus pipelines overloaded with Phase II candidates are putting the squeeze on big pharma, and "partnering, all across the value chain" is only part of the solution, said Steven Paul, president of Lilly's research labs division.

For a company to renew its entire roster of drug prospects every decade or so is necessary and "feasible, but not the way we are currently running our R&D," he said.

Paul's remarks came during a Sept. 24 workshop panel at Windhover Information's Pharmaceutical Strategic Alliances Conference in New York.

Phase II attrition rates range between 70 percent and 75 percent. The figures for new molecular entities in Phase III is "still pretty lousy" as well, Paul said - and not always because of some unexpected, previously unseen toxicity or side effect.

"You shouldn't be losing Phase III molecules for lack of efficacy," he said, but it's happening.

Groundbreaking research is bound to bring more blowups, noted panelist Paul Stoffels, Johnson & Johnson's chairman of global R&D.

Blockbuster drugs on the market came out of the "second or third or fourth try on the same target," Stoffels said. "We're coming in with a whole field of first-in-class drugs."

Still, Martin Mackay, president of global R&D for Pfizer, said his firm is making progress in the effort to sift winners from the Phase II crop.

The problem of losing molecules due to flaws in physical characteristics Pfizer has "if not nailed, pretty much so," Mackay said, while other causes still take a toll.

Registration issues are getting harder to predict, too, said G. Mikael Dolsten, president of R&D for Wyeth Pharmaceuticals, as FDA tightens the screws and scientists debate how clinically meaningful is a demonstrated benefit.

Paul called for better research done earlier, so companies can understand as soon as possible what need the would-be drug will address, and how.

"We write a draft launch label when we're doing a target - we don't even have a lead yet," he said.

"Sometimes you don't know what the product profile is going to be until you see the label," Paul conceded, but this only makes envisioning the label more important.

Devising a diagnostic to use with the compound worked well with the breast-cancer drug Herceptin (trastuzumab). The approach is not practical "unless you start from the get-go," as Genentech did, Paul said.

"It might not be a diagnostic" that holds the key to efficacy, he said. "It might be a demographic" - a sub-population for which the drug's value "goes beyond genetics" and is due to reasons "we don't know and may never know," Paul said.

Panelists agreed that one fix for pharma R&D woes could be a risk-sharing deal like the one between Elan and Wyeth, involving bapineuzumab for Alzheimer's disease.

Bapineuzumab, a monoclonal antibody designed to spark an immune response and clear beta amyloid plaque in the brain, provided few thrills in Phase II (¹ (Also see "Wyeth’s Pipeline Star Bapineuzumab Looks More Long Shot Than Sure Bet" - Pink Sheet, 30 Jul, 2008.)).

Another is the portfolio approach, which Paul likened to buying tickets for rides at a fair.

Companies should "balance the five- or six-ticket rides with the one- or two-ticket rides," putting R&D effort into major market long shots as well as smaller products with high unmet need, he said.

"The little ones - frankly, I don't know that they're any easier or harder than big ones," Paul said.

Pfizer's Mackay said broad-scale partnering should include everything the company has to offer, and suggested a model used by those in other industries.

Mackay proposed selling such services as high throughput screening (if the firm has unused capacity) to get capital while sharing risk through deals, a strategy "very different, I would submit, from the way we do business now."

The Windhover conference continues through Sept. 25.

- Randall Osborne ([email protected])