Photographer: Victor J. Blue/Bloomberg

Deep in ‘the Cave,’ Lilly’s CEO Decided Alzheimer’s Drug’s Fate

They called it “the cave.”

In an unmarked room in the middle of Eli Lilly & Co.’s Indianapolis campus, researchers holed up for months in 2012 poring over data from the company’s failed trials of its experimental Alzheimer’s disease treatment, solanezumab, looking for hidden signs the drug had worked.

The company’s scientists knew what they were up against. The disease has proven impervious to all previous drugs that tried to stop it from erasing the mind. One of Lilly’s earlier treatments had failed -- in fact had made the disease worse -- burning millions of dollars of research costs along the way. The trials of solanezumab certainly hadn’t shown that it slowed Alzheimer’s enough to get approval from the U.S. Food and Drug Administration.

But amid reams of results from two trials, a pattern began to form: One group of patients -- those who started taking the drug when their disease was only mild or moderate -- was deteriorating slower than expected. The team took the data to Chief Executive Officer John Lechleiter.

“We had hundreds and hundreds of different slides,” said Eric Siemers, distinguished medical fellow on the development team for the drug. “We distilled it down to 20 maybe. We got halfway through the data and he looks at us and goes, ‘Well, you’ve got a real effect here.’”

That sign-off from the CEO prompted the go-ahead for Lilly to double down and spend millions more.

The outcome of that bet is about to get clearer. Next month, Lilly will release data at the Alzheimer’s Association International Conference that show how patients performed in an extension of the two trials. The glimmer of efficacy seen in the cave three years ago will either strengthen, or vanish like so many that had gone before it.

Move Fast

The drugmaker hasn’t waited to find out. A trial in patients with mild disease -- the best chance for the drug’s real success -- should start wrapping up late next year. The gamble may be worth it. There are an estimated 5.3 million Americans with the disease, and the number is growing. Analysts predict the first drug to slow the disease down will garner billions of dollars in annual sales.

It would also be a much-needed rebound for Lilly. Best-known for psychiatric drugs like the antidepressant Prozac, its top products have lost their exclusive sales rights, and revenue hit a more than seven-year quarterly low this year.

Alzheimer’s has always been a scary place for drug companies. In 2010, Lilly halted development of semagacestat. Like solanezumab, the drug was meant to prevent plaque tangles in the brain that are thought to damage neurons. Stop the plaques, stop the disease. Semagacestat worked by inhibiting a type of protein that allowed for the formation of beta amyloid, which ultimately builds up into plaques.

The Protein

But the protein that semagacestat blocked had other, helpful functions as well. Patients got worse, not better.

“We did have a lot of discussions when sema’ ran into problems,” said Richard Mohs, head of earlier stage drug testing. “Would that mean there would be problems with solanezumab?”

That didn’t end up being the case. Solanezumab attacks the plaque formation process further along, targeting beta amyloid directly to prevent the clumps from forming.

Even the small success in 2012 was a breakthrough for those in the field. At one session in the cave, Lilly’s scientists convened to decide if the data were compelling enough to move forward. Paul Aisen is a leader in the field, and directs the Alzheimer’s Disease Cooperative Study. He was consulting on the trial for Lilly, and was last to speak, Paula Garrett, the drug’s senior director of marketing, recalled.


“He started to tear up and he said, ‘I had resigned myself to think that I would never see data like this in my lifetime, and you’ve proved me wrong.’”

Still, the drug didn’t meet the study’s primary goals: It failed to restore thinking, memory or the ability to perform daily activities in the broader population of patients studied.

And there’s darkness in the cave. Keep cutting the data, and shadows appear.

“It’s one thing to data mine a single study, but to combine studies to data mine, you’re really mixing apples and oranges to come up with something,” said Howard Fillit, founder and chief science officer of the Alzheimer’s Drug Discovery Foundation. “The further down the road you get here, the more speculative and uncertain you are about what we’re seeing in the data.”

It’s been enough for investors, though. Lilly shares have risen 23 percent this year, partly on the promise of solanezumab, compared with an 7.6 percent gain for the Standard & Poor’s 500 Pharmaceuticals Index.


Other drugmakers have followed, too. Biogen Inc. said late last year that its drug BIIB037 showed promising early results and that it would be moving into a final-stage trial. Roche Holding AG said it was considering another study for a beta amyloid-targeting drug that didn’t work in a previous trial. Merck & Co. has a drug as well. And Pfizer Inc. and Johnson & Johnson have products as well.

Lilly’s biggest advantage may be time -- by starting the new, 2,100-patient trial of early stage patients in 2013, its drug could come to market two years ahead of Biogen’s.

“I’d say Lilly is one of the better companies that are really tackling the Alzheimer’s disease problem head-on and giving it everything they can,” said Colin Masters, a laureate professor at the University of Melbourne and one of the world’s top Alzheimer’s experts.

Yet praise doesn’t pay. For now, none of the assets in Lilly’s Alzheimer’s portfolio is producing profit. It will be at least 2017 before solanezumab moves from the data room to the doctor’s office.

In the meantime, study data will start to flow back to Lilly’s researchers that will show, once and for all, if solanezumab really works. The scientists will head back into a secure room, likely with a sign on the door like the one at the cave’s exit: “Put on your poker face.” Until the company is sure, nothing can leak out.

Before it's here, it's on the Bloomberg Terminal. LEARN MORE