(Reuters) - The failure of Eli Lilly and Co's (>> Eli Lilly and Co) Alzheimer's treatment to slow declines in mental capacity of patients with even mild symptoms cast serious doubt over whether a long-held approach to the memory-robbing disease could provide meaningful benefit to millions of people at risk.

Many investors and families had been counting on solanezumab to become the first approved medicine to slow progression of the disease, which the Alzheimer's Association estimates will strike as many as 28 million Americans by midcentury.

But its third and latest failure in a large Phase III trial, this time in patients seen as more likely to be helped, could turn researcher and investor focus beyond drugs that directly target beta amyloid protein in the blood and brain, long thought to be an underlying cause of Alzheimer's.

"This will increase the interest in tau-lowering drugs for sure," said Dr. Sam Gandy, director of Mount Sinai School of Medicine Center for Cognitive Health, referring to a different protein being targeted in earlier-stage research, including by Lilly, Johnson & Johnson (>> Johnson & Johnson) and others.

Lilly shares fell 10.5 percent after the U.S. drugmaker on Wednesday announced long-awaited results of its Expedition 3 study, showing patients treated with solanezumab did not experience a significantly greater slowing in cognitive decline than those given placebos.

The news caused a 3.8 percent drop in shares of Biogen Inc (>> Biogen Inc), whose aducanumab also targets beta amyloid believed to cause toxic brain plaques.

Incoming Chief Executive Officer David Ricks called the results "heart-breaking."

Lilly has spent more than $3 billion on Alzheimer's research over the past 27 years and will keep pushing for effective treatments, he said. He noted solanezumab was being studied in a separate trial called "A4" among people with no Alzheimer's symptoms but who have the beta amyloid plaques.

Dr. Reisa Sperling, a professor of Neurology at Harvard Medical School who leads the A4 trial, said she continued to believe beta amyloid is an important factor in the development of Alzheimer's. But she expressed concern that some patients could drop out of her study, and some might not enroll, due to weakened faith in the beta-amyloid link to Alzheimer's.

Lilly said it would take a $150 million charge in the fourth quarter for the failed trial and on Dec. 15 would update its 2016 financial outlook and provide a 2017 forecast.

Ricks affirmed Lilly's forecast for annual revenue growth of at least 5 percent between 2015 and 2020. Lilly has strong growth prospects without solanezumab because of new treatments for diabetes, cancer and other conditions, he said.

"Lilly is having one of the best new-product cycles in its history," said Suntrust Robinson Humphrey analyst John Boris, who predicted its earnings would grow in the "mid-teens" percentage range next year.

Some analysts had said solanezumab, if approved, could generate up to $10 billion in annual sales and boost Lilly's earnings for years.

Biogen is racing to complete Phase III trials of aducanumab, which is designed to clear beta amyloid that has already formed plaques.

STILL HOPE FOR TREATMENTS

As with cancer, many experts believe combinations of medicines, each having different mechanisms, would be needed to greatly slow Alzheimer's progress or stop it in its tracks.

Gandy said amyloid drugs may work best in patients who have yet to develop plaques readable on scans. "My guess is that they would be part of a cocktail with something else, that they won't in and of themselves be enough," he said.

One of the biggest hopes is a class of experimental drugs called BACE inhibitors, which work by blocking production of beta amyloid. Lilly and others, including Biogen and Merck & Co (>> Merck & Co., Inc.), are conducting late-stage trials of such drugs.

In two original 18-month studies completed in 2012, solanezumab failed to slow cognitive decline or loss of physical functioning for 1,000 patients with mild to moderate disease in each trial.

But the combined data for just mildly affected patients suggested solanezumab caused significant slowdowns of 34 percent in mental decline and 18 percent in loss of functional abilities versus placebo.

For Expedition 3, Lilly extended the pair of large trials by another two years and enrolled only mildly impaired patients. Researchers continued to provide solanezumab to patients who had taken it during the initial studies and also allowed the placebo groups to switch to solanezumab.

(Reporting by Ransdell Pierson; Additional reporting by Bill Berkrot; Editing by Lisa Von Ahn and Richard Chang)

By Ransdell Pierson