The failure of a promising Alzheimer’s drug in clinical trials highlights the gap between diagnosis – where real progress has recently been made – and treatment of the disease.

It was not just that the drug, made by Eli Lilly, did not work – maybe that could be explained by saying the patients’ illness was too far advanced when they received it. It was that the drug actually made them worse, the company said. And the larger the dose they took, the worse were patients’ symptoms of memory loss and inability to care for themselves. Not only that, the drug also increased the risk of skin cancer.

So when Lilly announced Tuesday that it was ending its large clinical trials of that drug, semagacestat, researchers were dismayed.

"Obviously, this is disappointing news, to say the least," said Dr. Steven Paul, an Alzheimer’s researcher and a recently retired executive vice president at Lilly.

Beyond the setback for Lilly, the study raises questions about a leading hypothesis of the cause of Alzheimer’s and how to treat it. The idea, known as the amyloid hypothesis, says the disease occurs when a toxic protein, beta amyloid, accumulates in the brain. The idea is that if beta amyloid levels are reduced, the disease might be slowed, halted or even prevented if treatment starts early enough.

The Lilly drug, like most of the more than 100 Alzheimer’s drugs under development, blocks an enzyme, gamma secretase, needed to make beta amyloid. It was among the first shown to breach the blood-brain barrier and reduce levels of beta amyloid in the brain. And, company studies showed, it did reduce amyloid production.

"We did get enough in the brain to have an effect," said Dr. Eric Siemers, medical director of Lilly’s Alzheimer’s disease team. "Unfortunately, the effect was not what we wanted."

Now researchers are focused on what went wrong, and why.

Some, like Dr. Lon Schneider, an Alzheimer’s researcher at the University of Southern California, say the drug’s failure may mean the field is rushing off a cliff in its near single-minded focus on blocking the production of amyloid. Schneider, like most leading Alzheimer’s researchers, consults for a number of drug companies, including Lilly.

The Lilly study’s failure, he said, "chips away at that approach to testing the amyloid hypothesis.

"We don’t know what the drug targets for Alzheimer’s disease are," Schneider said. "We don’t know because we don’t know the causes of Alzheimer’s."

At the very least, said Dr. P. Murali Doraiswamy, an Alzheimer’s researcher at Duke University, the Lilly result "clearly tells us that our current views may be too simplistic."

Doraiswamy said he was not abandoning the amyloid hypothesis. But, he said, "this is a time of major soul-searching in the field.

"What worries me is that we don’t know if this was a toxicity unique to Lilly’s drug and this late-stage population or whether it also applies to similar anti-amyloid therapies given at earlier stages of the disease," Doraiswamy said.

The bad news came on the heels of what researchers see as a resurgence of hope in this challenging field. With new cooperation in research they have made advances in diagnosing Alzheimer’s, a disease that used to be uncertain until autopsy.

And those new diagnostic tests are still exciting, researchers said.

PET scans of amyloid plaques in the brain and tests of cerebrospinal fluid can show amyloid accumulation long before people have symptoms of Alzheimer’s disease and, as recently reported, appear to identify people at high risk of developing the disease. Researchers believe the best time to try to alter the course of the disease is before memory loss. By then, brain cells are dead or dying and are unlikely to be restored.

At this point, though, when there is no treatment, those tests are primarily a benefit for companies testing new therapies and researchers trying to understand the disease’s progress. .

The long journey of semagacestat began more than a decade ago when Lilly scientists discovered it could block gamma secretase in laboratory experiments. Years of work followed, showing it appeared safe, that it got into the brains of people, that it reduced the production of amyloid in the brain.

Finally, in 2008, Lilly began two large studies of semagacestat, enrolling more than 2,600 people with Alzheimer’s disease. The company did not expect its drug to reverse the disease – patients’ brains were too ravaged for that, said Richard Mohs, Lilly’s team leader in Alzheimer’s research. But it did hope to slow the disease’s progression.

Now, with the abrupt end of the studies, patients will continue to be followed but no one will be taking any more of the drug.

"The fact that people got worse means there is biology we don’t understand," Mohs said.

There are several possible explanations.

One is that the drug altered the functioning of other proteins in the brain and body – it now appears that gamma secretase is involved in the production of about 20 proteins in addition to beta amyloid. Companies, including Lilly, are developing drugs that block gamma secretase from making amyloid but have little effect on other proteins.

One company, Bristol-Myers Squibb, says that is what its drug does. Its drug is now being tested in two clinical trials. In one, the participants have Alzheimer’s. In the other, they have lesser memory impairment and have brain amyloid PET scans and tests of cerebrospinal fluid showing amyloid is accumulating in their brains, indicating that they are likely to develop Alzheimer’s.

"We still like the amyloid hypothesis," said Charlie Albright, a Bristol-Myers group director in neuroscience biology. The Lilly drug failure "doesn’t affect our enthusiasm about going forward."

Another possibility is that the enzyme is decreasing production not just of a dangerous form of amyloid, known as a beta 42, but also of another form, a beta 40, that may protect the brain. Companies are developing so-called selective gamma secretase inhibitors, Paul said, which only block the production of a beta 42.

Lilly and other companies are also testing the use of monoclonal antibodies to reduce amyloid levels.

And companies are pursuing a more difficult target – blocking a protein, tau, that accumulates in dead and dying nerve cells after the disease is under way.

But Alzheimer’s experts worry about the future. The research is extremely expensive – Lilly spent hundreds of millions of dollars on its failed drug – and it can take a decade or more to know if a drug works. It can take even longer if drugs are tested in people with mild symptoms of Alzheimer’s disease or in people who are at high risk but have no symptoms yet – a direction many think is necessary to really make a difference.

"Failures certainly don’t build energy and enthusiasm," said Dr. Samuel Gandy, an Alzheimer’s researcher at Mount Sinai Medical Center. "The market is still there, but failures do take their toll."