Eli Lilly announced on August 18 that it was ending its large clinical trials of the drug semagacestat in patients with mild-to-moderate Alzheimer’s disease (AD). Semagacestat is a drug that inhibits the activity of the enzyme gamma secretase. Gamma secretase catalyzes the processing of beta amyloid, a protein that slowly accumulates in the brains of patients with AD. The beta amyloid hypothesis of AD has been a guiding force in the search for disease-modifying treatments for AD for more than three decades. Simply stated, this hypothesis posits that if beta amyloid levels are reduced, the disease might be slowed, halted or even prevented if treatment starts early enough.

The premise of Lilly’s trials was that inhibition of gamma secretase would reduce the amount of beta amyloid protein in the brains of patients with mild-to-moderate AD, and would lead to improved cognitive outcomes. Unfortunately, instead of making patients better, those who received semagacestat got worse. As the dose went higher, patients’ symptoms, including memory loss and the inability to engage in activities of daily living, worsened. Moreover, the drug also increased the risk of skin cancer.

High profile trials have failed before, but this incident did not just condemn a single drug, it could change the strategy used to investigate all disease-modifying therapies for AD.

This is not the first setback for the beta amyloid hypothesis, just the latest. In the past few years, no fewer than three investigational drugs, all with mechanisms of action aimed at reducing brain amyloid burden, have done poorly in last-stage clinical trials. Now, even staunch proponents of the beta amyloid hypothesis are beginning to have doubts as to continuing in this direction.

There is no compelling reason to shelve the beta amyloid hypothesis of AD, but it is likely that AD is more complex than just the effects of beta amyloid protein. Hundreds of metabolic processes converge and diverge in the human brain; thus it may be shortsighted to believe that beta amyloid is the final common pathway for AD.

As of today, we have better tools to diagnose and assess AD than to treat it, but the search continues. The beta amyloid hypothesis remains exactly that, a hypothesis, not the cause of AD. We should continue on this path, but it is unlikely to be the only one leading to a viable AD treatment. Who knows? The answer may be just a few miles down the road.