Read our COVID-19 research and news.

Eli Lilly's corporate headquarters in Indianapolis.

Eli Lilly's corporate headquarters in Indianapolis.

Eli Lilly

Why the big change to Lilly’s Alzheimer’s trial is not evidence its drug has failed again

When pharmaceutical company Eli Lilly in Indianapolis last week announced a major change to its closely watched clinical trial for the Alzheimer’s drug solanezumab, some in the scientific community and drug development industry cried foul. To critics, the company’s decision to eliminate changes in a person’s daily ability to function as a primary measure of solanezumab’s efficacy and focus solely on a cognitive test seemed like a last-ditch attempt to keep a doomed drug from failing its third trial. Lilly’s stock plunged by nearly 5%, apparently reflecting that sentiment.

Largely lost in the online “chatter,” however, was that Lilly’s move reflects a growing scientific consensus about how the early stages of Alzheimer’s disease progress, says Dennis Selkoe, a neurologist at Brigham and Women’s Hospital in Boston, who is not involved in the Lilly trial. “From the point of view of a neurologist who’s seen hundreds of patients, [Lilly’s decision] makes clinical sense,” he says.

Solanezumab is an antibody designed to bind to and promote the clearance of the β-amyloid protein, which forms plaques around the neurons of people with Alzheimer’s. Not everyone agrees that these plaques are at the root of the disease—a concept called the amyloid hypothesis, of which Selkoe is a major proponent—but fighting them is the foundation of nearly all current efforts in Alzheimer’s drug development. By helping destroy the plaques in people with early stages of Alzheimer’s, Lilly hopes solanezumab can slow the disease’s progression.

A major challenge of such trials is how to measure the drug’s benefits, Selkoe says. Although people with early Alzheimer’s may show mild memory impairment and problems with attention and focus, they can often follow recipes, make a cup of coffee, or drive a car, Selkoe says. Such abilities are unlikely to change much over the course of an 18-month clinical trial, and any drug aimed at improving these daily functions is unlikely to show any difference from placebo, he says.

Regulatory agencies such as the U.S. Food and Drug Administration (FDA) have traditionally required—and still officially stipulate—that Alzheimer’s drugs demonstrate effectiveness in both cognitive tests and functional measures. Recognizing that functional measures might not be appropriate in early or mild stages of the disease, however, FDA and other agencies have begun to relax their initial standards for drugs aimed at preventing, rather than reversing, Alzheimer’s pathology and symptoms, Selkoe says.

In two previous trials of solanezumab, called EXPEDITION and EXPEDITION II, Lilly used both a cognitive test and a functional measurement to track the response of people with both mild and moderate Alzheimer’s disease. Both trials failed to show significant benefits over placebo in either measure. Combing through data from the second trial, however, Lilly noticed that participants with mild Alzheimer’s seemed to do better than controls in the cognitive portion of the testing, says Eric Siemers, a neurologist employed by Lilly. 

That led Lilly to its current, billion-dollar gamble on a study of roughly 2100 people with only mild Alzheimer’s, which will conclude in October. The last-minute decision to ditch the functional measure in this third trial, EXPEDITION III, was not based on any insight into current data from the trial, as some have suggested, Siemers says—investigators are still blind to which participants have received the drug and which have received a placebo.

Instead, the change reflects the recognition—admittedly belated—that there is no well-validated functional assessment for mild Alzheimer’s patients, according to Siemers. “As we did more analyses, and talked to more people in the field, we got to the point where we didn’t know which functional measure is best for this group,” he says. To address that gap, the company is still collecting functional data using two separate assessments, and hopes to see whether one or the other is better at detecting any subtle changes, Siemers adds.

Scott Small, a neurologist at Columbia University who is not involved in any clinical trials of Alzheimer’s drugs, says he finds that’s a “reasonable” explanation. “If that’s where the clinical neuroscience is moving, why include [function] as an endpoint?” he says.

Still uncertain is how Lilly’s clinical trial shift will be judged by FDA and other global regulatory agencies when the final data become available, sometime in 2017. The company did submit the changes to all relevant regulatory agencies.  But because a change in endpoint does not affect how the study itself is being conducted prior to data analysis, Lilly was not required to seek official FDA or other regulatory approval to make the change, Siemers says.

Selkoe believes FDA will look kindly on solanezumab if it shows convincing evidence of cognitive benefit without functional data—particularly if those positive results are paired with encouraging data from PET scans that seek to show a reduction in amyloid plaques. “We’ve known for a long time that the cardinal manifestation of this disease [in the early stages] is minor trouble with everyday memory—it’s a very subtle process when it begins,” he says. As such, Selkoe concludes, “we’re going to have to be more sophisticated” about testing treatments, “and that’s what’s Lilly is showing evidence of.”