SAN DIEGO – Solanezumab may have not have slowed the relentless march of Alzheimer’s disease (AD), but it was a valuable proving ground of the amyloid hypothesis, experts said during a wide-ranging discussion of Lilly’s failed EXPEDITION 3 trial.
Lilly representatives and EXPEDITION investigators released the study’s full results at the Clinical Trials in Alzheimer ’s disease meeting. – findings that should be read as tremendously encouraging rather than a defeat, according to Paul Aisen, MD, director of the Alzheimer’s Therapeutic Research Institute at the University of Southern California, Los Angeles.
“We have here a negative study that confirms a beneficial treatment,” said Dr. Aisen, who was also an EXPEDITION 3 investigator. “We have a treatment that engages its target, binds to soluble amyloid, and, by virtue of that mechanism, is slowing cognitive and functional decline,” not only in EXPEDITION 3, but in its predecessors EXPEDITION and EXPEDITION 2.
“This is not a refutation of the amyloid hypothesis but a confirmation of it.”
Nevertheless, the trial must be read as a failed one, he admitted. There was no statistically significant separation between solanezumab and placebo on the ADAS-Cog14, a combined assessment of cognition and function that was the study’s primary endpoint. The active group experienced 11% less cognitive decline than did the placebo group, but the p-value remained tantalizingly below the level of significance, at 0.095.
“But what if solanezumab had hit at 0.05 instead of 0.095?” Dr. Aisen asked. “In fact, it would still be a small effect size,” which would have thrown into question the drug’s clinical utility. “Going into this, we thought we might see a 30% slowing of decline on the ADAS-Cog, and it was disappointing to only get 11%. But that is also what we saw on the key secondaries. Overall the effect size looks to be about 12%-13%, and that’s just too small.”
EXPEDITION 3 was the last of a triad of solanezumab studies, all of which posted intriguing signals of cognitive and functional benefit in patients with mild-moderate AD. It was based on subgroup analyses of EXPEDITION 1 and 2, both of which failed to meet their primary endpoints. But when researchers pooled the mild patients from both studies, they found that solanezumab was associated with a 34% slowing of cognitive decline on the ADAS-Cog14. This translated to a clinical change of less than 2 points on the scale, however.
Lilly very carefully drafted EXPEDITION 3 to come as close to recreating those findings as possible but still stumbled over results that were numerically positive for the antibody but not statistically significant or clinically meaningful.
Lawrence S. Honig, MD, professor of neurology at Columbia University Medical Center, N.Y., and principal investigator of the EXPEDITION 3 study, presented the study’s full results to a packed audience on Dec. 8.
The study comprised about 2,000 patients with imaging-confirmed amyloid brain plaques and mild-moderate AD. They were randomized to placebo or monthly injections of 400 mg solanezumab for 80 weeks. The global study was conducted in 11 countries and 210 study sites.
Dr. Honig detailed the key secondary endpoints of cognition and function, and also revealed biomarker data.
While the ADAS-Cog failed to meet statistical significance, changes in the Mini Mental State Exam score did, with a 13% slowing of decline compared to placebo (P = .014). There was also a significant 5% difference in the Clinical Dementia Rating scale-sum of boxes test (P = .004).
Outcomes were mixed in measures of function. The Alzheimer’s Disease Cooperative Study activities of daily living (ADCS-ADL) and its related measure, the ADCS-ADL inventory instrumental items, posted significant results with 15% and 14% differences, respectively, relative to placebo (P = .009 and .019, respectively).
But results on the Functional Activities Questionnaire, an informant measure of more complex activities, were not significant, with only a 7% separation from placebo and a P value of .140.
Biomarkers trended the right way, Dr. Honig noted. Solanezumab did what it was supposed to: bind soluble amyloid beta. This resulted in a 500- to 800-fold increase in the protein in plasma relative to placebo. There were no changes in amyloid brain plaques as measured by PET imaging, but this was no surprise, Dr. Aisen said, since the antibody doesn’t recognize fibrillar amyloid.
“What we expect to see with biomarkers differs based on the epitope targeted. Solanezumab ignores plaques. It targets the middle of the peptide, binding to soluble AB. Now how that helps AD is something of a debate, but it is important tor recognize that it does not attack plaques. Instead, by tying up monomeric AB, it may change the dynamic exchange of various species of amyloid around plaques; the toxicity of amyloid is thought to reside as much in oligomeric species as in the fibrillar deposits. I see this [plasma AB increase] as confirming that it’s tying up monomeric amyloid species and that the result is a slowing of disease progression. I believe it is supportive of the amyloid hypothesis.”
Solanezumab had no significant effect on tau, either in cerebrospinal fluid or imaging, nor did it change the progression of ventricular enlargement, a marker of whole brain atrophy.
The antibody was quite safe, with 17% of patient reporting an adverse event, compared to 19% of placebo patients. There were 9 deaths in the solanezumab arm and 16 in the placebo arm; about 4% of each group discontinued treatment because of an adverse event.
Although the data discussion was framed in the most hopeful light possible, no one on the panel attempted to massage it into a more clinically positive form. On a webcast in late November, Eric Siemers, MD, senior medical director of the Alzheimer’s Disease Global Development Team at Eli Lilly, said the company was disappointed but would not bring solanezumab forward for approval for mild or moderate AD patients. He echoed that sentiment during the panel discussion.
“We didn’t expect this to be a cure for this disease, but we did hope it would be the first drug to slow its progress. So yes, we are very disappointed.”
He and Dr. Aisen confirmed, however, that two other trials using solanezumab in a different population will go forward uninterrupted. The Anti-Amyloid Treatment in Asymptomatic Alzheimer’s study (A4 study) is investigating its effect in cognitively health elders with Alzheimer’s risk factors, and the Dominantly Inherited Alzheimer’s Network (DIAN) study of patients with autosomal dominant mutations in Alzheimer’s genes.
There was also brief discussion of dosing. Some audience members suggested that a higher dose than EXPEDITION’s 400 mg might have bumped up efficacy, and asked if Lilly would reconsider the dosing schedule in the A4 and DIAN studies.
“There has been a lot of discussion around that,” said Dr. Siemers. “But it’s not as easy and straightforward as you think.”
Nevertheless, Dr. Aisen is excited about solanezumab’s potential in these trials that target the disease at its earliest phase, even before cognitive symptoms develop. “I expect all antiamyloid treatments would work better when neurodegeneration is not extensive. Any of the antiamyloid antibodies would theoretically be more effective at a preclinical stage of AD than even in the mild dementia stage.”
Maria Carrillo, PhD, chief science officer of the Alzheimer’s Association, said that EXPEDITION 3 is far from a path to nowhere. Instead, she urged the research community, patients, and families to double down on their commitment to tackling the disease.
“These results stress the urgency for pushing forward harder. This is not a time to slow down. It’s a time to ramp up our efforts. This is not the time to sit back and say, ‘The amyloid hypothesis has been the wrong pathway and we need to drop it.’ But we also need to pursue other pathways, to broaden our approach and to broaden the armamentarium our clinicians will need to combat this disease.”
“This is not a win, true. But it gets us a little closer to one.”