DUBLIN – Shares in AC Immune SA were off more than 40% Sept. 23 on news that its tau-directed antibody, semorinemab, which is partnered with Genentech, failed to demonstrate efficacy in a phase II trial in Alzheimer’s disease.

The placebo-controlled Tauriel study, which recruited 457 patients with prodromal or mild disease, failed to meet the primary efficacy endpoint, which comprised a reduction in the decline in the Clinical Dementia Rating-Sum of Boxes (CDR-SB) score. It also missed two key secondary endpoints, based on the Alzheimer’s Disease Assessment Scale-Cognitive Subscale 13 (ADAS-Cog13) and the Alzheimer’s Disease Cooperative Study Group – Activities of Daily Living Inventory (ADCS-ADL). A phase II study in 260 patients with moderate Alzheimer’s disease is ongoing. Top-line data are expected in the second quarter of 2021.

Andrea Pfeifer, CEO, AC Immune

At this point, it is unclear whether the outcome was due to a molecule effect, a class effect or a target effect. “I honestly don’t know,” Andrea Pfeifer, CEO of Lausanne, Switzerland-based AC Immune, told BioWorld. As well as being unhappy at the outcome, she was also unhappy at the lack of data from partner Genentech, a subsidiary of Basel, Switzerland-based Roche Holding AG. “We have not seen more than half the data,” she said. It is, therefore, impossible to draw any firm conclusions without having imaging data, biomarker data and patient-level data to hand. Genentech plans to report the data at an upcoming medical congress.

But there is no escaping the overall result. “This is clearly disappointing news for the company and for the Alzheimer’s community given that tau has been increasingly viewed as a promising target for treating AD, and investors (and the AD community) have been focused on this event to potentially validate the anti-tau antibody approach,” SVB Leerink LLC analyst Marc Goodman wrote in a flash note.

“The key question,” he added, “is whether there should be a negative read-through into the anti-tau antibody class or other anti-tau assets. We need to see additional data, including evidence on tau biomarker movement from Tauriel, to understand if there was any target engagement, which may still support a tau hypothesis even if the trial missed, and until then, it is hard to tell why the study failed – if it was due to suboptimal target engagement or a different factor in the tau-targeting approach.”

Tau is a microtubule-associated protein which, among other functions, provides structural support and stability to eukaryotic cells, particularly to neurons. In Alzheimer’s and other neurodegenerative diseases or tauopathies, hyperphosphorylated tau proteins seed the formation of pathological aggregates, which form neurofibrillary tangles that damage neuronal tissue. Because tau protein aggregation correlates more directly with the severity of the Alzheimer’s pathology, some drug developers took an optimistic view that it may make a better target than amyloid beta (a-beta) deposition, which has proved a notoriously difficult process to target with a drug. After a collective effort over several decades, one a-beta antibody, aducanumab, discovered by Neurimmune AG and developed by Biogen Inc. and Eisai Co. Ltd., is currently under FDA review, although the supporting data are by no means cut and dried.

A recent paper from a group led by Bradley Hyman, at Massachusetts General Hospital and Harvard Medical School, suggests that targeting tau may be every bit as complicated. In detailed investigations of the tau profiles of 32 Alzheimer’s patients, they identified “unexpected biochemical and biophysical differences in seed-competent tau across individuals with otherwise typical Alzheimer’s disease.” Moreover, the different forms of tau correlated with differences in the aggressiveness of the clinical course of the disease. Their work appeared online in Nature Medicine on June 22, in a paper, titled “Tau molecular diversity contributes to clinical heterogeneity in Alzheimer’s disease.”

Although she cautioned that this work was done in vitro, Pfeifer said she “fully agrees” that tau species are heterogeneous. At the same time, semorinemab was developed as a pan-inhibitor of all tau species. “It should work.” Although the jury is still out – just about – on the antibody, AC Immune is targeting tau through a couple of other approaches. A small-molecule drug, ACI-3024, which is partnered with Eli Lilly and Co., of Indianapolis, targets tau intracellularly. A vaccine, ACI-35.030, which is designed to elicit an antibody response against phospho-tau, is undergoing a dose-escalation phase Ib/IIa trial. It is partnered with Johnson & Johnson Co., of New Brunswick, N.J. Interim data are expected early next year. For its part, Genentech recently in-licensed another anti-tau antibody, bepranemab (UCB-0107), from Brussels-based UCB SA. (See Tau-directed therapies in clinical development for Alzheimer’s disease, below.)

AC Immune and Genentech also remain allies on a long-standing prevention trial of an a-beta-targeting antibody, crenezumab, which is being tested in a Colombian cohort with a genetic predisposition to Alzheimer’s. Data are expected in 2022. Given the long-term nature of the study – it is running for five years – it could shed crucial light on a key question that has dogged drug development in Alzheimer’s – whether two-year or 18-month studies of investigational agents are long enough to demonstrate effects on disease processes that can run for decades. The same drug has already failed two phase III trials in symptomatic patients.

Shares of AC Immune (NASDAQ:ACIU) closed Sept. 23 at $4.80, down $3.92.

No Comments