Alzheimer’s Tau Vaccine Shows Immune Response, Safety

An active peptide vaccine designed to target pathological tau was safe and elicited immune responses in people with mild Alzheimer’s disease, though showed no effects on cognitive decline in the overall population of the phase II placebo-controlled ADAMANT trial.

Exploratory analyses suggested the vaccine, AADvac1, also slowed accumulation of plasma neurofilament light (NfL), a marker of neurodegeneration, reported Petr Novak, MD, PhD, senior clinical research scientist at Axon Neuroscience in Bratislava, Slovakia, and co-authors, in Nature Aging.

Plasma NfL rose by 28% (4.93 pg/mL) in the placebo group over the 2-year study, but by only 13% (2.09 pg/mL) in the AADvac1 group (adjusted mean difference -2.79 pg/mL; 95% CI -4.65 to -0.93, P=0.0035; Cohen’s d = -0.48).

While AADvac1 showed no cognitive benefit overall compared with placebo, post hoc analyses suggested it may slow decline in some Alzheimer’s patients. The most pronounced effects on clinical and functional outcomes were seen in a subgroup of people who were most likely positive for amyloid and tau pathology.

“To the extent of my knowledge, this is the first time that a tau-targeted immunotherapy showed clear evidence of impact on the neurodegenerative process and a strong indication of clinical effect in patients with a confirmed Alzheimer’s disease biomarker profile,” Novak told MedPage Today.

Immunotherapy is being tested in several trials to decrease levels of toxic tau proteins and help slow cognitive decline. “Tau and tau-related pathways are very important targets for the development of Alzheimer’s disease therapeutics,” observed Jeffrey Cummings, MD, ScD, of the University of Nevada, Las Vegas, who wasn’t involved with the study.

“Tau is closely correlated with cognitive decline, and tau therapeutics may have a broad window of application from preclinical Alzheimer’s disease to Alzheimer’s dementia,” Cummings told MedPage Today. “Tau vaccines are very early in the developmental cycle, but have the promise of application early and may avoid some of the side effects seen with other agents such as the ARIA [amyloid-related imaging abnormalities] observed with amyloid therapeutics.”

The 24-month, double-blind ADAMANT study randomized 196 patients, 117 to AADvac1 and 79 to placebo, in eight European countries from June 2016 to May 2017. Nearly 17% of participants dropped out, leaving 100 in the vaccine group and 63 in the placebo group at the end of the study. The last safety visit was in June 2019.

The trial’s primary objective was to evaluate safety and tolerability. Secondary objectives were to assess immunogenicity and efficacy of AADvac1 in slowing cognitive and functional decline.

The study enrolled people who had a diagnosis of probable Alzheimer’s based on National Institute on Aging/Alzheimer’s Association criteria, a Mini Mental State Examination (MMSE) score of 20 to 26, and medial temporal lobe atrophy on magnetic resonance imaging (MRI) or cerebrospinal fluid biomarker levels of amyloid and tau that were consistent with Alzheimer’s disease. Participants had a mean age of 71, and 55% were female. All participants were white.

Eleven doses of AADvac1 were administered at 40 μg per dose: six subcutaneous injections monthly, followed by five booster shots quarterly. The vaccine led to high levels of antibodies; responder rates were 96.5% at the end of the first six doses and 98.3% overall.

Serious adverse events occurred in 17.1% of AADvac1-treated and 24.1% of placebo-treated participants. Adverse events were seen in 84.6% of vaccine-treated and 81.0% of placebo-treated participants. Vaccine-treated participants had more injection-site reactions.

Six AADvac1-treated patients had confusion, mostly transient. “As confusion naturally occurs in patients with Alzheimer’s disease, a larger study is required to evaluate whether this is an adverse reaction or a chance observation; currently it constitutes a potential risk of moderate importance,” Novak and co-authors wrote.

For the whole study sample, the adjusted mean point difference on the Clinical Dementia Rating-Sum of Boxes (CDR-SB) scale was -0.36 (95% CI -1.31 to 0.59), with a custom cognitive battery adjusted mean z-score difference of 0.0008 (95% CI -0.17 to 0.17).

In a subgroup of 109 participants who most likely had Alzheimer’s pathology, those treated with AADvac1 had 27% slower cognitive decline measured by the CDR-SB (P=0.048) than placebo and 30% slower functional decline measured by the ADCS-MCI-ADL, an activities of daily living scale for people with mild cognitive impairment (P=0.039). This subgroup was identified using a multimodal classifier that combined structural MRI, demographics, and clinical data to find participants most likely positive for amyloid and tau pathology.

The researchers noted that the post hoc analyses come with important limitations: they were not pre-specified in the study protocol and not corrected for multiplicity testing. “These results must therefore be interpreted with caution and will require confirmation in future clinical development,” the team wrote.

“We plan on running a phase IIb study in patients with biomarker evidence of both tau and amyloid pathology to confirm the findings in a larger, well-defined sample,” Novak said. “Ideally, we’d initiate the study around the end of this year, or start of the next year.”

“Given that aducanumab was approved by the FDA based on a surrogate biomarker outcome using the accelerated approval pathway, should this study be positive, we could explore possibilities for an accelerated approval,” he added.