Literature Review

BACE-1 inhibition worsens cognition in patients with prodromal Alzheimer’s disease

View on the News

Failed trials suggest amyloid is not a practical therapeutic target

“Some trials fail because the experimental treatment proves to be no different than a control or standard intervention,” David Knopman, MD, wrote in an accompanying editorial (N Engl J Med. 2019 Apr 11;380:1476-8). “Others fail because of unacceptable side effects. In this issue of the Journal, an article by Egan et al. and a letter to the editor by Henley et al. (N Engl J Med. 2019 Apr 11;380:1483-5) describe a third reason for failure – a treatment worsens the target symptoms.

Certainly, beta-site amyloid precursor protein-cleaving enzyme 1 (BACE-1) inhibition makes sense when viewed in the light of the current understanding of Alzheimer’s disease neuropathology. The amyloid cascade hypothesis holds that toxic amyloid-beta fragments accumulate in the brain, form dense neuritic plaques, and lead to neuronal death and cognitive decline.

“The model is rooted in the inseparability of Alzheimer’s disease from abundant amyloid-beta pathologic features,” Dr. Knopman wrote. But, “Over the past 2 decades, the amyloid-beta–lowering strategy has been put to the test in trials of antiamyloid antibodies, none of which have been successful.”

Therefore, hitting amyloid at the source – the transmembrane cleavage domain – seemed important and, potentially, effective. But three BACE inhibitors (verubecestat, atabecestat, and lanabecestat) have shown similarly negative cognitive effects. “Together, these results suggest that preserved BACE-1 activity may be critical to normal synaptic functions. These observations place a limitation on how amyloid-beta lowering can be accomplished.”

It is possible that decreasing the level of BACE inhibition might ameliorate off-target effects and neuronal compromise but still be enough to reduce the generation of toxic amyloid-beta fragments, Dr. Knopman said. But, “Adjustments in the dose to a narrow window of BACE-1 inhibition would be difficult to accomplish in a clinical trial until there are peripheral biomarkers that reflect the activity of the agent in the brain.”

Thus far, most of the studied antiamyloid drugs have indeed reduced amyloid-beta levels, but none of those reductions affected cognition. A rethinking of amyloid-beta’s place in dementia progression may be in order.

“The dissociation between amyloid-beta lowering and cognitive benefits with both BACE-1 inhibition and antiamyloid antibody therapy is troubling. To be blunt, amyloid-beta lowering seems to be an ineffective approach, and it is time to focus on other targets to move therapeutics for Alzheimer’s disease forward.”

Dr. Knopman is a clinical neurologist and Alzheimer’s researcher at the Mayo Clinic, Rochester, Minn.



More bad news for Alzheimer’s research. Two more BACE inhibitors fall far short of the finish line.

Not only did both verubecestat and atabecestat fail to improve cognition and function patients with prodromal Alzheimer’s disease, they also made things worse. Declining cognitive scores, falls, suicidal ideation, and liver enzyme abnormalities were all seen in clinical trials.

The news doesn’t bode well for the therapeutic target of BACE (beta-site APP cleaving enzyme) inhibition. BACE is one of the enzymes that trims the amyloid precursor protein (APP). Inhibiting it does reduce the amount of toxic amyloid-beta in cerebrospinal fluid, and amyloid plaque in the brain. But none of these molecules has shown clinical benefit in dementia patients, whether their disease is mild, or moderate or – now – prodromal. And it is now apparent that inhibiting BACE also produces serious off-target problems.

“BACE-1 inhibition certainly seemed to have a sound rationale assuming the basis for amyloid’s role in Alzheimer’s disease pathogenesis is a gain of toxicity,” Richard J. Caselli, MD, of the Mayo Clinic, Rochester, Minn., said in an interview. “That APP is important for Alzheimer’s pathogenesis still seems clear but whether amyloid-beta toxicity is the driving force is no longer clear. Further, interruption of the APP system disrupts more than amyloid-beta peptide, possibly explaining the adverse cognitive effects of BACE-1 inhibition shown exhibited now by three different BACE-1 inhibitors.”


Researchers got their first dose of bad news regarding verubecestat at the 2017 Clinical Trials in Alzheimer’s Disease meeting. There, Michael F. Egan, MD, Merck’s associate vice president of clinical neuroscience, discussed the molecule’s failure to slow cognitive decline in patients with mild to moderate Alzheimer’s disease. There was plenty of biomarker evidence that the drug did block amyloid-beta production, but there also was a plethora of concerning adverse events, Dr. Egan said in an interview.

However, verubecestat still was being pursued in patients with prodromal Alzheimer’s disease. In February, Merck stopped the trial after a futility analysis and announced that the company was terminating studies of verubecestat in that population as well. In the April 11 issue of the New England Journal of Medicine, Dr. Egan and his colleagues report the full extent of verubecestat’s failure in prodromal patients, and the accompanying adverse events.

At the time of termination, 1,454 patients had been enrolled. Of these, 485 received 12 mg/day, 484 received 40 mg/day, and 485 received placebo. About half of each group completed 104 weeks of treatment in the study, which was designed to extend up to 5 years.

The primary outcome was change in the Clinical Dementia Rating Scale–Sum of Boxes score (CDR-SB). Seven secondary outcomes examined other cognitive and functional end points, along with changes in hippocampal volume on MRI and amyloid burden as determined in PET imaging.

Not only did verubecestat fail to slow cognitive decline, it appeared to exacerbate it. The mean change on the CDR-SB was 1.65 in the 12-mg group, 2.02 in the 40-mg group, and 1.58 in the placebo group, favoring placebo.

“In an exploratory analysis according to time point, scores on the CDR-SB were also higher [signifying more impairment of cognition and daily functioning] in the 40-mg group than in the placebo group at 13, 26, and 52 weeks ... suggesting but not confirming the possibility of worse performance at these earlier time points in the high-dose verubecestat group,” the investigators said.

Verubecestat also was associated with more conversions to Alzheimer’s disease. Per 100 patient-years, the Alzheimer’s disease event rates were 24.5 in the 12-mg group, 25.5 in the 40-mg group, and 19.3 in the placebo group. Compared with placebo, those taking 12-mg doses were 30% more likely to develop Alzheimer’s disease and those taking 40-mg doses were 38% more likely. The findings suggest that “verubecestat may have accelerated the progression to diagnosis of dementia due to Alzheimer’s disease,” the investigators said.

The negative impact of verubecestat was apparent quite early in the study. “In exploratory analyses, both dose levels of verubecestat were associated with poorer outcomes on the [Composite Cognition Score-3 Domain] and the ADAS-Cog [Alzheimer’s Disease Assessment Scale–Cognitive Subscale] measures of cognition that, relative to placebo, appeared worse at week 13 and did not appear to progress thereafter.”

Results of the secondary end points, including the ADAS-Cog and the Mini-Mental State Exam, also indicated that verubecestat may have worsened cognitive performance.

Imaging outcomes were positive, however. Hippocampal volume was 6,448 mL in the 12-mg group, 6,469 mL in the 40-mg group, and 6,435 mL in the placebo group. Brain amyloid increased in the placebo group, as expected, and decreased in the active groups. The small group of patients who underwent cerebrospinal fluid sampling showed reductions of more than 60% in amyloid-beta and soluble APP-beta associated with verubecestat. These results show that the molecule was indeed hitting its intended target, but that doing so was not clinically beneficial.

Adverse events were more common in the verubecestat groups. These included rash, dermatitis, urticaria, sleep disturbance, weight loss, and cough. Hair coloring changed in 2.5% of patients in the 12-mg group and 5% of the 40-mg group, but in none of the subjects taking placebo.

Patients taking verubecestat were more likely to sustain falls and injuries and to express suicidal ideation.

The results of this trial differ from the study of verubecestat in mild to moderate Alzheimer’s disease, the investigators noted. Those patients did not decline cognitively as did those with prodromal disease.

“Patients at an earlier stage of the disease may be more sensitive to the effects of substantial BACE-1 inhibition, perhaps because of a role of BACE-1 in normal synaptic function. It is also possible that the effects of verubecestat are due to inhibition of BACE-2,” they said.


Next Article:

Amyloid brain imaging changed clinical management in 60% of MCI and dementia patients

Related Articles