And another one bites the dust. Yesterday, Janssen, the pharmaceutical division of Johnson & Johnson, announced that it would end clinical development of the BACE inhibitor atabecestat. Serious elevations in liver enzymes were detected in some patients taking the drug. “Janssen has concluded that the benefit-risk ratio is no longer favorable to continue development of atabecestat for people who have late-onset preclinical stage Alzheimer’s disease,” according to a company press release

Janssen has stopped screening, randomization, and dosing in the Phase 2/3 EARLY trial and also in a Phase 2 long-term safety study. More than 600 people in these trials have received either drug or placebo. Janssen will update the study protocol to monitor patients after they discontinue treatment.

“Certainly, this is another disappointment,” said Paul Aisen, University of Southern California, San Diego. “The EARLY trial was a very exciting project and it’s disappointing that it will not continue to its conclusion.” Aisen is one of the clinical leaders of the trial.

Aisen does not see this latest development as an indictment of the study design, or of BACE inhibitors as a class. He noted there are other BACE inhibitors being tested that seem to be free of hepatotoxicity. “We need to continue to advance early intervention and early anti-amyloid therapies, including anti-BACE approaches, in sporadic and in familial AD, and I don’t think this announcement changes that,” he said.

This news comes three months after Merck stopped development of its BACE inhibitor, verubecestat, for prodromal AD, citing lack of efficacy (Feb 2018 news). There was no indication of serious liver toxicity from verubecestat. However, in 2013 Eli Lilly halted development of its BACE inhibitor, LY2886721, because of liver abnormalities. Aisen acknowledged that hepatotoxicity has been seen with some, but not all, drugs in this class. “It occurs with a number of drugs, including some already on the market,” he said.

Whether liver toxicity turns out to be a function of BACE inhibition or an off-target effect remains to be seen. Novartis, AstraZeneca/Eli Lilly, and Biogen/Eisai also have BACE inhibitors in clinical trials.—Tom Fagan


  1. While last week’s announcement is disappointing, we believe that BACE inhibitors still hold promise in stopping the accumulation of amyloid and preventing the clinical onset of Alzheimer’s disease—particularly in healthy people who are at increased risk of the disease but do not yet have extensive amyloid accumulation.

    Since Merck’s recently discontinued drug, verubecestat, was not associated with adverse liver effects, the side effect observed with atabacestat may not be a general property of all BACE inhibitors. At this time, we have no reason to believe the reported safety concerns will occur with CNP520, which is currently being tested as part of the Alzheimer’s Prevention Initiative Generation Program. Since our review and selection of CNP520 for the API Generation Study 1, we have been encouraged about its pharmacodynamic effects, toxicology findings, and the reported initial safety data.

    We look forward to continuing the Generation Program trials and remain excited about the chance to give CNP520 a better chance to demonstrate an effect in unimpaired persons at AD risk. As in all trials, our participants will be continuously monitored for any unexpected risks. We remain firmly committed to finding ways to prevent the clinical onset of Alzheimer’s disease, and we remain excited about the chance to put this promising BACE inhibitor to the test.

  2. A BACE1 inhibitor failure is an opportunity for selective BACE1 inhibition.

    Janssen recently announced that it terminated atabecestat development. The announcement further stated that serious elevations in liver enzymes were detected in some patients taking the drug. This is not the first failure of a BACE1 inhibitor in clinical trials, nor is the first connected with changes in liver function. Multiple other BACE1 inhibitors are still in the pipeline, none of which have any reported liver anomalies, and we can still hope that at least one of them will turn out to be viable. Alzforum raised the possibility that BACE1 modification may, in and of itself, require caution because altering liver BACE1 activity may have its own unwanted consequences. We have also raised this possibility in the past (Lahiri et al., 2014; Feb 2017 comment). We propose trying a selective BACE1 inhibition to tackle the on-target, off-site (OnTOS) issue.

    As is already known, BACE1 is active in the liver as the primary cleavage enzyme for β-galactoside α2,6-sialyltransferase 1 (ST6Gal1), and its cleavage is necessary for the sialyltransferase to be secreted. ST6Gal1 is vital in response to radiation, oxidative, and mechanical damage to the liver and other tissues (Kitazume et al., 2009). A ST6Gal1 SNP (rs10937275) associates with drug-induced liver toxicity (Aithal and Grove, 2015), and hepatotoxic plant extracts downregulate expression of ST6Gal1 when fed to rats (Guo et al., 2007). Disruption of BACE1 and ST6GalI by tumor necrosis factor-α alter tight junctions (Kawata et al., 2017). 

    In essence, BACE1 inhibition could be perfectly on-target from a molecular standpoint but still be toxic in the long run if this on-target activity occurs “off-site,” such as in liver instead of the intended on-site brain. On-target, off-site (OnTOS) activity could explain the specific liver-related difficulties of BACE1 inhibition as a therapeutic strategy.

    “Body-wide” alteration of BACE1 levels may be a suboptimal strategy for treating AD, and it may be worthwhile to target more tissue or organ-specific treatments. For example, levels of specific miRNA species can differ broadly by organ, including those that regulate BACE1 in brain versus liver. If a given BACE1-regulating miRNA has low levels in the brain, when associated with AD, but high levels in liver, then effects in the liver will be essentially buffered by higher background. The activity of miRNA adheres to Michaelis-Menten kinetics (Chen et al., 2015), which means that “total” effects of exogenous miRNA plus background levels can quickly reach saturation, and “full effect” of miRNA-based therapies would only appear in specifically deficient tissues.

    While we have every hope for the current crop of anti-BACE1 therapies to be successful, the emerging complexity of modifying the enzyme’s metabolism suggests that it would be worthwhile to look into alternative pathways. Indeed, given that AD is a multifaceted and complex disorder (Maloney et al., 2016), perhaps it may be more fruitful to investigate attacking more than one pathway at a time for a given patient, much like it is currently acceptable to other diseases.


    . Lessons from a BACE1 inhibitor trial: off-site but not off base. Alzheimers Dement. 2014 Oct;10(5 Suppl):S411-9. Epub 2014 Feb 12 PubMed.

    . Genome-Wide Association Studies in Drug-Induced Liver Injury: Step Change in Understanding the Pathogenesis. Semin Liver Dis. 2015 Nov;35(4):421-31. Epub 2015 Dec 16 PubMed.

    . Comparison of gene expression profiles altered by comfrey and riddelliine in rat liver. BMC Bioinformatics. 2007 Nov 1;8 Suppl 7:S22. PubMed.

    . Ablation of neuropsin-neuregulin 1 signaling imbalances ErbB4 inhibitory networks and disrupts hippocampal gamma oscillation. Transl Psychiatry. 2017 Mar 7;7(3):e1052. PubMed.

    . Properties and kinetics of microRNA regulation through canonical seed sites. J RNAi Gene Silencing. 2015;11:507-14. Epub 2015 Feb 16 PubMed.

    . Epigenetics of dementia: understanding the disease as a transformation rather than a state. Lancet Neurol. 2016 Jun;15(7):760-74. Epub 2016 May 9 PubMed.

Make a Comment

To make a comment you must login or register.


Therapeutics Citations

  1. Atabecestat
  2. Verubecestat
  3. LY2886721

News Citations

  1. Merck Axes Verubecestat for Prodromal AD, Researchers Say ‘Go Earlier’

External Citations

  1. company press release
  2. Phase 2/3 EARLY trial
  3. Phase 2 long-term safety study

Further Reading