Back to the Top


Name: Verubecestat
Synonyms: MK-8931, MK-8931-009 , BACE inhibitor
Therapy Type: Small Molecule (timeline)
Target Type: Amyloid-Related (timeline)
Condition(s): Alzheimer's Disease
U.S. FDA Status: Alzheimer's Disease (Discontinued)
Company: Merck


MK-8931 is a small-molecule inhibitor of BACE1 and BACE2. BACE1 is the β-secretase enzyme that cleaves the APP protein to release the C99 fragment of APP, which gives rise to various species of Aβ peptide during its subsequent cleavage by γ-secretase. The rationale of BACE inhibition is that it represents an upstream interference with the amyloid cascade, regardless of which species or aggregation states of Aβ then exert toxicity in the brain. BACE inhibition is sometimes envisioned as long-term maintenance therapy to limit Aβ production after an initial round of immunotherapy to remove existing amyloid deposits. 

MK-8931 was developed preclinically with extensive use of a translational rhesus monkey model, in which a catheter implanted into the cisterna magna at the base of the neck enabled repeated CSF sampling for long-term monitoring studies without ill effects on the animal (see Dec 2007 conference news; Aug 2006 conference news).

To support the clinical development of MK-8931, Merck was using the FDA-approved amyloid PET tracer flutemetamol and, in 2013, forged an agreement with Luminex Corporation to develop a companion diagnostic device to measure Aβ and tau concentration in the CSF.


Phase 1 included four public studies of a total of 68 healthy controls, patients with mild to moderate Alzheimer's disease, and people with renal insufficiency, held in Japan and the United States, to gather initial data on safety, tolerability, and pharmacology (e.g. Min et al., 2019). One trial studied how renal insufficiency—a common condition in the aged—would alter clearance of the drug and inform dosing in future trials. Phase 1 trials tested single doses up to 450 mg and multiple doses from 12 to 150 mg/day. At the 2012 AAIC conference in Vancouver, Canada, MK-8931 was reported to have been generally safe, without discontinuations due to side effects, and to have reduced CSF Aβ concentration in AD patients.

Two Phase 1/2 dose-ranging trials further evaluated the tolerability and pharmacology of single and multiple doses, respectively, in 88 healthy adults. These results, too, were presented at the same conference to show good tolerability without withdrawals due to side effects, with dose-proportional increases in plasma and CSF exposure of the drug, and dose-dependent reduction in Aβ40 across the 2.5 to 550 mg/day administered to study volunteers. These studies used repeated CSF sampling, which found that CSF Aβ was reduced by up to 90 percent (see Jul 2012 conference news). An additional study confirmed similar PK and tolerability in healthy elderly volunteers (Forman et al., 2019).

In November 2012, Merck started EPOCH, an 18-month Phase 2/3 trial comparing 12, 40, or 60 mg/day of MK-8931 given as once-daily tablets to placebo in people with mild to moderate AD. EPOCH started out treating 200 people in Phase 2 and, after an interim safety analysis, expanded to Phase 3 with a total of 2,221 participants. This trial included conventional cognitive and functional primary outcomes, as well as substudies for biomarker outcomes indicating changes in brain amyloid, CSF tau levels, and brain volume. In response to questions about why this drug was being tested in mild to moderate AD rather than earlier-stage disease, Merck’s Johan Luthman said that the overall Phase 3 program included plans to test the drug across all disease stages, starting with mild to moderate, and that a trial in prodromal AD was planned (see Dec 2012 news). In December 2013, Merck announced that EPOCH had passed an interim safety evaluation and was proceeding to full enrollment and Phase 3. By October 2016 the trial was fully enrolled; data collection for the primary outcome was expected to wrap up in summer 2017. However, on 14 February 2017, Merck announced a premature end to this trial following an interim analysis (see Feb 2017 news). Results were presented at CTAD later that year (see Dec 2017 conference news), and formally published (Egan et al., 2018). A subsequent paper further detailed the safety data, noting particularly that while psychiatric side effects did not get worse over time, falls and injuries did (Egan et al., 2019).

In November 2013, Merck began the APECS trial in 1,500 participants with prodromal AD, aka mild cognitive impairment due to AD (aMCI). These patients have measurable cognitive deficits and a positive PET scan with the newly FDA-approved amyloid tracer flutemetamol, but are not functionally impaired. APECS compared 12 and 40 mg once-daily doses to placebo; treatment was to last for two years. APECS used change from baseline on the Clinical Dementia Rating Sum of Boxes (CDR-SB), a continuous measure, as its primary outcome. Secondary outcomes evaluated a range of newer measures, including a cognitive composite, CSF tau, brain imaging of hippocampal volume and amyloid load, and others. This trial was being conducted in more than 90 locations worldwide; it completed enrollment in November 2016 and was expected to complete data collection for its primary outcome in 2019. 

In October 2016, Merck started an additional Phase 1 study in the United States to compare the liver metabolism of verubecestat in 32 people with hepatic insufficiency to people with normal liver function. In November 2016, Merck published data on verubecestat's discovery, in vitro characteristics, activity, and safety profile in rats and monkeys, as well as on pharmacokinetic modeling used for dose finding, and initial Phase 1 data for exposure to drug of up to one week (see Nov 2016 news on Kennedy et al., 2016).

In February 2018, APECS was discontinued (see Feb 2018 news) and Merck no longer listed verubecestat in its research pipeline. At the CTAD conference later that year, Merck reported that APECS participants on 40 mg verubecestat scored worse than the placebo group on the CDR-SB and ADAS-Cog13 starting at 13 weeks. The effect was small, with a Cohen’s d of 0.2, and did not progress over time. The 12 mg treatment group also performed slightly worse than controls, with the difference reaching significance at scattered time points. Both treatment groups scored worse than the placebo group on a functional measure, the ADCS-ADL, and reported more anxiety, depression, and sleep problems than controls (Nov 2018 conference news). The data have been published (Egan et al., 2019).

Merck and other companies continue to publish further analyses from BACE inhibitor trials (Dec 2020 news). Among the findings: In APECS, worsening on many secondary and exploratory cognitive tests was accompanied by improvement in verbal fluency (Wessels et al., 2020). In both EPOCH and APECS, verubecestat treatment came with more hippocampal and whole brain atrophy than placebo (see Dec 2019 news and Sur et al., 2020). The volume loss appeared by 13 weeks but did not progress. It occurred predominantly in amyloid-rich regions, but did not correlate with changes in amyloid load, cognitive decline, or biomarkers of neurodegeneration.

In EPOCH, another study found that verubecestat did not affect biomarkers of retinal degeneration, which did not track with changes in brain volume (Sergott et al., 2021).

For all trials on this drug, see

Clinical Trial Timeline

  • Phase 1
  • Phase 2/3
  • Phase 3
  • Study completed / Planned end date
  • Planned end date unavailable
  • Study aborted
Sponsor Clinical Trial 2009 2010 2011 2012 2013 2014 2015 2016 2017 2018 2019 2020 2021 2022 2023 2024 2025 2026 2027 2028 2029 2030 2031 2032 2033 2034
Merck NCT01496170
Merck NCT01537757
Merck NCT01739348
Merck NCT01953601

Last Updated: 21 Jan 2021


  1. It is time for the Alzheimer’s field to abandon the "amyloid hypothesis" as a rigid model for developing treatment. The problem is that amyloid is not related to dementia—not temporally, pathologically, or therapeutically. Amyloid deposition (highly associated with ApoE genotype and all forms of early Alzheimer's disease) begins decades before dementia and some individuals with substantial amyloid burdens have no dementia. Further, the deposition of amyloid does not occur predominantly in brain areas associated with the clinical picture of dementia, and no amyloid measures are closely correlated with dementia. Beyond this, many recent clinical trials have shown that removing amyloid does not have a substantial impact on the disease. The issue is that there is a pathological process that causes dementia and a side-stream part of that process results in amyloid deposition. Examining the first genetic factor associated with Alzheimer's disease, Down’s syndrome, the problem is clear, there is a 50 percent excess production of the amyloid protein precursor (APP). So, the key for Down’s syndrome is to cut down the production of APP. But how APP is managed otherwise is the critical question. However, the amyloid deposition itself is basically a harmless scar with respect to the Alzheimer-associated dementia.

    Attention to the roles of β- and γ-secretases is important, but these proteins are important normal enzymes in the brain. A critical perspective is that these enzymes play a central role in neuroplasticity through their actions on APP, and inhibiting these enzymes should not be expected to have a beneficial effect clinically. If regulation of APP is critical, the augmentation of α-secretase should be the top priority. And there are several approaches that may stimulate this enzyme, which are relevant to Alzheimer' disease, including increasing cholinergic activity with cholinesterase inhibitors, one of only two beneficial drug strategies found for Alzheimer's disease. But inhibition of BACE or γ-secretase will decrease the normal non-α catabolism of APP and likely produce serious problems without benefiting Alzheimer's disease.

    There are several early genetic variants of Alzheimer's disease, many involving presenilins, which affect γ-secretase activity. Every time that a γ-cleavage takes place on the APP β product, two complementary proteins are created: the amyloid, which likely plays a role in the removal of old synapses, and the APP intracellular domain (AICD). The AICD is in a better position inside the cell to initiate the next pathological step in the cascade of processes that lead to dementia because AICD appears to have a role in phosphorylating tau, normally to cause synapse retraction. Still, it is unclear if the relevance of γ-cleavage is overactivity of γ-secretase or an aberrant cleavage of the APP β product. The production of Aβ42 is considered more associated with Alzheimer's disease, but the AICD-42 has been hard to study, even though it is a more likely culprit in the cascade leading to dementia. Important considerations are whether APOE genotype, statins, or NSAIDs, which all have major relationships with Alzheimer’s risk, are all working at this exact point (affecting membrane thickness and the exact site of the γ-cleavage on the APP β product).

    Note that the same concerns about the role of plaques in Alzheimer's disease are also true for the neurofibrillary tangles. This pathological hallmark of Alzheimer's disease has a closer association with the dementia than plaques, but the tangles themselves are also just the scars resulting from the core processes that lead to dementia. There is no reason to think that removal of hyper-phosphorylated tau deposits would help prevent or reverse dementia. Before any treatment endeavors are made there needs to be a much more complete understanding of the Alzheimer’s pathological process.

Make a Comment

To make a comment you must login or register.


News Citations

  1. Wave of New BACE Inhibitors Heading to Phase 2
  2. Q&A With Merck’s Johan Luthman
  3. Merck Pulls Plug on Phase 2/3 BACE Inhibitor Trial
  4. Verubecestat Negative Trial Data: What Does it Mean for BACE Inhibition?
  5. Paper Alert: Verubecestat Preclinical and Phase 1 Data Published
  6. Merck Axes Verubecestat for Prodromal AD, Researchers Say ‘Go Earlier’
  7. Bump in the Road or Disaster? BACE Inhibitors Worsen Cognition
  8. New Data from Past BACE Inhibitor Trials Shed Light on Side Effects
  9. Picking Through the Rubble, Field Tries to Salvage BACE Inhibitors
  10. San Diego: Merck Reports BACE Inhibition in Primates
  11. Madrid: γ-secretase Dimers, A New Model, A Drug in Clinic

Paper Citations

  1. . Pharmacokinetics and Pharmacodynamics of the BACE1 Inhibitor Verubecestat (MK-8931) in Healthy Japanese Adults: A Randomized, Placebo-Controlled Study. Clin Pharmacol Ther. 2019 May;105(5):1234-1243. Epub 2019 Jan 18 PubMed.
  2. . Safety, Tolerability, and Pharmacokinetics of the β-Site Amyloid Precursor Protein-Cleaving Enzyme 1 Inhibitor Verubecestat (MK-8931) in Healthy Elderly Male and Female Subjects. Clin Transl Sci. 2019 Sep;12(5):545-555. Epub 2019 Jun 19 PubMed.
  3. . Randomized Trial of Verubecestat for Mild-to-Moderate Alzheimer's Disease. N Engl J Med. 2018 May 3;378(18):1691-1703. PubMed.
  4. . Further analyses of the safety of verubecestat in the phase 3 EPOCH trial of mild-to-moderate Alzheimer's disease. Alzheimers Res Ther. 2019 Aug 7;11(1):68. PubMed.
  5. . The BACE1 inhibitor verubecestat (MK-8931) reduces CNS β-amyloid in animal models and in Alzheimer's disease patients. Sci Transl Med. 2016 Nov 2;8(363):363ra150. PubMed.
  6. . Randomized Trial of Verubecestat for Prodromal Alzheimer's Disease. N Engl J Med. 2019 Apr 11;380(15):1408-1420. PubMed.
  7. . Cognitive outcomes in trials of two BACE inhibitors in Alzheimer's disease. Alzheimers Dement. 2020 Nov;16(11):1483-1492. Epub 2020 Oct 13 PubMed.
  8. . BACE inhibition causes rapid, regional, and non-progressive volume reduction in Alzheimer's disease brain. Brain. 2020 Dec 1;143(12):3816-3826. PubMed.
  9. . Retinal Optical Coherence Tomography Metrics Are Unchanged in Verubecestat Alzheimer's Disease Clinical Trial but Correlate with Baseline Regional Brain Atrophy. J Alzheimers Dis. 2021;79(1):275-287. PubMed.

External Citations

  1. research pipeline

Further Reading


  1. . The BACE1 inhibitor verubecestat (MK-8931) reduces CNS β-amyloid in animal models and in Alzheimer's disease patients. Sci Transl Med. 2016 Nov 2;8(363):363ra150. PubMed.
  2. . Synthesis of Verubecestat, a BACE1 Inhibitor for the Treatment of Alzheimer's Disease. Org Lett. 2016 Nov 18;18(22):5780-5783. Epub 2016 Nov 4 PubMed.
  3. . Discovery of the 3-Imino-1,2,4-thiadiazinane 1,1-Dioxide Derivative Verubecestat (MK-8931)-A β-Site Amyloid Precursor Protein Cleaving Enzyme 1 Inhibitor for the Treatment of Alzheimer's Disease. J Med Chem. 2016 Dec 8;59(23):10435-10450. Epub 2016 Nov 18 PubMed.
  4. . Future Therapeutics in Alzheimer's Disease: Development Status of BACE Inhibitors. BioDrugs. 2016 Jun;30(3):173-94. PubMed.
  5. . Pharmacokinetics and Pharmacodynamics of the BACE1 Inhibitor Verubecestat (MK-8931) in Healthy Japanese Adults: A Randomized, Placebo-Controlled Study. Clin Pharmacol Ther. 2019 May;105(5):1234-1243. Epub 2019 Jan 18 PubMed.
  6. . Chronic Verubecestat Treatment Suppresses Amyloid Accumulation in Advanced Aged Tg2576-AβPPswe Mice Without Inducing Microhemorrhage. J Alzheimers Dis. 2017;59(4):1393-1413. PubMed.
  7. . BACE-1 Inhibitors: From Recent Single-Target Molecules to Multitarget Compounds for Alzheimer's Disease. J Med Chem. 2017 Aug 8; PubMed.
  8. . Verubecestat for Prodromal Alzheimer's Disease. N Engl J Med. 2019 Jul 25;381(4):388. PubMed.
  9. . Verubecestat for Prodromal Alzheimer's Disease. Reply. N Engl J Med. 2019 Jul 25;381(4):389. PubMed.
  10. . Lessons that can be learnt from the failure of verubecestat in Alzheimer's disease. Expert Opin Pharmacother. 2019 Dec;20(17):2095-2099. Epub 2019 Aug 18 PubMed.
  11. . BACE inhibitors in clinical development for the treatment of Alzheimer's disease. Expert Rev Neurother. 2018 Nov;18(11):847-857. Epub 2018 Oct 24 PubMed.
  12. . Partial reduction of amyloid β production by β-secretase inhibitors does not decrease synaptic transmission. Alzheimers Res Ther. 2020 May 26;12(1):63. PubMed.