Eli Lilly and Company announced yesterday that they have terminated the Phase 2 trial of their BACE inhibitor LY2886721 due to liver abnormalities that showed up in four out of 45 patients during routine testing (see company press release). Communications manager Eva Groves said they do not believe the toxicity was related to the drug’s mechanism of action. Lilly does not currently have other BACE inhibitors in clinical development. They will further analyze the trial data before deciding on the next steps for the program, Groves said.

LY2886721 was selected for inclusion in the Dominantly Inherited Alzheimer Network (DIAN) prevention trial (see ARF related news story), but no patients had been dosed, Groves said. In Phase 1 trials, LY2886721 appeared safe and lowered Aβ42 in cerebrospinal fluid by more than two-thirds (see ARF related news story; ARF news story). In 2010, Lilly scuttled an earlier BACE inhibitor in Phase 1 testing due to off-target effects on the eye that appeared in animals given high doses (see ARF related news story).

Other scientists agreed that the liver toxicity likely represents off-target effects. Robert Vassar at Northwestern University, Chicago, Illinois, wrote to Alzforum, “I know of no evidence that would indicate liver toxicity as a mechanism-based side effect of BACE inhibition. For example, livers of BACE knockout mice are normal.” He added that any toxic molecule will prominently affect the liver. “This is a general challenge faced by all compounds in therapeutic development for all indications,” he said. Representatives from other pharmaceutical companies would not comment on whether their experimental BACE inhibitors have shown any liver toxicity in clinical trials. Last week, Satori Pharmaceuticals shut down, abandoning their development of γ-secretase modulators, which had been plagued by renal toxicity (see ARF related news story).––Madolyn Bowman Rogers.


  1. It is a pity that this compound has failed. BACE1 is currently one of the most promising AD drug targets and was also planned to be tested in the DIAN prevention trial. More BACE1 inhibitors need to be tested in the clinic—and, in fact, additional clinical studies are currently running. I doubt that other BACE1 inhibitors will show the same toxicity, but this has to be monitored. From all we know about BACE1 and BACE2 expression patterns, their substrates, and the phenotype of the corresponding knockout mice, I would not expect liver toxicity to be a mechanism-based side effect of BACE inhibitors. Potentially, the toxicity results from side effects of the compound or from unfavorable pharmacodynamics/pharmacokinetics, as has also been seen with other compounds in clinical trials for other medical conditions.

  2. Lilly’s BACE1 inhibitor trial is off-base but not off-target.

    The amyloid β-site cleaving enzyme (BACE1), which is the rate-limiting enzyme in the pathway that produces amyloid-β peptide (Aβ) from the Aβ precursor protein (APP) (Vassar et al., 1999), is considered a promising target for Alzheimer’s disease (AD) prevention or therapy. Thus, presuming the validity of the amyloid hypothesis, drug inhibition of BACE1 activity would appear to be an ideal anti-AD strategy.

    Unfortunately, a recent Phase 2 trial of the LY2886721 BACE1 inhibitor from Eli Lilly may have at least called this into question due to signs of liver toxicity in test subjects. Eli Lilly has stated that they believe this to be due to a secondary effect, not related to the drug’s mechanism of action. At first blush, this is a reasonable conclusion. After all, BACE1 knockout mice are viable and grow to apparently normal adulthood (Luo et al., 2001). However, deeper examination of BACE1 activity on substrates other than APP may be cause for additional caution.

    In addition to APP processing at the β-cleavage site, BACE1 is the major enzyme that cleaves β galactoside α2,6-sialyltransferase (ST6Gal I) in the liver (Kitazume et al., 2005; Kitazume et al., 2009). Cleavage and subsequent activation of ST6Gal I is necessary for glycoprotein sialylation in response to radiation stress. This protein modification is a necessary step in resistance to radiation-induced cellular damage. Elimination of glycoprotein sialylation results in greater radiation-induced cytotoxicity (Lee et al., 2008). Notably, BACE1 knockout mice have one-third as much plasma ST6Gal I than control mice (Kitazume et al., 2005).

    We propose a model based on these different threads of protein processing. In a normal organism, hepatic BACE1 activity forms a vital step in resistance to the sort of damage typical of ionizing radiation. This would include reactive oxidizing species (ROS), DNA damage, and other aberrations that are, as is general knowledge, also present as a consequence of aging. We propose that relatively sudden inhibition of BACE1 activity that is not restricted to the brain but is able to also reach the liver may result in apparent toxicity due to disruption of necessary anti-toxic and cellular repair activities of sialylated glycoproteins. In the case of knockout mice, two factors would interfere with detecting this risk. First, BACE1 knockout mice begin their lives without a functioning BACE1 gene. However, this does not prevent other enzymes that otherwise would be minor actors from compensating sufficiently to permit apparently normal life functions, albeit at reduced overall ST6Gal I cleavage.

    Second, a BACE1 knockout mouse generally does not live long enough to amass the sort of cumulative damage or age-related loss of protections. On the other hand, a middle-aged or older human very likely does not have sufficient compensating enzymes already in place and has already amassed oxidative and other age-related damage that might very well be only kept at bay by current levels of sialylated glycoproteins, levels that would be reduced if BACE1 activity is suddenly inhibited by drug treatment. Lilly did not publish tests of the effects of LY2886721 on ST6Gal I processing in animals or human plasma prior to clinical trials. We do not claim that this necessarily explains the setback in the LY2886721 trial, but it does suggest that it may be imprudent to dismiss, so confidently, primary drug activity out of hand as a problematic factor. Importantly, this model also suggests concrete steps by which it could potentially be excluded.

    See also:

    Rogers MB. Lilly Halts Phase 2 Trial of BACE Inhibitor Due to Liver Toxicity. Alzheimer Research Forum. 2013. Accessed 6/20/2013.


    . In vivo cleavage of alpha2,6-sialyltransferase by Alzheimer beta-secretase. J Biol Chem. 2005 Mar 4;280(9):8589-95. PubMed.

    . Molecular insights into beta-galactoside alpha2,6-sialyltransferase secretion in vivo. Glycobiology. 2009 May;19(5):479-87. PubMed.

    . Protein sialylation by sialyltransferase involves radiation resistance. Mol Cancer Res. 2008 Aug;6(8):1316-25. PubMed.

    . Mice deficient in BACE1, the Alzheimer's beta-secretase, have normal phenotype and abolished beta-amyloid generation. Nat Neurosci. 2001 Mar;4(3):231-2. PubMed.

    . Beta-secretase cleavage of Alzheimer's amyloid precursor protein by the transmembrane aspartic protease BACE. Science. 1999 Oct 22;286(5440):735-41. PubMed.

Make a Comment

To make a comment you must login or register.


News Citations

  1. DIAN Trial Picks Gantenerumab, Solanezumab, Maybe BACE Inhibitor
  2. Wave of New BACE Inhibitors Heading to Phase 2
  3. BACE Inhibitors Barrel Forward—Next Hurdles: Safety, Efficacy
  4. Barcelona: Out of Left Field—Hit to The Eye Kills BACE Inhibitor
  5. Satori Pharmaceuticals Shuts Down, Abandons γ-Secretase Modulators

External Citations

  1. company press release

Further Reading