The Alzheimer’s field begins 2021 with a glimmer of good news: Donanemab, a monoclonal antibody that recognizes a pyroglutamated form of Aβ, slowed cognitive decline in people with early AD. That was the upshot of the topline results of the Phase 2 TRAILBLAZER-ALZ study, announced by Eli Lilly & Company on January 10. In addition to stemming cognitive decline, the monoclonal also wiped out Aβ plaques, lowering them into the range seen in healthy volunteers.

  • Donanemab slowed decline on the Integrated Alzheimer's Disease Rating Scale by a third.
  • The drug reduced Aβ plaques into the normal range.
  • A larger Phase 2 trial is ongoing.

“It’s a good day for the field,” said Eric Siemers of Siemers Integration, LLC, who worked at Lilly when the trial began but has since retired. “We now have one more monoclonal antibody signaling in the right direction.”

Cynthia Lemere of Brigham and Women’s Hospital in Boston welcomed the news. “Lilly’s Phase 2 positive findings lend further support to the amyloid hypothesis and validate Aβ as a therapeutic target for early-stage AD,” she wrote (full comment below).

Donanemab is unique among the cadre of anti-Aβ monoclonal antibodies developed so far. Rather than latching onto a specific amino acid epitope within the original peptide, donanemab recognizes Aβ(p3-42), an enzymatically modified form of Aβ adorned with pyroglutamate at its N-terminus. This form of Aβ is only found within amyloid plaques (Dec 2012 newsJul 2014 news). 

In a Phase 1b trial, the antibody had demonstrated prowess at plaque removal (Aug 2018 conference news; Dec 2019 conference news). This spurred Lilly to test whether banishing Aβ plaques would also slow cognitive decline in people with AD.

In 2017, the Phase 2 TRAILBLAZER-ALZ began enrolling people in the early symptomatic phase of AD. Participants had to score between 20 and 28 on the mini-mental state examination, and have had memory loss stretching back at least six months. The trial enrolled participants with evidence of amyloid plaques, whose neurofibrillary tangle accumulation fell within a low-to-medium range, defined by an SUVR of 1.15 to 1.36 by flortaucipir PET. Mark Mintun of Eli Lilly told Alzforum that using tau PET in this way helped select participants who were in the process of decline, but at an early enough stage to make them still likely to benefit from this treatment. The trial was initially designed as a combination study, testing the effectiveness of donanemab alone or in combination with Lilly’s BACE1 inhibitor, LY3202626, but in 2018 Lilly discontinued this latter portion.

The topline results disclosed yesterday are based on data from 272 participants randomized 1:1 to receive placebo or donanemab for 76 weeks. People in the treatment arm received 700 mg of donanemab once a month for the first three doses, then 1400 mg monthly thereafter. Notably, participants were switched to placebo once their plaque load dipped into the normal range, i.e. falling below 25 centiloids for two consecutive measures or below 11 centiloids at any one measure.

The trial met its primary endpoint, slowing decline on the Integrated Alzheimer’s Disease Rating Scale in the donanemab group by 32 percent relative to placebo. Developed by scientists at Eli Lilly, the iADRS tests both cognition and function, and is essentially a combination of the ADAS-Cog13 and the ADCS-iADL (Wessels et al., 2015). The company disclosed no numeric results of secondary clinical outcomes, but did claim that participants in the donanemab group trended better than those on placebo on every secondary measure.

Confirming findings from earlier trials, donanemab cleared amyloid plaques. In the treatment group, amyloid plummeted by an average of 84 centiloids, from 108 to 24. Anything less than 25 centiloids is considered within the normal range of amyloid accumulation. Lilly scientists did not say how many participants dropped below this threshold and were therefore switched to placebo.

Consistent with the Phase 1 data, donanemab did trigger amyloid-related imaging abnormalities—edema (ARIA-E), which occurred in 27 percent of treated participants, including 6 percent in whom it was symptomatic.

Lilly is currently enrolling participants into an open-label extension, which will continue to study safety, cognition, Aβ and tau deposition, while also validating the accuracy of home video-based cognitive assessments.

A larger placebo-controlled Phase 2 study, TRAILBLAZER-ALZ-2, began in June of last year. The 18-month trial will enroll 500 participants with early AD but, unlike its predecessor, will not exclude participants with high levels of neurofibrillary tangle accumulation. Mintun told Alzforum that the boost in statistical power gained from the larger trial allows the inclusion of participants with more advanced disease.

The larger trial will also move participants to placebo once their amyloid drops into the normal range. Mintun said that because donanemab’s target—pyroglutamated Aβ—resides specifically within Aβ plaques, there is no need to continue dosing once that target is gone. What about if amyloid creeps back up? Mintun said that given the slow, decades-long process of amyloid accumulation that occurs during the preclinical phase of AD, this is unlikely to happen within the course of an 18-month trial. Therefore, there are no plans, either within the placebo-controlled portion of the trial or in an extension phase, to switch plaque-cleared participants back to donanemab.

How does this compare to BAN2401 aka lecanemab—an antibody that recognizes Aβ protofibrils found in both plaques and oligomers? In a large Phase 2 trial, lecanemab also virtually wiped out plaques (Jul 2018 conference news). However, participants were off drug for anywhere between nine and 60 months before starting an open-label extension study. During that time, plaques remained at bay but cognition started to slide, suggesting that even in the absence of full-blown plaques, continued dosing of BAN2401 was necessary to maintain cognition (Dec 2019 conference news). This could be due to lecanemab’s binding to toxic oligomers, scientists have proposed. Unlike lecanemab, donanemab is not thought to bind oligomers. However, Siemers suggested donanemab may indirectly get rid of other toxic Aβ species by removing their reservoir, i.e. plaques. Whether draining this Aβ swamp is sufficient to hold off cognitive decline for years to come remains to be seen.

In its press release, Lilly referred to the second donanemab Phase 2 trial as “pivotal” (see press release). Does that imply positive results could be used to apply for regulatory approval, sans Phase 3? Mintun said both TRAILBLAZER-ALZ trials were designed as registration-level studies. “We believe that if both studies reinforce each other and describe a clearly efficacious drug, that would be compelling evidence,” Mintun said. Lilly will discuss data available thus far with regulators, including the FDA, to determine next steps.

Donanemab’s development path is slim compared to what Lilly did with solanezumab, an anti-Aβ antibody binding mostly Aβ monomers. Lilly ran three large Phase 3 trials of that antibody. If donanemab’s 32 percent slowing of decline sounds familiar, Lilly’s solanezumab ordeal may be the reason. The first two trials—Expedition 1 and 2—tested solanezumab in people with mild to moderate AD, without PET to select amyloid-positive participants. Both trials missed their primary endpoint, but a secondary analysis of participants with mild AD found solanezumab had slowed their progression by 34 percent (Siemers et al., 2015; Aug 2015 news). Lilly then ran a third Phase 3 trial—Expedition 3—that included only people with mild AD who had amyloid accumulation. In a blow, that trial missed its primary endpoint, with solanezumab slowing decline on the ADAS-Cog by only 11 percent, which was not statistically significant (Dec 2016 conference newsJan 2018 news). 

Mintun said that despite the glimmer of hope offered by the post hoc analyses of Expedition 1 and 2, all three Expedition trials fell short of their primary endpoints. In this initial Phase 2 trial, donanemab met the primary endpoint.—Jessica Shugart


  1. I was excited to read Eli Lilly’s press release today describing these positive Phase 2 results. Pyroglutamate-3 Aβ, an N-terminally truncated and modified form, has been shown to act as a seed for amyloid aggregation and is neurotoxic. Removal of this specific pathological Aβ species via immunotherapy has been shown to be protective against AD in non-clinical studies by Eli Lilly and my lab in collaboration with Vivoryon Therapeutics (formerly Probiodrug AG). Lilly’s Phase 2 positive findings lend further support to the amyloid hypothesis and validate Aβ as a therapeutic target for early stage AD.

  2. Eli Lilly’s announcement is exciting. It clearly supports our hypothesis of 1999, that pyroglutamate-3 Aβ is one of the most neurotoxic and Aβ-seeding species, not simply a "bystander" in Aβ deposits. The results prove our concept that the enzyme-driven formation of N-terminal pyroglutamate is one major driver of neurodegeneration in the early Alzheimer’s syndrome.

    It would be interesting to test the combination of such a specific antibody with a QC inhibitor; that would fight the genesis of pyroglutamate-3 Aβ by glutaminyl cyclase as well as trap already-formed material by antibodies.

    Some of the papers we published in that respect are Morawski et al., 2014; Frost et al., 2015; and Grochowska et al., 2017.


    . Glutaminyl cyclase in human cortex: correlation with (pGlu)-amyloid-β load and cognitive decline in Alzheimer's disease. J Alzheimers Dis. 2014;39(2):385-400. PubMed.

    . An anti-pyroglutamate-3 Aβ vaccine reduces plaques and improves cognition in APPswe/PS1ΔE9 mice. Neurobiol Aging. 2015 Dec;36(12):3187-99. Epub 2015 Aug 31 PubMed.

    . Posttranslational modification impact on the mechanism by which amyloid-β induces synaptic dysfunction. EMBO Rep. 2017 Jun;18(6):962-981. Epub 2017 Apr 18 PubMed.

  3. I believe the key element for all three antibodies that reduce PET amyloid is that they evade being bound up by monomers and can get sufficient antibody onto the plaques. This is true for BAN2401/lecanemab, aducanumab, and now donanemab. Other Aβ monomer-binding antibodies become absorbed by the monomeric decoys and have little antibody remaining to bind to the fibrillar plaques and clear amyloid except at highest concentrations. Recall that only 1/1,000th of the injected dose typically accesses the nervous system. Binding monomer in blood may reduce this further. If, as in earlier reports, the half-life of donanemab is short due to anti-idiotype responses, it suggests that antibody treatment may work even if intermittent. I look forward to seeing a peer-reviewed report on these data.

  4. We were pleasantly surprised by Lilly’s announcement of the positive Phase 2 results. Although we have to wait for the full disclosure of all the data, it seems from the press release that the design of the trial ticked all the boxes of a proper Phase 2 study (Cummings, 2019) and that the combined results of positive biomarker findings and the clinical endpoint warrants a larger follow-up study. We believe that these results provide further validation for Aβ as a therapeutic target for early stage AD.

    We offer two observations. First, the team seems to have made the right decision to have chosen a combined functional and cognitive outcome, using the iADRS (Wessels et al., 2015), which shows the power of such a combination, since both of its elements, ADAS-COG13 and ADCS-ADL, by themselves as secondary outcomes were not positive. Since the population was very early stage AD this is not surprising, as both these scales lack sensitivity to change in patients with MMSE above 20. For this population, the CFC has been designed, combining the Amsterdam-IADL with episodic memory ADAS-cog elements, combined with NTB and other cognitive measures of attention, working memory, and executive function (Jutten et al., 2020). It shows high sensitivity to change and is currently used in several Phase 2 programs in early AD.

    Second, the target of donanemab is pyroglutamate-3 Aβ, a very toxic subspecies of Aβ, which is found predominantly in plaques but also in oligomeric form in equilibrium with the plaque, and is formed through the enzymatic action of glutaminyl cyclase (QC). This post-translationally modified neuro- and synaptotoxic form of Aβ is not found in healthy individuals, but in individuals with pathologically defined preclinical and clinical AD (Thal et al., 2015). Pyroglutamate-3 Aβ can act as a seed for toxic oligomers, fibrils and plaques with mixed Aβ peptide composition and correlates with QC expression in AD patients.

    As Hans Ulrich Demuth describes in his comment, his work, and that of others, led to the discovery of PQ192 (varoglutamstat), a potent small-molecule inhibitor of QC. In a Phase 2a study, we showed (Scheltens et al., 2018) that varoglutamstat had proven target engagement, lowering QC in CSF up to 90 percent, and had beneficial effects on some cognitive tests, EEG markers of synaptic activity, as well as CSF neurogranin and YKL 40, without the side effects associated with a monoclonal antibody. This compound is now studied in a larger Phase 2B study in Europe (Viviad NCT04498650) and soon also in an NIA-sponsored study in the USA (VIVA-MIND NCT03919162) led by Howard Feldman. Earliest readouts will be available in 2023.

    Once positive, it will be extremely interesting to combine the small molecule varoglutamstat with the monoclonal antibody donanemab, perhaps in an alternating fashion once amyloid clearance on PET has been achieved.

    Disclosures: Philip Scheltens serves as principal investigator for Vivoryon’s European clinical trials, and has worked since January 2021 for the life sciences investment firm LSP. LSP is an investor in Vivoryon. Arno de Wilde works for the life sciences investment firm LSP. LSP is an investor in Vivoryon.


    . The "rights" of precision drug development for Alzheimer's disease. Alzheimers Res Ther. 2019 Aug 31;11(1):76. PubMed.

    . A Combined Measure of Cognition and Function for Clinical Trials: The Integrated Alzheimer's Disease Rating Scale (iADRS). J Prev Alzheimers Dis. 2015 Dec 1;2(4):227-241. PubMed.

    . The Cognitive-Functional Composite is sensitive to clinical progression in early dementia: Longitudinal findings from the Catch-Cog study cohort. Alzheimers Dement (N Y). 2020;6(1):e12020. Epub 2020 Apr 17 PubMed.

    . Neuropathology and biochemistry of Aβ and its aggregates in Alzheimer's disease. Acta Neuropathol. 2015 Feb;129(2):167-82. Epub 2014 Dec 23 PubMed.

    . Safety, tolerability and efficacy of the glutaminyl cyclase inhibitor PQ912 in Alzheimer's disease: results of a randomized, double-blind, placebo-controlled phase 2a study. Alzheimers Res Ther. 2018 Oct 12;10(1):107. PubMed.

Make a Comment

To make a comment you must login or register.


Therapeutics Citations

  1. Donanemab
  2. LY3202626

News Citations

  1. Could Antibodies Against Pyroglutamate Safely Break Down Plaques?
  2. Anti-Amyloid Therapies Combine Forces to Knock Out Plaques
  3. Four Immunotherapies Now Banish Amyloid From the Brain
  4. Amyloid Clearance: Check. Cognitive Benefit: Um … Maybe.
  5. BAN2401 Removes Brain Amyloid, Possibly Slows Cognitive Decline
  6. Aducanumab, Solanezumab, Gantenerumab Data Lift Crenezumab, As Well
  7. CTAD: Solanezumab Seen to Nudge AD Ever so Slightly
  8. End of the EXPEDITION: Solanezumab Results Published

Paper Citations

  1. . A Combined Measure of Cognition and Function for Clinical Trials: The Integrated Alzheimer's Disease Rating Scale (iADRS). J Prev Alzheimers Dis. 2015 Dec 1;2(4):227-241. PubMed.
  2. . Phase 3 solanezumab trials: Secondary outcomes in mild Alzheimer's disease patients. Alzheimers Dement. 2015 Aug 1; PubMed.

External Citations

  1. press release

Further Reading