15 Jan 2023

Part 1 of 2
The year 2022 was a turning point in Alzheimer's research. It began with the Aduhelm debacle fresh on people's mind. It ended with reassurance that amyloid removal indeed helps people with the disease, when lecanemab's Phase 3 data showed a slowing of cognitive decline. This result, coupled with evidence that blood tests will diagnose AD at earlier stages than were trialed thus far, fed hopes that anti-amyloid therapy may eventually be able to nip AD pathogenesis in the bud.

Far beyond this particular approach, research in the field advanced across the board in 2022, be it multi-omics analyses using human tissues, better animal models, deeper functional studies of gene variants, or dissections of neuroimmune cell biology. As you gear up for a productive 2023, take stock of last year's main developments. Read our summary of therapeutics development and biomarker research below, and more basic research topics in Part 2.

Therapeutics

The FDA's accelerated approval, on January 6, 2023, of the anti-Aβ antibody lecanemab, aka Leqembi, for early Alzheimer's disease has swung the field's spotlight sharply toward the future, making the 2022 developments leading up to the U.S. agency's decision look suddenly small in the rearview mirror. Capping this monoclonal antibody's 15-year-long development, Eisai and Biogen in May 2022 submitted a biologic license application based on Phase 2 data. By July, the agency had committed to priority review. September brought top-line news for lecanemab's Phase 3 trial and, two months later at CTAD, the data followed, paired with a peer-reviewed paper and plans to apply for full approval in the U.S. and other countries in 2023. For lecanemab, a large Phase 2 trial using an adaptive Bayesian design had generated data that were confirmed in Phase 3. "Confirmatory" is what Phase 3 is meant to be but, after two decades of skimpy Phase 2 trials followed by flops in Phase 3, Alzheimerologists had not seen this traditional sequence of events in a long time. Leqembi's clinical benefit in symptomatic patients is real but small—a mere foothold to expand upon with additional research. The ARIA risks are real, as well. Even so, while 2022 saw the first FDA-approved anti-amyloid antibody, aducanumab, get mired in controversy (see below), it also saw the final development steps of the second such antibody go off without a hitch.

Diverging Trajectories. People on lecanemab worsened more slowly on the CDR-SB than did people on placebo, resulting in a quarter less progression at 18 months. [Courtesy of Eisai.]

Not so for another closely watched Phase 3 program. Expectations that Roche's gantenerumab would prevail in two large trials were shakier because it lacked prior data showing cognitive benefit, but even so, ample biomarker data had shown this antibody removes amyloid from the brain. Alas, in November 2022, gantenerumab went under in a maelstrom of factors that added up to yet another case of "too little too late." Trial participants had started out with more brain amyloid than those in lecanemab's Phase 3, less of their amyloid was removed than had been expected, and results merely trended in favor of treatment. Disheartening as this was, it fit the emerging picture that robust amyloid removal must precede a clinical benefit. What role the antibody's formulation, or its injection under the skin rather than intravenously, played are still open questions. All trials have ended, except for an open-label extension of a DIAN trial that began in 2012. Gantenerumab's 22-year development is over, as attention shifts toward its successor, trontinemab, now in Phase 1.

Similarly to gantenerumab, the tail end of a nine-year long trial in an autosomal-dominant population—presymptomatic PSEN-1 mutation carriers in Colombia—delivered the nail in the coffin for crenezumab. This anti-Aβ antibody had proven ineffective in symptomatic LOAD in 2019, but it took until 2022 to learn that while it trended in the right direction on most outcomes in early ADAD, it still fell short of statistical significance. [Correction posted January 19, 2023: According to Roche, evaluation of crenezumab in the API is concluding. However, according to AC Immune, the overall crenezumab development program has not been terminated].

For Aduhelm, 2022 piled on more woes. The year started with the CMS denying broad insurance coverage. It continued with an unfavorable comparison with competitor donanemab, which pulls amyloid from the brain faster. It ended with the U.S. House of Representatives scolding both the FDA and Biogen. And yet, this drug is down but not out. Also in 2022, a Phase 3 trial to verify clinical benefit started recruiting at 47 sites; it may yet resurrect Aduhelm.

Next up before the FDA could be donanemab. Lilly completed rolling submission for an accelerated approval application for its anti-pyroglutamate Aβ antibody last summer, starting a priority review that will render a decision this February. This application rests on Phase 2 data, but one Phase 3 trial is ready to read out this spring, a second is enrolling across Asia and Europe, and a prevention trial in a presymptomatic population is up and running as well. In October 2022, a Phase 3 started up for a follow-up antibody called remternetug. Scientists suspect it reduces donanemab's penchant to induce auto-antibodies, hence rendering itself ineffective over time. Across companies, scientists are working on better therapeutic antibodies, for example more potent versions, such as PRX012, which has higher affinity for Aβ. Most of those projects had not quite broken through yet in 2022.

Could Benefit of Plaque Removal Grow in Time?

Drugs targeting tau in Alzheimer's and related diseases posted no big successes in 2022, though that does not mean research is dormant. For example, a second wave of anti-tau antibodies target this protein in a different way than did the initial N-terminally directed batch of immunotherapies that proved ineffective. Some new antibodies go after tau's mid or MTBR regions, or after oligomer-specific epitopes (e.g. E2814, JNJ-63733657),while other efforts deploy new tau constructs as vaccines. A provocative question bubbled up, too: Given that tau forms physiological phospho-aggregates in hibernating mammals, which dissolve like so much snow come spring, could it be that the tau proteopathy scientists document in AD is a consequence of metabolically dead neurons—hence too late a phenomenon to constitute a drug target?

New biology in neurodegeneration and inflammation has generated potential drug targets, some of which have reached early clinical trials. For example, TPN-101, a small-molecule LINE1 reverse transcriptase inhibitor initially studied in HIV, is being evaluated in Phase 2 trials for three different neuroimmune diseases, including ALS and PSP. In another neuroimmune approach, an LRRK2 inhibitor appeared to enter the brain and rescue lysosome activity in Phase 1; its follow-up compound is in a 640-patient Phase 2 trial for Parkinson's, as well as in a phase 3 trial for LRRK2 mutation carriers. Restoring lysosomal function in the context of neuroinflammation is also the goal of DNL593, currently in Phase 1/2. This replacement therapy comprises a progranulin molecule hitched to a transferrin antibody that ferries its cargo into the brain after intravenous injection. An intrathecal, viral gene therapy attempts to deliver the ApoE2 protein to the brains of ApoE4 homozygote AD patients. Called LX1001, it reported initial safety and exposure results in 2022. The anti-C1q complement antibody ANX005 showed target engagement in a trial for Huntington's and, in 2022, entered a trial in ALS. XPro1595, a pegylated protein that selectively shields TNFα from type 1 but not type 2 TNFα receptors, is in Phase 2 for Alzheimer's disease. Finally, the anti-inflammatory sterol metabolite NE3107, which has been tested unsuccessfully in other conditions since 2007, was in trials for Alzheimer's and Parkinson's diseases by 2022.

New biology is not brexpiprazole's claim to fame. Sold as Rexulti, this atypical antipsychotic has been around for a decade to alleviate schizophrenia and depression. In 2022, it cleared a Phase 3 trial for agitation in Alzheimer's, hence this troubling symptom of advanced Alzheimer's dementia may soon join its list of approved indications.

Biomarkers

Markers of both the biofluid and brain imaging kind are essential tools for intervention trials and cohort studies, and 2022 was a good year on this front. Blood tests, especially, are hurtling toward broad use much faster than scientists had envisioned as recently as a few years ago.

Tau

Finding good tau markers used to be a tough nut to crack. It broke open three years ago when scientists reported that certain phospho-tau epitopes in the CSF and plasma correlated tightly with amyloid plaque load in the brain. The race was on to see how p-tau181, p-tau217, p-tau231, and other forms in the CSF and plasma, relate to AD pathology in the clinical cohorts the field has been assembling around the world. In 2022, consensus started to form that in AD, plasma p-tau231 inches up first and soon plateaus, whereas p-tau181 and p-tau217 continue to climb. The latter leads the pack in best correlating with both amyloid and tau PET. In the Down’s syndrome model of Alzheimer's, too, plasma p-tau217 tracks with both amyloid PET and cognition.

Neuropathology validation against postmortem plaques and tangles has begun for these blood tests. At present, plasma p-tau217 outperforms the other p-taus, Aβ42/40, GFAP, and NfL. In community cohorts, both mass spec-based assays and immunoassays for plasma p-tau217 performed similarly.

Thus far, the data suggest that p-taus may help diagnose Alzheimer’s and stage its progression. A p-tau217 “clock” may come in handy here. Once plaques have formed, CSF p-tau217 reaches a tipping point and, thereafter, climbs at a constant rate in most people over the next 30 years. In theory, then, a single CSF test could predict when a person would start having symptoms.

From Chaos, Order. Aligning people's p-tau217 trajectories by their chronological age reveals no clear pattern, but anchoring them by the p-tau217 tipping point reveals a consistent rate of change for nearly everyone. Watch the movie. [Courtesy of Suzanne Schindler.]

In 2022, a handful of research-grade p-tau217 assays were moving through the standardization steps that precede regulatory approval. At the same time, the wealth of p-taus that have emerged has invigorated basic research into how this protein is being cleaved and post-transcriptionally modified in neurodegenerative disease.

As the year drew to a close, a turning point may have come for total tau, too, which has long been seen as a marker of neurodegeneration. Scientists are now able to distinguish, in blood, tau made and released by brain cells from tau made and released by peripheral cells. This may lead to a robust blood test for AD-specific neurodegeneration. For primary tauopathies, which lack specific blood tests, the dearth of a microtubule-binding region fragment of tau in the CSF might signal accumulation of neurofibrillary tangles.

Will all this progress sideline tau PET? By December 2022, the short answer seemed to be: yes, for large-scale trial screening or routine clinical diagnosis, but no for trial outcomes, difficult diagnostic cases, and manifold research questions. For example, evidence is trickling in that various amyloids can raise brain levels of p-tau181 and p-tau217 without subsequent deposition of tangles; PET could differentiate such cases. Further, while p-tau217 predicts that tangles will form in cognitively normal, amyloid-positive people, tau PET better predicts neurodegeneration and cognitive decline. Given the rapid progress on p-tau, a growing chorus of scientists is calling for a revision of the ATN amyloid-tau-neurodegeneration marker classification system.

Aβ

In the meantime, scientists continue to refine biofluid tests for amyloid based on the ratio of soluble Aβ42 and Aβ40. In May, the FDA approved Fujirebio’s fully automated, Lumipulse G-based immunoassay for CSF Aβ42/40, and in December, ditto for Roche’s CSF p-tau181/Aβ42 test for AD. Those two are formally approved diagnostic aids for Alzheimer's diagnosis, available at hospitals and commercial labs throughout the U.S.

Several plasma tests that still lack FDA approval are being used in trials and by dementia specialists. At this point, scientists recommend that they be limited to people who have symptoms of dementia and are being seen in specialized clinics. Next up are studies in community-based cohorts. They will focus on how diseases and medications common among the elderly affect these new tests, and thus help chart the path toward use in primary care.

None of the Aβ monomer tests measure the concentrations of toxic oligomers. A new affinity probe for β-sheets might help in this regard, as might a new assay for soluble oligomers.

Destined to Decline. Compared to cognitively intact people without AD pathology (A-T-, red lines) or people with only plaques (A+T-, light blue lines), people with plaques and tangles in their medial temporal lobes (A+T_MTL+, green lines) or neocortices (A+T_NEO+, navy line) progressed more quickly to MCI (left). Only the latter were likelier to progress to all-cause dementia (right). [Courtesy of Ossenkoppele et al., medRxiv, 2022.]

For amyloid PET, the European AMYPAD study in 2022 confirmed prior conclusions from U.S. and E.U. studies that these scans improve clinical diagnosis of AD. Post PET, physicians changed the diagnosis of one in three study volunteers. These real-world studies strengthened the case that brain amyloid predicts cognitive decline, though having both plaques and tangles boosts that risk much more. Who is at risk? An amyloid prevalence meta-analysis of almost 20,000 people from across the AD clinical spectrum was sobering. Among cognitively normal 50-year-olds, 17 percent already had amyloid in their brain, as did a third of 70-year-olds.

2022, a Year in Review

Quick Links

Tools

Therapeutics

Biomarkers

Tau

Aβ

Comments

Make a Comment

References

News Citations

Therapeutics Citations

External Citations

Further Reading

Annotate