22 Nov 2024

Most Alzheimerologists agree that lecanemab and donanemab are a start, but insufficient to treat this disease. At the Clinical Trials in Alzheimer’s Disease conference, held last month in Madrid, scientists not only debated how to make anti-Aβ monoclonals safer and more effective (Part 2 of this series; Part 1 of this series; Part 6 of this series), but also shared findings from Phase 1 trials evaluating new approaches. Gene silencing looked like it might shush APP and tau, and a gene therapy cranked up the protective ApoE2 protein while possibly nudging AD biomarkers. Arrows aimed at tau—an antisense oligonucleotide and a monoclonal antibody—might take down tangles. Meanwhile, immune checkpoint inhibitors and intranasal insulin are budding to target other aspects of the disease.

In her keynote presentation in Madrid, Catherine Mummery of University College London showed an image of a picturesque valley with the snowy peak of Nanga Parbat looming in the distance. Mummery noted the progress the field has seen over the past few years in its approach. “We are in the foothills of our treatment journey,” she said, “If we want to get to those dizzying heights of cures, or at least of halting the disease, then we need to redouble our efforts.”

Looong Trek to Summit. At CTAD, Nanga Parbat served as inspiration for trialists working to build a therapeutic approach to conquering an imposing, scary disease.

Down With APP
For starters, how else can scientists thwart amyloidosis? One approach is mivelsiran, previously known as ALN-APP. Developed by Alnylam of Cambridge, Massachusetts, with Regeneron in Tarrytown, New York, this small interfering RNA targets the APP transcript for destruction. Conjugated to a lipid chain that greases its passage into cells, this siRNA takes down APP and all its downstream metabolites, Aβ42 and Aβ40 among them.

Mivelsiran is in Phase 1 in AD, and in Phase 2 for cerebral amyloid angiopathy. Previously, Mummery had presented interim safety and biomarker data from a single-ascending-dose trial in people with early onset MCI or mild AD, who received an intrathecal infusion of placebo or 25 mg, 35 mg, 50 mg, or 75 mg of mivelsiran. A 70 percent drop in APP metabolites in CSF on the two higher doses stayed 30 percent below baseline by six months (Nov 2023 conference news).

In Madrid, Mummery reported 12-month findings for the two highest dose groups. No ARIA occurred, though adverse events related to lumbar puncture did. On both 50 mg and 75 mg, CSF sAPPb remained 30 percent below baseline. Mummery also showed six-month data for Aβ40 and Aβ42, in which the 75 mg group remained 27 and 40 percent reduced relative to baseline, respectively. This finding was not dose-dependent, however.

Durable Effect. Metabolites APPβ (top), Aβ40 and Aβ42 (bottom) plummeted after treatment with 35, 50, or 75 mg mivelsiran. [Courtesy of Alnylam Pharmaceuticals.]

According to Alnylam’s Tim Mooney, mivelsiran’s endurance is due to stabilizing tweaks the scientists made to the siRNA backbone. Once the construct enters cells, it can repeatedly silence mRNAs before it gets degraded. “It can punch above its weight,” Mooney said, adding that future work will define the required dosing frequency.

Both this study, as well as multi-dose studies, are ongoing among people with EOAD. Mummery considers mivelsiran’s potential for infrequent dosing a plus. After all, who wants monthly infusions near their spinal cord? “Giving such a drug once every nine to 12 months is a different prospect,” she told Alzforum.

Importantly, mivelsiran is also being evaluated to treat CAA. Not only is this cerebrovascular condition a common menace in its own right, but it often co-occurs in people with AD (Jäkel et al., 2021). CAA is difficult to diagnose but people with manifestations of it, such as superficial siderosis and microbleeds, are at increased risk for hemorrhage while on anti-amyloid antibodies. Both people with CAA alone, or with both conditions, need disease-modifying treatments. Could mivelsiran fit the bill?

In Madrid, Jin-Moo Lee, a stroke specialist who heads the neurology department of Washington University, St. Louis, laid out the rationale and design of the ongoing Phase 2 trial. Called cAPPricorn-1, it is enrolling people diagnosed with sporadic CAA as gauged by an updated version of the Boston criteria, which use several MRI features to infer cerebrovascular signs of CAA (Charidimou et al., 2022). People with MCI or mild AD in addition to CAA are eligible, though people with moderate or severe AD are not.

A separate “descriptive” cohort, which won’t count toward the primary endpoints, will enroll people with Dutch-type CAA. This rare, familial form of the disease worsens rapidly. It is caused by the E693Q APP mutation, located in the Aβ sequence itself. Its carriers typically suffer from recurrent strokes starting in midlife; some are members of DIAN.

Participants in this trial will receive a single intrathecal infusion of mivelsiran or placebo, followed by a two-year double-blind follow-up period. The primary endpoint is change in how many microbleeds arise in the cerebral lobes. Secondary endpoints include hemorrhagic and cerebrovascular signs of disease progression, most measured by MRI. Change in CSF sAPPα and sAPPβ is on the list, as are other fluid and imaging biomarkers designated exploratory endpoints, Mooney told Alzforum.

Depending on how both trials go, mivelsiran could be tried in late-onset AD, for example in ApoE4 homozygotes, or people who take medications that up their ARIA risk in response to anti-Aβ antibodies. Mummery suspects mivelsiran will work best to prevent amyloidosis, as it turns off the spigot of production.

Down With Tau
Only in the earliest stages of Alzheimer’s, years before symptoms, do anti-amyloid therapies stand a chance of stopping this disease in its tracks. For most people, halting progression will require hitting other parts of the cascade, likely more than one. “We need to target different modes of action if we want to make a greater difference beyond just nudging the needle,” Mummery told Alzforum.

Notable here is recent progress in targeting tau, which starts to misbehave earlier in AD progression than previously thought. A highlight at CTAD was a promising signal for bepranemab, the first tau immunotherapy that appears to remove tangles (Nov 2024 conference news).

Until then, the lone tau drug to have done that was Biogen’s antisense oligonucleotide BIIB080. It dampens tau expression and, in a Phase 1 trial, reduced tangles as per tau-PET (Apr 2023 conference news; Nov 2023 conference news). At CTAD, Mummery presented tau-PET trajectories of individual participants in this trial. Across Braak regions, nearly every participant saw their tau accumulation slow over the course of the trial. BIIB080 is now being evaluated in a Phase 2 trial, which is fully enrolled. Meanwhile, Eli Lilly recently threw its hat in the ring, starting a with an siRNA against tau called LY3954068.

Down with ApoE4
As the strongest AD risk gene, ApoE makes for a tantalizing therapeutic target. One approach attempts to reduce the negative effects of the high-risk ApoE4 isoform by forcing expression of the protective isoform, ApoE2. This is the rationale behind Lexeo’s LX1001, an adeno-associated virus (AAV) carrying the ApoE2 gene. At CTAD, Kim Johnson, a site investigator at Duke University in Durham, North Carolina, showed interim findings from an ongoing Phase 1/2 trial evaluating LX1001 among homozygous ApoE4 carriers with MCI or mild to moderate AD.

The 15 participants are divided into four little cohorts of three to five people, who received a single ascending dose of LX1001, ranging from 5.7 x 10¹² vector genomes for cohort 1, to 1.4 x 10¹⁴ vector genomes for cohort 4. The drug entered their CNS via intrathecal injection between their top two cervical vertebrae. At a previous CTAD, interim data from five participants in the two lowest dose groups looked as if the treatment safely stoked CSF ApoE2 production (Dec 2022 conference news). In Madrid, Johnson reported 12-month follow-up data for cohorts 1 to 3, and six-month data from cohort 4.

By this point, LX1001 was still largely safe and well-tolerated across cohorts, and no one developed ARIA, Johnson said. Twelve participants developed CSF pleocytosis, aka elevated white blood cells. This condition has been reported in response to other AAV therapies. It peaked at three months, receded by six months, and remained at bay at 12 months. Four serious adverse events cropped up during the study, of which one—mild hearing loss—was deemed potentially related to the drug. Johnson told Alzforum that the participant still has hearing loss, which is mild enough to be corrected with hearing aids.

Three months after injection, CSF ApoE2 ramped up in a LX1001 dose-dependent manner, holding steady at six and 12 months. CSF ApoE4 remained stable at the lower doses, but appeared to rise in cohort 4. The CSF ratio of ApoE2 to ApoE4 rose at three months, stabilizing thereafter (image below). CSF ApoE2 skyrocketed in one participant from cohort 3 at three and six months. This outlier declined to undergo lumbar puncture or PET at 12 months.

ApoE2 to the Rescue? CSF ApoE2 (left) rose three months after injection, stabilizing later. CSF ApoE4 stayed the same on the lower doses (middle). The CSF ApoE2/ApoE4 ratio increased dose-dependently (right). [Courtesy of Kim Johnson, Duke University.]

Amyloid-PET showed no significant change in amyloid accumulation. This measure was too variable to say whether the drug had any impact in such a small sample size, Johnson said.

What about tau biomarkers? For nine of the 13 participants, CSF p-tau181 and total tau had decreased relative to baseline by their last follow-up visit. P-tau217 and p-tau231 were measured in cohorts 2 and 3, where both dropped relative to baseline in four of the six participants.

Tau-PET scans at baseline and six months for cohorts 3 and 4, and out to 12 months for cohort 3, suggested that tau deposition stabilized, and for some, appeared to go down. The outlier with highest CSF ApoE2 had the steepest tau-PET reduction at six months. 

Gil Rabinovici, University of California, San Francisco, asked if the tau-PET scans had corrected for volume loss, to ensure the findings weren’t an artifact of ongoing brain shrinkage. Alzforum asked the company the same question, and Lexeo's Sandy See Tai clarified that Lexeo does not perform partial volume correction. "However, we use concurrent MRIs to each PET scan to define the target regions of interest for the PET SUVR analysis, therefore, atrophy that has occurred since baseline is taken into account," See Tai wrote to Alzforum.

Johnson acknowledged that, despite the encouraging CSF biomarker and tau-PET results, the study is too small to draw conclusions. Complete data from the 52-week trial will be available soon, and participants will enter a long-term extension to measure CSF ApoE2, CSF biomarkers, and amyloid- and tau-PET annually for four years. 

Even as LX1001 moves through its trial, a new mouse study by Lexeo’s founder, Ronald Crystal of Weill Cornell Medical College in New York, claims that adding a protective Christchurch mutation to the AAV-packaged ApoE2 might strengthen the protection (Günaydin et al., 2024).

Wrangling Inflammation and Metabolism
Dysfunction in these two areas is well recognized in AD. Tweaking inflammation is dicey, as scientists are still exploring which immune responses are beneficial versus detrimental. What’s more, immune responses change as disease progresses.

One controversial approach seeks to take the breaks off the immune system with checkpoint inhibitors such as antibodies against programmed cell death-1 receptor and its ligands PD-L1 and PD-L2. Preclinical work from Michal Schwartz’s lab at the Weizmann Institute of Science in Rehovot, Israel, supports this approach (Jan 2016 news; Feb 2019 news). To carry it into humans, Schwartz co-founded ImmunoBrain Checkpoint, which is running early dosing studies of an anti-PD-L1 antibody in people with early AD, with Mummery as its principal investigator.

The studies are testing six doses of an anti-PDL1 antibody in 40 people with early AD. They receive an intravenous infusion of placebo or the antibody once every quarter for 12 months. In Madrid, Mummery said that administration of the fifth dose is happening now. As expected for a checkpoint inhibitor, there have been mild immune-related adverse events, but no worrying safety signals so far, she said.

Also at CTAD, Suzanne Craft of Wake Forest School of Medicine in North Carolina reported biomarker data from a one-month trial of intranasal insulin alone or in combination with empagliflozin. Insulin exerts metabolic and immune effects in the brain; “Empa” is a sodium-glucose co-transporter 2 (SGLT2) inhibitor that reduces glucose uptake and improves vascular function. Craft hopes that the two drugs together may act additively or synergistically. She reported that a new intranasal device, called Apgar CPS, delivered insulin more effectively into the brain than was seen in a previous trial; radiolabeled insulin-PET scans are used to confirm exposure.

The month-long trial included 47 people with a clinical diagnosis of MCI or early AD, who were randomized to placebo, 40 IU of intranasal insulin four times per day, 10 mg of empa per day, or both. At baseline and four weeks, CSF and plasma biomarkers, MRI, and cognition were measured. No treatment-related adverse events cropped up.

A few possible signals from this trial: exploratory findings included improved scores on the PACC5, and more intact white matter among people on intranasal insulin alone or with empa. People on empa—with or without insulin—had a drop in CSF total tau. Craft showed changes in CSF and plasma markers of inflammation, as measured by Alamar’s Nulisa platform. In CSF, insulin stoked brain-derived neurotrophic factor while quelling the synapse-targeting complement protein C1qa. Empa reduced pro-inflammatory proteins including IL-6 and TNFα.

Surprisingly to Craft, both treatments nudged plasma markers. Intranasal insulin reduced pro-inflammatory proteins such as interferons, chemokines, and adaptive immune regulators, whereas Empa docked plasma levels of the immune regulators NAMPT and SIRPa, both of which have been implicated in AD risk. These effects were strongest among those with higher baseline p-tau217, suggesting people with amyloid are likely to benefit from the treatment.

These responses could be due to chance in this small sample. Or they could reflect immune communication between the CNS and periphery via the nasal olfactory plexus, Craft believes. Positioned behind the nasal cavity, this structure is an immune hub and part of the glymphatic system (for review, see Chae et al., 2024). Craft hopes her approach will find use alone or with anti-Aβ immunotherapy.

For her part, Mummery envisions combination therapies that address multiple mechanisms contributing to AD. Each therapy might be deployed for different stages of disease, defined by the field’s ever-growing arsenal of biomarkers.

Working toward this future, Mummery heads the U.K.’s Dementia Translational Research Collaboration (D-TRC), a government-funded initiative to create a network of clinical trial sites across the U.K. (Feb 2024 community news). Sites in this network will receive training and resources to run Phase 1 and 2 trials for innovative therapies, including gene therapies that require a deep technical expertise. Mummery told Alzforum that so far, 30 sites have applied to join the network. Companies interested in running early stage trials in the U.K. can tap the network to access sites equipped with the facilities to run such studies. In Madrid, D-TRC investigators got together with interested biotech partners to hash out plans for future trials.—Jessica Shugart

Comments

No Available Comments

Make a Comment

To make a comment you must login or register.

References

News Citations

1. Donanemab: Small Tweak in Titration, Big Gain in Safety? 8 Nov 2024
2. Trontinemab Data Strengthen Hope for Brain Shuttles 8 Nov 2024
3. Fully Loaded: Secondary Prevention Studies of Lecanemab, Donanemab 15 Nov 2024
4. Moving Forward: RNA-Targeted Attempts at Taking Down Tau, APP 9 Nov 2023
5. Finally, Therapeutic Antibodies Start to Reduce Tangles 14 Nov 2024
6. First Hit on Aggregated Tau: Antisense Oligonucleotide Lowers Tangles 7 Apr 2023
7. In Small Trial, Gene Therapy Spurs ApoE2 Production 14 Dec 2022
8. Cancer Drug Clears Plaque, Improves Mouse Memory 21 Jan 2016
9. Cancer Treatment Takes Aim at Tauopathy 5 Feb 2019
10. New U.K. Trials Network to Accelerate Dementia Studies 16 Feb 2024

Therapeutics Citations

1. ALN-APP
2. BIIB080
3. LY3954068
4. Nasal Insulin
5. Empagliflozin

Mutations Citations

1. APP E693Q (Dutch)
2. APOE C130R (ApoE4)
3. APOE R176C (ApoE2)

Paper Citations

1. Jäkel L, De Kort AM, Klijn CJ, Schreuder FH, Verbeek MM. Prevalence of cerebral amyloid angiopathy: A systematic review and meta-analysis. Alzheimers Dement. 2021 May 31; PubMed.
2. Charidimou A, Boulouis G, Frosch MP, Baron JC, Pasi M, Albucher JF, Banerjee G, Barbato C, Bonneville F, Brandner S, Calviere L, Caparros F, Casolla B, Cordonnier C, Delisle MB, Deramecourt V, Dichgans M, Gokcal E, Herms J, Hernandez-Guillamon M, Jäger HR, Jaunmuktane Z, Linn J, Martinez-Ramirez S, Martínez-Sáez E, Mawrin C, Montaner J, Moulin S, Olivot JM, Piazza F, Puy L, Raposo N, Rodrigues MA, Roeber S, Romero JR, Samarasekera N, Schneider JA, Schreiber S, Schreiber F, Schwall C, Smith C, Szalardy L, Varlet P, Viguier A, Wardlaw JM, Warren A, Wollenweber FA, Zedde M, van Buchem MA, Gurol ME, Viswanathan A, Al-Shahi Salman R, Smith EE, Werring DJ, Greenberg SM. The Boston criteria version 2.0 for cerebral amyloid angiopathy: a multicentre, retrospective, MRI-neuropathology diagnostic accuracy study. Lancet Neurol. 2022 Aug;21(8):714-725. PubMed.
3. Günaydin C, Sondhi D, Kaminsky SM, Lephart H, Leopold PL, Hackett NR, Khanna R, Crystal RG. AAVrh.10 Delivery of Novel APOE2-Christchurch Variant Suppresses Amyloid and Tau Pathology in Alzheimer's Disease Mice. Mol Ther. 2024 Nov 6; Epub 2024 Nov 6 PubMed.
4. Chae J, Choi M, Choi J, Yoo SJ. The nasal lymphatic route of CSF outflow: implications for neurodegenerative disease diagnosis and monitoring. Anim Cells Syst (Seoul). 2024;28(1):45-54. Epub 2024 Jan 29 PubMed.

Further Reading

No Available Further Reading