Abstract
Alzheimer's disease (AD) is the major cause of dementia that is now threatening the lives of billions of elderly people on the globe, and recent progress in the elucidation of the pathomechanism of AD is now opening venue to tackle the disease by developing and implementing “disease-modifying therapies” that directly act on the pathophysiology and slow down the progression of neurodegeneration. A recent example is the success of clinical trials of anti-amyloid b antibody drugs, whereas other therapeutic targets, e.g., inflammation and tau, are being actively investigated. In this dual perspective session, we plan to have speakers from leading pharmas in the field representing distinct investments in the AD space, which will be followed by the comment from scientific leadership of the Alzheimer's Association who will speak on behalf of all stakeholders. Neuroscientists participating in the Society for Neuroscience may be able to gain insights into the cutting edge of the therapeutic approaches to AD and neurodegenerative disorders, and discuss future contribution of neuroscience to this field.
Introduction
The global prevalence of Alzheimer's disease (AD) is growing rapidly, given the continuous rise in the number of older adults (Alzheimer's Association, 2023). In 2023, there are 6.7 million adults ≥65 years of age living with AD and other dementia in the United States, a figure that is projected to double by 2050 (Alzheimer's Association, 2023). In addition to those living with AD and other dementias, ∼8–11% of Americans ≥65 years of age may have mild cognitive impairment (MCI) because of AD (Alzheimer's Association, 2023). It is important to note that while most prevalence estimates of AD and MCI are based on individuals ≥65 years of age, there are likely many more individuals who fall along the AD continuum (Alzheimer's Association, 2023). A report from the American Academy of Neurology in 2018 estimated that 15.8% of people in the United States ≥60 years of age had all-cause MCI (Alzheimer's Association, 2018; Petersen et al., 2018). Furthermore, many individuals may be in the preclinical stage of AD, where they do not yet show symptoms (e.g., memory loss) but display AD-related brain changes as measured by biomarkers (Alzheimer's Association, 2018). In line with this, a study in 2017 showed that 38.4 million Americans ≥30 years of age had elevated amyloid β (Aβ) in the brain but did not yet have MCI (Alzheimer's Association, 2018; Brookmeyer et al., 2018). The estimated number of people of any age falling along the AD continuum is likely higher than the current reported estimates (Alzheimer's Association, 2023).
Recent progress in the elucidation of the pathomechanism of AD is now opening venues to tackle the disease by developing and implementing “disease-modifying drugs” that directly act on the pathophysiology and slow down the progression of neurodegeneration. A recent example is the success of clinical trials of anti-amyloid β antibody drugs fully approved by the Food and Drug Administration (FDA), whereas other therapeutic targets (e.g., inflammation and tau) are being actively investigated. In this dual-perspective session, we plan to have two speakers from leading pharmas in the field representing distinct investments in the AD space, which will be followed by comment from the scientific leadership of the Alzheimer's Association, who will speak on behalf of all stakeholders (e.g., patients, scientists, policymakers, and industries).
Development of disease-modifying therapies based on central role of amyloid β in early Alzheimer's disease pathophysiology
In the past, therapeutic management of AD involved symptomatic therapies (cholinesterase inhibitors and memantine), which provide only modest clinical benefit and have no impact on disease progression (Birks, 2006; McShane et al., 2006). Two anti-amyloid antibodies, lecanemab and aducanumab, now have either accelerated or traditional approval from the FDA, and another drug, donanemab, has recently met the primary clinical end point in a phase 3 clinical trial (Sims et al., 2023). Although each of these agents is distinct, they were all developed based on the central role of amyloid early in AD pathophysiology and the Aβ theory, which hypothesizes that decreasing soluble aggregated forms of Aβ (oligomers and protofibrils) and fibrillar amyloid would impact disease progression and slow cognitive decline (Iwatsubo, 2022).
Lecanemab, a humanized IgG1 monoclonal antibody directed against aggregated soluble and insoluble forms of Aβ, has demonstrated robust brain amyloid plaque reduction and slowing of clinical decline in early AD phase 2 and phase 3 studies. Lecanemab has high selectivity for soluble aggregated species of Aβ compared with monomeric amyloid, with moderate selectivity over fibrillar amyloid, a profile thought to convey an advantage in selectively targeting the most toxic pathologic Aβ species. Lecanemab was developed using the principles of the “4Rs: for precision medicines: right hypothesis, right population, right dose, and right end points. The right hypothesis is based on the mechanism of action in which lecanemab binds with high affinity to soluble Aβ protofibrils, which have been shown to be more toxic to neurons than monomers or insoluble fibrils (Sehlin 2011, 2012; Magnusson 2013; Tucker 2015; Söderberg et al., 2023). The right population is early AD with confirmed amyloid pathology (i.e., MCI because of AD or mild Alzheimer's dementia) before extensive irreversible neocortical neuronal and synaptic loss. The right dose was established through an innovative Bayesian phase 2 dose-ranging study. Finally, the end points in the development program included clinical outcomes [CDR-SB (Clinical Dementia Rating-Sum of Boxes), ADAS-cog14 (Alzheimer's Disease Assessment Scale-Cognitive Subscale test 14), ADCOMS (AD Composite Score), and ADCS-MCI-ADL (Alzheimer's Disease Cooperative Study–Activities of Daily Living Scale)], imaging biomarkers (amyloid PET and tau PET), plasma and CSF biomarkers, safety, and health-related quality-of-life measures. These included validated and well accepted AD clinical study end points and provided a comprehensive assessment of the impact of lecanemab on clinical disease progression and the underlying biology of AD.
The phase 2 study was a large, 18 month, proof-of-concept, dose-finding study (ClinicalTrials.gov ID, NCT01767311) with an open-label extension (OLE) conducted in 856 patients with MCI because of AD and mild AD dementia (collectively defined as early Alzheimer's disease). The study identified that lecanemab, 10 mg/kg, i.v., biweekly, without titration, is the optimal dose to test in phase 3 for amyloid clearance, downstream biomarker effects, as well as clinical efficacy and safety (Swanson et al., 2021; McDade 2022; Honig et al., 2023). There was dose-dependent, time-dependent, and statistically significant reduction of amyloid and improvement of related biomarkers. Amyloid clearance correlated with clinical benefit at the population and patient level.
The lecanemab phase 2 study had a gap period (Gap) with no treatment between the randomized core treatment phase (Core) and the single-arm OLE phase. Individuals were started on lecanemab, 10 mg/kg, i.v., biweekly after a period of 9–59 months off-treatment (mean, 24 months). The clinical and biomarker treatment differences of lecanemab versus placebo at the end of the Core were maintained while off-treatment during the Gap up to the beginning of the OLE in early AD patients, consistent with a disease-modifying effect (McDade 2022).
The lecanemab phase 2 proof-of-concept study provided a robust framework to optimally design the confirmatory phase 3 Study, Lecanemab for People with Early Alzheimer's Disease (Clarity AD; ClinicalTrials.gov ID, NCT03887455), intended to verify the efficacy and safety of lecanemab in patients with early AD. Clarity AD was a double-blind, randomized, placebo-controlled study of 1795 patients with early AD randomized to 10 mg/kg, i.v., lecanemab or placebo for 18 months of treatment. Lecanemab produced highly statistically significant slowing in multiple measures of cognitive and functional decline and slowing of decline of quality of life, and was accompanied by effects on biomarkers consistent with impact on disease biology. The clinical meaningfulness of the treatment differences with lecanemab treatment relative to placebo is supported by the convergence of evidence across all outcomes that are important to patients and caregivers, as follows: (1) consistent results across scales of cognition and function (26–37% slowing), across domains within scales, and across clinically relevant subgroups; (2) delay in progression by slope analysis (delay of 5.3 months over the 18 month trial), and by time-to-event analysis of progression to the next stage of AD (hazard ratio, 0.69; 95% CI, 0.57–0.83, p = 0.0001); (3) 23–50% slowing of decline in health-related quality of life measures and 38% slowing of caregiver burden; and (4) lecanemab effects on biomarkers of amyloid, tau, neurodegeneration, and gliosis that provide a biological basis for the treatment effects consistent with disease modification.
Important adverse events associated with lecanemab treatment were infusion-related reactions (26.4%) and amyloid-related imaging abnormalities [amyloid-related imaging abnormalities involving edema and hemorrhages (ARIA-E), 12.6%; symptomatic ARIA-E, 2.8%; ARIA with hemosiderin (ARIA-H), 17.3%; symptomatic ARIA-H 0.7%; van Dyck 2023]. ARIA usually occurs early in treatment (90% within 6 months) and is usually asymptomatic (3% rate of symptomatic ARIA), although serious and life-threatening events, including seizure and status epilepticus, rarely can occur (serious symptoms have been reported in 0.7%). When present, reported symptoms associated with ARIA may include headache, confusion, visual changes, dizziness, nausea, and gait difficulty. Focal neurologic deficits may also occur. Symptoms associated with ARIA usually resolve over time. The risk of ARIA, including symptomatic and serious ARIA, is increased in apolipoprotein E (ApoE) ε4 homozygotes. There was no increase in isolated ARIA-H (i.e., ARIA-H in patients who did not also experience ARIA-E) for lecanemab compared with placebo. In addition to ARIA, intracerebral hemorrhages >1 cm in diameter have occurred in patients treated with lecanemab [0.7% vs 0.1% in placebo; van Dyck et al., 2023; see lecanemab package insert at: https://www.accessdata.fda.gov/drugsatfda_docs/label/2023/761269s000lbl.pdf)].
Future research will further probe the disease-modifying effects of lecanemab, evaluate the therapy in novel settings and/or therapeutic combinations, and explore alternate dosing regimens/formulations.
The next generation of Alzheimer's therapeutics: building on the success of anti-Aβ antibodies toward increasingly effective disease treatment
The recent approval of lecanemab, followed by encouraging data from clinical studies of additional Aβ targeting antibodies, provides the first examples of therapeutics capable of modifying the course of disease and represents a significant milestone in AD treatment (Sims et al., 2023; van Dyck et al., 2023). Importantly, these clinical studies also provide validation of the “amyloid hypothesis,” which proposes that amyloid plaques composed largely of the Aβ peptide are an essential initiator and driver of cognitive decline in AD (Selkoe and Hardy, 2016). Although the amyloid hypothesis has long been supported by a combination of genetic and pathologic data, some doubt existed with regard to the relevance of amyloid plaques in symptomatic AD patients based on discouraging results in previous clinical trials testing other therapeutics designed to modify Aβ levels (Hardy and Selkoe, 2002; Hampel, 2021). As noted earlier in this article, lecanemab reported a 27% slowing in the course of decline in individuals with mild AD, which is accompanied by a significant change in multiple disease-related biomarkers that further support that this molecule is impacting disease (van Dyck et al., 2023). Donanemab, an antibody directed at an N-terminal pyroglutamate Aβ epitope that is present only in established plaques, was recently reported as having similar results, further validating this therapeutic approach (Sims et al., 2023). Although these results are highly encouraging, the efficacy of anti-Aβ molecules reported to date is relatively modest and comes with some potential of ARIA-related safety findings in a subset of patients that may be difficult to manage as these drugs begin to be prescribed. The incidence and timing of ARIA vary among treatments. While greater cortical volume loss relative to placebo (with preservation of hippocampal volume) is seen with amyloid-lowering drugs, volume loss is not associated with worsening in any neurodegenerative biomarkers or clinical outcomes (Bateman, 2022; Sims et al., 2023). This likely reflects pseudoatrophy related to fluid shifts, mobilization of amyloid, and reduced amyloid-associated dystrophic neurites and inflammation/gliosis. Based on these factors, the approval of lecanemab and potentially additional anti-Aβ antibodies is not the finish line for treatment of AD, but instead represents a launching point to stimulate development of the next generation of therapeutics founded on our evolving understanding of disease genetics and pathology to enable the field to continue building off this meaningful success.
A diverse set of therapeutic strategies for AD are currently in development, though the recent clinical successes will require the field to shift to a new normal that includes AD patients being treated with anti-Aβ antibodies. Given this, perhaps the most tractable opportunity in the near term is to further improve Aβ targeting therapeutics to enable efficacy or safety profiles superior to lecanemab and donanemab. Indeed, multiple groups are already pursuing this strategy with the hopes that improving the binding properties of Aβ-directed antibody to specific oligomeric Aβ species will result in enhanced activity (Perneczky et al., 2023). Another major limitation of anti-Aβ antibodies is the poor CNS uptake, which is limited to ∼0.1% or the injected dose, as is the case for all therapeutic antibodies because of the specialized vasculature known as the blood–brain barrier (BBB) that prevents perfusion of antibodies into the CNS. To address this challenge, an antibody transport vehicle (ATV) platform that improves the CNS uptake of antibodies by up to 30-fold with significantly improved biodistribution was developed (Kariolis et al., 2020). The ATV platform is based on engineering binding to transferrin receptor (TfR) or CD98 into the Fc domain of the antibody without the addition of appended sequences. Both TfR and CD98 are highly expressed at the BBB, and binding to one of these receptors enables transcytosis of ATVs into the brain (Zuchero, et al., 2016). Addition of the ATV to an anti-Aβ molecule has the potential to improve plaque engagement while reducing the amount of antibody present in perivascular regions. Consistent with this idea, preclinical studies in which gantenerumab was coupled to another TfR-based platform revealed that these molecules display significantly improved efficacy compared with the parent molecule (Weber et al., 2018).
It is also formally possible that the effect size observed in recent clinical studies with anti-Aβ antibodies represent the maximum clinical benefit achievable through targeting this mechanism, particularly in individuals with symptomatic AD. Should this be the case, targeting distinct mechanisms will be required to further impact the course of cognitive decline. Human genetic studies have demonstrated that variants in a number of microglia-enriched genes contribute to AD risk, suggesting that this cell type plays a causative role in disease (Efthymiou and Goate, 2017). There has been a specific interest in TREM2, a cell surface microglial receptor for which multiple coding variants have been identified (Lewcock et al., 2020). As functional studies suggest that the risk alleles are loss of function, antibodies that agonize TREM2 coupled with the TV platform (ATV:TREM2) were developed. In mice, treatment with ATV:TREM2 is sufficient to shift a microglial cell state to a more metabolically active cell state, to boost glucose metabolism, and to promote phagocytosis (van Lengerich et al., 2023). The microtubule binding protein tau has also garnered significant interest based on the correlation between neurofibrillary tangle pathology and cognitive decline, although therapeutic strategies have proven challenging based on an incomplete understanding of the mechanistic basis by which tau contributes to AD. However, approaches to knock down tau expression with antisense oligonucleotides (ASOs) has demonstrated encouraging initial clinical data. Recent data demonstrating that oligonucleotides can be effectively distributed throughout the CNS following intravenous administration through conjugation to the TV platform suggests that tau and potentially other ASO targets can be addressed using this delivery strategy (Barker et al., 2023). Last, coding variants in ApoE are the largest contributor to genetic risk of developing late-onset AD, though the mechanisms by which ApoE impacts disease are not fully defined and have the potential to be pleiotropic. Significant efforts exist to further define these mechanisms and translate them into therapeutics (Raulin et al., 2022).
Future clinical studies to examine these and related therapeutic approaches to determine whether they are capable of slowing disease progression as a single agent or in combination with an anti-Aβ molecule will be an important next step for the AD field. As additional targeted AD therapies continue to be developed, it is paramount that these studies are also used to gain a better understanding of the mechanisms that drive disease. Progress on this front was essential for the success of the current antiamyloid approaches and is expected to be similarly essential in the development of next-generation therapeutics. Further advancement of the use of genetic, imaging, fluid, and other biomarkers to identify the appropriate patient population, appropriate dose, and impacts on disease is expected. Development of more patient-friendly formulations and therapeutics, including subcutaneous active vaccination or novel oral treatments, may be able to address limitations of the current emerging intravenous Aβ antibody approaches. Many challenges lie ahead in the quest toward more effective treatments for AD, but the recent clinical successes combined with an improved understanding of disease mechanisms create optimism about the path ahead.
Alzheimer's disease treatment landscape: the future of research, treatment, and care
The emergence of novel FDA-approved AD therapies that change the underlying course of the disease and provide demonstrable clinical benefit (Budd Haeberlein et al., 2022; van Dyck et al., 2023) has opened up a new landscape in AD treatment and diagnosis. These therapies are specifically targeted at individuals with MCI or mild dementia because of AD with confirmed evidence of elevated amyloid-β in the brain (Budd Haeberlein et al., 2022; van Dyck et al., 2023). This emphasizes the critical need for early and accurate diagnosis of the disease to ensure that eligible individuals receive the highest benefit at the earliest stage of the disease (Kallmyer et al., 2021). To this end, a biomarker-based approach holds more significant promise than a syndrome-based approach (Jack et al., 2018). Several biomarkers have shown promise in aiding the diagnosis of AD and providing valuable insight into AD pathology. This includes biomarkers of functional impairment, neuronal loss, and protein deposition that can be assessed with brain imaging (e.g., MRI and PET) or CSF and are being used in research studies and specialized memory clinics (Frisoni et al., 2017; Jack et al., 2018). However, their application in primary care settings is hindered by limited accessibility, invasiveness, contraindications, and high costs (Hansson et al., 2022). In contrast, blood-based biomarkers (BBMs) have recently shown significant potential in improving the diagnostic workup of AD patients (Bjorkli et al., 2020; Hansson et al., 2022). They offer a less invasive, more feasible, and cost-effective option for widespread application in clinical practice and for evaluating the effectiveness of therapies in clinical trials (Hansson et al., 2022). However, additional data are needed before BBMs can be used as stand-alone diagnostic tests and before considering broad use in primary care (Hansson et al., 2022). Less expensive and less invasive biomarkers, such as BBMs, combined with clinical, demographic, and other information, will likely play an important role in the progress of AD diagnosis and clinical trials (Jack et al., 2018; Hansson et al., 2022)
Interpreting the clinical meaningfulness of the novel FDA-approved AD therapies is an important topic for researchers and clinicians to develop and prescribe treatments, and fuels discussions of the benefits and risks of new drugs. In 2022, a workgroup of experts convened by the Alzheimer's Association discussed what is considered a meaningful benefit or slowing of AD while considering the needs of patients and families (Petersen et al., 2023). They emphasized that slowing the progression of the disease has measurable and meaningful benefits for patients and families, especially in early AD when cognition is still mostly intact. It is worth noting that a statistically significant benefit in an 18 month AD trial may have an even greater impact when projected over succeeding years (Dickson et al., 2023; Petersen et al., 2023). Expectations for AD clinical trials may need to be adjusted to align with the complex and progressive nature of the disease (Dickson et al., 2023; Perneczky et al., 2023; Petersen et al., 2023).
While initially patient access to Leqembi has been blocked by the US Center for Medicare and Medicaid Services (CMS) coverage policies, it has recently been announced that CMS will cover AD treatments that receive traditional FDA approval if clinicians participate and enter data in a simple registry. Such action of the leaders of CMS demonstrates that they have listened to Alzheimer's disease experts, advocates, people living with the disease, and families in creating their plan for covering traditionally approved treatments. Moving forward, the plans of CMS must be implemented effectively to minimize clinician burden and, in so doing, maximize patient access. People living with AD and other dementias have the right to make informed decisions about their treatment options in collaboration with their physicians and families.
The long-term outcomes of novel FDA-approved AD therapies can ultimately be evaluated in real-world practice beyond the confines of the clinical trial setting (Perneczky et al., 2023). Such real-world data collection is essential to improve the applicability and generalizability of clinical trial findings through larger datasets with longer follow-ups that include more diverse populations (Chodankar, 2021). To this end, the Alzheimer's Association announced in 2021 the establishment of the Alzheimer's Network for Treatment and Diagnostics (ALZ-NET; see https://www.alz-net.org/). ALZ-NET is a network for providers who will collect longitudinal diagnostic and therapeutic clinical data, including cognitive and functional assessments, in real-world practice. It will provide a cohesive and integrated approach for high-quality data collection from various clinical practices, health records, and a broad patient population to fill gaps in knowledge regarding long-term health outcomes for individuals evaluated for and treated with novel AD therapies (Alzheimer's Association, 2023).
While current AD antiamyloid drugs hold significant promise in slowing the progression of disease, they are only one component of successfully halting neurodegenerative disease. It is well established that various processes, from amyloid and tau protein alterations to vascular diseases, α-synuclein modifications, and inflammatory processes, among others, contribute to AD pathology (De Strooper and Karran, 2016). Such complexity calls for diversifying AD drug development as a combination of therapies targeted at key underlying pathways may be more effective than monotherapy. Recent trends in the AD drug development pipeline suggest that 28% of the current AD clinical trials are primarily targeted at amyloid and tau pathology, while the remaining 72% are targeted at novel disease processes, with inflammation and synaptic plasticity being the most well represented ones (Cummings et al., 2023). Furthermore, growing evidence from epidemiological studies and clinical trials suggests that there is great promise in risk reduction strategies and healthy lifestyle habits to slow cognitive decline and reduce the risk of dementia (Ngandu et al., 2015; Livingston et al., 2020). The US POINTER study, led by the Alzheimer's Association, is the first study to be conducted in a large group of Americans evaluating whether lifestyle interventions that simultaneously target multiple risk factors can protect cognitive function in older adults at increased risk for cognitive decline. If the US POINTER Study proves effective, this study will lead the way in the development of an accessible and sustainable community-based program for prevention.
In summary, AD and other dementias continue to pose significant global burdens on public health, society, and economies. The AD drug development pipeline is more diverse than ever before, targeting novel pathways such as inflammation and synaptic plasticity. Given the complex pathology of AD, a combination of therapies and risk reduction strategies targeted at key underlying pathways will likely be the most effective approach toward meaningful and powerful therapeutic and prevention strategies.
Footnotes
The authors declare no competing financial interests. M.C.I. and J.W.L. are employees of Eisai Inc and Denali Therapeutics, respectively.
- Correspondence should be addressed to Takeshi Iwatsubo at iwatsubo{at}m.u-tokyo.ac.jp