Commercial

ALZHEIMER DISEASE  ​

Alzheimer disease (AD) is a neurodegenerative disorder leading to progressive, irreversible destruction of neurons and loss of cognitive function and memory. Over time, individuals progress to severe dementia, loss of independence, and death. Extracellular deposits of amyloid beta, referred to as amyloid plaques, are considered a hallmark of the disease. Beta-amyloid monomers lead to formation of beta oligomers and fibrils and are deposited as plaques and then interact with tau fibrils, leading to formation of neurofibrillary tangles. These pathophysiological changes and clinical manifestations of AD are progressive and occur along a continuum, and accumulation of amyloid beta may begin 20 years or more before symptoms arise. 

ADUCANUMAB (ADUHELM)

Aducanumab (Aduhelm) is a human IgG1 antiamyloid beta antibody targeting amyloid aggregates. The drug is administered by intravenous (IV) infusion every 4 weeks. Binding of antibody is intended to lead to clearance of amyloid from the brain. On June 7, 2021, the US Food and Drug Administration (FDA) approved aducanumab (Aduhelm) for the treatment of AD. The drug was approved under accelerated approval based on reduction in amyloid beta plaques observed in individuals treated with aducanumab (Aduhelm). Continued approval for this indication may be contingent on verification of clinical benefit in a confirmatory trial.

RANDOMIZED CONTROLLED TRIALS (RCTs)

The evidence for aducanumab (Aduhelm) includes a dose-finding and proof of concept Phase 1 trial (PRIME) and two Phase 3 pivotal trials (ENGAGE [study 301] and EMERGE [study 302]). PRIME was a multicenter, randomized, double-blind, placebo-controlled, dose-ranging, staggered study conducted in the United States with the primary objectives of safety and tolerability. The Phase 3 studies were multicenter, global, randomized, double-blind, placebo-controlled studies of identical design with the primary objective of efficacy and safety. In all three studies, the diagnosis of AD was confirmed by presence of amyloid pathology measured by [18F]-florbetapir positron emission tomography (PET) imaging. The pivotal trials ensured enrollment of individuals at an earlier stage of their disease; mild cognitive impairment (MCI) due to AD or mild AD dementia based on an entry criteria of baseline Mini-Mental State Examination (MMSE) score of 24 to 30; baseline Clinical Dementia Rating (CDR) global score of 0.5, and Repeatable Battery for the Assessment of Neurological Status (RBANS) delayed memory index score less than or equal to 85. Per the protocol design, most participants had a diagnosis of MCI due to AD (81.6%), while 18.4% of participants had mild AD dementia. Approximately two thirds of the study population in the Phase 3 trials are apolipoprotein E (ApoE) ε4 carriers. The trial had approximately 90% power to detect a true mean difference of 0.5 in change from baseline Clinical Dementia Rating Scale–Sum of Boxes​ (CDR‐SB) at week 78. The range for CDR-SB is 0 to 18, with higher scores indicating greater disease severity.

The Phase 3 studies randomly assigned individuals to aducanumab (Aduhelm) low dose (3 or 6 mg/kg for ApoE ε4 carriers and noncarriers, respectively), aducanumab (Aduhelm) high dose (10 mg/kg), or placebo every 4 weeks for 18 months, followed by an optional, dose-blind, long-term extension period. Although aducanumab (Aduhelm) ​10 mg/kg was hypothesized to be the most efficacious dose, due to safety concerns and limited understanding of amyloid-related imaging abnormalities (ARIA), both studies included an initial titration period of up to 6 months to the maximum target dose. At the beginning of the study, ApoE ε4 carriers were initially titrated up to a maximum of 6 mg/kg in the high-dose group, which was later adjusted to 10 mg/kg. Both pivotal trials were terminated prior to their planned completion. Study endpoints were analyzed based on a prespecified statistical analysis plan. Because of the early termination and consequent administrative censoring, data were missing for up to 45% of individuals randomly assigned in the two trials. Approximately 60% of individuals had the opportunity to complete week 78 of the trial before the trials were terminated for futility. 

​

Study 302 (N=1638 randomly assigned individuals) met the primary endpoint in individuals treated with high-dose aducanumab (Aduhelm) with an absolute difference of −0.39 in favor of aducanumab (Aduhelm) on the 18-point CDR-SB scale (a relative 22% less decline in high-dose aducanumab [Aduhelm]​ group compared to placebo; P=0.0120). The reported minimal clinically important difference (MCID) is generally considered to be 1 to 2 points on a scale from 0 to 18. Results in the low-dose aducanumab (Aduhelm) group were not statistically significant compared with placebo (absolute difference −0.26, relative difference −15%; P=0.0901). The prespecified multiplicity adjustment protocol prioritized testing the low dose on the primary endpoint before testing secondary endpoints for the high dose. Therefore, the P values reported in this study should be considered nominal and no statistically valid conclusions can be made for any of the secondary endpoints for either of treatment arms. Results in the low-dose aducanumab (Aduhelm) group were not statistically significant compared with placebo (absolute difference, −0.26; relative difference, −15%; P=0.0901) and therefore no statistically valid conclusions can be made for any of the secondary endpoints for either of the treatment arms.

Study 301 (N=1647 randomly assigned individuals) did not meet its primary endpoint of a reduction relative to placebo in the CDR-SB score. For the high-dose arm, an absolute difference of 0.03 and a relative difference of 2% favored placebo (P=0.8330). For the low-dose arm, an absolute difference of −0.18 and a relative difference of 12% favored aducanumab (Aduhelm) (P=0.8330). Because of the pre-specified plans to control for type I error for multiple comparisons, no statistically valid conclusions can therefore be made for any of the secondary endpoints.

Results of pre-specified exploratory responder analysis were reported. Two thresholds for defining a responder were used: change from baseline in CDR-SB at week 78 was less than or equal to 0.5 or less than or equal to 1.5. Explanation for choosing these thresholds was not provided or whether thresholds represent important clinical meaningful change. All participants with missing data at week 78 were classified as nonresponders. In study 302, proportions of responders for placebo versus high dose at week 78 was 18.8% and 25.7%, respectively (using CDR-SB cutoff ≤0.5) and 32.2% and 39.1%, respectively (using CDR-SB cutoff ≤1.5). In study 301, proportions of responders for placebo versus high dose at week 78 was 25.7% and 20.2%, respectively (using CDR-SB cutoff ≤0.5) and 39.9% and 36.8%, respectively (using CDR-SB cutoff ≤1.5). These results should be considered exploratory and are not statistically robust and the statistical significance of the results in 302 is not robust to minor recategorization.

Change in brain amyloid signal was measured by PET and quantified by a composite standard uptake value ratio (SUVR) in a subset of sites and individuals (n=488) at week 78. In study 302, adjusted mean change from baseline to week 78 relative to placebo showed a dose-dependent reduction in amyloid beta by −0.179 and −0.278 in the low- and high-dose arms, respectively. In study 301, adjusted mean change from baseline to week 78 relative to placebo showed a dose-dependent reduction in amyloid beta by −0.167 and −0.232 in the low- and high-dose arms, respectively. Whereas aducanumab (Aduhelm) showed statistically significant dose-dependent changes from baseline in amyloid beta plaques, there are no satisfactory data, clearly establishing individual changes in amyloid correlate with or predict long-term cognitive and functional changes as measured by CDR-SB. The FDA statistical review reported no individual-level correlation in study 302 between reduction in amyloid plaque and long-term clinical change among the high-dose cohort or full 10 mg/kg dosed subgroup.

Change from baseline in markers of downstream AD tau pathophysiology and neurodegeneration were reported for a small subset of individuals collected from a voluntary nondirectly randomized sample (n=45 in study 302 and n=33 in study 301). Whereas the prescribing label reports a statistically significant lowering of both phosphorylated tau and total tau in the treatment arms, aducanumab (Aduhelm) is not known to directly target tau pathways. Therefore, it is difficult to clinically interpret the observed findings on an off-target exploratory biomarker from a small voluntary nondirectly randomized sample.

Safety

Data with limited follow-up are available to analyze safety because the Phase 3 trials were stopped prematurely due to futility. Pooled safety data from the two Phase 3 clinical trials showed that about 35% (compared to 3% in the placebo arm) of individuals on aducanumab (Aduhelm) experienced ARIA, whose clinical effects can range from asymptomatic to severe. Although the majority of individuals were asymptomatic or had symptoms such as headache, confusion, or dizziness that resolved with temporary stoppage of the drug, 6.2% of participants receiving the high dose of aducanumab (Aduhelm) discontinued the drug due to ARIA. The incidence of amyloid related imaging abnormalities-edema (ARIA-E) was higher in ApoE ε4 carriers than noncarriers (42% and 20%, respectively). The majority of ARIA-E radiographic events occurred early in treatment (within the first eight doses), although ARIA can occur at any time. Among individuals treated with a planned dose of aducanumab (Aduhelm) 10 mg/kg who had ARIA-E, the maximum radiographic severity was mild in 30%, moderate in 58%, and severe in 13% of individuals (refer to the prescribing label for classification of severity of ARIA). Resolution occurred in 68% of individuals with ARIA-E by 12 weeks, 91% by 20 weeks, and 98% overall after detection. Ten percent of all individuals who received aducanumab (Aduhelm) 10 mg/kg had more than one episode of ARIA-E.

An increase in falling adverse events was observed in the high-dose as compared to placebo across the two Phase 3 studies (15% vs. 12%, respectively). FDA statistical review reported a hazard ratio of 1.33 (P=0.016), suggesting a 33% relative increase in hazard of falling for the 10 mg/kg dose compared to placebo. A quantitative integration of benefit and risk was not done, but if the high dose increases the number of falls, it could be a significant risk for the AD population.

Outcomes

Data supporting individual-centric clinical and humanistic outcomes related to cognition (e.g., memory, orientation, judgment/problem-solving, ability to perform cognitive tasks, and everyday functioning) are not interpretable due to conflicting evidence from two identical Phase 3 RCTs. Study 302 met the primary endpoint of statistically significant change in CDR-SB score in the high-dose arm. Observed magnitude of effect (0.39 points in CDR-SB) is of uncertain clinical benefit. Study 301 failed to meet the same CDR-SB endpoint. In fact, the high-dose arm's change in CDR-SB score was numerically worse than placebo at 78 weeks. Aducanumab (Aduhelm) was approved on the basis of statistically significant dose-dependent changes in amyloid beta plaques. However, no correlation between reduction in amyloid plaque and change in CDR-SB score was observed in the 10 mg/kg dosed subgroup. Further, lowering of phosphorylated tau and total tau levels as supportive evidence in the biomarker framework is difficult to interpret as tau levels were an off-target biomarker and results were exploratory from a small voluntary nondirectly randomized sample.

Amyloid beta has not been established as a valid surrogate outcome measure to evaluate clinical benefit in individuals with MCI or mild dementia due to AD. To establish surrogacy, the relationship between treatment, a surrogate and health outcome(s) have to be established. In this case, to establish PET amyloid levels as a surrogate outcome, the following would be required: (1) preceding clinical trials demonstrate that the antiamyloid treatment mitigates cognitive decline; (2) the treatment effect on mitigation of cognitive decline from previous trials is mediated by reduction of amyloid beta levels; (3) the current antiamyloid treatment has an effect on amyloid beta levels; and (4) amyloid beta levels are associated with cognitive decline. Current evidence demonstrates that aducanumab (Aduhelm) results in a dose-dependent reduction in amyloid beta while the remaining relationships are not supported by the existing evidence.

Durability and External Validity

The intended double-blind duration of the two RCTs was 78 weeks followed by an 18-week safety follow-up period after the final dose. Because the trial was terminated early due to futility, the available data are limited. Because of the early termination and consequent administrative censoring, data were missing for up to 45% of individuals at week 78 in the trials. The average follow-up for ApoE ε4 carriers exposed to a full dose of 10 mg/kg was only 50 weeks rather than 78 weeks. Cognitive decline in MCI due to AD and mild AD generally occurs over years, and thus the follow-up duration may not be sufficient to conclude whether a drug is effective for this disease or whether the safety profile might change with longer follow-up. Further, a statistically significant difference was only reported at week 78 and not any other earlier timepoints. Pooled safety data showed that about 35% of individuals on aducanumab (Aduhelm) experienced ARIA as well as an increase in the risk of falling. Although ARIA was detected early by frequent magnetic resonance imaging (MRI) monitoring in the clinical trials, it may be challenging to implement routine monitoring in a real-world setting, particularly when it involves individuals older than the trial participants. Thus, ARIA may pose greater risks to individuals who may be older, have more comorbidities, and are less carefully monitored outside of clinical trials.

Of 3285 individuals enrolled, fewer than 1% were Black or African American and 3.2% were Hispanic or Latino. Additionally, the average age was 70 years old, although trials allowed for enrollment up to 85 years of age. Given that older African Americans and Latinos are disproportionately more likely to have AD than White Americans, and more than one third of individuals with AD in the United States are over the age of 85, there is limited generalizability of these results to the broader US population.

Study Conduct

Pivotal trial protocols minimized functional blinding by mandating use of an independent rater who was blinded to individual management (including occurrence of ARIA and subsequent monitoring). However, individuals and caregivers could become aware of the occurrence of ARIA due to differential management including additional MRIs and dose modification. The CDR-SB and Activities of Daily Living Scale for use in Mild Cognitive Impairment (ADCS‐ADL‐MCI) rating scales require more individual and caregiver input and could therefore be susceptible to biased estimates if respondents knew they were on therapy. Further, differential rates of ARIA between study 301 and 302 could have contributed to discordant results because of the impact of differential functional unblinding in the three studies.​

LECANEMAB-IRMB (LEQEMBI)

Lecanemab-irmb (Leqembi) is the second amyloid beta–directed antibody indicated for the treatment of AD and should only be initiated in individual​s with MCI or mild dementia stage of disease. There are no safety or effectiveness data on initiating treatment at earlier or later stages of the disease than were studied. This indication was approved under accelerated approval based on reduction in amyloid beta plaques, and continued approval may be contingent on clinical benefit verification in a confirmatory trial. The efficacy of Leqembi was evaluated in one Phase 2b study (Study 1) and one Phase 3 study (CLARITY AD). Results from Study 1 demonstrated a reduction in amyloid beta plaque, which served as the basis for FDA approval under the accelerated approval pathway. On January 6th, 2023, it was announced that data from the confirmatory CLARITY AD trial were submitted to the FDA as part of a supplemental Biologics License Application (sBLA) to convert the accelerated approval to a traditional approval. The FDA has yet to announce when a decision on traditional approval can be expected. 

STUDY 1

Methods 

Individuals 50 to 90 years of age were eligible to participate if they met criteria for MCI due to AD or mild AD dementia and had confirmed evidence of beta-amyloid by PET scan or cerebrospinal fluid (CSF) beta-amyloid 1-42. Additionally, individuals had to have objective impairment in episodic memory on Wechsler Memory Scale-IV Logical Memory II, MMSE score greater than or equal to 22. Individuals were also required to have a CDR global score of 0.5 or 1.0 with a Memory Box score of greater than or equal to 0.5. Individuals were excluded from the study for any neurologic condition (other than AD), history of transient ischemic attacks, stroke, or seizures, or significant pathological findings on brain MRI. The study included a 2-month screening period, an 18-month (78-week) placebo-controlled treatment period, and a safety follow-up period of 3 months after the final dose. During the placebo-controlled treatment period, individuals were randomly assigned to placebo or one of five lecanemab-​irmb (Leqembi) dosing regimens: three arms with biweekly (once every 2 weeks) dosing (2.5, 5, and 10 mg/kg) and two arms with monthly (once every 4 weeks) dosing (5 and 10 mg/kg). Of note, during the study, the protocol was amended so that ApoE ε4 carriers were no longer randomly assigned to the 10 mg/kg every-2-weeks dose arm. ApoE ε4 carriers who had been receiving lecanemab-​irmb (Leqembi) 10 mg/kg every 2 weeks for 6 months or less were discontinued from the study drug. This decision was based on data indicating that ApoE ε4 homozygotes on the highest dose of lecanemab-​irmb (Leqembi) (10 mg/kg once every 2 weeks) had the highest risk of developing symptomatic ARIA-E. The primary endpoint was change from baseline at 12 months on Alzheimer’s Disease Composite Score (ADCOMS). Secondary endpoints included change from baseline at 18 months in brain amyloid by PET Standard Uptake Value ratio (SUVr), score on the ADCOMS, CDR-SB, and ADASCog14. 

Results 

Lecanemab-​irmb (Leqembi) 10 mg/kg once every 2 weeks was identified as the target dose. Lecanemab-​irmb (Leqembi) reached a 64% probability of being better than placebo with 25% less decline at 12 months in the ADCOMS, missing the pre-specified 80% probability threshold. Therefore, the primary endpoint was not met. A total of 315 individuals were enrolled in the amyloid PET substudy and of those, 277 had results evaluated at week 79. Lecanemab-​irmb (Leqembi) 10 mg/kg once every 2 weeks had a statistically significant decrease in brain amyloid plaque as measured by PET quantified by a composite SUVr when compared to placebo at Week 79 (−0.310; P<0.001). The FDA concluded, based on the surrogate endpoint of reduction in amyloid plaque burden, that there were data to support the accelerated approval of lecanemab-​irmb (Leqembi).

​CLARITY AD

Methods 

Eligible individuals were aged 50 to 90 years with MCI or mild AD dementia with confirmed evidence of beta amyloid by PET scan or CSF. Inclusion and exclusion criteria were similar to Study 1. The trial design included a screening period, followed by an 18-month (78-week) placebo-controlled treatment period, and a safety follow-up period of 3 months. Individuals were randomly assigned in a 1:1 fashion to either lecanemab-​irmb (Leqembi) 10 mg/kg IV once every 2 weeks or placebo for 18 months. The primary efficacy endpoint was change from baseline at Week 78 on the CDR-SB scale; higher scores indicate greater impairment. Secondary endpoints included change from baseline at 18 months in amyloid on PET scan, score on the ADAS-Cog14, ADCOMS, and ADCS-ADL-MCI. 

Results 

Of the 1795 participants, 898 received lecanemab-​irmb (Leqembi) and 897 received placebo. Baseline characteristics were similar between the two groups. Of note, 31% of participants were ApoE ε4 noncarriers. At 18 months, the adjusted mean change from baseline in CDR-SB score was 1.21 in the lecanemab-​irmb (Leqembi) arm and 1.66 in the placebo arm (treatment difference of −0.45; 95% confidence interval [Cl], −0.67 to −0.23; P<0.001). The substudy of amyloid burden on PET included 698 participants and found the adjusted mean change from baseline at 18 months was −55.48 centiloids in the lecanemab-​irmb (Leqembi) arm and 3.64 centiloids in the placebo arm (treatment difference, −59.12 centiloids; 95% CI, −62.64 to −55.60; P<0.001). Although lecanemab-​irmb (Leqembi) safety is similar to that of aducanumab (Aduhelm), lecanemab-​irmb (Leqembi) carries a warning regarding ARIA. In Study 1, symptomatic ARIA occurred in 3% of individuals treated with lecanemab-​irmb (Leqembi). Clinical symptoms resolved in 80% of individuals during the observation period. When including asymptomatic radiographic events, 12% of lecanemab-​irmb (Leqembi)–​treated individuals versus 5% of individuals on placebo had observed ARIA. ARIA-E was seen in 10% of lecanemab-​irmb (Leqembi) individuals versus 1% in the placebo group. ARIA-H was seen in 6% and 5% of individuals treated with lecanemab-​irmb (Leqembi) and placebo, respectively. ARIA was higher in lecanemab-​irmb (Leqembi)–​treated individuals who were ApoE ε4 homozygotes when compared to heterozygotes and noncarriers. Lecanemab-​irmb (Leqembi) prescribing information gives consideration for testing ApoE ε4 status to gauge risk of developing ARIA before starting lecanemab-​irmb (Leqembi) treatment. Additionally, the prescribing information outlines monitoring and dosing interruption protocols for ARIA. Lecanemab-​irmb (Leqembi) can cause infusion-related reactions and may require infusion rate reductions or premedication at subsequent dosing. The most common adverse reactions in individuals treated with lecanemab-​irmb (Leqembi) (incidence approximately 10% and a higher rate compared to placebo): infusion-related reactions (20%), headache (14%), and ARIA-E (10%).

It needs to be highlighted that uncertainty remains around the amyloid hypothesis and that adequate data for lecanemab-irmb (Leqembi)​ are unavailable to show a correlation between amyloid removal and treatment effect, or differences in outcomes by achieving or not achieving amyloid negativity. A substantial percentage of individuals reached amyloid negativity (by PET scan) in the Phase 3 trial, but 7.8% of individuals in the placebo arm also achieved amyloid negativity.

It can be suggested that this finding demonstrates the complexity of AD pathophysiology and that the role of amyloid in AD and factors that may impact clinical outcomes are not fully understood.

 

It should be noted that the ARIA risk with real-world use may be greater than that in clinical trials due to issues like limited clinical expertise and accessibility issues affecting monitoring. Additionally, there are questions of whether the trial results can be generalized to the broader mild AD population, as the average age of participants in the Phase 3 trial was just over 71 years of age and included participants with some comorbidities; however, two thirds of the AD population in the United States are 75 years of age and older and likely have significant comorbidities.

It also needs to be acknowledged that there is a disagreement among experts about clinical meaningfulness of the magnitude of change in the cognitive outcomes of these trials; despite the demonstrated statistical significance of the reduction in cognitive decline, there is no certainty that treatment with lecanemab-​irmb (Leqembi) will yield a meaningful change in the individual's status for the benefit to outweigh the risk to treatment.

 

Based on the current information available, there is insufficient evidence that lecanemab-​irmb (Leqembi) provides a meaningful clinical benefit in individuals with AD, and that potential benefits of therapy outweigh the risks of treatment. Therefore, demonstration of a clinical benefit is warranted in ongoing clinical trials.

DONANEMAB-AZBT (KISUNLA)

The efficacy of donanemab-azbt (Kisunla) for traditional approval was evaluated in one Phase III randomized, double-blind, placebo-controlled, multicenter, pivotal study (TRAILBLAZER-ALZ2) in individuals with mild cognitive impairment due to AD and mild AD dementia (n = 1736). The primary efficacy endpoint was the change from baseline in the integrated Alzheimer's Disease Rating Scale (iADRS) at 76 weeks, an assessment of cognition and daily function with scores ranging from 0 to 144 (lower scores indicate greater impairment). A key secondary endpoint included the change from baseline at 76 weeks in the Clinical Dementia Rating Scale–sum of boxes (CDR-SB), also an assessment of cognition and daily function with scores ranging from 0 to 18 (higher scores indicate greater impairment). For the low/medium tau population, the least-squares mean (LSM) change from baseline at Week 76 in the iADRS score was −6.02 in the Kisunla arm and −9.27 in the placebo arm (treatment difference 3.25; P<0.001).

In the combined (low/medium and high tau) population, the LSM change from baseline at Week 76 in the iADRS score was −10.19 in the donanemab-azbt (Kisunla) arm and −13.11 in the placebo arm (treatment difference 2.92; P<0.001). In the low/medium tau population, the placebo-adjusted LSM change from baseline at 76 weeks for CDR-SB was −0.67, and in the combined population, the placebo-adjusted LSM change from baseline at 76 weeks for CDR-SB was −0.70. However, this slowing of progression did not achieve clinical significance. The authors of TRAILBLAZER-ALZ2 note that the minimal clinically important difference for the iADRS is a change of five points for those with AD with mild cognitive impairment and nine points for those with AD with mild dementia, and it is one to two points for the CDR-SB.

Additionally, one Phase II, randomized, double-blind, placebo-controlled, multicenter study (TRAILBLAZER-ALZ) was conducted in individuals with mild cognitive impairment due to AD and mild AD dementia (n = 257). The change from baseline in the iADRS score at 76 weeks was −6.86 in the donanemab-azbt (Kisunla) arm and −10.06 in the placebo arm (treatment difference 3.20; P=0.04). The placebo-adjusted change from baseline at 76 weeks for the CDRSB score was −0.36 and failed to show a significant difference between the two trial groups.

Donanemab-azbt (Kisunla) can cause ARIA-E and amyloid related imaging abnormalities-hemosiderin deposition (ARIA-H), which includes microhemorrhage and superficial siderosis, which can be observed on MRI. A recent (within 1 year) MRI of the brain should be obtained prior to initiating treatment with Kisunla. The safety of donanemab-azbt (Kisunla) has not been evaluated in individuals with prior cerebral hemorrhage larger than 1 cm in greatest diameter, more than four microhemorrhages, more than one area of superficial siderosis, severe white matter disease, and vasogenic edema. Enhanced clinical vigilance for asymptomatic ARIA is recommended during the first four doses of treatment with donanemab-azbt (Kisunla), particularly during titration, because the majority of ARIA was observed during this time. MRIs of the brain should be obtained prior to the second, third, fourth, and seventh infusions of donanemab-azbt (Kisunla) to evaluate for the presence of asymptomatic ARIA. In addition to ARIA, intracerebral hemorrhages larger than 1 cm in diameter have occurred in individuals treated with donanemab-azbt (Kisunla). Symptomatic ARIA occurred in 6% of individuals treated with donanemab-azbt (Kisunla) (n = 52/853) in the pivotal trial, and clinical symptoms associated with ARIA resolved in approximately 85% of affected individuals (n = 44/52). Including asymptomatic radiographic events, ARIA was observed in 36% of individuals treated with donanemab-azbt (Kisunla) versus 14% of individuals treated with placebo in the pivotal trial. ARIA-E and ARIA-H were observed in 24% and 31% of individuals treated with donanemab-azbt (Kisunla) versus 2% and 13% of individuals receiving placebo.

SUMMARY

​ADUCANUMAB (ADUHELM)

The evidence for aducanumab (Aduhelm) includes two RCTs and one dose-finding and proof-of-concept Phase 1 trial. These clinical investigations were for individuals with early AD (MCI or mild dementia due to AD). Outcomes of interest to establish safety and effectiveness include​ disease-specific survival, change in disease status, functional outcomes, health status measures, quality of life, and treatment-related mortality and morbidity. ENGAGE (study 301) and EMERGE (study 302) were identical, randomized, double-blind, placebo-controlled studies that enrolled individuals with early AD. The majority of individuals had a diagnosis of MCI due to AD (81.6%) and approximately two thirds were ApoE ε4 carriers. The primary clinical outcome was change in mean score on the CDR-SB. Both trials were terminated early following a prespecified interim analysis for futility. In study 301, there was no treatment benefit observed in either the high- or low-dose arms at week 78. In study 302, a statistically significant difference in change from baseline in CDR-SB was observed in the high-dose arm (difference vs. placebo −0.39 [95% CI, −0.69 to −0.09]) but not the low-dose arm at week 78. The observed change of 0.39 was well below the range of 01 to 02 points reported as the MCID in published literature. Approval by the FDA was based on the reduction in amyloid beta plaques, which was observed in both trials and at all doses. However, there are no satisfactory data clearly establishing that individual changes in amyloid correlate with or predict long-term cognitive and functional changes. In the absence of clinical data convincingly demonstrating a clinical effect, it cannot be concluded that the observed reduction in amyloid will translate into a clinical benefit to individuals. Cognitive decline in early AD generally occurs over years, and thus the follow-up duration may not be sufficient to conclude whether a drug is effective for this disease or whether the safety profile might change with longer follow-up. Pooled safety data showed that about 35% of individual​s on aducanumab (Aduhelm) experienced ARIA as well as an increase in the risk of falling. A confirmatory, prospective, and adequately powered trial is necessary to assess the net health benefit of aducanumab (Aduhelm) in individuals with early AD. There is no consistent evidence of a clinically meaningful improvement in the net health outcome and there are concerns with the radiological findings.

LECANEMAB-IRMB (LEQEMBI)

van Dyck et al. (2023) conducted an 18-month, multicenter, double-blind, placebo-controlled, parallel-group, phase 3 trial evaluating lecanemab-​irmb (Leqembi) in persons with early AD. The study (Clarity AD; NCT03887455) included persons 50 to 90 years of age with MCI or mild dementia due to AD with evidence of amyloid on PET or by CSF testing. Participants were randomly assigned in a 1:1 ratio to receive IV lecanemab-​irmb (Leqembi) (10 mg per kilogram of body weight every 2 weeks) or placebo. The primary end point was the change from baseline at 18 months in the score on the Clinical Dementia Rating-Sum of Boxes (CDR-SB; range, 0–18, with higher scores indicating greater impairment). Key secondary end points were the change in amyloid burden on PET, the score on the 14-item cognitive subscale of the Alzheimer's Disease Assessment Scale (ADAS-cog14; range, 0–90; higher scores indicate greater impairment), the Alzheimer's Disease Composite Score (ADCOMS; range, 0–1.97; higher scores indicate greater impairment), and the score on the Alzheimer's Disease Cooperative Study-Activities of Daily Living Scale for Mild Cognitive Impairment (ADCS-MCI-ADL; range, 0–53; lower scores indicate greater impairment). A total of 1795 participants were enrolled, with 898 assigned to receive lecanemab-​irmb (Leqembi) and 897 to receive placebo. The authors found that the mean CDR-SB score at baseline was approximately 3.2 in both groups. The adjusted least-squares mean change from baseline at 18 months was 1.21 with lecanemab (Leqembi)​ and 1.66 with placebo (difference, –​0.45; 95% confidence interval [CI], −0.67 to −0.23; P<0.001). In a substudy involving 698 participants, there were greater reductions in brain amyloid burden with lecanemab (Leqembi) than with placebo (difference, 

−​59.1 centiloids; 95% CI, −62.6 to −55.6). Other mean differences between the two groups in the change from baseline favoring lecanemab (Leqembi) were as follows: for the ADAS-cog14 score, −1.44 (95% CI, −2.27 to −0.61; P<0.001); for the ADCOMS, −0.050 (95% CI, −0.074 to −0.027; P<0.001); and for the ADCS-MCI-ADL score, 2.0 (95% CI, 1.2–2.8; P<0.001). Lecanemab (Leqembi) resulted in infusion-related reactions in 26.4% of the participants and ARIA-E in 12.6%. The incidence of amyloid-related imaging abnormalities-hemosiderin deposition (ARIA-H) was 17.3%. The incidence of ARIA, including symptomatic ARIA, was numerically lower than in similar clinical trials, but differences in the drugs used and in trial design do not allow direct comparisons. ARIA-E generally occurred in the first 3 months, was mild and asymptomatic, did not lead to discontinuation of lecanemab (Leqembi) or placebo if mild, and resolved within 4 months. The authors note that the incidences of both overall and symptomatic ARIA-E were highest among ApoE ε4 homozygotes. The authors concluded that lecanemab (Leqembi) reduced markers of amyloid in early AD and resulted in moderately less decline on measures of cognition and function than placebo at 18 months but was associated with adverse events. The authors acknowledge study limitations, which include data for only 18 months of treatment. The authors state that longer trials are warranted to determine the efficacy and safety of lecanemab (Leqembi) in early AD. Additional trials of lecanemab (Leqembi) include a 5-year phase 2 long-term extension trial (NCT01767311) and a 4-year phase 3 long-term extension trial (NCT03887455) in early AD, the 4-year AHEAD 3-45 trial (NCT04468659) in preclinical AD, and the 4-year DIAN-TU (Dominantly Inherited Alzheimer Network Trials Unit) Next Generation trial (NCT05269394) in dominantly inherited AD.

The manufacturer of lecanemab (Leqembi) submitted the New Drug Application (NDA) based on data from a Phase 2 and 3 randomized, placebo-controlled study (Clarity AD). lecanemab (Leqembi) was initially FDA approved through an accelerated program, and continued approval was contingent on verification of clinical benefit via a confirmatory trial. In July 2023, the FDA changed this approval to a traditional/standard approval. A published Phase 2 study did not meet the primary efficacy endpoint of change from baseline in the ADCOMS. The key secondary endpoint measuring change on the CDR-SB was also not met. With regard to the safety results, amyloid-related imaging abnormalities-edema (ARIA-E) incidence was 10% at the highest doses for the overall population and 14.3% for ApoE ε4–positive subjects (Swanson et al. 2021). The published Phase 3 study met its primary endpoint measuring CDR-SB, but clinical significance regarding delay in dementia progression (−0.45 of 18) was not evident because a minimum change of 1 point is considered clinically significant. Safety results showed a 22% total incidence of ARIA with lecanemab (Leqembi) compared to 10% in the placebo group. There were no deaths considered related to lecanemab (Leqembi) during the double-blind portion of the study. However, two deaths during the open-label phase are considered related to lecanemab (Leqembi) by study personnel and being investigated by the manufacturer. Long-term efficacy and safety of lecanemab (Leqembi) are unknown. ARIA can be observed on MRI as brain edema or sulcal effusions (ARIA-E) or microhemorrhage and superficial siderosis (ARIA-H). ​

In Study 1, lecanemab (Leqembi) demonstrated statistically significant reduction in amyloid plaque burden but did not achieve the primary efficacy endpoint. Although CLARITY AD demonstrated a statistically significant reduction of worsening in the CDR-SB score compared with placebo, the absolute difference of 0.45 is less than MCID of 1 to 2 points cited in the literature. Based on the currently available literature for lecanemab (Leqembi), there are insufficient data to establish clinically meaningful effectiveness that leads to improved health outcomes for the treatment of AD. Additionally, when taking into consideration the higher incidence of ARIA compared to placebo, there is insufficient evidence to suggest that the clinical benefit of lecanemab (Leqembi)​ outweighs the potential risks of therapy. 

DONANEMAB-AZBT (KISUNLA)

The current donanemab-azbt (Kisunla) efficacy information is insufficient to determine if the medication demonstrates any clinically meaningful benefits. In the absence of additional clinical trials, there is not enough information to support approval.

With the approval of donanemab-azbt (Kisunla), we now have two Phase III clinical trials of anti-amyloid therapies (i.e., Clarity AD and TRAILBLAZER-ALZ-2) that have demonstrated an association between amyloid clearance and slowing of cognitive decline in the early AD population; however, the available evidence, particularly in light of the history of mixed results with previously trialed and failed anti-amyloid therapies in the pipeline, does not unequivocally suggest that amyloid clearance will definitively improve cognitive outcomes. It also remains unclear whether the modest effect on cognition and function demonstrated in the trial will be clinically relevant when used in the real world. There is disagreement among experts about clinical meaningfulness of the magnitude of change in the cognitive outcomes of these trials, and despite the demonstrated statistical significance of the reduction in cognitive decline, we cannot say with certainty that treatment with donanemab-azbt (Kisunla) and other anti-amyloid therapies will yield a meaningful change in the individual's status for the benefit to outweigh the risk to treatment. Ultimately, uncertainty remains around the amyloid hypothesis and we do not have adequate data for anti-amyloid therapies to show a correlation between amyloid removal and treatment effect, or differences in outcomes by achieving or not achieving amyloid negativity.

Any cognitive benefits seen with donanemab-azbt (Kisunla) and other anti-amyloid therapies must be weighed against the potential harms of treatment, especially in light of the risk of ARIA-E and/or cerebral microhemorrhage. We must also recognize that ARIA risk with real world use may be greater than that in clinical trials due to issues like limited clinical expertise and accessibility issues affecting monitoring. ARIA can occur spontaneously in individuals with AD, with three main risk factors being exposure to anti-amyloid treatment, presence of pre-treatment microhemorrhages, and APOE ε4 carrier status. According to the labeling, testing for ApoE ε4 status is recommended prior to initiating treatment to inform the risk of ARIA development, although it is not a requirement. An FDA-authorized test to detect ApoE ε4 alleles to identify those at risk of ARIA is not currently available, and those that are available may vary in accuracy and design.

The cognitive decline associated with MCI and mild AD dementia often spans many years. The limited follow-up duration of the donanemab-azbt (Kisunla) Phase III trial may be insufficient to conclude how effective donanemab-azbt (Kisunla) is for treating early AD in the long term. Ultimately, longer trials are warranted to determine the true efficacy and safety of donanemab-azbt (Kisunla) in early AD. With the lack of information surrounding long-term use, safety, and the real-world effects of donanemab-azbt (Kisunla), a number of questions arise including the appropriate duration of treatment, if or at what point effectiveness will start to decline, and how discontinuation of donanemab-azbt (Kisunla) upon reaching undetectable levels of amyloid beta plaque will affect the trajectory of AD progression.​