Expanding the Drug Arsenal

 

A lot of things go bioloically wrong in an Alzheimer's patient, but reverse any one of those breakdowns and you might reverse the disease too. Here are a few new drugs being marketed to treat Alzheimer's and their results so far to help you make an educated decision.
A Side-by-side Comparison of Donepezil (Aricept) to Lecanemab (Leqembi)

 

Here’s a summary of Figure 1 from Espay et al.’s “Lecanemab and Donanemab as Therapies for Alzheimer’s Disease: An Illustrated Perspective on the Data”.

 

Purpose of the Figure: Comparing Donepezil and Lecanemab: Efficacy Slopes & Absolute Change
The figure plots the change on the ADAS‑Cog (0–90 scale; higher = worse) for two cognitive interventions versus placebo — donepezil (a standard symptomatic therapy) and lecanemab (an anti-amyloid disease‑modifying monoclonal antibody). The Y-axis scale is identical for both to enable direct visual comparison

Donepezil:

Improvement over time versus baseline, demonstrating an actual symptomatic benefit: cognition stabilizes or slightly improves during the 24-week trial.

 

The minimum clinically important difference (MCID) on ADAS‑Cog is ~3 points, and donepezil achieved improvements near that threshold.

 

 

 

Lecanemab:

No cognitive improvement—both treatment and placebo groups declined over the ~18 months.

 

The divergence between curves indicates slower decline in the lecanemab group, but the absolute difference (~0.5 points on ADAS‑Cog equivalent via CDR‑SB) is modest and well below the 3-point MCID threshold.

 

 

 

Bottom line:

In recent clinical trials fo the mAb, lecanemab, and donanemab, after 18 months of biweekly infusions, the difference between the curves of people on antibody and those on the placebo differed by only half the difference found in the donepezil trial after 6 months. For all participants, in both the treatment and placebo arms, cognition declined or worsened over the 72 week period.

 

It is often argued that the magnitude of the difference at the end of a clincal trial is less relevant than the situation years later. The hope is the separation between the placebo and drug will continue to grow. While one may hypothesize that continued diverenge will continue between the treatment and placebo arms, in the absense of follow-up studies, evidence for this hope is lacking. Unfortunately, long-term divergence between treatment and control never materialized for donepezil.

 

 
 
Overview: Two Anti-Amyloid Approaches:

The pivotal trials of Lecanemab (Leqembi) and donanemab (Kisunla)

 

Lecanemab (Leqembi) and Donanemab (Kisunla) are monoclonal antibodies targeting amyloid‑ß plaques in early Alzheimer's disease (mild cognitive impairment or mild dementia). These are disease‑modifying treatments approved in the U.S. (July 2023 for lecanemab; July 2024 for donanemab), with varying regulatory status in Europe and other regions.
 
Trial Designs & Populations

Clarity‑AD (Lecanemab): Randomized, placebo‑controlled phase III with ~1,795 participants, treated for 18 months at 10 mg/kg biweekly

 

TRAILBLAZER‑ALZ 2 (Donanemab): Phase III in ~1,736 subjects; initial dosing ramp‑up (700 mg ×3 then 1400 mg monthly) until predefined amyloid clearance; treated up to ~76 weeks 

 

Participants were amyloid‑positive and diagnosed with early Alzheimer's (MCI or mild dementia). Subgroup analyses considered APOE ε4 status, age bands, sex, tau burden, et

 

 

ON EFFICACY

Lecanemab (Figure 2 below) is repoted to have slowed decline by 27%. This figure was calculated using the Clnical Dementia Rating Scale Sum of Boxes (CDR-SB), an 18-point scale that measures cognitive function (higher is worse). The average patient began with a score of 3.2, the equivalent of mild cognitive impairment. After 18 months, the scores increased (worsened), with 0.45 points less decline with lecanemab. For start, it should be noted that 0.45 is arguable less than half the change patient will be able to perceive (Lansdall et al, 2023). In addition, the percentage is expressed as the absolute difference in scores compared to the change in the placebo group, which makes the percent change misleading in how big of an impact lecanemab had. 

 

 

If we use the calculations of Sims et al, donanemab (Figure 3 below) slowed decline by 33% in the ideal subgroup of low/medium tau participants. The trial used as outcomes measures the iADRS, a 144 point composite scale measuring cognition plus activities of daily living (lower is worse). The average participant began with a score of 104 (mild dementia). After 76 weeks, the scores decreased 3.4 points less in the donanemab arm.  Here again, the percentage difference sounds larger than it really is because it is being compared relative to the placebo group.

 

However one looks at the data, the average chances of ~3 on the iADRS and ~0.5 on the CDR-SB represent changes imperceptible to patients, their families and their physicians.

 

 

ON SAFETY

With a disease as devastating as AD, even a small effect size should be considered by clinicians if treatments were safe. However, these mAb are not safe. In both trials, adverse events afflicted sizable numbers of participants. With lecanemab, 45% of the participants had treatment-related adverse events, with nearly one in four patients developing brain swelling and/or bleeding, which proved to be severe in some persons. Severe bleeding occurred to a greater extent compared with placebo (five vs one in the lecanemab trial; seven vs two in the donanemab trial), including three fatal cases. With donanemab, 89% of patients had treatment-related adverse events and more than one in three patients developed brain swelling and/or bleeding. Brain swelling is a major contributor to the acceleration of brain atrophy, a feature of most mAb (Alves et al., 2023).

 

In evaluating a therapy, two useful numbers are “the number needed to harm” (NNH) and “the number needed to treat” (NNT). The NNH is the number of patients that need to receive the treatment for one to experience an adverse outcome (Andrade, 2015). When accounting for all toxic effects, that number is ∼3 for both lecanemab and donanemab. The NNT is the number of patients that need to receive the treatment for one to experience a beneficial response. As discussed above, no individual patient-level data was reported during these trials, making this number impossible to accurately calculate. Thus, authors have marked the NNT with a “?” sign in Figures 2 and 3. 

 

 

BOTTOM LINE:

The new mAb do not improve cognition but rather reduce the decline by an arguably smaller magnitude than donepezil and below the individual threshold of perception for most patients, carrying a worrisome safety profile. The advantage of these drugs seems to be related more to the theoretical possibility that they are disease modifying (in keeping with the prevailing framework of amyloid toxicity), rather than in their actual risk/cost–benefit trade-off. Before these expensive drugs are administered broadly, both the safety profile of subgroups and the efficacy relative to the benchmark drug donepezil should ideally be tested. If this does not happen, we risk adopting a potentially less effective drug with significant potential harm over the benchmark drug on the market, donepezil. Excess deaths occur with mAb infusions along with frequent brain swelling and bleeding. This concern is all the more acute given the acceleration of brain volume loss they induce (Alves et al., 2023).

 

 

Final Thoughts

While lecanemab and donaemab offer a measurable slowing of decline and clear biomarker effects, the modest benefit, substantial risk profile, and high costs mean they are not universal game-changers.Drug companies are not given up on these drugs, and are now studying their effects in even earlier stages of Alziember's disease coupled with lifestyle interventions. Also since the costs of these therapies is ~$25,000 a year, a substantial impact on our healthcare networks would be unavoidable if even only 10% of at-risk patients were to be subjected to these infusions.