Therapeutic time window of disease-modifying therapy for early Alzheimer's disease

Abstract

Introduction: Recently approved disease-modifying therapies (DMT) for early Alzheimer's disease (AD), including lecanemab and donanemab, require patients to meet specific eligibility criteria for treatment. These criteria define a limited "therapeutic time window," after which patients become ineligible as the disease advances. Understanding factors influencing this window may help clinicians optimize patient management and reduce lost treatment opportunities.

Methods: We analyzed longitudinal data from two observational cohorts, the National Alzheimer's Coordinating Center (NACC) and the Alzheimer's Disease Neuroimaging Initiative (ADNI). At each visit, individuals were deemed eligible if they were amyloid-positive and had a Mini-Mental State Examination (MMSE) score of 22-30 (lecanemab) or 20-30 (donanemab), plus a Clinical Dementia Rating-Global Score (CDR-GS) of 0.5 or 1. We then applied survival analyses and Cox proportional hazards models to estimate time-to-ineligibility based on baseline cognitive status.

Results: Across both datasets, higher baseline CDR-GS and MMSE were associated with a lower risk of becoming ineligible (pooled hazard ratio of 1.601 for CDR-GS of 1 vs. 0.5, and pooled hazard ratio of 0.660 per 1-point increase in MMSE score above the lower limit of eligibility). The estimated 75% survival time for patients with baseline CDR-GS 0.5 was over 12 months, suggesting only 25% would become ineligible within 12 months. For those with CDR-GS 1, the estimated 50% survival time was approximately 12 months, depending on the data, indicating that half might become ineligible within 1 year.

Discussion: We quantitatively outlined the duration of the therapeutic time window for early AD patients who qualify for lecanemab or donanemab, which is significantly influenced by baseline CDR-GS and MMSE scores. These findings will support more proactive patient management, ensuring timely evaluations and prioritization of patients at higher risk of ineligibility, particularly where DMT access is limited.

Highlights: We examined the "therapeutic time window" eligibility for disease-modifying therapy.Longitudinal data from National Alzheimer's Coordinating Center (NACC) and Alzheimer's Disease Neuroimaging Initiative (ADNI) were used to quantify eligibility duration.Higher Clinical Dementia Rating-Global Score (CDR-GS) or lower Mini-Mental State Examination (MMSE) at baseline were associated with shorter window length.Our results will help optimize the management of the wait time for disease-modifying therapies (DMT) treatment.

Keywords: ADNI; Alzheimer's disease; NACC; disease‐modifying therapy; therapeutic window.

FIGURE 1

Kaplan–Meier survival curves until becoming ineligible. Kaplan–Meier survival curves illustrating the time until participants became no longer eligible are presented, focusing on the first 24‐month timeframe. The figures are organized by data source (A–C for NACC and D–F for ADNI), drug type (A, D), and CDR‐GS at baseline (0.5 vs. 1.0, B, C, E, F). ADNI, Alzheimer's Disease Neuroimaging Initiative; CDR‐GS, Clinical Dementia Rating Global Score; MMSE, Mini‐Mental State Examination; NACC, National Alzheimer's Coordinating Center.

FIGURE 2

Cox‐proportional hazard results. The results of the Cox‐proportional hazard analysis for events within 24 months from baseline are shown with the logarithm of HR. While age, sex, CDR‐GS, and MMSE were commonly included in the models, variables in the shaded area are those imputed one by one. * means the significant change (i.e., 95% CI of ln(HR) not spanning 0). Poorer cognitive status at baseline (e.g., a smaller MMSE) was consistently associated with a shorter survival time, irrespective of dataset or target drug. In addition, tau‐status as represented in CSF or tau‐PET was associated with the longer eligibility‐survival only in the ADNI cases, while not in NACC cases. We could not consider the effect of taking existing symptomatic drugs (e.g., donepezil or memantine), because taking the medication is a time‐dependent factor (e.g., the medication can be initiated after the start of observation in some cases), it could not be included in the analysis. ADNI, Alzheimer's Disease Neuroimaging Initiative; CDR‐GS, Clinical Dementia Rating Global Score; CI, confidence interval; CSF, cerebrospinal fluid; HR, hazard ratio; MMSE, Mini‐Mental State Examination; NACC, National Alzheimer's Coordinating Center; PET, positron emission tomography.

FIGURE 3

Therapeutic window length by baseline MMSE and CDR‐GS. The eligibility time length as the 50% survival time by baseline CDR‐GS and MMSE (in x‐axis). Eligibility time length was left blank for some baseline MMSE where 50% survival time could not be estimated validly. We applied spline fit using a GLM with a Poisson distribution. For patients with a CDR‐GS of 0.5 at baseline, survival time increased in conjunction with the baseline MMSE score. Meanwhile, this trend was not consistently evident in patients with a CDR‐GS of 1 at baseline, where the fitted spline curves showed a slope closer to 0. The influence of MMSE score on the therapeutic time window for those with CDR‐GS of 1 was apparently lower than that for those with CDR‐GS of 0.5. Although its reasons remain uncertain, one possible way to explain this observation is that individuals at CDR‐GS of 1 have already reached a more advanced clinical stage, where disease progression tends to be faster overall. While a slightly higher MMSE might help preserve function a bit longer, it has a comparatively smaller impact than it would at CDR‐GS of 0.5, because the underlying neurodegenerative processes are already more pronounced. In other words, once someone has moved into the CDR‐GS of 1 category, subtle differences in MMSE do not translate into as large a difference in time to losing eligibility, whereas at CDR‐GS of 0.5, even a small increase in baseline MMSE can significantly delay progression to ineligibility. ADNI, Alzheimer's Disease Neuroimaging Initiative; CDR‐GS, Clinical Dementia Rating Global Score; CI, confidence interval; GLM, generalized linear model; MMSE, Mini‐mental State Examination; NACC, National Alzheimer's Coordinating Center.

FIGURE 4

Spline estimates of 25%, 50%, and 75% survival time. Spline estimates of 25%, 50%, and 75% survival time along with their 95% CIs. While eligibility time length distributes similarly by baseline MMSE regardless of dataset or the drug in patients with baseline CDR‐GS of 0.5, they differ by the dataset (ADNI vs. NACC) in patients with baseline CDR‐GS of 1. The estimated 75% survival time for patients with baseline CDR‐GS 0.5 was more than 12 months, suggesting only 25% of them will experience becoming ineligible as of 12 months. Meanwhile, the estimated 50% survival time for patients with baseline CDR‐GS 1 was approximately 12 months depending on the data, suggesting half of them might experience becoming ineligible within 1 year. ADNI, Alzheimer's Disease Neuroimaging Initiative; CDR‐GS, Clinical Dementia Rating Global Score; CI, confidence interval; MMSE, Mini‐Mental State Examination; NACC, National Alzheimer's Coordinating Center.