# paper.pdf ## Metadata - PDFFormatVersion=1.6 - Language=en - IsLinearized=true - IsAcroFormPresent=false - IsXFAPresent=false - IsCollectionPresent=false - IsSignaturesPresent=false - CreationDate=D:20260421095519+05'30' - Creator=LaTeX with hyperref package - ModDate=D:20260421095544+05'30' - Producer=Acrobat Distiller 26.0 (Windows) - Title=Divergent patterns of cognitive decline in preclinical Alzheimer's disease: Implications for secondary prevention trials - Custom.WPS-ARTICLEDOI=10.1002/alz.71366 - Custom.WPS-JOURNALDOI=10.1002/(ISSN)1552-5279 - Custom.WPS-PROCLEVEL=3 - xmp:createdate=2015-08-18T08:05:36+05:30 - xmp:modifydate=2016-06-21T14:17:29-04:00 - xmp:metadatadate=2016-06-21T14:17:29-04:00 - pdf:producer=Acrobat Distiller 9.0 (Windows) - pdfx:wps-articledoi=10.1002/alz.71366 - pdfx:wps-journaldoi=10.1002/(ISSN)1552-5279 - pdfx:wps-proclevel=3 - jav:journal_article_version=VoR - prism:doi=10.1002/alz.71366 - dc:format=application/pdf - dc:title=Divergent patterns of cognitive decline in preclinical Alzheimer's disease: Implications for secondary prevention trials - xmpmm:documentid=uuid:29a731ae-3a49-3af2-b81c-6b28db02684c - xmpmm:instanceid=uuid:5fe1ab79-dc2d-3e14-a6f3-86bb438dd7c7 ## Contents ### Page 1 Received: 26 January 2026 Revised: 5 March 2026 Accepted: 11 March 2026 DOI: 10.1002/alz.71366R E S E A R C H A R T I C L E Divergent patterns of cognitive decline in preclinical Alzheimer’s disease: Implications for secondary prevention trials Runpeng Li1 Oliver Langford1 Philip S. Insel2 Reisa A. Sperling3 Rema Raman1 Paul S. Aisen1 Michael C. Donohue11 USC Epstein Family Alzheimer’s Therapeutic Research Institute, University of Southern California, San Diego, California, USA 2 Department of Psychiatry, University of California, San Francisco, California, USA 3 Center for Alzheimer Research and Treatment, Brigham and Women’s Hospital, Massachusetts General Hospital, Harvard Medical School, Boston, Massachusetts, USACorrespondenceMichael C. Donohue, USC Epstein Family Alzheimer’s Therapeutic Research Institute, University of Southern California, 9860 Mesa Rim Rd, San Diego, CA 92121, San Diego, CA, USA. Email: mdonohue@usc.eduFunding informationEpstein Family Foundation; Avid Radiopharmaceuticals; GHR Foundation; Foundation for the National Institutes of Health; Eli Lilly and Company; Alzheimer’s Association; National Institute on Aging, Grant/Award Number: U24AG057437Abstract INTRODUCTION: Biomarkers identify Alzheimer’s disease pathology in cognitively unimpaired adults, but the timing and rate of cognitive decline vary widely. This study aimed to identify subgroups of cognitive decline and baseline predictors of heterogeneity in preclinical progression.METHODS: Data were drawn from the Anti-Amyloid Treatment in Asymptomatic Alzheimer’s Disease Study, which enrolled amyloid beta-positive (Aβ+) participants, and the Longitudinal Evaluation of Amyloid Risk and Neurodegeneration (LEARN) Study, which enrolled amyloid beta-negative (Aβ−) individuals. Latent class mixed- effects models identified cognitive trajectory classes. Associations between class membership and demographic, clinical, and biomarker variables were evaluated. The primary outcome was change in the Preclinical Alzheimer Cognitive Composite.RESULTS: Three trajectory classes were identified: stable, slow decliners, and fast decliners. Higher phosphorylated tau at 217 (p-tau217), smaller hippocampal volume, and elevated tau positron emission tomography were associated with declining classes. About 70% of Aβ+ individuals were stable.DISCUSSION: Latent class modeling reveals substantial heterogeneity in preclinical trajectories with important implications for prevention trial design.K E Y W O R D Sclinical trial design, cognitive decline, latent class mixed-effects model, preclinical Alzheimer’s diseaseHighlights∙ Latent class analysis of A4 and LEARN data identified distinct patterns of preclinical cognitive trajectories among biomarker-positive individuals.∙ Recently developed biomarkers, including plasma p-tau217 and tau PET, improved prediction of latent class membership. This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited. © 2026 The Author(s). Alzheimer’s & Dementia published by Wiley Periodicals LLC on behalf of Alzheimer’s Association.Alzheimer’s Dement. 2026;22:e71366. wileyonlinelibrary.com/journal/alz 1 of 12https://doi.org/10.1002/alz.71366 ### Page 2 2 of 12 LI ET AL .∙ Most Aβ+ individuals showed cognitive stability, which reduces statistical power for traditional cognitive endpoints.∙ Declining individuals contributed disproportionately to statistical power despite representing a minority of the Aβ+ population.∙ Findings highlight the need for power analyses and/or trial simulations that account for heterogeneity in preclinical cognitive decline1 BACKGROUNDElevated levels of brain amyloid in cognitively unimpaired older adults are associated with subsequent cognitive decline and increased risk of clinical progression. 1–5 The Anti-Amyloid Treatment in Asymptomatic Alzheimer’s Disease (A4) Study, a randomized trial of solanezumab in cognitively unimpaired individuals with elevated amyloid PET, demon- strated group-level decline on a cognitive composite but no treatment benefit. 2 Follow-up analyses from A4 and its companion observational study of amyloid beta-negative (Aβ−) individuals, the Longitudinal Evaluation of Amyloid Risk and Neurodegeneration (LEARN) Study, further confirmed that higher baseline amyloid PET and plasma phos- phorylated tau (p-tau217) levels are associated with faster cognitive decline and increased functional progression. 5 Despite these consistent group-level associations, cognitive tra- jectories among amyloid beta-positive (Aβ+) cognitively unimpaired individuals remain highly variable. Conventional longitudinal models assume that individual cognitive trajectories are randomly scattered about a single mean trend for given covariate values, potentially obscuring meaningful heterogeneity. To better characterize this het- erogeneity, we applied a latent class mixed-effects model (LCMM) 6 to A4 and LEARN data. This approach identifies unobserved subgroups of participants with distinct longitudinal patterns of change and allows for the evaluation of baseline biomarkers and demographic factors as predictors of class membership. Latent class growth model approaches have been applied several times in the literature to identify distinct classes of late-life cognitive trajectories. A systematic review by Wu et al. 7 identified 37 investi- gations, and these consistently identified three classes: stable, slow decline, and fast decline. However, many of these papers lacked more recently developed biomarkers of Alzheimer’s pathology. One excep- tion is the study by Teipel et al., 8 which included amyloid positron emission tomography (PET) as a predictor in a cohort of N = 265 individuals followed for 2 years. Villeneuve et al. 9 identified latent classes of Amsterdam Instrumental Activities of Daily Living ques- tionnaire using amyloid PET and fluorodeoxyglucose (FDG) PET inN = 289 participants over 3 years. Here, we consider latent classes of cognition predicted by apolipoprotein E (APOE) genotype, amyloid PET, phosphorylated tau 217 (p-tau217), hippocampal atrophy, and tau PET in a cohort of N = 1629 individuals followed for up to 7 years in A4 and LEARN. Furthermore, we explore practical implications for classification algorithms and clinical trials.2 METHODS 2.1 Conduct of A4 and LEARN studiesThe A4 and LEARN studies have been described elsewhere, 2,5 but we briefly summarize key elements. The A4 Study was a multicenter, ran- domized, double-blind, placebo-controlled secondary prevention trial enrolling cognitively unimpaired older adults aged 65 to 85 years with elevated brain amyloid on florbetapir PET. Participants were recruited from 67 sites in the United States, Canada, Australia, and Japan and underwent standardized screening, baseline clinical assessment, biomarker acquisition, and neuropsychological testing. After eligibility confirmation, participants were randomized to receive solanezumab or placebo and were assessed longitudinally with regular cognitive eval- uations, safety monitoring, and follow-up imaging and fluid biomarker collection per protocol. The LEARN Study was conducted in parallel as an observational cohort enrolling individuals who met all A4 screening criteria except for amyloid PET positivity. LEARN participants completed identical clinical and cognitive assessments, allowing direct comparison of trajectories between Aβ+ and Aβ− individuals. Both studies followed harmonized procedures for data collection, visit scheduling, and quality control to ensure comparability of longitudinal outcomes. All participants of both studies provided written informed consent, and study protocols were approved by the Institutional Review Boards or Ethics Committees at each participating site.2.2 Statistical analysis2.2.1 Latent class mixed-effects models LCMMs extend traditional mixed-effects models by identifying unob- served subgroups, or latent classes, of individuals who follow dis- tinct longitudinal trajectories. In the context of cognitive decline in Alzheimer’s disease (AD), LCMMs allow for the estimation of population-level trends while capturing individual variability and uncovering hidden subpopulations that may progress at different rates. Analyses were conducted using harmonized data from the A4 and LEARN studies, accessed through the A4LEARN R data package (ver- sion 1.1.20250808 10,11 ). This package provides curated datasets and metadata derived from the A4 and LEARN clinical studies for repro- ### Page 3 LI ET AL . 3 of 12ducible statistical analysis. Code to reproduce the results from this paper are available from the A4LEARN GitHub repository. 12 The dependent variable was the Box-Cox-transformed Preclinical Alzheimer Cognitive Composite (PACC) 13 score at each visit. The Box- Cox transformation parameter (𝜆) was selected to maximize the like- lihood that the transformed data approximated normality. Time was modeled as a continuous variable representing years since baseline. Natural cubic spline basis functions with one, two, or three degrees of freedom were evaluated to model non-linear trajectories, with boundary knots at baseline and the maximum follow-up time and inte- rior knots at the median or tertiles of observation time. The choice of degrees of freedom and the number of classes (also one, two, or three) was determined by model fit criteria (Bayesian Information Cri- terion [BIC] and Integrated Complete Likelihood [ICL]). 14 Each model included class-specific spline-based time effects, as well as subject- specific random intercepts. The effect of baseline covariates was shared across latent classes and included randomized to solanezumab (1 for the active group, 0 for placebo or LEARN), plasma p-tau217 con- centration, florbetapir cortical standardized uptake value ratio (SUVr), APOE 𝜀4 carrier status, sex, age, education, hippocampal atrophy, and PACC test stimulus version administered (which alternated per study protocol). Plasma p-tau217 concentrations were measured using an immunoassay developed by Eli Lilly and reported in arbitrary units per milliliter (U/mL), where “U” denotes a generic assay unit proportional to signal intensity. Absolute calibration against mass concentration (e.g., pg/mL) was not available. Hippocampal atrophy measures were derived by first residualizing with respect to intracranial volume and then standardizing to mean zero and variance one (i.e., z-scoring). The LCMM includes two linked submodels. The longitudinal sub- model captures within-class trajectories of cognitive change over time. Each class had its own set of spline coefficients, allowing class-specific shapes of decline. The class membership submodel defines the proba- bility of belonging to each latent class as a multinomial logistic function of baseline covariates (which includes all covariates listed above except spline terms and PACC test version). In a secondary analysis, we fit the model to the subset of participants with baseline flortaucipir (tau) PET imaging. Tau PET was summarized as the mean SUVr across eight cortical regions: entorhinal cortex, inferior temporal, inferior parietal, posterior cingulate, caudal middle frontal, middle temporal, superior parietal, and frontal pole. This com- posite measure was included as an additional covariate in both the longitudinal and class membership submodels to assess whether tau pathology improved prediction of cognitive decline patterns or latent class assignment. To assess the reliability and potential predictive utility of the latent class membership model, we examined the posterior probabilities of class assignment for each participant. High posterior probabilities indicate greater certainty in latent class classification. We summa- rized these distributions across classes and computed the mean posterior probability within each group as an index of classification confidence.RESEARCH IN CONTEXT1. Systematic review: Prior research identified distinct patterns of cognitive decline in aging populations, but there is limited understanding of the predictive value of recently developed biomarkers of Alzheimer’s pathology. We found no studies that examined how latent class het- erogeneity informed the design or power of secondary prevention trials. 2. Interpretation: Our findings demonstrate marked het- erogeneity in cognitive change even among individuals with biomarker evidence of Alzheimer’s pathology. Most Aβ+ adults remained stable across the observed inter- val, while a minority showed slow or rapid decline. Baseline plasma p-tau217, hippocampal volume, and tau PET signal were strongly associated with class membership. 3. Future directions: Future work should validate these tra- jectory classes across independent cohorts, incorporate longer follow-up, and evaluate genetic, lifestyle, and new biomarkers. Applying class membership to refine enrich- ment strategies may improve the design and efficiency of secondary prevention trials. 2.2.2 Power analysis stratified by latent class Historically, the design of clinical trials for preclinical Alzheimer’s dis- ease has relied on the assumption of a homogeneous study population. To evaluate the impact of patient heterogeneity on trial design, we performed power calculations based on latent classes of cognitive decline. This approach assesses how variable rates of disease pro- gression affect the statistical power required to detect treatment effects. Effect size estimates were derived from longitudinal models using natural cubic splines applied to PACC scores. 15 Separate models were fit for (1) the stable class, (2) the decliner classes, and (3) a ref- erence group of Aβ− stable individuals. The models included fixed effects for time (two degrees of freedom), APOE 𝜀4 carrier status, sex, age, education, plasma p-tau217, florbetapir PET, and PACC test version. Residuals were assumed to be normally distributed with an unstructured covariance matrix. For each group, we extracted mean PACC and residual variance from the model at two time points, 2 years and 4 years, represent- ing potential trial durations. These estimates were used to compute the power to detect treatment effects. Power was approximated using two-sample t-test calculations, assuming 500 participants per arm, attrition rates of 10% at 2 years and 20% at 4 years, and a two-sided alpha of 0.05. ### Page 4 4 of 12 LI ET AL .2.2.3 Ten-fold cross-validation of latent class predictions To further evaluate prospective discrimination, we performed 10-fold cross-validation stratified by p-tau217 and latent class. For each of the 10 folds, 90% of the data were used to re-train the LCMM, and the re- trained model was used to predict latent classes for the 10% held-out test set using only baseline data. Model performance was summarized using the area under the precision-recall curve (AUPRC) for discrim- ination of each class versus all others. This approach can be more informative than receiver operating characteristic (ROC) curves when group sizes are imbalanced. 16 2.2.4 Regression tree analysis for latent class characterization To evaluate whether latent classes could be distinguished by base- line characteristics through a combination of binary decision rules, we conducted a post hoc classification tree analysis. 17,18 Class assign- ments from the optimal LCMM were used as the outcome variable, with baseline demographic and clinical characteristics as predictors (ran- domized to solanezumab, plasma p-tau217 concentration, florbetapir PET, APOE 𝜀4 carrier status, sex, age, education, hippocampal atrophy, and PACC). We separately considered a regression tree with tau PET as an additional predictor. Hyperparameter tuning was performed using 10-fold cross- validation. The optimal hyperparameters were selected based on maximum cross-validated accuracy. The final tree structure was visu- alized to illustrate the hierarchy of decision rules for class assignment. Variable importance was calculated based on the total reduction in node impurity attributed to splits on each predictor. This approach provides an interpretable framework for understanding the multivari- ate profile of each latent class and assessing whether classes can be reliably separated using clinically available baseline information. All analyses were conducted using R (version 4.5.2; R Foundation for Statistical Computing) with the lcmm package (version 2.2.1).3 RESULTS 3.1 Study participantsA total of N = 1629 participants were included in the analytic sample for the base model (without tau PET), comprising N = 1110 from the A4 Study and N = 519 from the LEARN Study. The mean (standard deviation [SD]) age at baseline was 71.46 (4.69) years, and 60.2% were female. The median follow-up time was 6.0 years (interquartile range 3.9 to 7.0). Baseline characteristics stratified by latent class member- ship are presented in Table 1. A total of N = 427 individuals had tau PET data and were submitted to the tau PET model. The tau PET sub- set included N = 372 from the A4 Study and N = 55 from the LEARN Study. Baseline characteristics for the tau PET subset are presented in Table S1.3.2 Characterization and prediction of latent classesIndividual and base LCMM (excluding tau PET) mean trajectories are shown in Figure 1. Model selection criteria identified the model with two degrees of freedom and three latent classes, which we labeled as stable (77%), slow decliner (16%), and fast decliner (7%). The next best fitting model had three classes and three degrees of freedom (ΔBIC = 22.2, Δ ICL = 22.2). The best fitting tau PET LCMM had two degrees of freedom and three latent classes (Figure S1), and the next best fitting model had two classes and two degrees of freedom (ΔBIC = 376.4, Δ ICL = 336.9). The two models largely agreed on indi- vidual classifications. Only 7/427 (1.6%) individuals were reclassified from one model to the next, and none of those were fast declin- ers. Posterior class probabilities (Figure S2) indicated good separation (mean posterior probability, 0.94), indicating model confidence in class assignment. The three classes are significantly separated starting immediately at baseline. From the base model, participants classified as fast declin- ers started with a mean −0.98 (95% confidence interval [CI] −1.44 to−0.52) PACC points and declined to a mean of −15.8 (95% CI −16.7 to−15.0) at 6 years. In contrast, stable individuals started with a mean 0.52 (95% CI 0.37 to 0.67) PACC points and improved to a mean of 1.16 (95% CI 1.00 to 1.31) at 6 years. Slow decliners were intermedi- ate between the other two classes starting with a mean −0.13 (95% CI −0.44 to 0.17) PACC points and declining to a mean −4.74 (95% CI−5.22 to −4.26). Clinical relevance of the PACC latent classes was supported by dif- ferences in functional outcomes. The proportion of participants who were observed to progress on the Clinical Dementia Rating (CDR) Global score increased monotonically across latent classes, with the lowest rate among stable individuals (22.8%) and progressively higher rates among slow (71.9%) and fast decliners (88.2%) (Table 1). However, discordance between PACC latent class labels and observed CDR pro- gression is common. While CDR Global scores greater than zero are possible without an indication of memory issues, this explains very little of the discordance. Among CDR progressors, the proportion of individ- uals with a CDR Memory Box score of zero was 8.4% of PACC stable individuals, 5.5% of slow decliners, and 4.8% of fast decliners. The distribution of continuous baseline variables by latent class is shown in Figure 2. Figure S3 summarizes standardized coefficients and 95% CIs from both components of the LCMM, and Table S2 shows these coefficients on the original scale of the covariates. Figure S3A,C depicts odds ratios (ORs) for the class membership submodels from the base model and tau PET model. Continuous predictors are standardized so that ORs reflect the change in odds of belonging to the fast or slow declining class compared with the stable class per SD increase in each continuous baseline predictor. ### Page 5 LI ET AL . 5 of 12TA B L E 1 Baseline demographic, clinical, and biomarker characteristics by latent class of cognitive decline.Stable (N = 1257) Slow decliner (N = 253) Fast decliner (N = 119) Total (N = 1629) p valueGroup < 0.001 LEARN 481 (38.3%) 35 (13.8%) 3 (2.5%) 519 (31.9%) Placebo 386 (30.7%) 128 (50.6%) 49 (41.2%) 563 (34.6%) Solanezumab 390 (31.0%) 90 (35.6%) 67 (56.3%) 547 (33.6%) Age, mean (SD), years 70.78 (4.29) 74.06 (5.24) 73.10 (5.12) 71.46 (4.69) < 0.001 Sex, No. (%) 0.016 Male 494 (39.3%) 117 (46.2%) 37 (31.1%) 648 (39.8%) Female 763 (60.7%) 136 (53.8%) 82 (68.9%) 981 (60.2%) Race, No. (%) 0.773N missing 10 2 1 13 American Indian or Alaska Native 6 (0.5%) 0 (0.0%) 1 (0.8%) 7 (0.4%) Asian 25 (2.0%) 4 (1.6%) 4 (3.4%) 33 (2.0%) Native Hawaiian or other PI 0 (0.0%) 0 (0.0%) 0 (0.0%) 0 (0.0%) Black or African Am. 30 (2.4%) 6 (2.4%) 1 (0.8%) 37 (2.3%) White 1179 (94.5%) 240 (95.6%) 112 (94.9%) 1531 (94.7%) Unknown or Not Reported 7 (0.6%) 1 (0.4%) 0 (0.0%) 8 (0.5%) Ethnicity, No. (%) 0.535 Hispanic or Latino 35 (2.8%) 8 (3.2%) 5 (4.2%) 48 (2.9%) Not Hispanic or Latino 1211 (96.3%) 241 (95.3%) 114 (95.8%) 1566 (96.1%) Unknown or not reported 11 (0.9%) 4 (1.6%) 0 (0.0%) 15 (0.9%) Education, mean (SD), years 16.62 (2.74) 16.58 (2.70) 16.32 (2.85) 16.59 (2.74) 0.512 APOE ε4, No. (%) < 0.001 Non-carrier 732 (58.2%) 93 (36.8%) 31 (26.1%) 856 (52.5%) Carrier 525 (41.8%) 160 (63.2%) 88 (73.9%) 773 (47.5%) PACC, mean (SD) 0.74 (2.37) −0.90 (2.47) −2.23 (3.10) 0.27 (2.61) < 0.001 Amyloid PET, mean (SD), CL 37.40 (34.81) 72.10 (42.88) 88.62 (37.75) 46.53 (40.23) < 0.001 P-tau217, mean (SD), U/ml 0.20 (0.09) 0.33 (0.16) 0.45 (0.26) 0.24 (0.15) < 0.001 Hippocampal atrophy, mean (SD), z-score −0.20 (0.96) 0.60 (0.96) 0.95 (0.87) 0.01 (1.03) < 0.001 Tau PET, mean (SD), SUVr < 0.001N missing 946 179 77 1202 Mean (SD) 1.08 (0.07) 1.13 (0.08) 1.23 (0.13) 1.10 (0.09) CDR Prog., No. (%) < 0.001 CDR Non-progressor 970 (77.2%) 71 (28.1%) 14 (11.8%) 1055 (64.8%) CDR Progressor 287 (22.8%) 182 (71.9%) 105 (88.2%) 574 (35.2%)Note: Baseline characteristics of participants classified by the LCMM of PACC trajectories. Classes represent distinct longitudinal cognitive patterns: stable, slow decliner, and fast decliner. Continuous variables are presented as mean (SD), categorical variables as No. (%). Amyloid PET and tau PET SUVr values represent mean cortical uptake relative to cerebellar reference region. Hippocampal atrophy is residualized for intracranial volume and z-scored. Education reported in years of formal schooling. CDR progressors were observed to have a CDR Global score greater than zero at two consecutive visits or the last visit. Abbreviations: APOE ε4, apolipoprotein E ε4 allele; CDR, Clinical Dementia Rating; LCMM, latent class mixed-effects model; LEARN, Longitudinal Evaluation of Amyloid Risk and Neurodegeneration; PACC, Preclinical Alzheimer Cognitive Composite; PET, positron emission tomography; p-tau217, phosphorylated tau 217; PI, Pacific Islander; Prog., Progressor; SD, standard deviation; SUVr, standardized uptake value ratio; U/mL, arbitrary units proportional to assay signal. ### Page 6 6 of 12 LI ET AL .F I G U R E 1 Individual and mean Preclinical Alzheimer Cognitive Composite (PACC) trajectories by latent class. (A) Spaghetti plot of individual participant trajectories on PACC, colored by latent class derived from the latent class mixed-effects model (LCMM). Each line represents one participant’s observed scores over time. (B) Estimated mean PACC trajectories for each latent class, with shaded regions indicating 95% confidence intervals. Higher PACC scores indicate better cognitive performance. From the base model (Panel A), higher plasma p-tau217 levels (OR 3.2, 95% CI 2.4 to 4.1), elevated amyloid PET Centiloid (OR 1.4, 95% CI 1.0 to 1.9), and greater hippocampal atrophy scores (OR 3.9, 95% CI 2.7 to 5.6) were each associated with increased odds of belong- ing to faster-declining classes. Interestingly, p-tau217 and hippocampal atrophy had numerically stronger effects than amyloid PET. Also of interest, fast decliners were numerically younger (mean 73.1 years, SD 4.7), on average, than slow decliners (mean 74.1 years, SD 5.2), and increased age was associated with membership in the slow declining group (OR 1.4, 95% CI 1.1 to 1.7) but not the fast declining group (OR 0.93, 95% CI 0.69 to 1.2). A possible explanation for the non‑monotonic relationship between age and latent class severity (slow decliners are oldest followed by fast decliners, then stable) is that age exerts two opposing influences in this cohort. On the one hand, older age is asso- ciated with a greater likelihood of belonging to the slow declining class, consistent with the well‑established role of age as a risk factor for cognitive decline. On the other hand, some comparatively younger amyloid‑positive individuals may have more aggressive or rapidly pro- gressive disease biology, which attenuates the age association with membership in the fast declining class. In addition, survivor bias may contribute to the slow declining class’s having the highest mean age, as individuals with more aggressive forms of preclinical AD are less likely to be represented at older ages. Tau PET model results are largely sim- ilar, but confidence intervals are wider, likely due to smaller available sample (N = 427 vs N = 1629). Increased tau PET was associated with increased odds of slow (OR = 1.8, 95% CI 1.2 to 2.8) and fast declining groups (OR = 3.6, 95% CI 2.1 to 6.1). Figure S3,B,D presents standardized coefficients from the longitudi- nal submodels, representing associations between each covariate and the PACC outcome. Higher PACC scores correspond to better cog- nitive performance, and continuous variables are standardized and oriented so that positive coefficients indicate better performance. We see that females had increased risk of being fast decliners (OR 2.5, 95% CI 1.5 to 4.4) but had better PACC performance for a given class mem- bership (0.30 PACC points, 95% CI 0.24 to 0.35). APOE 𝜀4 carriage was associated with increased risk of both slow (OR 1.9, 95% CI 1.3 to 2.8) and fast (OR 2.9, 95% CI 1.6 to 5.3) declining class versus stable class, but negligible remaining effect on PACC. Education had no effect on class membership but was associated with better PACC performance (0.12 PACC points per SD of education, 95% CI 0.095 to 0.15).3.3 Power analysis stratified by latent classPower analyses stratified by latent class revealed marked differences in cognitive trajectories and variability across classes (Table S3). The Aβ− stable group had the highest mean (SD) PACC scores over time:+1.14 (2.11) at 2 years and +1.20 (2.25) at 4 years. We considered this the maximum possible treatment benefit (Figure 3). The stable Aβ+group had means of +0.88 (2.19) at 2 years and +0.96 (2.34) at 4 years. With 500 participants per arm, a trial in Aβ+ stable individuals would have only 44% power at 2 years and 30% at 4 years to detect even 100% of the maximum benefit (0.26 PACC points at 2 years and 0.24 at 4 years). In contrast, a trial in Aβ+ declining individuals would have 93% power at 2 years and 98% at 4 years to detect just 20% of the maximum benefit (0.79 PACC points at 2 years and 1.40 at 4 years). These findings suggest that stable Aβ+ individuals, although they may have the greatest potential long-term benefit from a disease- ### Page 7 LI ET AL . 7 of 12F I G U R E 2 Distribution of baseline demographic and biomarker variables by latent class of cognitive decline. Sina plots show the distribution of selected baseline variables among latent classes identified by the latent class mixed-effects model (LCMM) of Preclinical Alzheimer Cognitive Composite (PACC) trajectories: stable, slow decliner, and fast decliner. Variables include age, baseline PACC score, plasma phosphorylated tau 217 (p-tau217), amyloid positron emission tomography (PET) standardized uptake value ratio (SUVr), tau PET SUVr (subset with tau PET available), and hippocampal atrophy (volume normalized to intracranial volume). Each dot represents an individual participant; box plots represent quartiles. Higher PACC scores indicate better cognitive performance. U/mL = arbitrary units proportional to assay signal. modifying drug, contribute little to the statistical power of a preclinical AD trial to detect an effect on the PACC. Instead, power is primarily driven by declining individuals, who represent a minority of the Aβ+population. Enrolling large numbers of stable individuals may dilute the overall treatment effect and reduce the trial’s ability to detect mean- ingful cognitive benefits. These results underscore the need to account for latent classes of decline when designing preclinical AD trials. Again, the goal of this analysis was not to suggest conducting trials in homo- geneous populations. Rather, we aimed to highlight the complexity and potential inefficiency introduced when multiple latent classes with distinct trajectories are present within a trial cohort.3.4 Predictive performance and cross-validationThe overall cross-validated accuracy was 0.80 (95% CI 0.79 to 0.82) for the base model and 0.756 (95% CI 0.73 to 0.78) for the tau PET model. However, this accuracy rate is inflated by the large number of ### Page 8 8 of 12 LI ET AL .F I G U R E 3 Trends for latent class-specific clinical trials. Mean trajectories were estimated from longitudinal models using natural cubic splines (two degrees of freedom) applied to Preclinical Alzheimer Cognitive Composite (PACC) scores for (A) two decliner classes, (B) amyloid-positive stable participants, and (C) a reference group of amyloid-negative stable individuals representing the maximum possible treatment benefit. For decliners, power calculations assume a treatment effect equal to 20% of the maximum possible benefit at 2 and 4 years. For amyloid-positive stable participants, power calculations assume a treatment effect equal to 100% of the maximum possible benefit at 2 and 4 years. Dashed lines represent projected mean PACC scores under treatment; crosses indicate time points used for power calculations. stable individuals (1257/1629 = 77.2%), which are relatively easy for the model to identify. AUPRCs from the base model were 0.95, 0.41, and 0.49 for non-, slow, and fast decliners versus the rest and 0.70 for either slow or fast decliner versus stable (Figure S4). AUPRCs from the tau PET model were 0.92, 0.31, and 0.56 for non-, slow, and fast decliners versus the rest and 0.71 for either slow or fast decliner versus stable (Figure S5).3.5 Post hoc regression treeThe optimally tuned classification tree without tau PET achieved a mean balanced accuracy of 0.63 (standard error 0.007) in 10- fold cross-validation (Figure S6). Variable importance analysis identi- fied baseline p-tau217 (56.2%), baseline PACC (15.5%), amyloid PET (13.1%), and hippocampal atrophy (10.0%) as most important, though amyloid PET was not included in the tree, likely due to its correla- tion with p-tau217. The tree structure revealed a hierarchical decision process, with participants with baseline p-tau217 < 0.29 classified as stable individuals, though this misclassifies N = 155 decliners. Those with p-tau217 ≥ 0.29 are further parsed by hippocampal atrophy, PACC, and P-tau217. Only one leaf contains a majority of fast declin- ers, and it contains only 1% of the initial sample. Note that although P‑tau217 was selected instead of amyloid PET in the regression tree, amyloid PET is still recognized as an important predictor. This occurs because the tree algorithm selects the single best classifier among correlated variables (such as p‑tau217 and amyloid PET). Meth- ods that aggregate multiple trees (e.g., random forests) could retain both biomarkers and leverage any independent predictive value each contributes. While the tree demonstrates the relative importance and utility of the predictors, it also demonstrates a high degree of misclassification (31.4%). Cross-validated balanced accuracy is improved by the addition of tau PET (mean 0.69, standard error 0.02; Figure S7). Tau PET has the second largest variable importance (22.5%), between p-tau217 (29.8%) and hippocampal atrophy (17.4%). The tree with tau PET results in a larger proportion of leaves that are majority fast decliners (8.5% vs 1% without tau PET).3.6 Longitudinal biomarker progression across latent classesTo further characterize biological differences among the latent trajec- tory groups, we compared longitudinal changes in amyloid PET, plasma p-tau217, tau PET (medial temporal and neocortical), and hippocam- pal volume across the four classes (Figure 4). As expected, slow and fast decliners showed the most rapid biomarker progression across modalities, consistent with their steeper cognitive decline. In contrast, Aβ− stable individuals demonstrated minimal change over time in all biomarkers, supporting their classification as biologically and clini- cally stable. Notably, Aβ+ stable individuals, despite showing PACC trajectories that closely resembled Aβ− stable individuals, exhibited clear evidence of biomarker progression, including increasing amyloid burden, rising plasma and PET tau signals, and accelerating hippocam- pal atrophy. These findings suggest that many Aβ+ stable individuals ### Page 9 LI ET AL . 9 of 12F I G U R E 4 Longitudinal biomarker trajectories by Preclinical Alzheimer Cognitive Composite (PACC) latent class. Each panel displays individual biomarker trajectories (light lines) and the estimated mean trajectory (bold black line) from a linear mixed-effects model with random slopes and natural cubic spline fixed effects (2 degrees of freedom). Columns correspond to amyloid beta (Aβ)− stable, Aβ+ stable, slow decliner, and fast decliner. Rows represent five longitudinal biomarkers: amyloid positron emission tomography (PET) (Centiloids), plasma phosphorylated tau 217 (p-tau217) (U/mL), medial temporal lobe (MTL) tau PET SUVr, neocortical tau PET SUVr, and hippocampal volume. While Aβ− and Aβ+stable individuals demonstrate similarly stable cognitive trajectories, the Aβ+ stable individuals exhibit clear biomarker progression across amyloid, tau, and neurodegeneration measures, suggesting that many may be in an earlier stage of preclinical disease and could transition into declining classes with longer follow-up. may represent an earlier stage of preclinical disease and could tran- sition into declining classes with extended follow-up, highlighting the temporal dissociation between biomarker progression and short-term cognitive change.4 DISCUSSIONIn this LCMM analysis of cognitively unimpaired older adults, we identified distinct trajectories of PACC performance and evalu- ### Page 10 10 of 12 LI ET AL .ated the extent to which plasma and imaging biomarkers prospec- tively distinguished individuals who remained cognitively stable from those who showed slow or rapid decline. Several findings high- light challenges for prognostic modeling in the preclinical stage of AD and have direct implications for secondary prevention trial design.4.1 Available predictors only partially identified future declineAlthough plasma p-tau217 and amyloid PET were each associated with latent class membership, neither biomarker alone, or in combina- tion, achieved high accuracy in predicting of cognitive decline. A key observation was the high frequency of cognitive stability among Aβ+participants. As shown in Table 1, 776 of 1110 Aβ+ A4 participants (69.9%) were classified as stable individuals, indicating that substantial amyloid burden does not necessarily translate into measurable cogni- tive change over the timescales typical of prevention trials. However, this is consistent with a pre-symptomatic phase of disease believed to last as long as 15 years. Compared to the 30.1% of Aβ+ individuals classified as decliners, a small minority of Aβ− LEARN participants were classified as declin- ers. Thirty-eight of 519 (7.3%) Aβ− individuals were classified as slow or fast decliners, suggesting that some causes of decline may reflect non-AD processes, early AD pathology below PET detection thresh- olds at baseline that accumulated over the 5-year period, or cognitive measurement variability. These findings reinforce that amyloid positiv- ity, while necessary for defining the AD pathophysiologic continuum, is an insufficient standalone predictor of short- to intermediate-term cognitive decline.4.2 Relative contributions of plasma p-tau217 and tau PETPlasma p-tau217 demonstrated stronger prognostic associations than amyloid PET, consistent with biomarker models positioning tau abnormalities closer to symptom onset. This observation aligns with prior reports showing improved prognostic accuracy with plasma tau markers. 5 However, classification tree analyses indi- cated that no single p-tau217 threshold meaningfully enriched for decliners without simultaneously excluding many true declin- ers, underscoring the limitations of threshold-based enrichment strategies. Tau PET provided incremental predictive value beyond plasma p- tau217 and amyloid PET, although gains were modest. Because tau PET captures later-stage pathologic changes, these results are biologically plausible. Nevertheless, the logistical and financial limitations of tau PET constrain its feasibility as a routine screening tool in large-scale prevention trials.4.3 Implications for secondary prevention trial designParticipant selection. Because most Aβ+ individuals remain cognitively stable, eligibility based solely on amyloid positivity will enroll many stable participants, reducing power for cognitive endpoints. Plasma p- tau217 may improve enrichment but cannot reliably isolate decliners. More complex multimodal strategies or repeated biomarker assess- ments may be needed for adequate prognostic discrimination.Designing trials for heterogeneous cohorts. Traditional power calcula- tions assume a homogeneous population and uniform treatment effect size, overlooking substantial heterogeneity in cognitive trajectories. Fast decliners may show larger measurable effects but often represent individuals further along the disease continuum and exhibit greater variability, while stable individuals pose a different challenge. Their stability limits detectable cognitive benefit even if biological effects occur, requiring treatments to produce gains beyond natural improve- ment. Our analysis shows that variability across classes directly influ- ences minimum detectable effect sizes, meaning that single-effect-size assumptions risk producing underpowered studies or making ineffi- cient resource allocations. More realistic simulation-based approaches that incorporate class-specific trajectories and differential treatment effects can improve trial design by informing enrichment strate- gies, optimizing sample sizes, and aligning outcomes with subgroup trajectories.Added value of tau PET. Tau PET provided additional predictive infor- mation, but its incremental value must be weighed against feasibility. Tau-PET-based staging may help identify individuals closer to clini- cal transition, though widespread implementation in prevention trials remains challenging.Outcome selection. Limited cognitive change in stable individuals raises questions about the sensitivity of traditional cognitive outcomes at this early stage. Digital measures of learning curves and biologi- cal endpoints such as plasma or tau PET may offer more responsive markers of treatment effect and serve as secondary or exploratory out- comes. Given evidence that tau PET correlates more strongly with cog- nitive decline than amyloid biomarkers, longitudinal tau PET or plasma p-tau217 may be useful for monitoring treatment response or detect- ing early trajectory divergence. Future work applying latent class or trajectory models to biomarker data may clarify whether biomarker- defined progressors can be identified earlier or more reliably than cognitive progressors.Early intervention. Anti-amyloid therapies show greater benefit in symptomatic individuals with less tau pathology, supporting the theory that earlier intervention yields greater benefit. However, if pre-symptomatic decline unfolds gradually over a decade or more, trials starting early may struggle to demonstrate cognitive benefit within feasible windows. Refining and validating sensitive digital and biomarker-based endpoints will be essential for advancing preclinical therapy development. Reliable measures of target engagement and disease progression could improve enrichment, increase power, and ### Page 11 LI ET AL . 11 of 12enable detection of treatment effects even among cognitively stable participants.4.4 Overall interpretationIn summary, data-driven cognitive trajectories among initially unim- paired older adults exhibited substantial heterogeneity, and widely used biomarkers, including plasma p-tau217 and amyloid PET, were only partially effective in predicting who would decline over the time frame of a prevention trial. These limitations highlight the need for improved prognostic tools, potentially incorporating longitudinal biomarker dynamics, multimodal risk models, or digital assessments. Until then, secondary prevention trials must account for substan- tial heterogeneity in cognitive trajectories when planning enrollment, power calculations, trial simulations, and analytic strategies.4.5 Strengths and limitationsThis study has several strengths. We leveraged a large, deeply phe- notyped cohort of cognitively unimpaired older adults with longitudi- nal cognitive assessments and harmonized biomarker data using the A4LEARN R package to standardize and reproducibly analyze data across the A4 and LEARN cohorts. The LCMM approach allowed us to identify data-driven cognitive trajectories as opposed to relying on predefined thresholds or single-outcome definitions of decline. The inclusion of plasma p-tau217, amyloid PET, and, within a subset, tau PET enabled evaluation of multiple biomarkers spanning distinct phases of AD pathophysiology. This study also has limitations. First, although latent class meth- ods improve characterization of heterogeneity, class assignments are probabilistic and sensitive to model specification. Second, follow- up duration, while substantial, may still be insufficient to capture long-term cognitive change among stable individuals. Third, tau PET was available only in a subset, limiting statistical power for com- parisons involving this biomarker and potentially reducing general- izability. Fourth, although we examined a broad panel of covariates, unmeasured factors, including comorbidities, lifestyle variables, and scanner-related differences, may influence both biomarker levels and cognitive outcomes. Finally, this was a highly selected sample of indi- viduals who were either eligible for the A4 study or ineligible owing only to low amyloid PET signal, and they may not fully represent community-dwelling older adults.5 CONCLUSIONSAmong cognitively unimpaired older adults, cognitive trajectories showed marked heterogeneity, and widely used biomarkers, includ- ing plasma p-tau217 and amyloid PET, were only partially effective at identifying who would decline. Tau PET provided modest additional predictive value but remains impractical for widespread enrichment. These findings highlight the challenge of prospectively identifying decliners in preclinical AD and underscore the need for improved mul- timodal prognostic tools. Secondary prevention trials should account for substantial heterogeneity in cognitive trajectories when developing enrollment strategies, powering cognitive endpoints, and interpreting treatment effects.ACKNOWLEDGMENTSThe authors would like to thank the A4 and LEARN Study teams and site principal investigators and staff. Special gratitude goes to the A4 and LEARN participants and their study partners, without whom these studies would not have been possible. The A4 and LEARN Studies and this analysis were supported by a public-private-philanthropic partner- ship that included funding from the National Institute on Aging of the National Institutes of Health (NIH) (R01 AG063689, U19AG010483, U24AG057437, and R01AG079142), Eli Lilly (also the supplier of active medication and placebo), the Alzheimer’s Association, the Accel- erating Medicines Partnership through the Foundation for the NIH, the GHR Foundation, the Davis Alzheimer Prevention Program, the Yugilbar Foundation, an anonymous foundation, and additional pri- vate donors to Brigham and Women’s Hospital, with in-kind support from Avid Radiopharmaceuticals, Cogstate, Albert Einstein College of Medicine, and the Foundation for Neurologic Diseases. Other support was provided by grants from the Epstein Family Foundation.CONFLICT OF INTEREST STATEMENTDr. Aisen reported personal fees from Merck, Biogen, Roche, AbbVie, ImmunoBrain Checkpoint, Bristol Myers Squibb, and Neurimmune and grants from Eisai outside the submitted work. Dr. Sperling reported consulting fees from AbbVie, AC Immune, Acumen, Alector, Apel- lis, Biohaven, Bristol Myers Squibb, Genentech, Janssen, Nervgen, Oligomerixg, Prothena, Roche, Vigil Neuroscience, Ionis, and Vaxxin- ity outside the submitted work. Dr. Donohue reported personal fees from Roche (consultant) and Johnson & Johnson (spouse is full- time employee) outside the submitted work. Dr. Raman reported grants from American Heart Association, Gates Ventures, Eisai, and Alzheimer’s Association. No other disclosures were reported. Author disclosures are available in the Supporting Information.CONSENT STATEMENTAll human subjects provided informed consent.ORCIDMichael C. Donohue https://orcid.org/0000-0001-6026-2238REFERENCES1. Petersen RC, Wiste HJ, Weigand SD, et al. Association of elevated amyloid levels with cognition and biomarkers in cognitively normal people from the community. JAMA Neurology. 2016;73(1):85-92. 2. Sperling RA, Donohue MC, Raman R, et al. Trial of solanezumab in pre- clinical Alzheimer’s disease. N Engl J Med. 2023;389(12):1096-1107. doi:10.1056/NEJMoa2305032 ### Page 12 12 of 12 LI ET AL .3. Donohue MC, Sperling RA, Petersen R, et al. Association between elevated brain amyloid and subsequent cognitive decline among cog- nitively normal persons. JAMA. 2017;317(22):2305-2316. 4. Parent C, Rousseau LS, Predovan D, Duchesne S, Hudon C, Longitudi- nal association between ß-amyloid accumulation and cognitive decline in cognitively healthy older adults: a systematic review. Aging Brain. 2023;3:100074. 5. Sperling R, Donohue M, Rissman R, et al. Amyloid and Tau prediction of cognitive and functional decline in unimpaired older individuals: lon- gitudinal data from the A4 and LEARN studies. J Prev Alzheimers Dis. 2024;11. doi: 10.14283/jpad.2024.122 6. Proust-Lima C, Philipps V, Liquet B, Estimation of extended mixed models using latent classes and latent processes: the R package lcmm.J Stat Softw. 2017;78:1-56. 7. Wu Z, Phyo AZZ, Al-Harbi T, Woods RL, Ryan J, Distinct cognitive trajectories in late life and associated predictors and outcomes: a systematic review. J Alzheimers Dis. 2020;4(1):459-478. 8. Teipel SJ, Cavedo E, Lista S, et al. Effect of Alzheimer’s disease risk and protective factors on cognitive trajectories in subjective memory complainers: an INSIGHT-preAD study. Alzheimers Dement. 2018;14(9):1126-1136. 9. Villeneuve SC, Houot M, Cacciamani F, et al. Latent class analy- sis identifies functional decline with Amsterdam IADL in preclini- cal Alzheimer’s disease. Alzheimers Dement. 2019;5:553-562. doi:10. 1016/j.trci.2019.08.009 10. A4 Study Team. A4 study data portal. Accessed December 15, 2025. https://www.a4studydata.org 11. Donohue MC, Hussen K, Langford O, Gallardo R, Jimenez-Maggiora G, Aisen PS, Alzheimer’s Clinical Research Data via R Packages: the alzverse. Alzheimers Dement. 2026;22:e71152.251002318. Published online. 12. Donohue MC, Li R, Langford O, A4LEARN vignettes and arti- cles. Accessed December 15, 2025. https://github.com/atri-biostats/ A4LEARN/tree/main/vignettes/articles 13. Donohue MC, Sperling RA, Salmon DP, et al. The preclinical Alzheimer cognitive composite: measuring amyloid-related decline. JAMA Neurol. 2014;71(8):961-970. doi:10.1001/jamaneurol.2014.803 14. Biernacki C, Celeux G, Govaert G, Assessing a mixture model for clus- tering with the integrated completed likelihood. IEEE Trans Pattern Anal Mach Intell. 2002;22(7):719-725. 15. Donohue MC, Langford O, Insel PS, et al. Natural cubic splines for the analysis of Alzheimer’s clinical trials. Pharm Stat. 2023;22(3):508- 519. 16. Saito T, Rehmsmeier M, The precision-recall plot is more informative than the ROC plot when evaluating binary classifiers on imbalanced datasets. PloS one. 2015;10(3):e0118432. 17. Therneau T, Atkinson B, rpart: Recursive Partitioning and Regression Trees. 2026. doi:10.32614/CRAN.package.rpart 18. Breiman L, Friedman J, Olshen RA, Stone CJ, Classifica- tion and Regression Trees. Chapman and Hall/CRC; 1984. doi:10.1201/9781315139470SUPPORTING INFORMATIONAdditional supporting information can be found online in the Support- ing Information section at the end of this article.How to cite this article: Li R, Langford O, Insel PS, et al. Divergent patterns of cognitive decline in preclinical Alzheimer’s disease: Implications for secondary prevention trials. Alzheimer’s Dement. 2026;22:e71366. https://doi.org/10.1002/alz.71366