Translating the U.S. POINTER cognitive‐slope benefit into predicted clinical progression: an external‐cohort analysis with exploratory economic thresholds

Abstract INTRODUCTION Cognitive‐slope benefits from multidomain lifestyle trials are difficult to interpret clinically or economically. We translated the U.S. Study to Protect Brain Health Through Lifestyle Intervention to Reduce Risk (U.S. POINTER) structured‐versus‐self‐guided cognitive‐slope benefit into predicted clinical progression and exploratory economic thresholds. METHODS In the Alzheimer's Disease Neuroimaging Initiative (ADNI), Anti‐Amyloid Treatment in Asymptomatic Alzheimer's Disease (A4), and Longitudinal Evaluation of Amyloid Risk and Neurodegeneration (LEARN), an early‐slope landmark Cox model using cognition through the earlier of Month 24 and the selected landmark served as the reference translation; a secondary joint longitudinal‐survival model used cognition through event or censoring. Both mapped a +0.029 standard deviation/year cognitive‐slope shift onto progression to mild cognitive impairment or dementia. Joint‐model hazard ratios were converted to absolute effects using observed Kaplan‐Meier survival. Sensitivity analyses examined information window, endpoint, and time horizon. Outcomes included 5‐year risk differences, numbers needed to treat, restricted mean survival time gains, and exploratory net monetary benefit. RESULTS Effects were directionally consistent but varied substantially with the information used. Reference landmark 5‐year risk differences were 0.42 percentage points in ADNI and 0.16 in LEARN. Secondary joint‐model estimates were larger: 2.71 percentage points in ADNI and 3.94 in A4; LEARN was not reportable. With risk set and pre‐landmark information aligned in ADNI, risk differences were similar for the landmark (0.42) and latent‐slope (0.49) models but increased to 1.60 when cognition through event or censoring was used. Economic margins remained model‐ and assumption‐dependent. CONCLUSION Model‐based translations of the U.S. POINTER cognitive‐slope benefit suggested small to modest predicted delays in clinical progression, with substantial variation across cohorts and modeling approaches. These exploratory estimates do not establish the intervention's actual effect on progression or its cost‐effectiveness.