A4 and LEARN
Anti-Amyloid Treatment in Asymptomatic Alzheimer's Disease (A4) Study and Longitudinal Evaluation of Amyloid Risk and Neurodegeneration (LEARN) Study
Shared data package of the A4 solanezumab prevention trial and its amyloid-negative companion study LEARN: 7,133 longitudinal 3T MRI sessions with T2*-GRE, plus amyloid and tau PET, from cognitively unimpaired adults aged 65 to 85, with central MRI safety reads for microhemorrhages.
Overview
A4 and LEARN are two linked studies of cognitively unimpaired older adults run by the Alzheimer's Therapeutic Research Institute at USC and Brigham and Women's Hospital. A4 was a phase 3, placebo-controlled trial of the anti-amyloid antibody solanezumab in people with elevated brain amyloid on PET (preclinical Alzheimer's disease). LEARN followed people who were screened for A4 but had no elevated amyloid, on a similar schedule. Both are released together as one data package with clinical, cognitive, biomarker and imaging data, used for work on preclinical Alzheimer's disease, amyloid and tau progression, and trial design.
Composition
A4 screened 6,763 people and randomized 1,169; LEARN enrolled 538. The final imaging release counts 7,133 MRI scans from 1,787 participants (A4 1,163, LEARN 537, 87 screen failures), 5,661 florbetapir amyloid PET scans from 4,477 participants and 1,576 flortaucipir tau PET scans from 452 participants. Most screen failures only have amyloid PET. MRI was planned at screening and at weeks 12, 84 and 168, with further scans in the open-label extension.
Acquisition
Data come from 67 to 68 sites in the US, Canada, Japan and Australia, scanned on one qualified 3 T scanner per site (Siemens 67%, GE 23.5%, Philips 8.6% of MRI). The protocol, derived from ADNI, includes 3D MPRAGE, axial T2*-GRE (0.9 x 0.9 x 4 mm, TE 20 ms), axial FLAIR, axial T2 TSE, DWI and resting-state fMRI. Images are shared as NIfTI with BIDS-style JSON sidecars; T1, T2*, FLAIR and T2 TSE series were de-faced and re-faced with mri_reface.
Annotations
No voxel labels. The package includes analysis datasets from central MRI safety reads for microhemorrhages and superficial siderosis, NeuroQuant brain volumes and amyloid and tau PET SUVR values. In the trial, ARIA with microhemorrhage or siderosis was reported in 29.2% of the solanezumab and 32.8% of the placebo group.
Known limitations
- Participants are 65 to 85 years old and about 94% white, so the cohort is not representative of the wider population.
- People with substantial vascular disease or four or more microhemorrhages at screening were excluded from A4.
- MRI safety reads are per-scan visual findings; lesion masks or coordinates are not described in the publications.
- The website states that the repository is under review for possible changes in line with US administration directives.
Cohort
Aggregate numbers from the sources below. Bars are relative to the largest value.
Modality
subjects
- MRI 1,787
PET tracer
subjects
- [18F]florbetapir 4,477
- [18F]flortaucipir 452
Condition
subjects, values can overlap
- Amyloid-related imaging abnormalities 361
- Cerebral microbleeds 347
- Superficial siderosis 38
License and access
Our reading of the license, not legal advice. Before you use the data, read the original license and confirm that your use is allowed. We take no responsibility for how you use a dataset. Full disclaimer
A research proposal is reviewed and approved
A4/LEARN Data Use Agreement
One-time, non-transferable license for the purposes stated in an approved data use application (scientific investigation, teaching or planning clinical research). The data must not be commercialized or redistributed, AI tools must keep inputs and model weights confidential, publications must credit A4/LEARN, and the data must be destroyed after five years.
The same data package is also distributed through LONI IDA and Synapse; the agreement text checked here is the one published on a4studydata.org.
What you can do
- No
- Conditional
- Not stated
- Conditional
What you can share
- No
- Not stated
- Not stated
What you must do
- Yes
- Share alike No
- Yes
- Not stated
- Manuscript review No
- Release code No
- No
- Yes
Limits
- Yes
- Conditional
Commercial license: Not stated No commercial license offer is published; the agreement states the data package shall not be commercialized.
Citation
Jimenez-Maggiora GA, Schultz AP, Donohue MC, Qiu H, Sperling RA, Aisen PS. Advancing open science in Alzheimer's disease clinical trials: Sharing of A4 and LEARN neuroimaging data. Aperture Neuro 6(SI 1) (2026). doi:10.52294/001c.160128. The data use agreement also requires the A4/LEARN acknowledgment statement.
All numbers
Every number on this page, as stored in stats.csv, with its source.
| Measure | Breakdown | Value | Source |
|---|---|---|---|
| Subjects | modality=MR participants with at least one MRI in the final dataset: A4 1163, LEARN 537, screen failures 87 | 1,787 | jimenezmaggiora2026 Table 4 |
| Subjects | tracer=florbetapir participants with at least one amyloid PET: A4 1167, LEARN 538, screen failures 2772 | 4,477 | jimenezmaggiora2026 Table 4 |
| Subjects | tracer=flortaucipir participants with at least one tau PET: A4 393, LEARN 55, screen failures 4 | 452 | jimenezmaggiora2026 Table 4 |
| Studies | total all MRI, amyloid PET and tau PET scans in the final longitudinal dataset (overall N) | 14,370 | jimenezmaggiora2026 Table 2 |
| Studies | modality=MR final longitudinal MRI scans: A4 6091, LEARN 955, screen failures 87 | 7,133 | jimenezmaggiora2026 Table 2 |
| Studies | tracer=florbetapir final amyloid PET scans: A4 2100, LEARN 789, screen failures 2772 | 5,661 | jimenezmaggiora2026 Table 2 |
| Studies | tracer=flortaucipir final tau PET scans: A4 1402, LEARN 170, screen failures 4 | 1,576 | jimenezmaggiora2026 Table 2 |
| Subjects | condition=aria A4 safety population of 1163 treated participants (572 solanezumab, 591 placebo): ARIA adverse events after the first dose, sum of the two disjoint groups (167 + 194) | 361 | sperling2023 Table 3 |
| Subjects | condition=cerebral_microbleeds A4 safety population: microhemorrhage adverse events, sum of the two disjoint groups (158 + 189) | 347 | sperling2023 Table 3 |
| Subjects | condition=superficial_siderosis A4 safety population: superficial siderosis adverse events, sum of the two disjoint groups (19 + 19) | 38 | sperling2023 Table 3 |
| Minimum age | total eligibility age range of A4 and LEARN screening | 65 | yaari2022 Methods |
| Maximum age | total eligibility age range of A4 and LEARN screening | 85 | yaari2022 Methods |
Sources
The keys used in the table above.
- jimenezmaggiora2026 Jimenez-Maggiora et al. 2026, Sharing of A4 and LEARN neuroimaging data, Aperture Neuro paper
- jimenezmaggiora2024 Jimenez-Maggiora et al. 2024, Sharing of baseline A4 and LEARN data, J Prev Alz Dis paper
- sperling2023 Sperling et al. 2023, Trial of solanezumab in preclinical Alzheimer's disease, NEJM (PMC author manuscript) paper
- yaari2022 Yaari et al. 2022, ARIA and other MRI findings in A4 and LEARN screening, J Prev Alz Dis paper