alzhub.org
Get help now — free helplines

You can call any of these for free. You do not need a diagnosis, an appointment, or a crisis to call — questions and plain exhaustion are reasons enough.

  • Alzheimer's Association 24/7 Helpline 800-272-3900 24 hours a day, 365 days a year Free, confidential support and information for people with dementia, caregivers, families and the public. Master's-level clinicians. Also Spanish.
  • 988 Suicide & Crisis Lifeline 988 24/7 Call or text 988, or chat at 988lifeline.org. For anyone in emotional distress or a mental-health crisis, including caregivers.
  • Alzheimer's Foundation of America (AFA) Helpline 866-232-8484 9:00 am - 9:00 pm ET, seven days a week Licensed social workers providing support, info and referrals. Also text message and web chat. Text: 646-586-5283

In a medical emergency, call 911. More numbers, and what each line does, on the help page.

Home › Research › Research Library › Timeline

Timeline of Alzheimer's research, 1906 to 2026

One case report in 1906, a protein in 1984, a gene in 1991, an image of amyloid in a living brain in 2004, two approved antibodies and a blood test by 2025. This page follows how each step made the next possible, and says where a step was smaller than its reputation.

23 milestones, 34 cited sources, 34 studies from this site's library linked below. Where a paper is not in the library, the source is its PubMed entry. Updated 15 November 2026.

Alzheimer's research did not move in a straight line. Between the first description of the disease and the first drug shown to slow it lay 117 years, most of them spent establishing what the disease is. Reading the milestones in order shows a pattern: a measurement becomes possible (counting plaques, isolating a protein, imaging amyloid, assaying p-tau in blood), a hypothesis is built on it, and a trial tests the hypothesis a decade or two later. The hypotheses were not all right, and the trials that failed are part of the record here.

A disease is described, and its parts are found 1906 to 1984

For its first eight decades Alzheimer's was a pathologist's finding: a name, two lesions seen under the microscope, and no way to see either in a living person. The work of this era was to connect what a person lost to what the brain showed after death.

  1. 1906

    Alois Alzheimer describes a woman's illness and her brain

    Alzheimer reported a patient who had been admitted in her early fifties with memory loss, disorientation and delusions, and whose brain, examined after her death, held two things he had not seen together: clusters of deposits in the cortex (plaques) and tangled, thickened nerve fibres (tangles). He published the case in 1907. It was a single patient.

    Source: Alzheimer A, Stelzmann RA, Schnitzlein HN, Murtagh FR. An English translation of Alzheimer's 1907 paper. Clin Anat 1995

    In the library:
  2. 1968

    Plaques are counted, and the count tracks the decline

    Blessed, Tomlinson and Roth compared the amount of plaque change in the cortex of elderly people with how impaired they had been in life. The two moved together. That turned a descriptive finding into a measurable one, and it is the reason later work treats plaques as something to quantify and, eventually, to remove.

    Source: Blessed G, Tomlinson BE, Roth M. Br J Psychiatry 1968

  3. 1976

    A chemical deficit is found: acetylcholine

    Davies and Maloney reported a selective loss of the central cholinergic neurons in Alzheimer's brains. It gave the rationale for the cholinesterase-inhibitor drugs. When one of them, donepezil, was tested against placebo in 468 people with mild to moderately severe disease, it produced statistically significant improvements on three cognitive and global scales (1998). Symptom relief, not a change in the disease's course.

    Source: Davies P, Maloney AJ. Lancet 1976

    In the library:
  4. 1984

    The protein in the deposits is isolated; the first diagnostic criteria are written

    Glenner and Wong purified the protein from the amyloid in the walls of brain blood vessels in Alzheimer's disease and found it matched nothing known. The same peptide, amyloid-beta, is the main component of the plaques. In the same year the NINCDS-ADRDA work group set out clinical criteria for probable, possible and definite Alzheimer's. Their report states plainly that the diagnosis cannot be determined by laboratory tests, which are used to exclude other causes. For the next thirty years that sentence stayed true.

    Source: Glenner GG, Wong CW. Biochem Biophys Res Commun 1984; McKhann G et al. Neurology 1984 (library record)

    In the library:

Genes and the amyloid hypothesis 1986 to 1999

Once the plaque protein and the tangle protein were known, the question became which came first. Families with early-onset disease supplied the answer for one form of the illness, and a risk gene supplied it for the common late-onset form.

  1. 1986

    Tangles are shown to be built from abnormal tau

    Grundke-Iqbal and colleagues showed that the microtubule-associated protein tau in Alzheimer's tangles is abnormally phosphorylated. Tau is the second pillar of the disease and, forty years on, the p-tau217 in the newest blood tests is a phosphorylated form of this same protein.

    Source: Grundke-Iqbal I et al. Proc Natl Acad Sci USA 1986

  2. 1991

    A mutation in the amyloid precursor protein gene runs in a family

    Goate and colleagues found a missense mutation in the gene for the amyloid precursor protein that segregated with familial Alzheimer's disease. It was the first direct genetic link between the amyloid pathway and the disease. In 1995 Sherrington and colleagues cloned a second gene with missense mutations in early-onset families, presenilin 1, and a 1996 library study of presenilin-1 found a doubling of late-onset risk for one genotype (odds ratio 1.97).

    Source: Goate A et al. Nature 1991; Sherrington R et al. Nature 1995

    In the library:
  3. 1992

    The amyloid cascade hypothesis is stated

    Hardy and Higgins proposed that amyloid-beta deposition is the initiating event and the tangles, cell loss and dementia follow from it. It organised the next thirty years of drug development and is still argued over: a 2025 paper in this library is titled "a conclusion in search of support", and a 2022 one proposes a probabilistic revision.

    Source: Hardy JA, Higgins GA. Science 1992

    In the library:
  4. 1993

    APOE4 is the major risk gene for late-onset disease

    In 42 families with late-onset disease, risk rose from 20% to 90% and mean age at onset fell from 84 to 68 years as the number of APOE-epsilon-4 alleles went up. Having two copies was virtually sufficient to cause the disease by age 80 in those families. Families are not the general population, and APOE4 is a risk, not a diagnosis.

    Source: Corder EH et al. Science 1993 (library record)

    In the library:
  5. 1999

    Immunisation clears plaques in mice

    Schenk and colleagues showed that immunising a transgenic mouse (the PDAPP strain) with amyloid-beta reduced Alzheimer-like pathology. It showed that the immune system could be aimed at plaques, which is the idea behind the antibody drugs that followed.

    Source: Schenk D et al. Nature 1999

Seeing the disease in the living, and the first big failures 2003 to 2015

The 2000s made the pathology visible in living people and showed how long it precedes symptoms. The same decade tested the amyloid hypothesis in humans and mostly lost.

  1. 2003

    An amyloid vaccine trial is stopped

    In a phase 2 trial of 372 patients with mild to moderate Alzheimer's, injections of aggregated amyloid-beta (AN1792) were halted after four early reports of meningoencephalitis. A later pathology study of two vaccinated patients, one of whom developed it, found plaque material focally absent: the plaques could be cleared, at a price. Whether clearing plaques helps patients was a separate question, not answered in a phase 3 trial until the 2023 results below.

    Source: Orgogozo JM et al. Neurology 2003

    In the library:
  2. 2004

    Amyloid is imaged in a living brain

    Klunk and colleagues reported imaging brain amyloid in Alzheimer's disease with Pittsburgh Compound-B, a PET tracer. In 2007 a library study of 13 people with mild cognitive impairment and 14 controls found higher tracer uptake in the MCI group in frontal, parietal and temporal cortex, an early sign that the pathology is present before dementia.

    Source: Klunk WE et al. Ann Neurol 2004

    In the library:
  3. 2010

    A cohort built to test biomarkers: ADNI

    The Alzheimer's Disease Neuroimaging Initiative characterised 819 volunteers at baseline (229 cognitively normal, 398 with MCI, 192 with mild Alzheimer's) and followed them with scans and fluid markers. Its open data underlie much of the biomarker literature in this library.

    Source: Petersen RC et al. Neurology 2010 (library record)

    In the library:
  4. 2011

    Three stages: preclinical, MCI, dementia

    The National Institute on Aging and the Alzheimer's Association replaced the 1984 criteria with recommendations that describe a preclinical stage, mild cognitive impairment, and dementia, and that allow biomarkers to support the diagnosis. The definition began to move from what a person can do toward what is in the brain.

    Source: McKhann GM et al. Alzheimers Dement 2011 (library record); Sperling RA et al. Alzheimers Dement 2011 (library record)

    In the library:
  5. 2013

    TREM2 puts microglia on the list

    Jonsson and colleagues reported a variant of TREM2, a receptor on the brain's immune cells, that raises the risk of Alzheimer's. Immune-system genes have featured prominently in large genetic studies since. A TREM2 antibody tested in 381 people missed its main endpoint at all three doses in 2026 (see below).

    Source: Jonsson T et al. N Engl J Med 2013

    In the library:
  6. 2014

    Two phase 3 trials of bapineuzumab show no benefit

    Bapineuzumab, an anti-amyloid antibody, was tested in two large phase 3 trials in mild to moderate Alzheimer's and produced no significant differences between treatment and placebo on the primary outcomes. It showed that removing plaque late in the disease was not enough, and pushed trials toward earlier stages.

    Source: Salloway S et al. N Engl J Med 2014 (library record)

    In the library:
  7. 2015

    FINGER: a lifestyle programme slows cognitive decline slightly

    In 1,260 Finnish adults aged 60 to 77 at raised risk, two years of diet, exercise, cognitive training and vascular monitoring gave a small advantage on a cognitive battery (0.20 against 0.16). It was the first large trial to show an effect of this kind, and the template for the 2025 US POINTER study.

    Source: Ngandu T et al. Lancet 2015 (library record)

    In the library:

A biological definition, a prevention estimate, and a first approval 2016 to 2022

This stretch redefined the disease by its biology, put a number on how much might be preventable, and produced the first drug approved for a mechanism rather than a symptom.

  1. 2016

    A/T/N: amyloid, tau, neurodegeneration

    A/T/N sorted seven biomarkers into three binary categories: amyloid (A), tau (T) and neurodegeneration (N). Two years later the 2018 NIA-AA research framework defined Alzheimer's biologically, by those markers rather than by symptoms alone, and said explicitly that it was for research, not everyday clinical diagnosis.

    Source: Jack CR et al. Neurology 2016 (library record); Jack CR et al. Alzheimers Dement 2018 (library record)

    In the library:
  2. 2017

    The Lancet Commission names modifiable risk factors

    The Commission's 2017, 2020 and 2024 reports each revised the list. The 2024 report names 14 risk factors, adding high LDL cholesterol in midlife and untreated vision loss, and estimates that roughly 45% of cases worldwide are theoretically attributable to them. These are population estimates from observational studies, not proof that changing one factor will prevent one person's dementia.

    Source: Livingston G et al. Lancet 2017 (library record); Livingston G et al. Lancet 2020 (library record)

    In the library:
  3. 2021

    Aducanumab: accelerated approval on a surrogate marker

    On 7 June 2021 the FDA granted accelerated approval to aducanumab, the first drug approved for a mechanism thought to modify the disease, on the strength of plaque reduction (71% in EMERGE, 59% in ENGAGE) rather than proven cognitive benefit. An advisory committee had voted against, and the approval was contingent on a confirmatory trial.

    Source: Biogen/FDA announcement, 7 June 2021 (library record)

    In the library:

Antibodies that slow decline, blood tests, and a larger genetic map 2023 to 2026

The current period is the first in which two drugs have shown a statistically significant slowing of decline in phase 3 trials, and the first in which a blood test is cleared to help diagnose the disease. Each comes with limits that the same records state.

  1. 2023

    Lecanemab: from accelerated to traditional approval

    The FDA gave lecanemab accelerated approval on 6 January 2023, then converted it to traditional approval on 6 July after the Clarity AD trial confirmed benefit. In 1,795 people over 18 months the decline on the CDR-SB scale was 1.21 points on lecanemab against 1.66 on placebo, a difference of 0.45 on an 18-point scale. Brain swelling (ARIA-E) occurred in 12.6%. A statistically significant result, and a small one.

    Source: van Dyck CH et al. N Engl J Med 2023 (library record); FDA, 6 July 2023 (library record)

    In the library:
  2. 2024

    Donanemab approved; the diagnosis is redefined again

    TRAILBLAZER-ALZ 2 (1,736 people, 76 weeks) slowed decline on the iADRS scale by 22.3% overall and 35.1% in people with low or medium tau, with ARIA-E in 24.0% against 2.1% on placebo and three treatment-related deaths. The FDA approved donanemab (Kisunla) on 2 July 2024. Days earlier, on 27 June, the Alzheimer's Association workgroup proposed revised criteria that define the disease by biomarkers alone and add blood-based markers. In Europe the regulators first refused lecanemab (July 2024), then recommended it for people with one or no copy of APOE4 (November 2024).

    Source: Sims JR et al. JAMA 2023 (library record); FDA, 2 July 2024 (library record); Jack CR et al. Alzheimers Dement 2024 (library record)

    In the library:
  3. 2025

    The first blood test is cleared; a lifestyle trial succeeds at scale

    On 16 May 2025 the FDA cleared the Lumipulse G pTau217/amyloid-beta 1-42 plasma ratio for symptomatic adults aged 55 and over in specialised care. A meta-analysis of 113 studies found p-tau217 the best marker (88.1% sensitivity, 88.7% specificity) while rating about 90% of them at high risk of bias. The same year US POINTER, in 2,111 older adults, found a structured lifestyle programme improved global cognition slightly more than a self-guided one (0.243 against 0.213 standard deviations a year). The EU authorised donanemab in September, restricted to one or no APOE4 copy.

    Source: FDA press release, 16 May 2025; Therriault J et al. Lancet Neurol 2025 (library record); Baker LD et al. JAMA 2025 (library record)

    In the library:
  4. 2026

    A 91-locus genetic map, and a TREM2 antibody that did not work

    A consensus meta-analysis of genome-wide studies combining nearly one million people brought the number of Alzheimer's-linked genetic regions to 91, 16 of them new. People in the top tenth of a genetic risk score were about twice as likely to show severe tangle and plaque pathology at autopsy. In the same year a phase 2 trial of a TREM2 agonist antibody in 381 people with early disease engaged its target but did not meet its primary endpoint. Genetics keeps pointing at new targets, and not every target pays off.

    Source: Bellenguez C, Lambert JC et al. Nat Genet 2026 (library record); Mummery CJ et al. Nat Med 2026 (library record)

    In the library:

What this timeline is not

It is a selection, not a census: milestones were chosen where a source can be named and where the step changed what was possible next. Many important papers, and most of the work of the thousands of researchers behind them, are not here. Dates are those of the cited paper or regulatory announcement. Statements about effect size repeat what the source reports; the library's own scoring of each study is on its page and explained under how we score studies. For the questions that remain open, see open research questions; for the current evidence on the two approved antibodies, anti-amyloid drugs; and for the blood tests, blood tests for Alzheimer's.

Last updated