Skip to main content

Pathways to Alzheimer's disease

Periodic Reporting for period 2 - PATHAD (Pathways to Alzheimer's disease)

Reporting period: 2018-06-01 to 2019-11-30

The amyloid cascade hypothesis on Alzheimer's disease (AD) states that abnormal accumulation of amyloid clumps in the brain is the cause of the disease. However, this is most likely an oversimplification. Amyloid accumulation in plaques outside the neurons is an important feature of AD but exactly how this is associated with neuronal dysfunction and memory problems is unknown.

AD is a very common disease and we need to find ways to prevent or treat its underlying cause. In the current project, we are developing new tools to measure the disease in people at increased risk of AD, in the general population, as well as in patients in different clinical stages of the disease. We are also developing biomarkers for other pathologies that may appear in the ageing brain and that may interact with the more classical AD pathologies (amyloid clumps outside neurons and tau threads inside neurons).

The ultimate goal is to develop easy-to-use objective tests for AD and related pathologies, which would work both in primary healthcare settings and to facilitate screening of individuals who may be eligible to enter clinical trials. The biomarker tools could also be used to evaluate treatment effects in the clinical trials. The latter activity will hopefully speed up the way clinical trials can be conducted so that we eventually will see a disease-modifying therapy against AD.
We started out by evaluating the classical cerebrospinal fluid (CSF) biomarkers for Alzheimer's disease (AD), namely amyloid beta 42 (Abeta42), total tau (T-tau) and phosphorylated tau (P-tau), in the Swedish dementia registry and in a pan-European multi-centre project called EMIF-AD. A large number of patients and control individuals are part of these projects. We could see that around 20-30% of clinically diagnosed AD patients did not have biomarker evidence of AD pathology. This means that they must have other brain changes that contribute to the symptoms.

We next set out to develop biomarker tools for such other pathologies. To that end, we are working on the proteins that build up Lewy bodies (alpha-synuclein, a-syn) and so-called TDP-43 inclusions. The former is seen in Parkinson's disease and dementia with Lewy bodies but is often also seen in AD. The latter may be seen in frontotemporal dementia and amyotrophic lateral sclerosis but can also be present in AD brains. When examining currently available CSF a-syn and TDP-43 tests, none proved good enough, and our research team are now working on improving the assays. We have successfully developed novel biomarkers for microglial and astrocytic activation, as well as synaptic dysfunction in AD.

It is possible to live many years with biomarker evidence of amyloid clumps in the brain without experiencing any symptoms. Our research has shown that the major risk gene for AD, APOE epsilon4, is associated with earlier age of onset of amyloid accumulation in the brain. APOE epsilon4 carriers (around 20% of the general population) often start to accumulate amyloid in midlife. Then it takes 10-30 years before symptoms appear. This is a window of opportunity to intervene with disease-modifying therapies. Neuronal dysfunction in AD can be monitored using tau biomarkers. Our research shows that amyloid markers change first, then tau markers and other markers of neurodegeneration (most notably neurofilament light and the synaptic markers), then markers of microglial activation and neuroinflammation. When neurodegeneration markers turn positive, symptoms soon appear. The more abnormal neurodegeneration markers, the more intense the disease process is. Successful drug candidates in clinical trials should normalize these biomarker abnormalities; a result we are still waiting for.

The most important development in the project is that we have successfully transferred several of the CSF tests into sensitive and reliable easy-to-use blood tests. We will use these tests to examine the disease mechanisms in and risk factors for AD in clinical cohorts in much greater detail than what has been possible before. Another aim is to start using the most promising biomarkers (neurofilament light and p-tau) as blood tests in clinical laboratory practice during 2020.
We have learnt that most AD patients are APOE epsilon4-positive and these are typically amyloid-positive. APOE epsilon4-negative AD patients, on the other hand, are often amyloid-negative. These patients must have other pathologies that cause their clinical phenotype, e.g. alpha-synuclein and/or TDP-43 pathologies. We have shown that currently available assays for these proteins do not reflect the corresponding pathologies and specific tests for pathology-enriched forms of the proteins are needed (ongoing work).

We have also learnt that amyloid-positive individuals most often stay symptom-free until biomarkers for neurodegeneration (total-tau and neurofilament light) turn positive. This symptom- and neurodegeneration-free phase of amyloid positivity may last 10-20 years and is a potential window of opportunity for secondary prevention.

In addition, we have found that a subgroup of AD patients have biomarker evidence of blood-brain barrier dysfunction. This subgroup is characterized by vascular risk factors, e.g. type II diabetes, and thus probably have cerebrovascular dysfunction contributing to their symptoms.

Microglial activation is another feature of AD. Compared with CNS infections and neuroinflammatory diseases, such as multiple sclerosis, microglial biomarkers suggest only low-grade activation in AD. Whether this is beneficial or detrimental may depend on disease stage and the details of this will require further study in longitudinal cohorts (ongoing work).

Using ultrasensitive measurement tools, we have developed blood tests for amyloid, tau and neurofilament light (NfL). Plasma Abeta42/40 ratio is a promising blood biomarker for amyloid build-up in the brain but the most convincing results have been obtained for neurofilament light (more than 20 publications) and phospho-tau181 (unpublished). Plasma or serum NfL is a reliable blood test for neuronal injury/neurodegeneration irrespective of cause and holds promise as a potential screening test that could be used in primary healthcare to help the doctor to decide which patients should be referred for more extensive testing at a memory clinic.

Finally, we have developed a number of improved human-derived neuronal models that could be used to screen for drugs in a manner that should be translatable to animal and human studies, as evidenced by supportive biomarker work on the cell media.

Please see the following link for an overview paper with a poster regarding the latest biomarker discoveries discussed above: