Explore Our Blog

Alzheimer’s Drug Development & Challenges to the Amyloid Hypothesis

In 2018, the FDA published a draft guidance on how to develop drugs to treat early Alzheimer’s Disease (AD). This document came in the wake of several high-profile late‑stage failures for AD drugs. These failures have challenged the decades old amyloid hypothesis of AD pathophysiology. In this post, we’ll explore the amyloid hypothesis and some recent findings that challenge it. We’ll also review the FDA’s recommendations for designing AD clinical trials.

The Amyloid Hypothesis

The amyloid hypothesis is based partly on research done by Selkoe and Glenner in the 1980s and was formally proposed by Hardy and Higgins in 1992. It identifies post-translational processing of the amyloid precursor protein (APP) by secretases as the source of short insoluble peptides (Aβ), which are proposed to be the root cause of AD.

In normal brains, these peptides are degraded and removed, but under pathological conditions, the peptides can aggregate to form plaques. The plaques interfere with normal neuronal function, causing cell death and neurodegeneration. Early support for the amyloid hypothesis came from genetic studies in families with a history of AD. These families were found to carry mutations in either APP or the secretase genes that led to increased levels of Aβ production.

Using the amyloid hypothesis as a starting point, many researchers created mouse models of AD in which high levels of Aβ were produced, forming numerous plaques in the brain. Surprisingly, although the mice developed plaques, they did not show evidence of nerve cell death or cognitive impairment. To be fair, animal models rarely replicate all of the elements of human disease, but this was not the only evidence that cast doubt on the amyloid hypothesis.

Three positron emission tomography (PET) AD diagnostic imaging agents have been approved in recent years, and these agents have enabled visualization of Aβ plaques in living patients. The hope was that these agents would facilitate earlier diagnosis of AD and provide a window of opportunity for treatment before clinical symptoms emerged. However, clinicians have been surprised to observe that plaques can be found in many normal patients and that some AD patients have very few plaques. Thus, it may be that plaques are simply a normal consequence of aging and not necessarily a hallmark of AD.


Many AD drug development programs have seen setbacks due to their inability to meet efficacy endpoints in late stage trials. Despite empirical evidence that challenges the amyloid hypothesis, many pharmaceutical companies have embarked on development programs that target Aβ plaques or the secretases that cleave APP. A number of pharmaceutical companies have developed monoclonal antibodies specific for Aβ plaques. All of them, however, have encountered difficulties.

Another popular target for AD drug development has been beta-secretase cleaving enzyme 1 (BACE1), an enzyme that cleaves APP to produce Aβ peptides. A Phase III trial of Merck’s BACE1-inhibitor, verubecestat, was ended prematurely for failing to have a positive clinical effect in patients with mild to moderate AD. Taken together, the clinical results of anti-Aβ antibodies and BACE1 inhibitors continue to challenge the amyloid hypothesis and the underlying assumptions about AD pathology.

In fairness, these and many other AD trials have focused on late stage AD patients, which present a number of additional challenges that are not present in patients with earlier stages of the disease. Many experts believe that these drugs will have a bigger impact in early stage patients, and as a result, the development of verubecestat continues.

FDA Guidance

Recognizing this shift in focus towards early AD, the FDA’s recently published draft guidance makes several recommendations regarding diagnostic criteria and staging for AD. It defines four conceptual stages of AD, separated by pathophysiological changes, abnormalities on sensitive neuropsychological measures, and functional impairment. The guidance encourages sponsors to define the anticipated AD stages for both enrollment and outcome assessment.

The FDA also uses the guidance to emphasize the expectation that biomarkers will play a role in the identification and assessment of patients. However, it concedes that a lack of information on biomarkers in AD prevents them from being used as an indicator for a persistent effect on disease course. Because no consensus exists for which biomarkers would best support clinical findings in trials in early AD, the FDA also encourages sponsors to analyze each biomarker independently.

Finally, the guidance indicates the FDA’s openness to novel approaches for assessing the pathophysiological and functional aspects of AD. For example, while not specifically addressed in the guidance, sponsors may find that dual-modality imaging devices, such as Positron Emission Tomography/Magnetic Resonance (PET/MR), are the most beneficial tools when evaluating early AD patients. PET/MR would theoretically allow for simultaneous molecular imaging (using one of the aforementioned probes) with functional imaging (using either 18F‑FDG or fMRI) and anatomical imaging.

The ability to combine information from multiple levels of biological complexity would not only provide a more complete picture of patient response to drugs, but would also yield important information on the pathology of the disease.

Updates to the Amyloid Hypothesis

Despite the high-profile failures of AD drug trials, the amyloid hypothesis has by no means been disproven. Instead, most scientists agree that the story of AD is more complex than implied by the original hypothesis. An increasingly large body of evidence suggests that Aβ plaques alone do not cause neurodegeneration, but rather it is an unbalanced immune response to Aβ that leads to inflammation and cell death in the brain.

Microglia, the macrophages of the central nervous system, have been identified in recent years as key determinants in the pathology of AD, with the emergence of complex relationships between Aβ plaques and microglial genes such as TREM2 and ASC. These observations have identified new targets for AD drug development, but they have also created a new dilemma: should AD be managed by activating microglia to clear more Aβ plaques, or should we inhibit microglial action to dampen the inflammatory response?

The turn towards the immune system in AD research is indicative of a wider trend linking neurological and psychiatric diseases to immunological causes. Several reports over the previous decade have linked the immune system to diseases like schizophrenia, challenging age-old paradigms about the brain being an immune-privileged organ. Hopefully, these new lines of investigation will lead to a more complete understanding of amyloid and AD. From that, new and improved therapies that can offer real relief to the millions of patients with AD.


Alzheimer’s Disease has proven to be an extremely difficult target for drug developers. While the amyloid hypothesis sparked tremendous hope in both the scientific and lay communities, more than 25 years after it was introduced only a handful of AD drugs have been approved, and there is still no cure. Meanwhile, researchers continue to look for novel ways to diagnose and treat AD.

If you are developing a therapy for AD or another difficult-to-treat neurological disorder, contact Nuventra today to learn how our team of experienced consultants can help you overcome the drug development and regulatory challenges that these programs present.


Additional Resources