A drug trial for a Merck & Co Inc. (MSD) medication verubecestat has been stopped because it was very unlikely to show a significant difference in Alzheimers disease progression compared to a placebo control. This was a phase III clinical trial in patients with early dementia.

The setback comes after months of positive media attributed to the medication, with results indicating the drug was indeed succeeding in its intended effect of halting the production of amyloid protein. Despite the reduction in amyloid protein, and therefore an end in the production of amyloid plaques, there was no change to the trial patients’ disease progression. This deals a double blow, denying sufferers of the disease a new treatment option as well as throwing into question the amyloid hypothesis of the disease.

MSD has a continuing clinical trial in Prodromal Alzheimer’s — the drug is being used in patients who do not yet have any form of Alzheimer’s but whose memory is failing. Some analysts think that use in these early patients may stop progression. The fault with trials that have reported so far, they think, is that patients with any symptoms of the disease have already lost too much functioning to be helped

The news of MSD’s failed drug trial falls within three months of Eli Lilly’s potential Alzheimer’s medication solanezumab also ending due to negative results. Solanezumab operates by binding soluble amyloid-β (beta-amyloid), the peptide responsible for the formation of amyloid plaques. This was another attempt at addressing the disease through prevention of plaque formation — as in the case of verubecestat, it did not slow down disease progression.

Both medications were developed using the amyloid hypothesis of Alzheimer’s disease. In short, this considers that amyloid protein is the primary cause of disease. Amyloid protein is potentially involved in communication between synapses — an electrical/chemical junction between neurons. The amyloid protein, the hypothesis says, can be broken down into smaller molecules that can bind to themselves, forming plaques that can cause damage to the brain.

This is a more detailed version of the hypothesis: a cleaved product of APP (amyloid precursor protein), which is thought to play a role in regulating synapse function, beta-amyloid, is formed during APPs passage across the cell membrane, forming a polypeptide with amino acid residues ranging from 37 to 49. The 42 amino acid form, Aβ-42 is chemically “stickier” and can aggregate with other Aβ-42 molecules to form amyloid plaques. These plaques form outside of the neurons and cause damage through inflammatory response and disruption of cell to cell communication.

Amyloid plaques have been found in autopsies of patients known to have severe Alzheimer’s disease, and there are a number of genes implicated in increasing amyloid burden – these include mutations to the APP gene, Presenilin-1 as well as APOE-ε4. All of these have been inserted into transgenic mouse models to study the disease, producing excess amyloid plaques and Alzheimer’s symptoms. This evidence supports the amyloid hypothesis.

Issues occur when attempting to translate the data produced during clinical trials conducted on mice to human trials. The genes inserted into the mice represent familial Alzheimer’s disease, in which the disease can be attributed to a passed-down genetic mutation in a gene often related to amyloid protein. Familial Alzheimer’s however, only represents around two to three percent of all Alzheimer’s cases in humans.

It may be that in up to 98 percent of all Alzheimer’s cases, there is no genetic alteration to the production of amyloid protein. The amyloid hypothesis may not fully explain the disease. This could explain why drugs that halt amyloid production such as verubecestat and solanezumab seem to have no effect on disease progression.

In the mouse models themselves, typically inserting a single gene causes Alzheimer’s disease. The results of using anti-amyloid drugs are often positive in initial animal trials in these genetically-modified mice. This is entirely due to having a single issue present that can be addressed with a single medication. In humans — and in the alternative to familial Alzheimer’s, sporadic Alzheimer’s disease – the causes are less well documented. The causes may involve genetic mutations occurring across a lifetime, environmental factors, or prion proteins. This may be why focused trials on specific factors have remained fruitless for over a decade.

A number of other pharmaceutical companies are currently trialling medicines based exclusively on the amyloid hypothesis, many of these are BACE (an enzyme involved in amyloid production) inhibitors. The companies include Novartis in collaboration with Amgen, AstraZeneca along with its research partner Eli Lilly and also Janssen and Biogen. Following the failed trials of MSD’s BACE inhibitor, verubecestat, a shadow of doubt has been cast on designing trials based on the hypothesis.

There is a large amount of research suggesting alternative avenues for clinical targets other than amyloid related proteins and enzymes. An alternative leading hypothesis is the Tau hypothesis. This involves the phosphorylation (the addition of a phosphate molecule) of tau protein within the neuron. Phosphorylation leads to aggregation and tangled, filament-like tau protein within the neuron. This stops the cell from effectively transporting molecules along the neuron itself eventually causing it to die.

In order to properly address the many potential causes of sporadic Alzheimer’s disease, it may be more productive to follow a hypothesis based on more than one cause. Many of these attribute the disease to a combination of both amyloid and tau dysregulation.

The “dual pathway” model suggests that both amyloid and tau may be linked by separate mechanisms but linked by a common cause. The “serial model of causality” suggests that initial aggregation of amyloid-beta causes biochemical alterations leading to the hyperphosphorylation of tau, which progresses to cause synaptic loss and cell death.

Each of these models explains the failures associated with medical trials focusing on amyloid alone. Tau may be causing symptoms separately from amyloid, or, by the time the amyloid production has stopped, a cascade has already begun in which tau has already been phosphorylated.

Alzheimer’s disease is becoming an ever increasing phenomenon, with statistics indicating 46.8 million people suffering from the disease worldwide as of 2015. Projections indicate this figure has the potential to double every 20 years.

Alzheimer’s disease is now the sixth leading cause of death in the USA — one in three older Americans die with it or another kind of dementia. There may be some good news: a recent study suggested that the overall proportion of senior citizens suffering from dementia has fallen consistently and dramatically over the past 12 years. The finding needs to be treated with some caution: it is a big, well-run study but all the patients lived in one American town (making this a highly isolated study) and the authors cannot explain fully why the decline happened. Even odder, more obese patients had the greatest benefit.

Whatever the percentages, the actual number of patients is rising rapidly as the population ages. This is a prime concern in modern medicine. Despite the setbacks of a number of failed medications, many more are in the pipeline over the coming few years.

“We must continue to pursue this disease. There’s good reason to have hope and optimism that with the pursuit of excellent science and some rethinking about [Alzheimer’s], a treatment will become available for this devastating disease in our lifetimes.” says Professor Bryce Vissel, director of the Centre of Neuroscience and Regenerative Medicine at University of Technology Sydney.