Insights on Neurodegenerative Diseases

“With the increasing use of biomarkers, we should be able to soon dramatically improve the success rate of drug development for neurodegenerative diseases. Biomarkers that are predictive of therapeutic effect could be employed in early development, allowing for smaller and shorter proof of concept trials. This should increase the success of drug development in the field, by allowing us to screen more drugs, more quickly,” suggests Gary Romano, Head of Development, Alzheimer’s Disease & Neurology, Johnson and Johnson.

Romano is a speaker at the marcus evans Evolution Summit 2015 taking place in Munich Germany, 4 – 6 November.

Why do neurodegenerative diseases have the highest failure rates in clinical trials?

One reason is that the animal models that we use for central nervous system diseases are imperfect models of the human disease. This is because many of the human neurodegenerative diseases that we wish to study do not occur in lower species. In recent years, transgenic animal models have been widely employed to study disease mechanisms, but these are incomplete models which typically only represent one or two components of the disease

pathophysiology. As a result, such models are not very predictive of drug effects in human disease.

A second reason is that neurological diseases, in particular neurodegenerative diseases, are very slow, insidiously progressive and can take years, sometimes decades to develop in humans. The pathophysiology in Alzheimer’s disease, for example, begins as much as two decades before patients start to develop overt symptoms. This makes it challenging to measure the progression of the disease in humans. Clinical trials in this area are necessarily long (often 2 or more years) and large (requiring hundreds of patients per treatment arm), making it very challenging to measure a drug effect on that slow rate of natural disease progression.

A third reason is that up until very recently we have had to diagnose patients with neurodegenerative diseases on the basis of clinical symptoms. For example, Alzheimer’s disease diagnosis was possible only after patients developed overt clinical or cognitive impairment, which we now know are very late manifestations of the disease process. This has been a big disadvantage for neurological drug development since it is generally more successful to intervene early in a disease process. For example, in heart disease it is much better to identify patients early using biomarkers such as cholesterol levels and to treat before they develop late manifestations such as stroke and heart attacks. Fortunately, this is now changing in neurology with the use of biomarkers to identify subjects early in the neurodegenerative disease process, before widespread neuronal injury occurs.

What is needed to increase neurodegenerative diseases success in clinical trials?

For all of these reasons, the cost of clinical trials in neurodegenerative diseases has been very high and as a

result the number of trials that are conducted is very limited. What would help tremendously would be if we had biomarkers that were predictive of treatment effects in patients with the disease. This would greatly reduce the cost of trials by making them smaller and shorter in duration. For example, if we could use a predictive biomarker to determine whether a drug is slowing the progression of Alzheimer’s disease, we could conduct smaller trials, exposing fewer patients to ineffective or possibly harmful compounds, and we could do so much faster and less expensively.

This would enable us to test many more mechanisms and compounds, which would in turn increase the likelihood of finding successful treatments for these devastating diseases.