Moscow, Russia – March 28, 2017. On the 1st April, Tatyana Karelina, ISBM leading specialist in neurodegenerative field, will present a poster entitled "Amyloid toxicity and different mechanisms of action for amyloid-targeted treatments studied by quantitative systems pharmacology model" at the 13th International Conference on Alzheimer's and Parkinson's Diseases and Related Neurological Disorders to be held on March 29 - April 2, Vienna, Austria. The research was made in collaboration with Timothy Nicholas, Pfizer. The aim of the presentation is to show how quantitative systems pharmacology (QSP) modeling helps to analyze modern Alzheimer’s disease therapies and hypotheses.
Amyloid beta protein is toxic to neurons, causes inflammatory processes and forming amyloid plaques. In modern clinical practice such deposits in brain are thought to be markers of Alzheimer’s disease. Pharmacologists working with the problem hope that if one could prevent the protein aggregation in plaques (or destroy them) there would be no pathology. However, it still remains not clear how exactly amyloid beta affects the development of the Alzheimer’s disease and which form of the protein is dangerous.
"We developed a model describing all aspects of amyloid protein dynamics: its production, distribution, and aggregation in plaques in brain. Using this model, one can predict how the amyloid concentration in the body tissues will change with different therapies of Alzheimer's disease for long periods of time and at different stages of the disease. This will help clinical pharmacologists to evaluate the efficiency of long-term anti-amyloid therapy for extended periods on the basis of relatively short clinical trials", Tatyana Karelina claims.
Tatyana Karelina's group focused on the four main hypotheses of the connection of Alzheimer's disease with beta-amyloid protein:
Based on the clinical trial data of the existing anti-amyloid drugs for the treatment of Alzheimer's disease, the investigators tried to answer the question: is it possible to understand, which hypothesis is correct? The results of the simulation have shown that three of the four hypotheses could be rejected, and hence the hypothesis of the insoluble form of amyloid toxicity was regarded as the main one.
At one of the work stages Tatyana and her colleagues made calculations for different mechanisms of Alzheimer’s disease drugs actions. All selected therapies were aimed to reduce the concentration of insoluble form of beta-amyloid in the human body, but this effect was achieved in different ways: either the destruction of this protein was activated, or its aggregating into plaques was blocked, or its synthesis was slowed down.
After simulation it’s clear that the amyloid plaques behave quite differently in all listed cases. When the protein production is inhibited, the number of short fibrils may still increase. If one tries to block the polymerization (protein aggregation into the filaments), a positive effect will unlikely be achieved, the disease will progress, perhaps, at a slower rate. And only if the protein degradation is activated, for example, with the help of antibodies, then the concentration of amyloid insoluble form will slowly decrease.
"Our model discovers some dynamic features of the system that greatly impede clinical trials. It takes a long time for modern therapies to show any efficiency. But moreover, even if the therapy has some effect on the amyloid, it may not bring a positive result", says Tatyana Karelina.
At present, Tatyana's group continues to develop a mathematical model of Alzheimer's disease. In the near future a large article will be published, in which one can learn more about this work.
← | December 2025 | → | ||||
Mo | Tu | We | Th | Fr | Sa | Su |
---|---|---|---|---|---|---|
1
|
2
|
3
|
4
|
5
|
6
|
7
|
8
|
9
|
10
|
11
|
12
|
13
|
14
|
15
|
16
|
17
|
18
|
19
|
20
|
21
|
22
|
23
|
24
|
25
|
26
|
27
|
28
|
29
|
30
|
31
|