Comprehensive coverage

Lethal learning: a new interpretation of adaptive mechanisms of cancer cells

An article by the researchers of the Center for Biological Networks at the Technion presents an innovative explanation for the limitations of anti-cancer treatments

A conceptual illustration of the chess battle between the cancer cells and the anti-cancer drugs. Most cancer deaths are related to the development of drug resistance and the formation of metastases. In both processes, the cancer cells face difficult and new challenges and a strain: finding a bypass for the drug, or a way to adapt to a foreign and hostile tissue. Unfortunately, the cells often manage to overcome these challenges. How does this happen? This is a complex process, and among other things, the cells go through a process of learning, which involves trial and error, while searching among a large number of configurations and internal states. The theory of learning theory constitutes a mental and mathematical framework for understanding these processes. A review of the current literature shows that basic concepts of learning theory provide an explanation for results that do not correspond to the popular concept. Linking the search process to a concrete mathematical model demonstrates its feasibility. (Illustration: Harel Spring)
A conceptual illustration of the chess battle between the cancer cells and the anti-cancer drugs. Most cancer deaths are related to the development of drug resistance and the formation of metastases. In both processes, the cancer cells face difficult and new challenges and a strain: finding a bypass for the drug, or a way to adapt to a foreign and hostile tissue. Unfortunately, the cells often manage to overcome these challenges. How does this happen? This is a complex process, and among other things, the cells go through a process of learning, which involves trial and error, while searching among a large number of configurations and internal states. The theory of learning theory constitutes a mental and mathematical framework for understanding these processes. A review of the current literature shows that basic concepts of learning theory provide an explanation for results that do not correspond to the popular concept. Linking the search process to a concrete mathematical model demonstrates its feasibility. (Illustration: Harel Spring)

Despite the enormous progress in cancer medicine in recent decades, two physiological phenomena still claim many victims among patients: cancer metastases and resistance to drug treatment. Metastasis - the migration of cancer cells from the primary tumor to distant organs - is responsible for about 90% of cancer deaths. The development of drug resistance is a different phenomenon, which is also very common and deadly. The common explanation for these phenomena is based on the well-known Darwinian principles: random formation of mutations during the division of cancer cells and natural selection dictated by the drug treatment and the target tissues of the metastases. However, many empirical observations challenge this approach, and the treatments based on it do not significantly increase the patient's survival expectancy but only by two to three months on average. 

Researchers at the Technion are now presenting in an article in the journal iScience, a new and different concept regarding metastasis and resistance. The article is signed by doctoral student Asil Shumer and the supervisors Prof. Naama Brenner (Wolfson Faculty of Chemical Engineering) and Prof. Omri Barak (Rappaport Faculty of Medicine). The three are also members of the Center for Biological Networks. This work is a continuation of a theoretical study published by Prof. Brenner in 2017 in the journal Nature Communications.

Metastasis or drug resistance is a kind of learning process

The Technion researchers refer to the said complications - metastases and resistance - as an instance of learning process. In this process, the learners are the cancer cells themselves and the learning helps them overcome the stress they are in as a result of the drug treatment or the nature of the new environment they have reached from the primary tumor. This learning is not based on a defined role of specific proteins or pathways but on Network the interactions between them. It is based on Trial and error Level Network which are increased by the strain. Since the cell is not equipped with a "contingency plan" to deal with this, it performs a random option search. In their article, the researchers propose an analogy between learning processes in the brain and artificial networks and the processes that occur within a single cell at the network level. They claim that learning theory provides a mental framework and mathematical tools for describing and understanding the phenomenon and show how concepts from learning theory shed light on incomprehensible phenomena in cancer processes.

The link between biological processes and learning theory is part of an innovative concept that has been developing in recent years at the Center for Biological Networks at the Technion. This concept refers to biological systems as learning systems that are able to respond ad hoc and adapt by improvising new responses in the face of unexpected challenges. Experiments conducted by the center's researchers in sperm cells, nerve cells, synapses, flies and hydra revealed aspects of complexity in these systems: multiple time scales, high-dimensional dynamics, closed-loop interactions, multiple solutions and microscopic irregularity. All of these are common phenomena that characterize a wide variety of biological systems - phenomena that form the basis of adaptation processes and are not understood theoretically enough. At the center, these properties of complex dynamic systems are studied and, in cooperation, they try to connect experimental observations to the developing theory in the fields of control and learning. The center has research members from different faculties at the Technion: electrical engineering, physics, medicine and chemical engineering.

The research was supported by the National Science Foundation. Asil Shumer's doctoral research is supported by the Israel Academy of Sciences through an Adams scholarship. Shomer completed a bachelor's degree in biochemical engineering at the Technion (Wolfson Faculty of Chemical Engineering) and two advanced degrees under the guidance of faculty members from the Center for Biological Networks. She is doing her doctorate under the guidance of Prof. Naama Brenner (Wolfson Faculty of Chemical Engineering) and Prof. Omri Barak (Rappaport Faculty of Medicine). 

for the article in iScience click here

More of the topic in Hayadan:

Leave a Reply

Email will not be published. Required fields are marked *

This site uses Akismat to prevent spam messages. Click here to learn how your response data is processed.