A "bottleneck" for cancer
Thousands of different genetic mutations are to blame for the development of cancer, but a new analysis of almost 10,000 samples found that, regardless of the origin, all tumors can be divided into 112 subtypes and that regulatory proteins controlling the transcriptional state of cancer cells are characteristic of each subtype. Now, instead of looking for drugs targeting the mutated genes of smaller subsets of patients, it will be possible to create new classes of drugs designed to target the main regulatory proteins, which will significantly increase the proportion of patients who can respond to the same treatment.
The study was organized by Andrea Califano, Corey Abate-Shen and Mariano Alvarez, who are doctors of medical sciences at Columbia University.
Analysis of thousands of samples of all types of cancer also showed that the key genetic programs necessary for the survival of cancer cells are controlled by only 24 master regulator blocks (MR blocks), each of which contains several such proteins working in concert.
As part of personalized medicine, scientists are looking for one of thousands of possible genetic mutations, or mutational patterns, that could trigger the disease, hoping that this will help create drugs aimed at the activity of related proteins. But instead of looking for drugs against each individual mutation, a new study shows that only a few dozen different drugs may be needed to target MR blocks.
This hypothesis has already been tested in a number of clinical studies, including in relation to breast cancer, pancreatic cancer and neuroendocrine tumors, as well as in the framework of the Columbia Precision Oncology Initiative, a large–scale program to evaluate genomic, immunotherapeutic and MR-based treatment units in 3,000 patients with eight aggressive types of tumors.
Personalized therapy benefits a small number of cancer patients
Most cancer patients receive the same treatment, which has been tested on thousands of patients. But when it turns out to be ineffective, patients can choose an individual approach, which includes the identification of specific mutations in the tumor and the selection of drugs aimed at them.
According to Califano, few patients actually benefit from this approach, because most tumors do not have drug-treatable mutations, and the few who have them often do not respond to treatment or quickly relapse after the initial response. Large-scale studies have shown that only 5-10% of patients benefit, but in most of them the tumor eventually progresses to a drug-resistant form. Therefore, alternative strategies are urgently needed, such as targeting the BRAF oncogene with the vemurafinib inhibitor, which provides a short-term response in patients with melanoma and mutations in this gene. Unfortunately, relapse occurs within a few months, so there is practically no effect on survival.
Califano and his colleagues focused on a different approach to therapy. Using advanced mathematical and physical methods for modeling complex biological systems, including molecular interactions that implement the biological logic of the cell, the research team collected data from thousands of tumor samples to understand how genetic mutations affect the activity of all proteins in a malignant cell. Of course, genes are important because they represent a scheme for creating proteins, but it is proteins that control functions in the cell, including the transformation of a normal cell into a cancerous one.
MR blocks ultimately determine the fate of the cell, although they themselves rarely undergo mutations. They are necessary and sufficient to maintain cancer cells in almost all tumors. If we imagine the main regulators as a narrow funnel opening, it turns out that the upper part of the funnel collects the effects of all genetic mutations in the cell and transmits them to this "neck".
Schematic representation of how MR blocks ("bottleneck") direct genetic changes in the cell, initiating programs that stimulate the progression of cancer.
Would it be more efficient and effective to simply close the narrow opening of the funnel by targeting one or more main regulators than to target all the mutated proteins fed into it?
Main Controller units
The main regulators have already been identified in several specific cancers, and the new study looked for them in 20 different types of cancer, as well as any overlap they may have in several cancers.
To achieve this goal, Califano's group has developed a computational tool called Multi-Omics Master-Regulator Analysis (MOMA) for analyzing gene expression in tumors. With the help of MOMA, almost 10 thousand tissue samples from the collection of the Cancer Genome Atlas of the National Cancer Institute were processed.
The analysis revealed 407 master regulators in various cancerous tumors and found that they are organized into 24 unique and closely interconnected modules, or master regulator blocks (MR blocks). Each MR block contains only a few master regulators working in concert to control certain signs of cancer cell behavior. For example, MR-block:2, the most frequently activated block in the most aggressive tumors, contains 14 regulators of cell growth, DNA repair, cell division and cell proliferation. Activation of this block has been found to be a predictor of poor outcomes in many different types of cancer. On the contrary, it was found that the MR block:24 is associated with inflammatory programs and immune response and, thus, is a predictor of a favorable outcome in melanoma.
On average, two to six MR blocks are activated in each individual tumor.
MR-blocks on the fly
The Califano group also demonstrated that the activity of MR blocks in cell lines can be modulated by drugs that favorably affect the behavior of cells in several types of cancer.
Targeting MR blocks, rather than individual mutated proteins, is potentially able to prevent the development of cancer cell resistance, since individual MR blocks cover the effect of an extremely large number of potential mutations in the above regulatory links. In other words, if the drug targets MR blocks, it is very difficult for the cell to avoid treatment. To do this, it must reprogram itself, and this most often, although with some exceptions, leads to cell death.
The study shows that in the future, each cancer case can be differentiated depending on the presence of specific MR blocks, and the drugs necessary for their targeting are prescribed either separately or in combination. The good news is that the tumor must aberrantly activate and inactivate many genetic programs in order to survive. Thus, even targeting only one of several MR blocks is likely to cause cancer cell death.
Despite the fact that the technology that makes it easy to identify activated MR blocks in a patient's cancer cells already exists, only a few drugs have been developed specifically to disable them. The research team created algorithms to evaluate the ability of existing drugs to inhibit or activate individual MR blocks. For example, there are already four FDA-approved experimental drugs that can activate MR-block:14 in prostate cancer, which dramatically reduces the ability of cells to migrate and metastasize.
Drugs developed specifically for the defeat of MR blocks can potentially exceed the effectiveness of existing treatment methods. Currently, a number of joint studies are underway aimed at creating this new class of inhibitors, despite the fact that until recently the main regulators were considered to be largely "unatackable" proteins. Some drug candidates are already being tested in preclinical and clinical trials, and the interim results have far exceeded the authors' expectations.
Article by E.O.Paull et al. A modular master regulator landscape controls cancer transcriptional identity is published in the journal Cell.
Aminat Adzhieva, portal "Eternal Youth" http://vechnayamolodost.ru Based on the materials of Columbia University Irving Medical Center: Big Data Analysis Finds Cancer's Key Vulnerabilities.