Biology: New approaches to better classify and treat brain tumors

Augusta, Georgia (January 19, 2021) – Looking at RNA, you can see what genes are telling cells to do. Scientists say that looking directly at RNA in brain tumor cells seems to provide objective and efficient evidence for better classification. Tumors and the most effective treatments.

Glioblastoma is the most common type of brain tumor in adults with a wide range of possible outcomes, from generally more curable astrocytomas and oligodendroglioma to typically more lethal gliomas. There are three subtypes.

Medical College of Georgia Scientists Report to Journal Science report Their method of producing what is called the transcriptome profile of tumors is particularly good at recognizing some of the most serious of these tumors, says MD / PhD student Paul MHTran. I have.

Gliomas are currently categorized by histology, primarily shape or morphology, and pathologists identify them when observing cancer cells under a microscope to identify the genetic mutations that cause known cancers.

“We are adding a third method,” said Jin-Xiong She, a prominent scholar of genomic medicine at the University of Georgia Research and the corresponding author of the study, director of the Center for Genome Medicine and MCG Biotechnology. The doctor says. Tran, who holds a PhD in her lab, is the lead author.

Most patients are performing both current classification methods, but the cancer is glioblastoma even if no mutation studies have been performed and even if two pathologists observe the same brain. As with traditional pathology, which turns out, there may be inconsistent findings between the two groups, tumor cells under a microscope, scientists say.

To find out more directly what cancer cells are doing, they choose to look at relatively unexplored gene expression, more specifically one-step downstream RNA that indicates where the cell is heading. I was scared. DNA expression is the same as RNA because DNA makes RNA, which makes proteins and determines cell function. One way cancer can thrive is to alter gene expression, with some up and others down or off.

They believed that the new approach would provide additional insights into tumors, continue to evaluate the effectiveness of existing classification methods, and likely identify new therapeutic targets.

“RNA is a snapshot of what is currently high and what is low in glial cells when they are removed from the body,” says Tran. “They are actually looking at how many copies of RNA-related genes are made, and that gene expression usually determines everything from hair color to weight,” she says. “The transcriptome profile counts the copy number of each gene in the cell.”

Glial cells, whose job is to support neurons, have tightly regulated gene expression, which is exactly what is possible. In the case of cancer, one of the first things that happens is the number of RNA copies of each gene that the cell is changing and the accompanying changes in important cell function. “You change gene expression and something is different,” she says.

Transcriptome profiling, like any other method, is initiated using a tumor sample from a surgeon, but then RNA is extracted through an automated process to read the gene expression levels of various genes. Can be put in a device that can. The large amount of data generated is then sent to a machine learning algorithm developed by Tran. This algorithm calculates the most probable glioma subtype and associated prognosis.

They started with The Cancer Genome Atlas (TCGA) program and the Repository of Molecular Brain Neoplasia Data (REMBRANDT). These are two datasets that have already been tested for RNA and also provide relevant clinical information, including outcomes for more than 1,400 patients. Glioma. Tran, her, and their colleagues used algorithms to discover patterns of gene expression and used those patterns to classify all glioma patients without other inputs. We then compared the three major glioma subtypes that emerged with standard classification methods.

She says their transcriptome classification was about 90% consistent with the traditional approach of observing cells under a microscope and about 93% consistent with observing gene mutations. They found that there was a discrepancy of about 16% between the two standard measurements.

“All three methods disagree with about 10-15% of patients,” she says, but is the most accurate of the three methods because it is superior to the other methods in predicting survival. The analysis states that it must be theirs.

And the discrepancies they find during the classification method can be important for some patients, despite close percentages.

“Our method may have several advantages, because we have found that some patients have a really poor prognosis that can be identified by our method, rather than other approaches,” Tran said. say.

As an example, patients with mutations in a gene called IDH or isocitrate dehydrogenase most commonly suffer from astrocytomas or oligodendroglioma, which are generally more responsive to treatment. It has a higher survival rate than glioblastoma. However, they can also progress to so-called secondary glioblastoma, which may not be detected by the other two methods, even for some low-grade glioblastomas with this IDH mutation. I also discovered that. IDH mutations are rare in primary glioblastoma, Trang said.

Using standard techniques to view time snapshots, these astrocytomas that progress to the more deadly glioblastoma were classified as less serious tumors in 27 patients. “The progression is known, but our technology is good at identifying those cases,” says Tran.

Further analysis also found that approximately 20% of patients with poor prognosis have mutations in the promoter region of the TERT gene. The TERT gene is best known for making telomerase, an enzyme that allows chromosomes to maintain a healthy length. This length is known to decrease with age. TERT function is known to be hijacked by cancer and enable infinite cell proliferation that is characteristic of cancer. They say that this mutation is usually absent in gliomas that begin as more aggressive glioblastomas, suggesting that mutations in the TERT promoter are important for the progression of gliomas. ..

“Inhibitors that target the TERT gene may prevent some of these cases from worsening their prognosis,” says Tran.

These findings also show the strengths of different classification methods, in which case the mutation classification observes cancer cells under a new transcriptome method, rather than these most aggressive brain tumors. It suggests that you may not pick up the old approach, which is excellent for making this important distinction.

“It is known that certain proportions of low-grade gliomas can progress to glioblastomas, some of which can sometimes be misidentified by the original technology. “Tran says. “Using our gene expression method, we found them even though some of them have IDH mutations.”

They write that all these variations have groups like the World Health Organization that are looking for better ways to determine IDH patients with poor prognosis. Other variations include that some glioblastomas with the normal IDH gene carry one of the worse prognosis of glioblastoma, but act like astrocytes and are better. There are subgroups of glioblastoma that tend to carry prognosis.

The MCG team was able to more accurately indicate which patients had a poor prognosis, so the next steps include finding out why and perhaps what they can do.

In addition to the accuracy of the prognosis, the second way to assess tumor classification is whether it points you to better treatment options, she says. He states that most drugs and many of our behaviors, such as exercise and what we eat, alter RNA expression.

“Currently, if someone provides us with RNA expression data from any patient in the world, we can quickly see which glioma subtype is most likely to be.” Tran says. They say that the fact that devices that can examine RNA expression are becoming more widely available should make transcriptome profiling more widely available.

Gliomas are tumors of glial cells, including astrocytes, oligodendrocytes, and microglial cells, whose normal task is to surpass and surround and support neurons.

According to scientists, the identification of intracellular IDH gene mutations has already made the standard glioma classification more systematic. Mutations can be identified by staining or sequencing biopsy slides.

Much progress has also been made in using machine learning to automate and objectify the diagnosis and subtyping of cancers, including glioblastoma. Glioblastoma has been characterized using transcriptome-based analysis, but not all gliomas as in current studies.

Like most genes, the IDH gene usually does a lot of work in the body, including the processing of glucose and other metabolites of different cell types. However, mutations can destroy cells, produce factors such as reactive oxygen species, damage DNA, and cause cancer and other illnesses. These mutations can be caused by age and / or environmental exposure. IDH inhibitors are in clinical trials for a variety of cancers, including glioma.

Gaining insight into the key DNA methylation that occurs in cancer has also led to altered gene expression, including silencing of tumor suppressor genes and the generation of additional gene mutations that cause cancer.

###

Read the entire study.