Investigators from the UCLA Health Jonsson Comprehensive Cancer Center developed an artificial intelligence (AI) model, based on tumor-associated epigenetic factors, which they say can predict patient outcomes successfully across different types of cancer.

Studying multiple cancer types, the researchers found that by examining tumor gene expression patterns of epifactors—epigenetic factors that influence how genes are turned on or off—they could categorize tumors into distinct groups. These clusters allowed the team to predict patient outcomes across various cancer types better than traditional measures such as cancer grade and stage.

The findings, described in Communications Biology, also lay the groundwork for developing targeted anticancer therapies aimed at regulating epigenetic factors, such as histone acetyltransferases and SWI/SNF chromatin remodelers. “Our research helps provide a roadmap for similar AI models that can be generated through publicly available lists of prognostic epigenetic factors,” said study first author, Michael Cheng, a graduate student in the Bioinformatics Interdepartmental Program at UCLA. “The roadmap demonstrates how to identify certain influential factors in different types of cancer and contains exciting potential for predicting specific targets for cancer treatment.”

The researchers, co-led by Hilary Coller, PhD, professor of molecular, cell, and developmental biology and a member of the UCLA Health Jonsson Comprehensive Cancer Center and the Eli and Edythe Broad Center of Regenerative Medicine and Stem Cell Research at UCLA, reported their findings in a paper titled “Pan-cancer landscape of epigenetic factor expression predicts tumor outcome.” In their paper the team concluded, “Our study provides an epigenetic map of cancer types and lays a foundation for discovering pan-cancer targetable epifactors.”

Epigenetics refers to protein factors and processes that allow the establishment and maintenance of different states, including states of gene activity, at the same genomic locus, the authors explained. “Epigenetic processes include changes in DNA methylation, modifications of histone proteins, chromatin accessibility, and higher order chromatin architecture. These changes in state are mediated by chromatin-associated protein factors (epigenetic factors or epifactors) such as those that add, remove, and read chromatin DNA and histone modifications, and remodel the chromatin.”

Coller added, “Traditionally, cancer has been viewed as primarily a result of genetic mutations within oncogenes or tumor suppressors. However, the emergence of advanced next-generation sequencing technologies has made more people realize that the state of the chromatin and the levels of epigenetic factors that maintain this state are important for cancer and cancer progression. There are different aspects of the state of the chromatin— like whether the histone proteins are modified, or whether the nucleic acid bases of the DNA contain extra methyl groups—that can affect cancer outcomes. Understanding these differences between tumors could help us learn more about why some patients respond differently to treatments and why their outcomes vary.” And as the authors continued, “Epifactors themselves can be genetically altered in tumors, which can cause widespread epigenetic dysregulation.”

While previous studies have shown that mutations in the genes that encode epigenetic factors can affect an individual’s cancer susceptibility, little is known about how the levels of these factors impact cancer progression. This knowledge gap is crucial in fully understanding how epigenetics affects patient outcomes, noted Coller. The authors also  noted, “Genetic mutations and epigenetic changes can have a cooperative effect on cancer development, and one may predispose cancer cells to the other …”

To see if there was a relationship between epigenetic patterns and clinical outcomes, for their reported study the researchers turned to The Cancer Genome Atlas (TCGA), and analyzed the expression patterns of 720 epigenetic factors to classify tumors from 24 different cancer types into distinct clusters. “We clustered the patient tumors from each cancer type using the non-negative matrix factorization (NMF) algorithm based on the epifactor genes with the most variable expression among the patient tumors,” they wrote.

“These epifactors encode proteins involved in the addition, removal, and recognition of DNA methylation and histone marks, and chromatin remodeling,” they wrote. And interestingly, they pointed out, the majority of these epifactors—556 out of 720—weren’t known to be genetically altered in cancer tissues.

Out of the 24 adult cancer types evaluated, the team found that for 10 of the cancers, the clusters were associated with significant differences in patient outcomes, including progression-free survival, disease-specific survival, and overall survival.

This was especially true for adrenocortical carcinoma, kidney renal clear cell carcinoma, brain lower grade glioma, liver hepatocellular carcinoma and lung adenocarcinoma, where the differences were significant for all the survival measurements. The clusters with poor outcomes tended to have higher cancer stage, larger tumor size, or more severe spread indicators.

“We saw that the prognostic efficacy of an epigenetic factor was dependent on the tissue-of-origin of the cancer type,” said Mithun Mitra, PhD, co-senior author of the study and an associate project scientist in the Coller laboratory. “We even saw this link in the few pediatric cancer types we analyzed. This may be helpful in deciding the cancer-specific relevance of therapeutically targeting these factors.”

The team then used epigenetic factor gene expression levels to train and test an AI model to predict patient outcomes. “Using machine learning, we developed a neural network model for the five-cancer group combined that was highly predictive of outcome,” they noted. This model was specifically designed to predict what might happen for the five cancer types that had significant differences in survival measurements.

The scientists found the model could successfully divide patients with these five cancer types into two groups: one with a significantly higher chance of better outcomes and another with a higher chance of poorer outcomes. “A pan-cancer machine learning model deploying epifactor expression data for these five cancer types successfully separated the patients into poor and better outcome groups,” they stated. They also saw that the genes that were most crucial for the AI model had a significant overlap with the cluster-defining signature genes.

“The pan-cancer AI model is trained and tested on the adult patients from the TCGA cohort and it would be good to test this on other independent datasets to explore its broad applicability,” said Mitra. “Similar epigenetic factor-based models could be generated for pediatric cancers to see what factors influence the decision-making process compared to the models built on adult cancers.”

The authors also noted, that many epifactors, such as enzymes involved in DNA methylation, histone methylation, and histone acetylation, have been suggested as targets for anti-cancer therapy. “Our extensive and unbiased survey of 720 epifactor genes revealed several novel genes that may represent possible drug targets,” they stated. In particular, histone acetyltransferases were enriched among prognostic genes and associated with improved patient outcome. “This finding would support a possible benefit for histone deacetylase inhibitors that are being approved for cancer treatment,” they noted. In addition, the SWI/SNF family of chromatin remodelers was also enriched among the prognostic genes across the 24 adult cancer types. “Our findings thus suggest these two protein families as possible targets for epigenetics-based cancer therapy,” the investigators wrote.

Noting limitations of their study, the authors concluded, “Our analyses add to our growing understanding of the clinical differences between cancer types based on their tissue-of-origins and spatial locations in the body, and the epigenetic contributors to patient outcome for tumors of the same site. The results from this pan-cancer study can be used as a foundation for rational drug design targeted at epigenetic regulators.”

Previous articleGene Therapy Briefs: Activist Investor Reported to Take $1B Stake in BioMarin
Next articleThe First CRISPR Drug: Vertex Pharmaceuticals’ Casgevy Wins U.K. Approval for Sickle Cell Disease