AI Algorithm Can Identify Potential Therapeutic Targets for Cancer

In a study published this week in Nature Cancer, researchers at Sylvester Comprehensive Cancer Center at the University of Miami Miller School of Medicine described an artificial intelligence (AI) algorithm designed to identify potential therapeutic targets for glioblastoma multiforme (GBM) and other cancers.

According to the press release discussing the study, GBM is an aggressive, and often fatal, type of brain cancer with a five-year survival rate of less than 10 percent. Numerous drugs are being developed as potential therapies, but identifying the molecular mechanisms that drive the disease and applying these to precision medicine approaches remains a challenge.

To address this, the researchers sought to develop a method to better identify protein kinases associated with tumor progression. The most active kinases, which the researchers call “master kinases,” are those that are targeted by drugs and other therapeutics in current cancer treatments.

The research team turned to machine learning (ML) to help identify and experimentally validate two particular kinases associated with tumor progression in two subtypes of GBM and some other subtypes of lung, breast, and pediatric cancers.

The algorithm, known as Substrate PHosphosite-based Inference for Network of KinaseS (SPHINKS), builds on the researchers’ previous work in glioblastoma classification. In a study published in the British Journal of Cancer in March 2021, the research team reported that by capturing key tumor cell traits and grouping GBM patients based on their likelihood of survival and their tumor’s vulnerability to drugs, the algorithm revealed a new glioblastoma classification.

In this week’s study, the researchers independently confirmed these classifications through various omics platforms, including genomics, proteomics, lipidomics, acetylomics, and metabolomics. Using these omics datasets, SPHINKS creates a complete set of biological interactions, known as an interactome, to determine which kinases drive tumor growth and treatment resistance in each glioblastoma subtype.

These findings highlight how multi-omics data and algorithms can be used to predict the targeted therapies that will provide the best treatment options based on each patient’s glioblastoma subtype, the press release states.

“We can now stratify glioblastoma patients based on biological features that are common between different omics,” said Antonio Iavarone, MD, deputy director of Sylvester Comprehensive Cancer Center and senior author of the study, in the press release. “Reading the genome alone has not been enough. We have needed more comprehensive data to identify tumor vulnerabilities.”

The study suggests that SPHINKS and related approaches can be readily incorporated into labs using a clinical classifier and online portal developed by the researchers alongside the algorithm.

“This classifier can be used in basically any lab,” said Anna Lasorella, MD, professor of biochemistry and molecular biology at Sylvester Comprehensive Cancer Center and co-senior author of the study, in the press release. “By importing the omics information into the web portal, pathologists receive classification information for one tumor, ten tumors, however many they import. These classifications can be applied immediately to patient care.”

Moving forward, the researchers hope to leverage the technology in a new type of clinical trial.

“We are exploring the concept of basket trials, which would include patients with the same biological subtype but not necessarily the same cancer types,” Iavarone explained. “If patients with glioblastoma or breast or lung cancer have similar molecular features, they could be included in the same trial. Rather than doing multiple trials for a single agent, we could conduct one combined trial and potentially bring more effective drugs to more patients faster.”