Researchers have developed an artificial intelligence (AI) algorithm, known as SPHINKS, capable of performing advanced computational analyses to identify potential therapeutic targets for patients with glioblastoma multiforme. The platform may also have applicability in other cancers, according to a new study published by Migliozzi et al in Nature Cancer.
The findings could have profound implications for the future treatment of patients with glioblastoma—which is aggressive and typically fatal—as well as those with certain breast, lung, and pediatric cancers.
Despite breakthroughs for many other cancers, patients with glioblastoma may face poor prognoses—with a 5-year survival rate below 10%. Although numerous drugs are being developed as potential therapeutic options for patients, clinicians have long expressed the need for a way to identify the molecular mechanisms that drive each patient’s disease and are applicable to precision cancer medicine.
Protein kinases are currently the key targets used in precision cancer medicine to tailor treatment to a patient’s specific cancer properties. The most active kinases—labeled “master kinases” by the researchers in the new study—are those for which clinicians direct targeted drugs as a hallmark of current cancer treatment.
Novel SPHINKS Algorithm
“Our work represents translational science that offers immediate opportunities to change the way [patients with glioblastoma] are routinely managed in the clinic,” explained senior study author Antonio Iavarone, MD, Deputy Director of the Sylvester Comprehensive Cancer Center at the University of Miami Miller School of Medicine. “Our algorithm offers applications to precision cancer medicine, giving oncologists a new tool to battle this deadly disease and other cancers as well,” he highlighted.
The SPHINKS algorithm—which stands for Substrate Phosphosite–based Inference for Network of Kinases—deployed deep-machine learning to help the researchers identify and experimentally validate that the master protein kinases PKCδ and DNAPKcs were associated with tumor progression in two glioblastoma subtypes. The researchers noted that the two master kinases may also be used as potential therapeutic targets for other cancers.
In addition to identifying the master kinases, the researchers used patient-derived tumor organoids to demonstrate that targeted drugs interfering with the activity of master kinases can thwart tumor growth.
Previous findings—published by Anna Lasorella, MD, Professor of Biochemistry and Molecular Biology at the University of Miami Miller School of Medicine, and Dr. Iavarone in the British Journal of Cancer—had identified a new glioblastoma classification by capturing key tumor cell traits and grouping patients with glioblastoma based on their likelihood of survival and their tumor’s vulnerability to medication. In the new study, these classifications were independently confirmed through the multiomics platforms of genomics, proteomics, lipidomics, acetylomics, metabolomics, and others.
The SPHINKS algorithm leverages deep-machine learning to refine these omics datasets and create an interactome to pinpoint the kinases that generate abnormal growths and treatment resistance in each glioblastoma subtype. The findings uncovered that multiomics data can generate new algorithms that predict which targeted therapies can provide the best therapeutic options based on each patient’s glioblastoma subtype.
“We can now stratify [patients with glioblastoma] based on biological features that are common between different omics,” Dr. Iavarone emphasized, stressing that “Reading the genome alone has not been enough. We have needed more comprehensive data to identify tumor vulnerabilities.”
The SPHINKS algorithm and related methods can be readily incorporated into molecular pathology labs, according to the researchers. Their research included a clinical classifier that can help assign the appropriate glioblastoma subtype to each patient. The authors believe this approach can produce insightful information that could benefit as many as 75% of patients with glioblastoma.
“This classifier can be used in basically any lab,” said senior study coauthor Dr. Lasorella. “By importing the omics information into the [online] portal, pathologists receive classification information for 1 tumor, 10 tumors, however many they import. These classifications can be applied immediately to patient care.”
While the SPHINKS algorithm was first tested in glioblastoma, the algorithm may be equally applicable to several other cancer types. The researchers found the same cancer-driving master kinases in breast, lung, and pediatric brain tumors—and believe their findings could be the impetus for 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,” Dr. Iavarone revealed. “If patients with glioblastoma or breast or lung cancers 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,” he concluded.
To access the SPHINKS algorithm through the online portal, visit lucgar88.shinyapps.io.
Disclosure: The research in this study was supported by grants from the National Institutes of Health, the National Cancer Institute P30 Supplement GBM CARE-HOPE, the Chemotherapy Foundation, and the Italian Association for Cancer Research Project. For full disclosures of the study authors, visit nature.com.The content in this post has not been reviewed by the American Society of Clinical Oncology, Inc. (ASCO®) and does not necessarily reflect the ideas and opinions of ASCO®.