Industry leading news about artificial intelligence in healthcare and diagnostics.

AI Algorithm Uncovers Risk Behind Thousands of Variants in Key Cancer Gene

Researchers at the Hadassah Medical Organization’s Cancer Research Institute in Jerusalem have developed a predictive artificial intelligence (AI) algorithm that identifies all the potential mutations of a key cancer gene that is present in half of all tumors. This scientific breakthrough may improve precision medicine care, enable better genetic screening and counseling, and can be applied to other types of genes.

The researchers trained an AI algorithm using large databases of tumor and normal genomics. The predictive algorithm can predict the impact of every single missense mutation of the TP53 gene, which is mutated in 50% of cancers.

The Hadassah research team partnered with Thierry Soussi, PhD, professor of molecular biology at Sorbonne University in Paris and an expert on the TP53 gene, to develop the experimental perimeters for the AI algorithm, known at TP53_PROF. PROF stands for “prediction of functionality.”

Algorithm Catalogs the Risk Behind Thousands of Variants

Researchers reported they were able to identify the risk for all 2,314 possible variants of the TP53 gene to a high degree of accuracy. Prior to this study, only 190 variants of the gene were understood. The results of this research were published in Briefings in Bioinformatics on Jan. 18.

Hadassah Medical OrganizationThis new discovery “can lead to more effective precision medicine for oncological patients by the creation of customized cancer treatment decision support systems that are able to identify and discern the important mutations that require treatment from the total number of somatic mutations of the tumor,” Hadassah Medical Organization noted.

While precision medicine applications are one benefit of this research, it also has great potential to be applied to other cancer-related genes.

“The research and technology behind this breakthrough not only provide life-saving screening for carriers of previously unknown cancerous mutations who may be at increased risk, but it is also critical for genomic analysis of somatic mutation profiles in all tumors," said Michal Lotem, MD, Head of the Center for Melanoma and Cancer Immunotherapy in the Department of Oncology at The Hebrew University of Jerusalem.

Attendees at the Artificial Intelligence in Healthcare and Diagnostics Conference can expect to learn more about AI’s role in cancer detection and treatment. Joseph Mossel, Co-Founder and CEO at Ibex Medical Analytics based in Tel Aviv, will present on AI-based cancer detection solutions for prostate cancer and breast cancer. The conference takes place May 10-11 in San Jose, Calif.

Genomic Profiling May Boost Precision Treatments for Cancer

This predictive AI algorithm comes at a time in precision medicine when there is an increased focus on sequencing the genetics of each patient’s cancer. A recent partnership between Illumina and Jean Perrin Center at Clermont-Ferrand University Hospital in France illustrates this progress.

Illumina, a genetic sequencing technology vendor based in San Diego, has partnered with the Jean Perrin Center to trial the clinical utility of comprehensive genomic profiling in patients with late-stage cancer. This method of genetic testing provides a more complete genetic analysis of individual cancers, allowing for more precise treatments.

Personalized treatments that target cancer mutations have improved prognosis greatly, helping us direct patients to the best option for them," said Pascal Pujol, MD, PhD, in a news release announcing the initiative. Pujol is President of the French Society for Predictive and Personalized Medicine and Head of the Oncogenetics Department at Montpellier University Hospital in France.

"It is critical to ensure that the use of comprehensive analysis based on large DNA and RNA panels becomes part of routine practice so that patients can benefit from the best possible care,” he added.

—Caleb Williams

Related Resources:

What is a missense mutation?

TP53_PROF: a machine learning model to predict impact of missense mutations in TP53

Predictive-AI Algorithm Used to Successfully Identify Common Cancerous Gene Mutations Demonstrated at Hadassah Cancer Research Institute

Illumina partners with centers across France to advance a precision medicine approach for patients with late-stage cancer