New Study Combines AI And Multiomics Analysis To Understand Cancer Resistance Mechanisms

Original source here.

Two prestigious research institutes, Baylor Scott and White Health and the Translational Genomics Research Institute (TGen), announced today a new research study aimed at analyzing cancer resistance mechanisms using several multiomics technologies, in collaboration with Israeli startup OncoHost.

Baylor Scott and White Health is the largest not-for-profit health system in the state of Texas, with 51 hospitals and more than 1,100 access points serving more than three million Texans. The Translational Genomics Research Institute (TGen) is a Phoenix, Arizona-based nonprofit organization dedicated to conducting groundbreaking research with life-changing results. TGen is part of City of Hope, a world-renowned independent research and treatment center for cancer, diabetes and other life-threatening diseases.

OncoHost has developed a precision oncology platform, a plasma-based, proprietary proteomic analysis tool that guides decision-making in the choice of first-line treatment in cancer patients, analyzes resistance mechanisms, and suggests potential treatment combinations to overcome this resistance.

“It’s detective work,” Michal Harel, Ph.D., Director of Science and Innovation at OncoHost, told me recently. “There are a lot of hidden things in the data—you see a trends, you get a hint, and you say this probably can be generalized, be statistically significant. Then you try to figure out what could be the theory, the biology behind that, and find out if there is any clinical relevance.”

Recent results from the ongoing trial of OncoHost’s platform show that it accurately predicts clinical benefit probability at twelve months based on one simple pre-treatment blood test, outperforming the predictive capabilities of the existing PD-L1 biomarker. The platform also identifies and highlights personalized drug combinations to mitigate active resistance mechanisms to overcome treatment resistance. OncoHost’s PROphet® platform is set to launch in the U.S. in the first quarter of 2023.

Last November, Harel presented at the Society for Immunotherapy of Cancer’s (SITC) 37th Annual Meeting a novel predictive model for clinical benefit of immunotherapy in non-small cell lung cancer (NSCLC) patients that outperforms PD-L1-based predictive models.

According to Harel, combining machine learning and proteomics analysis, OncoHost’s platform provides two clinical utilities: it can successfully predict clinical benefit at twelve months and, combined with PD-L1 testing, can serve as a decision-making tool for physicians on whether to administer immunotherapy alone or in combination with chemotherapy, with a significant effect on overall survival.

The new, multi-year study will utilize Baylor Scott and White’s Texas Immuno-Oncology Biorepository (TIOB)—which collects, catalogs and stores samples of biological material for cancer research—to analyze resistance mechanisms in cancer using multiomics tools including proteomics, single cell analysis, Ct-DNA, and microbiome analysis combined with bioinformatics and machine learning tools.

A comprehensive analysis will be conducted on lung cancer patients at various stages of the disease, analyzing elements such as the host response, the patient’s microbiome, tumor DNA and immune system activity. Testing will look at the patient’s urine, blood, stool, tissue, cells, DNA, RNA, and proteins. The study will include some 350 patients who will receive follow-ups for up to five years.

In a statement, the principal investigator of the study Dr. Ronan Kelly, Director of Oncology at the Baylor Scott and White Charles A. Sammons Cancer Center and the W.W. Caruth, Jr. Chair in Immunology at Baylor University Medical Center Dallas, described its approach as “a very comprehensive profiling process, one of the most comprehensive for cancer patients to date.” Professor Michael Berens, Director of the Cancer and Cell Biology Division at TGen pointed out that “collaborations like these can take us to greater heights in the fight against cancer.” Dr. Ofer Sharon, CEO of OncoHost, said that “together we aim to identify early resistance to treatment using multiomics technologies, which are further revolutionizing biomedical research.”

“For many years, precision medicine has been looking at the tumor and the treatment. We are looking at the tumor, the treatment and the patient, all in one. This helps physicians create a personalized treatment plan,” says Harel.