New AI tool characterizes the diversity of individual cells in tumors

New AI tool characterizes the diversity of individual cells in tumors

A multinational team of researchers led by the Garvan Institute of Medical Research has developed and tested a new AI tool to better characterize the diversity of individual cells within tumors and open the doors to more targeted therapies for patients.

The results of the development and use of the AI ​​tool called Aanet were published today inCancer detectiona journal of the American Association for Cancer Research.

Not all tumor cells are the same

Tumors aren't made up of just one type of cell - they're a mix of different cells that grow and respond to treatment in different ways. This diversity or heterogeneity makes cancer more difficult and, in turn, can lead to poorer outcomes, particularly in triple-negative breast cancer.

Heterogeneity is a problem because we currently treat tumors as if they were made of the same cell. This means that we give a therapy that kills most of the cells in the tumor by targeting a specific mechanism. But not all cancer cells can share this mechanism. While the patient may have an initial reaction, the remaining cells may grow and the cancer may come back. “

Associate Professor Christine Chaffer, co-senior author of the study and co-director of the Cancer Plasticity and Dormancy Program at Garvan

But while heterogeneity is a problem, researchers don't know enough to characterize it: "So far, researchers have not been able to clearly explain how neighboring cells in a tumor differ from one another and how we can classify these differences into meaningful ways to better treat tumors.allcells in this tumor with the right therapies,” adds Associate Professor Chaffer.

A new tool characterizes five new cancer cell groups

To solve this problem, the team developed and trained a powerful new AI tool called Aanet that can detect biological patterns in cells within tumors.

They then used the AI ​​tool to uncover patterns in gene expression of individual cells within tumors, focusing on preclinical models of triple-negative breast cancer and human samples of ER-positive, HER2-positive and triple breast cancer. In doing so, they identified five different cancer cell groups within a tumor with different gene expression profiles, which indicate major differences in cell behavior.

"Using our AI tool, we were able to consistently discover five new groups of cell types within individual tumors, called 'archetypes'. Each group showed different biological pathways and propensities for growth, metastasis and markers of poor prognosis. Our next steps are to see how these groups may change over time, such as

This is a first for cancer research. Co-leader, Associate Professor Smita Krishnaswamy of Yale University, who led the development of the AI tool: "Thanks to technological advances, the last 20 years have seen an explosion of data at the single-cell level. With this data, we found that not only the cancer cancer cells commercially. The first time that another studies on a different case. Meaningful archetypes through which diversity can be analyzed to make meaningful associations with spatial tumor growth and metabolomics Find signatures.

New classification to drive better, targeted treatments

The researchers say that using AANET to characterize the different groups of cells in a tumor according to their biology opens doors for a paradigm shift in the treatment of cancer.

"Currently, the selection of cancer treatment for a patient is based largely on the organ that the cancer has such as breast, lung or prostate and all the molecular markers. However, this assumes that all the cells in that cancer are the same. Instead, we now have a tool to characterize the heterogeneity that characterizes the heterogeneity of the patient, like any group that deals with a biological stage. We know they will targeteveryof these different groups through their biological pathways. This has the potential to significantly improve outcomes for this patient,” says Associate Professor Chaffer.

Regarding the application of Aanet, co-senior author of the study and Chief Scientific Officer of Garvan Professor Sarah Kummerfeld explains: "We envision a future where doctors combine this AI analysis with traditional cancer diagnostics to develop more personalized treatments that target all cell types in the unique tumor of cell types. We can improve a technical and biological improvement of patients. Our study can address technological and biological biology. We can address technological and biological bio-application to technological and biological biology have diseases such as autoimmune diseases.

This research was supported in the following sources.

In Australia: The Nelune Foundation, Tour de Cure, Estee Lauder, The Kinghorn Foundation, The Paramor Family Foundation, University of New South Wales Research Grant, the Ramaciotti Biomedical Research Award, the ARC Development Project Grant and NHMRC Ideas Grants and Investigator Grant.

In the USA: Gruber Foundation Science Fellowship and the Eric and Wendy Schmidt Center at the Broad Institute of MIT and Harvard, the National Science Foundation, the Yale Cancer Center Pilot Grant, and Sloan Fellowship.

Sources: