Oregon Health & Science University researchers have developed a\u00a0first-of-its-kind method\u00a0to predict cancer patient survival using advanced\u00a0molecular\u00a0data from individual cells.\u00a0<\/p>\n
Survival analysis is central to clinical oncology.\u00a0Modern cancer studies can now measure gene activity in single cells from a patient’s tumor and link this information to how long patients live. However, until now, there has\u00a0not been\u00a0a\u00a0good way\u00a0to use this detailed cell-level data to directly predict survival.\u00a0<\/p>\n
The study,\u00a0published\u00a0today\u00a0in\u00a0Cancer Discovery, describes a\u00a0method\u00a0called\u00a0scSurvival\u00a0that uses single-cell genetic data to\u00a0identify\u00a0which cells inside a tumor are most strongly linked to patient survival. Unlike traditional methods that average\u00a0signals\u00a0across an entire tumor, the\u00a0new approach\u00a0pinpoints harmful and helpful cell populations that can drive disease progression.\u00a0The research team\u00a0presented these findings today at the American Association for Cancer Research conference.\u00a0<\/p>\n
“This is the first kind of single-cell survival analysis that directly links individual tumor cells to patient outcomes,” said\u00a0Tao Ren, Ph.D.,\u00a0co-lead\u00a0author of the study\u00a0and postdoctoral\u00a0fellow\u00a0specializing in mathematics\u00a0in\u00a0the OHSU School of Medicine. “It allows us to see which cells are really driving disease progression instead of treating all cells the same.”\u00a0<\/p>\n
Co-lead author\u00a0Faming Zhao, Ph.D.,\u00a0said the approach helps solve a long-standing problem in cancer research.\u00a0<\/p>\n
“Tumors are very complex, and important signals can be lost when data are averaged across thousands or millions of cells,”\u00a0said Zhao, a postdoctoral\u00a0fellow\u00a0specializing in cancer biology\u00a0in\u00a0the OHSU School of Medicine. “By looking at survival at single-cell resolution, we can better understand why patients with the same cancer can have very different outcomes.”\u00a0<\/p>\n
Expertise\u00a0across fields\u00a0<\/p>\n
In tests using melanoma<\/a> and liver cancer data, the tool more accurately predicted patient outcomes than standard methods. It also uncovered specific immune and tumor cell states tied to better or worse survival. For example, the researchers found certain immune cells that appear to help patients respond better to immunotherapy<\/a>, while others were linked to poorer outcomes.\u00a0<\/p>\n Senior author\u00a0Zheng Xia, Ph.D.,\u00a0associate professor of biomedical engineering in the\u00a0OHSU School of Medicine\u00a0and a member of the\u00a0OHSU Knight Cancer Institute,\u00a0said the work reflects both technical innovation and close collaboration across disciplines.\u00a0<\/p>\n “This study was made possible by strong collaboration\u00a0at the Knight Cancer Institute\u00a0between computational scientists, cancer biologists and clinicians,” Xia said. “By bringing together expertise from different fields, we were able to use artificial intelligence to develop a new way to study survival using single-cell data.”\u00a0<\/p>\n Xia said the model goes beyond traditional machine learning approaches by capturing complex biological patterns that were previously difficult to study.\u00a0<\/p>\n “This work uses artificial intelligence to develop a new way to study survival using single-cell data,” Xia said. “The model is more complex than traditional machine learning approaches, and it allows us to capture information that was not accessible before.”\u00a0<\/p>\n Understanding these differences matters for patients, researchers said, because tumors\u00a0are made up of many cell types that behave differently. Treatments that work for one patient may fail in another if harmful cell populations are missed.\u00a0<\/p>\n While\u00a0scSurvival\u00a0is not yet used in clinical care, the researchers say it could eventually help doctors better\u00a0identify\u00a0high-risk patients and support the development of more precise, targeted cancer therapies.\u00a0<\/p>\n The\u00a0open-source\u00a0scSurvival\u00a0program and its tutorials are freely available at\u00a0GitHub,\u00a0Zenodo\u00a0and\u00a0Code Ocean.\u00a0<\/p>\n In addition to Xia, OHSU co-authors on this study include\u00a0Canping\u00a0Chen, M.S.,\u00a0Le Zhou, Ph.D.,\u00a0Gordon Mills, M.D., Ph.D.,\u00a0Lisa Coussens, Ph.D.,\u00a0FAACR, FAIO.\u00a0<\/p>\n Source:<\/p>\n Oregon Health & Science University<\/a><\/p>\n Journal reference:<\/p>\n