Predicting the prognosis of patients with ovarian cancer just became more precise thanks to AI software developed by researchers at Imperial College London and the University of Melbourne in Australia.
The software, which analyzes tumors on CT scans, was up to four-times more accurate at predicting ovarian cancer deaths compared to standard methods, according to research published Feb. 15 in Nature Communications.
The method may be able to stratify patients based on small differences in a cancer’s texture on CTs, rather than classifying based on cancer type or stage.
"The long-term survival rates for patients with advanced ovarian cancer are poor despite the advancements made in cancer treatments. There is an urgent need to find new ways to treat the disease,” said Eric Aboagye, with Imperial College of London and lead author of the research, in a news release. “Our technology is able to give clinicians more detailed and accurate information on the how patients are likely to respond to different treatments, which could enable them to make better and more targeted treatment decisions."
In this study, Aboagye et al. used the AI software—TEXLab—to calculate the aggressiveness of tumors in CT scans and tissue samples from 364 women diagnosed with ovarian cancer between 2004 and 2015. To assess for prognosis, four biological hallmarks of the tumors were analyzed: structure, shape, size and genetic makeup. Patients were given a Radiomic Prognostic Vector (RPV) score which indicates how severe their disease was.
Compared with blood tests and current prognostic scores used by clinicians, the AI software was up to four times more accurate at estimating survival. Additionally, five percent of patients with high RPV scores had a survival rate of less than two years. Higher scores were also concurrent with chemotherapy resistance and poor surgical outcomes; therefore RPV may be used as a biomarker to predict response to treatments, according to the study.
"Artificial intelligence has the potential to transform the way healthcare is delivered and improve patient outcomes. Our software is an example of this and we hope that it can be used as a tool to help clinicians with how to best manage and treat patients with ovarian cancer,” study co-author Andrea Rockall, honorary consultant radiologist at Imperial College Healthcare NHS Trust, said in the same release.