Complications in the timely, accurate screening of ovarian cancer mean many women receive a diagnosis when in the advanced stages of the disease. This frequently makes ovarian cancer a fatal diagnosis. While researchers have made significant advances in the development of various biomarker-driven tests for ovarian cancer, such procedures have yet to offer a reliable diagnosis without being performed multiple times. In the January 2024 publication of The Lancet Digital Health, Guangyao Cai and colleagues explored a novel application of these detection techniques by leveraging artificial intelligence (AI) to improve their diagnostic accuracy. The authors suggest that AI-enhanced ovarian cancer testing offers an avenue for cost-effective, widely available, and accurate diagnostic procedures. GlobalData epidemiologists forecast an increase in the diagnosed incident cases of ovarian cancer from over 585,600 cases to approximately 623,600 cases between 2024 and 2028 among adult women in the 16 major markets (16MM: US, France, Germany, Italy, Spain, UK, Japan, Australia, Brazil, Canada, China, India, Mexico, Russia, South Africa, and South Korea). If the AI-assisted diagnosis of ovarian cancer as proposed by Cai and colleagues is widely adopted by physicians in these markets, the incident cases of ovarian cancer may increase due to higher diagnosis rates. However, the number of prevalent cases may also grow due to improved survival attributed to fewer diagnostic delays.

Cai and colleagues explored the application of AI to ovarian cancer diagnosis through a retrospective study between 2012 and 2021. Drawing from three major hospitals based in China, the authors trained 20 AI predictive models using 6,778,762 laboratory results screening for ovarian cancer. Each predictive model was trained to produce an accurate diagnosis of ovarian cancer, and the model’s decision-making was based on 52 biomarkers. The predictions made by the model were validated against three datasets of laboratory results. The performance of the model in accurately predicting positive cases exceeded laboratory tests with and without the conventional carbohydrate antigen 125 (CA125) and human epididymal protein 4 (HE4) biomarkers most commonly associated with ovarian cancer. Importantly, these capabilities included the improved prediction of early-stage ovarian cancer, a result which frequently eludes traditional testing methods due to deficiencies in the specificity and sensitivity of current biomarker tests. In order to ensure access to the model for public review and potential utilization, Cai and colleagues have published it as an open-source tool on the internet.

The predictive model developed by Cai and colleagues serves as a promising first step toward strengthened capabilities in the accurate, timely diagnosis of ovarian cancer. While the authors acknowledge the need for further exploration of the model’s utility in clinical settings, as well as refinement of the model’s algorithms through its application, its publication online for free use can make it a powerful tool that physicians can leverage in conjunction with traditional methods of detection without the need for additional equipment or procedures. As expected for other clinical applications of AI, the increased adoption of AI-enhanced ovarian cancer detection could have resounding effects on patients and clinicians, but also on the ovarian cancer epidemiological landscape more broadly. While the incident cases of ovarian cancer could increase due to higher diagnosis rates, strengthened early detection could shift the stage distribution of diagnosed cases toward earlier stages, and potentially lead to improved odds of survival.