Researchers at the University of Tokyo have demonstrated at a population scale that insulin resistance is a risk factor for 12 types of cancer, according to a study published in Nature Communications.
The study applied a machine learning tool to data from roughly half a million participants in the UK Biobank. The tool, called AI-IR (artificial intelligence-derived insulin resistance), uses nine standard clinical parameters collected during routine health checkups to predict insulin resistance.
Direct measurement typically requires specialized testing found only in advanced diabetes clinics, making large-scale evaluation impractical. The model combines these parameters into a single metric to detect insulin resistance that body mass index (BMI) alone cannot explain. Clinicians often use BMI as a proxy, but it produces false positives and false negatives: some people with obesity remain metabolically healthy, while others with normal BMI develop insulin resistance.
AI-IR outperformed BMI, metabolic syndrome criteria, and other established markers in predicting diabetes incidence. In validation using independent cohorts from the United States and Taiwan, the tool achieved strong predictive performance when compared with directly measured insulin resistance.
Applying AI-IR to the UK Biobank population, researchers found that participants classified as AI-IR positive but without diabetes still faced elevated cancer risk compared to those who tested negative. “While a possible link between insulin resistance and cancer has been suggested, large-scale evidence has been limited due to the difficulty of evaluating insulin resistance in the clinic,” said Yuta Hiraike, a researcher at the University of Tokyo Hospital who led the study. “But with AI-IR, we have provided the first population-scale evidence that insulin resistance is a risk factor for cancer.”
Because the inputs come from standard health checkups, the tool could enable widespread screening for diabetes, cardiovascular disease, and cancer without requiring specialized clinic visits. The team is now investigating how genetic differences influence insulin-resistance-related cancer risk, aiming to link large-scale human data with molecular biology to develop strategies for overcoming insulin resistance.
