Background: Diabetes is a well-established risk factor for multiple cancers, with underlying hyperglycemia and insulin resistance serving as potential carcinogenic pathways. However, the evidence about the association between fasting plasma glucose (FPG) or insulin resistance indicators and malignant tumors remains largely unclear in large-scale populations. Besides, the optimal FPG cut-off value for cancer prevention in non-diabetic and diabetic individuals need to be explored.
Objectives: We aimed to explore the association between diabetes, blood glucose level and insulin resistance with cancer risk. Besides, we aimed to identify the optimal blood glucose level for diabetes and non-diabetic participants.
Methods: The prospective cohort study enrolled participants aged 45 years and older who underwent health examinations between August 2020 and December 2021 in Wuxi. Participants were categorized into non-diabetic and diabetic groups based on diabetes history and FPG levels. FPG were categorized according to WHO classification: FPG < 6.1 mmol/L defined as normal glucose, 6.1 ≤ FPG < 7.0 mmol/L as impaired fasting glucose (IFG) for non-diabetic participants. FPG < 7.0 mmol/L as well-controlled glucose, while FPG ≥ 7.0 mmol/L as poorly controlled glucose for diabetes patients. The study outcome was defined as newly diagnosed malignant tumors based on information-based follow-up. Cox model was used to evaluate the association between FPG/ MetS-IR and malignant cancer risk. Piecewise Linear Regression model (PLR) was used to identify the optimal cutoff value for cancer prevention, by selcting the 10th to 90th percentiles of the FPG distribution based on the Akaike Information Criterion (AIC) while considering practical significance.
Results: The study included 402,573 participants with a mean age of 68.76 ± 8.01 years. The median follow-up was 3.12 years (IQR: 2.55 - 3.23), with a cumulative follow-up of 1,158,800 person years. A total of 13,189 new malignant tumors were diagnosed during follow-up. Among non-diabetic individuals, impaired fasting glucose was associated with an increased risk of overall cancer (HR = 1.08, 95% CI: 1.02 - 1.13), and gastric cancer, colorectal cancer, liver cancer, gallbladder cancer and pancreatic cancer (HR ranged from 1.13 to 1.99, respectively). MetS-IR was associated with increased risks of overall cancer (HR=1.08, 95% CI: 1.01 - 1.15), colorectal cancer, female breast cancer, endometrial cancer and lymphoma (HR ranged from 1.38 to 2.03, respectively). Among diabetic individuals, similar results were observed for poor glycemic control or higher level of MetS-IR. We also observed that, the optimal cut-off value of FPG for cancer prevention should be 5.7 and 6.8 mmol/L in non-diabetic and diabetic individuals, respectively.
Conclusion: FPG exhibits a U-shaped association with the risk of overall caner, indicating FPG should be kept in optimal level for cancer prevention in both non-diabetic and diabetic individuals. MetS-IR also could be used to identify high-risk cancer populations and formulating precise prevention strategies.

The risk effect values of non-diabetic and diabetic groups with reference to normal FPG and well-controlled FPG