Associations between sex steroid hormones and female malignancies: A cross-sectional analysis of NHANES data.

Sex steroid hormones, including estradiol (E2), estrone (E1), follicle-stimulating hormone (FSH), and luteinizing hormone (LH), have been implicated in the development of female malignancies (FMs), yet their independent and combined effects remain unclear. We conducted a cross-sectional analysis using data from the National Health and Nutrition Examination Survey (NHANES). Weighted logistic regression, restricted cubic spline modeling, and chi-square trend tests were applied to evaluate dose-response associations. Machine learning algorithms, including Bayesian Kernel Machine Regression, Extreme Gradient Boosting, Shapley Additive Explanations, Least Absolute Shrinkage and Selection Operator regression, and Random Forest, were used to explore nonlinear relationships and assess variable importance in an exploratory manner. Sensitivity analyses confirmed the robustness of the findings. A total of 2108 participants were included. Higher FSH and LH levels were associated with increased odds of FMs, whereas higher E2 levels were associated with lower odds. In exploratory machine learning analyses, FSH and E1 were consistently ranked as important predictors of FMs. In this cross-sectional analysis, LH and FSH were associated with increased odds of FM, while higher E2 levels were inversely associated with malignancy risk. Exploratory machine learning analyses highlighted additional nonlinear patterns involving E2 and E1.