Abstract Body mass index (BMI) may misclassify obesity-related cancer (ORC) risk, as metabolic dysfunction can occur across BMI levels. We hypothesized that metabolic dysfunction at any BMI increases ORC risk compared to normal BMI without metabolic dysfunction. Postmenopausal women (n=20,593) in the Women’s Health Initiative with baseline metabolic dysfunction biomarkers (blood pressure, fasting triglycerides, high-density lipoprotein-cholesterol, fasting glucose, Homeostatic Model Assessment for Insulin Resistance (HOMA-IR) and high sensitive C-reactive protein (hs-CRP)) were included. Metabolic phenotype (metabolically healthy normal weight (MHNW), metabolically unhealthy normal weight (MUNW), metabolically healthy overweight/obese (MHO), metabolically unhealthy overweight/obese (MUO)) was classified using four definitions of metabolic dysfunction: (1) Wildman criteria, (2) National Cholesterol Education Program (NCEP) Adult Treatment Panel III (ATP III), (3) HOMA-IR, and (4) hs-CRP. Multivariable Cox proportional hazards regression, with death as a competing risk, was used to assess the association between metabolic phenotype and ORC risk. After a median (IQR) follow-up duration of 21 (IQR 15-22) years, 2,367 women developed an ORC. The risk of any ORC was elevated among MUNW (HR 1.12, 95% CI: 0.90-1.39), MHO (HR 1.15, 95% CI: 1.00-1.32), and MUO (HR 1.35, 95% CI: 1.18-1.54) compared with MHNW using Wildman criteria. Results were similar using ATP III criteria, hs-CRP alone, or HOMA-IR alone to define metabolic phenotype. Individuals with overweight or obesity with or without metabolic dysfunction were at higher risk of ORCs compared with metabolically healthy normal weight individuals. The magnitude of risk was greater among those with metabolic dysfunction, although the confidence intervals of each category overlapped.

Metabolic phenotype and risk of obesity-related cancers in the Women's Health Initiative