Serum Osteoprotegerin and Risk of Diabetes: The Hong Kong Osteoporosis Study

A recent study showed that osteoprotegerin (OPG), an important regulator of bone resorption, was able to stimulate human beta-cell proliferation (1), thus implying an important role in diabetes development and treatment. Before conducting clinical trial, it is valuable to evaluate the association in observational study. Our objective was to evaluate the relationship between serum OPG and risk of diabetes. In this observational study, 1,953 participants of the Hong Kong Osteoporosis Study (2) who were free of diabetes at baseline were included in the analysis. Serum OPG level was used in multivariate Cox regression analysis to predict diabetes risk. During a median follow-up of 10.3 years (range 0.1–18.2 years) and 18,983 person-years, 235 (12%) participants developed diabetes. In the crude model (Table 1), participants in quartiles 3 (HR=1.85; 95% CI: 1.25-2.72) and 4 (HR=2.39; 95% CI: 1.64-3.5) of serum OPG levels had a higher risk of incident diabetes when compared to quartile 1 (Trend P-value<0.001). One natural-log unit of serum OPG was positively associated with incident diabetes (HR=1.89; 95% CI: 1.49-2.39). However, such association became insignificant after adjustment of age, sex, and BMI. Similar null association was observed after further adjustment for confounding factors (Table 1). In conclusion, serum OPG was not associated with risk of diabetes. However, whether denosumab can be used to treat diabetes or improve glycemic control requires further study.

Table 1. Association of serum OPG and incident diabetes.

Model 1: Adjusted for age, sex, and BMI; Model 2: Further adjusted for lifestyle factors (smoking, drinking, education, physical activity); Model 3: Further adjusted for serum calcium, serum phosphate, serum parathyroid hormone, prevalent osteoporosis/osteopenia at the hip, serum albumin, serum alkaline phosphatase, prevalent chronic kidney disease, and season.

(1) Kondegowda NG et al. (2015) Cell Metab 22: 77-85

(2) Cheung CL et al. (2012) Age (Dordr) 34: 1239-1248