Activins, follistatins and inhibins in postmenopausal osteoporosis: A proof of concept, case-control study☆

Background: Bone metabolism has been proposed to be affected by the activins-follistatins-inhibins (AFI) hormonal system. We aimed to evaluate AFI in patients with osteoporosis and osteopenia compared with postmenopausal and premenopausal controls. Methods: In this case-control study, circulating levels of the AFI system were evaluated, individually and jointly, between postmenopausal women with osteoporosis (BMD T-score <=-2.5; n = 25) or osteopenia (BMD T-score >-2.5 and <=-1; n = 25) and postmenopausal women with normal BMD (T-score >-1.0; n = 25) or premenopausal women with normal BMD (Z-score >-1.0; n = 25), with and without adjustment for potential confounders. Results: In the sum of participants, AFI molecules and their ratios followed an opposite pattern of correlations for age and BMI vs. BMD. In unadjusted models, FSTL3 concentrations were higher, whereas activin B, inhibin A and inhibin B and the ratios of activin B/follistatin and activin B/FSTL3 were lower in the three postmenopausal groups compared with the premenopausal group. Activin A/follistatin and activin AB/follistatin ratios were lower in the osteoporosis group than the other three groups. After adjustment for BMI and age, inhibin B (p = 0.005), and the ratios of activin A/follistatin (p = 0.009), activin B/follistatin (p = 0.040) and activin AB/follistatin (p = 0.003) were lower in the osteoporotic group compared with the other groups. In fully adjusted logistic regression analysis log(inhibin B) (p = 0.041), log(activinA/follistatin) (p = 0.014), log(activinB/follistatin) (p = 0.025) and log(activinAB/follistatin) (p = 0.021), but not FSTL3, remained independently associated with the presence of osteoporosis. Conclusions: Lower inhibin B and higher ratios of activins A, B, and AB to follistatin are associated with lumbar spine BMD and the presence of osteoporosis independently from age or BMI.