Serum and Pulmonary Expression Profiles of the Activin Signaling System in Pulmonary Arterial Hypertension

Background: Activins are novel therapeutic targets in pulmonary arterial hypertension (PAH). We therefore studied whether key members of the activin pathway could be used as PAH biomarkers. Methods: Serum levels of activin A, activin B, α-subunit of inhibin A and B proteins, and the antagonists follistatin and follistatin-like 3 (FSTL3) were measured in controls and in patients with newly diagnosed idiopathic, heritable, or anorexigen-associated PAH (n=80) at baseline and 3 to 4 months after treatment initiation. The primary outcome was death or lung transplantation. Expression patterns of the inhibin subunits, follistatin, FSTL3, Bambi, Cripto, and the activin receptors type I (ALK), type II (ACTRII), and betaglycan were analyzed in PAH and control lung tissues. Results: Death or lung transplantation occurred in 26 of 80 patients (32.5%) over a median follow-up of 69 (interquartile range, 50–81) months. Both baseline (hazard ratio, 1.001 95% CI, 1.000–1.001; P =0.037 and 1.263 95% CI, 1.049–1.520; P =0.014, respectively) and follow-up (hazard ratio, 1.003 95% CI, 1.001–1.005; P =0.001 and 1.365 95% CI, 1.185–1.573; P <0.001, respectively) serum levels of activin A and FSTL3 were associated with transplant-free survival in a model adjusted for age and sex. Thresholds determined by receiver operating characteristic analyses were 393 pg/mL for activin A and 16.6 ng/mL for FSTL3. When adjusted with New York Heart Association functional class, 6-minute walk distance, and N-terminal pro-B-type natriuretic peptide, the hazard ratios for transplant-free survival for baseline activin A <393 pg/mL and FSTL3 <16.6 ng/mL were, respectively, 0.14 (95% CI, 0.03–0.61; P =0.009) and 0.17 (95% CI, 0.06–0.45; P <0.001), and for follow-up measures, 0.23 (95% CI, 0.07–0.78; P =0.019) and 0.27 (95% CI, 0.09–0.78, P =0.015), respectively. Prognostic values of activin A and FSTL3 were confirmed in an independent external validation cohort. Histological analyses showed a nuclear accumulation of the phosphorylated form of Smad2/3, higher immunoreactivities for ACTRIIB, ALK2, ALK4, ALK5, ALK7, Cripto, and FSTL3 in vascular endothelial and smooth muscle layers, and lower immunostaining for inhibin-α and follistatin. Conclusions: These findings offer new insights into the activin signaling system in PAH and show that activin A and FSTL3 are prognostic biomarkers for PAH.