Association of the biomarker bone morphogenetic protein 10 with adverse clinical outcomes

Background
The biomarker bone morphogenetic protein 10 (BMP10) belongs to the transforming growth factor β (TGF β) superfamily. BMP10 is primarily expressed in the cardiomyocytes of the right atrial chamber of the heart, which is a distinctive feature that sets it apart from other cardiac biomarkers. Furthermore, BMP10 is involved in cardiac development and structural changes in adults. BMP10 was recently identified as an essential biomarker in atrial fibrillation (AF), the most prevalent cardiac arrhythmia. BMP10 was associated with rhythm status in patients with AF undergoing cardioversion, and furthermore, an experimental study showed that BMP10 identifies patients with a higher risk of AF recurrence after catheter ablation (CA). The latter study revealed that BMP10 is regulated by the paired-like homeodomain transcription factor 2 (PITX2), an important gene strongly associated with AF and AF recurrence after CA. The identification of patients with a high risk for AF recurrence is fundamental in clinical routine, however, the validation of BMP10's association with AF recurrence after CA had not been performed. Moreover, AF patients remain at increased risk for mortality and morbidity compared to patients without AF. Identifying those high-risk patients is crucial, but current risk assessment tools are insufficient. The potential role of BMP10 in identifying such high-risk AF patients was unclear.
Recent evidence also suggests a role of BMP10 in pulmonary hypertension (PH), a severe and complex disease characterized by abnormally high pressure in the pulmonary arteries. Elevation of the pulmonary vascular resistance (PVR; pre-capillary component) and elevation of the left atrial filling pressure (pulmonary arterial wedge pressure, PAWP; post-capillary component) contribute to an increase of the mean pulmonary arterial pressure (mPAP). An invasive hemodynamic assessment by means of right heart catheterization (RHC) is currently the only way to confirm diagnosis and distinguish between pre- and post-capillary PH. Pulmonary vascular remodeling is a major factor in the pathophysiology of PH. BMP10 is a ligand for two important receptors in PH, bone morphogenetic protein receptor type 2 (BMPR2) and activin receptor-like kinase-1 (ALK1). BMP10 was found to contribute to the tonic quiescent signals to the pulmonary vascular endothelium trough those receptors and it also directly acts on vascular smooth muscle cells for induction and maintenance of their contractile state. Furthermore, BMP10 was designated as a new gene in heritable PAH. Whether BMP10 is associated with PVR because of its role in pulmonary vascular remodeling and therefore reflects the pre-capillary component of PH was unknown.
Aims
As a blood-based biomarker, BMP10 can provide an option for the diagnosis, monitoring, and prognosis of disease. However, further research is necessary to understand the role of BMP10 in patients with AF and PH and to assess its value in the clinical setting. This was thus the overall aim of this PhD project. The specific aims were: 1. To validate the association of BMP10 with AF recurrence after CA in a large cohort of AF patients, 2. To explore the association of BMP10 with AF-related adverse outcomes, namely all-cause death and major adverse cardiovascular events (MACE), 3. To explore the association of BMP10 with hemodynamic parameters which define the diagnosis of PH (mPAP), and the pre- and post-capillary component of PH (PVR, PAWP).
Methods
We measured BMP10 concentrations in three observational studies: Swiss-AF-PVI, Swiss-AF, and PH Study. For the first specific aim, we analyzed patients from Swiss-AF-PVI, an ongoing prospective cohort study since 2010, which includes patients ≥ 18 years undergoing CA for AF. Venous blood samples were obtained before CA. Patients received follow-ups at 3, 6, and 12 months after CA, including a 12-lead electrocardiogram (ECG) and a 7-day Holter ECG to assess AF recurrence. For the second specific aim, we analyzed patients from Swiss-AF, an ongoing prospective multicenter cohort study since 2014, which includes patients ≥ 65 years (and a limited number of younger patients) with diagnosed AF. Venous blood samples were obtained at baseline, and patients received yearly follow-up visits to assess death and MACE. For the third specific aim, we analyzed patients from the PH Study, a prospective single-center cross-sectional study, which includes patients ≥ 18 years undergoing RHC for suspected or established PH between 2021-2022. Venous blood samples were obtained during RHC. We performed multivariable cox proportional hazard and linear regression analyses to assess the association of BMP10 with the different outcomes of interest.
Results
In the first analysis, we included 1112 AF patients undergoing CA (mean ± SD age 61 ± 10 years, 74% male, 60% paroxysmal AF). 374 patients (34%) experienced AF recurrence within 12 months of follow-up. In the multivariable cox proportional hazard model, a per-unit increase in log-transformed BMP10 was associated with a hazard ratio (HR) of 1.98 (95% CI 1.14 - 3.42, p = 0.01) for AF recurrence. In the second analysis, we included 2219 AF patients (median follow-up 4.3 years [interquartile range 3.9, 5.1], mean ± SD age 73 ± 9 years, 73% male). In the multivariable cox proportional hazard models, the HR associated with 1 ng/mL increase of BMP10 was 1.60 (95% CI 1.37 - 1.87, p < 0.001) for all-cause death, and 1.54 (95% CI 1.35 - 1.76, p < 0.001) for MACE. The concordance index of this model was 0.78 (95% CI 0.76 - 0.81) for all-cause death, and 0.73 (95% CI 0.72 - 0.75) for MACE. For the third analysis, we included 127 patients with suspected or established PH undergoing RHC (mean ± SD age 66 ± 13 years, 46% male). PH was diagnosed in 93 (73%) patients. In the multivariable linear regression model, a per standard deviation increase of log-transformed BMP10 was associated with a beta coefficient of 5.05 (95% CI 2.58 - 7.52, p < 0.001) for mPAP, 0.29 (95% CI 0.15 - 0.43, p < 0.001) for PVR, and -0.10 (95% CI -1.29 - 1.08, p = 0.86) for PAWP.
Conclusion
BMP10 was strongly associated with AF recurrence after CA as well as with all-cause death and MACE in AF patients. Randomized controlled trials are necessary to investigate if a BMP10-guided risk prediction and management can improve outcomes. In patients undergoing RHC, BMP10 was strongly associated with mPAP and PVR (reflecting the pre-capillary component of PH), but not with the left atrial filling pressure (PAWP), which reflects the post-capillary component of PH. Further studies should investigate the combination of BMP10 with non-invasive imaging for the diagnosis and monitoring of PH.