Abstract
BACKGROUND: This study aimed to develop a comprehensive prediction model integrating polygenic risk scores (PRS) and clinical factors to identify individuals at high risk for incident idiopathic pulmonary arterial hypertension (IPAH).</p>
METHODS: A PRS was constructed using summary statistics derived from the largest genome-wide association study for pulmonary arterial hypertension in Europeans and validated in the UK Biobank (732 cases, 458,258 controls). After excluding individuals with IPAH at baseline, 316,073 participants (316 incident IPAH patients) were split into training and testing sets (7:3). Variable selection was performed using least absolute shrinkage and selection operator in the training set, and a prediction model for incident IPAH was established using the Cox proportional hazards model.</p>
RESULTS: A gradient increase in IPAH risk estimates was observed across polygenic risk strata (P for trend = 0.0035). Compared with individuals with low PRS, those with high PRS had a significant 38.5 % increased risk (OR: 1.385, 95 % CI: 1.105, 1.736). Eighteen predictors were selected and included in the comprehensive prediction model, achieving C-statistics of 0.803 (95 % CI: 0.774, 0.831) and 0.785 (95 % CI: 0.734, 0.836) in the training and testing sets, respectively. Following risk stratification using the prediction model, the high-risk group exhibited an 8.941-fold higher risk of incident IPAH compared to the low-risk group. Moreover, 14 of 18 independent factors of incident IPAH were further identified.</p>
CONCLUSION: The integrated prediction model effectively identifies individuals at high risk for IPAH, facilitating early detection and personalized interventions to reduce disease risk.</p>