Identify malignant pulmonary nodules-associated circRNAs and develop a prediction model to estimate the probability of malignancy in pulmonary nodules.

Background: The increasing incidence of asymptomatic pulmonary nodules (PNs) underscores the need for accurate malignancy estimation to guide early-stage lung cancer management. This study aimed to identify risk factors for malignant PNs and develop a risk prediction model. Methods: This study enrolled 691 patients with PNs (444 training, 247 validation) and 62 healthy controls. Clinical, imaging, and serum data were collected. Transcriptome sequencing was performed to identify circular RNAs (circRNAs) associated with malignant PNs. Additionally, Tumor-associated antigens and tumor-associated autoantibodies (AAbs) were measured. Univariate logistic regression analysis was employed to identify risk factors for malignant PNs, followed by multivariate logistic regression to establish a risk prediction model. Finally, receiver operating characteristic (ROC) curve analysis was performed to evaluate the diagnostic performance of the risk prediction model. Results: Among five tumor-associated antigens, only neuron-specific enolase (NSE) levels were significantly higher in malignant versus benign PNs (P<0.001). AAb positivity rates and number of positive AAbs were elevated in malignant PNs. Transcriptome sequencing revealed hsa_circCFLAR_008 was upregulated in malignant PNs (P=0.04). The risk prediction model was established [area under the curve (AUC), 0.8241], and was validated with a concordance statistic (C-statistic) of 0.8344 in an independent cohort. Conclusions: A risk prediction model for malignant PNs was established. Hsa_circCFLAR_008 enhanced the diagnostic performance of the model and served as a novel biomarker for malignant PNs.