Pulmonary nodules diagnosis using breath test of volatile organic compound (VOC) is in its infancy. The accuracy of VOC analysis in diagnosing malignant pulmonary nodules varies cross the published studies. The diagnosis accuracy of VOC alone is generally poor. We speculate that the accuracy of diagnosing malignant pulmonary nodules will be improved by combining breath test with chest computed tomography (CT). This study aims to establish a predictive model of malignant pulmonary nodule using bio-markers from exhaled breath and image-markers from chest CT with retrospective data from multi centers. The sensitivity, specificity and accuracy of the model will be validated prospectively.

Endogenous volatile organic compounds (VOCs) can be derived from many different metabolic pathways. VOCs can be transported to the alveoli through the blood circulation and expelled by exhalation. Changes in VOCs production, clearance, and alterations in lung air-blood exchange functions can lead to aberrant VOCs profiles in the exhaled breath. Testing exhaled breath has the advantages of being completely non-invasive and easy to collect, and has been considered as a perfect approach for disease diagnoses and therapeutic monitoring. Many clinical studies have found that VOCs in exhaled breath are closely related to disease status. Specific VOCs alterations have been identified in many tumors, especially lung cancer. Pulmonary nodules diagnosis using breath test of volatile organic compound (VOC) is in its infancy. The accuracy of VOC analysis in diagnosing malignant pulmonary nodules varies cross the published studies. The diagnosis accuracy of VOC alone is generally poor. We speculate that the accuracy of diagnosing malignant pulmonary nodules will be improved by combining VOC analysis with chest computed tomography.

In this study, we use a highly sensitive mass spectrometry to detect exhaled VOCs of patients with pulmonary nodules. The chest CT will be used for detecting the imaging characteristics of pulmonary nodules. The pathological diagnosis of pulmonary nodules after surgical resections is selected as golden standard.

This study aims to establish a predictive model of malignant pulmonary nodule using bio-markers from breath test and image-markers from chest CT with retrospective data from multi centers. The sensitivity, specificity and accuracy of the model will be varied prospectively.

Inclusion Criteria:

1. Age > 18 years old;
2. Pulmonary nodules with planned surgical resection;
3. Signed informed consent and agreed to participate in this study.

Exclusion Criteria:

1. Preoperative radiotherapy, chemotherapy, targeted therapy or other anti-tumor therapy
2. The lack of chest computed tomography within two weeks before surgery
3. A history of malignant disease within 5 years.

To establish pulmonary nodule diagnosis model of breath test combined with computed tomography, we reviewed the prospective database established during the study of NCT04419207 for participant recruitment. 200 participants with malignant pulmonary nodules and 200 patients with benign pulmonary nodules will be included for establishing the diagnosis model.

To validate the efficacy of the established model in distinguishing malignant pulmonary nodules from benign nodules, we will prospectively recruit 500 patients with pulmonary nodules in multi centers: Peking University People's Hospital (200 participants), The First Affiliated Hospital of Zhengzhou University (200 participants), and Beijing Haidian Hospital (100 participants).