Xia He1, Ning Xue2, Xiaohua Liu1, Qingxia Xu2, Wanli Liu1, Shulin Chen3,4, Xuemiao Tang1, Songguo Peng1, Yuanye Qu2, Lina Jiang5. 1. State Key Laboratory of Oncology in South China, Collaborative Innovation Center for Cancer Medicine, Guangdong Key Laboratory of Nasopharyngeal Carcinoma Diagnosis and Therapy, Sun Yat-sen University Cancer Center, 651 Dongfeng Road East, 510060, Guangzhou, People's Republic of China. 2. Department of Clinical Laboratory, Affiliated Cancer Hospital of Zhengzhou University, Zhengzhou Key Laboratory of Digestive Tumor Markers, Henan, 450008, Zhengzhou, People's Republic of China. 3. State Key Laboratory of Oncology in South China, Collaborative Innovation Center for Cancer Medicine, Guangdong Key Laboratory of Nasopharyngeal Carcinoma Diagnosis and Therapy, Sun Yat-sen University Cancer Center, 651 Dongfeng Road East, 510060, Guangzhou, People's Republic of China. chenshl@sysucc.org.cn. 4. Research Center for Translational Medicine, the First Affiliated Hospital, Sun Yat-sen University, 58 Zhongshan Road 2, Guangdong, 510080, Guangzhou, People's Republic of China. chenshl@sysucc.org.cn. 5. Department of Radiology , Affiliated Tumor Hospital of Zhengzhou University , Henan, 450008, Zhengzhou, People's Republic of China.
Abstract
BACKGROUND: This study aimed to establish and validate a novel clinical model to differentiate between benign and malignant solitary pulmonary nodules (SPNs). METHODS: Records from 295 patients with SPNs in Sun Yat-sen University Cancer Center were retrospectively reviewed. The novel prediction model was established using LASSO logistic regression analysis by integrating clinical features, radiologic characteristics and laboratory test data, the calibration of model was analyzed using the Hosmer-Lemeshow test (HL test). Subsequently, the model was compared with PKUPH, Shanghai and Mayo models using receiver-operating characteristics curve (ROC), decision curve analysis (DCA), net reclassification improvement index (NRI), and integrated discrimination improvement index (IDI) with the same data. Other 101 SPNs patients in Henan Tumor Hospital were used for external validation cohort. RESULTS: A total of 11 variables were screened out and then aggregated to generate new prediction model. The model showed good calibration with the HL test (P = 0.964). The AUC for our model was 0.768, which was higher than other three reported models. DCA also showed our model was superior to the other three reported models. In our model, sensitivity = 78.84%, specificity = 61.32%. Compared with the PKUPH, Shanghai and Mayo models, the NRI of our model increased by 0.177, 0.127, and 0.396 respectively, and the IDI changed - 0.019, -0.076, and 0.112, respectively. Furthermore, the model was significant positive correlation with PKUPH, Shanghai and Mayo models. CONCLUSIONS: The novel model in our study had a high clinical value in diagnose of MSPNs.
BACKGROUND: This study aimed to establish and validate a novel clinical model to differentiate between benign and malignant solitary pulmonary nodules (SPNs). METHODS: Records from 295 patients with SPNs in Sun Yat-sen University Cancer Center were retrospectively reviewed. The novel prediction model was established using LASSO logistic regression analysis by integrating clinical features, radiologic characteristics and laboratory test data, the calibration of model was analyzed using the Hosmer-Lemeshow test (HL test). Subsequently, the model was compared with PKUPH, Shanghai and Mayo models using receiver-operating characteristics curve (ROC), decision curve analysis (DCA), net reclassification improvement index (NRI), and integrated discrimination improvement index (IDI) with the same data. Other 101 SPNs patients in Henan Tumor Hospital were used for external validation cohort. RESULTS: A total of 11 variables were screened out and then aggregated to generate new prediction model. The model showed good calibration with the HL test (P = 0.964). The AUC for our model was 0.768, which was higher than other three reported models. DCA also showed our model was superior to the other three reported models. In our model, sensitivity = 78.84%, specificity = 61.32%. Compared with the PKUPH, Shanghai and Mayo models, the NRI of our model increased by 0.177, 0.127, and 0.396 respectively, and the IDI changed - 0.019, -0.076, and 0.112, respectively. Furthermore, the model was significant positive correlation with PKUPH, Shanghai and Mayo models. CONCLUSIONS: The novel model in our study had a high clinical value in diagnose of MSPNs.
Authors: Peter M Burch; Warren E Glaab; Daniel J Holder; Jonathan A Phillips; John-Michael Sauer; Elizabeth G Walker Journal: Toxicol Sci Date: 2017-03-01 Impact factor: 4.849