A model of malignant risk prediction for solitary pulmonary nodules on ¹⁸F‐FDG PET/CT: Building and estimating

Abstract

Background: To develop a model of malignant risk prediction of solitary pulmonary nodules (SPNs) using metabolic characteristics of lesions. Methods: A total of 362 patients who underwent PET/CT imaging from January 2013 to July 2017 were analyzed. Differences in the clinical and imaging characteristics were analyzed between patients with benign SPNs and those with malignant SPNs. Risk factors were screened by multivariate nonconditional logistic regression analysis. The self-verification of the model was performed by receiver operating characteristic (ROC) curve analysis, and out-of-group verification was performed by k-fold cross-validation. Results: There were statistically significant differences in age, maximum standardized uptake value (SUV max ), size, lobulation, spiculation, pleural traction, vessel connection, calcification, presence of vacuoles, and emphysema between patients with benign nodules and those with malignant nodules (all P < 0.05). The risk factors for malignant nodules included age, SUV max , size, lobulation, calcification and vacuoles. The logistic regression model was as follows: P = l/(1 + e -x ), x = − 5.583 + 0.039 × age + 0.477 × SUV max + 0.139 × size + 1.537 × lobulation -1.532 × calcification + 1.113 × vacuole. The estimated area under the curve (AUC) for the model was 0.915 (95% CI: 0.883-0.947), the sensitivity was 89.7%, and the specificity was 78.9%. K-fold cross-validation showed that the training accuracy was 0.899 AE 0.011, and the predictive accuracy was 0.873 AE 0.053. Conclusions: The risk factors for malignant nodules included age, SUV max , size, lobulation, calcification and vacuoles. After verification, the model has satisfactory accuracy, and it may assist clinics make appropriate treatment decisions.

Clinical and CT Radiomics Nomogram for Preoperative Differentiation of Pulmonary Adenocarcinoma From Tuberculoma in Solitary Solid Nodule

“…Vacuole sign refers to the presence of a small air-like low-density shadow within the nodule with smooth edges and no more than 5 mm. Vacuole sign is often seen in malignant nodules, which is an important sign of early lung cancer, but also occasionally seen in benign nodules ( 28 , 29 ). There were statistically significant differences in nodule size between the two groups, and Chu et al.…”

Clinical Analysis of Video-Assisted Thoracoscopic Surgery for Resection of Solitary Pulmonary Nodules and Influencing Factors in the Diagnosis of Benign and Malignant Nodules

“…Malignant SPN imaging usually shows a gravelly or diffuse distribution with a small extent, while benign SPN imaging shows smooth, intact margins with clear contours [ 21 , 22 ]. Studies [ 23 ] have pointed out that the signs of malignant lesions in pulmonary nodules include pleural traction, burr sign, lobular sign, tumor vascular sign, bronchus truncation sign, etc., and the results of this study are generally consistent with them.…”

Prediction of malignancy for solitary pulmonary nodules based on imaging, clinical characteristics and tumor marker levels

“…Vacuole sign, unlike air bronchogram, is the area of low attenuation owing to small patent air containing bronchi in nodules. The presence of vacuole sign is also reported to be correlated with malignant pulmonary nodules ( Snoeckx et al ., 2018 ; Yu et al ., 2020 ). In most studies, nodule size is found to be important in the prediction of benign and malignant SPNs ( Yu et al ., 2020 ; Wang et al, 2018a , b ).…”

“…The presence of vacuole sign is also reported to be correlated with malignant pulmonary nodules ( Snoeckx et al ., 2018 ; Yu et al ., 2020 ). In most studies, nodule size is found to be important in the prediction of benign and malignant SPNs ( Yu et al ., 2020 ; Wang et al, 2018a , b ). However, in the present study, it was not identified as a parameter of the model, which might be associated with study populations and sample size.…”

“…Third, it was of great significance to enhance the recognition of patients at high risk of malignancy through this prediction model. Although the high proportion of nonsmokers (67.4%) in our study was similar as several Chinese studies, it was not representative for other countries ( Gould et al ., 2013 ; Truong et al ., 2014 ; Yu et al ., 2020 ). Additionally, the establishment of the model was based on a retrospective study with relatively small sample sizes; the model was not simply enough for clinician use and lack of external validation.…”