Development and validation of a 18F-FDG PET/CT-based clinical prediction model for estimating malignancy in solid pulmonary nodules based on a population with high prevalence of malignancy

Abstract Objective To develop a prediction model based on 18F-FDG positron emission tomography/computed tomography (PET/CT) for solid pulmonary nodules (SPNs) with high malignant probability. Methods We retrospectively reviewed the records of CT-undetermined SPNs, which were further evaluated by PET/CT between January 2008 and December 2015. 312 cases were included as training set and 159 as validation set. Logistic regression was applied to determine independent predictors and a mathematical model was deduced. Area under the receiver operating characteristic curve (AUC) was compared with other models. Model fitness was assessed based on the ACCP guidelines. Results There were 215 (68.9%) and 127 (79.9%) malignant lesions in training and validation set, respectively. Eight independent predictors were identified: age [odds ratio (OR): 1.030], male gender (OR: 0.268), smoking history (OR: 2.719), lesion diameter (OR: 1.067), spiculation (OR: 2.530), lobulation (OR: 2.614), cavity (OR: 2.847), the maximum standardized uptake value (SUVmax) of SPN (OR: 1.229). Our AUCs (training set: 0.858, validation set: 0.809) was superior to those of previous models (Mayo: 0.685, p=0.0061; PKUPH: 0.646, p=0.0180; Herder: 0.708, p=0.0203; ZU: 0.757, p=0.0699). The C-index of the nomogram was 0.858. Our model reduced the diagnosis of indeterminate nodules (26.4% vs. 79.2%, 53.5%, 39.6%, 34.0%, respectively) while improved sensitivity (81.3% vs. 16.4%, 49.2%, 62.5%, 68.0%, respectively) and accuracy (65.4% vs. 16.4%, 39.6%, 52.8%, 58.5% respectively). Conclusion Our model could raise man accurate diagnoses and was recommended in identifying malignant SPNs with high malignant probability as our data pertain to a very high prevalence cohort only.