[F]FDG PET/CT is useful in discriminating invasive adenocarcinomas among pure ground-glass nodules: comparison with CT findings-a bicenter retrospective study.

Purpose: Predicting the malignancy of pure ground-glass nodules (GGNs) using CT is challenging. The optimal role of [F]FDG PET/CT in this context has not been clarified. We compared the performance of [F]FDG PET/CT in evaluating GGNs for predicting invasive adenocarcinomas (IACs) with CT.

Methods: From June 2012 to December 2020, we retrospectively enrolled patients with pure GGNs on CT who underwent [F]FDG PET/CT within 90 days. Overall, 38 patients with 40 ≥ 1-cm GGNs were pathologically confirmed. CT images were analyzed for size, attenuation, uniformity, shape, margin, tumor-lung interface, and internal/surrounding characteristics. Visual [F]FDG positivity, maximum standardized uptake value (SUV), and tissue fraction-corrected SUV (SUV) were evaluated on PET/CT.

Results: The histopathology of the 40 GGNs were: 25 IACs (62.5%), 9 minimally invasive adenocarcinomas (MIA, 22.5%), and 6 adenocarcinomas in situ (AIS, 15.0%). No significant differences were found in CT findings according to histopathology, whereas visual [F]FDG positivity, SUV, and SUV were significantly different (P=0.001, 0.033, and 0.018, respectively). The size, visual [F]FDG positivity, SUV, and SUV showed significant diagnostic performance to predict IACs (area under the curve=0.693, 0.773, 0.717, and 0.723, respectively; P=0.029, 0.001, 0.018, and 0.013, respectively). In the multivariate logistic regression analysis, visual [F]FDG positivity discriminated IACs among GGNs among various CT and PET findings (P=0.008).

Conclusions: [F]FDG PET/CT demonstrated superior diagnostic performance compared to CT in differentiating IAC from AIS/MIA among pure GGNs, thus it has the potential to guide the proper management of patients with pure GGNs.