Narae Lee, Hyukjin Yoon, Ie Ryung Yoo
18F-fluorodeoxyglucose (FDG) positron emission tomography (PET)/computed tomography (CT) is widely used for oncologic staging, restaging, and response assessment. Its CT component frequently reveals pulmonary lesions that are not the principal target of the examination. In patients with known malignancy, such lesions may have a higher pre-test probability of malignancy than incidental pulmonary lesions detected in the general population and cannot be reliably characterized by CT morphology or FDG uptake alone. Small lesion size, partial-volume effects, respiratory motion, and low tumor cellularity may render metastases metabolically occult, whereas infectious and inflammatory processes may show intense FDG uptake and mimic malignancy. This review presents a case-based, lesion-by-lesion approach integrating thin-section CT morphology, concordance or discordance between FDG uptake and CT findings, interval evolution, and oncologic and clinical context. Illustrative cases demonstrate PET-occult pulmonary metastases, inflammatory mimics, coexisting infection and metastasis, synchronous or metachronous primary lung cancer, and subsolid nodules representing either atypical metastases or primary lung adenocarcinoma. Subsolid or cavitary morphology, waxing-and-waning interval changes, and low FDG uptake should not prompt premature benign labeling. Systematic evaluation of the CT component, followed by lesion-based integration of imaging and clinical data, is essential for characterizing these lesions and may help reduce both false-positive and false-negative interpretations.