Purpose: We sought to distinguish lung adenocarcinoma (ADC) from squamous cell carcinoma using a machine-learning algorithm with PET-based radiomic features. Methods: A total of 396 patients with 210 ADCs and 186 squamous cell carcinomas who underwent FDG PET/CT prior to treatment were retrospectively analyzed. Four clinical features (age, sex, tumor size, and smoking status) and 40 radiomic features were investigated in terms of lung ADC subtype prediction. Radiomic features were extracted from the PET images of segmented tumors using the LIFEx package. The clinical and radiomic features were ranked, and a subset of useful features was selected based on Gini coefficient scores in terms of associations with histological class. The areas under the receiver operating characteristic curves (AUCs) of classifications afforded by several machine-learning algorithms (random forest, neural network, naive Bayes, logistic regression, and a support vector machine) were compared and validated via random sampling. Results: We developed and validated a PET-based radiomic model predicting the histological subtypes of lung cancer. Sex, SUVmax, gray-level zone length nonuniformity, gray-level nonuniformity for zone, and total lesion glycolysis were the 5 best predictors of lung ADC. The logistic regression model outperformed all other classifiers (AUC = 0.859, accuracy = 0.769, F1 score = 0.774, precision = 0.804, recall = 0.746) followed by the neural network model (AUC = 0.854, accuracy = 0.772, F1 score = 0.777, precision = 0.807, recall = 0.750). Conclusions: A machine-learning approach successfully identified the histological subtypes of lung cancer. A PET-based radiomic features may help clinicians improve the histopathologic diagnosis in a noninvasive manner.

⁸⁹Zr is an emerging radionuclide that plays an essential role in immuno-positron emission tomography (PET) imaging. The long half-life of ⁸⁹Zr (t_1/2 = 3.3 days) is favorable for evaluating the in vivo distribution of monoclonal antibodies. Thus, the use of ⁸⁹Zr is promising for monitoring antibody-based cancer therapies. Immuno-PET combines the sensitivity of PET with the specificity of antibodies. A number of studies have been conducted to investigate the feasibility of ⁸⁹Zr immuno-PET imaging for predicting the efficacy of radioimmunotherapy and antibody therapies, imaging target expression, detecting target-expressing tumors, and the monitoring of anti-cancer chemotherapies. In this review, we summarize the current status of PET imaging using ⁸⁹Zr in both preclinical and clinical studies by highlighting the use of immuno-PET for the targets of high clinical relevance. We also present ⁸⁹Zr-PET applications other than immuno-PET, such as nanoparticle imaging and cell tracking. Finally, we discuss the limitations and the ongoing research being performed to overcome the remaining hurdles.

Vascular ¹⁸F-FDG uptake marker represents inflammation in atherosclerotic lesions, but whether inflammation can be reversed by risk-modifying interventions has not, to our knowledge, been demonstrated. In this study, we evaluated the change of vascular ¹⁸F-FDG uptake in response to lifestyle intervention on serial PET/CT scans and further assessed how the findings relate to atherogenic risk reduction. Methods: A total of 60 healthy adults underwent ¹⁸F-FDG PET/CT scans and atherogenic risk-factor assessment at baseline and again after 17.1 ± 8.3 mo of practicing lifestyle modification. The PET/CT images were evaluated for the presence of vascular ¹⁸F-FDG lesions, and vessel–to–blood-pool ¹⁸F-FDG ratios were measured. Indices from summed ratios of positive lesions were compared and correlated to atherogenic risk factors. Results: At follow-up, significant reductions in diastolic blood pressure (P < 0.05), total cholesterol (P < 0.05), and low-density lipoprotein level (P < 0.05) and an increase in high-density lipoprotein (HDL) level (P < 0.0001) were demonstrated. On the initial PET/CT scan, 50 of 60 subjects showed 1 or more ¹⁸F-FDG–positive lesions (5.9 ± 5.0/subject), leading to a total of 352 vascular sites. On follow-up, ¹⁸F-FDG–positive lesions were significantly reduced to 2.1 ± 2.2 sites per subject (P < 0.0001) and a total of 124 sites (64.8% reduction). Follow-up ¹⁸F-FDG–positive rates were significantly reduced for the aorta and iliac arteries. In addition, significant reductions in the whole-body ¹⁸F-FDG index from 1.39 ± 1.23 to 0.53 ± 0.59 (P < 0.0001) and carotid ¹⁸F-FDG index from 0.08 ± 0.16 to 0.03 ± 0.06 (P = 0.01) were shown. The whole-body ¹⁸F-FDG index correlated with total cholesterol (P < 0.05) and HDL level (P < 0.05), and the magnitude of reduction in the ¹⁸F-FDG index closely correlated to the amount of increase in plasma HDL level (P = 0.005). Conclusion: Our study demonstrated that vascular ¹⁸F-FDG uptake is reversed in response to atherogenic risk reduction by lifestyle intervention and that the magnitude of improvement correlates to increases in plasma HDL levels. Thus, serial ¹⁸F-FDG PET/CT may be useful for monitoring improvements in the inflammatory component of atherosclerotic lesions in response to risk modification.

For ¹⁸F-FDG PET to be widely used to monitor atherosclerosis progression and therapeutic response, it is crucial to better understand how macrophage glucose metabolism is influenced by the atherosclerotic microenvironment and to elucidate the molecular mechanisms of this response. Oxidized low-density lipoprotein (oxLDL) is a key player in atherosclerotic inflammation that promotes macrophage recruitment, activation, and foam cell formation. We thus explored the effect of oxLDL on macrophage ¹⁸F-FDG uptake and investigated the underlying molecular mechanism including the roles of hypoxia-inducible factor-1α (HIF-1α) and reactive oxygen species (ROS). Methods: RAW264.7 macrophages were stimulated with native LDL, oxLDL, or lipopolysaccharide. Cells were assessed for ¹⁸F-FDG uptake, lactate production, membrane glucose transporter 1 (GLUT1) expression, and hexokinase activity. ROS generation, Nox expression, and HIF-1α activity were also measured. Results: oxLDL (20 μg/mL) induced a 17.5 ± 1.7-fold increase in macrophage ¹⁸F-FDG uptake by 24 h, which was accompanied by increased lactate production, membrane GLUT1 expression, and hexokinase activity. oxLDL-stimulated ¹⁸F-FDG uptake was completely blocked by inhibitors of Src or phosphoinositide 3-kinase. ROS generation was increased to 262.4% ± 17.9% of controls by oxLDL, and N-acetyl-l-cysteine completely abrogated both oxLDL-induced ROS production and ¹⁸F-FDG uptake. oxLDL increased Nox2 expression, and nicotinamide adenine dinucleotide phosphate oxidase inhibition totally blocked increased ROS generation and ¹⁸F-FDG uptake by oxLDL. Finally, there was a clear ROS-dependent increase of HIF-1α accumulation by oxLDL, and silencing of HIF-1α completely abolished the metabolic effect of oxLDL. Conclusion: oxLDL is a strong stimulator of macrophage ¹⁸F-FDG uptake and glycolysis through upregulation of GLUT1 and hexokinase. This metabolic response is mediated by Nox2-dependent ROS generation that promotes HIF-1α activation.