HMGA1 is a Prognostic Biomarker and Correlated with Glycolysis in Lung Adenocarcinoma.

Purpose: Lung cancer is one of the leading causes with high morbidity and mortality. High mobility group A1 (HMGA1) protein participates in the process of tumorigenesis. This study seeks to explore the specific role of HMGA1 in prognostic value based on The Cancer Genome Atlas (TCGA) database of Lung adenocarcinoma (LUAD) and glycolysis progression in LUAD cells. Patients and Methods: In this research, we compared HMGA1 mRNA expression between tumor tissues and normal samples and evaluated the correlations with clinical characteristics in LUAD patients based on the data of TCGA database. The survival outcome with overall survival (OS), disease-specific survival (DSS) and clinicopathologic characteristics associated were performed using the Kaplan-Meier method and Cox regression. In addition, gene-set enrichment analysis (GSEA) was carried out to explore the biological pathways that related to HMGA1. Cell experiments including cell proliferation assay and glycolysis proteins were performed with A549 and H1299 cells. Results: Our results revealed that HMGA1 mRNA expression was higher in LUAD tissues than in normal tissues. Increased HMGA1 expression in LUAD was associated with Gender (p<0.01), Pathologic stage I&II vs stage III&IV (p<0.001), T1&T2 vs T3&T4 stage (p<0.05), N0 vs N2 stage (p<0.01). Furthermore, multivariate analysis revealed that HMGA1 was an independent risk factor of OS and DSS for LUAD patients (p<0.05). HMGA1 were positively correlated with glycolysis gluconeogenesis pathway and glycolysis markers (HK2, GLUT1, PKM2, LDHA) based on GSEA and Gene Expression Profiling Interactive Analysis (GEPIA) database. At the cellular level, the results of qRT-PCR and western blot assays showed that si-HMGA1 markedly decreased the expression of glycolysis markers. HMGA1 promoted cell glycolysis progression via PI3K/AKT pathway transfected with HMGA1-plasmid and the treatment with 20 µM LY294002. Relevant animal experiments were also synchronously validated and si-HMGA1 groups down-regulated xenograft growth including the weights and size in tumor xenografts. Conclusions: In conclusion, our results suggested that HMGA1 was significantly correlated with poor survival for LUAD tissues and involved in the process of glycolysis in LUAD cells.

Imaging Intratumoral Heterogeneity of Diffuse Pancreatic Neuroendocrine Tumor With Multitracer PET/CT.

ABSTRACT: Diffuse involvement of pancreatic neuroendocrine tumor (PNET) is a rare presentation. Here, we report a case of suspected autoimmune pancreatitis with 18 F-FDG and 18 F-FAPI-42 PET/CT showing increased tracer uptake in the entire pancreas, which was eventually confirmed by biopsy pathologic analysis as diffuse PNET. 18 F-AlF-NOTA-octreotide PET/CT imaging showed heterogeneous tracer uptake in the entire pancreas.

Additional Findings of 18 F-AIF-FAPI-42 PET/CT in a Patient With Mycosis Fungoides-Type Cutaneous T-Cell Lymphoma : Comparisons With 18 F-FDG PET/CT.

ABSTRACT: A 44-year-old woman presented with extensive skin patches and pruritus persisting for 3 years. Histopathological examination of the skin from the right abdomen confirmed mycosis fungoides-type cutaneous T-cell lymphoma. Staging PET with 18 F-FDG PET/CT) showed increased uptake in the skin on the right abdomen and left hip. Subsequently 18 F-FAPI-42 PET/CT revealed additional foci of abnormal uptake on the skin of the chest and back.

Inhaled nanoparticles for treating idiopathic pulmonary fibrosis by inhibiting honeycomb cyst and alveoli interstitium remodeling.

Idiopathic pulmonary fibrosis (IPF) is a progressive lung disease with high mortality. The Food and Drug Administration-approved drugs, nintedanib and pirfenidone, could delay progressive fibrosis by inhibiting the overactivation of fibroblast, however, there was no significant improvement in patient survival due to low levels of drug accumulation and remodeling of honeycomb cyst and interstitium surrounding the alveoli. Herein, we constructed a dual drug (verteporfin and pirfenidone)-loaded nanoparticle (Lip@VP) with the function of inhibiting airway epithelium fluidization and fibroblast overactivation to prevent honeycomb cyst and interstitium remodeling. Specifically, Lip@VP extensively accumulated in lung tissues via atomized inhalation. Released verteporfin inhibited the fluidization of airway epithelium and the formation of honeycomb cyst, and pirfenidone inhibited fibroblast overactivation and reduced cytokine secretion that promoted the fluidization of airway epithelium. Our results indicated that Lip@VP successfully rescued lung function through inhibiting honeycomb cyst and interstitium remodeling. This study provided a promising strategy to improve the therapeutic efficacy for IPF.

Gallbladder's Adenocarcinoma With Enteroblastic Differentiation Revealed by 18 F-FDG PET/CT.

ABSTRACT: Gallbladder's adenocarcinoma with enteroblastic differentiation (GAED) is extremely rare. A 43-year-old woman complained of pain in the right upper abdomen, and enhanced CT showed a cystic and solid mixed mass in the hepatic hilar region. Adenocarcinoma with enteroblastic differentiation was pathologically identified. 18 F-FDG PET/CT revealed a lesion in the gallbladder neck with increased FDG uptake, accompanied by a cystic and solid mixed mass in the hepatic hilar region with liver and lymph node metastases. Gallbladder biopsy was also carried out, and GAED was confirmed. 18 F-FDG PET/CT may assist in the evaluation of GAED and guide biopsy.

18 F-FAPI-42 PET/CT Findings in a Patient With Left Ventricular Mural Thrombus.

ABSTRACT: 18 F-FAPI-42 PET/CT is a novel imaging tool targeting fibroblast activation protein (FAP). We describe the 18 F-FAPI-42 PET/CT findings of a left ventricular mural thrombus in a 50-year-old man who had chest tightness. The 18 F-FAPI-42 PET/CT showed annular uptake at the apex of the left ventricle, but there was no uptake of 18 F-FDG. This case showed that abnormal 18 F-FAPI-42 uptake in the heart may be associated with mural thrombus and should be evaluated clinically.

An Autologous Macrophage-Based Phenotypic Transformation-Collagen Degradation System Treating Advanced Liver Fibrosis.

In advanced liver fibrosis (LF), macrophages maintain the inflammatory environment in the liver and accelerate LF deterioration by secreting proinflammatory cytokines. However, there is still no effective strategy to regulate macrophages because of the difficulty and complexity of macrophage inflammatory phenotypic modulation and the insufficient therapeutic efficacy caused by the extracellular matrix (ECM) barrier. Here, AC73 and siUSP1 dual drug-loaded lipid nanoparticle is designed to carry milk fat globule epidermal growth factor 8 (MFG-E8) (named MUA/Y) to effectively inhibit macrophage proinflammatory signals and degrade the ECM barrier. MFG-E8 is released in response to the high reactive oxygen species (ROS) environment in LF, transforming macrophages from a proinflammatory (M1) to an anti-inflammatory (M2) phenotype and inducing macrophages to phagocytose collagen. Collagen ablation increases AC73 and siUSP1 accumulation in hepatic stellate cells (HSCs) and inhibits HSCs overactivation. Interestingly, complete resolution of liver inflammation, significant collagen degradation, and HSCs deactivation are observed in methionine choline deficiency (MCD) and CCl4 models after tail vein injection of MUA/Y. Overall, this work reveals a macrophage-focused regulatory treatment strategy to eliminate LF progression at the source, providing a new perspective for the clinical treatment of advanced LF.

PTPRH promotes the progression of non-small cell lung cancer via glycolysis mediated by the PI3K/AKT/mTOR signaling pathway.

BACKGROUND: The protein tyrosine phosphatase H receptor (PTPRH) is known to regulate the occurrence and development of pancreatic and colorectal cancer. However, its association with glycolysis in non-small cell lung cancer (NSCLC) is still unclear. In this study, we aimed to investigate the relationship between PTPRH expression and glucose metabolism and the underlying mechanism of action. METHODS: The expression of PTPRH in NSCLC cells was evaluated by IHC staining, qRT‒PCR and Western blotting. The effect of PTPRH on cell biological behavior was evaluated by colony assays, EdU experiments, Transwell assays, wound healing assays and flow cytometry. Changes in F-18-fluorodeoxyglucose (18F-FDG) uptake and glucose metabolite levels after altering PTPRH expression were detected via a gamma counter and lactic acid tests. The expression of glycolysis-related proteins in NSCLC cells was detected by Western blotting after altering PTPRH expression. RESULTS: The results showed that PTPRH was highly expressed in clinical patient tissue samples and closely related to tumor diameter and clinical stage. In addition, PTPRH expression was associated with glycometabolism indexes on 18F-FDG positron emission tomography/computed tomography (PET/CT) imaging, the expression level of Ki67 and the expression levels of glycolysis-related proteins. PTPRH altered cell behavior, inhibited apoptosis, and promoted 18F-FDG uptake, lactate production, and the expression of glycolysis-related proteins. In addition, PTPRH modulated the glycometabolism of NSCLC cells via the phosphatidylinositol-3-kinase (PI3K)/protein kinase B (AKT)/mammalian target of rapamycin (mTOR) signaling pathway, as assessed using LY294002 and 740Y-P (an inhibitor and agonist of PI3K, respectively). The same results were validated in vivo using a xenograft tumor model in nude mice. Protein expression levels of PTPRH, glycolysis-related proteins, p-PI3K/PI3K and p-AKT/AKT were measured by IHC staining using a subcutaneous xenograft model in nude mice. CONCLUSIONS: In summary, we report that PTPRH promotes glycolysis, proliferation, migration, and invasion via the PI3K/AKT/mTOR signaling pathway in NSCLC and ultimately promotes tumor progression, which can be regulated by LY294002 and 740Y-P. These results suggest that PTPRH is a potential therapeutic target for NSCLC.

NOX4-derived ROS Regulates Aerobic Glycolysis of Breast Cancer through YAP Pathway.

Background: NOX4 is highly expressed in breast cancer and is closely associated with cell invasion and metastasis. The involvement of NOX4 in glycolysis in breast cancer remains unclear. The aim of this study was to investigate the role and mechanism of NOX4 in glycolysis in breast cancer. Methods: NOX4 expression in breast cancer cells was detected by qRT-PCR and western blotting. siRNAs and plasmids were used to silence or enhance the expression of NOX4. The mRNA and protein expression of HK2, GLUT1, PKM2, LDHA, and YAP was detected by qRT-PCR and western blotting, and the 18F-FDG uptake rate was detected by γ-radiometer. Detection of reactive oxygen species (ROS) in cells was performed using a commercial ROS kit. After transfection, CCK8, EDU and Transwell experiments were conducted to detect cell proliferation and migration ability. MicroPET imaging was used to detect the effects of NOX4 on tumor metabolism. Immunohistochemistry was used to detect the expression of NOX4, glycolytic enzymes HK2, GLUT1, PKM2, LDHA, the proliferation index KI67, and the activation of YAP pathway molecule. Results: In this study, the expression of NOX4 in MDA-MB-231 and MDA-MB-453 was higher than in MCF10A. qRT-PCR and western blotting experiments showed that NOX4 downregulation decreased the expression of glycolytic enzymes HK2, GLUT1, PKM2, LDHA, and 18F-FDG uptake. Conversely, the overexpression of NOX4 enhanced the expression of HK2, GLUT1, PKM2, LDHA, and 18F-FDG uptake. Proliferation and migration experiments showed that after down-regulation of NOX4, cell proliferation and migration ability decreased, while NOX4 overexpression promoted cell proliferation and migration ability. Additionally, ROS content and YAP expression decreased after NOX4 down-regulation, while ROS content and YAP expression increased following NOX4 overexpression, which was reversed by N-acetyl cysteine (NAC), a ROS inhibitor. Furthermore, exposure to NAC and Peptide17, a YAP inhibitor, blocked the increase in glycolytic enzyme expression, and the enhancement of proliferation and migration caused by NOX4 overexpression. In addition, in animal experiments, the results of the MicroPET imaging showed that the glucose metabolism rate of the NOX4 inhibitor group was significantly lower than that of the control group. ROS levels in the NOX4 inhibitor group was lower than that in the control group. Immunohistochemistry showed that the expression of HK2, GLUT1, PKM2, LDHA, KI67, and YAP in the NOX4 knock-down group were decreased. Conclusions: NOX4 affects breast cancer glycolysis through ROS-induced activation of the YAP pathway, further promoting the proliferation and migration of breast cancer cells.

LncRNA AP000695.2 promotes glycolysis of lung adenocarcinoma via the miR-335-3p/TEAD1 axis.

AP000695.2 is a novel long non-coding RNA (lncRNA). Its aberrant high expression is remarkably associated with poor prognosis of patients with lung adenocarcinoma (LUAD). However, its role and underlying mechanism in LUAD remains unclear. Previous bioinformatics analysis indicated that AP000695.2 may be closely related to the glycolysis of LUAD. This study aims to verify and explore the mechanism of AP000695.2 in glycolysis of LUAD. Overexpression plasmid and siRNA are used to construct cell models of upregulation and downregulation of AP000695.2, respectively. AP000695.2 is highly expressed in lung cancer cell lines as revealed by qPCR. Western blot analysis, FDG uptake, lactate production assay and ECAR determination results show that high expression of AP000695.2 facilitates glycolysis of LUAD cells. CCK-8, EdU staining, Transwell and wound healing assays show that high expression of AP000695.2 promotes cell growth and migration of LUAD. The relationship between AP000695.2 and miR-335-3p is confirmed by bioinformatics analysis and dual-luciferase reporter assays. Through the dual-luciferase reporter assay, TEA domain transcription factor 1 (TEAD1) is identified as a target gene of miR-335-3p. Rescue experiments are applied to verify the relationship among AP000695.2, miR-335-3p and TEAD1. Our study indicates that AP000695.2 is involved in the mechanism of LUAD through functioning as a ceRNA to competitively sponge miR-335-3p, thereby regulating the expression of TEAD1. In the in vivo models, AP000695.2 depletion restrains tumor growth and glycolysis. AP000695.2 promotes the glycolysis of LUAD by regulating the miR-335-3p/TEAD1 axis, and it may serve as a potential target of anti-tumor energy metabolism therapy.

Hypoxia-induced ALDH3A1 promotes the proliferation of non-small-cell lung cancer by regulating energy metabolism reprogramming.

Aldehyde dehydrogenase 3A1 (ALDH3A1) is an NAD+-dependent enzyme that is closely related to tumor development. However, its role in non-small-cell lung cancer (NSCLC) has not been elucidated. This study aimed to clarify the mechanism of ALDH3A1 and identify potential therapeutic targets for NSCLC. Here, for the first time, we found that ALDH3A1 expression could be induced by a hypoxic environment in NSCLC. ALDH3A1 was highly expressed in NSCLC tissue, especially in some late-stage patients, and was associated with a poor prognosis. In mechanistic terms, ALDH3A1 enhances glycolysis and suppresses oxidative phosphorylation (OXPHOS) to promote cell proliferation by activating the HIF-1α/LDHA pathway in NSCLC. In addition, the results showed that ALDH3A1 was a target of ß-elemene. ALDH3A1 can be downregulated by ß-elemene to inhibit glycolysis and enhance OXPHOS, thus suppressing NSCLC proliferation in vitro and in vivo. In conclusion, hypoxia-induced ALDH3A1 is related to the energy metabolic status of tumors and the efficacy of ß-elemene, providing a new theoretical basis for better clinical applications in NSCLC.

Highlighting Fibroblasts Activation in Fibrosis: The State-of-The-Art Fibroblast Activation Protein Inhibitor PET Imaging in Cardiovascular Diseases.

Fibrosis is a common healing process that occurs during stress and injury in cardiovascular diseases. The evolution of fibrosis is associated with cardiovascular disease states and causes adverse effects. Fibroblast activation is responsible for the formation and progression of fibrosis. The incipient detection of activated fibroblasts is important for patient management and prognosis. Fibroblast activation protein (FAP), a membrane-bound serine protease, is almost specifically expressed in activated fibroblasts. The development of targeted FAP-inhibitor (FAPI) positron emission tomography (PET) imaging enabled the visualisation of FAP, that is, incipient fibrosis. Recently, research on FAPI PET imaging in cardiovascular diseases increased and is highly sought. Hence, we comprehensively reviewed the application of FAPI PET imaging in cardiovascular diseases based on the state-of-the-art published research. These studies provided some insights into the value of FAPI PET imaging in the early detection of cardiovascular fibrosis, risk stratification, response evaluation, and prediction of the evolution of left ventricular function. Future studies should be conducted with larger populations and multicentre patterns, especially for response evaluation and outcome prediction.

Asymptomatic Prostate Cancer Metastasis in Rectal Mucosa Revealed by 18 F-PSMA-1007 PET/CT.

ABSTRACT: Prostate cancer metastasis to the rectal mucosa, a relatively rare metastatic site, leads to a higher clinical stage and poorer prognosis. A 65-year-old man with prostate cancer underwent 18 F-prostate-specific membrane antigen (PSMA) PET/CT for staging. Intense 18 F-PSMA uptake occurred at the primary lesion, bladder, adjacent seminal vesicle, and rectum. PET/CT imaging revealed increased homogeneous round activity of the rectal wall. The final diagnosis was prostate cancer metastasis to the rectal mucosa. This case suggested that 18 F-PSMA PET/CT may assist in locating rare metastases of prostate cancer, with potential value for early staging.

MRLA-Net: A tumor segmentation network embedded with a multiple receptive-field lesion attention module in PET-CT images.

The tumor image segmentation is an important basis for doctors to diagnose and formulate treatment planning. PET-CT is an extremely important technology for recognizing the systemic situation of diseases due to the complementary advantages of their respective modal information. However, current PET-CT tumor segmentation methods generally focus on the fusion of PET and CT features. The fusion of features will weaken the characteristics of the modality itself. Therefore, enhancing the modal features of the lesions can obtain optimized feature sets, which is extremely necessary to improve the segmentation results. This paper proposed an attention module that integrates the PET-CT diagnostic visual field and the modality characteristics of the lesion, that is, the multiple receptive-field lesion attention module. This paper made full use of the spatial domain, frequency domain, and channel attention, and proposed a large receptive-field lesion localization module and a small receptive-field lesion enhancement module, which together constitute the multiple receptive-field lesion attention module. In addition, a network embedded with a multiple receptive-field lesion attention module has been proposed for tumor segmentation. This paper conducted experiments on a private liver tumor dataset as well as two publicly available datasets, the soft tissue sarcoma dataset, and the head and neck tumor segmentation dataset. The experimental results showed that the proposed method achieves excellent performance on multiple datasets, and has a significant improvement compared with DenseUNet, and the tumor segmentation results on the above three PET/CT datasets were improved by 7.25%, 6.5%, 5.29% in Dice per case. Compared with the latest PET-CT liver tumor segmentation research, the proposed method improves by 8.32%.

FAPI PET/CT in Diagnostic and Treatment Management of Colorectal Cancer: Review of Current Research Status.

FAPI PET/CT is a novel imaging tool targeting fibroblast activation protein (FAP), with high tumor uptake rate and low background noise. Therefore, the appearance of FAPI PET/CT provides a good tumor-to-background ratio between tumor and non-tumor tissues, which is beneficial to staging, tumor description and detection. Colorectal cancer has the biological characteristics of high expression of FAP, which provides the foundation for targeted FAP imaging. FAPI PET/CT may have a potential role in changing the staging and re-staging of colorectal cancer, monitoring recurrence and treatment management, and improving the prognosis of patients. This review will summarize the application status of FAPI PET/CT in colorectal cancer and provide directions for further application research.

Metastatic Papillary Thyroid Cancer Mimicking Hypopharyngeal Carcinoma.

ABSTRACT: Papillary thyroid cancer (PTC) metastasizing to the hypopharynx is extremely rare. Here, we describe FDG PET/CT findings of lesions in the posterior hypopharyngeal wall and left parapharyngeal space in a 58-year-old man who complained of blood in the sputum. The patient had a history of postoperative PTC. Therefore, hypopharyngeal carcinoma with lymph node metastasis was suspected. However, metastasis of PTC was pathologically confirmed after surgery. Hypopharyngeal metastasis of PTC is rare, which should be differentiated from hypopharyngeal carcinoma with metastasis.

The long noncoding RNA HOXA11-AS promotes lung adenocarcinoma proliferation and glycolysis via the microRNA-148b-3p/PKM2 axis.

BACKGROUND: Lung cancer is the most common malignancy in the world and a growing number of researches have focused on its metabolic characteristics. Studies have shown that the long non-coding RNA (lncRNA) HOXA11-AS is aberrantly expressed in many tumors. However, the role of HOXA11-AS in lung adenocarcinoma (LUAD) glycolysis and other energy metabolism pathways has not been characterized. METHOD: The mRNA levels of HOXA11-AS, microRNA-148b-3p (miR-148b-3p), and pyruvate kinase M2 (PKM2) were detected using qRT-PCR. The expression levels of proteins were measured using immunohistochemistry and western blot. The CCK-8, EdU, and colony formation assays were used to assess proliferation. Glycolytic changes were assessed by measuring lactate production, ATP production, and 18 F-FDG uptake. Bioinformatics analysis and dual-luciferase reporter assays were used to characterize the relationship between HOXA11-AS, miR-148b-3p, and PKM2. Proliferation and glycolytic changes were analyzed in xenograft tumor experiments using Micro-PET imaging after downregulation of HOXA11-AS in vivo. RESULTS: The expression of HOXA11-AS was markedly increased in LUAD, and was strongly associated with a poor prognosis. In addition, HOXA11-AS promoted proliferation and glycolysis in LUAD, and miR-148b-3p inhibited proliferation and glycolysis in LUAD. Mechanistically, HOXA11-AS positively regulated PKM2 expression by binding to miR-148b-3p, thereby promoting LUAD proliferation and glycolysis. In addition, HOXA11-AS inhibited LUAD xenograft growth and glycolysis via upregulation of miR-148b-3p expression and downregulation of PKM2 expression in vivo. CONCLUSIONS: These results showed that HOXA11-AS enhanced LUAD proliferation and glycolysis via the miR-148b-3p/PKM2 axis. The findings in this paper expanded our understanding of the molecular mechanisms of LUAD tumorigenesis and glycolysis and showed that HOXA11-AS could be useful as a diagnostic and prognostic marker for LUAD. 18 F-FDG PET/CT can be used to visually evaluate the therapeutic effect of targeting HOXA11-AS.

Upstream stimulatory factor 2 inhibits erastin-induced ferroptosis in pancreatic cancer through transcriptional regulation of pyruvate kinase M2.

Ferroptosis is considered as a potential target in cancer treatment including chemotherapy and radiotherapy, however, its regulatory mechanism on pancreatic cancer (PC) is not fully understood. Herein, we explored the role of upstream stimulatory factor 2 (USF2) and pyruvate kinase M2 (PKM2) in ferroptosis in PC cells. USF2 and PKM2 were highly expressed in PC tissues and USF2 was positively correlated with PKM2. PC cell lines BxpC-3 and AsPC-1 were transfected with small interfering RNAs against USF2/PKM2 or USF2 overexpressing plasmids or co-transfected with small interfering RNAs against PKM2 and USF2 overexpressing plasmids. Twenty-four hours after cell transfection, ferroptotic cell death was induced by incubation with 20 µmol/l erastin for 24 h. Ferroptotic cell death was promoted by USF2 knockdown and inhibited by USF2 overexpression. USF2 knockdown increased lipid reactive oxygen species and malonaldehyde generation and decreased glutathione concentration and glutathione peroxidase 4 expressions, indicating the enhanced lipid peroxidation. USF2 knockdown also increased ferrous iron levels and ferritin heavy chain expressions and reduced solute carrier family 7 member 11 expressions. However, USF2 overexpression reversed these changes. Furthermore, dual-luciferase reporter assay, chromatin immunoprecipitation assay and DNA pull down assay validated that USF2 transcriptionally regulated PKM2 expression through binding to its promoter. Interestingly, PKM2 also negatively regulated ferroptosis and PKM2 knockdown markedly impaired the effects of USF2 on lipid peroxidation and ferroptotic cell death. This study demonstrated that USF2 negatively regulated ferroptosis in PC cells through transcriptional regulation of PKM2, providing new evidences for uncovering the regulatory mechanism of ferroptosis on PC.

Using stacked deep learning models based on PET/CT images and clinical data to predict EGFR mutations in lung cancer.

Purpose: To determine whether stacked deep learning models based on PET/CT images and clinical data can help to predict epidermal growth factor receptor (EGFR) mutations in lung cancer. Methods: We analyzed data from two public datasets of patients who underwent 18F-FDG PET/CT. Three PET deep learning ResNet models and one CT deep learning ResNet model were trained as low-level predictors based on PET and CT images, respectively. A high-level Support Vector Machine model (Stack PET/CT and Clinical model) was trained using the prediction results of the low-level predictors and clinical data. The clinical data included sex, age, smoking history, SUVmax and SUVmean of the lesion. Fivefold cross-validation was used in this study to validate the prediction performance of the models. The predictive performance of the models was evaluated by receiver operator characteristic (ROC) curves. The area under the curve (AUC) was calculated. Results: One hundred forty-seven patients were included in this study. Among them, 37/147 cases were EGFR mutations, and 110/147 cases were EGFR wild-type. The ROC analysis showed that the Stack PET/CT & Clinical model had the best performance (AUC = 0.85 ± 0.09), with 0.76, 0.85 and 0.83 in sensitivity, specificity and accuracy, respectively. Three ResNet PET models had relatively higher AUCs (0.82 ± 0.07, 0.80 ± 0.08 and 0.79 ± 0.07) and outperformed the CT model (AUC = 0.58 ± 0.12). Conclusion: Using stack generalization, the deep learning model was able to efficiently combine the anatomic and biological imaging information gathered from PET/CT images with clinical data. This stacked deep learning model showed a strong ability to predict EGFR mutations with high accuracy.