Automated detection and classification of coronary atherosclerotic plaques on coronary CT angiography using deep learning algorithm.

Background: Coronary artery disease (CAD) is the leading cause of mortality worldwide. Recent advances in deep learning technology promise better diagnosis of CAD and improve assessment of CAD plaque buildup. The purpose of this study is to assess the performance of a deep learning algorithm in detecting and classifying coronary atherosclerotic plaques in coronary computed tomographic angiography (CCTA) images. Methods: Between January 2019 and September 2020, CCTA images of 669 consecutive patients with suspected CAD from Nanjing Drum Tower Hospital Clinical College of Nanjing University of Chinese Medicine were included in this study. There were 106 patients included in the retrospective plaque detection analysis, which was evaluated by a deep learning algorithm and four independent physicians with varying clinical experience. Additionally, 563 patients were included in the analysis for plaque classification using the deep learning algorithm, and their results were compared with those of expert radiologists. Plaques were categorized as absent, calcified, non-calcified, or mixed. Results: The deep learning algorithm exhibited higher sensitivity, specificity, positive predictive value (PPV), negative predictive value (NPV), and accuracy {92% [95% confidence interval (CI): 89.5-94.1%], 87% (95% CI: 84.2-88.5%), 79% (95% CI: 76.1-82.4%), 95% (95% CI: 93.4-96.3%), and 89% (95% CI: 86.9-90.0%)} compared to physicians with ≤5 years of clinical experience in CAD diagnosis for the detection of coronary plaques. The algorithm's overall sensitivity, specificity, PPV, NPV, accuracy, and Cohen's kappa for plaque classification were 94% (95% CI: 92.3-94.7%), 90% (95% CI: 88.8-90.3%), 70% (95% CI: 68.3-72.1%), 98% (95% CI: 97.8-98.5%), 90% (95% CI: 89.8-91.1%) and 0.74 (95% CI: 0.70-0.78), indicating strong performance. Conclusions: The deep learning algorithm has demonstrated reliable and accurate detection and classification of coronary atherosclerotic plaques in CCTA images. It holds the potential to enhance the diagnostic capabilities of junior radiologists and junior intervention cardiologists in the CAD diagnosis, as well as to streamline the triage of patients with acute coronary symptoms.

Identification of lncRNA, miRNA and mRNA expression profiles and ceRNA Networks in small cell lung cancer.

BACKGROUND: Small cell lung cancer (SCLC) is a highly lethal malignant tumor. It accounts for approximately 15% of newly diagnosed lung cancers. Long non-coding RNAs (lncRNAs) can regulate gene expression and contribute to tumorigenesis through interactions with microRNAs (miRNAs). However, there are only a few studies reporting the expression profiles of lncRNAs, miRNAs, and mRNAs in SCLC. Also, the role of differentially expressed lncRNAs, miRNAs, and mRNAs in relation to competitive endogenous RNAs (ceRNA) network in SCLC remain unclear. RESULTS: In the present study, we first performed next generation sequencing (NGS) with six pairs of SCLC tumors and adjacent non-cancerous tissues obtained from SCLC patients. Overall, 29 lncRNAs, 48 miRNAs, and 510 mRNAs were found to be differentially expressed in SCLC samples (|log2[fold change] |> 1; P < 0.05). Bioinformatics analysis was performed to predict and construct a lncRNA-miRNA-mRNA ceRNA network, which included 9 lncRNAs, 11 miRNAs, and 392 mRNAs. Four up-regulated lncRNAs and related mRNAs in the ceRNA regulatory pathways were selected and validated by quantitative PCR. In addition, we examined the role of the most upregulated lncRNA, TCONS_00020615, in SCLC cells. We found that TCONS_00020615 may regulate SCLC tumorigenesis through the TCONS_00020615-hsa-miR-26b-5p-TPD52 pathway. CONCLUSIONS: Our study provided the comprehensive analysis of the expression profiles of lncRNAs, miRNAs, and mRNAs of SCLC tumors and adjacent non-cancerous tissues. We constructed the ceRNA networks which may provide new evidence for the underlying regulatory mechanism of SCLC. We also found that the lncRNA TCONS_00020615 may regulate the carcinogenesis of SCLC.

The diagnostic value of quantitative parameters on dual-layer detector-based spectral CT in identifying ischaemic stroke.

Objective: To investigate the diagnostic value of quantitative parameters of spectral computed tomography (CT) in ischaemic stroke areas. Methods: The medical records of 57 patients with acute ischaemic stroke (AIS) who underwent plain computed tomography (CT) head scans, CT angiography (CTA), and CT perfusion (CTP) were retrospectively reviewed. The ischaemic areas (including the core infarct area and penumbra) and non-ischaemic areas in each patient were quantitatively analyzed using F-STROKE software. Two independent readers measured the corresponding values of the spectroscopic quantitative parameters (effective atomic number [Zeff value], iodine density value, and iodine-no-water value) in the ischaemic area and contralateral normal area alone. The differences in spectroscopic quantitative parameters between the two groups were compared, and their diagnostic efficacy was obtained. Results: The Zeff, iodine-no-water value, and iodine density value of the ischaemic area all showed significant lower than those of non-ischaemic tissue (P < 0.001). For differentiating the ischaemic area from non-ischaemic tissue, the area under the curve (AUC) of the Zeff value reached 0.869 (cut-off value: 7.385; sensitivity: 93.0%; specificity: 70.2%), the AUC of the iodine density value reached 0.932 (cut-off value: 0.235; sensitivity: 91.2%; specificity: 82.5%), and the AUC of the iodine-no-water value reached 0.922 (cut-off value: 0.205; sensitivity: 96.5%; specificity: 78.9%). Conclusion: The study showed the spectral CT would be a potential novel rapid method for identifying AIS. The spectral CT quantitative parameters (Zeff, iodine density values, and iodine-no-water values) can effectively differentiate the ischaemic area from non-ischaemic tissue in stroke patients.

Iodine-based extracellular volume for evaluating myocardial status in patients undergoing percutaneous coronary intervention for acute myocardial infarction by using dual-layer spectral detector computed tomography: a comparison study with magnetic resonance.

Background: The myocardial status of patients who undergo percutaneous coronary intervention (PCI) must be evaluated accurately to enable treatment plans to be made for potential complications such as abrupt vessel closure, stent deformation, and myocardial chronic ischemia. This study examined the modality and clinical feasibility of iodine-based extracellular volume (ECV) assessment of the myocardium versus cardiovascular magnetic resonance (CMR) imaging in patients undergoing PCI. Methods: In all, 21 patients who underwent PCI were prospectively enrolled in the study. All patients underwent follow-up cardiac dual-layer spectral detector computed tomography (SDCT) and CMR imaging after PCI. Myocardial ECV was quantified by either computed tomography (ECVCT) or magnetic resonance (ECVMR) using iodine or T1-weighted mapping, respectively. The quality of SDCT and CMR images was independently assessed by two radiologists using a 4-point scale (1= poor and 4= excellent). Any patient with an image quality (IQ) score <2 was excluded. Consistency between radiologists was evaluated using intraclass correlation coefficients (ICC). Correlations between ECVCT and ECVMR values were analyzed using Pearson's test, and consistency was analyzed with Bland-Altman plots. Results: Nineteen of 21 patients completed both cardiac CT and CMR examinations, while three patients were excluded after IQ assessment (two with poor CMR IQ; one with a discontinuous coronary artery on CT images). The mean (±SD) IQ scores for CT and CMR images were 3.81±0.40 and 3.25±0.58, respectively, and interobserver agreement was good (ICC =0.93 and 0.92 for CT and CMR, respectively). The mean (±SD) ECVCT and ECVMR values were 35.93%±9.73% and 33.89%±7.51%, respectively, with good correlation (r=0.79, P<0.001). Bland-Altman analysis showed a difference of 2.04% (95% CI: -9.56%, 13.64%) between the ECVCT and ECVMR values. Conclusions: There is high correlation between iodine-based ECVCT and ECVMR values, which indicates that ECVCT is clinically feasible for evaluating the status of myocardial recovery in patients undergoing PCI.

Clinical feasibility of using effective atomic number maps derived from non-contrast spectral computed tomography to identify non-calcified atherosclerotic plaques: a preliminary study.

Background: To assess the clinical feasibility of using effective atomic number (Zeff) maps derived from non-contrast-enhanced computed tomography (NCECT) scans obtained by dual-layer spectral computed tomography (DLCT) to identify non-calcified atherosclerotic plaques. Methods: A total of 37 patients with 86 non-calcified atherosclerotic plaques confirmed by contrast-enhanced CT (CECT) were enrolled in this retrospective study. Both spectral-based-images (SBI) and conventional images (CI) were reconstructed from NCECT and CECT scans. The presence of plaques on NCECT Zeff maps and CIs were independently assessed by 2 radiologists. In CECT scans, plaques and regions of interest (ROIs) in vessel lumens were assessed with CT attenuation and Zeff values, and the proportion of plaques was determined as Area (plaque)/Area (vessel). The CT and Zeff values for plaques and blood were recorded from both CECT and NCECT scans. Contrast-to-noise ratios (CNRs) of the plaques were calculated and compared using CT attenuation and Zeff values. Finally, interobserver agreement was evaluated. Results: A total of 47 of the 86 (54.7%) plaques were identified on Zeff map images derived from the NCECT scans while only 7 (8.1%) plaques were identified on the CI. There was no significant difference between the mean vessel ROI area measured on CIs and that measured on Zeff map images (502.19 vs. 498.14 mm2; P=0.28), while the mean plaque ROI area was larger (81.45 vs. 75.46 mm2). The observer consensus of vessel and plaque ROI area measurements using both methods was excellent, with interclass correlation coefficients (ICCs) of 0.99 and 0.94, respectively. For the 7 plaques detected both by NCECT CI and Zeff mapping, the CT attenuation and Zeff blood values were both larger than the plaque values [42.00 vs. 25.67 Hounsfield unit (HU); 7.33 vs. 7.19 HU; both P<0.05]; the plaque ROI area measurement on the NCE Zeff map was smaller than that on the CE CI (48.73 vs. 77.76 mm2), but was much larger than that on the NCE CI (18.39 mm2). For all 47 plaques detected by NCE Zeff mapping, the CT attenuation and Zeff values of blood and plaques on the NCECT images showed no significant differences (42.53 vs. 35.14 HU; P=0.18; 7.32 vs. 7.31, P=0.71); however, the CNR of Zeff was significantly higher than the CT attenuation value (1.69 vs. 1.12; P<0.05) derived from the NCECT scans. Inter-reviewer agreement was good (ICC =0.78). Conclusions: Zeff map images derived from NCECT SBI with DLCT provide a potentially feasible approach for identifying non-calcified atherosclerotic plaques, which might be clinically useful for the screening of asymptomatic at-risk patients.

RNA-Seq profiling of circular RNAs in human small cell lung cancer.

Aim: We aimed to explore the circular RNA (circRNA) profile of small-cell lung cancer (SCLC). Materials & methods: Total RNA was extracted from six paired SCLC tumors and adjacent noncancerous tissues. Next-generation sequencing was performed to identify the circRNA expression profile of SCLC. Results: We found that five circRNAs were significantly upregulated and 30 circRNAs were significantly downregulated in the SCLC tissues. We confirmed the five upregulated and four randomly selected downregulated circRNAs using real-time quantitative PCR. Notably, circ-STXBP5L was selectively upregulated in SCLC samples, but undetectable in the normal control tissues. Bioinformatics analysis demonstrated that circ-STXBP5L may participate in SCLC carcinogenesis by regulating numerous cancer-related pathways. Conclusion: This study may provide new insights into the early diagnosis and development of targeted therapies for SCLC.

Overexpressed D2 Dopamine Receptor Inhibits Non-Small Cell Lung Cancer Progression through Inhibiting NF-κB Signaling Pathway.

BACKGROUND/AIMS: Non-small cell lung cancer (NSCLC) is one of the deadliest cancers worldwide. Dopamine receptor D2 (DRD2) has multiple roles in clinical progression of NSCLC and functional maintenance of cancer cells. However, little is known about the molecular mechanism. Here, we clarified whether DRD2 inhibits lung cancer progression and identified the underlying downstream signaling. METHODS: DRD2 mRNA and protein levels were detected in clinical specimens by qRT-PCR and immunohistochemistry, respectively. MTT and colony formation assays were applied to analyze cell proliferation. The underlying molecular mechanism was identified by dual luciferase, western blot, qRT-PCR, cAMP detection, immunoprecipitation, and chromatin immunoprecipitation assays. A murine NSCLC model was used to clarify the role of DRD2 in tumor cell proliferation. RESULTS: We found that DRD2 ablated tumor cell growth. DRD2 expression in NSCLC tissues was lower than in adjacent normal lung tissues. Moreover, DRD2 mRNA and protein levels in NSCLC were negatively correlated with the tumor size, TNM status, and patient overall survival. In vitro experiments showed that disruption of DRD2 promoted the proliferation of NSCLC cell lines A549 and SK-MES-1 by inhibiting the NF-κB signaling pathway. Furthermore, DRD2 overexpression not only blocked lipopolysaccharide-induced A549 and SK-MES-1 cell proliferation and growth, but also inhibited the tumorigenesis in murine xenograft models. CONCLUSION: These results indicate that DRD2 may be a potential therapeutic target for lung cancer patients with high DRD2 expression by ablating the NF-κB signaling pathway.

Expression of the enhancer of zeste homolog 2 in biopsy specimen predicts chemoresistance and survival in advanced non-small cell lung cancer receiving first-line platinum-based chemotherapy.

OBJECTIVES: Enhancer of zeste homolog 2 (EZH2) plays a key role in tumorigenesis and cancer progression through epigenetic gene silencing and chromatin remodeling. The objective of this study was to investigate the correlation between EZH2 expression and platinum-based chemotherapy response as well as survival of patients with advanced non-small cell lung cancer (NSCLC). MATERIALS AND METHODS: We identified 360 consecutive stage IIIB and IV NSCLC patients who underwent first-line platinum-based chemotherapy. Immunohistochemical analysis of EZH2 on the paraffin-embedded pre-treatment tumor samples was performed and correlated with chemotherapy response and survival. RESULTS: EZH2 was positive in 204 of 360 patients (56.7%). Of the 204 positive EZH2 patients, 72 (35.3%) responded to chemotherapy with either complete response, or partial remission. Of 156 negative EZH2 patients, 90 (57.7%) exhibited a response to chemotherapy. The difference in response to therapy between positive and negative EZH2 patients was statistically significant (p<0.01). Univariate survival analysis indicated that patients with positive EZH2 had a significantly lower disease-free survival (DFS) and overall survival (OS) than those patients with negative EZH2 expression. Multivariate Cox regression analysis demonstrated that positive EZH2 expression was an independent prognostic factor for both DFS and OS. Kaplan-Meier survival curves further confirmed that positive EZH2 expression correlates with poor survival in NSCLC patients. CONCLUSIONS: Our results indicate that advanced NSCLC patients with positive expression of EZH2 exhibited resistance to cisplatin-based chemotherapy. EZH2 may be a predictive and prognostic factor for cisplatin-based therapy response and disease survival in advanced NSCLC.

Elevated expression of Cripto-1 correlates with poor prognosis in non-small cell lung cancer.

Human Cripto-1 (CR-1) plays an important oncogenic role during tumorigenesis and is overexpressed in a wide range of carcinomas, yet little is known about CR-1 in non-small cell lung cancer (NSCLC). The aims of this study were to detect CR-1 expression in NSCLC and to analyze its association with prognosis of NSCLC patients. The expression of CR-1 messenger RNA (mRNA) and protein in 35 cases of NSCLC and corresponding noncancerous tissue samples was examined by quantitative real-time reverse transcription-polymerase chain reaction (qRT-PCR) and Western blotting. Immunohistochemistry was performed to detect the expression of CR-1 in 128 NSCLC tissues. The expression levels of CR-1 mRNA and protein in NSCLC tissues were significantly higher than those in corresponding noncancerous tissues (P < 0.001). A high level of CR-1 expression was correlated with poor tumor differentiation (P = 0.002), tumor-node-metastasis (TNM) stage (P = 0.004), and lymph node metastasis (P = 0.001). The results of the Kaplan-Meier analysis indicated that a high expression level of CR-1 resulted in a significantly poor prognosis of NSCLC patients. Multivariate Cox regression analysis revealed that CR-1 expression level was an independent prognostic parameter for the overall survival rate of NSCLC patients. Our data suggest that the high expression of CR-1 may play an important role in the progression of NSCLC, and CR-1 expression may offer a valuable marker for predicting the outcome of patients with NSCLC.