Deep Learning to Predict EGFR Mutation and PD-L1 Expression Status in Non-Small-Cell Lung Cancer on Computed Tomography Images

Objective. The detection of epidermal growth factor receptor (EGFR) mutation and programmed death ligand-1 (PD-L1) expression status is crucial to determine the treatment strategies for patients with non-small-cell lung cancer (NSCLC). Recently, the rapid development of radiomics including but not limited to deep learning techniques has indicated the potential role of medical images in the diagnosis and treatment of diseases. Methods. Eligible patients diagnosed/treated at the West China Hospital of Sichuan University from January 2013 to April 2019 were identified retrospectively. The preoperative CT images were obtained, as well as the gene status regarding EGFR mutation and PD-L1 expression. Tumor region of interest (ROI) was delineated manually by experienced respiratory specialists. We used 3D convolutional neural network (CNN) with ROI information as input to construct a classification model and established a prognostic model combining deep learning features and clinical features to stratify survival risk of lung cancer patients. Results. The whole cohort (N = 1262) was divided into a training set (N = 882, 70%), validation set (N = 125, 10%), and test set (N = 255, 20%). We used a 3D convolutional neural network (CNN) to construct a prediction model, with AUCs of 0.96 (95% CI: 0.94–0.98), 0.80 (95% CI: 0.72–0.88), and 0.73 (95% CI: 0.63–0.83) in the training, validation, and test cohorts, respectively. The combined prognostic model showed a good performance on survival prediction in NSCLC patients (C-index: 0.71). Conclusion. In this study, a noninvasive and effective model was proposed to predict EGFR mutation and PD-L1 expression status as a clinical decision support tool. Additionally, the combination of deep learning features with clinical features demonstrated great stratification capabilities in the prognostic model. Our team would continue to explore the application of imaging markers for treatment selection of lung cancer patients.

Ringed Fluorodeoxyglucose Uptake Predicted Poor Prognosis After Resection of Pulmonary Pleomorphic Carcinoma

Background:Pulmonary pleomorphic carcinoma (PPC) is a relatively rare and poorly differentiated non-small cell carcinoma. This study aimed to investigate the clinicopathological features including programmed cell death ligand 1 (PD-L1) expression status in patients with PPC who underwent curative resection.Methods:We retrospectively studied 29 consecutive patients who had undergone anatomical lung resections for PPC. Perioperative and pathological variables, including radiological findings, were investigated to define prognostic factors.Results:Overall survival (OS) rates were 71.8% at 1 year and 60.0% at 5 years. Disease-free survival (DFS) rates were 54.8% at 1 year and 43.6% at 5 years. Univariate analysis revealed that ringed fluorodeoxyglucose (FDG) uptake on positron emission tomography/computed tomography (PET/CT) (p=0.003), a cavity in the tumor on CT (p=0.004), and tumor size (>40mm) (p=0.014) were poor prognostic factors for OS. Regarding DFS, ringed FDG uptake (p=0.002), a cavity on CT (p<0.001), tumor size (p=0.007), and pleural invasion (p=0.014) were poor prognostic factors. PD-L1 expression was not a prognostic factor. Conclusion:Early relapse was frequently observed. This study showed for the first time that ringed FDG uptake on PET/CT is a poor prognostic factor of PPC. PD-L1 expression status was not related to the prognosis. Trial registration:The study was approved by the Kobe City Medical Center General Hospital’s ethics board (No. 20112) on August 20, 2020.

Assessing PD-L1 Expression Status Using Radiomic Features from Contrast-Enhanced Breast MRI in Breast Cancer Patients: Initial Results

The purpose of this retrospective study was to assess whether radiomics analysis coupled with machine learning (ML) based on standard-of-care dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) can predict PD-L1 expression status in patients with triple negative breast cancer, and to compare the performance of this approach with radiologist review. Patients with biopsy-proven triple negative breast cancer who underwent pre-treatment breast MRI and whose PD-L1 status was available were included. Following 3D tumor segmentation and extraction of radiomic features, radiomic features with significant differences between PD-L1+ and PD-L1− patients were determined, and a final predictive model to predict PD-L1 status was developed using a coarse decision tree and five-fold cross-validation. Separately, all lesions were qualitatively assessed by two radiologists independently according to the BI-RADS lexicon. Of 62 women (mean age 47, range 31–81), 27 had PD-L1− tumors and 35 had PD-L1+ tumors. The final radiomics model to predict PD-L1 status utilized three MRI parameters, i.e., variance (FO), run length variance (RLM), and large zone low grey level emphasis (LZLGLE), for a sensitivity of 90.7%, specificity of 85.1%, and diagnostic accuracy of 88.2%. There were no significant associations between qualitative assessed DCE-MRI imaging features and PD-L1 status. Thus, radiomics analysis coupled with ML based on standard-of-care DCE-MRI is a promising approach to derive prognostic and predictive information and to select patients who could benefit from anti-PD-1/PD-L1 treatment.

Magnetic resonance imaging-based radiomics signature for preoperative prediction of Ki67 expression in bladder cancer

Purpose The Ki67 expression is associated with the advanced clinicopathological features and poor prognosis in bladder cancer (BCa). We aimed to develop and validate magnetic resonance imaging (MRI)-based radiomics signatures to preoperatively predict the Ki67 expression status in BCa. Methods and materials We retrospectively collected 179 BCa patients with Ki67 expression and preoperative MRI. Radiomics features were extracted from T2-weighted (T2WI) and dynamic contrast-enhancement (DCE) images. The synthetic minority over-sampling technique (SMOTE) was used to balance the minority group (low Ki67 expression group) in the training set. Minimum redundancy maximum relevance was used to identify the best features associated with Ki67 expression. Support vector machine and Least Absolute Shrinkage and Selection Operator algorithms (LASSO) were used to construct radiomics signatures in training and SMOTE-training sets, and diagnostic performance was assessed by the area under the curve (AUC) and accuracy. The decision curve analyses (DCA) and calibration curve and were used to investigate the clinical usefulness and calibration of radiomics signatures, respectively. The Kaplan-Meier test was performed to investigate the prognostic value of radiomics-predicted Ki67 expression status. Results 1218 radiomics features were extracted from T2WI and DCE images, respectively. The SMOTE-LASSO model based on nine features achieved the best predictive performance in the SMOTE-training (AUC, 0.859; accuracy, 80.3%) and validation sets (AUC, 0.819; accuracy, 81.5%) with a good calibration performance and clinical usefulness. Immunohistochemistry-based high Ki67 expression and radiomics-predicted high Ki67 expression based on the SMOTE-LASSO model were significantly associated with poor disease-free survival in training and validation sets (all P < 0.05). Conclusions The SMOTE-LASSO model could predict the Ki67 expression status and was associated with survival outcomes of the BCa patients, thereby may aid in clinical decision-making.

Involvement of Ischemia-driven 5LPOX-RvE1-ChemR23 axis in the resolution of post-CABG inflammation in Coronary Arteries

Aims: Expression status of pro-resolving lipid mediators (PLM) and receptors in the post-CABG coronary arteries are largely unknown. Here, we aim to investigate the expression of PLMs in the atherosclerotic post-CABG swine LAD compared to without CABG (LAD-AS), and in isolated coronary artery smooth muscle cells (CASMCs) cultured under ischemia. Methodology: The arteries of interest were harvested from post-CABG atherosclerotic swine and the histomorphology and the expression status of key PLM mediators were quantified using immunostaining. SMCs were cultured under ischemia and confirmed the expression on PLM mediators at transcript and protein level.Results: The histomorphometric analysis revealed considerable alterations in the tissue architecture in LAD-CABG and LAD-AS arteries compared to control. PLM synthetic enzyme 5LPOX was significantly upregulated in LAD-CABG and LAD-AS whereas the other mediators including 12LPOX, 15LPOX, COX2, ChemR23, GPCR18, GPCR120 were decreased in LAD-CABG than control. LPOX enzymes and PLM receptors were upregulated in ischemic CASMCs with respect to control. Western blot showed the upregulation of 5LPOX, and ChemR23. Additionally, higher level of RvE1 was observed in ischemic control CASMCs which was decreased following reperfusion. Conclusion: These findings suggest that the CASMCs withstand the ischemia-triggered proinflammatory episodes by increasing the secretion of RvE1 mediated through 5LPOX and ChemR23 signaling.

Impact of Serum γ-Glutamyltransferase on Overall Survival in Men with Metastatic Castration-Resistant Prostate Cancer Treated with Docetaxel

Background: Reports on the prognostic significance of serum γ-glutamyltransferase (GGT) in men with metastatic castration-resistant prostate cancer (mCRPC) are limited. In addition, GGT expression status in cancer tissues has not been well characterized regardless of cancer types. Methods: This retrospective study included 107 consecutive men with mCRPC receiving docetaxel therapy. The primary endpoints were associations of serum GGT with overall survival (OS) and prostate-specific antigen (PSA) response. The secondary endpoint was an association of serum GGT with progression-free survival (PFS). Additionally, GGT expression status was immunohistochemically semi-quantified using tissue microarrays. Results: A total of 67 (63%) men died during follow-up periods (median 22.5 months for survivors). On multivariable analysis, high Log GGT was independently associated with adverse OS (HR 1.49, p = 0.006) as were low hemoglobin (HR 0.79, p = 0.002) and high PSA (HR 1.40, p < 0.001). In contrast, serum GGT was not significantly associated with PSA response or PFS. Moreover, incorporation of serum GGT into established prognostic models (i.e., Halabi and Smaletz models) increased their C-indices for predicting OS from 0.772 to 0.787 (p = 0.066) and from 0.777 to 0.785 (p = 0.118), respectively. Furthermore, there was a positive correlation between serum and tissue GGT levels (ρ = 0.53, p = 0.003). Conclusions: Serum GGT may be a prognostic biomarker in men with mCRPC receiving docetaxel therapy. GGT overexpression by prostate cancer cells appears to be responsible for the elevation of GGT in the serum.

Intertumoral Heterogeneity of Primary Breast Tumors and Synchronous Axillary Lymph Node Metastases Reflected in IHC-Assessed Expression of Routine and Nonstandard Biomarkers

Breast cancer (BC) remains a significant healthcare challenge. Routinely, the treatment strategy is determined by immunohistochemistry (IHC)-based assessment of the key proteins such as estrogen receptor (ER), progesterone receptor (PR), human epidermal growth factor receptor 2 (HER2), and Ki-67. However, it is estimated that over 75% of deaths result from metastatic tumors, indicating a need to develop more accurate protocols for intertumoral heterogeneity assessment and their consequences on prognosis. Therefore, the aim of this preliminary study was the identification of the expression profiles of routinely used biomarkers (ER, PR, HER2, Ki-67) and additional relevant proteins [Bcl-2, cyclin D1, E-cadherin, Snail+Slug, gross cystic disease fluid protein 15 (GCDFP-15), programmed death receptor 1 (PD-L1), and phosphatase of regenerating liver 3 (PRL-3)] in breast primary tumors (PTs) and paired synchronous axillary lymph node (ALN) metastases. A total of 67 tissue samples met the inclusion criteria for the study. The expression status of biomarkers was assessed in PTs and ALN metastases using tissue microarrays followed by IHC. In 11 cases, the shift of intrinsic molecular BC subtype was noticed between PTs and paired ALN metastases. Moreover, a significant disproportion in E-cadherin presence (p = 0.0002) was noted in both foci, and the expression status of all proteins except for HER2 demonstrated considerable variance (k = 1, p &lt; 0.0001). Importantly, in around 30% of cases, the ALN metastases demonstrated discordance, i.e., loss/gain of expression, compared to the PTs. Intertumoral synchronous heterogeneity in both foci (primary tumor and node metastasis) is an essential phenomenon affecting the clinical subtype and characteristics of BC. Furthermore, a greater understanding of this event could potentially improve therapeutic efficacy.