Are there features that can predict the unresectability of pleural mesothelioma?

INTRODUCTION: The current clinical staging of pleural mesothelioma (PM) is often discordant with the pathologic staging. This study aimed to identify clinical and radiological features that could help predict unresectability in PM. METHODS: Twenty-two descriptive radiologic features were retrospectively evaluated on preoperative computed tomography (CT) and/or positron emission tomography/CT (PET/CT) performed in patients with presumably resectable PM who underwent surgery. Measurements of maximum and sum pleural thickness at three levels of the thorax (upper, middle, and lower) were taken and stratified based on the cutpoints provided by the International Association for the Study of Lung Cancer (IASLC). Clinical and radiological features, including clinical-stage, were compared between resectable and unresectable tumors by univariate analysis and logistic regression modeling. RESULTS: Of 133 patients, 69/133 (52%) had resectable and 64/133 (48%) had unresectable PM. Asbestos exposure (p = 0.005), neoadjuvant treatment (p = 0.001), clinical T-stage (p < 0.0001), all pleural thickness measurements (p < 0.05), pleural thickness pattern (p < 0.0001) and degree (p = 0.033), lung invasion (p = 0.004), extrapleural space obliteration (p < 0.0001), extension to subphrenic space (p = 0.0004), and two combination variables representing extensive diaphragmatic contact and/or chest wall involvement (p = 0.002) and mediastinal invasion (p < 0.0001) were significant predictors at univariate analysis. At multivariable analysis, all models achieved a strong diagnostic performance (area under the curve (AUC) > 0.8). The two best-performing models were one that included the upper-level maximum pleural thickness, extrapleural space obliteration, and mediastinal infiltration (AUC = 0.876), and another that integrated clinical variables and radiological assessment through the clinical T-stage (AUC = 0.879). CONCLUSION: Selected clinical and radiologic features, including pleural thickness measurements, appear to be strong predictors of unresectability in PM. CLINICAL RELEVANCE STATEMENT: A more accurate prediction of unresectability in the preoperative assessment of patients with pleural mesothelioma may avoid unnecessary surgery and prompt initiation of nonsurgical treatments. KEY POINTS: About half of pleural mesothelioma patients are reported to receive an incorrect disease stage preoperatively. Eleven features identified as predictors of unresectability were included in strongly performing predictive models. More accurate preoperative staging will help clinicians and patients choose the most appropriate treatments.

Noncompaction cardiomyopathy: a substrate for a thromboembolic event.

BACKGROUND: Noncompaction cardiomyopathy (NCC) is a rare genetic cardiomyopathy characterized by a thin, compacted epicardial layer and an extensive noncompacted endocardial layer. The clinical manifestations of this disease include ventricular arrhythmia, heart failure, and systemic thromboembolism. CASE PRESENTATION: A 43-year-old male was anticoagulated by pulmonary thromboembolism for 1 year when he developed progressive dyspnea. Cardiovascular magnetic resonance imaging showed severe biventricular trabeculation with an ejection fraction of 15%, ratio of maximum noncompacted/compacted diastolic myocardial thickness of 3.2 and the presence of exuberant biventricular apical thrombus. CONCLUSION: Still under discussion is the issue of which patients and when they should be anticoagulated. It is generally recommended to those presenting ventricular systolic dysfunction, antecedent of systemic embolism, presence of cardiac thrombus and atrial fibrillation. In clinical practice the patients with NCC and ventricular dysfunction have been given oral anticoagulation, although there are no clinical trials showing the real safety and benefit of this treatment.

Conventional and radiomic features to predict pathology in the preoperative assessment of anterior mediastinal masses.

OBJECTIVES: The aim of this study was to differentiate benign from malignant tumors in the anterior mediastinum based on computed tomography (CT) imaging characteristics, which could be useful in preoperative planning. Additionally, our secondary aim was to differentiate thymoma from thymic carcinoma, which could guide the use of neoadjuvant therapy. MATERIALS AND METHODS: Patients referred for thymectomy were retrospectively selected from our database. Twenty-five conventional characteristics were evaluated by visual analysis, and 101 radiomic features were extracted from each CT. In the step of model training, we applied support vector machines to train classification models. Model performance was assessed using the area under the receiver operating curves (AUC). RESULTS: Our final study sample comprised 239 patients, 59 (24.7 %) with benign mediastinal lesions and 180 (75.3 %) with malignant thymic tumors. Among the malignant masses, there were 140 (58.6 %) thymomas, 23 (9.6 %) thymic carcinomas, and 17 (7.1 %) non-thymic lesions. For the benign versus malignant differentiation, the model that integrated both conventional and radiomic features achieved the highest diagnostic performance (AUC = 0.715), in comparison to the conventional (AUC = 0.605) and radiomic-only (AUC = 0.678) models. Similarly, regarding thymoma versus thymic carcinoma differentiation, the model that integrated both conventional and radiomic features also achieved the highest diagnostic performance (AUC = 0.810), in comparison to the conventional (AUC = 0.558) and radiomic-only (AUC = 0.774) models. CONCLUSION: CT-based conventional and radiomic features with machine learning analysis could be useful for predicting pathologic diagnoses of anterior mediastinal masses. The diagnostic performance was moderate for differentiating benign from malignant lesions and good for differentiating thymomas from thymic carcinomas. The best diagnostic performance was achieved when both conventional and radiomic features were integrated in the machine learning algorithms.

Radiogenomics in personalized management of lung cancer patients: Where are we?

With the rise of artificial intelligence, radiomics has emerged as a field of translational research based on the extraction of mineable high-dimensional data from radiological images to create "big data" datasets for the purpose of identifying distinct sub-visual imaging patterns. The integrated analysis of radiomic data and genomic data is termed radiogenomics, a promising strategy to identify potential imaging biomarkers for predicting driver mutations and other genomic parameters. In lung cancer, recent advances in whole-genome sequencing and the identification of actionable molecular alterations have led to an increased interest in understanding the complex relationships between imaging and genomic data, with the potential of guiding therapeutic strategies and predicting clinical outcomes. Although the integration of the radiogenomics data into lung cancer management may represent a new paradigm in the field, the use of this technique as a clinical biomarker remains investigational and still necessitates standardization and robustness to be effectively translated into the clinical practice. This review summarizes the basic concepts, potential contributions, challenges, and opportunities of radiogenomics in the management of patients with lung cancer.

CT Radiomic Features for Predicting Resectability and TNM Staging in Thymic Epithelial Tumors.

BACKGROUND: To explore the performance of a computed tomography based radiomics model in the preoperative prediction of resectability status and TNM staging in thymic epithelial tumors. METHODS: We reviewed the last preoperative computed tomography scan of patients with thymic epithelial tumors prior to resection and pathology evaluation at our institution between February 2008 and June 2019. A total of 101 quantitative features were extracted and a radiomics model was trained using elastic net penalized logistic regressions for each aim. In the set-aside testing sets, discriminating performance of each model was assessed with area under receiver operating characteristic curve. RESULTS: Our final population consisted of 243 patients with: 153 (87%) thymomas, 23 (9%) thymic carcinomas, and 9 (4%) thymic carcinoids. Incomplete resections (R1 or R2) occurred in 38 (16%) patients, and 67 (28%) patients had more advanced stage tumors (stage III or IV). In the set-aside testing sets, the radiomics model achieved good performance in preoperatively predicting incomplete resections (area under receiver operating characteristic curve: 0.80) and advanced stage tumors (area under receiver operating characteristic curve: 0.70). CONCLUSIONS: Our computed tomography radiomics model achieved good performance to predict resectability status and staging in thymic epithelial tumors, suggesting a potential value for the evaluation of radiomic features in the preoperative prediction of surgical outcomes in thymic malignancies.

Pre-treatment CT imaging in stage IIIA lung cancer: Can we predict local recurrence after definitive chemoradiotherapy?

OBJECTIVES: The aim of this study was to delineate computed tomography (CT) features of stage IIIA non-small cell lung cancers on pre-treatment staging studies and identify features that could predict local recurrence after definitive concurrent chemoradiotherapy. MATERIALS AND METHODS: We retrospectively reviewed pre- and post-treatment CT scans for 91 patients with Stage IIIA non-small cell lung cancer undergoing definitive concurrent chemoradiotherapy. Pre-treatment CT qualitative features were evaluated by consensus. The primary endpoint was local recurrence as determined on post-treatment CT scans along with the radiotherapy fields. Local recurrence was defined as intrathoracic in-field and marginal as opposed to out-of-field failures. Competing risk regressions were used to examine associations between CT features and recurrence. RESULTS: The median follow-up was 51.5 months (range 2.4-111.2). Median overall survival was 25.6 months (95% CI: 20.4-30). At last follow-up, 72 (79.1%) patients had died, 48 (52.7%) had in-field recurrence, and 30 (32.9%) presented with out-of-field recurrence. On pre-treatment CT scans, tumors presenting as pulmonary consolidations (hazard ratio = 2.34, 95% CI: 1.05-5.22; p 0.038) were more likely to have in-field failure. Tumors with 50-100% necrosis (hazard ratio = 0.15, 95% CI: 0.02-1.06) were associated with decreased out-of-field failure (overall p = 0.038). However, these were rare features in our sample which limit the ability of these features to be associated with such outcomes. CONCLUSIONS: Pre-treatment CT features alone are limited in predicting locoregional recurrence. Larger studies using quantitative tools are needed to predict such outcomes.

Are there imaging characteristics that can distinguish separate primary lung carcinomas from intrapulmonary metastases using next-generation sequencing as a gold standard?

OBJECTIVES: Distinguishing separate primary lung carcinomas (SPLCs) from intrapulmonary metastases (IPMs) in non-small cell lung cancer (NSCLC) patients is a challenging dilemma in clinical practice. Next-generation sequencing (NGS) was recently shown to represent a robust molecular method for clonal discrimination in this setting. In this study, using clonal relationships established by comprehensive NGS as the ground truth, we investigated whether NSCLC patients with SPLCs versus IPMs exhibit distinct imaging characteristics. MATERIAL AND METHODS: This retrospective study included patients who underwent pre-treatment computed tomography (CT) and/or positron emission tomography/CT (PET/CT) imaging followed by surgical resection for >1 NSCLC. Nodular, parenchymal, pleural, and ancillary CT features, as well as maximum standardized uptake values (SUVs) on PET/CT were recorded. Rao-Scott chi-square, Wilcoxon rank-sum, and Fisher's exact tests were used in patient- and lesion-level comparisons. RESULTS: This study included 60 patients (median age = 69 years, 68 % female) with 127 individual tumors comprising 51 SPLC vs 23 IPM tumor pairs based on NGS profiling. SPLCs were associated with subsolid consistency (P = 0.005) and spiculated contours (P <  0.001), while IPMs were associated with greater difference of size between lesions (P = 0.017) or pure solid consistency of the smaller lesion (P = 0.011). Lymph node involvement was more frequent in IPMs than SPLCs (P = 0.036). SUV measurements were not useful for differentiation (P > 0.05). CONCLUSION: Selected preoperative CT features are distributed differentially in SPLCs and IPMs, suggesting that imaging may have a role in distinguishing clonal relationships of tumors in patients with >1 NSCLC.