Citations

e16550 Background: Improved computational power and modern algorithms have generated significant interest in radiomics for cancer diagnosis and staging. Here we assess the performance of deep learning (DL) models as a means for feature extraction in combination with supervised machine learning (ML) algorithms for accurate staging and chemotherapy response assessment of bladder cancer. Methods: Deidentified grayscale CT images from bladder cancer patients scheduled to undergo radical cystectomy were included in this retrospective study. These images were manually annotated with two regional masks (normal region and cancer region). Five DL models- namely, AlexNet, GoogleNet, InceptionV3, ResNet-50, and XceptionNet pre-trained on the ImageNet dataset, a public dataset, were then fine-tuned on our bladder CT scan data to extract features. Through feature selection process, the subset of the features was used to build ML classifiers for classification. The classification was performed using five different ML classifiers, namely k-Nearest Neighbor (KNN), Naïve-Bayes (NB), Support Vector Machine (SVM), Linear Discriminant Analysis (LDA), and Decision Tree (DT). The classification task was performed with 10-fold cross-validation, and each of the experiments contained a different but not mutually exclusive subset of samples. The evaluation metrics include accuracy, sensitivity, specificity, precision, and F1-score. Results: A total of 200 deidentified grayscale CT images of 100 patients with histologically proven bladder cancer, were included in this study. For experiment (1) normal vs. cancer, the LDA classifier on XceptionNet based features provides the best performance with an accuracy of 86.07%, sensitivity of 96.75%, specificity of 69.65%, precision of 83.07%, and F1-score of 89.39%. For experiment (2) non-muscle invasive Cancer (NMIBC) vs. muscle invasive bladder cancer (MIBC), the LDA classifier on XceptionNet based features provided the best performance with an accuracy of 79.72%, sensitivity of 66.62%, specificity of 87.39%, precision of 75.58%, and F1-score of 70.81%. For experiment (3) T0 lesion vs. MIBC, the LDA classifier on XceptionNet based features provides the best performance with an accuracy of 74.96%, sensitivity of 80.51%, specificity of 70.22%, precision of 69.78%, and F1-score of 74.73%. Conclusions: Our proposed model has shown good results in differentiating normal vs cancer and promising performance in differentiating T0 vs MIBC after chemotherapy treatment. We are expanding our dataset to further improve the performance in differentiating T0 vs MIBC. In addition, we will investigate the applicability of GAN for data augmentation to address data limit. We believe the hybrid DL and ML framework may facilitates radiologists' decisions and clinical decision-making in patients with bladder cancer.

The burden of liver diseases is increasing worldwide, with liver transplantation remaining the only treatment option for end‐stage liver disease. Regenerative medicine holds great potential as a therapeutic alternative, aiming to repair or replace damaged liver tissue with healthy functional cells. The properties of the cells used are critical for the efficacy of this approach. The advent of liver organoids has not only offered new insights into human physiology and pathophysiology, but also provided an optimal source of cells for regenerative medicine and translational applications. Here, we discuss various historical aspects of 3D organoid culture, how it has been applied to the hepatobiliary system, and how organoid technology intersects with the emerging global field of liver regenerative medicine. We outline the hepatocyte, cholangiocyte, and nonparenchymal organoids systems available and discuss their advantages and limitations for regenerative medicine as well as future directions.

Abstract Cyclooxygenase-2 (COX-2) inhibitors are rapidly emerging as a new generation of therapeutic drug in combination with chemotherapy or radiation therapy for the treatment of cancer. The mechanisms underlying its antitumor effects are not fully understood and more thorough preclinical trials are needed to determine if COX-2 inhibition represents a useful approach for prevention and/or treatment of breast cancer. The purpose of this study was to evaluate the growth inhibitory mechanism of a highly selective COX-2 inhibitor, celecoxib, in an in vivo oncogenic mouse model of spontaneous breast cancer that resembles human disease. The oncogenic mice carry the polyoma middle T antigen driven by the mouse mammary tumor virus promoter and develop primary adenocarcinomas of the breast. Results show that oral administration of celecoxib caused significant reduction in mammary tumor burden associated with increased tumor cell apoptosis and decreased proliferation in vivo. In vivo apoptosis correlated with significant decrease in activation of protein kinase B/Akt, a cell survival signaling kinase, with increased expression of the proapoptotic protein Bax and decreased expression of the antiapoptotic protein Bcl-2. In addition, celecoxib treatment reduced levels of proangiogenic factor (vascular endothelial growth factor), suggesting a role of celecoxib in suppression of angiogenesis in this model. Results from these preclinical studies will form the basis for assessing the feasibility of celecoxib therapy alone or in combination with conventional therapies for treatment and/or prevention of breast cancer.

Abstract The pre‐mRNA‐processing factor 8, encoded by PRPF8, is a scaffolding component of a spliceosome complex involved in the removal of introns from mRNA precursors. Previously, heterozygous pathogenic variants in PRPF8 have been associated with autosomal dominant retinitis pigmentosa. More recently, PRPF8 was suggested as a candidate gene for autism spectrum disorder due to the enrichment of sequence variants in this gene in individuals with neurodevelopmental disorders. We report 14 individuals with various forms of neurodevelopmental conditions, found to have heterozygous, predominantly de novo, missense, and loss‐of‐function variants in PRPF8 . These individuals have clinical features that may represent a new neurodevelopmental syndrome.

Background Intensive care unit (ICU) sedation guidelines recommend targeting a light sedation level, but light sedation has no accepted definition, and inconsistent levels have been proposed. Objective To determine Sedation-Agitation Scale and Richmond Agitation-Sedation Scale scores that best describe patients’ ability to follow voice commands. Methods This prospective, observational pilot study enrolled a convenience sample of ICU patients receiving mechanical ventilation. Pairs of trained investigators evaluated scores on the Sedation-Agitation Scale and Richmond Agitation-Sedation Scale and ability to follow commands before and up to 2 hours after sedation lightening in a blind, independent, simultaneous fashion. Positive predictive values (PPVs) and likelihood ratios (LRs) of Sedation-Agitation Scale and Richmond Agitation-Sedation Scale scores associated with light sedation (ability to follow at least 3 commands) were calculated. Results Ninety-six assessments (50 before and 46 after lightening of sedation) were performed in medical ICU patients. Scores best associated with ability to follow at least 3 commands were Sedation-Agitation Scale score of 4 (PPV, 0.88; 95% CI, 0.70-0.98; LR, 14.0) and Richmond Agitation-Sedation Scale score of −1 (PPV, 0.81; 95% CI, 0.61-0.93; LR, 10.7), superior to previously recommended thresholds of Sedation-Agitation Scale score of 3 (PPV, 0.62; 95% CI, 0.48-0.75; LR, 3.1) and Richmond Agitation-Sedation Scale score of −3 (PPV, 0.52; 95% CI, 0.39-0.64; LR, 2.0). Conclusions The level of sedation most associated with the ability to follow commands appears higher than previously recommended. Further study is needed regarding the effects of sedation level on ICU patients’ ability to follow commands and assessment of delirium, pain, and patient preferences.

Background Current mortality data for pulmonary arterial hypertension (PAH) in the United States are based on registries that enrolled patients prior to 2010. We sought to determine mortality in PAH in the modern era using the PHAR (Pulmonary Hypertension Association Registry). Methods and Results We identified all adult patients with PAH enrolled in the PHAR between September 2015 and September 2020 (N=935). We used Kaplan‐Meier survival analysis and Cox proportional hazards models to assess mortality at 1, 2, and 3 years. Patients were stratified based on disease severity by 3 validated risk scores. In treatment‐naïve patients, we compared survival based on initial treatment strategy. The median age was 56 years (44–68 years), and 76% were women. Of the 935 patients, 483 (52%) were ≤6 months from PAH diagnosis. There were 121 deaths (12.9%) during a median follow‐up time of 489 days (281–812 days). The 1‐, 2‐, and 3‐year mortality was 8% (95% CI, 6%–10%), 16% (95% CI, 13%–19%), and 21% (95% CI, 17%–25%), respectively. When stratified into low‐, intermediate‐, and high‐risk PAH, the mortality at 1, 2, and 3 years was 1%, 4% to 6%, and 7% to 11% for low risk; 7% to 8%, 11% to 16%, and 18% to 20% for intermediate risk; and 12% to 19%, 22% to 38%, and 28% to 55% for high risk, respectively. In treatment‐naïve patients, initial combination therapy was associated with better 1‐year survival (adjusted hazard ratio, 0.43 [95% CI, 0.19–0.95]; P =0.037). Conclusions Mortality in the intermediate‐ and high‐risk patients with PAH remains unacceptably high in the PHAR, suggesting the importance for early diagnosis, aggressive use of available therapies, and the need for better therapeutics.

ABSTRACT Background The second consensus criteria for the diagnosis of multiple system atrophy (MSA) are widely recognized as the reference standard for clinical research, but lack sensitivity to diagnose the disease at early stages. Objective To develop novel Movement Disorder Society (MDS) criteria for MSA diagnosis using an evidence‐based and consensus‐based methodology. Methods We identified shortcomings of the second consensus criteria for MSA diagnosis and conducted a systematic literature review to answer predefined questions on clinical presentation and diagnostic tools relevant for MSA diagnosis. The criteria were developed and later optimized using two Delphi rounds within the MSA Criteria Revision Task Force, a survey for MDS membership, and a virtual Consensus Conference. Results The criteria for neuropathologically established MSA remain unchanged. For a clinical MSA diagnosis a new category of clinically established MSA is introduced, aiming for maximum specificity with acceptable sensitivity. A category of clinically probable MSA is defined to enhance sensitivity while maintaining specificity. A research category of possible prodromal MSA is designed to capture patients in the earliest stages when symptoms and signs are present, but do not meet the threshold for clinically established or clinically probable MSA. Brain magnetic resonance imaging markers suggestive of MSA are required for the diagnosis of clinically established MSA. The number of research biomarkers that support all clinical diagnostic categories will likely grow. Conclusions This set of MDS MSA diagnostic criteria aims at improving the diagnostic accuracy, particularly in early disease stages. It requires validation in a prospective clinical and a clinicopathological study. © 2022 The Authors. Movement Disorders published by Wiley Periodicals LLC on behalf of International Parkinson and Movement Disorder Society

Abstract Antispike monoclonal antibody treatment of 180 B-cell-depleted patients with mild-to-moderate coronavirus disease 2019 (COVID-19) resulted in good outcomes overall, with only 12.2% progressing to severe disease, 9.4% requiring hospitalization, 0.6% requiring mechanical ventilation, no deaths within 30 days, and 1.8% developing persistent COVID-19. Antispike monoclonal antibodies appear effective in this immunocompromised population.