Integration Of Clinical Data Research Articles

The Rare Instance of Uterine Adenocarcinoma Metastasizing to the Skull Base, Manifesting as Blindness and Ptosis

Metastasis to the skull base is a rare but serious complication of various malignancies. This case report discusses a 50-year-old woman with a history of advanced endometrial adenocarcinoma who presented with progressive headache, left-eye blindness, and ptosis. Imaging revealed a mass along the left optic nerve, raising suspicions of mucormycosis, but biopsy confirmed metastatic endometrial adenocarcinoma. Despite undergoing skull base radiotherapy, the patient’s vision did not improve. This case underscores the importance of considering metastatic endometrial adenocarcinoma in differential diagnoses for skull base lesions, particularly when other conditions are ruled out. Accurate diagnosis requires careful integration of clinical, radiological, and histopathological data.

Drug-gene interactions in older patients with coronary artery disease

BackgroundOlder patients with coronary artery disease (CAD) are particularly vulnerable to the efficacy and adverse drug reactions, and may therefore particularly benefit from personalized medication. Drug–gene interactions (DGIs) occur when an individual’s genotype affects the pharmacokinetics and/or pharmacodynamics of a victim drug.ObjectivesThis study aimed to investigate the impact of cardiovascular-related DGIs on the clinical efficacy and safety outcomes in older patients with CAD.MethodsHospitalized older patients (≥ 65 years old) with CAD were consecutively recruited from August 2018 to May 2022. Eligible patients were genotyped for the actionable pharmacogenetic variants of CYP2C9, CYP2C19, CYP2D6, CYP3A5, and SLCO1B1, which had clinical annotations or implementation guidelines for cardiovascular drugs. Allele frequencies and DGIs were determined in the cohort for the 5 actionable PGx genes and the prescribed cardiovascular drugs. All patients were followed up for at least 1 year. The influence of DGIs on the cardiovascular drug-related efficacy outcomes (all-cause mortality and/or major cardiovascular events, MACEs) and drug response phenotypes of “drug-stop” and “dose-decrease” were evaluated.ResultsA total of 1,017 eligible older patients with CAD were included, among whom 63.2% were male, with an average age of 80.8 years old, and 87.6% were administrated with polypharmacy (≥ 5 medications). After genotyping, we found that 96.0% of the older patients with CAD patients had at least one allele of the 5 pharmacogenes associated with a therapeutic change, indicating a need for a therapeutic change in a mean of 1.32 drugs of the 19 cardiovascular-related drugs. We also identified that 79.5% of the patients had at least one DGI (range 0–6). The median follow-up interval was 39 months. Independent of age, negative association could be found between the number of DGIs and all-cause mortality (adjusted HR: 0.84, 95% CI: 0.73–0.96, P = 0.008), and MACEs (adjusted HR: 0.84, 95% CI: 0.72–0.98, P = 0.023), but positive association could be found between the number of DGIs and drug response phenotypes (adjusted OR: 1.24, 95% CI: 1.05–1.45, P = 0.011) in the elderly patients with CAD.ConclusionsThe association between cardiovascular DGIs and the clinical outcomes emphasized the necessity for the integration of genetic and clinical data to enhance the optimization of cardiovascular polypharmacy in older patients with CAD. The causal relationship between DGIs and the clinical outcomes should be established in the large scale prospectively designed cohort study.

Osteoarthritis Year in Review 2024: Molecular biomarkers of osteoarthritis

ObjectiveTo provide a comprehensive and insightful summary of studies on molecular biomarkers at the gene, protein, and metabolite levels across different sample types and joints affected by osteoarthritis (OA). MethodsA literature search using the PubMed database for publications on OA biomarkers published between April 1, 2023 and April 30, 2024 was performed. Publications were then screened, examined at length, and summarized in a narrative review. ResultsOut of the 364 papers initially identified, 44 publications met inclusion criteria, were relevant to OA, and were further examined for data extraction and discussion. These studies included 1 genomic analysis, 22 on protein markers, 6 on metabolite markers, 9 on inflammatory mediators, and 6 integrating multiple molecular levels. ConclusionsSignificant advancements have been made in identifying molecular biomarkers for OA, encompassing various joints, sample types, and molecular levels. Despite this progress, gaps remain, particularly in the need for validation, larger sample sizes, the integration of more clinical data, and consideration of covariates. For early detection and improved treatment of OA, continued efforts in biomarker identification are needed. This effort should seek to identify effective biomarkers that advance early detection, support prevention, evaluate interventions, and improve patient outcomes.

ADVANCING SKIN CANCER DETECTION USING MULTIMODAL DATA FUSION AND AI TECHNIQUES

Cancer remains a leading cause of death worldwide, driving the need for continuous advancements in early detection and treatment. Deep learning, a subset of artificial intelligence, has become a transformative tool in medical image analysis, significantly improving cancer diagnosis. This study explores various modalities used in lung cancer diagnosis, including medical imaging (e.g., radiology, pathology), genomics, and clinical data, addressing the specific challenges of each domain. The proposed Multimodal Fusion Deep Neural Network (MFDNN) effectively integrates these diverse data sources to enhance diagnostic accuracy. Additionally, it emphasizes the integration of clinical data and electronic health records, demonstrating the value of multimodal approaches for improving reliability in lung cancer diagnosis. Ethical considerations related to AI in clinical settings, along with the need for validation and regulatory guidelines, are also discussed.

HL7 FHIR in Health Research: A FHIR Specification for Metadata in Clinical, Epidemiological, and Public Health Studies.

This work presents the Fast Healthcare Interoperability Resources (FHIR®) specification of the NFDI4Health Metadata schema based on FHIR Version 4: We created 16 profiles to facilitate the integration of clinical, epidemiological, and public health study data. Despite challenges arising from the extensive MDS as well as missing concepts in semantic standards, it marks a significant advance in applying information technology standards to health research.

Optimizing Glioblastoma, IDH-wildtype Treatment Outcomes : A Radiomics and Support Vector Machine -Based Approach to Overall Survival Estimation.

Glioblastoma multiforme (GBM), particularly the IDH-wildtype type, represents a significant clinical challenge due to its aggressive nature and poor prognosis. Despite advancements in medical imaging and its modalities, survival rates have not improved significantly, demanding innovative treatment planning and outcome prediction approaches. This study utilizes a Support Vector Machine (SVM) classifier using radiomics features to predict the overall survival (OS) of GBM, IDH-wildtype patients to short (< 12 Months) and long (>=12 Months) survivors. A dataset comprising multi-parametric MRI (mpMRI) scans from 574 patients was analyzed. Radiomic features were extracted from T1, T2, FLAIR, and T1-Gd sequences. Low variance features were removed, and Recursive Feature Elimination (RFE) was used to select the most informative features. The SVM model was trained using a k-fold cross-validation approach. Furthermore, clinical parameters such as age, gender, and MGMT promoter methylation status were integrated to enhance prediction accuracy. The model showed reasonable results in terms of cross-validated AUC of 0.84 (95% CI: 0.80-0.90) with (p-value < 0.001) effectively categorizing patients into short and long survivors. Log-rank test (Chi-square statistics) analysis for the developed model was 0.00029 along with the 1.20 Cohen's d effect size. Most importantly, clinical data integration further refined the survival estimates, providing a more fitted prediction that considers individual patient characteristics by Kaplan-Meier curve with p-value<0.0001. The proposed method significantly enhances the predictive accuracy of OS outcomes in GBM, IDH-wildtype patients. By integrating detailed imaging features with key clinical indicators, this model offers a robust tool for personalized treatment planning, potentially improving OS.

Ontology-based modeling, integration, and analysis of heterogeneous clinical, pathological, and molecular kidney data for precision medicine.

Many data resources generate, process, store, or provide kidney related molecular, pathological, and clinical data. Reference ontologies offer an opportunity to support knowledge and data integration. The Kidney Precision Medicine Project (KPMP) team contributed to the representation and addition of 329 kidney phenotype terms to the Human Phenotype Ontology (HPO), and identified many subcategories of acute kidney injury (AKI) or chronic kidney disease (CKD). The Kidney Tissue Atlas Ontology (KTAO) imports and integrates kidney-related terms from existing ontologies (e.g., HPO, CL, and Uberon) and represents 259 kidney-related biomarkers. We have also developed a precision medicine metadata ontology (PMMO) to integrate 50 variables from KPMP and CZ CellxGene data resources and applied PMMO for integrative kidney data analysis. The gene expression profiles of kidney gene biomarkers were specifically analyzed under healthy control or AKI/CKD disease states. This work demonstrates how ontology-based approaches support multi-domain data and knowledge integration in precision medicine.

Preoperative prediction of microvascular invasion risk in hepatocellular carcinoma with MRI: peritumoral versus tumor region

ObjectivesTo explore the predictive performance of tumor and multiple peritumoral regions on dynamic contrast-enhanced magnetic resonance imaging (MRI), to identify optimal regions of interest for developing a preoperative predictive model for the grade of microvascular invasion (MVI).MethodsA total of 147 patients who were surgically diagnosed with hepatocellular carcinoma, and had a maximum tumor diameter ≤ 5 cm were recruited and subsequently divided into a training set (n = 117) and a testing set (n = 30) based on the date of surgery. We utilized a pre-trained AlexNet to extract deep learning features from seven different regions of the maximum transverse cross-section of tumors in various MRI sequence images. Subsequently, an extreme gradient boosting (XGBoost) classifier was employed to construct the MVI grade prediction model, with evaluation based on the area under the curve (AUC).ResultsThe XGBoost classifier trained with data from the 20-mm peritumoral region showed superior AUC compared to the tumor region alone. AUC values consistently increased when utilizing data from 5-mm, 10-mm, and 20-mm peritumoral regions. Combining arterial and delayed-phase data yielded the highest predictive performance, with micro- and macro-average AUCs of 0.78 and 0.74, respectively. Integration of clinical data further improved AUCs values to 0.83 and 0.80.ConclusionCompared with those of the tumor region, the deep learning features of the peritumoral region provide more important information for predicting the grade of MVI. Combining the tumor region and the 20-mm peritumoral region resulted in a relatively ideal and accurate region within which the grade of MVI can be predicted.Clinical relevance statementThe 20-mm peritumoral region holds more significance than the tumor region in predicting MVI grade. Deep learning features can indirectly predict MVI by extracting information from the tumor region and directly capturing MVI information from the peritumoral region.Key PointsWe investigated tumor and different peritumoral regions, as well as their fusion.MVI predominantly occurs in the peritumoral region, a superior predictor compared to the tumor region.The peritumoral 20 mm region is reasonable for accurately predicting the three-grade MVI.Graphical

The Chilean exposome-based system for ecosystems (CHiESS): a framework for national data integration and analytics platform.

The double burden of diseases and scarce resources in developing countries highlight the need to change the conceptualization of health problems and translational research. Contrary to the traditional paradigm focused on genetics, the exposome paradigm proposed in 2005 that complements the genome is an innovative theory. It involves a holistic approach to understanding the complexity of the interactions between the human being’s environment throughout their life and health. This paper outlines a scalable framework for exposome research, integrating diverse data sources for comprehensive public health surveillance and policy support. The Chilean exposome-based system for ecosystems (CHiESS) project proposes a conceptual model based on the ecological and One Health approaches, and the development of a technological dynamic platform for exposome research, which leverages available administrative data routinely collected by national agencies, in clinical records, and by biobanks. CHiESS considers a multilevel exposure for exposome operationalization, including the ecosystem, community, population, and individual levels. CHiESS will include four consecutive stages for development into an informatic platform: (1) environmental data integration and harmonization system, (2) clinical and omics data integration, (3) advanced analytical algorithm development, and (4) visualization interface development and targeted population-based cohort recruitment. The CHiESS platform aims to integrate and harmonize available secondary administrative data and provide a complete geospatial mapping of the external exposome. Additionally, it aims to analyze complex interactions between environmental stressors of the ecosystem and molecular processes of the human being and their effect on human health. Moreover, by identifying exposome-based hotspots, CHiESS allows the targeted and efficient recruitment of population-based cohorts for translational research and impact evaluation. Utilizing advanced technologies such as Artificial Intelligence (AI), Internet of Things (IoT), and blockchain, this framework enhances data security, real-time monitoring, and predictive analytics. The CHiESS model is adaptable for international use, promoting global health collaboration and supporting sustainable development goals.

Integrating Clinical Data and Medical Imaging in Lung Cancer: Feasibility Study Using the Observational Medical Outcomes Partnership Common Data Model Extension.

Digital transformation, particularly the integration of medical imaging with clinical data, is vital in personalized medicine. The Observational Medical Outcomes Partnership (OMOP) Common Data Model (CDM) standardizes health data. However, integrating medical imaging remains a challenge. This study proposes a method for combining medical imaging data with the OMOP CDM to improve multimodal research. Our approach included the analysis and selection of digital imaging and communications in medicine header tags, validation of data formats, and alignment according to the OMOP CDM framework. The Fast Healthcare Interoperability Resources ImagingStudy profile guided our consistency in column naming and definitions. Imaging Common Data Model (I-CDM), constructed using the entity-attribute-value model, facilitates scalable and efficient medical imaging data management. For patients with lung cancer diagnosed between 2010 and 2017, we introduced 4 new tables-IMAGING_STUDY, IMAGING_SERIES, IMAGING_ANNOTATION, and FILEPATH-to standardize various imaging-related data and link to clinical data. This framework underscores the effectiveness of I-CDM in enhancing our understanding of lung cancer diagnostics and treatment strategies. The implementation of the I-CDM tables enabled the structured organization of a comprehensive data set, including 282,098 IMAGING_STUDY, 5,674,425 IMAGING_SERIES, and 48,536 IMAGING_ANNOTATION records, illustrating the extensive scope and depth of the approach. A scenario-based analysis using actual data from patients with lung cancer underscored the feasibility of our approach. A data quality check applying 44 specific rules confirmed the high integrity of the constructed data set, with all checks successfully passed, underscoring the reliability of our findings. These findings indicate that I-CDM can improve the integration and analysis of medical imaging and clinical data. By addressing the challenges in data standardization and management, our approach contributes toward enhancing diagnostics and treatment strategies. Future research should expand the application of I-CDM to diverse disease populations and explore its wide-ranging utility for medical conditions.

Machine learning-based integration of omics and clinical data reveals an N-glycan biosynthesis signature predictive of the outcome in low-grade glioma: an in silico study

Machine learning-based integration of omics and clinical data reveals an N-glycan biosynthesis signature predictive of the outcome in low-grade glioma: an <i>in silico</i> study

Integration of individual preclinical and clinical anti-infective PKPD data to predict clinical study outcomes.

The AIDA randomized clinical trial found no significant difference in clinical failure or survival between colistin monotherapy and colistin-meropenem combination therapy in carbapenem-resistant Gram-negative infections. The aim of this reverse translational study was to integrate all individual preclinical and clinical pharmacokinetic-pharmacodynamic (PKPD) data from the AIDA trial in a pharmacometric framework to explore whether individualized predictions of bacterial burden were associated with the trial outcomes. The compiled dataset included for each of the 207 patients was (i) information on the infecting Acinetobacter baumannii isolate (minimum inhibitory concentration, checkerboard assay data, and fitness in a murine model), (ii) colistin plasma concentrations and colistin and meropenem dosing history, and (iii) disease scores and demographics. The individual information was integrated into PKPD models, and the predicted change in bacterial count at 24 h for each patient, as well as patient characteristics, was correlated with clinical outcomes using logistic regression. The invivo fitness was the most important factor for change in bacterial count. A model-predicted growth at 24 h of ≥2-log10 (164/207) correlated positively with clinical failure (adjusted odds ratio, aOR = 2.01). The aOR for one unit increase of other significant predictors were 1.24 for SOFA score, 1.19 for Charlson comorbidity index, and 1.01 for age. This study exemplifies how preclinical and clinical anti-infective PKPD data can be integrated through pharmacodynamic modeling and identify patient- and pathogen-specific factors related to clinical outcomes - an approach that may improve understanding of study outcomes.

OTHR-12. GENERATING A COMPREHENSIVE INTERACTIVE PEDIATRIC BRAIN TUMOR VISUALIZATION PORTAL FROM MOLECULAR AND CLINICAL DATA DERIVED FROM TWO MULTI-INSTITUTIONAL PROTOCOLS (SJMB03, SJYC07) AND NON-PROTOCOL PATIENTS TREATED AT ST JUDE CHILDREN’S RESEARCH HOSPITAL

Abstract BACKGROUND Pediatric central nervous system (CNS) tumors represent a diverse group of neoplasms with heterogeneous molecular and clinical characteristics. The need for well-organized, searchable datasets is paramount to enhance our understanding and improve the treatment approaches for these devastating diseases. To illustrate this, we have organized multifaceted data from patients into an interactive portal equipped with cutting-edge visualization tools. METHODS Clinical and molecular data from the SJMB03 and SJYC07 protocols and select published non-protocol St. Jude patients were collated and harmonized. A data dictionary was constructed to record demographic information, treatment details, histology, methylation classification, mutations, chromosomal aberrations, and outcomes. We implemented five visualization techniques - summary plots, sample view, scatter plots, sample matrix, and survival plots. These techniques are interconnected, enabling interaction and “crosstalk”. RESULTS Data from 710 patients with CNS tumors were incorporated: 386 Medulloblastomas (MB); 88 Atypical teratoid Rhabdoid Tumors (ATRT); 54 Ependymomas (EPN); 49 Pineoblastomas (PB); 28 Embryonal Tumors with Multilayered Rosettes (ETMR), 22 Infant-type hemispheric glioma (IHG), 16 High grade glioma (HGG), 13 Choroid Plexus Tumors (CPT); 11 CNS Sarcoma; 8 CNS tumor with BCOR internal tandem duplication (CNS BCOR); 6 CNS neuroblastoma FOXR2 activated (CNS NB FOXR2); 5 Low grade glioma (LGG); and 24 other entities or unclassified tumors. The portal features enabled the creation of custom cohorts. For example, patients with TP53 mutations (n=32; 15 MB, 7 CPT, 6 HGG, 1 CNS Sarcoma, 1 CNS NB FOXR2, 1 PB, 1 unclassified tumor) or DICER1 mutations (n=10; 7 PB, 2 CNS Sarcoma, and 1 ETMR) could be identified, cross-compared and analyzed adjusting for variables at users’ choice. CONCLUSION The integration of molecular and clinical data from multiple sources into an interactive visualization portal presents a comprehensive resource for research and clinical practice. Future work will focus on expanding the portal to integrate more cases/cohorts.

MDB-105. CONSTRUCTION OF AN INTERACTIVE MEDULLOBLASTOMA VISUALIZATION PORTAL FROM MOLECULAR AND CLINICAL DATA DERIVED FROM THREE MULTI-INSTITUTIONAL PROTOCOLS (ACNS0331, SJMB03, ACNS0332) TO COMPARE, CONTRAST, AND INVESTIGATE COHORTS AS A MEANS TO REFINE THERAPY

Abstract BACKGROUND Medulloblastoma (MB) patients display a heterogeneous mix of molecular and clinical characteristics that influence tumor behavior and survival. We hypothesized that if these multiple characteristics could be incorporated into an organized, searchable, and interactive dataset, this would advance our understanding and improve treatment of this disease. Consequently, we organized multifaceted datasets into an interactive portal equipped with cutting-edge visualization tools. METHODS Clinical and molecular data from the ACNS0331, SJMB03, and ACNS0332 protocols were collated and harmonized. A data dictionary was built to capture demographics, treatment, histology, methylation classification, mutations, chromosomal copy number variations and outcomes. To display the data, a software framework was applied. Five visualization methods were implemented (summary plots, sample view, scatter plots, sample matrix, and survival plots) in an interconnected manner such that these “cross-talk” with each other. RESULTS Data from 898 patients with MB were incorporated: 131 WNT, 151 SHH, 220 G3, 396 G4. The portal features enabled the creation of custom cohorts that could be instantly queried by any of the dictionary terms. For example, cohorts could be customized to include only cases with a mutation of the user’s choice and analyzed by molecular group, gender, age at diagnosis, or treatment given. Survival of patients with SHH-MB harboring TP53 mutations could be compared to patients with WNT-MB harboring TP53 mutations. Moreover, the influence of MB subtype (e.g. G34_I-VIII), focal amplifications (MYC, MYCN) and whole chromosome aberration phenotype on survival could be queried across treatment regimens accounting for metastatic status and/or additional variables of the user’s choice. CONCLUSION The integration of molecular and clinical data from multiple trials into an interactive visualization data portal presents an unprecedented resource for research and clinical practice. This portal facilitates answering highly specific questions about mutational frequency and patient outcome by molecular and clinical risk factors.

Predicting disease recurrence in breast cancer patients using machine learning models with clinical and radiomic characteristics: a retrospective study

BackgroundThe goal is to use three different machine learning models to predict the recurrence of breast cancer across a very heterogeneous sample of patients with varying disease kinds and stages.MethodsA heterogeneous group of patients with varying cancer kinds and stages, including both triple-negative breast cancer (TNBC) and non-triple-negative breast cancer (non-TNBC), was examined. Three distinct models were created using the following five machine learning techniques: Adaptive Boosting (AdaBoost), Random Under-sampling Boosting (RUSBoost), Extreme Gradient Boosting (XGBoost), support vector machines (SVM), and Logistic Regression. The clinical model used both clinical and pathology data in conjunction with the machine learning algorithms. The machine learning algorithms were combined with dynamic contrast-enhanced magnetic resonance imaging (DCE-MRI) imaging characteristics in the radiomic model, and the merged model combined the two types of data. Each technique was evaluated using several criteria, including the receiver operating characteristic (ROC) curve, precision, recall, and F1 score.ResultsThe results suggest that the integration of clinical and radiomic data improves the predictive accuracy in identifying instances of breast cancer recurrence. The XGBoost algorithm is widely recognized as the most effective algorithm in terms of performance.ConclusionThe findings presented in this study offer significant contributions to the field of breast cancer research, particularly in relation to the prediction of cancer recurrence. These insights hold great potential for informing future investigations and clinical interventions that seek to enhance the accuracy and effectiveness of recurrence prediction in breast cancer patients.

Multimodal fully automated predictive model for therapeutic efficacy of first-line cancer immunotherapy based on clinical information and imaging modalities including brain MRI and chest CT images in advanced non-small cell lung cancer.

1555 Background: The integration of imaging and clinical data through artificial intelligence (AI) holds promise for more accurately predicting the therapeutic efficacy of immune checkpoint inhibitors (ICI) ± platinum-based chemotherapy in individual patients with advanced non-small cell lung cancer (NSCLC). However, tumor segmentation still relies entirely on radiologist evaluation and has not yet achieved fully automated tumor evaluation for image analysis. Furthermore, while central nervous system (CNS) metastasis is a critical prognostic factor in patients with advanced NSCLC, radiomics has primarily focused on chest imaging, with few studies incorporating brain MRI analysis. Methods: This study included patients with advanced NSCLC treated with ICI ± platinum-based chemotherapy as first-line therapy from March 2017 to August 2023. We developed an ensemble model combining a logistic regression model that analyzes clinical information with a deep learning model that extracts image features. Each sub-model was independent, allowing for individual adjustments and the ability to be attached or detached. Both models were trained to predict response, 12-month progression-free survival (PFS), and 12-month overall survival (OS) using cross-entropy loss. The dataset was randomly divided into training, validation, and test sets in a 5:3:2 ratio. This random assignment was repeated four times, and the average area under the curve (AUC) was calculated. The test sets were further classified into high- and low-risk groups based on the predicted results, and log-rank tests were performed on their survival curves. Results: The study included 232 patients, with 89 (38%) receiving ICI alone and 143 (62%) receiving ICI + platinum-based chemotherapy. In the entire population, the overall response rate (ORR) for ICI ± platinum-based chemotherapy was 49.5%, with a median PFS of 7.1 months (95% confidence interval [CI]: 5.8-9.1) and a median OS of 24.5 months (95% CI: 21.2-30.3). The model achieved an AUC of 0.70 (95% CI: 0.65- 0.76) for predicting response, 0.65 (95% CI: 0.59-0.71) for 12-month PFS, and 0.69 (95% CI: 0.48-0.90) for 12-month OS. Test sets were classified into high-risk (N = 55) and low-risk (N = 60) groups. The median PFS was 6.8 months in the high-risk group and 14.7 months in the low-risk group (p &lt; 0.05). The median OS was 21.5 months in the high-risk group and 34.3 months in the low-risk group (p &lt; 0.05). Conclusions: We have demonstrated that a multimodal, fully automated ensemble model has significant predictive accuracy for the efficacy and survival outcomes of ICI ± platinum-based chemotherapy in patients with advanced NSCLC.

Advances in big data and omics: Paving the way for discovery in childhood epilepsies

The insights gained from big data and omics approaches have transformed the field of childhood genetic epilepsy. With an increasing number of individuals receiving genetic testing for seizures, we are provided with an opportunity to identify clinically relevant subgroups and extract meaningful observations from this large-scale clinical data. However, the volume of data from electronic medical records and omics (e.g., genomics, transcriptomics) is so vast that standardized methods, such as the Human Phenotype Ontology, are necessary for reliable and comprehensive characterization. Here, we explore the integration of clinical and omics data, highlighting how these approaches pave the way for discovery in childhood epilepsies.

Melanoma immunotherapy longitudinal database: A comprehensive approach to capture oncological and toxicity data.

e23314 Background: Evaluating immunotherapy approaches is consistently performed in melanoma, a forefront cancer in immunotherapy responsiveness. Although immune checkpoint inhibitors (ICIs) showed the highest significant survival benefits in melanoma, 50% of patients experience recurrence/refractory disease, and ~59% develop grade 3/4 immune-related adverse events (irAEs). Thus, understanding the disease's natural history, exploring biomarkers of response/toxicity, and identifying the efficacy/safety of immunotherapy using real-world data is warranted. Methods: We established a centralized pro-/retrospective database on Research Electronic Data Capture (REDCap), encompassing patients' demographics/comorbidities, disease risk factors, clinicopathological parameters, genomic/immune profiling, treatment modalities, tumor outcomes, irAEs, and patients’ reported outcomes (PROs). Additionally, to streamline data collection and minimize the required time, effort, and cost for manual retrieval, we developed an automated pathway for data capture using MD Anderson Cancer Center (MDACC) electronic data capture (EDC) tools which pull clinical data from EPIC. Data quality will be validated by comparing it to manually retrieved data from an ongoing study of melanoma patients treated with ICIs in the adjuvant setting (NCT04990726). Results: A total of 4074 immunotherapy-treated melanoma patients were identified. We automatically captured demographics and laboratory values at predefined time points from REDCap, along with PROs using validated electronic self-reported questionnaires already built in REDCap. We also automatically captured family history, comorbidities, pathology data, and cancer treatments from the Data Management Initiative (DMI) validated software that automatically captures data from 12 various institutional sources that complement each other, including Radiation Oncology Database, the Biospecimen Information Management System,...etc. Moreover, genomic data were automatically pulled from the Molecular and Clinical Data Integration Platform for 1850 patients. Automated capture of tumor response and irAEs according to the Response Evaluation Criteria in Solid Tumors 1.1. and Common Terminology Criteria for Adverse Events version-5, respectively are currently ongoing and will be presented. Conclusions: Integrating EDC tools demonstrated accurate real-time data capture and minimized the manual retrieval burden. Our immunotherapy-treated melanoma patients’ cohort provides a comprehensive and a valuable resource for exploring the safety signals and efficacy of various immunotherapeutic approaches beyond clinical trials.

A Multimodal Ensemble Deep Learning Model for Functional Outcome Prognosis of Stroke Patients.

The accurate prediction of functional outcomes in patients with acute ischemic stroke (AIS) is crucial for informed clinical decision-making and optimal resource utilization. As such, this study aimed to construct an ensemble deep learning model that integrates multimodal imaging and clinical data to predict the 90-day functional outcomes after AIS. We used data from the Korean Stroke Neuroimaging Initiative database, a prospective multicenter stroke registry to construct an ensemble model integrated individual 3D convolutional neural networks for diffusion-weighted imaging and fluid-attenuated inversion recovery (FLAIR), along with a deep neural network for clinical data, to predict 90-day functional independence after AIS using a modified Rankin Scale (mRS) of 3-6. To evaluate the performance of the ensemble model, we compared the area under the curve (AUC) of the proposed method with that of individual models trained on each modality to identify patients with AIS with an mRS score of 3-6. Of the 2,606 patients with AIS, 993 (38.1%) achieved an mRS score of 3-6 at 90 days post-stroke. Our model achieved AUC values of 0.830 (standard cross-validation [CV]) and 0.779 (time-based CV), which significantly outperformed the other models relying on single modalities: b-value of 1,000 s/mm2 (P<0.001), apparent diffusion coefficient map (P<0.001), FLAIR (P<0.001), and clinical data (P=0.004). The integration of multimodal imaging and clinical data resulted in superior prediction of the 90-day functional outcomes in AIS patients compared to the use of a single data modality.

Regional differences in predictive biomarker testing rates for patients with metastatic NSCLC in the Netherlands

BackgroundPredictive biomarker testing has a key role in the treatment decision-making for patients with non-small cell lung cancer (NSCLC) and is mandated by (inter)national guidelines. The aim of this study was to establish guideline-adherent biomarker testing rates in the Netherlands in 2019 and to examine associations of demographical, clinical, and environmental factors with guideline-adherent testing. MethodsThis study involved the integration of clinical data of the Netherlands Cancer Registry with pathology reports of the Dutch Nationwide Pathology Databank. Data extracted from these reports included sample type, diagnosis, and molecular testing status of predictive biomarkers. The study population comprised all patients diagnosed with metastatic non-squamous NSCLC in the Netherlands in 2019. ResultsIn the cohort of 3877 patients with metastatic non-squamous NSCLC under investigation, overall molecular testing rates for non-fusion predictive biomarkers (EGFR, KRAS, BRAF, ERBB2, MET) ranged from 73.9 to 89.0 %, while molecular testing for fusion-drivers (ALK, ROS1, RET, NTRK) ranged from 12.6 % to 63.9 %. Guideline-adherent testing of EGFR, KRAS, and ALK was performed in 85.2 % of patients, with regional rates spanning from 76.0 % to 90.8 %. Demographical and clinical factors associated with guideline-adherent biomarker testing included lower age (OR = 1.05 per one year decrease; p < 0.001), female sex (OR = 1.36; p = 0.002), diagnosis of adenocarcinoma (OR = 2.48; p < 0.001), availability of histological tumor material (OR = 2.46; p < 0.001), and clinical stage of metastatic disease (p = 0.002). Other factors associated with guideline-adherent biomarker testing included diagnosis at academic center (OR = 1.87; p = 0.002) and patient’s region of residence (p < 0∙001). ConclusionOptimization of the chain-of-care of predictive biomarker testing in patients with NSCLC in the Netherlands is needed to provide adequate care for these patients.