Objectives/Goals: To date, no study has jointly profiled RNA isoforms and modifications associated with Alzheimer’s disease neuropathological changes (ADNC) in a human cohort. Therefore, the overall objective of this study is to identify and characterize RNA isoforms and modifications that associate with ADNC by leveraging long-read direct RNA sequencing. Methods/Study Population: This translational study will analyze human prefrontal cortex tissue from a well-defined cohort of 12 individuals. The cohort consists of six individuals with severe Alzheimer’s disease neuropathological changes and six age- and sex-matched non-demented controls. Leveraging long-read direct RNA sequencing, which uniquely quantifies both RNA isoform and modifications simultaneously, can provide novel insights into how severe Alzheimer’s disease neuropathological changes drive changes in RNA isoform expression patterns and alter RNA modifications contributing to isoform dysregulation. Results/Anticipated Results: My preliminary data from human prefrontal cortex (n=2) revealed pseudouridine sites, 1 of ~170 known RNA modifications, in Alzheimer’s disease-risk genes (APP, CLU, SORT1, and BIN1), supporting the need to map RNA modifications across Alzheimer’s disease-relevant isoforms. My central hypothesis is that severe Alzheimer’s disease neuropathological changes (ADNCs) drive transcriptomic and epitranscriptomic dysregulation by altering RNA isoforms. Thus, my two working hypotheses are 1) severe ADNC drives transcriptomic dysregulation of RNA isoforms and 2) severe ADNC drives epitranscriptomic dysregulation of RNA modifications. Discussion/Significance of Impact: This project will be the first to jointly profile RNA isoforms and modifications accounting for Alzheimer’s disease neuropathological changes, offering candidate regulatory targets for therapeutic development. Importantly, this research will equip me with an expertise in RNA biology, neurodegenerative diseases, and bioinformatics.

Objectives/Goals: A clinical decision support tool (CDS) with an individualized prescribing approach was piloted to determine impact on opioid overprescribing rates for surgical patients at discharge. As a follow-up, clinicians were interviewed to determine current challenges to opioid prescribing and required components of an effective CDS for discharge opioids. Methods/Study Population: In the EHR-based randomized control trial, patients admitted to a qualifying surgical service for 24+ hours postoperatively were randomized to the individualized CDS tool with a taper based on inpatient opioid use or the standard discharge order set. Performance of the CDS tool was assessed by rate of opioid mismatch, that is, the difference between median inpatient and daily discharge dose in morphine milligram equivalents (MMEs). Pilot results were further investigated using a sequential explanatory design. Semi-structured interviews were held with discharging clinicians. A purposive sampling strategy with snowball sampling was used to identify providers. Data saturation was used to determine sample sufficiency. Interview transcripts were coded and thematically analyzed, with 40% group coded. Results/Anticipated Results: Among 61 randomized patients, median discharge MME (30 [IQR 22.5, 45] remained four times the median inpatient MME (7.5 [IQR 0, 30]) after the pilot period. Nine nurse practitioners and six general surgery residents were interviewed (n=15). Perceived challenges to opioid prescribing centered on three themes: limited protocolization for standard cases; lack of workflow efficiency; and systemic barriers (patient insurance and pharmacy). Clinicians identified at least one useful component of the tool for their current practice. Crucial CDS tool components identified by prescribing clinicians included seamless integration with the electronic health record, adequate clinician training, effective communication of treatment plan with the patient, and clinician buy-in to overcome current prescribing culture. Discussion/Significance of Impact: Multiple patient- and system-related factors impede standardization of opioid prescribing for postoperative patients. CDS tools in this area are a well-received idea, and the tool is now being redesigned to promote effective integration into clinician workflow with multiple stages of intervention, prior to resuming the randomized control trial.

Objectives/Goals: A key barrier to clinical trial (CT) access is knowing which trials are enrolling. Frontiers has developed a Trial Finder mobile app that displays streamlined, up-to-date trial availability that can be integrated to other institutions. We’re gathering usability feedback and exploring interest from other institutions in adopting the tool. Methods/Study Population: The Frontiers Trial Finder app is a unique tool that extracts trial data and status directly from a Clinical Trial Management System (CTMS). It includes helpful links to ClinicalTrials.gov and provides ways to connect with the study team. The app updates automatically every night, storing the extracted data in a separate stream that can be accessed through the app or online. Its interface is custom-designed in-house and fully customizable, allowing the integration of additional resources such as FAQs, links to partner affiliates, and more. During onboarding, we identified two key efforts: first, organizing the trial directory by defining parent groups and subgroups; and second, encouraging research personnel to keep CTMS trial information current, as it now serves both operational and marketing purposes. Results/Anticipated Results: To organize the trial directory, we used the institution’s divisions as primary categories and disease types as subcategories. Since the CTMS previously didn’t capture ”disease type,” we added this field to the CT activation process. For studies already underway, we generated a report of currently enrolling trials. CTMS and CTSA staff collaborated to manually assign a disease type to each study. Going forward, adding a disease type to the directory will be a standard part of CT activation. Another challenge was that study teams were not accustomed to updating the CTMS when a study remained active but was no longer enrolling – largely because the CTMS had not been used for marketing. To support this shift, we launched a communication campaign to ensure all study teams understood and adopted the new expectations. Discussion/Significance of Impact: Depending on the state of an institution’s CTMS, onboarding this tool may require institutional messaging effort. However, the long-term advantages of automated data exchange significantly improve the efficiency of connecting volunteers to trials. Clinicians and volunteers alike can benefit from reliable, up-to-date trial availability.

Objectives/Goals: There is growing interest in using physiological data produced from medical devices such as 12-lead electrocardiographs within the Observational Health Data Sciences and Informatic community. We have introduced new waveform extensions to support this need and illustrate its implementation using 12-lead ECG. Methods/Study Population: We identified 8,822 12-lead ECGs from Tufts Medical Center between January 1, 2018 and April 31, 2024. 6,702 of these ECGs were linked to patients in our Observational Medical Outcomes Partnership (OMOP) Research Data Warehouse (TRDW). The cohort of patients with linked ECGs were extracted into a study OMOP database in which the new Waveform Extension (WFExt) were implemented. A data extract/transform/load import process was developed that complied with the WFExt specifications and formed a new custom vocabulary and concepts to support waveform data that is not presently supported by standard vocabularies such as SNOMED and LOINC. Results/Anticipated Results: Using the OMOP Waveform Extensions allowed us to capture the full range of high-resolution physiological data contained in 12-lead electrocardiograms. The OMOP common data model is designed to capture low-resolution data contained in electronic health record systems. In emergency medicine, critical care, and neonatal care, medical devices play an important role in driving treatment decisions. Research therefore needs access to both clinical and medical device data in developing clinical decision support tools. In this project, we demonstrate the ability to connect clinical data such as outcomes in an OMOP database with device measurements, computerized interpretations, metadata, and provide linkage to the source device record to support signal processing and new algorithm development using the WFExt. Discussion/Significance of Impact: The OMOP WFExt mirrors work to support imaging data, and both extend the utility and research capacities among the many Clinical Translational Science Institutes that have already implemented an OMOP-based research warehouse. The WFExt is a standards-based and open science approach to support research and backed by a worldwide OMOP community.

Objectives/Goals: Standardize, centralize, and automate the collection of administrative data at the Northwestern Clinical and Translational Sciences (NUCATS) Institute CRU at Northwestern Memorial Hospital. Facilitate instantaneous data reporting and visualization for immediate insights across eight service lines. Methods/Study Population: The CRU offers nursing and laboratory services for clinical research studies. We have developed a comprehensive flowchart that outlines CRU workflows and data collection for all its service lines, from study enrollment to billing processes. Our goal is to pinpoint areas for improvement, focusing on minimizing manual labor and enhancing automation. To streamline operations, we implemented direct data entry into the EPIC system for all visits. Additionally, we established efficient pipelines for direct data extraction and reporting through Azure Databricks, leading to the creation of near real-time dashboards that display key metrics, which are deployed using Power BI. Results/Anticipated Results: The integration of a centralized data system has significantly enhanced workflow efficiency within the CRU. Automated reports and interactive visualizations provide near real-time insights, which greatly improve decision-making and operational transparency. By eliminating unnecessary manual data entries and establishing a centralized data repository, we achieved greater accuracy and granularity in our data management. This streamlined reporting workflows for over 2,000 visits from 250 studies annually. Overall, these enhancements enabled the CRU to operate more effectively, facilitating better resource management and improving service delivery for clinical research. Discussion/Significance of Impact: Streamlining and automating data management within the NUCATS CRU has reduced manual labor and helps ensure timely and accurate reporting. This allows us to better analyze current data and seasonal trends, which facilitates better decision-making and resource allocation.

Objectives/Goals: The goal of this study was to determine if there are the peripheral immune system changes in persons undergoing treatment with lecanemab, and whether those changes are associated with the incidence of amyloid-related imaging abnormalities (ARIA), in order to identify strategies for improving the safety profile of anti-amyloid antibodies. Methods/Study Population: Three pairs of age, sex, APOE genotype, and infusion-matched individuals were selected from a study population at Norton Neuroscience Institute Memory Center (NNI-MC), based on the development of ARIA. Peripheral blood mononuclear cells isolated from these case–control subjects were collected and cryopreserved. Thawed cells were run through a deep sequencing pipeline, including single-cell RNAseq, CITEseq TCR clonality (V(D)Jseq), metabolomics, and lipidomics. Multi-omic data was integrated and analyzed using R (scRNAseq, CITEseq, and V(D)Jseq) and MetaboAnalyst (Metabolomics and Lipidomics). Deidentified clinical data was obtained to correlate to sequencing findings, including cognitive exam scores, MRI scans, and relevant lab values. Results/Anticipated Results: We noted an increase in lymphocytes in ARIA-positive subjects, with expansion in CD8+ T-cells – particularly CD45-RA expressing T-effector memory (TEMRA) subsets. These TEMRA cells expressed large clonal expansions in cells programmed for antigen response, cytotoxicity, and metabolic shifts. We found that in ARIA+ subjects, there was a pro-inflammatory glycolytic shift, accompanied by CD8-driven lipidomic changes. To determine potential effects of TEMRAs on the CNS, we mapped our data onto previously published ARIA+ patient brain transcriptomes (van Olst et al., 2025). TEMRA CD8s from our dataset showed increases in in silico communication with the vascular endothelium, indicating priming of these cells for cerebrovascular engagement. Discussion/Significance of Impact: Together, this first-of-its-kind data in human subjects receiving lecanemab infusions may point to peripheral immunity as a driving factor – and potential biomarker and therapeutic target for – ARIA pathogenesis, allowing for increased safety of these drugs in the highest risk AD populations.

Objectives/Goals: This study aims to evaluate and establish a reliable strategy for identifying at-risk diabetic foot ulcer cases in hospitalized patients utilizing machine learning, enhancing the accuracy of care delivery and facilitating future research on DFU outcomes. “At-risk” is defined as risk for further serious complications, most notably, amputation. Methods/Study Population: Data analyzed for this work started with a previous computable phenotyping study that compared multiple computable phenotype strategies. The dataset created for subsequent ML analysis includes a random sample of 351 hospital admissions during the years 2022–2024. These hospitalizations were manually curated by SME’s, scored and annotated for their inclusion as hospitalized DFU patients, resulting in a “gold standard” dataset for “at-risk” patients. We analyzed six clinical variables hypothesized to be predictive of “at-risk” status (imaging [XRAY, MRI], specialty consultations, and clinical findings [infection, ulcer]). All possible combinations (n=63) were evaluated using precision, recall, F1-score, and accuracy metrics. The dataset contained 42 positive outcomes (12.0% prevalence). Results/Anticipated Results: The findings show, while INFECTION, comprised of two SNOMED codes, is the strongest single predictor, adding ULCER and XRAY provides optimal balance. This combination achieved an F1-score of 72.1% with 93.2% accuracy, 70.5% precision, and 73.8% recall, correctly identifying 31 of 42 positive cases. Adding additional variables beyond this three-feature model yielded diminishing returns with no meaningful improvement in predictive performance. Discussion/Significance of Impact: This research will then be used to proactively identify potential DFU patients for outreach and intervention. A parsimonious three-variable model (XRAY + infection + ulcer) provides optimal predictive accuracy for identifying patients, offering a clinically practical decision support tool with strong discriminative performance.

Objectives/Goals: While there are many efforts to teach artificial intelligence and data science (AI & DS) methods, the importance of problem formulation in translational research (TR) has received limited attention. The goal of this presentation is to revisit this challenging pedagogical topic and formalize approaches in inculcating its design principles. Methods/Study Population: AI & DS didactic courses cover problem formulation implicitly in their various course modules. Also, problem formulation is a prominent education theme of quantitative fields such as mathematics and physics. Also, data scientists working with TR usually spend considerable time in the data design aspects to quantitively solve the health problem at hand. We used three approaches to building a pedagogical framework for DS problem formulation in TR: (1) review existing literature on problem formulation in TR, applied mathematics and physics; (2) obtain structured responses around teaching problem formulation from AI & DS instructors; and (3) add exemplar DS in TR studies we actively supported. Results/Anticipated Results: Conceptualizing complex problems in a quantifiable manner can often be challenging. The typical processes involved in formulation included starting with a real-world and relatable context, iteratively defining the problem and chunking it into portions related to individual entities, abstracting each chunk to a quantifiable and working statement, identifying appropriate data resources and their constraints, and transforming data into reasonable representations. Strategies to teach problem formulation we identified through our analysis include active learning (problem-based learning and question formulation techniques), stimulating metacognition (problem defining, contextualizing, experimentation, and creativity), and group learning with think aloud and constructive feedback. Discussion/Significance of Impact: Instilling problem formulation skills, acquired over years of experience, in condensed timelines, requires formal pedagogical approaches. Ability to formulate a problem in multiple ways increases the likelihood of finding a tractable solution and can be covered in TR and DS courses and enable increased use of real-world data resources in research.

Objectives/Goals: To identify lipid signatures associated with adverse pregnancy outcomes (APOs) and postpartum weight retention (PPWR) in women entering pregnancy with overweight or obesity and to explore potential epigenetic mechanisms through ongoing DNA methylation (DNAm) analysis. Methods/Study Population: The GROWell: Goals for Reaching Optimal Wellness study was a blinded, randomized clinical trial to improve dietary support during pregnancy and postpartum. Data were collected via online questionnaires, electronic health records (EHRs), and participant-collected dried blood spots (DBS) at 10–16 weeks, 36–38 weeks, and 3 months postpartum. We analyzed DBS data from 49 participants using three lipidomic analytical techniques: weighted lipid correlation network analysis, differential lipid analysis, and partial least squares discriminant analysis. Lipid profiles were characterized and then tested for associations with APOs and PPWR at all three time points. DNAm analysis from the samples is ongoing to determine the epigenetic mechanisms associated with lipid dysregulation and pregnancy outcomes. Results/Anticipated Results: The lipid network analysis revealed positive correlations between specific lipid types and APOs, particularly at 10–16 weeks and postpartum. Notably, triglycerides and phosphatidylcholine were associated with an increased risk of APOs. Additionally, in the third trimester, triglycerides were positively correlated with weight-related variables. Postpartum analysis also showed a positive correlation between elevated triglyceride levels and a history of APOs. The network analysis also revealed positive associations between triglycerides and PPWR, which were most evident at the 36–38 week and postpartum time points. Ongoing methylation analysis is expected to identify genomic regions associated with pregnancy outcomes. Discussion/Significance of Impact: These findings underscore the significant role of lipids in pregnancy outcomes for individuals with overweight or obesity. Identifying lipidomic and methylome biomarkers of risk for APOs and PPWR may help with early detection and intervention, improving maternal and fetal health.

Objectives/Goals: We aimed to explore informaticist experiences, lessons learned, and recommendations arising from their experiences during the Department of Veterans Affairs (VA) transition from its homegrown electronic health record (EHR) to a commercially developed system. Methods/Study Population: We conducted semi-structured Microsoft Teams interviews with 16 informaticists from five VA facilities and a regional office, representing roles that included nurse informaticists, program analysts, change management managers as well as chief roles in health informatics, medical informatics, nursing informatics, and quality, safety, and value. Interviews explored role transitions, staffing impacts, and lessons learned during the Oracle Cerner EHR implementation. Transcripts underwent matrix-based rapid qualitative analysis, with ongoing thematic analysis guided by Nicholson’s role transition framework. Results/Anticipated Results: Informaticists underwent major role transitions during EHR implementation. Pre-transition, their roles were largely technical, centering on updating and maintaining the EHR and providing as-needed user support. During the transition, informaticists became advocates, change leads, and trainers for frontline clinicians, performing relational and educational rather than technical work. Post-transition, roles shifted once again, encompassing frontline support, workflow education, and help ticket triage. With little external guidance, informaticists engaged in adaptive role innovation and identity redefinition throughout the transition, often experiencing role overload. Discussion/Significance of Impact: Findings highlight the need to better prepare and support workforce members whose evolving roles are central to others’ adaptation. Implementation planning must account for the education and training needs of these intermediary roles, not just their functional contributions.

Objectives/Goals: TMEM106B forms age-dependent amyloid fibrils that are elevated in dementia patients and correlated with increased TDP-43 pathology. Cerebrovascular disease (CVD), a major dementia co-pathology, may contribute to TDP-43 dysfunction. Understanding the relationships among CVD, TMEM106B, and TDP-43 is crucial for developing targeted therapies. Methods/Study Population: We compared autopsy brains across TDP-43 pathologies and CVD to investigate potential differences in TMEM106B levels. Autopsy brain samples (N=58) were selected from the Biggs Brain Bank to include frontotemporal lobar degeneration with TDP pathology (FTLD-TDP), limbic-predominant age-related TDP-43 encephalopathy (LATE), TDP-43 negative Alzheimer’s diseases neuropathologic change (ADNC), and controls. Immunofluorescence staining using anti-TMEM106B (239-250aa) was performed on formalin-fixed posterior hippocampal sections to assess whether CVD burden had an impact on TMEM staining patterns. Quantitative digital pathology analyses were conducted on whole slide images (WSI) to quantify TMEM106B fibril accumulation. Results/Anticipated Results: Age was the strongest predictor of TMEM106B fibril accumulation (p < 0.001), and females showed higher fibril burden than males (p = 0.035). TDP-43 pathology was associated with increased TMEM106B accumulation (p = 0.03). Greater co-existing cardiovascular disease (CVD) pathology burden showed a trend toward reduced TMEM106B fibril accumulation, suggesting an attenuating, and possibly dose-dependent effect of CVD pathology on the observed TDP-43 and TMEM106B association (p = 0.058). Across ADNC (Braak tangle stage, Thal amyloid plaque phase, CERAD neuritic plaque density score), increasing CVD pathology burden did not modify TMEM106B fibril levels with increasing Braak stage (p = 0.64), Thal phase (p = 0.39), or CERAD score (p = 0.56), and TDP-43 did not increase TMEM106B in the presence of these AD lesions. Discussion/Significance of Impact: Age, sex, and TDP-43 are determinants of TMEM106B fibril. Increasing CVD pathology appears to blunt the TDP-43 and TMEM106B association, possibly due to hypoxia-driven lysosomal dysfunction. Additional co-pathologies do not appear to drive TMEM106B fibril accumulation. Future work will assess if lower TMEM106B may reflect compensatory change.

Objectives/Goals: The main objective of this study was to determine if dietary supplementation with the iron-binding protein lactoferrin is able to reverse age-related brain proteomic changes in a heterogenous mouse model of aging, with implications for the modulation of age-related brain disease. Methods/Study Population: UM-HET3 mice of both sexes were aged to 6 months on control chow (Young Control, N=11) or 24 months (Old Control, N=12) on control chow or a diet supplemented with 12,000 ppm of bovine lactoferrin for 9.5 weeks before sacrifice (Old Lactoferrin, N=13). At sacrifice, brain cortex was collected. Cortex homogenate was analyzed through SomaLogic for 1,500 specific proteins relevant to brain aging. Differential expression analysis was run to analyze the age and treatment effect separately, with an adjusted p-value cutoff of 0.05 to reveal effects of age (young control vs old control) and treatment (old control vs old lactoferrin) separately. Gene Ontology Biological Processes (GO-BP) Enriched Pathway analysis was run to identify the biological pathways altered by lactoferrin supplementation. Results/Anticipated Results: Differential expression analysis adjusted by sex identified 71 differentially expressed proteins in the age comparison and 99 differentially expressed proteins in the treatment comparison. Thirteen proteins were differentially expressed in both lists, which we classify as lactoferrin rescued proteins, those that were elevated in old untreated mice, but significantly lowered by treatment. These proteins include SYDC, CLIC5, NF2L2, STAT3, Peroxiredoxin.3, UFM1, Profilin.1, NUCB1, Laminin.2, Ezrin, PPT1, IDH, and PUR9. Pathway analysis revealed significant alteration in 30 pathways relevant to synapse, neuron, and cellular structure and function relevant to age-related neurodegenerative disease. Discussion/Significance of Impact: Lactoferrin may be a viable therapy for the modulation of age-related brain changes. Next, we will evaluate endogenous levels of lactoferrin as a biomarker for Alzheimer’s disease, moving toward a clinical trial to evaluate the utility of lactoferrin to serve as both a biomarker and intervention against age-related neurodegenerative disease.

Objectives/Goals: The objective of this research is to investigate the association between suboptimal self-reported antihypertensive medication adherence, increased blood pressure variability (BPV), and cognitive decline independent of average blood pressure (BP) levels in a post hoc analysis of the Systolic Blood Pressure Intervention Trial (SPRINT). Methods/Study Population: 7371 participants with medication adherence, blood pressure, and mild cognitive impairment follow-up data were included for analysis (36% female, mean age 68.3±9.22). BPV was calculated as the standard deviation of available systolic blood pressure measurements. Medication adherence was quantified using the “Percent Meds” variable, a self-report measure that quantifies the proportion of the last 10 days that a participant remembered to take their medications correctly. Suboptimal medication adherence was defined as 90% adherence or less. Survival analysis using Cox proportional hazards regression was used to assess the mediating effects of BPV and BP on the association between suboptimal medication adherence and increased MCI risk. Results/Anticipated Results: Proportional hazards regression showed suboptimal medication adherence to be associated with increased MCI risk (HR=1.62, P=.00004). Both systolic BP (HR: 1.14, P=.002) and systolic BPV (HR: 1.11, P=.004) displayed independent adverse effects on MCI risk. In parallel mediation, BPV (indirect effect log hazard=.01, 95%CI=.002-.02) and BP (indirect effect log hazard=.009, 95%CI=.001-.02) acted as partial mediators of the association between suboptimal medication adherence and increased MCI risk. All models were adjusted for age, sex, resting heart rate, and history of cardiovascular and chronic kidney disease. Discussion/Significance of Impact: Average BP and BPV are independently associated with increased MCI risk and represent distinct risk factors for MCI conversion. Both average pressure and variability in pressure link suboptimal medication adherence to increased MCI risk, suggesting that suboptimal medication adherence is a modifiable risk factor for MCI conversion.

Objectives/Goals: Clinical studies with triple-negative breast cancer (TNBC) are encouraging for immunotherapy combined with chemotherapy. Additional clinical advances may be accelerated by leveraging preclinical TNBC models including syngeneic mammary tumor cell lines. Here, we report two lines exhibit differential immunotherapy responses in vivo. Methods/Study Population: Spontaneous mammary tumors from C57BL/6J MMTV-Cre Trp53fl/+ animals were passaged serially in cell culture and in vivo in the mammary fat pad of fully wildtype animals. The resulting lines, MM001i and MM008i, lost Trp53 and formed 1000 mm3 [https://www.biorxiv.org/content/10.1101/2025.09.18.677171v1#ref-3] tumors in the mammary fat pad within 21–28 days. Tumors from each group were analyzed for their response to αPd-1 and αCtla-4 immunotherapy, infiltrating immune populations, and transcriptional profiles. Results/Anticipated Results: Despite originating from the same genetic background, these lines exhibit differential responses to immunotherapy. For αPd-1 therapy, MM001i is poorly responsive and MM008i is strongly responsive with near-complete tumor regression. In comparison, both MM001i and MM008i respond rapidly to αCtla-4 therapy. Both models express unique tumor antigens as evidenced by immunity to subsequent engraftments. Primary MM008i tumors exhibit greater T-cell infiltration, and CD8-positive T lymphocytes are required for αPd-1 responses. Cancer epithelial cells within MM008i tumors demonstrate a greater abundance of Pd-l1 expressing cells, correlating with observations in human TNBC where PD-L1 expression is a biomarker for αPD-1 response. Discussion/Significance of Impact: These TNBC models are promising for further mechanistic studies and testing future single and combinatorial therapies.

Objectives/Goals: KSHV is well documented to infect and establish latency in endothelial cells; however, infected cell types in the Kaposi Sarcoma (KS) tumor microenvironment are not well defined. Here, we present the first application of the Lunaphore COMETTM multiplexed immunofluorescence platform for spatial proteomic analyses of KS lesions. Methods/Study Population: Custom antibody panels were validated and optimized for characterization of the TME components, as compared to marginal structures. The Ab panels supported the detection, quantification, and comparison of >15 cell types in regions of interest in the KS tissues in a single automated run. For data analysis, we developed a rule-based decision tree for accurate cell classification and used the resulting cell populations to perform a nearest-neighbor analysis. We assessed spatial proximity among cell types located within, adjacent to, or distal from the areas with a large number of cells expressing KSHV LANA. Results/Anticipated Results: The highest KSHV load was detected in CD34+ cells with or without CD31, consistent with an endothelial progenitor phenotype (EPCs). We also defined 12 cell types with KSHV-LANA detected in <1% of each cell type. Among them, tumor-infiltrating CTLs were proximal to or overlapping with infected EPCs, suggesting they have cytolytic activities. In addition, we observed areas with T-cells detected at high frequency near normal vascular endothelial cells (VECs), indicating immune cell trafficking toward the lesion in these marginal areas. While supporting previous findings that T-cells are not infected by KSHV, our data also revealed areas having high viral Ag load and abnormal vasculature, versus surrounding areas with much lower Ag load, normal vasculature, and robust immune infiltration. Discussion/Significance of Impact: Our approach not only demonstrated its ability to characterize the KS TME at a previously unapproachable level of complexity but has also revealed gaps in lineage assignments in substantial proportions of the tumor, which will be subjected to further characterization.

Objectives/Goals: The goal of this project is to integrate human and animal data to explore the relationship between social participation and cognitive functioning after traumatic brain injury (TBI). Further, we aim to determine whether reduced social participation is not only a consequence of TBI but a potential driver of post-injury cognitive deficits. Methods/Study Population: This study uses a translational framework to examine how social participation influences cognitive outcomes after traumatic brain injury (TBI) using complementary human and animal models. Longitudinal TBIMS data will be analyzed to assess associations between social participation (PART-O: productivity, out and about, social relations) and cognition (BTACT: memory, fluency, reasoning, processing speed) using linear regression (potentially a Random Intercept Cross-Lagged Panel Model to capture bidirectional effects over time). Further, mice with mild, moderate, or severe TBI will be housed in isolated, standard, or enriched social environments. Post-injury cognitive performance (NOR, Y-maze) and neuronal proliferation and survival in the prefrontal cortex and hippocampus will be evaluated. Results/Anticipated Results: We anticipate that both humans and animals with greater social participation will demonstrate superior cognitive performance over time. In the animal model, we further expect that mice housed in social or enriched environments will exhibit reduced cell death and enhanced neuronal survival in the prefrontal cortex and hippocampus compared to socially isolated counterparts. Discussion/Significance of Impact: Together, these findings may inform future rehabilitation strategies that integrate social participation as a core component of cognitive recovery supporting more personalized and multidisciplinary approaches to chronic TBI care.

Objectives/Goals: Our objectives were to determine enteric microbiome characteristics in fluoroquinolone-resistant Enterobacterales (FQRE)-colonized and non-FQRE-colonized hematopoietic cell transplant (HCT recipients, and to understand the enteric microbiome-specific factors associated with the development of FQRE BSI in FQRE-colonized HCT recipients. Methods/Study Population: This was a prospective cohort study of patients undergoing HCT at Weill Cornell Medicine from November 2016 to August 2019. All participants received levofloxacin prophylaxis during post-HCT neutropenia. Stool samples were collected prior to levofloxacin prophylaxis initiation and during the week after HCT. Our overall study population consisted of 26 FQRE-colonized HCT recipients and 69 non-FQRE-colonized HCT recipients. Samples underwent selective quantitative culture for FQRE, as well as 16S rRNA sequencing. Raw sequence data were processed and aligned to a custom SILVA database. Analysis of 16S rRNA sequencing was conducted in R using phyloseq. Results/Anticipated Results: We saw significant differences in microbiome composition in comparing pre-HCT samples from FQRE-colonized HCT recipients to non-FQRE-colonized HCT recipients, including significantly higher relative abundances of Bacteroidales and Enterobacterales in colonized recipients. MaAslin2 demonstrated significantly differential abundance of Bacteroides and Escherichia-Shigella ASVs. We observed a significant decrease in Enterobacterales relative abundance in non-FQRE-colonized recipients after HCT, but no such change in FQRE-colonized recipients. FQRE-colonized recipients who developed FQRE BSI had a lower relative abundance of Bacteroidales as compared to those who did not. No significant difference in FQRE colonization density as measured by quantitative FQRE stool culture was observed. Discussion/Significance of Impact: FQRE-colonized HCT recipients represent a unique subgroup of HCT recipients. Colonized HCT recipients with FQRE BSI had lower relative abundance of Bacteroidales compared to those without BSI. Further work is needed to determine the mechanism behind this association and to further characterize the microbiome in this subgroup of HCT recipients.

Objectives/Goals: Improvements in acute stroke treatment, including critical care management, have increased survival rates post-stroke. However, many stroke survivors have significant neurological deficits, and stroke remains a leading cause of long-term disability. We aim to study the chronic progression and long-term sequelae of ischemic stroke pathology. Methods/Study Population: We examined long-term neurobehavioral recovery and progressive neuropathology of young (3 months) and middle-aged (14 months) adult male and female C57Bl/6 mice at 1, 3, and 6 months after a 60-minute transient middle cerebral artery occlusion (MCAO) or sham surgery, as well as histopathological analysis of post-mortem brain samples from patients with a chronic infarct. Behavioral examination included the tail suspension test, the novel object recognition test (NORT), and the fear conditioning test (FCT). Molecular and pathological analyses included bulk transcriptomic screening, immunohistochemistry (IHC), flow cytometry, and atrophy analyses via MRI and microCT scanning. Results/Anticipated Results: In mice, depression-like behavior persisted for 6 months post-stroke (p<0.0001), while cognitive function progressively worsened from 3 to 6 months post-stroke as seen by NORT (6 months, p=0.0063). Memory retention deficits at 6 months were seen in the FCT (p=0.0084). Significant brain atrophy was evident by MicroCT and MRI at 2 and 6 months post-stroke. IHC showed significant demyelination (mouse p=0.0361; human p=0.0001) and increased microglial (p=0.0059) and astrocyte activation. TUNEL co-labeling with a neuronal marker showed neuronal apoptosis in human samples of chronic stroke. Disease-associated microglia (DAM) displayed senescent-like phenotypes. Nanostring showed neuroinflammatory pathway upregulation up to 6 months PS. Human stroke brains had more amyloid plaque (p=0.06). Discussion/Significance of Impact: Cellular senescence and chronic neurodegenerative signatures likely result in poor cognitive function that may progressively worsen after stroke. Studying this neurodegenerative etiology in the long-term post-stroke may offer viable targets for delayed stroke treatment, which would fill a severely unmet need for stroke survivors.

Objectives/Goals: Tibial bone stress injuries are running injuries that have a high risk of recurrence. Tibial load contributes to the development of bone stress injuries; however, this variable is clinically inaccessible. The objective of this study is to establish a relationship between tibial load during running and clinical tests to support safe return to running. Methods/Study Population: Thirty-two healthy recreational runners (13 F, 19 M, age 23.6 y, height 1.77 m, mass 73.71 kg, weekly volume 16.91 miles) ran at a self-selected pace on an instrumented treadmill for 5 minutes, while a 7-camera motion capture system collected whole body kinematics. Inverse dynamics were used to calculate ankle joint reaction forces and moments which were then used in a musculoskeletal model to calculate tibial axial force. Single leg heel-raise endurance, hand-held dynamometry ankle strength, and Y-balance tests were also assessed. Linear mixed models will be used to determine relationships between tibial force during running and strength and balance measures. Results/Anticipated Results: Data analysis is ongoing. Based on preliminary data, it appears that peak tibial bone force during running may demonstrate a relationship with lower extremity strength and balance. Clustering participants into “high” and “low” tibial force production during running may help identify specific ranges of strength and balance assessment outcomes that are appropriate for clinical goals. Discussion/Significance of Impact: The results of this study will attempt to establish relationships between tibial bone loads during running and strength and balance assessments that are clinically feasible. Results have potential to directly impact clinical practice by providing objective measures that may be used as surrogates for tibial bone force in a clinical setting.

Objectives/Goals: Previous studies of nanoparticles in a penetrating spinal cord injury demonstrated their ability to modulate immune responses to reduce secondary injury. This study investigated nanoparticles in contusion injury models, which is more common clinically, to quantify their efficacy as a function of injury severity. Methods/Study Population: Female C57BlJ/6 mice were anesthetized, and a T9/T10 laminectomy was performed to expose the spinal cord. A stereotactic setup with an Infinite Horizon impactor was used to position the spinal cord to receive a mild (30 kdyne), moderate (50 kdyne), or severe (70 kdyne) contusive injury. Within two hours of injury, poly(lactide-co-glycolide) nanoparticles, formed via a single emulsion with sonication protocol, or phosphate-buffered saline (PBS) vehicle were injected intravenously via tail vein injection. Nanoparticles (NPs) or PBS were administered daily for 7 days. Function was evaluated using the Basso Mouse Scale (BMS) test for hind limb function starting 3 days post-injury and repeated weekly thereafter, as well as a horizontal ladder beam test once they had recovered to at least occasional plantar stepping. Results/Anticipated Results: In the moderate injury group, NP-treated mice had significantly higher BMS scores starting 6 weeks post-injury (p=0.009), with the NP-treated mice performing better than control to the end of the study. In the mild injury group, NP-treated mice scored significantly better on BMS starting at 3 days post-injury (p=0.005). This difference persisted through week 4, and then the vehicle-treated mice maintained a slightly lower score through the end of the study. In the severe injury group, no significant difference on BMS scores was observed between groups and animals did not recover enough function to be evaluated via ladder beam. Scoring of the ladder beam test for moderate and mild injury studies is underway. Animals treated with NPs are anticipated to have a significantly higher number of successful steps compared to control. Discussion/Significance of Impact: Spinal cord injury consists of a primary injury and secondary injury caused by inflammation at the injury site. Nanoparticles present an opportunity to reduce secondary injury to improve functional outcomes of individuals with SCI. This study of NPs in a graded severity contusion model further illustrates the feasibility and utility of NPs in SCI.