Objectives/Goals: To develop a comparative framework evaluating FDA regulatory pathways for ADCs in HER2-low and TNBC, mapping biomarker translation, approval timelines, and labeling language, and conducting qualitative and quantitative analyses to identify determinants of global therapeutic access. Methods/Study Population: This study conducted a targeted review of scientific, regulatory, and epidemiological databases to examine the US FDA’s regulatory pathways for ADCs in HER2-low and TNBC populations. Literature from PubMed and Scopus was analyzed to trace the translation of biomarkers from gene targets to therapeutic moieties. FDA databases were reviewed to map approval timelines, labeling language, and evidentiary standards shaping ADC authorization. ClinicalTrials.gov, SEER, and GLOBOCAN data were integrated to contextualize trial geography, disease burden, and access disparities. Comparative synthesis connected molecular biomarker validation to regulatory decision-making, enabling the identification of translational and policy determinants influencing time to patient access within the FDA landscape. Results/Anticipated Results: Preliminary findings indicate that advances in biomarker characterization have accelerated the development and regulatory progression of ADCs in breast cancer. HER2-low and Trop-2 expression have enabled the evolution of Enhertu and Trodelvy from histopathological to molecularly guided indications. FDA approvals demonstrated reduced timelines and more detailed biomarker-specific labeling, supported by surrogate endpoints such as progression-free survival and overall response rate. Analysis of approval data and clinical trial mapping highlights a temporal correlation between biomarker discovery, therapeutic innovation, and expedited regulatory authorization for targeted ADC therapies, serving as a pathway toward improved global access. Discussion/Significance of Impact: This study provides a structured analysis of FDA regulation of ADCs in breast cancer, linking biomarker translation to therapeutic approval. Findings underscore the need for evidence-based and equity-driven frameworks to accelerate global access and support ADC transition to first-line therapy.

Objectives/Goals: Given previously demonstrated relationships between decline in proteinuria to thresholds below 0.3-1.5 g/g and kidney failure, we sought to assess if this relationship was maintained when proteinuria change was measured on a continuous scale and to compare the difference in association between changes measured at 12- and 24-months post-index. Methods/Study Population: PARASOL is a consortium of academic scientists, patient advocacy organizations, and industry partners who came together to identify candidate surrogate endpoints for clinical trials in focal segmental glomerulosclerosis (FSGS). Data were harmonized among international cohorts and registries of adults and children with clinically adjudicated FSGS without other causes. Index was identified as the first observed urine protein-creatinine ratio (UPCR)≥1.5 g/g with associated eGFR≥30 ml/min/1.73m2. Change in UPCR from index to 12 and 24 months was evaluated using the difference in log(UPCR) from baseline and follow-up and transformed to percent change. Time-updated landmark Cox regression models were fitted, and the interaction between UPCR change and the 24-month timepoint indicator was assessed. Results/Anticipated Results: 1366 participants with at least 12 months of follow-up post-index and UPCR measurements 6-12 and/or 18-24 post-index were included. Mean age was 27 years (SD=21); 46% were pediatric at index. Median (IQR) eGFR and UPCR at index were 83 (57, 110) ml/min/1.73m2 and 3.7 (2.3, 7.0) g/g, respectively. A 50% decrease in UPCR from index was associated with a 28% and 32% reduction in the probability of kidney failure at 12 and 24 months, respectively, with no difference detected based on timing of UPCR (HR [95% CI] = 0.72 [0.65-0.80] at 12m, 0.68 [0.63-0.74] at 24m, interaction p-value = 0.46). In subgroups based on age, UPCR, and eGFR at index, results were consistent (HR range 0.66-0.72 at 12m and 0.58-0.78 at 24m), with only the eGFR≥60 ml/min/a.73m2 subgroup showing patterns of larger effects at 24m (HR = 0.62 vs. 0.72, p=0.08). Discussion/Significance of Impact: Continuous measures of proteinuria demonstrated consistent associations with kidney failure at 12- and 24-months post-index and across patient subgroups. If validated, these relationships support potential variation in timing of endpoint evaluation in clinical trials where there is mechanistic justification for earlier assessment.

Objectives/Goals: The CHIRP program aims to promote community-driven research on local health priorities, strengthen community–academic partnerships, and build capacity for sustainable and fair research through education, training, and mentoring, while supporting CTSA’s mission to advance translational science. Methods/Study Population: CHIRP, funded by NCATS/CCTS, supports community/academic partnered teams to conduct impactful research in community or clinical settings. Selected projects have undergone a competitive application process: eligibility screening by the CHIRP team, a panel review by both community and academic partners, and scoring using a 9-point NIH system. After final approval by CCTS leadership, selected teams consisting of at least one community partner and one academic partner participate in a mandatory 2-day Proposal Development Retreat where they receive training and resources to further refine their proposal. Upon IRB approval, teams receive funding of up to $20,000 and mentorship from experts during the grant period to support implementation of the proposed project. Results/Anticipated Results: Since its inception, CHIRP has funded 20 academic teams in 3 CTSA regions of Texas (Houston, Northeast Texas, and Rio Grande Valley) in partnership with 20 community organizations such as public schools, faith-based organizations, and community-based health provider organizations, providing $380,000 funding for clinical research or initiatives on health promotion and disease prevention strategies driven by community need. The program resulted in 10 peer-reviewed articles, 15 conference presentations, and multiple community forums. Currently, CHIRP is supporting its fifth cohort of fellows who completed the Proposal Development Retreat and a majority reported being “Very Satisfied” with the training. Discussion/Significance of Impact: The CHIRP model supports community–academic partnerships to drive translational research. The model has application for other sites wanting to train and support community-engaged research and strengthen the translational research pipeline.

Objectives/Goals: Randomized evaluation of interruptive EHR alerts in improving service delivery or patient health outcomes is usually carried out within the EHR system and in routine hospital operations. However, this study setting leads to increased risk of contamination, where factors outside the original design affect estimation and testing of treatment effect. Methods/Study Population: We focused on the effect of contamination on the design and analysis of individually randomized controlled trials (iRCTs) for the study of the effectiveness of EHR alerts. More specifically, we looked at post-randomization contamination due to clinician-induced correlations. To account for this kind of contamination, we extended the iRCT design to include clinicians’ random effects under the linear mixed effects model (LMEM) framework. We then proceeded to derive sample size and power formulas under different LMEM correlation structures. Using extensive simulation studies, we evaluated the derived formulas and assessed the effect of using a naive two-sample t-test that ignores post-randomization contamination. Results/Anticipated Results: We observed that the level of contamination on sample size, power, and type I error due to clinician-induced correlations was driven by three parameters: variances of the clinicians’ random intercepts and slopes, and the correlation between them. These parameters reflect the way and extent to which clinicians influence participant-level outcomes. For scenarios with common correlation structures from the literature, a test of the treatment effect based on a naive two-sample t-test led to inflated type I error. Using the naive test in the presence of contamination increased the risk of incorrectly concluding that the treatment effect is significant. We did observe valid naive tests with well controlled type I error but under extreme and unusual values of inter-cluster correlations. Discussion/Significance of Impact: Using iRCT as the baseline design for the interruptive EHR alert trials, a naive two-sample t-test may appear reasonable. However, our analytical and simulation findings indicate that a careful consideration of clinicians’ impact on patient-level outcomes in routine hospital operations is necessary for valid hypothesis tests and inference.

Objectives/Goals: To describe the early implementation of the FDA-approved Shield blood-based colorectal cancer screening test using a mobile health unit serving rural Louisiana, assessing population reach and acceptability of the test. Methods/Study Population: This is a secondary analysis of implementation data collected by the Partners in Wellness (PIW) Mobile Health Unit between February and August 2025. The mobile unit includes a designated blood-draw space where specimens were collected by a trained phlebotomist or a registered nurse following standardized procedures. Adults aged ≥45 years were eligible. De-identified demographic information and Shield test results were analyzed descriptively. All individuals with positive results were referred for colonoscopy, and those with inconclusive results were offered Cologuard stool-based testing. One participant declined the stool-based testing. Results/Anticipated Results: A total of 77 individuals completed Shield testing the mean age was 59.5 years (range 30.6–89.9). Most were female (98.7%), representing 31 rural cities and one urban city. Racial/ethnic distribution included African American (33), White (26), Hispanic/Latino (16), Pacific Islander (5), and Asian (1) participants. Test outcomes were negative (87%), positive (10%), and inconclusive (3%). All positive participants were referred for colonoscopy. Among inconclusive cases, two were offered stool-based testing and one declined. Implementation was well accepted and demonstrated operational feasibility within mobile health services. Discussion/Significance of Impact: Integrating an FDA-approved blood-based colorectal cancer screening test into a mobile health program demonstrates a scalable and innovative strategy to reach medically underserved populations. This model can transform rural screening delivery and advance cancer health equity.

Objectives/Goals: The Michigan Institute for Clinical & Health Research (MICHR) created an eight-phase adaptive Translational Science Framework to guide us in selecting and solving translational science problems. We operationalized the initial problem selection phases and assessed utility of various methods in shaping complex problems for solution design. Methods/Study Population: We identified three high-priority strategic domains where there was potential to improve the efficiency and effectiveness of research and related processes. Starting with broad problem statements, we engaged a wide array of stakeholders and end-users in a sequence of methods to reduce ambiguity, uncover root causes, and clarify actionable challenges. The methods we selected align with the four problem selection/problem analysis phases of our Translational Science Framework: contextualize, frame, map, and craft. Methods used included landscape and workforce assessments; purpose and scoping sessions; design research strategies; systems mapping; quantitative needs assessments; and translational science proposal pitches. Results/Anticipated Results: Across the three strategic domains, we successfully reframed and scaled ill-defined problem statements into translational science questions that are actionable, meaningful, feasible, and relevant. The outputs of each problem selection/problem analysis phase served as critical inputs for subsequent phases; for example, synthesis of landscape and workforce assessments in the contextualize phase provided critical substrate for structuring the design research and broad scale needs assessments in the frame phase. This funneling approach brought continuous focus and clarity to the problem statements. A variety of metrics embedded within the phases informed performance, progress, and decision-making, which enabled rapid adjustments in the face of evolving contexts and unforeseen challenges. Discussion/Significance of Impact: In the absence of translational science problem analysis, we risk not addressing actual needs, duplicating effort, and/or creating unwanted solutions. The methods and metrics within the analysis phases of our Translational Science Framework promote iterative refinement of intractable problems, with the goal of yielding impactful solutions.

Objectives/Goals: Synthetic data holds potential for inclusion in medical product development pipelines. Therefore, we explored the current regulatory and practice landscape to identify best practices used to ensure, and communicate to key stakeholders, synthetic data quality, relevance, and reliability in regulatory settings. Methods/Study Population: We identified areas in which synthetic data created using generative AI holds the most value for health researchers. We reviewed regulatory documents, published literature, and expert insights to examine how regulators currently use, define, apply, and govern synthetic data created using generative AI. Next, we identified data management tools, best practices, ethical considerations, and regulatory developments that are necessary for generating fit-for-use synthetic datasets. Using this information, we developed a risk-based credibility assessment framework that aligns with current governmental standards that can be useful for users of synthetic data derived from generative AI applications in regulatory settings. Results/Anticipated Results: Synthetic data created using generative AI holds value for health researchers across four key areas: Acting as a 1) privacy-enhancing technology, 2) data science ’sandbox’ for training and exploration, 3) mechanism to navigate legalities around data sharing and/or use, and 4) method to augment underrepresented subgroups in datasets. However, synthetic data raise ethical and legal concerns, particularly regarding privacy, consent, stakeholder engagement, and ownership. Regulators and health technology assessment bodies, including FDA, EMA, MHRA, and Canada’s Drug Agency, are exploring synthetic data to supplement medical datasets, validate external control arms, enhance model performance, and inform regulatory decision-making. Discussion/Significance of Impact: Our work underscores the need to continue cultivating transparent, ethical, and fit-for-use approaches to synthetic data generation using generative AI. Moving forward, effective synthetic data use and development requires a culture of learning and transparency among regulators, end users, and those involved in data generation and exchange.

Objectives/Goals: The goal of the NCATS CTSA Collaborative and Innovative Acceleration (CCIA) Award Initiative is to advance clinical and translational science through the collaboration, dissemination, and implementation of innovative solutions that have long-term, real-world impact. Methods/Study Population: To evaluate the real-world impacts of the CCIA Initiative, the Translational Science Benefit Model (TSBM) framework was used to create TSBM Impact Profiles for 55 CCIA awards issued between 2016 and 2022 using a secure large-language model (HHS-NIH GPT-5). Profiles were generated using data from final research performance progress reports (RPPRs) or, when those were unavailable, the most recent RPPRs in ERACommons. To guard against inaccuracies, 10% of AI-generated case studies were randomly selected and independently validated by subject matter experts. Descriptive statistics were used to summarize overall CCIA impacts. Results/Anticipated Results: A total of 49 fRPPR and 6 RPPRs were analyzed. TSBM Impact Profiles were efficiently, securely, and accurately generated using AI. Analysis showed that the CCIAs produced innovative solutions impacting all four TSBM domains and each of the nine subcategories. Evidence linked the CCIAs to 28 of the 30 TSBM benefits. The CCIAs had the greatest translational science impact in the Procedure & Guidelines category of the Clinical & Medical Benefits domain; most observed benefits were associated with improvements to Clinical Care Guidelines (25 awards), followed by Therapeutic- (23 awards) and Investigative Procedures (19 awards). Discussion/Significance of Impact: Analysis using the TSBM framework suggests that the CCIA Initiative has had wide-ranging, real-world impacts, indicative of the effective development and dissemination of innovative solutions beyond traditional scientific outputs.

Objectives/Goals: Attention deficit hyperactivity disorder (ADHD) affects 5.9% of youth worldwide and is associated with mental health disorders. In adults, HeartMath biofeedback improves ADHD, emotional regulation, and anxiety. Breathing practices promote similar outcomes. Our goal was to examine the synergistic effect of these therapies in children with ADHD. Methods/Study Population: A 10-week, 2-arm randomized controlled clinical trial was conducted in a rural health care clinic to examine the safety, feasibility, preliminary efficacy, and self-reported child and parent perceptions of a biofeedback plus breathing practices intervention (Healing Minds) in a pediatric population (n=12; M=5; F=7; 11.6 ±3.6 years). Children were randomized to waiting list placebo (P) or treatment (T) intervention [Weekly biofeedback (HeartMath) and breathing exercises (Qi gong, belly, box, and alternate nostril) by trained medical staff]. Symptoms and severity of ADHD (Vanderbilt; Conners), Screen for Child Anxiety-Related Disorders, Child & Adolescent Trauma Screen (CATS), CATS-Caregiver, Child & Youth Resilience Measure, and perceptions by interview were assessed at baseline, 5 and 11 weeks. Results/Anticipated Results: Twelve children were enrolled (6=T; 6=P) into the clinical trial. Nine completed the study [T (N=4); P (N=5)]. One moderate, unrelated adverse event was reported. Treatment attendance rates varied [90% (2), 70% (1), 60% (1), and 30% (1)]. Children self-reported reduced panic attacks, improved sleep, and feelings of self-control. Parents perceived a calmer home with fewer meltdowns, less overwhelming events, and expressed satisfaction with biofeedback, Qi gong, and belly breathing, but not box and alternate nostril breathing. Questionnaires were perceived as burdensome, and placebo waiting list delays were unacceptable to parents. School-related scheduling challenges and missing assessment data limited study implementation. Small sample sizes limited statistical power and the evaluation of outcomes. Discussion/Significance of Impact: Healing Minds shows promise as a beneficial adjunct to clinical care in rural children with ADHD. The program was shown to be safe and feasible. Children and parents self-reported beneficial effects. Future studies should include a patient-centered study design, focused breathing techniques, fewer assessments, and improved school coordination.

Objectives/Goals: Heart disease is a leading cause of death in Alabama. Non-Hispanic Black (NHB) women demonstrate greater arterial stiffness than Non-Hispanic White (NHW) women. We explored whether cardiorespiratory fitness (VO2max) accounts for population differences in arterial stiffness among postmenopausal women. Methods/Study Population: We analyzed data from 57 postmenopausal women (65% NHB; age: 62±8 years; BMI: 28±4 kg/m2; blood pressure: 136±17/80±9 mmHg). VO2max was measured using maximal treadmill testing (Modified Bruce Protocol) and arterial stiffness (pulse wave velocity, PWV) via SphygmoCor XCEL. We used bias-corrected bootstrapped mediation models (5,000 resamples) to estimate the indirect effect of ancestry on PWV via VO2max, adjusting for age, body mass index, lived experience, adverse childhood exposures, and neighborhood deprivation. Given the small sample size, these analyses are exploratory. Results/Anticipated Results: NHB women had significantly lower VO2max than NHW women (18.0±3.7 vs. 23.8±4.5 mL•kg−1•min−1). VO2max was inversely associated with PWV (B = -0.10, SE = 0.04, p = 0.02, 95% CI = −0.19 to −0.01). After adjustment, ancestry was not directly associated with PWV. Neighborhood deprivation (B = 0.16, SE = 0.07, p = 0.02, 95% CI = 0.03 to 0.29) and lived experience (B = 0.10, SE = 0.04, p = 0.03, 95% CI = 0.01 to 0.19) were associated with higher PWV, even after adjusting for VO2max and risk factors. In bootstrapped mediation models, the indirect effect of ancestry on PWV through VO2max was significant (B = 0.35, SE = 0.19, 95% BCa CI = 0.03 to 0.78), consistent with partial mediation. Discussion/Significance of Impact: Cardiorespiratory fitness appears to partly account for population differences in arterial stiffness, but this cross-sectional analysis with a modest sample is exploratory. Targeting both fitness and upstream socioeconomic factors may be needed to reduce differences in women’s vascular health.

Objectives/Goals: Problems with cognition are common and distressing for people with systemic sclerosis (SSc). We are developing a patient-centered cognitive rehabilitation intervention using the multiphase optimization strategy (MOST) framework. Our study aim is to identify patient needs, preferences, and priorities to help address their cognitive challenges. Methods/Study Population: We conducted semi-structured interviews with 15 individuals with SSc to capture in-depth perspectives on desired intervention features, potential barriers and facilitators to implementation, and overall relevance of the intervention to daily life. Transcripts were analyzed using the Rigorous and Accelerated Data Reduction technique in combination with thematic content analysis. Codes with similar content were merged and organized into subthemes and themes to understand the data. Insights from these interviews are used to inform and refine a conceptual model to guide intervention development in the MOST preparation phase, ensuring that resulting components are acceptable, feasible, and closely aligned with patients’ specific needs, preferences, and priorities. Results/Anticipated Results: Participants were 55 years old on average (100% female, 80% White). Forty percent had diffuse cutaneous SSc, and 53% were within 7 years of diagnosis. Participants reported cognitive challenges affecting short-term memory, attention, and processing speed. They highlighted the need for flexible online sessions (having multiple time options to choose from), extra time for discussion, and peer support (a space for connection and group discussion). Barriers to participation included symptom fluctuation, fatigue, and limited time. Facilitators included peer support and user-friendly resources. These insights will refine a conceptual model to guide development of an intervention that is acceptable, feasible, and tailored to the cognitive and lifestyle needs of people with SSc. Discussion/Significance of Impact: We will incorporate feedback from this study to create a patient-centered cognitive rehabilitation intervention that is tailored to people with SSc. Involving people with SSc in intervention development provides the foundation to build a robust, relevant intervention for people with cognitive challenges.

Objectives/Goals: Collaborative agreements are vital for building strong partnerships, particularly during the formation of new research teams. To support team scientists during critical stages of team formation, we developed a flexible, on-demand eLearning module on collaborative agreements with examples and guidelines for developing tailored agreements. Methods/Study Population: The on-demand Collaborative Agreement module empowers learners to complete a full overview of collaborative agreement concepts or to directly access topics of interest within the module. Designed for real-time applications, it includes conversation prompts, downloadable templates, and example scenarios to guide team building and maintenance. To date, the module has been used by clinical and translational science trainees, early-career investigators, faculty participating in mentor training programs, and PIs of a multi-site research project. Early feedback has prompted the development of a consultative service to support teams in developing their own collaborative agreements after team members complete the module. Results/Anticipated Results: We are collecting data through pre- and post-training assessments embedded within the module. Post-assessment also invites respondents to share contact information for follow-up evaluation of their adoption. As of October 2025, we have collected 21 pre-assessment responses and 14 post-assessment responses. We anticipate gathering more data prior to Translational Science 2026. We will summarize assessment data focused on the tool’s effectiveness, learner’s intention to adopt collaborative agreements, and their confidence in building their own agreements. Additionally, we will provide feedback on the value of support services post-training for team implementation of collaborative agreements. Discussion/Significance of Impact: The just-in-time online tool facilitates the use of collaborative agreements for individual learners, teams, and between mentors and mentees. It supports effective collaboration by clarifying shared expectations. Feedback for the module will inform enhancements as well as future translational science workforce development eLearning projects.

Objectives/Goals: To evaluate whether signup recency in the Indiana CTSI All IN For Health Volunteer Registry influences responsiveness to outreach for clinical trial recruitment, testing the hypothesis that newer records are more likely to yield referrals. Methods/Study Population: Outreach data from the TrialX Connect registry were analyzed. Each outreach was coded as referral versus no response. Volunteers were grouped by time since signup: <6 months (n=386), 6–12 months (n=282), 1–2 years (n=659), 2–3 years (n=760), and >3 years (n=13,592). Conversion rate was defined as referrals per unique volunteers outreached. Statistical comparisons were made across categories. Results/Anticipated Results: Conversion rates were stable across categories (2.14–3.36%). The 6–12 month cohort had the highest rate (3.36%), while 2–3 years had the lowest (2.14%). Newer (<6 months) volunteers converted at 2.28%. Most referrals (n=345) came from >3 year signups, reflecting their numerical dominance (13,592 signups, ~90% of registry). Discussion/Significance of Impact: Signup recency showed little impact on responsiveness (steady at ~2–3%). Older records produced most referrals due to larger volume, emphasizing the lasting value of long-term participants. Results suggest focusing on sustained engagement and retention over simply expanding the registry.

Objectives/Goals: Acute myeloid leukemia (AML) is an aggressive blood cancer that hides from the immune system, which led to ~140,000 deaths worldwide in 2024. AML often recurs or resists therapy; therefore, our goal is to develop and test TriKE-PACCs, which combine dual tumor antigen targeting with NK cell activation, to overcome antigen escape and heterogeneity in AML. Methods/Study Population: TriKE-PACC molecules were engineered by combining a TriKE [anti-CD16 single-domain antibody (sdAb), an IL-15 moiety, and an antitumor antibody domain single-chain variable fragment (scFv) or an sdAb] with a PACC arm containing a second antigen-binding scFv/sdAb linked to IL-15Rα. TriKE-PACC constructs were produced in Expi293 cells. To identify clinically relevant AML tumor targets, we analyzed the expression of tumor antigens in AML bone marrow samples obtained from 20 patients using flow cytometry. We modeled antigen escape and heterogeneity using CRISPR/Cas9-generated knockout AML lines. TriKE-PACC-mediated NK cell activity was measured via degranulation, cytokine production, and Incucyte-based cytotoxicity assays against AML cell lines and CRISPR knockout models. Results/Anticipated Results: Our examination of AML patient bone marrow samples revealed diverse antigen profiles, highlighting the need for modular TriKE-PACC designs. The dual-targeting TriKE-PACCs enhanced NK cell activation via CD107a expression and the production of IFNγ and TNFα in co-culture assays with AML cell lines. Notably, TriKE-PACCs retained activity against AML models lacking a targeted antigen, demonstrating the ability of the second tumor targeting arm to overcome antigen escape. Furthermore, NK cell-mediated killing was improved against antigen-deficient AML cells when treated with the TriKE-PACCs compared to standard TriKEs. Discussion/Significance of Impact: TriKE-PACCs offer a modular and customizable NK-engaging immunotherapy to address AML heterogeneity and resistance. Our data, collected utilizing primary patient samples, support the use of this platform for the future development of personalized, dual-targeted therapies to enhance the efficacy of AML treatments.

Objectives/Goals: Linking GAINS evaluates a 14-week community-based program combining genomic testing, ancestry education, and culturally tailored nutrition to improve cardiometabolic health, literacy, and behavior among African American adults in Baton Rouge, Louisiana. Methods/Study Population: Sixty African American adults in Baton Rouge will be randomized into genetically blinded (GB) or unblinded (GU) groups. Both complete baseline surveys, biometric assessments, and a 14-week health behavior program with screenings, nutrition education, cooking classes, and exercise. The GU group receives Color Genomics testing and counseling at baseline; GB receives counseling post-intervention. Follow-up measures assess literacy, diet, and cardiometabolic outcomes. Results/Anticipated Results: CHATGPT SAID: We anticipate improvements in genomic literacy, nutrition knowledge, and willingness to undergo genetic testing, particularly among the unblinded group. Participants are expected to adopt healthier eating and physical activity behaviors, leading to modest reductions in BMI and blood pressure and increased trust in genomics and preventive health research. Discussion/Significance of Impact: Linking GAINS shows how combining genomic testing with culturally tailored nutrition education can improve literacy and lifestyle behaviors, offering a scalable model for advancing equitable precision health in African American communities.

Objectives/Goals: Multivariable statistical models are commonly used as clinical decision aids. However, more research examining real-world implementation of models is needed. The goal of this mixed methods study is to explore to what degree clinician uptake of risk models is influenced by perceptions of the potential risks and benefits of taking action on model output. Methods/Study Population: Mixed qualitative and quantitative data for the ’low-risk/low-reward’ scenario arise from an implementation trial of a electronic medical records-based (EMR) model to alert primary care providers of potential cognitive problems in patients ages 65 and older. We evaluated adoption of the EMR model based on provider actions (cognitive screening/referrals/medications/diagnoses) following Epic best practice alert (BPA), and acceptability of the models through semi-structured qualitative interviews. Data for the ’high-risk/high-reward’ scenario will be gathered from clinical notes documenting the use of statistical models to support elective surgical decision-making in scenarios where such models have been regularly used at our institution. Mixed methods analyses will be holistic. Results/Anticipated Results: Early quantitative analyses following implementation indicated that primary care providers largely did not adopt the BPA, but did take action (referrals, diagnoses, medications) when automated cognitive screening was performed in preparation for the appointment. Qualitative analyses of interview data to understand provider perspectives are ongoing. In the high-risk/high-reward scenario, we expect that providers will more frequently utilize statistical models for decision-making and report positive attitudes regarding their use, when compared to the low-risk/low-reward scenario. Discussion/Significance of Impact: Though used in practice with increasing frequency, the implementation of personalized statistical prediction models has not been widely studied. Understanding in what scenarios providers feel risk models are useful and actionable can guide decision-making for the development and implementation of future models.

Objectives/Goals: Texas Southern University (TSU), a low-resourced HBCU and one of the largest in the USA, without an affiliated hospital, clinic network, or patient population, established a community-focused clinical research site to educate, recruit, and expand research participation, especially among those with unmet health needs. Methods/Study Population: The implementation process involved strategic planning in five key areas as below: * Collaboration: Partnered with mentor site to align site capabilities with industry standards * Capacity Building: EstablishCTMS, eReg, eSource, and REDCap workflow * Workforce Training: Hired and strengthened research personnel expertise through training, certification, and mentorship * Community Engagement: Enhanced trusted relationships with community members, key stakeholders, and advisory board members * Infrastructure Development: Defined site capabilities: regulatory processes, data management, IRB optimization, budgeting, contract management, and prospective study feasibility assessments. Developed standard operating procedures and clinical research registry for site readiness Results/Anticipated Results: A first of its kind Community-Focused Clinical Research Site in a low-resourced HBCU without an affiliated hospital, clinic network, or patient population was successfully established at TSU with mentorship by the University of Texas Medical Branch Galveston, PhRMA Foundation EQBMED, and RCMI Coordinating Center. A registry for potential research participants was successfully implemented in May 2025. Since its inception, 160 community members have indicated interest in clinical research registry participation. 56 community members are fully enrolled in the registry. Key challenges included bridging feasibility gaps at TSU, navigating the development of new clinical research infrastructure, and managing regulatory and contracting approval timeline. Discussion/Significance of Impact: Establishing a clinical research site in a low-resourced HBCU without an affiliated hospital requires cross-disciplinary collaboration and intentional community engagement. TSU’s approach offers a replicable model for integrating clinical research to expand access to community member especially among those with unmet health needs.

Objectives/Goals: This scoping review aims to synthesize current literature on post-deployment monitoring of AI-enabled digital health solutions within clinical practice. Findings identify existing approaches and gaps that inform guidance for post-deployment monitoring in clinical practice. Methods/Study Population: We conducted a scoping review in accordance with PRISMA-ScR guidelines to characterize the current landscape of post-deployment monitoring in healthcare systems. A PubMed search targeted peer-reviewed articles in English published between 2015 and mid-2025, including text or MeSH terms on 1) health system/hospital; 2) artificial intelligence; 3) post-deployment; and 4) evaluation/monitoring. We performed a thematic analysis to identify common challenges, gaps, and opportunities in AI oversight. Additionally, we reviewed guidelines addressing post-deployment AI monitoring. All analyses were conducted using Rayyan.ai and Microsoft Word. Results/Anticipated Results: Among the six studies included after the full-text review, five provide recommendations to ensure transparency, safety, and model performance. These recommendations encompassed monitoring model performance and real-time case report, post-market surveillance, adverse event reporting, end-user training, data standardization and documentation, and interdisciplinary collaboration. One study reports a framework for post-deployment impact grading. Currently, no guidelines addressing post-deployment of AI monitoring in health systems exist. Our findings highlight the urgent need for structured post-deployment processes to ensure AI in healthcare systems is safe, effective, and trustworthy. Discussion/Significance of Impact: The absence of post-deployment guidelines raises concerns. This review underscores the need for interdisciplinary collaboration to establish a post-deployment monitoring process with scientific rigor, scalability, and sustainability that aligns with operational realities.

Objectives/Goals: The objective is to determine the relationship between prior progestin intrauterine device (IUD) use, duration of use, and adverse pregnancy outcomes and also to examine the return-time to fertility and whether demographic factors modify the effect of IUD use on pregnancy outcomes. Methods/Study Population: This is a prospective cohort study of 600 pregnant adults between 10 and 28 weeks of gestation who have not delivered a fetus. We will compare participants with and without prior progestin IUD use. Participants are recruited via flyers posted in university-associated OB-GYN clinics and via the health portal, MyChart. Participants complete a comprehensive contraception, medical, and pregnancy history and receive compensation. Pregnancies are followed until 6 weeks post-delivery, and outcomes are collected via the electronic health record. Pregnancy outcomes will be analyzed using multivariable logistic regression with adjustment for confounders, with sensitivity analyses stratified by duration of prior progestin IUD use. Results/Anticipated Results: We will characterize the prevalence and duration of progestin IUD use in the cohort and evaluate demographic factors associated with IUD use compared to other contraceptive methods. We will compare the rates of placental abnormalities, as well as miscarriage, preterm birth, hypertensive disorders of pregnancy, including preeclampsia and gestational hypertension, and other complications, between participants with prior progestin IUD use versus non-use. Among participants with prior IUD use, we will examine whether the risk of adverse pregnancy outcomes is associated with the length of use. We will also examine the time between the last contraceptive use and conception. Discussion/Significance of Impact: An association between prior progestin IUD use and pregnancy outcomes would be a novel finding and will have clinical implications for how contraceptive history may inform counseling and reproductive care. We hope that this study will rationalize the need for well-powered, larger studies to investigate the robustness of our study findings.