OBJECTIVES/GOALS: This project seeks to understand how personalized medicine can optimize care for patients with colorectal cancer. It identifies opportunities for personalized medicine to improve clinical outcomes, and uses cost-effectiveness analysis to assess the clinical and financial impact of this approach. METHODS/STUDY POPULATION: This project uses two methods to understand the impact of personalized medicine. First, this project has used SEER-Medicare data in conjunction with Clinical Pharmacogenetics Implementation Consortium guidelines to identify medications used by patients with colorectal cancer that can be impacted by genetic variants. This data will then be combined with population genetic variant rates to understand the likely impact screening for a given variant will have on medication response and adverse events. Medication use frequencies and genetic variant rates are then used to populate cost-effectiveness models that simulate the clinical and financial outcomes, identifying optimal genes to screen. RESULTS/ANTICIPATED RESULTS: The first result will be a comprehensive overview of treatment patterns for patients with colorectal cancer in the United States, as well as the treatments used for disease-induced comorbidities. The second result will be the identification of genetic variants based on population rates and medication utilization that should be screened in this patient population. The final result will be a breakdown of the clinical and financial outcomes associated with implementing screening for the identified genes. Preliminary results from a two-gene cost-effectiveness analysis demonstrates that screening for variants in those genes improves both clinical and financial outcomes. DISCUSSION/SIGNIFICANCE OF IMPACT: This project demonstrates how current treatment approaches can be optimized via personalized medicine. It uses epidemiological methods to identify opportunities to integrate genetic findings from other diseases, and uses cost-effectiveness analysis to understand the impact of transforming care. CONFLICT OF INTEREST DESCRIPTION: Stocks-Aurinia, Syndax, Adaptimmune, Rigel pharma

OBJECTIVES/GOALS: There is an imperative need to initiate translational genetic studies of hidradenitis suppurativa (HS). Such work requires large cohorts and no HS registries exist. Precision medicine initiatives provide new resources and methods for efficiently constructing cohorts, but empirically informed best practice guidelines are needed. METHODS/STUDY POPULATION: Traditional methods for building cohorts rely on clinical encounters to identify patients and collect phenotype data. Precision medicine initiatives aim to decrease the time and cost of data collection by using alternative sources, including electronic health records (EHR) and remote collection of patient-reported data. The public’s use of the Internet to obtain and exchange health-related information coupled with the success of direct-to-consumer genetic companies suggests that it is feasible to remotely ascertain research participants for genetic studies. Importantly, Internet cohorts provide an opportunity to include research participants who are disconnected from healthcare, and thus remain hidden from research that relies on EHR or clinical services. RESULTS/ANTICIPATED RESULTS: First, to conduct studies in EHR we are developing an analytic pipeline for the automated extraction of an accurate HS diagnosis using natural language processing of clinical notes. In our preliminary work we are also using ICD codes to build cohorts in two EHR systems with and without linked genetic data. Second, we have developed Internet advertising campaigns for symptom-based recruitment. Informed consent and patient-reported data is collected on-line through a series of short surveys. Patients who complete the surveys and express interest in participating in genetic studies are sent saliva collection kits and return mailing material. Finally, we have established an HS biobank that has DNA from 300 participants identified through clinical services. Enrollment is on going. DISCUSSION/SIGNIFICANCE OF IMPACT: Our goal is to assemble an HS cohort that is large enough to power genetic discoveries. Our work is generating empirical evidence for precision medicine guidelines and will improve our knowledge about HS. The methods we are developing can be applied to efficiently create new cohorts for genetic studies of other diseases across different clinical areas.

OBJECTIVES/GOALS: There is no objective, biologic tool to detect concussion or guide clinical management. We previously showed that saliva microRNA (miRNA) levels differ in children with concussion and may predict symptom duration. The purpose of this study was to validate the utility of saliva miRNA and define longitudinal trends during the recovery period. METHODS/STUDY POPULATION: We collected concussion symptom burden (SCAT-5), cognitive performance (DANA), balance measures (ClearEdge), and saliva from 150 children (7-21 years) with concussion over 5 time-points: 0-2, 3-6, 7-14, 15-29, and 30-60 days post-injury. Saliva miRNA levels within the 443 concussion samples were quantified with RNA sequencing and compared to 218 samples from age- and sex-matched controls (healthy and post-exercise participants). Non-parametric ANOVA assessed RNA levels across time-points, and between concussions/controls. Machine learning was used to build logistic regression algorithms differentiating concussions/controls, and symptomatic/recovered concussion participants. Relationships between miRNAs and concussion phenotypes were explored with Spearman’s Rank correlations. RESULTS/ANTICIPATED RESULTS: Fifteen miRNAs differed across control and concussion participants (FDR < 0.05). Within concussion participants, all 15 miRNAs trended back toward control levels by 30-60 days post injury. A regression algorithm employing 6 of the 15 miRNAs differentiated control and concussion participants with an area under the curve (AUC) of 0.78 in a training set (n = 244) and 0.84 in a naïve test set (n = 24). Similarly, 6 miRNAs were able to differentiate symptomatic (SCAT-5 symptom score > 7) and asymptomatic concussion participants with an AUC of 0.73 in a training set (n = 219) and 0.76 in a naïve test set (n = 44). Furthermore, 5 miRNAs showed significant (R > 0.3; FDR < 0.05) associations with subjective and/or objective measures of concussion-related symptoms. DISCUSSION/SIGNIFICANCE OF IMPACT: Saliva miRNAs levels are altered in children with concussion, and display predictable longitudinal trends following injury. Saliva miRNA measurement represents a non-invasive, objective tool that could be rapidly assessed to provide biologic evidence for clinicians managing pediatric concussion. CONFLICT OF INTEREST DESCRIPTION: I serve as a paid consultant and scientific advisory board member for Quadrant Biosciences, who has funded a portion of this work and licensed the findings from the Penn State College of Medicine.

OBJECTIVES/GOALS: It has been previously shown that pediatric high-grade glioma (pHGG) survival is different between sexes. We set out to find out whether there are sex-specific differences in the genomic landscapes of pHGG that may underlie this sex disparity. METHODS/STUDY POPULATION: We downloaded Illumina 450k DNAm data from ArrayExpress and GeneExpressionOmnibus. The minfi package was used to process raw DNAm data. Sex chromosomes and CpGs that are common SNPs were removed. Surrogate variables (SVs) were estimated via the sva Bioconductor package. Differentially methylated CpGs were identified by fitting a multiple linear regression model for the DNAm level at each CpG, with independent variables being sex (a binary variable) and the estimated SVs. RNAseq data was downloaded from Cavatica, and differential gene expression analysis was carried out via the DESeq2 package. RESULTS/ANTICIPATED RESULTS: In the pediatric glioblastoma (GBM) DNAm data [58 female & 91 male IDH wt samples; ages 0.1–21 yrs;], we found 7,371 differentially methylated cytosines (DMCs) at FDR≤0.05. Of the DMCs, 289 had DNAm differences between male and female samples ≥10%. The majority of probes (68%) were in CpG islands, shelves, or shores. We also found 4 differentially methylated regions (DMRs) between sexes (FWER≤0.1). In the adult GBM DNAm samples [32 F & 32 M IDH wt samples; ages 22–75 yrs], we found only 117 DMCs at FDR≤0.05, and no DMRs. In the RNAseq dataset [68 F & 54 M pHGG samples, ages 0.08–30.6 yrs], we found 383 differentially expressed genes (at FDR≤0.05), and 16 of them (4%) overlapped a DMC. DISCUSSION/SIGNIFICANCE OF IMPACT: Our findings demonstrate that pHGG exhibits sex-specific methylome differences. Interestingly, this difference is greater in the pediatric population as compared to adults. The pHGG transcriptome also differs by sex, which may be related to differential DNAm in a minority of cases.

OBJECTIVES/GOALS: We use a tissue engineered, biomimetic, 3D model to study the pathogenesis of breast implant-associated anaplastic large cell lymphoma (BIA-ALCL) by comparing the effect of silicone implant shell on proliferation of patient-derived BIA-ALCL to its precursor T cells within the breast microenvironment. METHODS/STUDY POPULATION: Patient-derived breast tissue was processed for component adipocytes, ductal organoids, and stromal vascular fraction. These were suspended within 50 µl of 0.3% type I collagen matrix to which was added 200,000 cells/mL of either patient-derived BIA-ALCL cells or T progenitor cells. These were then plated into 6mm wells. As a control, both BIA-ALCL cells and T progenitor cells were suspended within type I collagen alone at the same seeding density without breast components. Before plating, wells were lined circumferentially with either textured, smooth, or no implant shell. These were 1cm by 2cm pieces dissected from the whole implant. Wells were imaged using confocal microscopy over 8 days. RESULTS/ANTICIPATED RESULTS: Unstimulated T progenitor cell count showed no significant increase in any of the conditions tested. The change in cell count over 8 days was 3.85% in each condition (p = 0.3352). A Tukey’s multiple comparison test comparing each condition revealed no significant increase in cell count over 8 days for all six conditions. Notably, our previous studies have shown proliferation of BIA-ALCL cells to be significantly more robust in the biomimetic platform compared to collagen-only groups, regardless of implant shell type (p < 0.01). BIA-ALCL cells grew nearly 30% faster in textured and smooth shell biomimetic groups compared to biomimetic wells lacking implant shell. DISCUSSION/SIGNIFICANCE OF IMPACT: Towards elucidating BIA-ALCL’s etiopathology, we show that silicone implant shell has a significant effect on proliferation of BIA-ALCL cells, but not their precursor T cells. If breast implant silicone shell is not a sufficient stimulus for T cell proliferation, co-stimulatory factors are required.

OBJECTIVES/GOALS: Traumatic interpersonal experiences are associated with higher rates of chronic pain, increased pain severity and poorer functioning. The objective of this ongoing project is to obtain prevalence rates for various forms of interpersonal trauma among individuals with chronic pain, and to explore the potential mediating effect of heightened sensory and social sensitivity on the experience of pain. METHODS/STUDY POPULATION: Patients at Michigan Medicine between the ages of 18 and 65 complete an online survey. Patients are being recruited through a tertiary-care, outpatient pain clinic, as well as through an online health research portal. We aim to recruit 700 participants; we currently have 59.6% of our goal (n = 417). Participants also have the option to be included in a registry from which we can recruit for future studies. Approximately 85% of our participants have agreed to be in the registry. RESULTS/ANTICIPATED RESULTS: Preliminary data show that, of the 263 (63.4%) participants for whom data on chronic pain is available, 167 (63.5%) report chronic or persistent pain over the previous 3 months. Of these, 54% reported some form of childhood abuse or neglect. Approximately 41% reported four or more adverse childhood experiences. Additionally, of the 122 participants (73%) who were in a current romantic relationship, 20% reported some form of physical violence victimization from their romantic partner. We anticipate that interpersonal trauma will be associated with poorer perceptions of social relationships, higher sensory sensitivity, and higher perceived stress. DISCUSSION/SIGNIFICANCE OF IMPACT: The PASE Project parent study will be used to better understand prevalence rates for various forms of interpersonal trauma in our chronic pain population. Future analyses and studies will explore alternative pathways linking interpersonal trauma to the experience of pain through sensory and social sensitivity, which will inform interventions aimed at reducing pain among patients with a history of trauma.

OBJECTIVES/GOALS: Cell-cell (CC) and cell-matrix interactions (CM) are known to affect drug sensitivity of cancer cells, but are not effectively recapitulated using 2D platforms. This research aims to determine how cell and matrix interactions confer drug resistivity in 3 distinct culturing models: 2D (no CM/limited CC), 3D spheroids (CC) and 3D fibronectin (both). METHODS/STUDY POPULATION: We examined four breast cancer cell types. The cells were derived from a nonmetastatic primary tumor (HMLE-E2) or overt bone-metastasis (BM). Transglutaminase 2 (TGM2), a matrix crosslinking protein, is overexpressed in metastatic bone tumors and may play a key role in matrix-conferred drug resistivity. In a gain-of-function model, TGM2 was upregulated in HMLE-E2 cells and compared to shTGM2 knockdown BM cells. Growth rates were analyzed using metabolic activity over 8 days, and drug sensitivity to Neratinib (0-1000 nM) was analyzed via cell titer. To account for the different transport properties of the 3 distinct culture environments, we developed a mathematical model for each condition, allowing us to normalize the drug sensitivity results across models to effectively compare true biological resistivity. RESULTS/ANTICIPATED RESULTS: We observed that increased cellular levels of TGM2 significantly increase the growth rate and drug resistivity of cells on fibronectin matrices. Interestingly, in 2D cultures, TGM2 expression was correlated with higher Neratinib resistivity but did not affect growth rates. In spheroid models without a significant matrix component, that rely solely on cell-cell junctions, high levels of TGM2 were correlated with lower survival rates. Lower levels of TGM2 are correlated with a more epithelial phenotype, and using our mathematical model we have identified significant transport differences between high and low TGM2 spheroids. We theorize that the low TGM2 spheroids have denser packing, which lowers the rate of diffusion and, thus reduces the effective concentration of the drug to the majority of the cells. DISCUSSION/SIGNIFICANCE OF IMPACT: Our studies indicate that the cellular response to drugs can be altered by changes in both transport properties of the tissue and the CM interactions. By systematically investigating the effects of CC interactions and CM interactions, we can use mathematical models to delineate physical means of drug resistivity from a biologically driven resistance.

OBJECTIVES/GOALS: Real-world studies have been gaining momentum in providing evidence of treatment effectiveness and hold great potential for facilitating the drug regulatory process. The U.S. Food and Drug Administration (FDA) has recognized this by providing a framework for using real-world data (RWD) to generate real-world evidence (RWE). The objective of this study is to assess the current level of RWE implementation in clinical studies. METHODS/STUDY POPULATION: Using keywords relevant to RWE, we reviewed studies on drugs, biologics, and medical devices published on PubMed in 2018. Information regarding the therapeutic area of focus, intervention type, study design, primary outcome, and data source was recorded. Further analyses of the three main therapeutic areas of study (oncology, cardiology, and infectious diseases) were performed to determine how RWE was being utilized. In addition, a broad “real-world” search was performed on Clinicaltrials.gov, from which we extracted relevant observational and Phase I, II, II/III, III, III/IV, and IV studies. A supplemental PubMed search was used to evaluate published studies in order to identify which field these trials were concentrated in and the outcome of interest.” RESULTS/ANTICIPATED RESULTS: After application of “real-world” search terms to PubMed, 995 hits were generated and of these, 311 studies were excluded. More than half of the studies were observational and retrospective in nature (64%) with 70% examining drug/biologic outcomes. RWE data sources were largely dominated by medical records and claims data. The primary uses of RWE across oncology, cardiology, and infectious diseases included supporting drug product effectiveness, assessing safety, and evaluating treatment patterns. Of the 207 RWE studies identified on ClinicalTrials.gov, 66 were cancer randomized controlled trials (RCTs), a majority of which were used for post-marketing safety evaluations. Further research will be conducted to determine the precise role of RWE in all studies (e.g. historical comparator, label expansion). DISCUSSION/SIGNIFICANCE OF IMPACT: By examining the use of RWE in regulatory decision making, we can inform stakeholders of the extent to which robust RWE studies complement evidence generated by RCTs. Thought to reflect a product’s performance in a broader and more diverse population, RWE can provide greater insight to clinical trial conduct and ultimately transform patient outcomes.

OBJECTIVES/GOALS: The goal was to understand the effectiveness of a novel clinical trial educational module and a corresponding initiative designed and disseminated by the Southern California Clinical and Translational Science Institute (SC-CTSI) to increase the quality of clinical trials conducted in academia. METHODS/STUDY POPULATION: The CRCs (Clinical Research Coordinators) for the initiative are asked to complete the online training. Possible study protocols are picked to be monitored by the CRCs. The monitor is instructed to study the protocol extensively and prepare for their monitoring visit. The trained monitor from the initiative then reaches out to the CRC of the study that is to be monitored and carries out the monitoring visit. Afterwards, the monitor sends initiative personnel the monitoring report, which is evaluated to see if the monitor checked everything they should have during the visit. The PI of the study is contacted with highlights from the monitoring report and improvements that they can make. RESULTS/ANTICIPATED RESULTS: The first study monitored was a site of a large NIH-sponsored study where the consent forms were signed electronically. It was found that the monitor could not access the consent forms. Therefore, the monitor could not do source data verification. The PI of the study said that they would be raising this issue with the NIH. During the monitoring visit of the second study chosen for the initiative, patient binders were specifically examined for informed consent and source documentation completeness. The charts of patients were also reviewed. The only deviation found was a missing signature in the Investigator Site File. For the last two studies, data will be reported. DISCUSSION/SIGNIFICANCE OF IMPACT: Monitors were not only able to monitor efficiently, but also able to point out deficiencies in the monitoring practices of large studies. This model could be expanded to other academic institutions to establish quality management systems to ensure data integrity and subject protection.

OBJECTIVES/GOALS: Navigating the NIH Single IRB Policy has been challenging for investigators, study teams, and Human Research Protection Programs (HRPP). In response, the Indiana Clinical and Translational Sciences Institute (CTSI) created an innovative Single IRB Project Manager role (sIRB PM), uniquely placed within the Indiana CTSI. METHODS/STUDY POPULATION: The Single IRB Project Manager role was created in 2018 by the Indiana CTSI in response to the NIH Single IRB Policy for Multi-Site Research. The role of the sIRB PM is to serve as a liaison between the Indiana University HRPP, lead site, coordinating center, and participating sites when Indiana University serves as the Single IRB. This model has proven useful to both the IRB and lead site, notably in the following ways:

• At study start-up, the sIRB PM can handle complicated communications among sites and the IRB at the same time the lead site is responsible for many other administrative tasks related to start-up. By absorbing the workload of IRB approval for multiple sites, the sIRB PM provides the lead site more capacity to handle other essential tasks.

• The sIRB PM translates new terminology and facilitates processes that are new for sites.

RESULTS/ANTICIPATED RESULTS: Early assessment of this program is predominantly positive. The sIRB PM currently supports 24 external sites. In an NIA-funded 13 site study, all sites were added within 9 months of initial IRB approval of the protocol. This role fills a gap that benefits:

• IRB staff by allowing them to fulfill their duties of screening and review while leaving some of the reliance organization to the sIRBPM.

• Lead PI by allowing them to focus on conducting the research instead of the many administrative tasks required for single IRB review.

• Participating sites by having a liaison to enter their amendments and reportable events into an otherwise closed IRB software system.

• All parties by having the sIRB PM manage document organization, storage, and distribution study-wide.

DISCUSSION/SIGNIFICANCE OF IMPACT: The CTSI sIRB PM role effectively shifts administrative work caused by the sIRB mandate by merging research coordinator experience with regulatory experience while building upon an existing strong relationship with the HRPP. Future focus is on process education, standardizing pricing structure, and ensuring sufficient budget support in grants.

OBJECTIVES/GOALS: Clinical research is the backbone of the medical community. However, there are few regulations to ensure clinical trial participants can understand their results, leading to volunteers feeling unvalued and unlikely to enroll in trials1. This study examines the need of lay summaries METHODS/STUDY POPULATION: To understand the current landscape of clinical trial summaries, literature searches were conducted using the University of Southern California Library database with keywords Title contains “lay language” OR “lay summary” AND any field contains “Trial” OR “clinical”, and Title contains “natural language processing” AND “clinical trial” OR “Summary”. Studies were deemed relevant if they discussed lay language summaries for health care realms or using Natural Language Processing (NLP) to increase comprehension. Papers published by the Center for Information and Study on Clinical Research Participation (CISCRP) were reviewed and their Associate Director was interviewed. RESULTS/ANTICIPATED RESULTS: Of 67 total results, 14 were determined to be relevant. Ten of the relevant results examined lay language summaries and their regulation and 4 were NLP studies. The European Medicines Agency set regulations mandating clinical trial summaries. However, researchers have difficulty validating to an appropriate reading level2. Difficulty and potential bias halted a U.S. mandate of lay summaries3. The nonprofit CISCRP has partnered with industry to develop unbiased clinical trial summaries resulting in all volunteers feeling appreciated and 91% understanding clinical trial results post summary1. Similarly, NLP software for annotating Electronic Health Records increased comprehension for 77% of patients4. DISCUSSION/SIGNIFICANCE OF IMPACT: In the U.S., a lack of regulations mandating lay summaries may be related to concerns by regulatory agencies that summaries in plain language may introduce bias3. Future looks into integration of NLP systems to clinical trials may create unbiased summaries and allow for FDA regulation.

OBJECTIVES/GOALS: Introduction: Between 2014 and 2019 the National Institute of Health (NIH) through the National Center for the Advancement of Translational Science (NCATS) has awarded about $2.7 billion to U.S. Academic Medical Centers to build a national network of clinical and translational science program hubs that serve to meet their key goals and initiatives. Today there are about 60 Clinical and Translational Science Award (CTSA) program hubs. Each CTSA program hub has a corresponding website highlighting its clinical and translational science centered programs and activities. These websites are a critical communication gateway to promote NCATS goals and initiatives. Objective: The objective of this research is to evaluate the NIH funded Clinical and Translational Science Award (CTSA) program hub websites for NCATS goals and initiative content alignment, navigability, and interactivity. METHODS/STUDY POPULATION: Methods: Each CTSA program hub website was systematically evaluated for information or tools that align with the five NCATS / CTSA Goals and eight CTSA nationally identified program initiatives. Each NCATS goal and CTSA initiative was subsequently ranked by information diversity level (text, tool, interactivity) and navigation level (click distance from the home page). RESULTS/ANTICIPATED RESULTS: Results: Four of the five NCATS goals are thoroughly and consistently represented among the CTSA Consortium with workforce development, patient and community engagement, and quality and efficiency of research being the top three. Informatics is thoroughly and consistently represented, but not always clearly identified on the home page. The most underrepresented goal is integration of special and underserved populations which was identified on only 60% of CTSA program hub websites. The most common focus of the eight CTSA program initiatives is the Trial Innovation Network in CTSA program hub websites. The Smart IRB comes in a distant second. The remaining six initiatives are severely underrepresented. DISCUSSION/SIGNIFICANCE OF IMPACT: Discussion: The identification of these gaps among the CTSA program hubs presents an understanding of content management and website functionality among the consortium from 3 principal approaches. First it creates an understanding of CTSA program hub content alignment with its funding source goals and initiatives. Such an understanding presents an opportunity to promote ways to create a better aligned consortium with improved collaboration pathways by the funding source through program hub website content standards. Second, it creates an opportunity for program hubs to understand and respond to the messaging their websites are presenting as it relates to the funding source. Third, it provides an opportunity to identify specific program initiatives and goals the CTSA institutions independently chose to highlight which can open a dialog to the better understanding the value of the program initiatives as they relate to the needs of CTSA program hubs. Ultimately, CTSA websites through content alignment should lead to an improved user experience.

OBJECTIVES/GOALS: Guided by the health belief model and social identity theory, we aim to identify socio-cultural and psychological factors that influence rural tobacco users a) participation in research and b) quitting tobacco use. We also explore how citizen scientists are perceived as disseminators of messages. METHODS/STUDY POPULATION: In Phase I of this multi-stage project, we are conducting in-depth interviews with approximately 30 tobacco users. Interviews are on-going, and have been conducted with 16 participants thus far from four rural counties in Florida. The interview consists of semi-structured questions and multiple validated questionnaires. Specifically, we ask a series of questions about participants’ barriers to participating in research, tobacco use history, and internet use and message preferences. Additionally, we include questionnaires on participants’ substance use, nicotine dependence, motivation to quit, and willingness to participate in research studies. RESULTS/ANTICIPATED RESULTS: Initial findings suggest that rural tobacco users have an overall positive perception of research, and many choose to participate in research for altruistic reasons (i.e. they want to help others). Further, participants noted described feeling stigmatized due to their tobacco use. Although most began smoking to fit in with their community, many now feel on the outs. Participants also reported logistical barriers to participating in research, including lack of transportation. DISCUSSION/SIGNIFICANCE OF IMPACT: Findings can inform the development of recruitment materials to resonate with rural adults, including by emphasizing the collective potential to help by participating. This interdisciplinary highlights areas for collaboration to enhance the reach of health education and public health messages.

OBJECTIVES/GOALS: There is universal recognition of the importance of team science and team leadership. We have developed a semi-quantitative translational science specific team leadership competency assessment tool and have begun implementation studies to assess the impact of personalized feedback on the team science leadership skills of KL2 Clinical Scholars. METHODS/STUDY POPULATION: To create the instrument, we employed a modified Delphi approach by conducting a thorough literature review on Leadership to concretize the relevant constructs, then used these extracted constructs as a springboard for the Rockefeller Team Science Educators (TSE’s) to discuss and refine the leadership domain areas, collectively create domain-specific survey items. Further discussion helped refined the number, grouping, and wording. Scholars also contributed feedback in item development. We piloted the Leadership Survey by having all of the Rockefeller TSEs rate Clinical Scholars, and having each Scholar rate themselves. Each item was answered using a six-point Likert scale where a low score indicated poor expression and a high score represented excellent expression of the specific leadership attribute. RESULTS/ANTICIPATED RESULTS: Incorporation into a REDCap data base made consenting and rating process by TSE’s and the Scholars straightforward. The a priori domains (Foundational Leadership Competencies, Professionalism, Team Building and Team Sustainability, Appropriate Resource Use and Study Execution, and Regulatory Accountability) had high internal validity and good internal factor structure. The congruence between TSE and Scholar self-ratings were uniformly high, and discordance was often a function of “confidence” and “modesty” on the part of the scholar, rather than deficiency. Supporting comments were informative about performance barriers and mechanisms for improvement. Return of results allowed for the exploration of training gaps. Scholars were surveyed to gauge their reaction to the formal feedback. DISCUSSION/SIGNIFICANCE OF IMPACT: This quantification of team science leadership constructs has allowed for A)- the articulation of constructs essential for successful Translational Scientists to acquire during their training, B)- identification of gaps in that training and skill set, and C)- mechanisms for bolstering any identified gaps in these essential leadership constructs. CONFLICT OF INTEREST DESCRIPTION: None

OBJECTIVES/GOALS: As part of the Cleveland CTSA, “Collaborative Working Retreats” have been developed for the purpose of being a catalyst to move groups of interdisciplinary investigators and stakeholders to collaborative research teams with feasible and actionable translational research projects. METHODS/STUDY POPULATION: Groups of interdisciplinary investigators with engaged stakeholder(s) were invited to apply. Selected groups participated in a 4-hour, professionally facilitated retreat, tailored to the unique needs of each team. In addition to the facilitator, a graphic recorder was utilized to capture ideas and aid in decision making by creating a visual narrative linked to the team’s overall vision. Teams were charged with generating three translational research projects and writing a formal Team Action Plan (TAP) by two months post retreat. Retreat participants were asked to complete a survey to evaluate the retreat, and structured interviews were conducted with team leaders 4-6 months post retreat. RESULTS/ANTICIPATED RESULTS: Six groups were awarded retreats, comprised of 48 investigators (representing all schools in the university and 3 of 4 affiliated hospital systems) and 28 stakeholders for a total of 76 participants. 45% completed the followup survey. 77% said they would recommend the service to other teams or would use it again themselves and 97% stated their team benefited from having a facilitator. At 2 month follow up, one team had completed the TAP and subsequently applied for federal funding. However, 4 of the remaining 5 teams indicated that they had made significant progress, attributing progress to their retreat time. Each teams’ progress is being tracked for 2 years, using a newly developed metric. DISCUSSION/SIGNIFICANCE OF IMPACT: Facilitated retreats appear to serve as an important catalyst for progression of translational research projects, providing needed time and support for brainstorming and planning. Lessons learned, pre-retreat work, and tools for tailoring retreat content and tracking progress will be presented.

OBJECTIVES/GOALS: Health care systems are complex, dynamic, and varied. Advances in artificial intelligence (AI) are enabling healthcare systems to use their own data to elicit patterns and design suitable interventions. To realize this potential, computer scientists and clinicians need an effective, practical, and replicable approach to collaboration METHODS/STUDY POPULATION: In this study, computer scientists partnered with clinicians to investigate predictors of avoidable emergency department use. The team sought an approach to computational medicine that could increase the relevance and impact of prediction to solve pressing problems in the health system. The team adopted an emergent architecture that engaged system leaders, computer scientists, data scientists, health services researchers, and practicing clinicians with deep ambulatory and inpatient knowledge to form the initial questions that shaped the prediction model; to understand nuances of coding and recording in source data and the implications for models; and to generate insights for promising points of intervention. The team recorded decisions and challenges as it progressed to analyze its function. RESULTS/ANTICIPATED RESULTS: Most avoidance models focus on a narrow time period around target events, or on high cost patients and events. This interdisciplinary team used their insights into the health system’s workflows and patient population to adopt a longitudinal approach to their prediction models. They used AI to build models of behavior in the system and consider prevention points across clinical units, time, and place. The holistic, systemwide focus enabled the team to generate insights that the system leaders and subsequently specific clinical units could apply to improve value and outcomes. A facilitated team process using learning system and cooperative network principles allowed a large and modular interdisciplinary team to build a transparent AI modeling process that yielded actionable insights into hypercomplex workflows. DISCUSSION/SIGNIFICANCE OF IMPACT: An architecture for involving diverse stakeholders in computational medicine projects can increase the relevance and impact of AI for solving care delivery problems in complex health systems. Translational science and computational medicine programs can foster this type of engagement and encourage a whole system perspective.

OBJECTIVES/GOALS: We examined how individual characteristics and characteristics of the socioeconomic and built environment were associated with care coordination’s effect on cardiovascular disease (CVD) risks to identify geographic areas that may benefit from supplementary clinic-community linkages. METHODS/STUDY POPULATION: We analyzed data with geocoded residential addresses and data from electronic health records for 9946 adults from a Centers for Medicare & Medicaid Services funded innovation project from 7/1/2013 to 3/30/2015. Variables included patient-level demographics, Elixhauser comorbidity index, total time with a nurse care manager, and neighborhood factors such as poverty indicators, walkability, and social capital index. Outcomes were change in CVD risk factors, hemoglobin A1C, blood pressure (BP), and low-density lipoprotein (LDL). Generalized linear models were used to assess the effect of nurse care management program on outcomes after controlling for confounding factors. RESULTS/ANTICIPATED RESULTS: We report preliminary models that include patient demographics (age, sex, race), health care utilization, nurse care manager contact time, Elixhauser comorbidity index, neighborhood education status, percent of population below 200% federal poverty level, median home value, walkability score of the residential address, and social capital index. After adjusting for all mentioned variables, in adults with HbA1C more than 7.5% at baseline, females had worsening HbA1C by 0.53% over the study period. Additionally, LDL values in females worsened over the study period by 4.8 mg/dL after adjusting for all variables. No clinically significant changes were noted for BP. DISCUSSION/SIGNIFICANCE OF IMPACT: Women’s HbA1C and LDL worsened despite nurse care management and may benefit from additional community-based interventions or interventionists. In future analyses, we anticipate that CVD risk will worsen for patients with higher fast food proximity and with greater geographic distance from their PCP.

OBJECTIVES/GOALS: Collaborations are at the core of translational science and team science. Differences by gender have been identified in various research contexts from recruitment to retention to promotion. This study assesses the relational associations of translational collaborations, and what role of gender. METHODS/STUDY POPULATION: In 2011 and 2013, clinical and basic sciences investigators at University of Rochester School of Medicine and Dentistry responded to an online survey nominating their research collaborators. Two study years were merged, and name lists were transformed into a collaboration network. Departments were classified into basic sciences (e.g. biochemistry) and clinical (e.g. urology). If respondent and partner were affiliated to different department classes, the collaboration was defined as translational. Multi-level GLM models were developed to assess the associates of the likelihood of translational vs. within discipline collaborations. Partner nominations were nested in respondents. RESULTS/ANTICIPATED RESULTS: 202 respondents were included in the multi-level GLM models. A collaboration was more likely to be translational if the respondent shared more collaborators with the partner (OR:1.13), and respondent was a central actor in collaboration network (OR: 1.2). Translational collaborations were less likely to be reported by clinicians (OR: 0.25). In the model to assess gender match, a collaboration was more likely to be translational if the respondent was male, and nominated a male partner. For both genders, collaboration with a partner of the opposite gender was more likely to be translational if respondent had more shared collaborators with the partner. DISCUSSION/SIGNIFICANCE OF IMPACT: Translational collaborations happen in teams. Gender homophily exits in translational collaborations, and is reduced by shared collaborators; implying the effect of personal connections and community membership. Community-building interventions may increase diversity in translational collaborations.

OBJECTIVES/GOALS: Vaginal delivery is typically avoided in extremely preterm breech fetuses due to the concern for head entrapment by the cervix. Development of a device to prevent head entrapment would be best addressed by a multidisciplinary approach incorporating engineering principles with clinical obstetrics. METHODS/STUDY POPULATION: Construction of a collaborative multidisciplinary team to address the clinical challenge of preventing head entrapment was initiated through a unique course at the Massachusetts Institute of Technology (Course 2.75, Medical Device Design). The course would provide a structured means by which students (senior undergraduate and graduate students in Mechanical Engineering) would be paired with a clinical advisor and faculty in their department. Weekly team meetings were scheduled to review the clinical context pertinent to the problem and review engineering principles needed to develop a solution. The course also provided a small monetary budget ($4K) for the students to purchase supplies. RESULTS/ANTICIPATED RESULTS: During the semester long course, several iterations of a prototype were designed. Each subsequent rendition was evaluated from both an engineering and manufacturing perspective, as well as clinical appropriateness. The weekly meetings allowed for rapid re-design and assured that all necessary parameters were considered by the entire team. Students also had access to lab facilities and additional mentorship that allowed for supplementary input beyond that generated by core team members. These interactions, along with those of their classmates working on other projects, provided a strong base for exploring subsequent device development. DISCUSSION/SIGNIFICANCE OF IMPACT: Successful medical device development requires a collaborative process and students can be ideal members of these teams as they reside in an environment that is conducive to exploration and novel idea generation. Course-based student led team science platforms can provide an excellent foundation for solving uniquely challenging medical problems.

OBJECTIVES/GOALS: This team science pilot program aims to elevate the quality of Parkinson’s disease modeling initiatives by strengthening connections between clinical researchers and computational teams. As many data science projects in Parkinson’s research would benefit from deeper clinical expertise, many clinical engagements would be improved by upfront integration of computational requirements. These team science programs, developed from design thinking methodologies, provide structured, sustainable, and scalable means for multi-disciplinary teams to come together and co-create translational science in PD. METHODS/STUDY POPULATION: Design Thinking (DT) could help yield an effective learning experience. DT is grounded in ethnographic research strategies and prototyping, relying heavily on grantee interviews and feedback. This approach is commonly used to navigate and design amidst complexity; its applications range from product to healthcare to instructional design. The following is an overview of the process as applied to this project: Discover: Once the core team (MJFF and project designers) has refined the key question they would like to answer, the team will begin gathering both primary and secondary data. This phase may include focus groups, one-on-one interviews, expert interviews, and immersive data-gathering. The purpose of this phase is to capture complexity and lay the groundwork to understand grantees’ perspectives and lexicon around their work. The deliverables of this phase are primarily unstructured research findings, such as transcribed interviews and secondary sources. Define: When sufficient data has been gathered, the core team will move into an initial round of synthesis and sense-making (making connections and assumptions to explain emerging themes in the data). This phase may include one to two in-person engagements with the core team. The purpose of this phase is to define the guiding principles for subsequent prototypes. It will also help reveal potential opportunity areas, both latent or apparent. The deliverables of this phase are agreed upon key themes, insights, and an informed “How Might We” question that will anchor the ideation process. Develop: Armed with informed themes, the core team will begin to brainstorm potential solutions. Following a set of brainstorming techniques, they will initially aim for quantity versus quality in order to allow potentially innovative and/or risky solutions to surface. Eventually, these ideas will be distilled into three robust and unique prototypes. Like the prior phase, ideation may also require one to two in-person engagements. The deliverables here are three unique prototypes; the reason for three is the ensure that the team does not anchor themselves in just one solution, but rather remains in an exploratory mindset as they solicit feedback on these prototypes from the grantees. Deliver: In this final phase, the core team revisits the grantees and presents the three prototypes. This phase may include conducting three small-scale pilots or simply just explaining the prototypes. Either way, it is important to solicit another round of feedback to ensure the solutions are indeed addressing the needs and context of grantees. Once completed, the core team will iterate a final pilot design and identify any remaining questions and assumptions they would like the pilot to inform. RESULTS/ANTICIPATED RESULTS: The team science pilot identifies five main opportunities to tighten collaboration, communication, and expectations across clinical and computational teams. Firstly, in-person events, held regularly in a central location, can act as an incubating space for these teams to partner, ideate, and pitch for grant funding. Secondly, co-developed guidelines for research questions would ensure consistent availability of clinically-relevant, computationally-feasible research topics. Thirdly, increasing the presence of Parkinson’s cohort data resources at computational conferences could introduce more diverse data and genetics interest in Parkinson’s research. Fourthly, a standard suite of research-facing, educational content (focused on both disease background and data basics) would ensure a strong baseline and launch-pad for PD modeling projects. Lastly, a fellowship program focused on early-stage researchers could establish a unique foundation to ground both clinical and computations fellows to collaboratively work on PD research as well as iterate on the aforementioned solutions. DISCUSSION/SIGNIFICANCE OF IMPACT: This team science program has the potential to upend collaborative silos in Parkinson’s research, accelerating disease modeling projects which otherwise stagnate or over-emphasize clinical v. computational aspects. By more effectively connecting team members with diverse backgrounds across clinical and computational roles, PD disease patterns can be discovered and validated ultimately resulting in improved patient care and therapeutic development. CONFLICT OF INTEREST DESCRIPTION: Several authors are staff members at The Michael J. Fox Foundation for Parkinson’s Research, the sponsor of this Team Science grant. All author and non-author contributors are grant recipients from The Michael J. Fox Foundation.