Increasing emphasis on the use of real-world evidence (RWE) to support clinical policy and regulatory decision-making has led to a proliferation of guidance, advice, and frameworks from regulatory agencies, academia, professional societies, and industry. A broad spectrum of studies use real-world data (RWD) to produce RWE, ranging from randomized trials with outcomes assessed using RWD to fully observational studies. Yet, many proposals for generating RWE lack sufficient detail, and many analyses of RWD suffer from implausible assumptions, other methodological flaws, or inappropriate interpretations. The Causal Roadmap is an explicit, itemized, iterative process that guides investigators to prespecify study design and analysis plans; it addresses a wide range of guidance within a single framework. By supporting the transparent evaluation of causal assumptions and facilitating objective comparisons of design and analysis choices based on prespecified criteria, the Roadmap can help investigators to evaluate the quality of evidence that a given study is likely to produce, specify a study to generate high-quality RWE, and communicate effectively with regulatory agencies and other stakeholders. This paper aims to disseminate and extend the Causal Roadmap framework for use by clinical and translational researchers; three companion papers demonstrate applications of the Causal Roadmap for specific use cases.

OBJECTIVES/GOALS: To investigate nursing homes (NHs) acquired by private equity (PE) firms and estimate the effect of PE NH acquisitions on NH care quality for NH residents and whether PE NH acquisitions exacerbate racial inequities in quality of care. METHODS/STUDY POPULATION: My research estimates the causal effect of NH PE acquisitions on NH access and quality of care for NH residents and whether acquisitions exacerbate racial disparities in NH care for about 115 PE-owned NHs in the US, measuring the quality of care at the facility survey year level, and using an array of NH-level data sources. I identified 115 PE-owned NHs (treatment group) and 665 non-PE-owned NHs between 2003 and 2010, using the Online Survey Certification and Reporting database to obtain facility characteristics. I compare facility characteristics (e.g., payer mix, staffing levels, and quality measures such as pressure ulcers, unexpected weight loss, acuity, and health deficiencies). I will then test whether effects differ by race, with hypotheses informed by Public Health Critical Race Praxis approach. RESULTS/ANTICIPATED RESULTS: Preliminary results show that staffing levels differ between PE and non-PE-owned NHs in a way that aligns with a shift in focus toward the Medicare population i.e. short stays. We also find that deficiencies increased in PE-owned NHs compared to non-PE-owned NHs. We expect that PE acquisitions may lead to slightly widening racial disparities in NH care quality. Results may show that PE-owned NHs have a higher share of low-rated, high-BIPOC facilities. In weak markets, PE-owned NHs may have a higher share of BIPOC residents compared to highly competitive markets. This is because PE managers may prioritize cost over quality by cutting services. However, since quality measures are self-reported, except for measures related to deficiencies, this predicted lower quality of care may not be evident in observed data. DISCUSSION/SIGNIFICANCE: Understanding how PE ownership impacts nursing home care quality for long-stay residents, especially those funded by Medicaid, can help develop intervention strategies to effectively mitigate racial inequities in NH care, as Medicaid funded NH residents are more likely to be Black, Indigenous, and people of color.

OBJECTIVES/GOALS: Our objective is to assess whether street race is a stronger predictor of sleep quality and sleep hours than self-reported race. We also seek to understand whether the association between race and sleep quality/hours can be explained by experiences of microaggressions. METHODS/STUDY POPULATION: This study uses data from the National Couples’ Health and Time Study (NCHAT), a population-representative sample of 20–60 year-olds (N=3,642) who were married or cohabiting during 2020-2021 when the COVID-19 pandemic disproportionately negatively impacted racial and ethnic minorities in the U.S. (Boserup et al., Yip et al.). During this time, incidents of racial trauma increased (Tessler et al.). Using NCHAT data we examine whether street race is a stronger predictor of sleep quality and sleep hours than self-reported race. We also seek to understand whether the association between race and sleep quality/hours can be explained by experiences of microaggressions. RESULTS/ANTICIPATED RESULTS: Results show that microaggressions mediate the link between identifying as Black and being perceived as Black or Asian and sleep quality/hours. Identifying as Black and being perceived as Black or Asian, compared with non-Latinx White respondents, is associated with more frequent microaggressions. More microaggressions are associated with poorer sleep quality and fewer sleep hours. Asian street race is a marginally better predictor of microaggressions than self-reported race. In all models, microaggressions are associated with poorer sleep quality and less sleep hours. DISCUSSION/SIGNIFICANCE: With a growing non-white population, the wellbeing of our future generations is in everyone’s best interest. Poor sleep increases the risk of cardiovascular disease, diabetes, obesity, and cancer. The United States spends $93 billion in excess medical care costs due to health disparities.

Knowledge graphs have become a common approach for knowledge representation. Yet, the application of graph methodology is elusive due to the sheer number and complexity of knowledge sources. In addition, semantic incompatibilities hinder efforts to harmonize and integrate across these diverse sources. As part of The Biomedical Translator Consortium, we have developed a knowledge graph–based question-answering system designed to augment human reasoning and accelerate translational scientific discovery: the Translator system. We have applied the Translator system to answer biomedical questions in the context of a broad array of diseases and syndromes, including Fanconi anemia, primary ciliary dyskinesia, multiple sclerosis, and others. A variety of collaborative approaches have been used to research and develop the Translator system. One recent approach involved the establishment of a monthly “Question-of-the-Month (QotM) Challenge” series. Herein, we describe the structure of the QotM Challenge; the six challenges that have been conducted to date on drug-induced liver injury, cannabidiol toxicity, coronavirus infection, diabetes, psoriatic arthritis, and ATP1A3-related phenotypes; the scientific insights that have been gleaned during the challenges; and the technical issues that were identified over the course of the challenges and that can now be addressed to foster further development of the prototype Translator system. We close with a discussion on Large Language Models such as ChatGPT and highlight differences between those models and the Translator system.

OBJECTIVES/GOALS: 1. Describe method of Community Based System Dynamics. 2. Describe CBSD as used in addressing structural racism in a previously redlined community. 3. Review example causal loop diagrams and simulation models developed using CBSD focused on structural racism as a social determinant of health. METHODS/STUDY POPULATION: Community Based System Dynamics (group model building and computer simulation development). Cuyahoga County, Ohio, a previously redlined community of more than 1.2 million people. RESULTS/ANTICIPATED RESULTS:Actionable Community identified leverage points for action to mitigate structural racism. Computer simulation models built on causal loop diagrams built by Community members with loved experience. DISCUSSION/SIGNIFICANCE: Structural and social determinants of health, such as racism, have profound impacts on the health of individuals and populations, however they remain challenging to address in pragmatic ways. CBSD is a novel method to engage community members not proximal to the impact of structural racism in generating maps of the complex, dynamic system they live in.

Precision Medicine/Health

The National Institutes of Health’s (NIH) K99/R00 Pathway to Independence Award offers promising postdoctoral researchers and clinician-scientists an opportunity to receive research support at both the mentored and the independent levels with the goal of facilitating a timely transition to a tenure-track faculty position. This transitional program has been generally successful, with most K99/R00 awardees successfully securing R01-equivalent funding by the end of the R00 period. However, often highly promising proposals fail because of poor grantsmanship. This overview provides guidance from the perspective of long-standing members of the National Heart, Lung, and Blood Institute’s Mentored Transition to Independence study section for the purpose of helping mentors and trainees regarding how best to assemble competitive K99/R00 applications.

OBJECTIVES/GOALS: Overall response rates for metastatic uveal melanoma (UM), regardless of treatment modality, are less than 20%, highlighting an urgent need for novel therapies. Herein, we present a UM patient-derived organoid (PDO) biobank as a novel platform for translational research. METHODS/STUDY POPULATION: Patients with primary choroidal or ciliochoroidal UM undergoing enucleation from 7/1/2019-9/30/2022 were invited to enroll. Tumor tissue was harvested within 30 minutes of globe removal. Cells were isolated using the human tumor isolation kit and gentleMACS dissociation protocol (Miltenyi Biotech). PDOs were placed on Cultrex-coated multiwell plates and cultured in supplemented RPMI media. DNA and RNA were isolated using kits from Zymo Research. Exon-enriched libraries and RNA were sequenced using an Illumina HiSeq 4000. Immunohistochemistry (IHC) assessed the following histone post-translational modifications: H3K4me1/3, H3K27Ac, and H3K27me. RESULTS/ANTICIPATED RESULTS: PDOs were established in 19 of 20 (95%) attempted cases. BAP1 protein expression was retained (n=7) or lost (n=12) in the primary tumors, with matching phenotype confirmed in PDOs. In 9 sequenced cases, a driving mutation was present in GNAQ (n=4), GNA11 (n=4), or CYSLTR2 (n=1). Morphology ranged from spindle-like to epithelioid clusters, mimicking primary tumor histopathology. Pigmentation increased with time in culture. Growth in culture was slow, and 1-2 months were allotted prior to passaging in most cases. Whole exome and RNA-sequencing confirmed distinct molecular profiles, with differential staining of active chromatin marks by IHC. DISCUSSION/SIGNIFICANCE: A biobank of primary UM PDOs with unique morphological and molecular characteristics has been established. These will serve as a model of human disease to facilitate translational research and investigate personalized treatments for patients with UM.

Advancing the new field of translational science and developing innovative solutions to overcome translational roadblocks are key priorities of the Clinical and Translational Science Awards (CTSA) Program of the National Center for Advancing Translational Science (NCATS). However, interpreting this emerging concept of “translational science” (TS) as a field of inquiry distinct from “translational research” (TR) and developing real-world investigations in TS can be challenging. The goal of this paper is to share the obstacles the Einstein-Montefiore CTSA hub has faced in generating institutional interest and research in TS and to present potential strategies for addressing them. The aim is to stimulate dialog within the wider CTSA community and beyond about the need to systematically examine how TS should be efficiently and effectively pursued, that is, the science of translational science. The collective sharing of experiences and innovative approaches to overcoming TS challenges that arise at CTSA hubs is critical if the field is to grow and gain wider recognition and acceptance by the scientific and broader communities.

OBJECTIVES/GOALS: Pharmacogenomic testing for major depressive disorder is an expanding area of precision medicine with multiple direct-to-provider tests available. While clinical evidence supporting testing is expanding, there has been little research on the views and experiences of patients and clinicians utilizing this novel intervention. METHODS/STUDY POPULATION: This ongoing study is conducting semi-structured interviews with clinicians and patients exploring their views of the benefits and limitations of pharmacogenomic testing. Qualitative interviews have been conducted with 10 patients and 10 clinicians who have experience with ordering or receiving results within the past 12 months. Interviews are being thematically coded following a modified grounded theory approach using the Dedoose software. Following the principles of exploratory sequential mixed methods design, findings will be used to develop a survey to be administered to prescribing clinicians in both primary care and psychiatry. The survey will examine clinician's knowledge, interest, and concerns about utilizing testing. RESULTS/ANTICIPATED RESULTS: Preliminary analysis of qualitative interviews indicates that both patients and clinicians find that the broader testing process has benefits beyond the test results themselves. Benefits identified by patients include an increased trust in the process of selecting medications, validation of their negative experiences with medications, and improved communication with their provider. Limitations identified by patients include difficulty in accessing test results, and gatekeeping for testing by providers. Benefits identified by clinicians include increased empathy with patients, medication adherence, and improved communication with patients about medication. Limitations identified by clinicians include difficulty with ordering and interpreting test results. DISCUSSION/SIGNIFICANCE: Medication selection is a difficult process for both patients and clinicians. Improvements to clinician-patient communication and medication adherence are important benefits to consider in the adoption of testing. Future research should include these dimensions in assessment of the benefits and limitations of testing.

OBJECTIVES/GOALS: We leverage the disparate incidence of Ewing sarcoma (ES) between European (EUR) and African (AFR) ancestry to study ES tumorigenesis in iPSC-derived cells from donors with a range of AFR ancestry via functional / molecular profiling. Integrated multi-omics analysis furthers explore local regulatory networks in pursuit of novel drug targets of ES. METHODS/STUDY POPULATION: In our pilot, eight induced pluripotent stem cell lines were obtained, differentiated into neural crest cells, and then transduced with a lentivirus expressing GFP-2A-EWS/FLI1. We compared wild type (WT) to EWS-FLI1-induced cells and then compared cell survival, gene expression, and EWS-FLI1 binding differences at varying levels of EUR / AFR ancestry admixture. We will build on this pilot data by expanding the number of cell lines and measuring chromatin state. Subsequently we will refine our understanding of the relationships between local ancestry, epigenetic and gene expression changes, and phenotype in tumor progression via integration of multi-omics datasets. Our systems genomics approach will utilize directed local regulatory networks in a Bayesian structure learning framework. RESULTS/ANTICIPATED RESULTS: Induction by EWS-FLI1 resulted in gene expression changes enriched in known ES gene sets. Higher %EUR ancestry correlated with prolonged maintenance of EWS-FLI1. We identified thousands of ancestry-linked changes to gene expression and EWS-FLI1 binding. Eighty of these genes are both differentially expressed and differentially bound based on AFR ancestry admixture level and may be some of the early critical targets that initiate the cascade of molecular changes in ES. We will identify novel drug targets, with potential cross functional use of known drugs. Once we have developed directed local regulatory networks, we will use them to test in silico potential perturbations due to small molecules or novel drugs and predict expression changes. DISCUSSION/SIGNIFICANCE: With a limited number of cell lines, we identify 80 ancestry-linked candidate loci for functional validation through genome engineering. As EWS-FLI1 itself has proven elusive to direct targeting, studying its immediate downstream effects has the potential for establishing new druggable biologic pathways for treatment of ES.

LGBTQIA2+ patients often experience discrimination and hostility in healthcare spaces. Negative perceptions of healthcare can contribute to poor health outcomes in LGBTQIA2+ patients. This population is rarely included in clinical trials through a lack of inclusion in study protocols, informed consent, and trials not addressing their needs and demographics. Many clinical institutions have created LGBTQIA2+-specific clinics; however, few have successfully developed a free clinic dedicated to this population. A Rainbow Clinic was formed at an established student-run free clinic, utilizing the existing infrastructure. Dissemination of this clinic’s creation can help others replicate similar initiatives.

REAP-2 is an interactive dose-response curve estimation tool for Robust and Efficient Assessment of drug Potency. It provides user-friendly dose-response curve estimation for in vitro studies and conducts statistical testing for model comparisons with a redesigned user interface. We also make a major update of the underlying estimation method with penalized beta regression, which demonstrates great reliability and accuracy in dose estimation and uncertainty quantification. In this note, we describe the method and implementation of REAP-2 with a highlight on potency estimation and drug comparison.

OBJECTIVES/GOALS: Familial Hypercholesterolemia (FH) is a common disorder that is vastly underdiagnosed and causes an increased risk for sudden cardiac death. Cardiology providers (CHCPs) are in an ideal position to care for patients with FH. This research aimed to understand the practice behaviors of CHCPs in the screening, diagnosis, and management of FH. METHODS/STUDY POPULATION: An explanatory mixed methods design was utilized for this study. Adaptation of an existing FH knowledge tool guided survey development. The results of the quantitative survey, along with the Knowledge to Action framework and Theory of Planned Behavior, guided development of the interview protocol. Convenience and snowball sampling recruited CHCPs in the Division of Cardiology at Columbia University Irving Medical Center (CUIMC). Descriptive statistical analysis was performed on survey data. Qualitative interviews were conducted with survey respondents who volunteered to participate. Interviews were audio recorded, transcribed, and analyzed thematically. A descriptive review of the educational materials offered by the Division of Cardiology was conducted to identify FH knowledge domains presented. RESULTS/ANTICIPATED RESULTS: CHCPs with MDs, at CUIMC for 6-10 years, in clinical practice for 1-5 years, and in Inpatient Services had the highest average total FH knowledge scores. CHCPs with RNs, at CUIMC for less than 1 year, in clinical practice for 6-10 years, and in Cath Lab had the lowest average FH knowledge scores. Twenty interviews were completed, and four themes emerged- variability in FH care; issues related to addressing FH at institutional, practice setting and individual levels; importance of identifying FH early; and intervention approaches to overcome barriers to caring for FH patients in cardiology. CHCPs with MDs or with experiential FH knowledge were the only CHCPs to describe FH care beyond the point of screening. The document review revealed that only MDs were provided four lectures over the course of 4 years pertaining to FH. DISCUSSION/SIGNIFICANCE: CHCPs with didactic or experiential FH knowledge provided care beyond screening. Future interventions should increase didactic and experiential FH knowledge by incorporating institutional, local, and national FH resources. Improving the FH care CHCPs provide, can reduce FH-related morbidity and mortality as well as improve FH health outcomes.

OBJECTIVES/GOALS: The effect of the Direct-Acting Antiviral (DAA) on hepatic histopathological features from patients treated for HCV has not been thoroughly evaluated. The goals of this retrospective study were to determines differences between the liver biopsies collected before and after DAA treatment and correlated the histopathology with clinical outcome. METHODS/STUDY POPULATION: Spectral imaging was used to evaluate differences in intrahepatic macrophage (CD68, CD14, CD16, MAC 387, and CD163) and T cell (CD3, CD4, CD8, CD45, and FoxP3) phenotypes in paired liver biopsies collected from the same patient before (n=10) and after (n=10) achieving SVR (Figure 1). Imaging analysis and machine learning algorithms were used to evaluate changes in these key immune cells. We also compared differential gene expression of over 700 genes using RNA isolated from liver biopsies with NanoString. RESULTS/ANTICIPATED RESULTS: Multispectral imaging analysis showed a significant increase of proinflammatory/M1-like (e.g., CD14+) and anti-inflammatory/M2-like macrophage (e.g., CD163+) phenotypes in pre-treatment versus post-treatment biopsies, respectively. Gene expression analysis revealed enrichment of inflammatory (HLA-B, STAT1, CXCL10) and interferon induced-antiviral (ISG15, OAS3, MX1 and IFIT1) genes in the pre-treatment vs the post-treatment group. Cell deconvolution analysis also showed a significant increase of M1-like macrophages in the pre-treatment group when compared to the post-treatment group or controls. Upregulation of genes associated with cell proliferation and differentiation (c-KIT and Fos) was observed in the post-treatment biopsies of patients with persistent inflammatory infiltrates. DISCUSSION/SIGNIFICANCE: Protein and gene expression profiles observed in patients before DAA therapy showed a strong macrophage-mediated inflammatory response against HCV infection in the liver that shifted significantly to a tissue remodeling microenvironment after treatment.

Despite the availability of vaccinations, severe acute respiratory syndrome coronavirus 2 (SARS-CoV-2) continues to cause Coronavirus Disease 2019 (COVID-19) infection with a spectrum of disease in the acute setting. Transmission, infection, and severe disease remain common. There is a critical need to establish treatment regimens in the ambulatory setting that can reduce symptom burden and potentially prevent progression to severe disease and death. Many existing medicines previously approved for other uses may have benefit but remain unproven in informative clinical trials.

Accelerating COVID-19 Therapeutic Interventions and Vaccines (ACTIV)-6 is a decentralized, placebo-controlled, double-blind, randomized, platform trial that has now enrolled more than 7500 participants and has reported on the effectiveness of ivermectin at two doses, fluticasone, and fluvoxamine for helping people with COVID-19. With additional repurposed therapies added to the platform, ACTIV-6 continues to enroll symptomatic outpatients aged ≥ 30 years with a confirmed positive PCR or antigen test for SARS-CoV-2. Potential participants are screened and enrolled online, through a call center, or facilitated by local study sites. Participants consent electronically and are randomized to placebo or to one of the open study drugs for which they are eligible at the time of enrollment. A shared, contemporary placebo approach is used. Participants receive study drug in the mail and remain on study for up to 180 days. While enrolled, electronic patient-reported outcome assessments are used to monitor symptoms, healthcare utilization, and mortality. The primary endpoint is time to recovery or a composite of hospitalization and mortality within 28 days. Symptoms, acute healthcare utilization, and the Patient-Reported Outcomes Measurement Information System-29 are collected for up to 180 days.

Using a decentralized trial approach allowed the ACTIV-6 platform to increase both reach and rate of enrollment. The decentralized approach did not simplify regulatory oversight, and we found unanticipated challenges in patient behavior and the study drug delivery process. Despite challenges, ACTIV-6 has enrolled thousands of participants from across the USA and continues to test the effectiveness of repurposed medicines for treating COVID-19. Our lessons learned contribute to the emerging understanding of how to optimize decentralized trials.

OBJECTIVES/GOALS: In this study, the ability of a pathologist to detect malignancy on digital pseudo-H&E slides obtained via structured illumination microscopy (SIM) imaging of fresh diagnostic breast biopsies was assessed. The speed of imaging and processing was also assessed for potential clinical implementation. METHODS/STUDY POPULATION: This study was conducted in accordance with an Ochsner Medical Center of New Orleans IRB. 200 patients undergoing either stereotactic or ultrasound-guided diagnostic breast biopsies were consented and an additional core from the suspicious lesion was collected for research use. Research biopsies were transported to the lab and stained with DRAQ5 and Eosin-Y and imaged with SIM before being submitted for histology processing. Imaging and digital processing times were recorded. The resulting SIM images and histology slides were given to a pathologist for blind review to assess accuracy. RESULTS/ANTICIPATED RESULTS: The ex-vivo structured illumination microscopy images and subsequent histology slides from 79 research cores have been assessed to date. Some samples were excluded from the total data set and not included in the final assessment due to technical failures of the imaging protocol. Of the current set, the pathologist has a specificity of 88% and a sensitivity of 65%, as well as an NPV of 88% and a PPV of 65%. Staining time for each biopsy was completed within 3 and a half minutes and imaging at 20x magnification took between 4 and 12 minutes, depending on size and implementation of autofocus to the imaging system. Image processing took approximately 5 minutes per biopsy and is a direct function of biopsy size. DISCUSSION/SIGNIFICANCE: Decreased time between cancer suspicion and treatment will improve the prognosis of breast cancer patients. SIM imaging of fresh breast biopsies could ultimately allow primary and secondary histology to be performed simultaneously and minimize histopathology time, thus allowing clinicians and patients to implement treatment course more quickly.