University Of Pennsylvania

NIH Research Projects · FY 2025 · 2024-01

TRANSLATION OF NONINVASIVE METABOLIC BIOMARKERS TO TARGETED THERAPY FOR CANCER PROJECT SUMMARY/ABSTRACT The genetic mutations that make a living cell become cancer produce faulty signaling cascades that alter the cell’s metabolic processes and metabolites. This metabolic reprogramming, called the Warburg effect, is present in most cancers. It is demonstrated primarily by increased lactate formation from glucose, despite proper oxy- genation and functional mitochondria. In addition, macromolecule precursors favoring cell growth and division are also increased (e.g., alanine for protein synthesis). Notably, new cancer therapies target these faulty cas- cades. Thus, successful treatment with these targeted therapies must modify the metabolite content in the tumor. For example, Bruton’s tyrosine kinase (BTK), hyperactive in mantle cell lymphoma (MCL), alters the B-cell receptor signaling cascade. Thus, the hyperactive BTK favors cancer processes, including the Warburg effect. Therefore, BTK inhibition (BTKi) is a potential treatment choice for MCL. Our preclinical studies showed lactate and alanine reductions in BTKi-sensitive human MCL cell lines in cul- ture and MCL xenografts. In comparison, MCL models unresponsive to BTKi do not show these metabolic changes. Notably, these metabolic changes during effective BTKi therapy precede the objective evidence of tumor response to therapy (i.e., reduction of viability in cultured cell lines or tumor volume in xenografts). Our overall goal is to assess the clinical translatability of our preclinical studies. Thus, our central hypothesis states that the altered tumor content of lactate and alanine in BTKi-sensitive tumors of MCL patients precede the reduction of tumor burden, supporting their potential use as predictive biomarkers of response. However, to test our hypothesis, we aim to assess the feasibility of measuring lactate and alanine in tumors of MCL patients. Our group has helped develop and improve a robust magnetic resonance spectroscopic imaging method, the Hadamard-selective multiple quantum coherence (Had-Sel-MQC). We adapted this method to measure lactate and alanine selectively, making it a simplified evaluation of the Warburg effect. Additionally, the Had-Sel-MQC anatomically localizes lactate and alanine, allowing their noninvasive assessment of tumors in their place in the body (i.e., in situ). Furthermore, we have transferred the Had-Sel-MQC to clinical whole-body magnetic reso- nance imagers and recently improved it for its application to human subjects (Lee, 2022, doi: 10.1088/2057- 1976/ac57ad, and Lee, 2022, doi: 10.1016/j.mri.2022.08.020). We will study 15 healthy volunteers and 30 MCL patients in this feasibility project. Our specific goal is to ensure that the MRSI exam achieves proper localization of tumor signals with the best spectral quality (i.e., improved B0-shimming and water and lipid suppression) irrespective of the body part studied. In addition, we will pay special attention to tumors in the torso to assess how cardiac and respiration movements affect the exam and determine the best way to minimize their effect.

NIH Research Projects · FY 2026 · 2024-01

Developing PARP-1 PET with companion tissue assay as a precision tool to guide PARPi therapy PROJECT SUMMARY When we give pharmaceuticals to patients with cancer, they deserve the best predictive measures we can employ to tell them whether the therapy has a significant likelihood of killing the tumor and extending their life. No one should have to suffer drug toxicity without benefit. If this proposal is successful, we will develop a noninvasive predictive biomarker of sensitivity to PARP inhibitors (PARPi) that could potentially be used for real- time dose modulation as well as patient selection. While our investigations will impact PARPi in any tumor type, we largely focus on breast cancer where PARPi are currently used in the setting of metastatic disease and the adjuvant setting. Here, the ability to predict response would also provide a critically needed biomarker to stratify patients for clinical trials testing neoadjuvant PARPi/chemotherapy combinations, especially in the triple- negative breast cancer population. BRCA1/2 and HRD testing indirectly select tumors that may be susceptible to PARPi, leveraging the concept of synthetic lethality, but do not directly measure drug-target density. The development of imaging-tissue markers for PARPi binding and PARP-1 expression is innovative in the context that currently used companion diagnostics for PARPi use genomic assays. Preliminary data in breast and ovarian cancer show that assay of the target (PARP-1), binding by PARPi, and measures of drug-target engagement provide complementary information that appears to add predictive value. In this proposal, we will develop a correlative tissue assay to a promising PARP- 1 radiotracer, using imaging, immunohistochemistry, and in vitro radioligand binding assay (Aim 1), test the assay for predictive value in historical clinical trial sets from patients who received PARPi with known clinical outcomes (Aim 2), and directly evaluate the hypothesis that tracer uptake predicts response in animal models of disease (Aim 3). The validated imaging agent and tissue companion diagnostic would have planned clinical use similar to fluoroestradiol imaging/estrogen receptor assessment. Clinicians could avoid futile therapy in cases where the target is not expressed and also potentially identify additional patients who would benefit from PARP-1 targeted therapy and combination strategies. Precision medicine is within our reach and needs to be employed for aggressive breast cancers. If successful, this research could lead to a new way to functionally characterize breast cancer, guide targeted therapy selection, and assess therapeutic response and resistance. The new imaging probe is immediately applicable to patients at our center and production has been standardized to facilitate multi-center use.

NIH Research Projects · FY 2026 · 2024-01

PROJECT SUMMARY/ABSTRACT Food insecurity—the economic and social condition of inconsistent access to healthy food—is associated with worse cardiovascular health. Whether food assistance related social policies can improve cardiovascular health among low-income adults in the US is unknown. The Supplemental Nutrition Assistance Program (SNAP) is the largest US government-funded program providing food assistance. In 2021, at the cost of approximately $114 billion, 41.5 million Americans received SNAP benefits. As SNAP is administered by individual states, there are variations in policies related to eligibility assessment and ease of application that significantly impact participation levels. Although SNAP participation has been associated with improved food security, poverty reduction, and lower healthcare expenditures, it's impact on cardiovascular health is unknown. The mechanisms by which SNAP, and poverty reduction programs in general, influence cardiovascular health is unknown, but may be mediated through mechanisms such as medication adherence and psychosocial stress. In a previously published analysis, we found that implementation of state level policies associated with higher SNAP participation was associated with slower growth in diabetes prevalence. However, whether SNAP participation, or the policies impacting this participation, impacts cardiovascular health at an individual level is unknown. During the COVID-19 pandemic, the amount of SNAP benefits received by participants increased, but some of these increases are scheduled to expire. How SNAP benefit amounts impact cardiovascular health related measures is therefore important to understand. Aim 1 of the proposed research will use data from a nationally representative survey of older adults – the Health and Retirement Study (HRS) – linked with Medicare or Medicaid administrative claims data to identify whether SNAP participation among low-income individuals with cardiovascular disease is linked with improvements in medication adherence and psychosocial stress as well as lower healthcare utilization. Aim 2, using Medicare and all-payer claims data, will examine whether changes in state-level SNAP policies associated with SNAP participation lead to changes in the incidence of acute cardiovascular events among low-income adults. We will also study whether specific policy categories are associated with a greater change in health outcomes. In Aim 3 we will examine how changes in SNAP benefit amounts influence cardiovascular health-related measures. We will examine the temporary increase in SNAP benefit amounts that occurred nationwide from 2009 to 2013 as a result of the Recovery Act of 2009 and evaluate whether these changes resulted in differences in medication adherence, healthcare utilization, and health outcomes among SNAP participants with cardiovascular disease. The proposed research aims will require robust causal inference techniques to disentangle the effect of SNAP from other potential confounders. This research will provide important insights to policymakers regarding the potential public health benefits of investments in food assistance programs.

NIH Research Projects · FY 2025 · 2024-01

PROJECT SUMMARY/ABSTRACT Language learning is a critical aspect of children’s development, laying the groundwork for future social and academic success. Understanding how language learning begins in infancy has long been a key project of developmental psychology and cognitive science. However, current theories of language acquisition make unrealistic assumptions about the nature of speech, infants’ entry point to their future native language. Current accounts assume that words in speech can be translated into consistent phonological (sound) representations by infants and adults alike. In reality, how a speaker pronounces a given word is fundamentally variable, so much so that many pronunciations, even in infant-directed speech, are unintelligible on their own and can only be understood by adults because adults can leverage their top-down linguistic knowledge about the surrounding sounds and words. The proposed studies represent one of the first forays into how infants manage this fundamental pronunciation variability while they are learning the system itself. Using a multimethod approach, the proposed research investigates the hypothesis that infants rely on temporal context to interpret reduced pronunciation variants. In normal discourse, including speech to children, first uses of words are often more clearly spoken than following uses. If infants can connect clear uses and closely- following reduced forms, this could help them recognize instances of words that would be unintelligible on their own. Characterizing this process will help make it possible to create better models of how infants begin to connect the speech signal to meaning. Study 1, a corpus analysis, analyzes the language input that parents provide by asking how often parents produce reduced pronunciation variants in different contexts and how these contexts line up with ideal word- learning opportunities. Studies 2 and 3 use eye-tracking procedures to experimentally investigate infants’ initial encoding of novel words and recognition of familiar words given naturalistic pronunciation variability. Together, these three studies will help guide scientific theories of word learning by providing new insight into this unsolved problem in children’s language acquisition and demonstrating powerful new methods for future research on how young children handle the messy reality of spoken language. The proposed research will provide the applicant with the ideal training to advance her career, empowering her to expand her previous work into a full research program, learn new computational and experimental methods, and gain critical professional development experience communicating her findings and mentoring students. These training goals are further supported by a strong institutional environment with many pertinent student resources and intellectual communities. This training will ultimately prepare the applicant to become a successful independent researcher.

NIH Research Projects · FY 2026 · 2024-01

Project Abstract In this proposal we will deploy state-of-the-art neural ensemble recordings across the reach and saccadic- attention cortices of NHP performing coordinated reach-and-saccade movements to test hypotheses about the multiregional brain mechanisms supporting coordinated visual behavior. There is growing evidence that an important way to conceptualize the behavioral significance of multiregional brain dynamics is as communication channels that mediate information transmission between sender brain regions and receiver brain regions. The communication channel allows populations of neurons in a sender region to dynamically influence the activity of populations of neurons in a receiver region and support behavioral flexibility. We channel the will coordinated seek to test the communication model of multiregional communication by asking how the model explains neural i nteractions between neurons in the saccade and reach regions that must communicate during coordinated visual behavior. We specifically investigate multiregional communication during behaviors in which visual fixation enhances reach and saccade movement performance.One way to demonstrate the importance of fixation to accurate reaching is to present a second target around the time of a coordinated look and reach or pointing movement, and instruct subjects to acquire the newly-presented target with a new saccadic eye movement. This sequence of two movements, the coordinated look-and-reach movement and the subsequent saccadic eye movement, reveals that the importance of target foveation to accurate reaching extends beyond the act of fixation. Accurate coordinated looking and reaching features a form of attentional suppression that inhibits saccades to newly-presented targets. As a result, gaze is said to be temporarily anchored to the target of the coordinated reach and, importantly, the longer the duration of fixation is extended, the more accurate the reach movement. Gaze anchoring therefore provides an opportunity to investigate the multiregional mechanisms of a flexible behavior in terms of communication between at least two separate systems in the brain - the reach system that controls the reach and the saccade system that is inhibited to suppress new saccades. We will first do experiments to test and elaborate the conceptual framework of communication channels for multiregional communication during gaze anchoring. Aims 1 and 2 will test complementary hypotheses involving subspaces and neural coherence for how communication between reach and saccadic-attentional cortex gives rise to coordinated behavioral performance. In Aim 3, we will then more closely examine the cognitive processes in play and select between alternative explanations for how coordinated visual behavior depends on attention. These experiments will advance understanding of how channel gain and modulation combine in the mechanisms of multiregional communication, will lead to new strategies for neural rehabilitation and neural prosthetic systems.

NIH Research Projects · FY 2025 · 2024-01

Project Abstract: Diabetic retinal disease (DRD) is the leading cause of vision loss among working age adults costing the U.S. an estimated $4.5 billion per year and disproportionately affects racial and ethnic minorities.2,3 As with any pressing public health concern, tracking the prevalence and incidence of the disease is paramount to understanding the impact on society and creating successful policies to counteract it. The previously preferred method of assessing U.S. DRD rates has been regional population-based studies, but even the most recent of these has not been updated in over 10 years.6–10 The National Eye Institute’s currently published DRD rates are derived from expert consensus agreed upon in 2001.2,15 The issue of comparisons across time and regions becomes even more pronounced when discussing U.S. racial and ethnic disparities. While known disparities exist among minorities in DRD screening rates, severity of DRD and associated vision loss,6–9 the primary study used to quote rates of DRD in Black/African Americans is the Salisbury Eye Study. This was undertaken in the 1990’s and focused solely on people living in Maryland.19 Data on US Hispanics comes only from Los Angles and Arizona and have not been updated in 20 years.20,21 Lastly, relying on regional and outdated data ignores the last 20 years of national macro-environmental trends that likely have had a direct impact on DRD rates. How the conflicting trends of improved systemic DM care and increasing DM incidence have impacted the overall rates of DRD, the time to DRD progression and whether racial and ethnic disparities have changed has yet to be assessed.22 We hypothesize that while national DRD prevalence is increasing (due to more people living with the disease), DRD incidence is decreasing due to better systemic DM care. In addition, similar to the reduced DRD incidence rates, we hypothesize that due to new tools for systemic monitoring and treatment, the time to DRD progression is increasing (i.e. improving). We also hypothesize that despite the improvements in DRD incidence, DRD racial and ethnic disparities are continuing to widen, exacerbated by increasing disparities in underlying systemic DM incidence rates. The goal of this grant will be to create accurate, current national prevalence and incidence measurements of DRD. Furthermore, understanding the trajectory of the disease, particularly differences between racial and ethnic minorities will allow for both more effective policy creation tailored to the groups most in need and the tracking of the effectiveness of these policies by providing a baseline for future assessments.

NIH Research Projects · FY 2025 · 2024-01

Contact PD/PI: HALPERN, SCOTT PROJECT SUMMARY The University of Pennsylvania (Penn) proposes to establish the Patient-Oriented Research and Training to Accelerate Learning, or Penn PORTAL, to build and diversify capacity in deploying patient- centered outcomes research (PCOR) within LHSs. Penn PORTAL will integrate operations leaders, mentors who are expert in all Learning Health System (LHS) competencies, community partners, and all health systems in the City of Philadelphia with several leading LHS research centers at Penn. We will develop and test interventions that are informed by our expertise in changing human behavior and using the most rigorous and ethical comparative effectiveness and implementation science research methods. Penn PORTAL will be led by 3 physician-scientists with diverse personal backgrounds, clinical disciplines, career stages, and methodologic expertise: Scott Halpern, Meghan Lane-Fall, and Mucio (Kit) Delgado. These MPIs will lead PORTAL’s Administrative Core and be joined on its Steering Committee by 7 senior investigators and operations leaders in overseeing and synchronizing the work of a Research Education Core (REC), Research & Data Analysis Core (RDAC), and Equity & Community Engagement (ECE) unit that will ensure principles of diversity, equity, and inclusion pervade all PORTAL activities. The REC will be co-led by Jen Myers, Director of Penn’s Center for Healthcare Improvement and Patient Safety, and Rachel Werner, Director of its Leonard Davis Institute of Health Economics. Drs. Myers and Werner bring complementary expertise in training embedded scientists with system leadership roles and independently-funded LHS investigators, enabling us to train both types of scientists in a 2-year program blending didactic and experiential learning and intensive mentorship that we developed during our current AHRQ/PCORI K12 award. The RDAC will be co-led by Kevin Johnson, an internationally recognized leader in building and leading informatics-informed LHSs, and Michael Harhay, a statistical epidemiologist and expert in developing methods for pragmatic clinical trials. The ECE will be led by Jaya Aysola, Director of the Center for Health Equity Advancement – a first-of-its-kind health equity center embedded within a large health system enterprise. Led by the ECE, PORTAL will partner with the Philadelphia Department of Health, several community organizations, and all Philadelphia health systems, the latter through an initiative funded by the region’s largest payer to combat racism and health inequities across the City. These partnerships will diversify the pool of LHS scientists we train and amplify our opportunities to implement effective interventions across all Philadelphia-area health systems. Finally, guided by an Internal Advisory Board that includes the Deans of Penn’s Schools of Medicine, Nursing, and Business (Wharton), and an External Advisory Board that includes 11 national LHS leaders, PORTAL will realize the ideals of an LHS imbued with an equity-first focus.

NIH Research Projects · FY 2026 · 2024-01

PROJECT SUMMARY/ABSTRACT Cardiovascular diseases are not only the leading cause of mortality in the world, but they also lead to almost twice as many deaths in Black adults compared to White adults in the United States. Similar differences in disease prevalence and mortality exist for many other heart, lung, blood, and sleep (HLBS) disorders. To address factors that play a role in disease prevalence, we cannot overlook the lack of genetic heterogeneity that has perturbed existing genomic studies on which we base the development of precision medicine technologies. Of the ~6,300 studies currently compiled in the genome-wide association studies (GWAS) catalog as of March 2023, ~95% of all GWAS participants are of European (EUR) ancestry with less than 1% of participants being of African American (AFR) ancestry. Attempts to translate genetic research findings into clinical practice may be not just incomplete, but dangerously mistaken and misapplied. With the advent of large and heterogeneous patient biobanks, there needs to be a targeted focus on studying genetic variations specific to genetically heterogeneous populations. The Penn Medicine BioBank serves as an ideal discovery vehicle for this purpose with a repository of genotype, whole-exome sequencing, and electronic health record (EHR) data as well as one of the largest AFR populations at any single-institutional medical biobank in the US, to our knowledge. We hypothesize that by using a genome-first AFR-specific approach, we will identify significant genetic associations for HLBS disorders that would have otherwise been missed by previous EUR-dominant studies and future multiancestry approaches. We will curate a list of AFR-specific predicted loss-of-function and missense variants, and then through HLBS-focused phenome-wide association studies, we will characterize the clinical manifestations of the diseases caused by these protein-altering variants. Furthermore, genome-wide association studies will be used to identify genetic loci with significant varying effects across ancestries and uncover variabilities in HLBS disease prevalence. To investigate these ancestry-specific risk loci, we will employ local ancestry inference to study ancestry-dosage effects and construct haplotypes to uncover drivers of the observed phenotype associations. This proposed project has the potential to broaden our understanding of genetic risk for HLBS disorders and significantly impact the development of diagnostic and therapeutic methods. Beyond that, our work embodies the vision that the advancement of personalized medicine should be effective and benefit all individuals.

NIH Research Projects · FY 2026 · 2024-01

ABSTRACT This new T32 research training program will provide postdoctoral fellows training in sleep and circadian research. It builds on a prior highly successful postdoctoral training program and is part of a comprehensive succession plan for T32 leadership at the University of Pennsylvania. The fellows can have PhD, MD, VMD, or MD/PhD degrees. The program provides support for each fellow for up to three years to provide the training they require to reach the next level of career advancement. The program is broad in scope. It is organized into specific tracks, each of which has a number of faculty who can act as mentors. The specific tracks are as follows: a) basic research in sleep and circadian mechanisms; b) clinical/translational research in sleep and its disorders; c) personalized medicine approaches that include development of biomarkers, genetics of complex traits, and big data approaches using electronic health records (EHR). Each track has relevant didactic courses and there are multiple Masters degree programs that fellows can pursue. There are also organized programs to teach academic survival skills—grant writing, giving scientific talks, and negotiating for a position. The program also offers the ability to collaborate with faculty at other institutions including Jackson Laboratory (mouse genetics), Geisinger Clinic (big data, genetics), and Kaiser Southern California (big data approaches). Penn developed the first medical-school wide center to facilitate sleep/circadian research in 1991 and the first multidisciplinary Sleep Medicine Division in 2001. These developments have led to Penn training the largest number of successful investigators in sleep/circadian research of any institution in the world. The program takes advantage of the infrastructure that has been developed at Penn over the past 30 years for research training. This includes the Office of Biomedical Postdoctoral Programs (BPP), the Master of Science in Clinical Epidemiology, the Master of Science in Translational Research, and the Master of Science in Biomedical Informatics.

NIH Research Projects · FY 2026 · 2024-01

PROJECT SUMMARY/ABSTRACT Pulse oximeters are used in nearly every pediatric hospitalization to non-invasively monitor arterial blood oxygen saturation. Pulse oximetry data are essential to clinical decision-making, informing decisions from the use of supplemental oxygen to hospital admission to transfer to intensive care. Research indicates that pulse oximeters may overestimate arterial oxygen saturation in hypoxemic patients with dark skin colors more than in those with light skin colors. Prior studies have been limited by availability of data and use subjective skin color measurement methods. In some studies, self-identified Black race has been associated with diminished pulse oximeter accuracy, without measuring skin color at all. The majority of existing studies were conducted in adults; however, characteristics of children's skin, such as skin thickness and melanin concentration, may contribute to different results in children. The overall objective of this study is to develop a comprehensive understanding of the relationship between skin color and pulse oximeter accuracy in children in the clinical setting. The results may inform recommendations for clinicians as well as policies and action from industry and the Food and Drug Administration. This multisite, prospective, observational study will be conducted in a population of children under 18 years old who are undergoing cardiac catheterization. Children undergoing cardiac catheterization are a unique population in which to study pulse oximeter accuracy because they have a wide range of baseline arterial oxygen saturations secondary to their cardiac conditions, allowing for evaluation of pulse oximeter accuracy across a range of arterial oxygen saturations. Children undergoing cardiac catheterization are continuously monitored with pulse oximeters and also have direct measurement of arterial blood oxygen as part of routine care. Data will be collected directly from participants and extracted from their medical record. Skin color will be objectively measured using colorimetry. The specific aims of the study are twofold; first, to quantify pulse oximeter accuracy in children with varying skin colors and second, to explore the mechanistic link between self-identified race and pulse oximeter accuracy. The second aim will assess whether skin color mediates the relationship between race and pulse oximeter accuracy that was demonstrated in earlier studies. The results of this study will contribute much needed knowledge about pulse oximeter accuracy in children of varying skin colors, with the goal of high-quality healthcare for all.

NIH Research Projects · FY 2026 · 2024-01

Project Summary We have identified a novel mode of autonomous programmed cell death that is controlled by p53 in small cell lung cancer (SCLC) — the most deadly form of lung cancer. We capitalized on a genetically engineered mouse model that we created to regulate endogenous p53 inactivation and its temporally controlled reactivation in established mouse models of SCLC. We found that p53 controls a canonical senescence program in approximately half of the tumors and a non-apoptotic cell death program in the rest. The mechanism of cell death is distinct from known forms of programmed cell death but is dependent on multiple members of the cyclophilin protein family of peptydyl-prolyl isomerases. We demonstrate that p53 regulates a distinct transcriptional program in dying SCLC tumors not active in senescing SCLC. Moreover, the activation of this transcriptional program is abolished by small molecule inhibitors of cyclophilins. In aim 1 of our project, we will interrogate the molecular underpinnings of this p53-cyclophilin interaction. At the cell physiological level, p53-mediated SCLC cell death is associated with features of paraptosis; a poorly defined mode of cell death not previously associated with p53 but characterized by excessive accumulation of cytoplasmic vacuoles, cell swelling, and plasma membrane dysfunction. Additionally, we observed massive induction of an endoplasmic reticulum-directed autophagy program (ER-phagy) during p53-mediated SCLC death. In aim 2 of our project we will interrogate the molecular features and dissect the molecular determinants of paraptosis and ER-phagy in this context. Finally, we present comparative genomics that implicates that the `classical' molecular subtype of SCLC is represented by SCLC tumors that die after p53 restoration and that the `variant' molecular subtype of SCLC is represented by SCLC tumors that senesce after p53 restoration. In aim 3 of our project, we will determine if the effects of p53 restoration are distinct based on either the cell of origin, or the collection of driver mutations as each of these impacts the molecular subtype that develops in SCLC. Together, our project will elucidate an important molecular mechanism for controlling SCLC, shed light on targetable vulnerabilities, and identify specific subsets of SCLC patients that may be susceptible to such approaches.

NIH Research Projects · FY 2025 · 2023-12

PROJECT SUMMARY People living with chronic obstructive lung disease, interstitial lung disease, and other forms of chronic lung disease frequently experience exacerbations requiring acute care hospitalizations. These severe exacerba- tions requiring hospitalization are associated with decreased quality of life for patients, increased morbidity and mortality, and increased burdens on health system resources. Through work funded by my K23 award, I have identified actionable risk mechanisms that contribute to hospitalization risk in this population. Notably, many of these mechanisms are present for months or even years among community-dwelling patients liv- ing with chronic lung disease prior to a hospitalization. These findings suggest an opportunity to identify patients with these actionable risks and refer them for appropriate interventions as early as possible. To accomplish this, my K23 supported the development of two prediction models that can efficiently analyze electronic health record data, including both structured (e.g. demographics, diagnoses, and laboratory tests) and unstructured sources (e.g. the text of clinical encounter notes), of large populations to identify those who may benefit from frailty-focused and depression-focused interventions. However, although the models I de- veloped have good predictive performance overall, they were found to have unequal performance by patient race and gender. Additionally, there is a large evidence gap around the feasibility and acceptability of using such machine learning models to guide clinical care in this setting. Therefore, the work proposed in this R03 will fill these important knowledge gaps and pave the way for a future randomized clinical trial of a predictive clinical decision support system to promote population health among people living with chronic lung disease. First, I will use state-of-the-art methods to recalibrate the prediction models so that they perform equitably by patient racial and gender subgroups. This aim is necessary, in light of known racial and disparities in care pro- cesses and outcomes by patient race and gender, to meet minimum ethical standards prior to further model deployment and testing. Second, I will conduct a pilot study of using these models in a real-world clinical workflow to prompt referrals among community-dwelling patients with chronic lung disease. A pulmonologist will review the predictions from these models, verify their clinical appropriateness, and make referrals if the patient and their outpatient clinical team is agreeable. I will assess the feasibility and acceptability of this workflow among patients, their caregivers, and outpatient clinicians. Additionally, I will measure several key process measures related to the proposed protocol in anticipation of a future clinical trial. The findings from these two aims will be sufficient to support a future, planned R01-level application for a clinical trial of this predictive clinical decision support system. Finally, securing this funding, with the data produced through the work outlined in this R03 proposal, will support my transition to an independent investigator.

NIH Research Projects · FY 2026 · 2023-12

Patients with hypertrophic cardiomyopathy (HCM) experience a high symptomatic burden, heart failure and lethal arrhythmias. While the HCM has been recognized as a disease of the sarcomere for >30 years, the heterogeneity of disease expression is substantial and not fully explained by these primary genetic variants. Recent genome wide association studies (GWAS) for HCM have uncovered common genetic variants as risk alleles for HCM. Many of these are conversely protective alleles for dilated cardiomyopathy (DCM), supporting the concept of polygenic regulation of contractile function. However, the mechanisms by which they exert their effects are unknown. Here, we focus on a subset of GWAS loci that strongly implicate protein quality control as a modifier of HCM. Three of the top 10 loci associated with HCM reside within or near genes that encode proteins of the heat shock protein 70 (HSP70) chaperone network: BAG3, HSPB7, and DNAJC18. Prior studies, and our own preliminary data, indicate that these cochaperones are involved in sarcomeric protein maintenance. We hypothesize that risk alleles in BAG3, HSPB7 and DNAJC18 are strong genetic modifiers of sarcomeric HCM, acting through distinct molecular mechanisms to coordinate sarcomere protein dynamics and cardiac contractility. Our approach involves focused human genetic association studies and functional studies in human model systems. In Aim 1, we will determine the impact of risk alleles in BAG3, HSPB7, and DNAJC18 on disease expression and clinical outcomes in patients with HCM, stratified by sarcomere genotype and genetic similarity. Combining cases and controls from several sources, including the Sarcomeric Human Cardiomyopathy Registry (SHaRe), Penn Medicine BioBank, TOPMED and All of Us, we will perform a case-control study to determine associations of BAG3, DNAJC18 and HSPB7 risk alleles with HCM, a case-only analysis to determine association with clinical outcomes, and additive and epistasis modeling to determine interaction effects. In Aim 2, the role of each co-chaperone in regulating sarcomeric protein dynamics and contractility will be explored through genetic knockdown experiments. We will determine effects of each variant on transcript/protein abundance and splice isoforms in cardiomyopathic human hearts. In parallel, we will perform genome editing of each variant in human stem cell-derived cardiac myocytes (hiPSC-CMs) and measure cellular phenotypes, including contractility. For the BAG3 coding variant, Cys151Arg, we will use proximity labeling to determine how this variant affects BAG3 localization and its interactome. Finally, we will use patient-derived hiPSC-CMs that have reduced levels of MyBP-C to determine whether chaperone manipulation stabilizes wild-type MyBP-C to rescue haploinsufficiency. Successful completion of these aims will impact the field by gaining an understanding of the clinical and biological relevance of HSP70 risk alleles to HCM. Our team is composed of highly experienced investigators in cardiomyopathy, cardiovascular genetics, and bioinformatics and is well poised to achieve these goals.

NIH Research Projects · FY 2026 · 2023-12

Complement, a part of innate immunity, eliminates invading microorganisms, apoptotic cells, and cellular debris to maintain the homeostasis of the cellular environment. The complement pathway is tightly regulated to prevent the consequences of inflammation. This regulation can be disrupted by disease, infection, aging, genetic variation or any combination. The long-term objective of this proposal is to uncover the effect of Alternative Pathway (AP) genetic variations on the local complement response in epithelial cells. Our experiments will be done on Retinal Pigment Epithelial (RPE) cells differentiated from Induced Pluripotent Stem Cells (iPSCs) obtained from patients with intermediate Age-related Macular Degeneration. In our first aim we will differentiate 8 iPSCs lines to RPE, four of which have CFH risk alleles and another four which have the risk alleles corrected to non-risk (isogenic controls). After the iPSCs have reached RPE maturity we will quantitate mRNA and protein levels and functional capabilities of complement system players (components, regulators and receptors) synthesized by iPSC-RPE cells. In addition, intracellular AP levels will be determined for iPSC-RPE and their isogenic controls. These results will inform the influence of genetic variation on secretion of complement proteins into the extracellular environment, their intracellular content and pathway differences. Because extracellular vesicles (EVs) are an important cargo for complement and non-complement proteins we will investigate in the second aim the differences in EV protein content secreted from iPSC-RPE cells from apical and basal membrane locations. Our working hypothesis is that the EV protein content and quantity will differ between cells which have the CFH risk variants and their isogenic controls which carry non-risk CFH variants. Secreted proteins also contribute to deposit formation under the cells, which we will investigate by scanning electron microscopy and immunostaining. Cells and their isogenic controls will be exposed to smoke extract and cytokines to determine if the EVs, deposit composition and complement system players (components, regulators and receptors) are influenced by CFH risk variants. Our preliminary results of single cell RNA sequencing (scRNA-Seq) of RPE have led to our central hypothesis that complement plays a crucial role in maintenance of healthy RPE. But the GWAS results on AMD suggest a dysfunction in the complement system. Overall, these studies will provide unique knowledge about the influence of genetic variation on the complement pathway by its focus on functional assays. Our interdisciplinary team is uniquely poised to address this fundamental question by combining our broad background of GWAS and genetic variants of AMD (Stambolian), complement genetics (Atkinson) and RPE complement and extracellular vesicle analysis (Rohrer). The proposal’s outcomes will fill a critical gap in our understanding of genetic variants on RPE complement activity and identify new potential therapeutic targets directed at the RPE. We predict that these results will also have implications for other epithelial cells that serve as barriers in the kidney and lung, also targets of complement activation in human diseases.

NIH Research Projects · FY 2025 · 2023-12

Abstract Repeat-expansion disorders are a collection of more than 40 rare but devastating diseases that largely effect the central nervous system. Currently, they are incurable. The genetic basis for these disorders is repeat expansions, the length of which determines disease severity and age of onset. In these disorders, repeat sequences cause pathology in a multimodal fashion. The translation of repeats located in coding regions can disrupt normal protein function and produce toxic repeat peptides. These neurotoxic peptides can be produced even when the repeats are in noncoding regions via repeat associated non-AUG translation. Another agent of pathology is the repeat RNA transcript itself, where aberrant binding of the repeat RNA to RNA-binding proteins leads to issues like deregulation of the miRNA machinery, aberrant translation, and mis-splicing events. Repeat RNAs also enable transcripts to undergo liquid-liquid phase separation (LLPS) and form nuclear RNA foci in cells. These foci can sequester RNA-binding proteins, which may contribute to the observed repeat RNA toxicity. It has been well established that RNA helicases regulate the formation of phase separated ribonucleoprotein (RNP) granules in cells, but no work exists to understand how helicases affect the RNA foci in repeat-expansion disorders. Considering first that repeat RNAs cause potentially pathological RNA foci and that second, helicases are important for the formation of other cellular RNP granules, we hypothesize that RNA helicases can also modulate the formation of RNA foci in repeat-expansion disorders. The proposed work is focused on the deleterious RNA foci that form in the repeat-expansion disorders caused by CAG trinucleotide expansions like Huntington’s Disease and several spinocerebellar ataxias. This proposed study aims to identify RNA helicases that modulate these deleterious RNA foci and understand the mechanism by which they do so. Knockout screens in human HAP1 cells expressing CAG repeat constructs which form RNA foci have identified candidate RNA helicases that affect RNA foci size and number. This work will utilize in vitro approaches to understand the mechanism underlying how these candidate helicases effect RNA foci. After the recent failure of two candidate therapies for Huntington’s Disease in clinical trials, it is especially necessary to consider new therapeutic targets for these disorders. This work will expand our understanding of RNA helicase function in repeat-expansion disorders and offer a new therapeutic tool to explore for combatting these diseases. The proposed fellowship will be conducted at the University of Pennsylvania in the lab of Dr. James Shorter, a leading researcher in the molecular basis of neurodegenerative diseases and protein disaggregases. Completion of the proposal in this lab will provide rigorous training in biochemical assay design and in vitro protein and RNA handling techniques.

NIH Research Projects · FY 2026 · 2023-12

PROJECT SUMMARY Rapid eye movement (REM) sleep is characterized by intense cortical activation, giving rise to its wake-like electroencephalogram (EEG). Disturbances in REM sleep are symptomatic of mood disorders including major depressive disorder (MDD), which is associated with dysfunction of the prefrontal cortex (PFC). However, to what extent altered cortical activity in mood disorders may directly cause changes in REM sleep remains largely unknown. Particularly, because REM sleep is thought to be primarily controlled by subcortical nuclei, the mechanisms by which the PFC may regulate REM sleep are unknown. A prominent biomarker for depression is an increase in phasic REM sleep, reflected in an increased density (frequency) of rapid eye movements during REM sleep. However, despite its clinical relevance, the mechanisms and brain structures underlying the regulation of phasic REM sleep are still largely unknown. Our preliminary results show that pyramidal neurons in the medial PFC (mPFC) promote REM sleep in mice through their projections to the lateral hypothalamus (LH). In addition, we found that activation of mPFC pyramidal neurons enhances the density of rapid eye movements during REM sleep. The central objective of this proposal is to identify the neural mechanisms by which the mPFC regulates the induction and maintenance of REM sleep and rapid eye movement density. This will be accomplished in two aims. First, we will combine in vivo calcium imaging, optogenetics, optrode recordings, and viral tracing to identify the activity, connectivity, and functional role in brain state regulation of the LH neurons innervated by the mPFC. Second, we will investigate the mechanisms by which the mPFC regulates rapid eye movements during REM sleep. We will combine pupillometry with optogenetic manipulation, in vivo calcium, and optrode recordings to examine whether the axonal projections from the mPFC to the pons induce rapid eye movements and test whether mPFC neurons and their postsynaptic pontine neurons are recruited during rapid eye movements. Unraveling the mechanisms by which the mPFC controls REM sleep and the rapid eye movement density is a critical precursor to resolving the causal relationship between changes in prefrontal circuits associated with depression and REM sleep disturbances. Our results have the potential to develop more refined biomarkers to infer from disturbances in the regulation of REM sleep and rapid eye movements information about the integrity of prefrontal circuits in health and disease.

NIH Research Projects · FY 2026 · 2023-11

The study of brain-immune interactions has contributed to a paradigm shift in the understanding of the pathophysiology of mental disorders and has provided new insights into their prevention and treatment. Notably, the blood-brain barrier (BBB) functions to restrict interactions between the brain and periphery, limiting the influence of peripheral inflammation on the brain. However, increasing evidence indicates impairment of the BBB in conditions like autism and schizophrenia, which combined with the consistently elevated levels of the cytokines IL-6 and IL-1β, could potentially affect the ontology and outcome of neuropsychiatric disease. Thus, this proposal addresses the connectivity of peripheral-neuroinflammation through brain vasculature, its impact on behavior, and how they interact in response to an environmental “second hit”. We will study this interplay within the brain- immune axis in the context of genetic susceptibility to neuropsychiatric disease. Besides providing a compelling rationale to devise immune therapeutic strategies, the demonstration of a brain-immune link will be crucial for understanding the mechanisms driving disease processes in mental illness. Our proposal makes use of a model of human genetic susceptibility to neuropsychiatric disease conferred by the 22q11.2 deletion syndrome (22qDS). Based on the inclusion of genes that could affect BBB function, our previous work demonstrated BBB compromise in 22qDS, a phenotype recapitulated in the mouse model of the disease. Furthermore, our preliminary data indicate that adult 22qDS model mice have elevated levels of IL-6 and IL-1β, cytokines that promote T cell responses characterized by the expression of the pro-inflammatory cytokine IL-17. We find enhanced IL-17-mediated inflammatory activation in the brain, modeling the exaggerated IL-17 responses seen in 22qDS and neuropsychiatric patients. Notably, 22qDS mice also exhibit behavioral deficits reminiscent of those seen in human patients. Therefore, we will test the hypothesis that inflammation across the brain-immune axis exacerbates a deficient BBB to promote neuroinflammation and neuropsychiatric disease in 22qDS. To address this, we will: (Aim 1) Define the links between these different phenotypes throughout development. (Aim 2) Determine how the BBB impacts brain-immune axis responses to environmental “second hits” known to trigger symptom onset and the potential of restoring BBB properties therapeutically. (Aim 3) Identify potential immune therapeutic strategies aiming at brain-immune axis dysfunction to treat behavioral deficits. As CHOP houses the world's largest 22qDS clinic and Penn is a leader in neuropsychiatric research, our environment is ideal for translating these studies into actionable science.

NIH Research Projects · FY 2026 · 2023-11

PROJECT SUMMARY Parasitic nematodes colonize over a billion people worldwide and are associated with a range of maladies. Among these parasites, the whipworm Trichuris trichiura is particularly difficult to manage with existing medication that target adult worms due to drug resistance and frequency of reinfection. Following ingestion by the host, Trichuris eggs hatch in the gastrointestinal tract where they mature into adulthood and produce eggs, which are then deposited back into the environment in feces to perpetuate the cycle. As such, the entire time the parasite inhabits the host is spent in the presence of a diverse community of bacteria that are part of the gut microbiota. We and others have shown that colonization by Trichuris trichiura and the model parasite Trichuris muris restructures intestinal bacterial communities in humans and mice, respectively. Consistent with interactions between parasitic nematodes and the gut microbiota, bacteria are required for T. muris to complete its life cycle in mice due to a role for bacteria in promoting egg hatching. The details of how the gut microbiota mediate egg hatching and the reproductive fitness of Trichuris remain obscure. In preliminary results, we identified bacteria that differentially affect T. muris egg hatching and established a novel C. elegans screen to identify bacterial gene products that have a conserved role in the reproductive fitness of free-living and parasitic nematodes. The main objective of this proposal is to identify mechanisms by which bacteria mediate Trichuris egg hatching and affect subsequent colonization of the host. First, based on our findings that T. muris and C. elegans share a requirement for an arginine-dependent byproduct of E. coli, we will identify this metabolite and determine how it impacts host reproduction using the two nematode models. We will specifically test the hypothesis that polyamines are the key arginine-dependent factors in this system. Then, we will delve deeper into how bacteria-egg interactions mediate hatching by determining the role of chitinase produced by the parasite and compare bacterial species that promote versus impede hatching. Lastly, we will determine whether bacteria displaying superior egg-hatching activity enhance T. muris infection of mice, and using a new assay we developed for T. trichiura, extend our observations to the human parasite. We believe these innovative approaches will improve our understanding of how parasites have adapted to their host environment, and ultimately reveal vulnerabilities in the Trichuris life cycle that can serve as therapeutic targets.

NIH Research Projects · FY 2024 · 2023-09

PROJECT SUMMARY / ABSTRACT Policymakers are pursuing strategies to integrate coverage for dual-eligibles—individuals with Medicare and Medicaid—to better coordinate care across these programs. Dual-Eligible Special Needs Plans (D-SNPs)— Medicare Advantage plans that exclusively serve duals—have emerged as the largest category of plans intended to serve as a platform for integration. However, D-SNPs vary in their level of integration with Medicaid. While many D-SNPs meet only a minimum level of integration (primarily related to care coordination), a growing number of plans have attained greater integration by covering Medicare and Medicaid spending for the same patients. CMS continues to refine integration standards for D-SNPs and policymakers have called for designing these standards to improve the quality, efficiency, and equity of care for duals—a priority given this population’s medical and social vulnerability. Yet little evidence exists about the extent to which differences among D-SNPs, including plan attributes and state policies affecting integration, are linked to quality, care patterns that drive spending, or disparities. Further, there is no evidence about how policy levers to expand enrollment in integrated D-SNPs affect care. One such lever is default enrollment—a mechanism that allows certain Medicaid managed care plans to automatically enroll Medicaid beneficiaries into integrated D-SNPs when they become dual eligibles (e.g., at age 65). Although only some D-SNPs are permitted to use default enrollment, there is interest in expanding this mechanism. To inform policy, this project will provide new evidence about variations in the performance of D-SNPs, including by factors affecting these plans’ level of Medicaid integration (Aim 1); assess plan performance and disparities in vulnerable subgroups of duals, including Black and Hispanic duals, duals with a disability, and those with behavioral health conditions (Aim 2); and investigate the effects of using default enrollment to integrate coverage, leveraging variation between plans permitted to use this mechanism and a quasi-experimental difference-in-differences design (Aim 3). To measure key aspects of plan performance relevant to integration, we will use Medicaid T-MSIS data, Medicare Advantage encounter data, and nursing facility assessment data linked at the beneficiary level (Aims 1-3) and CAHPS patient experience surveys (Aims 1-2). Continuous updating of these data allows us to monitor changes in performance as CMS phases in new D-SNP integration standards, enabling us to evaluate and inform evolving policy. This project is directly related to AHRQ priorities and priority populations through its focus on how the design of D-SNPs—an emerging managed care model for complex patients—affects quality, efficiency, and equity of care. To maximize this project’s impact, we will work with a Policy Advisory Committee of experts in integration policy from the government, research, and insurance sectors, building on the team’s track record of translating health services research to inform policy.

NIH Research Projects · FY 2025 · 2023-09

Project Summary Whether cocaine exposure supports long-term HIV expression in immune cells and whether inflammation mediates this effect are major knowledge gaps. A clear understanding of cocaine's impact on HIV expression in complex multicellular contexts is a prerequisite for the identification of novel targets for effective ART regimens, adjunctive neuroprotective therapies, and latency reversal strategies. Our overarching hypothesis is that long­ term HIV expression in macrophages and microglia contributes to neuropsychiatric damage, which is exacerbated by cocaine. During the second year of this award, we made tremendous progress in developing HID-Seq (HIV integrated proviral DNA [HID]/single-nuclear RNASeq) (Fig. 1). In brief, HID-Seq is a new approach that combines single-nuclear RNASeq (snRNASeq) with dCas9-Tn5 tagmentation. This is a scalable method that permits increased read depth that can simultaneously detect integrated HIV proviral DNA, HIV RNA species, and host transcriptome within the same cell, achieved through a multi-index barcoding strategy (Fig. 3, 4). HID-Seq will enable us to (A) determine the identity of discreet populations of productively infected (HIV DNA+/HIV RNA+), latenUnonproductively infected (HIV DNA+/HIV RNA, and uninfected (HIV DNA/HIV RNA) cells, and (B) to simultaneously link the presence of HIV DNA and RNA to functional cellular outcomes, such as inflammation, via the transcriptome in the same cell. We have achieved nearly all objectives in all 3 milestones for the R61 phase, as detailed in the Progress to Milestones section. The primary focus within the second period of R61 period was to complete the individual steps of HID-Seq, including the detection of HIV DNA and RNA (Milestone 1), the quantification of cells harboring HIV DNA and HIV RNA (Milestone 2), and the confirmation of read depth (Milestone 3). We have overcome the technical challenge in the detection of HIV DNA by revising our original design (version 1), where we would use dCas9-Tn5 fusion protein to cut and tag DNA, to version 2, where we use dCas9-Tn5 fusion protein together with free Tn5. This enabled us to increase the rate of tagmentation of the region of interest, i.e., where HIV proviral DNA is integrated into the cellular genome. • Milestones 1 and 2: In our experiment where we bulk sequenced over 5000 cells that were processed using dCas9-Tn5, we were able to detect that 2430 cells had HIV DNA, leading to a successful HIV DNA detection rate of 48.6%. We also optimized the single-nucleus RNASeq (snRNASeq) protocol, which we term HIV-snRNASeq, which enabled us to significantly increase the sensitivity of HIV RNA detection by 40-67-fold compared with the traditional methods. • Milestone 3:_We successfully detected HIV DNA in ~50% of the J-Lat cells with bulk sequencing 5000 cells, indicating a good read depth, by using this novel approach. We also developed the computational pipeline to facilitate the analysis of joint-profiling of the host-cell transcriptome and HIV genome at the single-cell level, i.e., the junction site including both the HIV and host-cell genome. We will be able to determine the identity of discreet populations of productively infected (HIV DNA+/HIV RNA+), latenUnonproductively infected (HIV DNA+/HIV RNA, and uninfected (HIV DNA/HIV RNA) cells and determine their transcriptomic profiles. With our data showing that we can detect HIV DNA and RNA with increased sensitivity and good read depth, we are confident that the ongoing analyses of the HID-Seq of J-Lat cells will reveal that we can detect the different states of cells based on the presence of HIV DNA and RNA

NIH Research Projects · FY 2025 · 2023-09

Housing and zoning policies that cause geographically clustered economic hardship and lack of economic opportunity often result in neighborhood disinvestment and increased neighborhood disorder. Neighborhood disorder—deterioration of the urban landscape, also known as blight—significantly impacts opioid misuse, including opioid initiation, injection drug use, opioid use disorder development and duration, and risk of opioid-involved overdose. Interventions to improve disorder have shown significant positive impacts on neighborhood crime and community health. No studies have yet investigated the degree to which neighborhood disorder interventions influence opioid misuse or associated overdose risk. The broad objective of this study is to investigate the impact of community infrastructure reinvestment programs on opioid misuse and opioid overdose. In Philadelphia, government, community, and academic partnerships have resulted in infrastructure-based approaches (PHS Philadelphia LandCare and Basic Systems Repair Program) to remediate vacant lots, abandoned buildings, and dilapidated homes. These programs, which were designed to improve neighborhood disorder in communities with little to no green space, may have distal effects on opioid misuse and associated overdose risk. Previous Philadelphia studies have shown significant reductions in crime and improved stress and depression for residents on blocks with blight remediation. This study will use spatial analysis methods and systematic social observation to address HEAL’s RFA DA-23-051 Preventing Opioid Misuse and Co-Occurring Conditions by Intervening on Social Determinants. Specific aims are: (1) Investigate whether street block-level indicators of opioid misuse differ on blocks with blight remediation activities using systematic social observation methods for primary data collection of litter from opioid misuse (e.g., syringes, tourniquets) and harm reduction paraphernalia (e.g., wound care, naloxone); (2) Examine if fatal and nonfatal opioid overdose rates improved with remediation activities using a longitudinal spatial panel approach with fatal overdose data from Philadelphia’s medical examiner and nonfatal overdose data from hospital emergency care visits; (3) Explore mechanisms by which remediation activities improve neighborhood indicators of opioid misuse and fatal and nonfatal overdose rates, including improved community social capital and social cohesion (measured by primary data collection of neighborhood structures related to these concepts), using systematic social observation and spatial analysis methods. This study will examine the impact of highly innovative place-based public health interventions on opioid misuse and fatal and nonfatal opioid overdose. Findings will provide support for expanding neighborhood disorder remediation projects in Philadelphia and promote similar public health-based policy and community-level health interventions for opioid misuse and overdose prevention in other cities.

NIH Research Projects · FY 2024 · 2023-09

COVID-19 has caused unprecedented loss of life and global disruption since its emergence in 2019. Caused by the SARS CoV-2 virus, this infection shows extreme heterogeneity, ranging from completely asymptomatic to deadly. One factor that has been linked to COVID-19 severity is the microbiome of the upper respiratory tract, specifically the oropharynx. Lower relative abundance of oral commensal taxa, such as Haemophilus, Neisseria, Prevotella, and Actinomyces, and lower alpha diversity are seen in severe COVID-19 patients compared to individuals with more moderate disease. The mechanism of this association is still unknown, and it is unclear in which direction causation occurs. We propose to further examine the association between the respiratory microbiome and COVID-19 by 1) increasing specificity of these associations to the species, strain, and gene level, 2) identifying how comorbidities shape the respiratory microbiome prior to SARS CoV-2 infection, and 3) identifying host pathways that may be involved in these associations. For this first aim, we will leverage a cohort of over 200 hospitalized COVID-19 patients (previously enrolled and specimens already in hand), using deep metagenomic sequencing for taxonomic and functional annotation. The increased specificity provided by this aim will pave the way for in vitro or animal model experiments, which require species or strain level associations for proper experimental design. The second aim will focus on respiratory tract microbiome profiles in individuals with obesity, diabetes, or old age (who do not and have not had COVID-19), three conditions that are strongly associated with elevated risk of severe COVID-19. The effect that these have on the respiratory microbiome is unknown, but one still untested possibility is that the microbiome mediates some of the effects of these conditions on disease severity. By studying microbiome alterations in these diseases prior to SARS CoV-2 infection we could identify a potential high risk microbiome that precedes severe COVID- 19. Finally, the third aim pulls data from a diverse set of databases to create a knowledge graph of microbe- disease-gene associations. Using knowledge graph completion, we will predict host genes that both associate with COVID-19 severity, and interact with bacteria in the upper airway. With this data, we can propose possible host mechanisms that mediate microbiome-COVID-19 associations, allowing for in vitro follow-up to move from correlation to causation. Ultimately, this work is a bridge between existing high level associations between COVID-19 and the upper respiratory microbiome, and future work targeting specific mechanisms and causal links. Having recently published a review on all studies of the airway microbiome in COVID-19, we believe that these aims address the most critical gaps in understanding currently in the literature.

NIH Research Projects · FY 2023 · 2023-09

Contact PD/PI: Fashaw-Walters, Shekinah Antoinette PROJECT SUMMARY Home health care is an essential healthcare service and benefit under the Medicare insurance programs, aimed to improve health outcomes, reduce avoidable hospitalizations, and decrease institutional placements. Racial and ethnic minority Medicare beneficiaries utilize home health care at higher-rates compared to the general Medicare population. Despite the growth and increased use of home health in recent years, there is little evidence about the role of home health agency organizational structures in home health quality and equity. There is also a gap in knowledge about minority-serving home health agencies. Most “high-minority” agencies are rated as lower quality by CMS; however, they may also provide culturally and linguistically appropriate care that should be associated with the best outcomes and highest quality of care. This paradox will be explored using a strengths-based approach, to understand factors associated with high-quality home health care outcomes for racial- and ethnic minority patients overall, and within the context of minority-serving home health agencies. The overall objective of the proposed project is to create home health process and outcome equity measures by calculating race- and ethnicity-specific quality ratings to examine the organizational characteristics of home health agencies with the highest ratings and those with the highest share of patients of color. The evidence generated from this project will be an important first step in (1) creating better measures of health equity that center marginalized populations, and (2) designing and evaluating health systems and policy interventions with the goal of eliminating inequities in essential home health services. Informed by Donabedian's Structure-Process-Outcomes framework, race- and ethnicity-specific home health quality ratings will be calculated for the process of care and outcome measures used in the Centers for Medicare and Medicaid Services (CMS) star ratings. This study will match, merge, and link national Medicare beneficiaries' administrative, home health care assessment, and inpatient claims data with home health agency level CMS Care Compare data to evaluate within-home health agency racial inequities in care quality outcomes through two aims: (1) Compose and demonstrate the utility of race and ethnic group-specific home health agency quality ratings, adapting the methodology used by CMS Care Compare. We will evaluate utility of the new race- and ethnicity-specific measures in comparison to the original, overall measures through the calculation of reliability and informativeness metrics. (2) Assess home health agency structure as a predictor of quality and equity in home health agency process and outcomes. Home health agency structural factors (e.g., minority-serving, size, tenure, ownership, rurality) will be used to predict race- and ethnicity-specific care quality process (i.e., timely initiation of care) and outcome measures (e.g., functional improvements, hospitalizations). This work is both timely and significant, with the potential to profoundly impact quality and health equity measurement within home health and other healthcare sectors. Project Summary/Abstract Page 6

NIH Research Projects · FY 2025 · 2023-09

Johnson, Kevin DP1 Details Project Name: Helping Doctors Doctor: Using AI to Automate Documentation and “De-Autonomate” Health Care Project Summary Clinical encounter documentation is one of the most time-consuming tasks of the ambulatory encounter, taking approximately two hours for every hour spent with patients. Clinical note lengths in the US are longer than those in other countries due to requirements to justify billing and complete quality care metrics. Not surprisingly, clinical encounter documentation has become a major source of clinician burnout over the past two decades. Although companies are selling technologies to transcribe what was discussed during the encounter, these costly solutions reproduce what is already suboptimal about how we create and use EHR information. There is a desperate need to reimagine the process of documenting as well as the content of documenting a clinical encounter. In this application I propose to develop a new generation of automated documentation algorithms—algorithms that can listen to the dialog between a patient and clinician, collect quantitative data about these observations, combine those with existing electronic health record (EHR) data and create relevant encounter summary information. These documentation algorithms will leverage the remarkable progress we have made in computer vision, natural language processing, machine learning to support image labeling, and other advances using EHR data. These novel computational approaches have yet to be explored as alternative approaches to summarizing medical data collected in real time. As a pediatrician and biomedical informatician who has acquired considerable expertise in real-world systems design, implementation and medical data analytics, this project leverages many of my skills, though it is a departure from my previous human-computer interface work. Rather, the goal of this project is to remove the burden of documentation from clinicians to the extent possible. To achieve this goal, I will work with a multidisciplinary group of collaborators, including computer scientists, technology engineers, and clinical domain experts. Specifically, I will: (1) collect and analyze exam room video and annotations of the encounters to identify salient characteristics of patients and their interaction with the clinician that led to specific diagnoses; (2) apply natural language processing, deep learning, and computer vision methods to learn and characterize patterns from vast streams of data using supervised and unsupervised learning methods. Combining these techniques to directly impact what is documented and how it is generated is a new area of investigation for me and an approach that promises to support innumerable other projects including identifying implicit bias in clinical encounters, enabling a new class of real-time decision making and improving the usefulness of encounter summaries.

NIH Research Projects · FY 2024 · 2023-09

Project Summary People insured by Medicaid with a co-occurring serious mental illness (i.e., depression, schizophrenia, and bipolar disorder) experience persistent and growing disparities following medical hospitalizations and interventions that focus on their transitional care needs are lacking. Our study focuses on the adaptation of the THRIVE, an equity-focused evidenced based intervention that supports Medicaid-insured individuals with multiple chronic conditions transitioning from hospital to home. THRIVE services include coordinating care, standardizing interdisciplinary communication, and addressing unmet clinical and social needs following hospital discharge. While evidence suggests robust improvements in post-discharge outcomes for THRIVE participants, it remains unclear whether the interventions equally benefit patients with and without SMI. The goals of this project are to adapt, implement and evaluate THRIVE to ensure suitability for all patients insured by Medicaid including those with co-occurring SMI. To meet this goal, we propose the conduct of a Type I Hybrid effectiveness-implementation stepped wedge cluster randomized controlled trial (CRCT)15 at a new site, the Hospital of the University of Pennsylvania Cedar Campus (HUP-Cedar). HUP-Cedar is a 100-bed minority serving hospital located in West Philadelphia where >50% of patients are Medicaid or dually Medicare and Medicaid-insured. This study represents a three-way partnership among health services researchers, community advisors, and health care leadership and staff at HUP-Cedar and Penn Medicine at Home and seeks as its primary objective to leverage participatory methods to adapt the THRIVE intervention and evaluate the context required for implementation. A second objective for this study is to examine differences in the utilization outcomes (e.g., readmissions) between THRIVE participants with SMI compared to those receiving usual care. Finally, we will employ mixed methods approaches through participant and clinician interviews and surveys to evaluate the acceptability, appropriateness, feasibility and cost of the intervention. Our study and engagement with community and health system partners is informed equity centered frameworks that will help to emphasize the systems, policies and delivery process from which inequities might arise. If our study hypotheses are supported, we will identify not only if an equity-oriented evidence-based intervention can be adapted and implemented to meet the needs of Medicaid insured patients including those with SMI, but we will also uncover the factors associated with those results. Our proposal is well-aligned with AHRQ's goal to develop and adapt EFEBI's that prioritize the voices and needs of the most underserved populations and change care processes to accelerate equity within healthcare systems. Collectively, the results of this study will provide the foundation for the next phase of our research, which includes a multi-site evaluation of equity-based implementation strategies to scale and rigorously evaluate THRIVE diverse hospital settings.