ACCEL Funding Lifecycle Resources

People with mental illness (PWMI) are overrepresented in and underserved by the criminal justice system. Justice involvement often begins when a PWMI makes a 911 call. Public safety operators (PSOs) deploy police officers who may lack the expertise and resources to connect PWMI to supportive services. PWMI may be better served by mental health clinicians through PSO referral. The proposed research evaluates the impacts of a 911 dispatch diversion program. Developed by the New Castle County Division of Emergency Communications (NCDEC), Intercept Zero allows PSOs to redirect callers presenting with an emergency related to a mental health crisis and those without emergencies and documented mental health histories to clinicians embedded in the 911 center. Previously, our work with the New Castle County Division of Police (NCCPD) has shown that an arrest diversion program called the Behavioral Health Unit (BHU) featuring the co-response of a clinician with a specially trained police officer decreased arrests, incarceration, and hospitalization for PWMI. It is urgent to determine whether deflections can occur at the point of a 911 call that represents the earliest stage of justice contact. The long-term goal is to understand how PWMI can be diverted from criminal processing and toward supportive services at the 911 stage. Our proposed research uses two programs featuring three approaches to assess Intercept Zero’s impacts. Program 1 uses time-series modeling to take two approaches that distinguish changes in call outcomes. The first approach aims to distinguish the effects of Intercept Zero on police deployment and arrest events. We compare observed events documented by NCDEC and time-series forecasts of these events to approximate a counterfactual world without the Intercept Zero program from April 2025 onward. We will construct this counterfactual using five years of dispatch and arrest data collected on behalf of NCCPD's BHU. Our second approach considers whether Intercept Zero affects 911 calls. We will use program records and apply a time-series regression model strategy to examine shifts in the length and frequency of 911 calls. Program 2 relies on cost analyses to take a third approach to estimate Intercept Zero's cost savings to health and justice systems. Using Delaware healthcare cost and public expenditure data, we will calculate changes in police deployment, arrest, and hospitalization costs. Together, these Programs will underscore the effectiveness of Intercept Zero as a paradigm for dispatch diversion.

Ewing sarcoma (EwS) is a rare and aggressive cancer primarily affecting children. While early-stage EwS can be cured, over 40% of patients experience disease recurrence after initial chemotherapy, leading to a low 5-year survival rate of 15%. Currently, there are no reliable biomarkers to predict which tumors will respond to treatment or recur, highlighting the urgent need for new biomarkers and therapies.

Intra-tumor heterogeneity, where a single tumor contains diverse cell populations, affects patient outcomes and treatment responses in various cancers. Advances in spatial biology now allow us to study this heterogeneity using RNA expression from archival tissue samples, which is particularly useful for rare cancers like EwS.

Our project aims to develop a tumor microarray (TMA) using archival EwS tissue to study gene expression heterogeneity. This approach overcomes the challenges of small sample sizes, the need for fresh tissue, and high costs associated with current methods.

Aim 1 (Feasibility): We will determine if TMA-based spatial transcriptomics can measure gene expression heterogeneity in EwS. We will assess the quality of sequencing and the correlation between sequencing quality and sample characteristics. We will also quantify gene expression heterogeneity within each sample.

Aim 2 (Exploratory): We will evaluate if spatial RNA analysis can accurately measure the expression of the driving fusion gene (EWSR1::FLI1/ERG) in EwS. We will compare protein and RNA expression of the fusion gene and analyze differences between high and low fusion-expressing cells.

Impact and Expected Outcomes: This project will provide a new method to study EwS, creating a valuable dataset to measure gene expression heterogeneity. The findings will support the development of predictive and prognostic biomarkers, addressing a critical need in this rare and aggressive pediatric cancer.

Prediabetes affects over one-third of Delaware residents, placing more than 70% at risk of developing type 2 diabetes (T2D) within a decade if left untreated. This public health burden disproportionately impacts diverse population segments, including racial/ethnic minority groups and those with lower income and education; groups who are less likely to benefit from current prevention strategies. The 12-month National Diabetes Prevention Program (DPP) is the first-line intervention to prevent T2D, but only 35% of participants achieve the recommended ≥5% weight loss, with non-responders more likely to reside in high-priority communities. Moreover, comorbid conditions such as hypertension remain inadequately addressed. Sleep health improvements and time-restricted eating (TRE) are two promising, evidence-based strategies shown to enhance metabolic and cardiovascular outcomes. Despite recommendations from national guidelines for diabetes prevention and management to include sleep health, their widespread implementation is limited by the lack of an effective delivery platform. Expanding the DPP to include these interventions (creating the "DPP++"; DPP + sleep health + TRE) could provide a scalable solution to enhance efficacy and improve access to diabetes prevention across all communities. This 12-month Clinical and Translational Research (CTR) pilot study will evaluate the clinical efficacy of DPP++ compared to the standard DPP among adults at risk for T2D already enrolled in DPP cohorts administered by the University of Delaware Nutrition Clinic. Focused community-engagement efforts that are overseen by the study steering committee will target traditionally overlooked populations to enroll in these cohorts. Between weeks 4–14 of the standard DPP, participants will be invited to join a nested randomized controlled trial. At week 16, consenting participants will be randomized to either the standard DPP (control) or DPP++. New cohorts will launch in October 2025, January 2026, and May 2026. With ~n=28 participants per cohort, 60% enrollment in the nested trial of n=84, and 20% attrition of n=50 in the nested trial, we anticipate n=40 randomized participants with complete data at the 24-week/6-month primary endpoint. Aim 1 of this study will compare the efficacy of DPP++ vs. standard DPP on metabolic health at 6 and 12 months. Aim 2 will assess DPP++ vs. standard DPP differences in cardiovascular health and quality of life. We hypothesize that participants randomized to the DPP++ condition will achieve greater weight loss (primary outcome) and improvements in waist circumference, blood pressure and quality of life as compared to controls at the 6-month primary endpoint. Completing this pilot will also support the capacity for integrating the sleep health and TRE intervention components into clinical practice. As such, completion of this study is expected to provide critical preliminary data to support a subsequent Patient-Centered Outcomes Research Institute (PCORI) application as part of the Broad Pragmatic Studies mechanism as well as provide effect sizes to support a rigorous power analysis for a parent R01 application. Ultimately, this line of inquiry is expected to inform subsequent updates to the DPP that occur every 5 years, enhancing its impact for priority populations and reducing the burden of T2D.

Individuals with chronic stroke are characterized by low levels of physical activity, which negatively impacts post-stroke health and the risk for recurrent stroke. Although walking speed and endurance are logical targets for increasing physical activity, improving them alone rarely translates to more activity. Individuals with chronic stroke also have low falls efficacy, which includes their confidence in performing daily activities without falling. Evidence suggests that, if low falls efficacy is not addressed, the benefits of improved walking capacity on walking activity won’t be realized. To date, falls efficacy has been expressed as a composite score from questionnaires, including the Activities-specific Balance Confidence (ABC) scale. A challenge to addressing falls efficacy after stroke is that it is likely comprised of distinct balance domains not reflected by a composite score. These balance domains include anticipatory control, walking balance, and reactive balance, and they can be considered independent targets for rehabilitation. A composite ABC scale score does not inform the specific balance domains that should be targeted to improve falls efficacy and physical activity. Our secondary analyses of existing data suggest that falls efficacy in chronic stroke is comprised of distinct balance domains that have specific, factor-based relationships with different aspects of sensor-based physical activity. While exciting, this preliminary data has weaknesses that must be addressed before we propose a larger study. In the current proposal, we will a) incorporate additional scales that improve how we evaluate walking-balance and reactive-balance efficacy, and b) incorporate measures of sedentary time, an important aspect that is associated with cardiovascular risk independent of walking activity levels. We will also test the feasibility of a remote study approach that enables us to recruit a broader range of function, personal factors, and environmental contexts. The results from his study will support subsequent applications to evaluate the relationships between skill-specific falls efficacy and physical activity, develop new measures of skill-specific falls efficacy, and target such efficacy with intervention in a manner not yet achieved.

Delaware faces a growing substance use disorder (SUD) crisis, with one of the highest overdose death rates in the nation—53.5 per 100,000 in 2023, nearly double the national average. Opioids are involved in nearly 90% of these deaths, and the burden is disproportionately borne by vulnerable populations: overdose mortality among non-Hispanic Black residents rose 72% from 2020 to 2022, and over 85% of deaths occurred among individuals with a high school education or less. Delaware also ranks fifth nationally for SUD diagnoses among pregnant women and third for Neonatal Abstinence Syndrome. Rural communities face structural barriers including limited treatment access, transportation challenges, and a shortage of addiction providers. Despite these alarming trends, Delaware lacks a comprehensive system to monitor SUD prevalence, treatment access, health outcomes, and related costs. This project aims to fill that gap by leveraging the Delaware Healthcare Claims Database (HCCD) to generate actionable insights into SUD trends and disparities across the state. Our central hypothesis is that SUD outcomes and treatment access vary significantly by demographic and geographic group, contributing to unequal health and economic burdens. Aim 1 will describe statewide trends in SUD prevalence, treatment utilization (e.g., medication-assisted treatment), emergency department visits, hospitalizations, and inflation-adjusted healthcare costs from 2016–2025. Estimates will be stratified by age, sex, race/ethnicity, insurance type, and geography. We will also compare healthcare utilization and costs between individuals with and without SUD to quantify excess burden. Aim 2 will examine disparities in treatment outcomes and costs using multivariable regression and spatial analysis. We will identify differences by demographic group and insurance status, assess intersectional disparities (e.g., socioeconomic status × sex), map high-burden ZIP codes or census tracts, and evaluate the role of comorbidities in driving healthcare costs. This pilot will create the foundation for a scalable, statewide SUD monitoring system that supports equitable, data-driven public health responses, guides policy and resource allocation. The analytic infrastructure developed through this project can also be adapted to address other pressing public health challenges across Delaware and beyond.

Gliomas are the most common central nervous system tumors in children and account for about 35% of childhood cancers. The death rate within 3 months after diagnosis was approximately 2.4%. Isocitrate dehydrogenase (IDH) enzymes play a key role in normal major metabolic processes and homeostasis. However, mutations in their isoforms, IDH1 and IDH2, are present in almost all grade 2 diffuse gliomas in adults and over 80% of low-grade gliomas (LGGs). Moreover, mutations in IDH1 and IDH2 occur early in glioma development and are associated with a much higher risk of malignant transformation, particularly in older children. Therefore, targeting gliomas with IDH mutations has been a novel treatment strategy. However, the current lack of noninvasive imaging of IDH1 and IDH2 mutations in a clinical setting is a pressing issue. Brain magnetic resonance imaging (MRI) followed by tumor biopsy testing is the current gold standard for diagnosing IDH mutant gliomas. However, MRI may show lower specificity in differentiating active tumors from treatment-related changes, such as necrosis. MRI also shows limitations in non-focal pediatric gliomas due to its difficulty distinguishing tumor and surrounding infiltration from normal tissue. Traditional tissue biopsy assessment is invasive with potential morbidity and does not fully reflect the heterogeneous and evolving tumor landscape in longitudinal settings. Therefore, the need for diagnostic and predictive tools for guiding therapy in children with gliomas is urgent. In August 2024, the FDA approved the first small molecular brain-penetrable drug, vorasidenib, for treating gliomas by targeting IDH1 and IDH2. Here we hypothesize that a fluorine-18 labeled vorasidenib, in which one of the fluorine-19 atoms is replaced by PET isotope fluorine-18, can be noninvasive, quantitative imaging of IDH mutation in gliomas in real time. We proposed two aims in the pilot study: Aim 1, develop a fluorine-18 labeled vorasidenib; Aim 2, evaluate [18F]vorasidenib in glioma xenograft animal model. Upon completing the pilot project, we expect to develop the first brain-penetrable and noninvasive imaging tool for accurate diagnosis, stratification, and prognosis of pediatric gliomas who benefit from IDH-targeting treatment. Our long-term goal is to validate our hypothesis in a clinical setting.

Meniscus tears are common and have deleterious impact on both current activities and future onset of knee osteoarthritis. This can be particularly debilitating in the young active population. Therefore, surgical repair of a torn meniscus is regularly performed. While the surgical procedure is well established, unfortunately, there is no consensus on the post-operative rehabilitation protocol. On the one hand, early weight-bearing can facilitate functional recovery. On the other hand, early weight-bearing could compromise the repair’s mechanical integrity. Addressing the unknown effects of post-op rehabilitation is critical. If this question remains unanswered, patients will suffer from unnecessary restriction after surgery, or may compromise the repair and risk re-tear of the meniscus.
Our long-term goal is to provide evidence-based guidelines for post-op rehabilitation protocols following meniscus repair that strikes the optimal balance between functional recovery and healing. Because the main function of the meniscus is to distribute mechanical loads, in order to provide these guidelines, the meniscus and repair site mechanics must be quantified during post-op protocols. Therefore, the objective of this proposal is to investigate in vivo knee biomechanics after meniscus repair: particularly the meniscus and cartilage displacement under conditions that simulate physiological loading in daily activities using a novel MRI-compatible loading device.<br />
We will use one of the most common scenarios for meniscus repair in a young active population: a vertical mattress suture repair for vertical longitudinal tear in the posterior horn. Post-op meniscus healing is usually 6 weeks after repair, and patients start high impact training (e.g. running, jumping…) at 3 months post-op. Therefore, we aim to assess knee mechanics at two time points: early after surgical repair (1 month), and when they are ready for high impact activities (3 months).
We have developed an MRI-compatible knee loading device that can apply physiological joint loading. We have successfully used this device to quantify meniscus and cartilage displacement in the knees of healthy volunteers, by applying compression with the knee in extension and flexion. We will use these innovative methods to pursue the following Aims:
Aim 1: Test the hypothesis that meniscus and cartilage mechanical function under physiological loading conditions is partially restored to intact values at 1-month post-surgery, and fully restored at 3-months under current post-operative rehabilitation protocols.
Aim 1A: Determine the normal variation between right and left knees in healthy controls who are age- and sex- matched to the surgical patients to establish the detectable change for the operative subjects.
Aim 1B: Measure meniscus and cartilage mechanics in surgical patients to determine whether the operative knee has reached functional equivalency comparing to the contralateral non-operative knee, with preserved repair integrity.
This study will establish, for the first time, in vivo quantification of knee biomechanics following loading that simulates physiological loading conditions to the knee joint. The next step in this project will be to expand the rehabilitation protocols being evaluated (either more conservative or aggressive), with the goal of determining a protocol that maximizes early weight-bearing and range of motion while preserving healing of the repaired meniscus.

Ovarian cancer (OVC) is the deadliest gynecologic cancer and the fifth leading cause of cancer-related death among US women. A critical limitation in fighting OVC is the inability to detect OVC early (stages I & II). Detection at early stages improves five-year survival to ~95% compared to ~57% overall. The five-year survival further drops to ~36% for black women, and is also significantly lower for women with high body mass index (BMI). Currently, no reliable diagnostic markers are available that identify OVC early and accurately. We utilize female chickens (hens) as a model to identify early and accurate diagnostic biomarkers for OVC. Chickens (unlike mice) develop OVC spontaneously, just like humans. Moreover, disease progression and disease staging parallels human OVC. Further, total number of ovulation events (a risk factor, if high) and age of onset (menopause) are also similar across both species. The overarching long-term goal of this research is to design and validate a diagnostic test similar to a mammogram or a pap smear that can be routinely performed to identify OVC before it can be clinically identified.

Pilonidal disease is a painful skin infection that occurs in the crease of the buttocks near the tailbone (gluteal cleft) and affects approximately 1% of the population between the ages of 15-30 years. Medical therapy of pilonidal disease is largely dependent upon continued hair removal, meticulous hygiene to the area, and recurrent courses of antibiotics with intermittent incision and drainage procedures. However, recurrent pilonidal disease after initial incision and drainage can be as high as 40% and wound complications after resection have been reported to be as high as 30%. Recurrence often leads to a substantial burden on patients with dependence on caregivers, long term disability, reduced quality of life, and social withdrawal.

Chronic hair removal to the gluteal cleft is a consistently recommended therapy to prevent pilonidal disease recurrence. Research has demonstrated the efficacy of laser hair depilation of the gluteal cleft to reduce pilonidal disease recurrence compared to standard of care in adult and adolescents. A recent randomized controlled trial (RCT) demonstrated laser hair depilation significantly decreased recurrence rates at 1 year from 33% to 10%. However, in exploratory secondary analyses, there was potential heterogeneity of treatment effect based on self-reported race and ethnicity. Although limited by small numbers of patients in many of the racial and ethnic groups, there was a significant reduction in disease recurrence among Non-Hispanic White patients with laser treatment but no change in recurrence among all other groups. This study will investigate the effects of laser hair depilation on pilonidal disease recurrence among those with darker skin types. We will perform a pilot RCT comparing laser hair depilation and mechanical/chemical depilation to mechanical/chemical depilation alone. The rate of recurrence of pilonidal disease at 1 year will be compared between the two groups. We hypothesize that laser hair depilation will lead to lower rates of disease recurrence at 1 year follow-up compared to standard of care. We will also compare differences in the 1-year morbidity associated with pilonidal disease between the two groups. We hypothesize that laser hair depilation will lead to less morbidity including less disability, higher health related quality of life, higher healthcare satisfaction, and fewer procedures, surgical excisions, and post-operative complications.

The estimate of the efficacy of laser hair removal to decrease pilonidal disease recurrence in darker skin adolescents/young adults from this pilot study is needed to design and perform a critical multi-institutional trial. A focused study of laser hair depilation in patients with darker skin types is essential to promote equitable health care across all patients with pilonidal disease by targeted inclusion of underrepresented populations with darker skin types in a therapeutic trial. Community engagement and partnership with Delaware communities that represent these populations will be critical to engage these underrepresented populations in research and to disseminate and implement research findings into these communities.

Hypoxic ischemic encephalopathy (HIE) is a common birth injury that results from a lack of oxygen to the fetal brain at birth. Inflammation is a major contributor to this injury and about 50% of babies affected by HIE are born to mothers with signs of infection or inflammation at birth. Maternal infection increases the risk of being diagnosed with HIE by 8-fold. Despite this increased risk, we know very little about how maternal infection impacts the fetal brain during birth. Our laboratory has developed a mouse model of HIE that incorporates maternal immune activation as a preceding risk factor. Using this model, we find that brain macrophages have significant upregulation in genes that regulate epigenetic modification. Notably, these mice have motor deficits and gait changes typical of those seen in cerebral palsy. In human studies of children with spastic cerebral palsy, stable epigenetic changes were identified as early as 4 years of age. Preliminary analysis using a set of epigenetic markers in a group of neonates affected by chorioamnionitis revealed complete separation between control and infection groups, suggesting that epigenetic signatures from maternal inflammation may be present as early as birth. In this proposal we hypothesize that circulating macrophages harbor epigenetic signatures of maternal inflammation in HIE and that these cells convey the risk of maternal inflammation to the fetal brain. We propose to collect circulating monocytes from infants diagnosed with HIE with or without exposure to maternal inflammation to perform methylome analysis and analysis of initial neurologic outcome from NICU hospitalization. We will also perform a retrospective chart review of cases of HIE with or without exposure to maternal immune activation to determine how maternal inflammation initial and 24-month neurologic outcomes. Completion of the proposed study will determine how maternal infection impacts fetal brain development and compounds brain injury in infants suffering HIE and will potentially identify biomarkers for those infants at the greatest risk of neurological injury. This may allow us to stratify those infants who would not otherwise qualify for therapies in the future and may allow us to design novel neurotherapeutics.

Preventive interventions that support the mental health of early childhood educators using known levers for change such as stress reduction, positive collegial relationships, and effective collaboration are needed. This project focuses on the development and refinement of one such intervention called the RELATE intervention. The objective of this project is to (a) refine and expand, alongside early educators, the RELATE intervention aimed at reducing depression symptomology for this population and (b) determine the feasibility of this intervention in the early childhood settings. This first aim is to refine and expand the RELATE intervention for early childhood teacher by using feedback from researchers and teachers, assistant teachers, administrators, and early childhood coaches from partner sites. The second aim is to determine potential supports and barriers to implementing the RELATE intervention in early childhood settings. Using the race-conscious adaptation of the Consolidated Framework for Implementation Research, ten classroom teams will participate in the RELATE intervention training and provide monthly written feedback as they implement the RELATE strategies and receive coaching. Additionally, two observations of the use of training strategies in the classrooms along with notes on barriers and supports will be conducted. Interviews will be conducted with classroom team members at the conclusion of the intervention to gain final insights into implementation supports and barriers. This project’s contribution is expected to be significant because early childhood educators report higher levels of depression than average workers and are disproportionately women of color who face additional stressors that may threaten their mental health. Moreover, the proposed research is innovative because it challenges the status quo of individual-focused mental health interventions by addressing workplace mental health supports through skills training and ongoing coaching.

Triple Negative Breast Cancer (TNBC) is an aggressive subtype of breast cancer that lacks effective treatment options and is often diagnosed at later stages. Given the poor outcomes associated with TNBC, prevention is critical. Although both genomic and non-genomic risk factors have been identified, including genetic variants and environmental or lifestyle exposures, little is known about how these factors interact. These interactions are believed to occur well before the development of cancer, yet little is known about how they influence risk in healthy women, particularly those living in communities with high exposure to metabolic, environmental, and behavioral risk factors. We hypothesize that gene-environment interactions shape immune and inflammatory responses that may increase risk of TNBC.

This study will investigate how genetic susceptibility and cumulative neighborhood, and individual-level exposures contribute to systemic inflammation and immune profiles that may signal elevated risk for TNBC. In Aim 1, we will develop a community-based recruitment infrastructure to engage healthy women from high- and low-risk neighborhoods and collect blood and exposure data. In Aim 2, we will conduct a pilot study to examine the relationship between environmental exposures, genetic polymorphisms, and inflammatory biomarkers in these women. This research will lay the groundwork for developing immune-genotype-based biomarkers of risk and inform early intervention strategies to prevent aggressive breast cancers before they develop.

Electroencephalography (EEG) remains a cornerstone of epilepsy diagnosis, but EEG analysis relies heavily on the visual detection of overt epileptiform discharges such as spikes or seizures or focal slowing, which are often absent during routine clinical recordings. This diagnostic limitation contributes to delays in treatment, misdiagnosis, and under-treatment, particularly in pediatric populations, where early and accurate diagnosis is essential for optimizing developmental outcomes. Machine learning (ML) offers a powerful framework for uncovering subtle patterns in the interictal EEGs of a pediatric population, but it is critical that the metrics and underlying features, even if subtle, are interpretable by neurologists. By focusing on interictal EEG, this work has the potential to transform current diagnostic paradigms. This proposal aims to support clinical decision-making by identifying EEG features that may be subtle to the human eye but are statistically robust across large datasets (Specific Aims 1 and 2). By identifying oscillatory and transient EEG patterns as phenotypic signatures of underlying network changes associated with seizure risk, clinicians are provided with actionable insights that enhance diagnostic precision and prognostic accuracy. The first and primary specific aim of this project is to systematically identify and validate EEG signatures of epilepsy embedded in background EEGs collected from pediatric populations through an innovative machine learning based methodology with the potential to reliably detect interpretable EEG patterns in background EEGs. The efficacy of the computer-aided diagnostic in increasing conclusive detection will be evaluated on a set of EEG recordings that were diagnostically inconclusive, requiring supplementary recordings or other clinical data for confirmation. The second specific aim is to analyze the spatial patterns of these signatures to distinguish focal from generalized epilepsy and to correlate with the epileptogenic zone. Finally, the third specific aim, directly implicated for individuals living with epilepsy, is to support seizure prediction from the EEG biomarkers. Towards this goal, a critical gap is to better understand the time changes of EEG biomarkers during long-term recordings with seizure events. Together, the project’s significance is based on the widespread prevalence of epilepsy, along with the challenges in diagnosing epilepsy from EEG without overt waveforms, differentiating focal and generalized epilepsy, and predicting upcoming seizures.

Hematopoietic cell transplantation (HCT) is commonly used for blood cancer patients to restore the ability to make blood cells after chemotherapy or radiation therapy. Allogeneic HCT (allo-HCT) using hematopoietic stem cells (HSCs) from donors has become increasingly popular in recent years due to improved outcomes. Despite this, relapse rates post-HCT can range from 30-80% depending on various patient- and disease-related factors. Relapse can be treated using an approach called donor lymphocyte infusion (DLI) in which T lymphocytes (T cells) from the HSC donor are collected and infused intravenously. The T cells are alloreactive, especially in cases of human leukocyte antigen (HLA) mismatch, and can attack the patient’s cancer cells. This so-called graft vs leukemia (GvL) effect can be highly efficacious and can cure or prevent relapse in some patients. The alloreactivity of T cells, however, can also target the patient’s healthy cells and tissues and cause graft vs host disease (GvHD), which severely limits the application and potential of DLI, especially for patients whose donors are HLA mismatched. To overcome this hurdle, we propose to selectively turn on T cell alloreactivity in tissues where cancer cells accumulate but not in tissues and organs most at risk of GvHD. Blood cancer cells mostly reside in lymphoid organs including bone marrow, lymph nodes, and spleen. GvHD, however, mainly affects the skin, liver and intestine. We have engineered a completely novel treatment called Lymphoid tissue-Activated Donor T cells (LADTs) that turn on alloreactivity upon sensing the B cells that are numerous in all lymphoid organs or pre-B cells that are only located in the bone marrow. Both B cells and pre-B cells are rare in skin, liver and intestine. After showing that LADTs can function in cell cultures, we are designing animal studies to determine their ability to eliminate cancer cells without causing GvHD. To optimize the design of these in vivo studies, we will determine the values of key experimental parameters such as the degree of HLA mismatch and the number of B cells and pre-B cells in bone marrow by collecting data from relapsed allo-HCT patients previously treated at Nemours and by analyzing their banked bone marrow samples (Aim 1). To test the proof-of-concept of LADT-based DLI in a clinically relevant manner, we will determine the ability of LADTs to kill cancer cells in bone marrow samples of allo-HCT patients collected at the time of relapse (Aim 2). The project, co-led by a physician specialized in HCT and a translational scientist experienced in T cell immunotherapy development, should significantly advance the research of this potentially curative approach to clinical stage.

Bronchopulmonary dysplasia (BPD) is the most common chronic respiratory disease in infants and has life-long morbidity risks for disorders such as chronic obstructive pulmonary disease (COPD). Standard-of-care treatment approaches are non-specific and non-localized, and accordingly, they fail to prevent morbidity risks and have adverse effects on neurodevelopment and immunity. Development of both more precise immunotherapies and more effective local treatment methods for affected infants would offer critically needed strategies to improve patient therapeutic outcomes while reducing side effects. Extracellular vesicles (EVs) are an emerging class of drug delivery vehicles with unique capacity for local cargo transport between cells, including efficient loading and transfer of nucleic acids. We recently discovered a new, nanoparticle (NP)-driven strategy to amplify and tune EV production in bone marrow macrophages (BMMs), with exciting potential as a localized treatment approach to generate and distribute therapeutically-relevant EVs in BPD lung in situ. Intriguingly, alveolar macrophage-derived EVs are known to provide signals to pulmonary epithelial cells that modulate inflammatory responses, with high untapped therapeutic potential in BPD given the vital role of macrophage-derived EVs (Mac-EVs) in immunoregulation and tissue repair. In this proposal we aim to 1. Understand the therapeutic benefits, localization behavior, and retention kinetics of Mac-EVs after lung delivery in murine models of BPD and 2. Develop NP-driven approaches to amplify Mac-EV production and maximize mRNA loading in Mac-EVs using BMMs and alveolar macrophages (MH-S) as versatile Mac-EV generation models. Our team of experts represents an innovative new collaborative group with an ideal fusion of engineering expertise in pulmonary drug delivery and nucleic acid delivery (Fromen and Sullivan teams, University of Delaware) with clinical and biomedical expertise in BPD treatment, pathogenesis, and preclinical model development (Alapati team, Nemours Hospital). Our proposal will establish both new fundamental understanding of the role of EVs in modulating BPD pathogenesis, as well as new enabling principles for the generation of translationally-relevant Mac-EV therapies, spanning the basic science to preclinical phases of the translational spectrum. These approaches will position our team for future advancement of a variety of advanced Mac-EV treatments for BPD, including future strategies aimed at harnessing NPs as a versatile tool relevant to both ex vivo and in situ / in vivo generation of therapeutically beneficial Mac-EVs in BPD and other lung diseases.

Trisomy X (47,XXX) occurs in 1 in 1,000 females and has been understudied and underdiagnosed, based on absent specific phenotypic markers. Emerging data show adults with Trisomy X have increased prevalence of metabolic dysfunction including obesity, dyslipidemia and type 2 diabetes. Studies in adult women also show altered ovarian function, increased risk of premature ovarian insufficiency, and earlier onset of menopause in addition to increased diagnoses of heart disease, renal anomalies and kidney disease. Critical gaps exist in characterizing physical features, ovarian, cardiometabolic and renal function during childhood, adolescence and young adulthood. Systematic data describing these are lacking. This study seeks to describe phenotypes and assess biomarkers of cardiometabolic, renal, and ovarian function during childhood and adolescence in girls and young women with Trisomy X using prior diagnoses and laboratory measures. Our overall hypothesis is that physical and biochemical markers will reveal variation and early signs of metabolic, renal and ovarian dysfunction during childhood and adolescence. Aim 1 assesses biomarkers of ovarian function (AMH, LH, FSH, inhibin B, estradiol) in girls with Trisomy X compared to age normative values. Aim 2 will assess biomarkers of cardiometabolic and renal function (glucose, glycosylated hemoglobin, lipid profiles, BUN, creatinine) in girls and young women with Trisomy X. In Aim 3, detailed phenotyping of participants to describe physical characteristics and growth patterns will occur. Data from this study will provide knowledge regarding multiple understudied aspects of Trisomy X, with focus on reproductive hormone, cardiometabolic and renal function during childhood and adolescence, an important and foundational phase of development. This will provide targets for subsequent investigation and early intervention to prevent comorbidities. Results can translate to improved clinical care by providing data for anticipatory guidance.

The recent defunding of SNAP-Ed represents a critical threat to nutrition education access for low-income children and families across the United States. SNAP-Ed (Supplemental Nutrition Assistance Program Education) has historically served as the primary federal mechanism for delivering evidence-based nutrition education to SNAP-eligible populations, reaching millions of children through schools, community centers, and youth programs. With this funding eliminated, communities face an urgent imperative to identify sustainable, effective approaches to maintain nutrition education delivery while preserving the evidence-based practices developed through decades of SNAP-Ed programming. While often supported for its strong evidence base, SNAP-Ed has also faced criticism for its programmatic rigidity, including requiring 6 weekly nutrition education series sessions, for example. It is unclear the extent to which these strict requirements accounted well for local contexts, cultural relevance, and practical realities of diverse implementation settings.

Cardiac surgery associated-acute kidney injury (CSA-AKI) is a frequently observed complication of surgical repair of complex congenital cardiac defects, with reported incidence of up to 55% in those patients requiring cardiopulmonary bypass (CPB). About 9% of these children develop severe AKI, with high risk for pathologic fluid overload. Moreover, CSA-AKI is independently associated with increased morbidity and mortality. Despite the unfavorable outcomes, there are no standard treatment strategies to prevent or ameliorate CSA-AKI. Since the predominant mechanism of AKI during CPB is renal vasoconstriction mediated by adenosine, one potential strategy for optimal peri-operative kidney function is to block this vasoconstriction response. Aminophylline is a nonselective adenosine receptor antagonist and phosphodiesterase inhibitor. In this study, we aim to investigate aminophylline as a potential therapy to prevent or ameliorate AKI during cardiac surgery. Aminophylline is safe, extensively used in newborns and young infants for treatment of bronchospasms, readily available and affordable. By blocking the afferent arteriole vasoconstriction, aminophylline has the potential not only to preserve urine output but also to prevent the drop of glomerular filtration rate (GFR). This unique effect cannot be replicated with any of the other available diuretic agents. We hypothesize that peri-operative aminophylline will preserve the kidney function and maintain the post-operative urine output. Furthermore, we hypothesize that aminophylline use will demonstrate reduced renal tubular damage as evidenced by urinary neutrophil gelatinase associated lipocalin (uNGAL) concentrations. We plan to test these hypotheses by conducting a randomized clinical trial in newborns and young infants undergoing cardiac surgery. The results of this preliminary study have the long-term potential to offer a new preventive treatment strategy for those very young children undergoing cardiac surgery to improve outcomes and ultimately save lives.

Nephrotic syndrome (NS) is a kidney disorder that causes severe protein loss including clotting factors in the urine and low blood albumin levels, leading to a very high risk of venous thromboembolism (VTE). To prevent these blood clots, many patients receive anticoagulation. Warfarin has historically been used, but it requires frequent blood testing and is difficult to manage due to food and drug interactions. Direct oral anticoagulants (DOACs) offer simpler dosing and fewer interactions and have become the standard for VTE prevention in many conditions. DOACs have also proven to have lesser bleeding events than Warfarin. Despite growing off-label use in NS, major clinical guidelines still advise against DOACs for prophylaxis because evidence is scarce and concerns remain about drug behavior in heavy proteinuria. We anticipate that our study will shed light on DOACs safety and efficacy in Nephrotic Syndrome patients at risk for thromboembolism, possibly influence future guidelines and help clinicians make informed decisions about anticoagulation options for this at-risk population.

This project seeks to address this gap by evaluating real-world outcomes among NS patients treated with DOACs compared to warfarin. Using 10 years of clinical data from a large integrated health system, we will compare VTE prevention, bleeding events, and treatment patterns in adults with NS who received prophylactic anticoagulation before any initial clotting event. We will also examine how kidney function, albumin levels, and NS subtypes influence these outcomes.

The results will provide timely evidence to inform clinical decision-making and improve safety for a vulnerable population. By clarifying whether DOACs are an effective and practical alternative to warfarin, this study may support future modifications to national guidelines and serve as a foundation for larger multicenter studies or prospective trials. Ultimately, this work aims to help clinicians tailor anticoagulation strategies that reduce preventable blood clots while minimizing harm for patients living with nephrotic syndrome.

Adolescent depression is a significant public health concern with approximately 20% of adolescents experiencing at least one major depressive episode in the past year (NIMH, 2023). The effects of adolescent depression can be catastrophic with 10% of high school students attempting suicide annually (CDC, 2023; Verlenden et al., 2024). Depression often begins during adolescence, or earlier, and follows a recurrent course associated with more negative outcomes and impairment along a range of important psychosocial domains that can persist into adulthood (Wilson & Dumornay, 2021). Integrated primary care (IPC) settings provide opportunities to prevent, screen, and treat depressive symptoms. IPC settings have been regarded as the “de facto” mental health system since the 1970s in part due to opportunities to promote service utilization (Hodgkinson et al. 2017), reduce stigma for mental health treatment (Miller-Matero et al., 2019), and reduce the negative impact of social determinants of health (Chakawa et al., 2020). The American Academy of Pediatrics (AAP) recommends that all youth 12 and older be screened for depression annually and those who screen positive or endorse suicidal ideation be referred to psychotherapy and initiate medication treatment (Hua et al., 2024). Preliminary research indicates that pediatric primary care providers (PPCPs) may not consistently adhere to national guidelines surrounding screening and treatment for adolescent depression. PPCPs may also fail to consistently connect adolescents with onsite and community mental health providers and may fail to initiate referrals to crisis services when suicidal ideation is endorsed (Hoffses, 2024). These findings have critical implications for adolescents seeking to connect to mental health services; if PPCPs do not provide information and resources to connect adolescent to mental health treatment, youth may never receive the care they need. The aims of this work are to identify barriers and facilitators reported by PPCPs in connecting adolescents with depressive symptoms to mental health treatment and identify implementation strategies to enhance PPCP uptake and adherence to AAP guidelines on connecting adolescents with a positive depression screen to mental health treatment. These findings are expected to have critical implications for adolescents seeking to connect to mental health treatment and are expected to inform the delivery of mental health services in IPC and reduce rates of depression for at risk adolescents. Results of this study will serve as preliminary data for future grant applications to test identified implementation strategies, with the goal of improving access to mental health services for adolescents.

More children are being diagnosed with autism than ever before. When parents of autistic children try to learn about this complex condition, they are met with several barriers. First, online informational content is readily accessible, but parents are likely to encounter autism information that is either inaccurate or that depicts autism in an overly negative light. Further, the autism information provided through widely accessible resources (e.g., via online sources) does not reflect the diverse social circumstances of families at Nemours, where over half of families have low access to socioeconomic resources. Overall, families of autistic children want autism educational resources that are 1) accurate and holistic in their portrayals of autism and 2) applicable to families from diverse social circumstances.

This community-based, mixed methods project has three aims. The first aim is to characterize, through qualitative interviews, the informational needs of autistic children’s families at Nemours. The second aim uses mixed methods to identify neurodiversity-affirming and socially contextualized methods to maximize families’ use and access to autism educational materials. Qualitative interviews will query families on strategies that optimize the usability and accessibility of autism educational materials. Quantitative questionnaires will ask families to rate existing autism educational materials. The third aim is to adapt, refine, and develop digital educational resources on autism that healthcare professionals can provide to families after their child is diagnosed with autism.

Purposive sampling will be used to ensure the recruitment and enrollment of families across the socioeconomic/socioecological spectrum. For Aims 1 and 2, we hypothesize data gathered may differ as a function of family socioeconomic/socioecological resources. Interview data will be analyzed using content and applied thematic analyses. Questionnaire data will be analyzed using descriptive statistics (means, SDs). We will engage with our community advisory board and healthcare stakeholders to consider the findings of Aims 1 and 2 and iteratively develop prototype educational materials designed to fill identified informational gaps (Aim 3).

The direct outcomes of this project will position me to subsequently examine families’ satisfaction and the efficacy of developed materials to increase families’ autism knowledge and empowerment. Further, this project has the potential to positively impact service delivery by developing an innovative solution that service providers can feasibly implement within an overburdened service system to meet families’ autism service needs.

Balance is needed to maintain posture and prevent falls in everyday life. The medial (MG) and lateral (LG) gastrocnemius and soleus (SOL) muscles produce torque at the ankle for postural realignment. Having reduced plantar flexor strength is also linked with poorer balance, which may also lead to falls. Evidence also suggests that during quiet stance, both heads of the gastrocnemius and soleus have different activation rates in young adults and may play a more significant role in balance compared with overall strength. Considering this, it is important to gain a better understanding of the relationship between individual calf muscle activation, strength and balance in older adults so interventions may be implemented if deficits arise due to aging or neurological conditions. As such, it is the purpose of this study to investigate the relationship between calf muscle function and postural sway in older adults. It is hypothesized that muscle activation will have a stronger positive relationship with postural sway performance compared with peak plantar flexor strength.

Spinal muscular atrophy (SMA) is a neuromuscular disorder resulting from mutation of the ubiquitous survival motor neuron one (SMN1) gene. SMA is the leading genetic cause of infant mortality. With the significant advances in our understanding of SMA in the past few decades, the mechanisms of neuropathology in this disease are not well understood. Similar to motor neuron degeneration, we reported lobule selective degeneration of cerebellar Purkinje cells (PCs) in SMA mice models. However, the cause for the selective vulnerability of PCs are unknown. Uncovering the causes of PC vulnerability will not only deepen our understanding of SMA pathology, but lead to more targeted treatment strategies to prevent neurodegeneration, and increase our understanding of cerebellar function. We hypothesize that specific properties of function, morphology, and/or gene expression are particularly dependent on optimal SMN function and susceptible to its loss in SMA.

The objective of this application is to extend our previous work by elucidating, at the single-cell level, the functional, morphological, and gene expression specializations of cerebellar PCs that are related to their vulnerability to low SMN expression with the novel three-in-one technique: patch-seq. The three specific aims of this study will provide evidence to test our hypothesis.

Specific aim 1: To identify physiological specializations in PCs associated with lobule selective degeneration in SMA mice. We will compare the membrane properties, synaptic transmission, and spontaneous firing

patterns of PCs between the least affected lobules (anterior - simple/III) and most affected (posterior - ANcr2/VII) lobules in control and SMA mice at very early-stages of PC degeneration.

Specific aim 2: To characterize the morphological specializations of PCs associated with lobule selective degeneration in SMA mice. During whole-cell patch-clamp studies carried out in aim 1, we will fill cells with biocytin allowing reconstruction and analysis of the microstructure of the recorded PCs.

Specific aim 3: To identify gene expression biomarkers associated with the functional and morphological specializations of lobule selective PC degeneration in SMA mice. We will use single cell RNA-seq and differential gene expression analysis to identify biomarkers associated with the lobule specific functional and morphological specializations correlated with vulnerability to low levels of SMN expression we expect to observe in aims 1 and 2.

Selective PC degeneration in SMA mice suggests that the pathology of SMA is broader than motor neuron degeneration. This project will be the first to unravel lobule specific functional and morphological specializations and defects in SMA, and the differences in gene expression of cerebellar PCs that underlie these properties using innovative patch-seq single-cell approaches. The results will provide critical information necessary for understanding both the molecular basis of specialization in cerebellar lobules, and the selective vulnerability of PCs in SMA, providing new targets for more effective treatment to extend life and preserve function in SMA patients, and potentially other neurological conditions affecting the cerebellum.

The reticulospinal tract (RST) is a secondary motor pathway, supporting the corticospinal tract (CST) in tasks such as locomotion, reaching, and grasping. The RST is relevant for motor recovery after CST lesions, such as stroke. There is currently a debate on whether increased RST drive is supportive of motor recovery, or a sign of maladaptive plasticity. A primary reason for the unresolved debate is the lack of methods for measuring function of the reticular formation (RF), a collection of nuclei in the brainstem that originate the RST.

Our group has recently developed StretchfMRI, a non-invasive imaging method that has enabled us to map motor function in the RF in-vivo for the first time. StretchfMRI is enabled by the unique integration of robotics, surface electromyography, and functional magnetic resonance imaging (fMRI). In this project, we aim to study the role of the RST in post-stroke individuals. Our central hypothesis is that the RF is hyper-excitable in post-stroke individuals. To test this hypothesis, we will measure stretch-evoked fMRI signal in the RF during extension perturbations applied to the hemiparetic and non-hemiparetic wrist of 24 post-stroke individuals. We will test whether stretch-evoked fMRI signal is greater for perturbations applied to their hemiparetic arm, compared to the non-hemiparetic arm, and establish the relationship between RF function and post-stroke motor impairment.

The main goal of this project is to understand how different neural pathways contribute to motor control, and whether secondary motor pathways may support post-stroke motor recovery in presence of damage to the corticospinal tract. The knowledge gained will support new interventions that modulate function in specific pathways to enhance motor recovery after stroke.

Those with mild cognitive impairment and Alzheimer’s disease have a reduced capacity to rapidly decide and move, which has been linked to fatal outcomes including broken hips from falling and car accidents. Mild cognitive impairment leads not only to impaired decision-making, but also movement deficits that predict the development of Alzheimer’s disease. Recent behavioral work has suggested a common mechanism that throttles the speed of both decisions and reaching movements, which is supported by converging neural evidence that finds an interaction between decision-making and movement (motor) circuits. Yet it remains unknown how the interplay between decision-making and motor neural circuits becomes impaired and impedes rapid responses for those with mild cognitive impairment. Here we test the central hypothesis that there is an impaired interaction between decision-making and motor neural circuits with mild cognitive impairment. To test this idea, we have developed a novel reaching paradigm that provides a real time readout of the evolving deliberation—prior to a final decision—via hand movement. Magnetic resonance elastography (MRE), a state-of-the-art imaging technique, will be used to quantify viscoelastic (stiffness and damping) properties in brain areas linked to decision making and motor control. Further, we will combine computational models of decision-making and motor control to capture human movement behavior with or without mild cognitive impairment in a theory driven manner. In Aim 1 we will use human reaching experiments to establish that mild cognitive impairment disrupts the interplay of decision making and motor control. We expect those with mild cognitive impairment will display less hand movement relative to neurotypical age matched controls. In Aim 2 we will use MRE to elucidate whether brain stiffness in decision-making and motor brain regions are related to altered movement behavior. Compared to neurotypical age matched controls, we expect that those with mild cognitive impairment will have lower brain stiffness in decision making and motor areas. Further, we expect lower brain stiffness will be related to decreased hand movement. The expected outcome is a mechanistic understanding of how impaired decision making and motor neural circuits impact movement for those with mild cognitive impairment. Our work will have a positive impact by providing sensitive biomarkers for early detection of Alzheimer’s disease and paving the way for informed and effective neurorehabilitation.