United States

Background
The WHO reports that elderly people have the highest risk of death or serious injury arising from a fall (WHO, 2021). This study utilized the 2020 National Inpatient Sample (NIS) database from the Healthcare Cost and Utilization Project to determine which diagnostic conditions were most likely to be seen in elderly fall patients. 

Methods
The 2020 NIS database was used to identify the ICD-10 diagnostic codes in elderly patients that were associated with a fall code. Analyses used the first 3 digits of the diagnostic code. 

Results
Analyses revealed that 4.7% of the in-patient sample (6,471,165 patients) had a diagnostic fall code. We referred to these patients as falls patients (FPs). The incidence of falls increased dramatically with age, as 7.6% of patients &gt;50 years of age (YOA) and 12.2% of patients &gt;70 YOA were FPs. Moreover, 90.6% of the FPs were &gt;50 YOA. As expected, 90.2% of the FPs also had a code for a musculoskeletal disorder or injury. Interestingly, 26.2% of FPs &gt;50 YOA were associated with dementia or a neurodegenerative disorder resulting in dementia. The most common diagnostic categories seen in FPs &gt;50 YOA were Circulatory (88.6% of patients), Psychiatric (55.6%), Neural (48.1%), and Heme/Immune diagnoses (47.1%). Interestingly, 73.1% of FPs &gt;50 YOA had either a Psychiatric or Neural code. The highest percentage of Circulatory codes seen in FPs &gt;50 YOA were Hypertension (44.1%), Atrial fibrillation (26.4%), Chronic heart disease (26.2%), Hypertensive chronic heart/kidney disease (22.7%), Heart failure (22.0%), and Hypotension (10.5%). The highest percentage of Psychiatric codes seen in FPs &gt;50 YOA were Dementias (22.4%), Depression (15.8%), Anxiety (15.6%), Nicotine dependence (12.5%), and Alcohol abuse (8.2%). The highest percentage of Neural codes seen in FPs &gt;50 YOA were Metabolic encephalopathy (12.3%), Chronic pain (11.3%), Insomnia (10.8%), Epilepsy (4.91%), Alzheimer’s (4.4%), and Parkinson’s (3.42%). 

Conclusion
As expected Circulatory codes were common in the FPs &gt;50 YOA, as the incidence of these diagnoses increases dramatically with age. It was, however, surprising to see the high incidence (73.1%) of Psychiatric and Neural codes in the FPs &gt;50 YOA, particularly codes relating to dementias. It is important to note that it was impossible in this study to determine if any falls occurred in the hospital, although we presume that most falls occurred outside of the hospital and may have been associated with hospital admission. Studies like this may help to develop fall prevention strategies by isolating specific diagnostic patient profiles in the elderly.

AMA Research Challenge 2024 

Diagnostic codes associated with falls, a focus on psychiatric and neural diagnoses in the elderly in-patient sample

Welcome to the **largest national, multi-specialty research** event featuring research from more than 750 medical students, residents, fellows, and international medical graduates.

Use the buttons below or navigation menu to access the event.
 

[![](https://assets.underline.io/markdown_image/1/image/20e1e77aa954efeac7a15d84c85eb383.jpg)](https://underline.io/events/468/sessions?searchGroup=all)
 
[![](https://assets.underline.io/markdown_image/1/image/068522d81aca0b68a369787c7dfccd1b.jpg)](https://underline.io/events/468/posters?searchGroup=all)
 
[![](https://assets.underline.io/markdown_image/1/image/d3971ece79d5e5575b178bdc2c8bbfae.jpg)](https://underline.io/events/468/help-desk)
 

#### Event Details

The 2024 American Medical Association Research Challenge includes a poster symposium hall and a semifinals hall with the top scored research. **Log in at any time throughout the event to:**

* **View posters and presentations** on a variety of topics and specialties
* **Score posters in the semifinals** to help decide the top five finalists (AMA member exclusive opportunity)
* **Connect with participants** and provide feedback to help advance their research

At the conclusion of this event, the AMA will announce the top five finalists who will advance to the AMA Research Challenge finals, present their research to an elite panel of judges, and compete to win a $10,000 grand prize presented by Laurel Road. 

### Grand Prize Sponsor

AMA would like to thank Laurel Road for sponsoring a $10,000 grand prize for the winner of the AMA Research Challenge. Laurel Road is not involved in winner selection process. The winner will be announced on February 20, 2025 during the AMA Research Challenge finals. Payment of the grand prize will be directly from Laurel Road. See the AMA Research Challenge Official Rules for terms and conditions.

[![](https://assets.underline.io/uploads/markdown_image/1/image/611e331d4a0eba0920f240cfc04b4d77.png)](https://www.laurelroad.com/)

You need to log in with the email address you registered with. Access credentials have been sent to your email. 

Please be sure to check your spam and other email folders if you do not see an email confirmation right away.

Please log in to explore this event.

**There is no registration fee for this event! **

If you have used the AMA Research Challenge platform previously, please click the ‘Log-in’ button below. If you do not remember your password, please click “forgot password” and follow prompts to reset your password. After resetting your password, please refresh your webpage to access the platform.

Register to explore research in a variety of topics. AMA members get exclusive access to score posters in the semifinals. 

**Please use this link [here](https://www.ama-assn.org/member-benefits/events/ama-research-challenge) to register for free.**

Registration Is Required

Join the AMA Research Challenge poster symposium and semifinals to explore top research posters and score posters in the semifinals (an AMA member exclusive opportunity).

Background
The WHO reports that elderly people have the highest risk of death or serious injury arising from a fall (WHO, 2021). This study utilized the 2020 National Inpatient Sample (NIS) database from the Healthcare Cost and Utilization Project to determine which diagnostic conditions were most likely to be seen in elderly fall patients. 

Methods
The 2020 NIS database was used to identify the ICD-10 diagnostic codes in elderly patients that were associated with a fall code. Analyses used the first 3 digits of the diagnostic code. 

Results
Analyses revealed that 4.7% of the in-patient sample (6,471,165 patients) had a diagnostic fall code. We referred to these patients as falls patients (FPs). The incidence of falls increased dramatically with age, as 7.6% of patients >50 years of age (YOA) and 12.2% of patients >70 YOA were FPs. Moreover, 90.6% of the FPs were >50 YOA. As expected, 90.2% of the FPs also had a code for a musculoskeletal disorder or injury. Interestingly, 26.2% of FPs >50 YOA were associated with dementia or a neurodegenerative disorder resulting in dementia. The most common diagnostic categories seen in FPs >50 YOA were Circulatory (88.6% of patients), Psychiatric (55.6%), Neural (48.1%), and Heme/Immune diagnoses (47.1%). Interestingly, 73.1% of FPs >50 YOA had either a Psychiatric or Neural code. The highest percentage of Circulatory codes seen in FPs >50 YOA were Hypertension (44.1%), Atrial fibrillation (26.4%), Chronic heart disease (26.2%), Hypertensive chronic heart/kidney disease (22.7%), Heart failure (22.0%), and Hypotension (10.5%). The highest percentage of Psychiatric codes seen in FPs >50 YOA were Dementias (22.4%), Depression (15.8%), Anxiety (15.6%), Nicotine dependence (12.5%), and Alcohol abuse (8.2%). The highest percentage of Neural codes seen in FPs >50 YOA were Metabolic encephalopathy (12.3%), Chronic pain (11.3%), Insomnia (10.8%), Epilepsy (4.91%), Alzheimer’s (4.4%), and Parkinson’s (3.42%). 

Conclusion
As expected Circulatory codes were common in the FPs >50 YOA, as the incidence of these diagnoses increases dramatically with age. It was, however, surprising to see the high incidence (73.1%) of Psychiatric and Neural codes in the FPs >50 YOA, particularly codes relating to dementias. It is important to note that it was impossible in this study to determine if any falls occurred in the hospital, although we presume that most falls occurred outside of the hospital and may have been associated with hospital admission. Studies like this may help to develop fall prevention strategies by isolating specific diagnostic patient profiles in the elderly.

poster

Abstract Title: 
Financial Conflicts of Interest of Physicians Followed by Leading Medical Journals on X (Twitter)
Background
Social media platforms like X (Twitter) have become important for disseminating medical research. However, financial conflicts of interest (FCOIs) among physicians on these platforms may influence content. This study aimed to assess FCOIs among physicians followed by top medical journals on X.

Methods
We identified US physicians followed on X by The New England Journal of Medicine, The Lancet, and JAMA. Using the Open Payments database, we analyzed general payments, research payments, and ownership interests received by these physicians from 2014-2021. We also examined payments to editorial board members and analyzed a subset of posts from physicians with FCOIs. We analyzed the posts using named-entity-recognition, topic modeling, and sentiment analysis. 

Results
Of 253 US physicians followed, 54.9% received general payments totaling $19.2 million. Specialists received significantly higher median payments than primary care physicians ($540 vs $0, p=0.0003). Among editorial board members with X accounts followed by their journals, 73.3% received general payments.When compared to general payments, only 9.5% of physicians received research payments. Analysis of physicians’ X activity showed those with FCOIs had significantly more followers than those without (64,490 vs 17,724, p=0.044). Topic modeling of Posts from physicians with FCOIs revealed frequent discussion of industry-related topics. Physicians with declared financial conflicts of interest tend to include academic links in their tweets slightly more often (20.93%) compared to those without FCOI (19.07%). Subspecialities of internal medicine and surgery tended to share more academic sources. Physicians with FCOIs tend to discuss more specific medical topics (e.g medical education, FDA and regulations, vaccines) compared to those without FCOIs, who often have "other" as their most common topic.The average sentiment scores vary between -0.037 and 0.576, with most scores being positive.

Conclusion
A majority of physicians followed by top medical journals on X have FCOIs, with some receiving substantial payments. This raises concerns about potential bias in social media content from these influential voices. Greater transparency is needed regarding FCOIs on social media platforms used for medical information dissemination. Medical journals should consider implementing policies for disclosing FCOIs when interacting with or promoting content from physicians on social media. Physicians should also provide reference to their FCOIs to avoid the appearance of bias and maintain public trust.


Financial Conflicts of Interest Among Physicians Followed by the Leading General Medical Journals on X

Objective: It is unclear if Medicare Advantage (MA) benefit design may influence when MA beneficiaries receive elective treatments such as spinal fusion surgery. We characterize the impact of MA insurance status on the timing of spinal fusion surgeries across calendar years.

Study Design: Cross-sectional event study.

Methods: Using the national sample of Medicare beneficiaries from 2016-19, we identified 175,181 Medicare beneficiaries who received a spinal fusion surgery from November through February each year. We conducted a descriptive analysis and a series of event study difference-in-difference linear regressions to answer our research questions. 

Results: Among Medicare enrollees who received spinal fusion surgeries between November and February, 2,460 (2.0%) beneficiaries in Traditional Medicare (TM) at the time of surgery and 1,978 (3.8%) beneficiaries in MA at the time of surgery switched plans from one calendar year to the next. Among all spinal fusion surgeries received by Medicare enrollees and controlling for confounders, beneficiaries enrolled in MA are 1.6 percentage points (95% CI [-2.2%, -1.0%]; P<0.001) less likely to have been at the beginning of the calendar year compared to TM surgeries. 

Conclusions: The budgetary impact of enrollees switching between MA and TM plans to gain access to spinal fusion surgery is likely low. MA enrollees may be accelerating care at the end of the calendar year due to benefit design differences between MA and TM, such as annual spending caps or out-of-pocket maximums. Dual beneficiaries are not subject to this incentive structure, providing further evidence that the benefit design of MA may be driving the timing of surgery.

Impact of Medicare Advantage Insurance Status on Spinal Fusion Surgery Rates Across Benefit Year Transition

Background
The Patient Protection and Affordable Care Act (ACA) mandated that hospitals conduct a community health needs assessment (CHNA) every 3 years to maintain 501(c)(3) tax exempt status. In this report, a hospital defines the community it serves and assesses the health needs of that community. Following this, they develop a community health implementation plan (CHIP) that addresses each priority documented in the CHNA. Multiple studies have documented discrepancies between CHNA and CHIP. Additionally, following the passing of the ACA, nonprofit hospitals only increased average spending on community benefits by 0.5%, indicating a missed opportunity for significant community health improvement. Likewise, the amount of benefit nonprofit hospitals provide to their communities is often much less than their received tax exemption. Previous studies used small sample sizes, limited by location or priority assessed. For a comprehensive understanding of this misalignment, a national sample needs to be evaluated. 

Methods
The study extracted CHNA and CHIP reports from nonprofit hospitals (n=450). The sample was stratified by USDA rural-urban continuum code (RUC), a nine-category system dividing communities by population and proximity to a metro area. Adult and youth priorities were developed through a content analysis by drawing priorities and health categories from previous research. A coding protocol was created, and interrater agreement was tested on a 20% stratified random sample for the adults. Youth data was double coded because of smaller sample size. The protocol defined 43 categories (e.g., preventative healthcare, mental health, and education). 

Results
A series of 2 (CHNA, CHIP) x 2 (metro, nonmetro) repeated measures were performed. Results revealed a main effect for adult priorities where hospitals CHNA did not match CHIP for three priorities: coordination of care, elder care, and Alzheimer’s. Of those priorities, two had significant interactions with community size. A significant mismatch existed for four youth priorities: maternal and child health, affordable housing, employment, and disability. From those priorities, two had significant interactions with community size. 

Conclusion
Generally, CHIPs demonstrate that hospitals execute a plan for most priorities identified in the CHNA. However, a mismatch exists in adult priorities targeting aging populations and youth priorities targeting social determinants of health. Discrepancies could exist for reasons including cost, feasibility, and hospital scope. The IRS’s goal is for non-profit hospitals to address unmet needs in their communities using the tax benefits they receive. These discrepancies between CHNA and CHIP may demonstrate that these goals are not being met for certain communities. 


Misalignment between Community Health Needs Assessments and Community Health Implementation Plans in Rural and Nonrural Hospitals

Abstract Title: Post-Procedural Outcomes in Tibial Angioplasty vs Femoropopliteal Angioplasty 

Authors: Ashley Newsholme, BS; Amina Namrouti, BS, Nicole Moscoso, BS, Grettel Castro, MPH, Rupa Seetharamaiah, MD; Georgeta Vaidean, MD,PhD 

Background: Peripheral artery disease is a prevalent condition affecting over 8.5 million Americans and can lead to chronic limb-threatening ischemia (CLTI), and Acute Limb Ischemia (ALI), necessitating surgical revascularization. Treatment for both CLTI and ALI commonly involves an endovascular lower extremity revascularization procedure (LEP), through either tibial or femoropopliteal angioplasty. Short-term complications include Major Adverse Cardiovascular Events (MACE) and Major Adverse Limb Events (MALE), but literature that compares the risk of MACE/MALE with type of LEP is lacking. Our aim is to determine if there is a difference in the 30-day postprocedural outcomes (MALE/MACE) in femoropopliteal (FA) versus tibial angioplasty (TA). 

Methods: This retrospective cohort study utilized the 2019-2021 National Surgical Quality Improvement Program (NSQIP) data. Adults above the age of 18 undergoing either procedure were included, while those with "missing" or "not documented" values were excluded. MALE and MACE were composite variables. MALE included major amputation, intervention of previously treated segment, severe procedural outcome, and loss of patency. MACE included myocardial infarction or stroke, and death. A univariate analysis was used to describe the sample, chi-square testing for baseline charcate4ristics and logistic regression analysis to obtain crude and multivariate adjusted odds ratios (OR) and 95% confidence intervals (CI). Adjusted variables include Age, Ethnicity, pre-procedural ABI, Diabetes, Smoking status and Dialysis for MALE. For MACE, the variables Age, Race, pre-procedural ABI, Diabetes Meletus, Smoking status, and Dialysis were adjusted. 

Results: Among the 8,319 patients, 5924 (71.21%) underwent FA while 2395 (28.79%) underwent TA. MALE incidence was 7.02% in FA and 8.14% in TA (p-value= 0.076). MACE incidence was 2.48% in FA and 3.26% in TA (p=0.048). Compared to FA, the association of TA with MALE revealed an OR:1.17, 95% CI: 0.98-1.40 in the crude model and OR:1.12, 95% CI: 0.91-1.37 in the multivariate adjusted model. Compared to FA, the association of TA with MACE revealed an OR:1.32, 95% CI: 1.00-1.75 on crude analysis and OR:1.16, 95% CI: 0.84-1.59) in the multivariate adjusted model. 

Conclusion: We found no significant difference in 30-day postprocedural outcome (MALE or MACE) between Tibial or Femoropopliteal angioplasty. 



Major Adverse Cardiovascular and Limb Events Post Tibial vs Femoropopliteal Angioplasty​

Purpose/Objective: 
Triple-negative breast cancer (TNBC) is an aggressive phenotype with patients having more limited survival compared with other breast cancer subtypes. The high rate of locoregional/distant recurrence after primary management implies that additional treatment is necessary to optimize the sensitivity of TNBC tumors to therapy including radiation. Our lab studies the role of VEGF/Neuropilin-2 (NRP2) in mediating cancer stem cell properties such as self-renewal and therapy resistance. We sought to identify a new therapeutic approach that can induce radiosensitization in TNBC. 

Materials/Methods: Cell viability in response to radiotherapy in patient-derived organoids (PDO) and xenografts (PDX) was determined using CellTiter-Glo. The VEGF/NRP2 pathway was disrupted in human BT549 and murine 4T1 models using RNA interference and a function blocking antibody (aNRP2-10) provided by aTyr Pharma. Nitrite levels, a surrogate marker for NOS2 activity, were measured using Measure-IT Nitrite Assay Kit. H2DCFDA was used to measure intracellular reactive oxygen species (ROS). Clonogenic survival assays were used to calculate radiosensitivity enhancement ratios (rER) in TNBC cell lines. For our in vivo models, 4T1 xenografts were given antibody treatment every 48 hours starting a day prior to 10Gy irradiation.

Results: We identified a novel radioresistant population expressing high levels of NRP2 and NOS2 and observed that NOS2 expression was dependent on VEGF/NRP2 via Gli1 transcription. Interestingly, reducing NRP2 expreession increased radiation-induced ROS because of reduced NOS2/NO levels. The downstream consequence of higher NOS2/NO was increased Nrf2 activation via KEAP1 S-nitrosylation; thus, inducing expression of antioxidant response elements. TNBC cells were radiosensitized by knocking down NRP2 (rER 1.19 – 1.34). We saw similar results when using aNRP2-10 vs IgG (rER 1.30 – 1.37). Ectopic expression of NOS2 and constitutively active Nrf2 rescued the radioresistance phenotype in the NRP2 knock-down cells (rER 0.73 – 0.82). Consistent with the cell line data, the clinically relevant models (PDOs and PDXs) had a synergistic reduction in cell viability in response to aNRP2-10 prior to radiotherapy. We also observed a significant reduction in tumor volume and increase in necrosis in vivo in response to VEGF/NRP2 inhibition during radiotherapy.

Conclusion: Inhibiting VEGF/NRP2 induces radiosensitization of TNBC by decreasing NOS2/NO levels and, consequently, Nrf2-mediated antioxidant gene expression. The role of Gli1 as a transcription factor for NOS2 is a novel finding that provides an unexpected link between VEGF/NRP2 and redox homeostasis in response to radiation. Current Nrf2 inhibitors are non-specific and have high toxicity; thus, making VEGF/NRP2 inhibition more favorable. Our data provide a novel therapeutic strategy for targeting TNBC breast cancer.


Enhancing the radiosensitivity of triple-negative breast cancer by targeting VEGF/Neuropilin-2

Background Follicular lymphoma is a common lymphoma of germinal center B cells. Accurate grading of follicular lymphoma is challenging, as it relies on accurate identification and quantitation of neoplastic centrocytes and centroblasts. within the tissue sections. Grading is consequential in determining plans and treatments for patients, however there is a high degree of inter-observer variability among practicing physicians. Artificial intelligence (AI) approaches present an opportunity to refine the grading process for follicular lymphoma and provide better reproducibility compared to that achieved by pathologists and ease the workload of physicians in a reviewing a high volume of slides. The aim of this study is to develop and incorporate an AI-driven follicular lymphoma grading model into routine pathology practice to improve patient care.
Methods 1313 H&E-stained Whole Slide Images (WSI) with diagnoses of follicular lymphoma from 441 distinct patients submitted to hematopathology service in our institution during 2019- 2021 were collected. The WSI were categorized into grades 1-2, 3A, and 3B based on clinical pathology report and then split into training (70%), validation (15%), and test (15%) datasets at individual patient level following by pre-processing. The WSIs were processed at resolutions of 20x, 10x, and 5x. A pre-trained ResNet-50 model was applied to the training set to extract features at the tile level. Multiple Instance Learning (MIL) with a Random Forest (RF) classifier was utilized to determine the slide's overall grade.
Results The neural network model classified the WSI into grades 1-2, 3A, or 3B with 85% accuracy overall. The tile-level ROC curves displayed AUROC scores of 0.95 for both 20x resolutions and 0.96 for the 5x resolution. Precisions were 87% for 20x, 84% for 10x, and 86% for 5x resolutions. For slide-level classification, we used concatenated features derived from the tile level. The slide-level AUROC revealed an AUC score of 0.94, accompanied by a precision of 80%.
Conclusion Our model has demonstrated efficacy and precision in predicting the grades of follicular lymphoma. Our model's accuracy surpasses reported inter-reader consistency among practicing pathologists. This model offers promise as an initial grading tool, facilitating pathologists' tasks by enabling them to validate preliminary findings. With use of this tool, we can ease the of grading and review of the high volume of slides that pathologists analyze daily.

Utilizing Deep Learning for Follicular Lymphoma Grading on Whole Slide Imaging

Background
The porcine model has been used in several preclinical contexts due to its anatomical and physiological homology with the human body. Here, we focus on the utility of the porcine knee in replicating human joint anatomy. Previous studies have identified the porcine stifle joint to have similarities to human with respect to the ACL, PCL, menisci, femoral condyle width, tibial plateau and interspinal width, and bone mineral density. However, there is mixed data on advantages of respective porcine breeds, sizes, and ages. This study aims to support existing data on the use of porcine knees for orthopedic research as well as add a new perspective on the use of adolescent porcine knees as an analogous preclinical model.

Methods
Yorkshire pigs aged 1-3 months were used to obtain sixty porcine stifle joints, which were dissected and analyzed for femur and tibia dimensions as well as medial and lateral meniscal height, width, and volume. Measurements were obtained via digital vernier caliper (precision of 0.01mm) and ImageJ extrapolation from standardized photographs. Scaling was performed with a reference ruler to ensure appropriate pixel-length ratio. Comparative analysis to human data was performed via PubMed literature search. Reliability was assessed with Intraclass Correlation Coefficients and unpaired Student's t-tests. 

Results
Joint measurements showed a mean of 64.67±4.12mm for medial femur AP width, 63.54±3.67mm for lateral femur AP width, and 61.73±3.23mm for femur medial-lateral width. The mean medial tibia AP width was 43.15±3.10mm, lateral tibia AP width was 40.07±2.17mm, and tibia medial-lateral width was 63.60±3.81mm. Compared to human measurements, this data in aggregate shows a 0.86 correlation coefficient, demonstrating a strong positive correlation between porcine and human anatomy. Additionally, the femoral intercondylar notch dimensions widened and became shallower with age, resembling the maturation patterns seen in human knee development.
For menisci measurements, there were notable differences between the medial and lateral measurements, with medial menisci averaging 13.31±1.86mm in width and lateral menisci at 15.38±1.9mm. The meniscal volumes, recorded at 4.30±0.13ml for medial and 5.9±0.29ml for lateral menisci, aligned closely with those reported in human studies, reported as 4.5ml and 5.0ml, respectively.

Conclusion
This investigation confirms that the porcine stifle joint closely mirrors human knee anatomy and additionally suggests that adolescent porcine samples may be used in preclinical orthopedic research. The identified age-related morphological variations further refine model selection, enhancing orthopedic research value. This information supports the porcine model as cost-effective and accessible for knee, ACL, and meniscus research.

Insights into Adolescent Porcine Meniscus Anatomy and Osteology for Knee Research

BACKGROUND: Immunotherapy in many tumors is limited by tumor heterogeneity and low mutational burden. While the vast majority of tumor-specific antigens (neoantigens) are identified through somatic mutations, recent studies indicate aberrant RNA-splicing (neojunctions) as a new source for targets. Our spatial splicing neoantigen identification platform (SNIPP) characterizes a novel class of clonally-expressed splicing-derived neoantigens that elicit a CD8+ T-cell-mediated tumor killing response. METHODS: SNIPP identified public neojunctions expressed in TCGA RNA-seq (positive sample rate (PSR) > 10%) and not in GTEx normal tissue RNA-seq data (PSR < 1%) across 12 cancer types. To characterize intratumorally-conserved neojunctions, we performed maximally-distanced multi-site biopsies (n=535) within glioma patients (n=56) and obtained RNA-seq data from each intratumoral site. This was additionally performed on publicly-available multi-site sequencing data from other cancer types. Two independent algorithms then predicted peptide processing likelihood and HLA-binding affinity of ASN candidates. Neoantigen-specific TCR sequences characterized from subsequent PBMC in vitro sensitization and 10x V(D)J scRNA-seq were transduced into CD8+ T-cells for immunogenicity and cytotoxicity assays against glioma cell lines. Differential gene expression (DESeq2) and gene set enrichment analyses (GSEA) is performed between tumor subtypes within cancers, and a Pearson correlation matrix is generated to identify correlated neojunctions with specific splicing factor expression. CRISPRi knockdown of select splicing factors followed by qPCR of their corresponding neojunctions are performed to determine the correlative mechanism for neojunction expression. RESULTS: Our pipeline identified 789 public neojunctions, with 32 neojunctions concurrently identified in transcriptomic and proteomic glioma data and predicted to be presented by HLA-A*02:01 with high confidence. IVS and subsequent 10x VDJ scRNA-seq identified TCR clonotypes reactive against neojunctions in RPL22 (n=7) and GNAS (n=1), the latter being highly intratumorally-conserved (detected in > 90% of spatially-mapped biopsies across 17/56 patients (26.78%)). TCR-transduced T-cells demonstrated recognition and immunogenic activation against endogenously processed and presented neoantigens in various pan-cancer cell lines. Furthermore, IDH1-mutant oligodendroglioma samples demonstrated significantly elevated expression of neojunctions over IDH1-mutant astrocytoma and IDH1wt subtypes. Differential gene expression (DESeq2) identified decreased expression of splicing factors due to oligodendroglioma-specific co-deletion of Chromosomes 1p/19q. siRNA knockdown of these splicing factors (e.g. SF3A3, SNRPD2) in IDH1wt glioma cells resulted in significantly increased expression of corresponding neojunctions. CONCLUSION: Our unique SNIPP neoantigen discovery platform identified novel public tumor-wide splice-derived neoantigens and reactive TCRs, and its pan-cancer application characterized candidates targetable across multiple diseases.

Tumor-wide RNA splicing aberrations generate immunogenic public neoantigens across cancers

Background
Hereditary Hemorrhagic Telangiectasia (HHT) is an autosomal dominant inherited syndrome of dysregulated angiogenesis presenting with arteriovenous malformations in multiple organ systems. Despite the significant complications associated with HHT, FDA-approved therapeutic options remain unavailable. In HHT, bypassing the ACVRL1 mutation is critical to restore downstream TGF-β signaling. Extracellular vesicle (EVs) mediated delivery of recombinant proteins offer a pioneering approach, in biocompatible fashion, to treat dysregulated angiogenesis in HHT. Engineered Bone Morphogenic Protein 2 and 7 extracellular vesicles (BMP2/7-EV) may offer a potential therapeutic approach for HHT by bypassing the defective ACVRL1 and reactivating SMAD signaling.
Methods
We utilized an ex ovo chicken chorioallantoic membrane (CAM) model to investigate the effects of BMP2/7-EVs. Fertilized eggs were incubated at 37C and transferred to weigh boats on developmental day 3. On day 9, each respective treatment was applied: 50 µL of either PBS (vehicle control), ALK1-Fc (500 ng/mL), BMP2/7-EV (1.78e8 particles/ml), or a combination of BMP2/7-EV (1.78e8 particles/ml) and ALK1-Fc (500 ng/mL) onto three different areas of each embryo. Microscopy at days 9 through 11 was used to document the vessel development. IKOSA software was used to generate images quantifying branch number. Statistical significance compared to control was determined by two-way ANOVA with Tukey post hoc analysis.
Results
ALK1-Fc treatment at concentrations above 100ng/mL induced hypervascularization, demonstrated by an increase vascular branching in the native ex ovo CAM model. This effect was most pronounced at 500ng/mL. BMP2/7-EV treatment of native ex ovo CAM model and ALK1-Fc CAM model mimicking HHT behavior showed a trend towards vascular regression although the differences were not statistically significant in this pilot study. 
Conclusion
The CAM assay model demonstrates promising potential for HHT research, providing a cost-effective and ethical alternative to mammalian models, allowing direct visualization of vascular changes, and offering a platform for testing angiogenic modulators. While the BMP2/7-EV treatment showed a trend towards normalizing vessel formation in the ALK1-Fc model mimicking HHT, further optimization and repeated experiments are necessary to draw definitive conclusions. This approach not only enhances our understanding of HHT pathophysiology but also paves the way for the development of targeted therapies for HHT patients, potentially improving their quality of life. Future studies will focus on refining the model, conducting more extensive studies to ensure the reproducibility of our results and exploring how various factors influence vessel formation in our model.

Engineered BMP2/BMP7 extracellular vesicles as an anti-angiogenic strategy for Hereditary Hemorrhagic Telangiectasia Using an ex vivo Chick Chorioallantoic Membrane Model

INTRODUCTION: 
Post traumatic and chronic osteoarthritis (OA) share the underlying feature of damaged hyaline cartilage and exposure of type-II collagen (CII) in the damaged tissue due to erosive proteolytic degradation triggered by low grade ongoing inflammation. Adipose tissue derived mesenchymal stem cells (ADSC) have shown efficacy in the treatment of osteoarthritis through their ability to mitigate the host inflammatory response through paracrine signaling and possess capability for chondrogenic differentiation. However in vivo, very few ADSC bind to the damaged cartilage or differentiate in the OA lesion. Our study offers a solution to these obstacles through the creation of injectable type I collagen (Col I) microspheres loaded with ADSC and linked to an E4D4 IgG monoclonal antibody that specifically targets the type II collagen (MabCII) unmasked in osteoarthritic hyaline cartilage. Upon binding to damaged cartilage, the ADSC carried by the microsphere rapidly migrate onto the damaged cartilage surface. We propose that MabCII linked to Col I microspheres can serve as an intraarticular carrier for transport and targeting large numbers of ADSC to osteoarthritic tissue while promoting chondrogenic differentiation.

METHODS: 
Isolation and culture of ADSC: The ADSC were isolated from subcutaneous fat tissue of 3-4 months old pigs as described previously (1). All animal protocols and experimental procedures were approved by the Institutional Animal Care and Use Committee (IACUC) at the University of Tennessee Health Science Center. The adipose tissue was digested with 200 μg/mL collagenase II solution (Sigma) to yield ADSC that were cultured in Dulbecco's Modified Eagle Medium-High Glucose (DMEM-HG) (Gibco) supplemented with fetal bovine calf serum (FBS) and antibiotics at 37°C and 5% CO2. The ADSC were characterized by flow cytometry and for multigenic differentiation before experimental use. 

Preparation of MabCII targeted Col I scaffold: 
Lyophilized purified native Col I from fetal calf skin was the gift of Dr. David Brand, Memphis VA Medical Center. The E4D4 IgG monoclonal antibody was isolated and characterized as described previously (2). The experimental group of MabCII targeted scaffolds were constructed using a two-step EDC/NHS crosslinking reaction (ThermoFischer) that bound hydroxyl groups on Col I to primary amine groups on MabCII in a 2:3 molar ratio. The efficiency of conjugation of MabCII conjugation on the Col I scaffold was measured by protein assay of the supernatant after pelleting the collagen by centrifugation at 1000g’s for 10 minutes. Cellular scaffolds were assembled in a solution of DMEM-HG + FBS using 4mg/ml homogenized MabCII-Col I and 16 million cells/ml ADSC. 

Experimental group and analysis: 
5ul droplets of the cells (80,000 cells) in a 4 mg/ml collagen suspension were plated on a non-tissue culture petri dish to form microsphere pellets and then cultured in DMEM-HG supplemented with FBS and antibiotics at 37°C and 5% CO2 for three days. At this time point, microsphere pellets were either collected for experimentation, or left on tissue culture plate and imaged under light microscopy after 4, 6, and 12 days (Figure 1). Control groups were established using pelleted ADSC (80,000 cells/ 5 microliters) without any Col I and ADSC integrated in a Col I scaffold without linked MabCII similarly cultured. To test the migration of ADSC from the scaffold onto damaged cartilage tissue, ADSC were stained with a fluorescent dye (Vivotrack680, PerkinElmer) before pellet formation. A 5ul MabCII linked scaffolded pellet was subsequently placed on to a 5 mm trypsinized cartilage explant and imaged after 6 days using a Keyence BZ-X800 microscope with a Cy5.5 filter (NIKON). To demonstrate targeting efficiency of our MabCII linked scaffolds, microspheres of fluorescent ADSC, pelleted alone or with Col I or MabCII/Col I were suspended in a 2 ml test tube on a rocker table with 5 mm damaged (trypsinized) cartilage explants. IVIS images were obtained at 4 and 24 hours and total fluorescence of ROI was calculated by Living Image 4.2 software. 

RESULTS: 
ADSC integrated in Col I microspheres and cultured demonstrated release from microsphere and growth on tissue culture plate (Figure 1). Additionally, over 6 days in culture, ADSC migrated from their microsphere attached to the cartilage explant and spread on damaged cartilage tissue as seen by the spread of fluorescence outside of the stationary ADSC/MabCII microsphere (Figure 1). This migration was again demonstrated by scanning electron photography of stationary 5ul ADSC + Col I pellets after 4 days of incubation on trypsinized cartilage explants. The images exhibited ADSC viability on day 4 by capturing active cell division on the damaged cartilage (Figure 2). Finally, IVIS imaging of quantified fluorescence showed targeting of only ADSC/Col I/ MabCII microspheres to damaged cartilage at 4 hours, while the explants incubated alone or with ADSC or ADSC/Col I exhibited no florescence at this time point (Figure 3). ADSC/Col I linked MabCII pellets remained fluorescent on damaged cartilage at 24 hours. 

DISCUSSION: 
Adipose tissue derived mesenchymal stem cell regenerative therapy for regrowth of damaged cartilage for osteoarthritis relies on ADSC’s ability to differentiate into chondrocytes on damaged hyaline cartilage tissue. We demonstrate in vitro that the use of a CII targeting MabCII linked to a Col I scaffold induced chondrogenesis of ADSC and increased the specificity and rapidity whereby ADSC can be delivered to osteoarthritic tissue. Our targeted pellets not only bound more rapidly to damaged cartilage, but they also remained attached after 24hrs and migrated onto the damaged cartilage. These results show promise for an in vivo study using our MabCII targeted Col I microspheres as an intra-articular injectable therapeutic treatment for osteoarthritis. 

SIGNIFICANCE: 
The creation of an injectable cellular scaffold with ability to target ADSC to type-II collagen on damaged hyaline cartilage increases the localization of mesenchymal stem cells to an osteoarthritic lesion at early stages. This increases ADSC therapeutic potential as a treatment for OA.

REFERENCES: 
1. Bhatti F, Kim SJ, Yi AK, Hasty K.A., Cho. H. “Cytoprotective Role of Vitamin E in Porcine Adipose-Tissue Derived Mesenchymal Stem Cells against Hydrogen-peroxide Induced Oxidative Stress” Cell and Tissue Research. 2018, 374(1):111-120. (PMID: 29951700). 2. Bedingfield SK, Colazo JM, Yu F, Liu DD, Jackson MA, Himmel LE, Cho H, Crofford LJ, Hasty KA, Duvall CL. Amelioration of post-traumatic osteoarthritis via nanoparticle depots delivering small interfering RNA to damaged cartilage. Nat Biomed Eng. 2021 Sep;5(9):1069-1083. (PMID: 34413494).

Premium content

Next from AMA Research Challenge 2024

Financial Conflicts of Interest Among Physicians Followed by the Leading General Medical Journals on X

Similar lecture

Do Patients with Pre-Existing psychiatric Diagnoses Experience Increased Rates of Post-Operative Complications Following Pelvic Fracture Surgery?