Penheiter, Sumedha G.

Abstract Background: The increasing volume and diversity of clinical genomic reports pose a significant challenge across healthcare institutions for the interpretation and proper implementation of precision oncology research. Reports from multiple different vendors (e.g., Invitae, Ambry Genetics, Foundation Medicine) are typically PDFs and exhibit substantial heterogeneity in panel design, gene coverage, and reporting standards, which hinders efficient retrospective data mining and patient cohort identification within the electronic medical records. We sought to develop a framework to interrogate all of these reports. Methods: We extracted genomic reports from patients with breast cancer treated with neoadjuvant chemotherapy and developed MolHarmonizer, a novel, scalable framework leveraging Python and Gemini LLMs, designed to process and harmonize genomic data from disparate multi-vendor reports. Gemini LLMs are employed explicitly for robust information extraction, normalization, and structuring of key genomic features, transforming unstructured data into a unified, queryable dataset. Results: Our MolHarmonizer framework successfully processed 1,147 genomic reports from 1703 breast cancer patients (2006-2023) from 23 different companies, demonstrating robust capability to extract and standardize critical actionable biomarkers. Data sources included Invitae (n=554), Ambry Genetics (n=189), Natera (n=95), Mayo Clinic (n=88), Tempus (n=63), Guardant Health (n=47), and Foundation Medicine (n=37), with others contributing less than 20 reports. Of the samples, 827/1147 (72.1%) were germline (blood/saliva). A majority of the patients were tested using the panels due to a personal/family history (n=925). Overall, 413/1147 (36.0%) reports identified at least one mutation. For breast cancer, 75 reports showed BRCA1/2 mutations (37 BRCA1, 37 BRCA2, and one patient with both BRCA1 and 2). Other mutations identified included: PIK3CA (n=40), TP53 (n=96), PTEN (n=21), ESR1 (n=13) and AKT1/2 (n=8). Conclusion: MolHarmonizer, a powerful framework leveraging Gemini LLMs, effectively addresses genomic data heterogeneity by automating biomarker extraction and harmonization. This enables rapid cohort identification and deep retrospective analyses for clinical insights, biomarker discovery, understanding disease history, facilitating novel pattern discovery, e.g., predicting BRCA1 mutations from WSI, and accelerating research within our neoadjuvant BC cohort. Future plans include expanding to include over 20,000 breast cancer patients, developing a user-friendly chatbot, and ensuring inter-institutional adaptability for a variety of complex diseases. Citation Format: Krishna Rani Kalari, Xiaojia Tang, Thanmayee Boyapati, Tanya L. Hoskin, Sumathilatha Myla, Sumedha G. Penheiter, Richard M. Weinshilboum, Liewei Wang, Hamid R. Tizhoosh, Karthik Vikram Giridhar, Matthew P. Goetz, Judy C. Boughey. Development of an LLM framework for analysis of heterogeneous breast cancer patients genomic reports [abstract]. In: Proceedings of the American Association for Cancer Research Annual Meeting 2026; Part 1 (Regular Abstracts); 2026 Apr 17-22; San Diego, CA. Philadelphia (PA): AACR; Cancer Res 2026;86(7 Suppl):Abstract nr 2756.

The FAITH! (Fostering African-American Improvement in Total Health) Cardiovascular Health and Wellness Program is more than a decade-long community-based participatory research initiative aimed at addressing cardiovascular health disparities among African-Americans in Minnesota. Founded in 2013, the program employs a culturally tailored, community-driven approach by partnering with African-American faith communities to promote cardiovascular health through education, digital health tools, and multilevel interventions targeting the social determinants of health. Grounded in community-based participatory research principles, FAITH! prioritizes equitable academic-community partnerships, co-learning, community capacity building, and shared ownership in all aspects of research and implementation. The program’s exemplary innovations include the NIH-funded FAITH! Trial, a randomized clinical trial, testing a mobile health intervention (the FAITH! App) co-created with the African-American community, and the Techquity by FAITH! study. Techquity by FAITH! evaluates the effectiveness of a culturally relevant, community-informed mHealth intervention supported by a Digital Health Advocate network to improve overall cardiovascular health and digital health literacy. During its evolution, FAITH! has addressed emergent public health crises, including the COVID-19 pandemic, by adapting programming to provide emergency preparedness resources, health education, and vaccine outreach. Key outcomes include sustainable church-based health ministries, increased research participation, and successful translation of research into practice. The program has also contributed to research workforce development by mentoring and training diverse early-career scholars and community leaders in community-based participatory research and cardiovascular health equity research. Lessons learned highlight the transformative impact of community-based participatory research in building trust, facilitating culturally relevant dissemination, and sustaining health equity initiatives. The FAITH! model demonstrates a scalable, community-led strategy for advancing cardiovascular health in underserved populations and provides a blueprint for future initiatives aiming to reduce racial health disparities.

Transforming growth factor β (TGF-β) coordinates a number of biological events important in normal and pathophysiological growth. In this study, deletion and substitution mutations were used to identify receptor motifs modulating TGF-β receptor activity. Initial experiments indicated that a COOH-terminal sequence between amino acids 482–491 in the kinase domain of the type I receptor was required for ligand-induced receptor signaling and down-regulation. These 10 amino acids are highly conserved in mammalian, Xenopus, andDrosophila type I receptors. Although mutation or deletion of the region (referred to as the NANDOR BOX, for nonactivating non–down-regulating) abolishes TGF-β–dependent mitogenesis, transcriptional activity, type I receptor phosphorylation, and down-regulation in mesenchymal cultures, adjacent mutations also within the kinase domain are without effect. Moreover, a kinase-defective type I receptor can functionally complement a mutant BOX expressing type I receptor, documenting that when the BOX mutant is activated, it has kinase activity. These results indicate that the sequence between 482 and 491 in the type I receptor provides a critical function regulating activation of the TGF-β receptor complex.

Transforming growth factor (TGF)-β family proteins form heteromeric complexes with transmembrane serine/threonine kinases referred to as type I and type II receptors. Ligand binding initiates a signaling cascade that generates a variety of cell type-specific phenotypes. Whereas numerous studies have investigated the regulatory activities controlling TGF-β signaling, there is relatively little information addressing the endocytic and trafficking itinerary of TGF-β receptor subunits. In the current study we have investigated the role of the clathrin-associated sorting protein Disabled-2 (Dab2) in TGF-β receptor endocytosis. Although small interfering RNA-mediated Dab2 knockdown had no affect on the internalization of various clathrin-dependent (i.e., TGF-β, low-density lipoprotein, or transferrin) or -independent (i.e., LacCer) cargo, TGF-β receptor recycling was abrogated. Loss of Dab2 resulted in enlarged early endosomal antigen 1-positive endosomes, reflecting the inability of cargo to traffic from the early endosome to the endosomal recycling compartment and, as documented previously, diminished Smad2 phosphorylation. The results support a model whereby Dab2 acts as a multifunctional adaptor in mesenchymal cells required for TGF-β receptor recycling as well as Smad2 phosphorylation.

Abstract Transforming growth factor-β (TGF-β) stimulates cellular proliferation and transformation to a myofibroblast phenotype in vivo and in a subset of fibroblast cell lines. As the Smad pathway is activated by TGF-β in essentially all cell types, it is unlikely to be the sole mediator of cell type–specific outcomes to TGF-β stimulation. In the current study, we determined that TGF-β receptor signaling activates phosphatidylinositol 3-kinase (PI3K) in several fibroblast but not epithelial cultures independently of Smad2 and Smad3. PI3K activation occurs in the presence of dominant-negative dynamin and is required for p21-activated kinase-2 kinase activity and the increased proliferation and morphologic change induced by TGF-β in vitro.