Faculty Bibliography

TOPIC/OBJECTIVE:The prevalence of bronchiectasis and nontuberculous mycobacterial (NTM) pulmonary disease are rising and increasingly managed in United States community and safety-net health systems. However, patients cared for in these settings remain underrepresented in epidemiologic studies and clinical trials that inform clinical practice guidelines. IMPORTANCE/OBJECTIVE:Although existing recommendations provide important standards of care, they often assume reliable access to subspecialty evaluation, advanced diagnostics, and costly therapies. In under-resourced systems, structural constraints shape how and when patients obtain subspecialty evaluation and evidence-based therapies. OVERVIEW/BACKGROUND:Drawing on previous evidence and experience from centers within the Bronchiectasis and NTM Care Center Network, we propose a practical framework for adapting guideline-concordant care to under-resourced systems. This framework focuses on four domains: diagnostic pathways that support earlier recognition, co-management models that prioritize high-risk patients and leverage disease expertise, pragmatic approaches to evidence-based therapies aligned with available resources, and linguistic and culturally responsive education to support adherence and shared decision-making. Aligning evidence-based recommendations with structural realities is necessary to improve care delivery and reduce inequality in bronchiectasis and NTM care.

Histone lysine methylation, primarily mediated by the enhancer methyltransferase KMT2D (MLL4), regulates gene expression through H3K4 mono- and di-methylation. Dysregulation of KMT2D disrupts enhancer activation and contributes to tumorigenesis and cellular plasticity across multiple cancers, including lung, prostate, bladder, head and neck, and pancreatic tumors. KMT2D functions in a context-dependent manner, acting as either a tumor suppressor or oncogenic driver, and modulates key phenotypic transitions-such as epithelial-to-mesenchymal, squamous, endothelial, and neuroendocrine states-that underlie metastasis and therapeutic resistance. Beyond its tumor-intrinsic roles, KMT2D loss remodels the tumor immune microenvironment by enhancing antigen presentation and effector T-cell infiltration, thereby sensitizing tumors to immune checkpoint blockade. Understanding how KMT2D interfaces with signaling pathways such as PI3K/AKT, TGF-β, and NOTCH to regulate plasticity and immunogenicity will be critical for leveraging its biomarker and therapeutic potential. This review summarizes current insights into KMT2D's roles in cancer progression, lineage plasticity, therapy resistance, and immune regulation, highlighting its emerging relevance in precision oncology.

Hirschsprung disease (HSCR) is a congenital enteric neuropathy caused by disrupted development of enteric neural crest-derived cells (ENCDCs). Although pathogenic coding variants in RET account for many cases, the largest genetic contribution to HSCR risk arises from a common noncoding variant (rs2435357) within a SOX10-bound RET enhancer (MCS+9.7) that reduces RET gene expression in vivo and triggers expression changes in other ENS genes in the human fetal gut. However, the ENS cell types affected by this enhancer and the mechanisms by which these transcriptional changes lead to HSCR remain unknown. Here, we investigated the role of this enhancer by generating mice carrying a deletion of the orthologous Ret mcs+9.7 enhancer (Δmcs+9.7). Single-cell RNA sequencing of E14.5 embryonic gut demonstrated that enhancer deletion reduced Ret expression by 8% without altering ENS cell composition. However, reduced Ret expression was restricted to differentiating neurons and inhibitory motor neuron lineages, revealing cell type-specific enhancer activity. To determine the functional consequences of further reducing Ret dosage, we generated compound heterozygous mice carrying both the enhancer deletion and a Ret coding null allele (+/Δmcs+9.7;+/CFP). These mice exhibited additive reductions in Ret expression, altered Sox10 expression, dysregulation of cell-cycle and neuronal differentiation programs, and selective depletion of developing inhibitory motor neuron lineages. These findings establish a cell type-specific role for the mcs+9.7 enhancer in modulating Ret dosage and reveal how subtle enhancer perturbations alter neural subtype specification without overt hypoganglionosis, suggesting that HSCR arises from a cascade of cellular defects triggered by >50% loss of Ret function.

BACKGROUND/UNASSIGNED:Individuals with opioid use disorder (OUD) who return to the community from incarceration face a high risk of fatal drug overdose. This mortality risk is related to significant barriers to accessing health care, social support, and resources needed to transition to and thrive within the community safely. Digital health tools have effectively supported substance use treatment and recovery, but their potential to support OUD during the high-risk period of reentry is underexplored. OBJECTIVE/UNASSIGNED:This study aimed to examine how adults with OUD returning from incarceration perceive and engage with digital health tools, and identify barriers and facilitators to their use. METHODS/UNASSIGNED:Semistructured interviews were conducted with a purposive sample of 39 adults recently released from New Hampshire prisons and jails. Participants were recruited from a randomized clinical trial (EXIT-CJS) comparing the effectiveness of extended-release buprenorphine, naltrexone, and enhanced treatment as usual on treatment retention. Interviews were audio recorded, transcribed, and analyzed using a deductive-inductive approach to content analysis. The COREQ (Consolidated Criteria for Reporting Qualitative Research) criteria were used to guide the study reporting. RESULTS/UNASSIGNED:Most participants preferred digital health over in-person care for both OUD treatment and other health care, mainly due to its convenience and resilience against reentry challenges. Yet, most participants also underscored the irreplaceable role of in-person therapeutic human connection in recovery and recognized this as a trade-off of stand-alone digital health. Participants described using digital health tools to overcome barriers that otherwise would have prevented them from receiving in-person care, such as transportation shortages, limited provider availability, and work scheduling conflicts. Accessing digital health was acknowledged as contingent on access to internet-capable devices, affordable connectivity, and digital literacy. CONCLUSIONS/UNASSIGNED:The findings suggest that maximizing digital health's impact in reentry will require a multipronged approach: investing in infrastructure to close the digital divide, addressing upstream structural barriers to accessing care, and partnering with impacted persons to co-design hybrid care models that enhance the feasibility of engaging in care during reentry while enhancing therapeutic human connection and respecting patient preferences.

Current predictive models for pregnancy and infant outcomes often focus on limited endpoints and rely on costly tests or imaging. Here we developed the Mother-Child AI Agent (MoChiAgent), an LLM-based clinical assistant that orchestrates multiple tools to integrate sequential electronic health record (EHR) data, including routine laboratory tests, for forecasting maternal and infant diseases. MoChiAgent's core predictive engine, MoChiFormer, was developed and internally evaluated using 4,401,599 longitudinal clinical visits and externally validated using independent maternal and infant cohorts consisting of 263,452 and 23,192 visits, respectively. MoChiFormer reconstructs missing laboratory values, reduces batch effects and learns EHR representations that support gestational, fetal and infant age estimation, health-trajectory modelling and stratification of current and future disease risk. Subsequently, a Knowledge Search Tool utilizes these forecasts to retrieve evidence-based intervention and treatment recommendations from curated medical literature and authoritative guidelines. For maternal health, MoChiFormer accurately identified key gestational conditions, achieving AUROCs of 0.89 for placental abruption, 0.89 for premature rupture of membranes, and 0.91 for preterm labour. Analysis of paired mother-infant data further revealed transgenerational risk associations, with infants born to mothers in specific clusters showing substantially elevated risks of neonatal jaundice (HR = 2.81, 95% CI 2.60-3.03) and haematological diseases (HR = 2.83, 95% CI 2.62-3.05). Integrating maternal gestational EHRs with infant records improved prediction of infant conditions, including chromosomal abnormalities and respiratory disorders. These findings suggest that MoChiAgent can provide clinically relevant, actionable decision-support information to enhance risk-stratified care for mothers and infants.

Castration-resistant prostate cancer (CRPC) progression despite treatment with potent androgen receptor (AR) antagonists such as enzalutamide is a major clinical challenge. The mitotic kinase BUB1B is a central member of an oncogenic seven-gene network that drives CRPC. Here, we characterized BUB1B as an actionable therapeutic target for treatment-resistant PC. High BUB1B expression correlated with PC progression and aggressiveness. BUB1B depletion blocked CRPC cell proliferation through accumulation in G2/M and mitotic delay. Conversely, ectopic expression of BUB1B conferred castration and enzalutamide resistance to androgen-dependent PC cells in vitro and in vivo. BUB1B promotion of CRPC was not dependent on AR canonical signaling as assessed through AR knockdown and PROTAC-mediated AR degradation. BUB1B kinase activity was necessary for CRPC progression, as only wild type (WT), but not two kinase-dead mutants, promoted castration-resistant growth, and BUB1B kinase activity was required to maintain castration resistance. Expression of the primary substrate of BUB1B, CENP-E, also promoted CRPC progression, while a phosphodeficient mutant of the BUB1B phosphorylation site was ineffective. Moreover, only a phosphomimic mutant of CENP-E rescued growth after BUB1B depletion in CRPC cells, indicating that BUB1B-dependent phosphorylation of CENP-E was sufficient for progression and necessary to sustain CRPC growth. Targeting CENP-E with the clinically tested, small-molecule GSK923295 phenocopied BUB1B knockdown, sensitized resistant cells to enzalutamide and suppressed growth of enzalutamide-resistant xenografts in vivo. These data support BUB1B kinase as a driver of castration and enzalutamide resistance, positioning CENP-E inhibition as an attainable approach for treatment-resistant disease.

BACKGROUND:Emerging therapies for acute PE highlight the need for a more granular risk classification. AIMS/OBJECTIVE:The objective of this study was to validate a novel risk model-the RIsk claSsification Adapted to SCAI shock stages in acute PE (RISA-PE), which applies the SCAI shock staging framework to right ventricular failure due to acute PE. METHODS:We retrospectively analyzed consecutive patients with acute intermediate- or high-risk PE treated with LBMT at a tertiary academic center (October 2020 to May 2025). Patients were stratified by RISA-PE stage (A-E) as previously described. Clinical data, including invasive pulmonary artery catheter hemodynamics and clinical outcomes were compared across stages. The primary endpoint was a composite of 30-day PE-related death, resuscitated cardiac arrest, or hemodynamic instability. RESULTS:) rising from 25% in Stage A to 75% in Stage C (p < 0.001). Clinical outcomes paralleled stage severity. Median hospital length-of-stay increased from 5 to 6 days (Stages A/B) to 19.9 days (Stage E, p < 0.001), and ICU stay from 2.5 to 3 days to 9.5 days, respectively (p < 0.001). The primary composite outcome occurred in 0% of Stage A, 5% of Stage B, 12% of Stage C, 15% of Stage D, and 50% of Stage E patients (p < 0.001). CONCLUSIONS:The RISA-PE staging system demonstrates strong correlation with RV dysfunction severity, invasive hemodynamics, and adverse outcomes in patients with acute PE treated with LBMT. RISA-PE provides a graded framework for identifying patients at highest risk of deterioration and may guide escalation of therapy.

BACKGROUND/UNASSIGNED:Radiation pneumonitis (RP) remains a significant treatment-related toxicity in patients with unresectable, locally advanced non-small cell lung cancer (NSCLC) undergoing chemoradiotherapy (CRT). Most existing predictive models rely on static baseline demographic or dosimetry variables and lack real-time clinical applicability. We developed a novel predictive framework that integrates longitudinal symptom data extracted from clinical notes using natural language processing (NLP) with clinical and dosimetry features to improve early RP prediction. METHODS/UNASSIGNED:We retrospectively identified 227 patients with locally advanced NSCLC treated with definitive CRT at a high-volume cancer center in the United States. We included all patients older than 18 years who were diagnosed between Jan 1, 2006, and Dec 31, 2022 with histologically or cytologically confirmed unresectable Stage 2 or 3 NSCLC and treated with conformal radiotherapy to a minimum dose of ≥45 Gy with or without chemotherapy. Of these, 31 RP events were identified through manual adjudication using radiologic criteria and chart review. NLP was used to extract the temporal relationship of 16 pre-specified symptoms with treatment from over 100,000 clinical notes spanning pre- and during-treatment intervals. We trained and validated machine learning models on combinations of baseline clinical data, radiation dosimetry, and NLP-derived symptom features. Model performance was evaluated using a nested cross-validation framework, with an outer cross-validation loop reserved for performance assessment and an inner cross-validation loop used for model training and integration, and summarized using area under the receiver operating characteristic curve (AUC) and partial AUC (pAUC) at high specificity thresholds. Clinical utility was evaluated using decision curve analysis (DCA). FINDINGS/UNASSIGNED:The best-performing model incorporated longitudinal NLP features and achieved a median AUC of 0.759 (90% confidence interval 0.753-0.766), significantly outperforming baseline models using only dosimetry (AUC 0.613) or clinical variables (AUC 0.635). NLP-based features such as cough trajectory, shortness of breath, and wheezing were among the most important predictors. Inclusion of NLP-derived symptom data improved early identification of high-risk patients, particularly in the clinically relevant high-specificity range (pAUC 0.021 vs. 0.010 for dosimetry alone). DCA showed that the calibrated MLP model provided greater net benefit than default strategies of treating all or no patients across clinically relevant threshold possibilities. INTERPRETATION/UNASSIGNED:In this early work, NLP-based extraction of longitudinal symptoms from routine clinical documentation meaningfully enhances RP prediction in patients undergoing CRT for NSCLC. This approach leverages existing electronic health record infrastructure to deliver real-time, scalable, and interpretable risk estimates, offering a pathway toward potential early intervention and personalized toxicity management. The model and DCA requires external and prospective validation before clinical deployment; as such, future work should focus on this validation and integration into clinical decision support systems. FUNDING/UNASSIGNED:AstraZeneca.

PURPOSE/UNASSIGNED:The addition of MEK inhibition (MEKi) to programmed cell death ligand 1 (PD-L1) blockade improves progression-free survival (PFS) in patients with advanced biliary tract cancer. Although MEK inhibitors may increase tumor cell immunogenicity, they can impair T-cell priming/effector function, limiting combination efficacy. We hypothesized that the addition of a CD27 agonist could restore T-cell function and enhance antitumor immunity in this combination. PATIENTS AND METHODS/UNASSIGNED:We conducted a randomized, phase II trial evaluating atezolizumab (840 mg, intravenously, days 1 and 15) in combination with the CD27 costimulatory monoclonal antibody [CDX-1127/varlilumab (3 mg/kg, intravenously, days 1 and 15)], with/without the addition of an MEK inhibitor [cobimetinib (60 mg, orally, daily, days 1-21, off days 22-28)] in unresectable biliary tract cancer following at least one metastatic therapy. Overall response rate (ORR) and PFS were coprimary endpoints. Treatment-related changes in CD8+ tumor-infiltrating lymphocytes (TIL) were the primary correlative outcomes. RESULTS/UNASSIGNED:The trial was closed early following interim preplanned ORR analysis. At closure, 57 patients had been enrolled [n = 29 in the cobimetinib + atezolizumab + varlilumab (CAV) arm; n = 28 in the atezolizumab + varlilumab (AV) arm]. A majority (67%) had intrahepatic cholangiocarcinoma, and 32% were immunotherapy experienced. Both regimens were well tolerated without new safety signals. Objective responses were rare [0% (CAV); 3.8% (AV)]. The median PFS (mPFS) was 2.40 (CAV) and 1.84 (AV) months [hazard ratio (HR), 0.67; 95% confidence interval (CI), 0.38-1.18]. Among immunotherapy-experienced patients, the mPFS was 3.62 (CAV) and 1.84 (AV) months (HR, 0.54; 95% CI, 0.18-1.62). Treatment with CAV increased intratumoral CD8+ T-cell density compared with treatment with AV. CONCLUSIONS/UNASSIGNED:The combinations of atezolizumab and varlilumab with/without cobimetinib were safe, but neither meaningfully improved outcomes in biliary tract cancer treated in the later lines. Correlative tissue studies validated preclinical work that MEKi increases CD8+ TILs.