Faculty Bibliography

BACKGROUND/UNASSIGNED:Coronary heart disease (CHD) is the leading cause of morbidity and mortality worldwide, yet current prediction tools do not fully capture sex differences in risk. Menopause further modifies women's risk of CHD, but underlying molecular mechanisms are not well understood. Proteomic profiling may provide new insights into these mechanisms. METHODS/UNASSIGNED:We analyzed 4,877 SomaScan™-based plasma proteins in 10,034 Atherosclerosis Risk in Communities (ARIC) participants. Using linear regression, we identified sex- and menopause-related differences in circulating protein concentrations. We performed pathway enrichment analyses to capture underlying biological mechanisms. To evaluate the associations between proteins and incident CHD and differences by sex or menopause status, we applied Cox regression. Finally, we used LASSO regression to identify proteins that improve CHD risk prediction beyond an established clinical model. RESULTS/UNASSIGNED:We observed significant differences in circulating proteins by sex and menopausal status, including early changes during the menopausal transition. We found protein associations with incident CHD differed markedly by sex and menopausal stage: eight proteins were associated with CHD in men, compared with 54 in women, and six and 17 proteins in peri‑ and postmenopausal women, respectively. In risk prediction analyses, adding sex-specific protein panels improved CHD prediction (ΔC-statistic=0.05 in men and 0.03 in women), with similar improvement across menopausal stages. CONCLUSIONS/UNASSIGNED:Proteomic profiles underlying CHD risk differ substantially by sex and menopausal stage. These findings suggest potential biological heterogeneity in CHD and support the study of sex- and menopause-related differences in underlying mechanisms.

INTRODUCTION/BACKGROUND:To estimate the life expectancy gains, cost effectiveness, and impact on life expectancy inequality achieved by additional investment in community health worker (CHW) programs in New York City Methods: Using a state-transition model of mortality-driving conditions adapted to New York City anchored to 2021, we examined the impact of additional investment in CHWs for people with obesity, hypertension, and diabetes on (1) change in life expectancy, (2) health-inclusive (lifespan) return on investment (LROI), and (3) change in inequality of life expectancy distribution between neighborhoods. Inequality analyses included neighborhood-level baseline life expectancy, median income, and plurality race or ethnicity. Analyses were conducted in 2026. RESULTS:Scaling up CHWs results in a 0.13-year life expectancy gain on the citywide level, with favorable LROI of 14.72 years per $100,000 invested. Neighborhood-level gains range from 0.05 years to 0.33 years, while LROI ranges from 11.76 to 29.62 life-years/$100,000. In general, life expectancy gains and LROI were higher for lower-income, lower life expectancy, and minoritized race neighborhoods. CONCLUSIONS:Additional investment in CHW programs in New York City has the potential to cost-effectively raise the citywide life expectancy while reducing inequalities between neighborhoods.

OBJECTIVES/OBJECTIVE:To determine whether integrating an artificial intelligence (AI) decision-support system (VivoAID) into optical coherence tomography (OCT) interpretation improves diagnostic accuracy and efficiency for novice users evaluating basal cell carcinomas (BCCs). METHODS:Two dermatology research fellows completed a structured OCT training module for BCC identification using the VivoSight Dx system. Each independently reviewed 45 de-identified OCT scans (31 BCC-positive and 14 BCC-negative, negative ones included squamous cell carcinomas, keratoacanthomas, actinic keratoses and scar tissue), with ground truth established by expert clinical interpretation and histopathology, in two sessions: one unassisted and one AI-assisted. Session order was counterbalanced between learners, and scans were re-randomized for AI sessions. In unassisted sessions, learners manually classified scans; in AI-assisted sessions, they reviewed AI-annotated scans, recorded agreement/disagreement, and noted additional findings. Time-to-decision per scan was measured. A linear mixed-effects model in SAS assessed the effect of AI on decision time while accounting for scan difficulty. RESULTS:Without AI, correct BCC identification was 62% for Learner 1% and 64% for Learner 2 (combined 63%). With AI assistance, accuracy increased to 78% and 76%, respectively (combined 77%), reflecting relative improvements of 25% and 17% (combined +21%). Mean decision time decreased from 45 to 28 s for Learner 1 and from 35 to 22 s for Learner 2, yielding a combined reduction from 40 to 25 s (~38% faster). AI use significantly reduced decision time by an average of 15.1 s per image (p < 0.0001). AI standalone accuracy versus expert opinion was 64%, comparable to novice baseline performance. CONCLUSIONS:AI-enhanced OCT with VivoAID significantly improved novice users' diagnostic accuracy and efficiency, despite AI performance alone being similar to baseline novice interpretation. These data support VivoAID's role as a decision-support and potential educational tool in dermatology. Findings are limited by small sample size, single-institution design, and repeated review of the same scans. TRIAL REGISTRATION/BACKGROUND:None.

BACKGROUND/UNASSIGNED:Treatment options for refractory gastric cancer are limited. Preclinical data suggest interplay between angiogenesis and immune regulation, with vascular endothelial growth factor impairing antitumor immunity through decreasing antigen presentation, effector function, and immune cell trafficking. This phase II study evaluated lenvatinib plus pembrolizumab in the refractory setting. PATIENTS AND METHODS/UNASSIGNED:Eligible patients had advanced gastric or gastroesophageal junction adenocarcinoma with measurable disease, Eastern Cooperative Oncology Group ≤ 1, and progression after ≥1 prior therapy, without prior immune checkpoint inhibitor exposure. Patients received pembrolizumab 200 mg i.v. every 21 days and lenvatinib 20 mg orally daily, with a 7-day lead-in. Primary endpoint was objective response rate. Using Simon optimal two-stage design, 29 patients were planned; the study would be considered positive if ≥6 responses were observed. Secondary endpoints included safety, progression-free survival (PFS), and overall survival (OS). RESULTS/UNASSIGNED:Enrollment was prematurely suspended at 24 patients due to evolving practice patterns. Median age was 63 years; 75% were male; 63% were White. Seven patients (29%) received ≥2 lines of therapy. Among 20 assessable patients, 5 (25%) achieved partial responses, including 2 with combined positive score (CPS) <1 tumors. Median duration of response was 31.6 weeks. Median PFS and OS were 12.9 and 21.4 weeks, respectively. Treatment was well tolerated with no new safety signals. CONCLUSIONS/UNASSIGNED:Lenvatinib plus pembrolizumab did not meet the predefined efficacy threshold in the overall cohort. However, treatment was well tolerated and encouraging responses were seen in patients with CPS <1 tumors. Although the sample size was limited, these findings suggest that further investigation of this treatment strategy in patients with CPS < 1 disease may be warranted.

Immune checkpoint inhibitors (ICIs) have transformed the treatment of advanced non-small cell lung cancer (NSCLC), yet a large proportion of patients do not respond. The lung microbiome is a key modulator of antitumor immunity, however, the functional contribution of active microbial communities and their interaction with host transcriptional programs in shaping ICI outcomes remain poorly defined. Here, we prospectively analyzed bronchoalveolar lavage (BAL) samples from 71 NSCLC patients prior to ICI initiation. We profiled the microbiome and host transcriptome in the BAL to look for microbial and host signatures in the lower airways associated with treatment response. We found that enrichment of the lower airway microbiome with oral commensals was associated with disease progression and worse prognosis of patients treated with ICI. Metatranscriptomics-derived microbial clustering revealed distinct functional profiles. The microbial cluster enriched with oral commensals, which was only present among patients with poor response to ICI, had differential regulation of pathways involved in the metabolism of short-chain fatty acids, methane, and pyruvate. This cluster also correlated with host immune pathways including upregulation of interferon signaling, neutrophil degranulation, and mitophagy, all of which could contribute to the poor response to immunotherapy. In a preclinical murine model of lung cancer with dysbiosis, inoculating lung tumor-bearing mice with oral commensals increased intratumoral neutrophils and Th17 responses and impaired ICI efficacy. Our findings support the concept that microbial communities in the lower airways influence tumor-immune interactions via defined metabolic-immune axes.

BackgroundLupus nephritis (LN) affects over 60% of patients with systemic lupus erythematosus and is a risk factor for progression to end stage kidney disease. Guidelines recommend high dose parenteral methylprednisolone "pulses" for induction, although intravenous pulse therapy has not been rigorously compared to oral therapy. In this observational study, we report the outcomes of 17 patients who initiated therapy for LN with high dose oral dexamethasone.MethodsLN patients treated with oral dexamethasone between December 2021, and January 2025, were identified and baseline characteristics were recorded. Patients were assessed for a complete renal response (CRR, uPCR < 500, eGFR > 80%), satisfactory renal response (SRR, uPCR < 700, eGFR > 80%), or partial renal response (PRR, uPCR ≤ half of baseline uPCR, eGFR > 80%) at 1 year after receipt of dexamethasone. Responses were compared to outcomes in the BLISS-LN, AURORA and REGENCY trials.ResultsSeventeen patients with LN treated with induction dexamethasone were identified. The average age was 40 years old, and 15 patients were female. Six patients were of African ancestry, four Hispanic ethnicity, and three were White. Fourteen patients had proliferative LN. 47% of patients achieved a CRR, 59% obtained a SRR, and 53% obtained a PRR. Renal outcomes compared favorably to outcomes in the active arms of the BLISS-LN, AURORA 1, and REGENCY studies. Biomarkers including C3, C4, and dsDNA antibodies improved throughout the period. Proteinuria declined in 14 of 17 patients.ConclusionPatients with LN initially treated with oral dexamethasone in place of intravenous pulse methylprednisolone achieved responses similar to outcomes in recent LN trials. Dexamethasone is a promising potential oral option to initiate treatment for LN obviating need for parenteral therapy.

AIMS/OBJECTIVE:Plaque erosion accounts for up to 40% of acute coronary syndromes. Unlike plaque rupture, erosion is characterized by thrombus formation on an intact fibrous cap following endothelial denudation. Rupture is associated with broad inflammatory activation, whereas erosion features accumulation of specific immune cells, particularly CD4+ and CD8+ T cells. Their functional programming remains insufficiently defined.The objective was to characterize intracoronary and peripheral T cell phenotypes, activation programs, and clonal architecture in acute coronary syndromes caused by plaque erosion compared with plaque rupture. METHODS AND RESULTS/RESULTS:We prospectively enrolled 133 consecutive patients with acute coronary syndromes in whom plaque erosion or rupture was classified by optical coherence tomography. Blood was sampled from the coronary segment proximal to the lesion and from the peripheral circulation. T cell phenotypes were assessed by flow cytometry and plasma proteomics. In a subgroup, intracoronary samples underwent CITE-seq, integrating single-cell transcriptomics with surface protein profiling, together with paired T cell receptor sequencing. RESULTS:Flow cytometry demonstrated enrichment of activated memory helper and cytotoxic T cells with anti-inflammatory features at erosion sites, in contrast to predominantly naive and pro-inflammatory T cells in rupture. CITE-seq revealed erosion-specific activation of pathways related to proliferation, adhesion, and tissue repair, whereas T cells in rupture displayed enhanced inflammatory pathways. Functional validation confirmed greater IL-2/IFN-γ induction in rupture-associated memory T cells and greater IL-10 induction in erosion-associated Tregs. T cell receptor sequencing showed greater clonal expansion in erosion, suggesting antigen-experienced clonal selection, whereas rupture exhibited broader and less focused clonal diversity. CONCLUSIONS:Plaque erosion and plaque rupture show fundamentally different T cell immune profiles. Erosion is marked by memory-skewed, clonally expanded regulatory and cytotoxic T cell subsets, while rupture is characterized by proinflammatory phenotypes, diverse T cell receptor repertoires, and heightened inflammatory signaling. These findings support morphology-tailored immunotherapeutic approaches, including established anti-inflammatory strategies for rupture and selective T cell-directed interventions for erosion.

BACKGROUND/UNASSIGNED:Dilated cardiomyopathy (DCM) frequently presents with heart failure in children. Whether a genetic cause of DCM is associated with disease severity or eventual outcome in pediatric DCM is unclear. METHODS/UNASSIGNED:A multiinstitutional prospective pediatric cohort study (0-18 years) with nonsyndromic DCM was undertaken from 2013 to 2016. A panel of 37 cardiomyopathy-associated genes was interrogated by exome sequencing with analysis for pathogenic or likely pathogenic (LP) variants. Clinical severity and outcomes of death, cardiac transplant, and continuing left ventricular dysfunction versus echocardiographic normalization were assessed. RESULTS/UNASSIGNED:=0.09) toward greater echo normalization, was noted in those without a pathogenic/LP variant by 10 years after diagnosis. CONCLUSIONS/UNASSIGNED:A genetic cause was detected in a minority of children and was not associated with the phenotypic severity of DCM at diagnosis. Echocardiographic normalization of left ventricular function over time was noted more frequently in patients who did not have an identified pathogenic/LP variant.

With the increasing prevalence of obesity in the United States, potential living liver donors (LLDs) are similarly impacted. Historically, LLDs with obesity are deemed too high risk for donation due to concern for hepatic steatosis (HS) or post-surgical complications. Evolving data show that obesity is not the absolute indicator of HS and metabolic risk. Instead, distribution of fat in the visceral area as well as body composition drive the risk for metabolic dysfunction associated steatotic liver disease (MASLD). Prior practice patterns of denial of obese LLDs with metabolic risk require a change in mindset from the liver transplant community with a thoughtful, individualized risk assessment of all LLDs. By losing weight utilizing various dietary and procedural strategies, these "gray zone" LLDs improve their metabolic risk profile and can move into a low risk "green zone", thus expanding the donor pool. Little data exist whether "gray zone" LLDs who succeed in weight loss have long-term consequences to their own health if they regain their weight or the clinical impact on their LDLT recipients. Ethical and financial questions remain as to the LDLT centers' direct responsibility to maintaining donor health. Future studies evaluating interventions and outcomes are needed.