Faculty Bibliography

Extracorporeal liver cross-circulation (ELC) using gene-edited porcine liver xenografts offers a potential bridge therapy for liver failure. We previously performed five ELC procedures in four brain-dead human decedents, during which the recipients developed severe thrombocytopenia. The porcine liver xenografts maintained their parenchymal structure, with immune cell infiltration and detectable IgM deposition on endothelial cells. Here, to investigate the underlying host-xenograft interactions, we performed longitudinal proteomic, lipidomic and metabolomic profiling of 64 blood samples, alongside spatial transcriptomics and histology of 25 porcine liver xenograft and 3 native human liver biopsies. Spatial transcriptomic analysis revealed progressive infiltration of human immune cells (predominantly inflammatory macrophages and neutrophils) in the xenografts, concurrent with the loss of porcine Kupffer-like macrophages and T cells. Distinct human and porcine complement dynamics were observed, with suppressed human but elevated porcine complement levels, accompanied by increased levels of porcine acute-phase proteins and coagulation factors. Moreover, multiomics analyses identified candidate cellular and molecular factors associated with thrombocytopenia. Human platelets colocalized with activated porcine endothelial cells, which showed increasing porcine vWF expression over time, as well as with immune cells (primarily macrophages and neutrophils) and hepatocytes in the xenografts. Integrative analyses indicated that ELC procedures provided hepatic support for apolipoprotein synthesis, bilirubin clearance, energy metabolism and detoxification, although circulating lipid levels remained low under anhepatic conditions. Collectively, these findings yield insights into the complex interplay between human and xenograft systems during ELC and inform strategies to improve liver xenograft biocompatibility for clinical translation.

BACKGROUND:The gut microbiota (GM) plays an important role in chronic kidney disease (CKD) progression, and dialysis modalities can differentially impact the GM composition and function. There is also limited information on the GM in Arab populations. AIM/OBJECTIVE:To investigate the distinct microbial profiles and functional alterations associated with hemodialysis (HD) and peritoneal dialysis (PD) in a Saudi Arabian cohort. METHODS:We performed whole-genome metagenomic sequencing on fecal samples from 189 participants (controls and CKD, HD, and PD patients). RESULTS:interactions. Conversely, the HD group showed partial recovery of microbial balance and beneficial metabolic functions, including increased short-chain fatty acid metabolism and reduced lipopolysaccharide biosynthesis. CONCLUSION/CONCLUSIONS:The findings of this study highlight the potential of the microbial profile as a robust biomarker for CKD classification and underscore the differential impacts of different dialysis modalities.

MOTIVATION/BACKGROUND:Large-scale prospective cohort studies collect longitudinal biospecimens alongside time-to-event outcomes to investigate biomarker dynamics in relation to disease risk. The nested case-control (NCC) design provides a cost-effective alternative to full cohort biomarker studies while preserving statistical efficiency. Despite advances in joint modeling for longitudinal and time-to-event outcomes, few approaches address the unique challenges posed by NCC sampling, non-normally distributed biomarkers, and competing survival outcomes. RESULTS:Motivated by the TEDDY study, we propose "JM-NCC", a joint modeling framework designed for NCC studies with competing events. It integrates a generalized linear mixed-effects model for potentially non-normally distributed biomarkers with a cause-specific hazard model for competing risks. Two estimation methods are developed. fJM-NCC leverages NCC sub-cohort longitudinal biomarker data and full cohort survival and clinical metadata, while wJM-NCC uses only NCC sub-cohort data. Both simulation studies and an application to TEDDY microbiome dataset demonstrate the robustness and efficiency of the proposed methods. AVAILABILITY/BACKGROUND:Software is available at https://github.com/Zhaoyn-oss/JMNCC and archived on Zenodo at https://zenodo.org/records/18199759 (DOI: 10.5281/zenodo.18199759). SUPPLEMENTARY INFORMATION/BACKGROUND:Supplementary data are available at Bioinformatics online.

Organ shortage remains a major challenge in transplantation, and gene-edited pig organs offer a promising solution^1-3. Despite gene-editing, the immune reactions following xenotransplantation can still cause transplant failure⁴. To understand the immunological response of a pig-to-human kidney xenotransplantation, we conducted large-scale multi-omics profiling of the xenograft and the host's blood over a 61-day procedure in a brain-dead human (decedent) recipient. Blood plasmablasts, natural killer (NK) cells, and dendritic cells increased between postoperative day (POD)10 and 28, concordant with expansion of IgG/IgA B-cell clonotypes, and subsequent biopsy-confirmed antibody-mediated rejection (AbMR) at POD33. Human T-cell frequencies increased from POD21 and peaked between POD33-49 in the blood and xenograft, coinciding with T-cell receptor diversification, expansion of a restricted TRBV2/J1 clonotype and histological evidence of a combined AbMR and cell-mediated rejection at POD49. At POD33, the most abundant human immune population in the graft was CXCL9+ macrophages, aligning with IFN-γ-driven inflammation and a Type I immune response. In addition, we see evidence of interactions between activated pig-resident macrophages and infiltrating human immune cells. Xenograft tissue showed pro-fibrotic tubular and interstitial injury, marked by S100A6⁵, SPP1⁶ (Osteopontin), and COLEC11⁷, at POD21-POD33. Proteomics profiling revealed human and pig complement activation, with decreased human component after AbMR therapy with complement inhibition. Collectively, these data delineate the molecular orchestration of human immune responses to a porcine kidney, revealing potential immunomodulatory targets for improving xenograft survival.

BACKGROUND:Dietary self-monitoring is central to effective personalized nutrition, providing critical data to inform tailored feedback and support behavior change. OBJECTIVE:To examine the impact of dietary self-monitoring adherence and the indirect effect of personalized scores to predict postprandial glycemic response (PPGR) on weight loss. METHODS:Post-hoc analysis of the Personal Diet Study that investigated the impact of a machine algorithm-based diet that integrates clinical and microbiome features (Personalized) compared to a standard, low-fat diet (Standardized) on weight loss. All participants received behavioral counseling and were encouraged to self-monitor dietary intake via a smartphone app. Personalized received algorithm-based scores (1 to 5) on predicted PPGR to foods logged (PPGR score; 1-2 indicating optimal; 3-5 suboptimal). Dietary self-monitoring adherence was the percentage of days logging ≥50% of target calories, classified as high or low. PPGR score quality was calculated by the proportion of optimal predicted PPGR scores per day; defined as "high-PPGR quality" days when this exceeded the group average. Mediation analysis assessed whether PPGR quality mediated the relationship between dietary self-monitoring adherence and weight loss. RESULTS:Participants with high self-monitoring adherence lost an average of 4.2% of their baseline weight, compared to 1.9% among those with low adherence (p=0.016). High self-monitoring adherence was associated with a greater likelihood of achieving ≥5% weight loss (aOR=3.67, 95% CI: 1.63-8.50). Within Personalized, high PPGR quality mediated 53.4% of the total effect of self-monitoring adherence on weight loss (p<0.001). CONCLUSION/CONCLUSIONS:Consistent self-monitoring coupled with personalized feedback may significantly enhance weight loss in a precision nutrition approach. CLINICAL TRIAL REGISTRATION/BACKGROUND:NCT03336411.

BACKGROUND/UNASSIGNED:The gut microbiome is implicated in obesity and type 2 diabetes (T2D), but how diet, body composition, glycemic status, and self-efficacy factors relate to the microbiome in high-risk individuals is not well characterized. The purpose of this post-hoc analysis is to examine the relationship between the gut microbiome and obesity-related metabolic factors, including body composition, resting energy expenditure (REE), and glycemic variability (GV). METHODS/UNASSIGNED: = 95) included the collection of fecal microbiome profile, demographics, socioeconomic status, physical and metabolic characteristics [fat mass, fat free mass (FFM), continuous glucose monitoring-derived GV, REE], self-efficacy and dietary intake. Gut bacterial alpha diversity, beta diversity, and genus level abundances were associated with these host and lifestyle variables using multivariate regression, permutational multivariate analysis of variance, and correlation analyses. RESULTS/UNASSIGNED:ratio was positively associated with total energy, sugar, and carbohydrate intake (g/day) and negatively associated with monounsaturated fat intake (g/day). Glycemic measures and self-efficacy were not associated with any genera. CONCLUSION/UNASSIGNED:In adults with prediabetes and obesity, the gut microbiome at baseline was most strongly associated with diet, with additional associations observed for body composition and selected host characteristics. These findings underscore diet as a key correlate of gut microbiome structure in a high-risk metabolic population and support further development of microbiome-informed precision nutrition strategies for obesity prevention and management.

IMPORTANCE/UNASSIGNED:The oral microbiota may be involved in the development of pancreatic cancer, yet current evidence is largely limited to bacterial 16S amplicon sequencing and small retrospective case-control studies. OBJECTIVE/UNASSIGNED:To test whether the oral bacterial and fungal microbiome is associated with the subsequent development of pancreatic cancer. DESIGN, SETTING, AND PARTICIPANTS/UNASSIGNED:This cohort study used data from 2 epidemiological cohorts: the American Cancer Society Cancer Prevention Study-II Nutrition Cohort and the Prostate, Lung, Colorectal, and Ovarian Cancer Screening Trial. Among cohort participants who provided oral samples, those who prospectively developed pancreatic cancer were identified during follow-up. Control participants who remained free of cancer were selected by 1:1 frequency matching on cohort, 5-year age band, sex, race and ethnicity, and time since oral sample collection. Data were collected from August 2023 to September 2024, and data were analyzed from August 2023 to January 2025. EXPOSURES/UNASSIGNED:The oral bacterial and fungal microbiome were characterized via whole-genome shotgun sequencing and internal transcribed spacer (ITS) sequencing, respectively. The association of periodontal pathogens of the red complex (Treponema denticola, Porphyromonas gingivalis, and Tannerella forsythia) and orange complex (Fusobacterium nucleatum, F periodonticum, Prevotella intermedia, P nigrescens, Parvimonas micra, Eubacterium nodatum, Campylobacter shower, and C gracilis) with pancreatic cancer was tested via logistic regression. The association of the microbiome-wide bacterial and fungal taxa with pancreatic cancer was assessed by Analysis of Compositions of Microbiomes With Bias Correction 2 (ANCOM-BC2). Microbial risk scores (MRS) for pancreatic cancer were calculated from the risk-associated bacterial and fungal species. MAIN OUTCOMES AND MEASURES/UNASSIGNED:Pancreatic cancer incidence. RESULTS/UNASSIGNED:Of 122 000 cohort participants who provided samples, 445 developed pancreatic cancer over a median (IQR) follow-up of 8.8 (4.9-13.4) years and were matched with 445 controls. Of these 890 participants, 474 (53.3%) were male, and the mean (SD) age was 67.2 (7.5) years. Three oral bacterial periodontal pathogens-P gingivalis, E nodatum, and P micra-were associated with increased risk of pancreatic cancer. A bacteriome-wide scan revealed 8 oral bacteria associated with decreased and 13 oral bacteria associated with increased risk of pancreatic cancer (false discovery rate-adjusted Q statistic less than .05). Of the fungi, genus Candida was associated with increased risk of pancreatic cancer. The MRS, based on 27 oral species, was associated with an increase in pancreatic cancer risk (multivariate odds ratio per 1-SD increase in MRS, 3.44; 95% CI, 2.63-4.51). CONCLUSIONS AND RELEVANCE/UNASSIGNED:In this cohort study, oral bacteria and fungi were significant risk factors for pancreatic cancer development. Oral microbiota hold promise as biomarkers to identify individuals at high risk of pancreatic cancer, potentially contributing to personalized prevention.

BACKGROUND:levels. However, it remains unclear whether the CARE program also improves diabetes self-efficacy and psychosocial outcomes in the same study sample. OBJECTIVE:This is a secondary analysis to examine the potential efficacy of the CARE program on secondary outcomes, including diabetes self-efficacy, self-care activities, beliefs in diabetes self-care activities, and diabetes distress among Chinese Americans with T2D. METHODS:level of 7% or higher. Participants were recruited from various health care settings in New York City, including community health centers, private primary care providers, and NYU Langone Health and its affiliates, and were randomly assigned to either the CARE intervention group (n=30) or a waitlist control group (n=30). The intervention consisted of 2 culturally and linguistically tailored educational videos per week for 12 weeks, covering diabetes self-care topics such as healthy eating, physical activity, and medication adherence. These videos were delivered via the WeChat app. In addition, community health workers provided support calls to assist them in setting goals, problem-solving, and addressing social determinants of health barriers every 2 weeks. Secondary outcomes included patient self-reported diabetes self-efficacy, self-care activities, beliefs in diabetes self-care activities, and diabetes distress. Outcomes were assessed at baseline, 3 months, and 6 months. RESULTS:Participants had a mean age of 54.3 (SD 11.5) years and 62% (37/60) were male, 78% (47/60) were married, 58% (35/60) were employed, 70% (42/60) had a high school education or lower, and 88% (53/60) reported limited English proficiency. Intervention participants demonstrated statistically significant improvements in self-efficacy at 3 months (estimated difference in change: 8.47; 95% CI 2.44-14.5; adjusted P=.02), diabetes distress at 6 months (estimated difference in change: -0.43; 95% CI -0.71 to -0.15; adjusted P=.009), and adherence to a healthy diet at both 3 months (estimated difference in change: 1.61; 95% CI 0.46-2.75; adjusted P=.02) and 6 months (estimated difference in change: 1.64; 95% CI 0.48-2.81; adjusted P=.02). CONCLUSIONS:The culturally and linguistically tailored intervention showed promise in improving self-efficacy and diabetes self-care activities among Chinese Americans with T2D, warranting validation through a large-scale randomized controlled trial. TRIAL REGISTRATION/BACKGROUND:ClinicalTrials.gov NCT03557697; https://clinicaltrials.gov/study/NCT03557697.

BACKGROUND:Minoritized populations face many barriers to accessing evidence-based diabetes intervention. OBJECTIVES/OBJECTIVE:To evaluate the feasibility, acceptability, and potential efficacy of a social media-based intervention to improve glycemic control among Chinese Americans with type 2 diabetes. DESIGN/METHODS:A pilot randomized controlled trial (RCT) with 3-month and 6-month follow-ups. PARTICIPANTS/METHODS:Chinese Americans (n = 60, mean age 54.3 years old) with limited education (70.0% with high school or less) and low income (50.0% with annual household income < $25,000), and 88.3% have limited English proficiency. INTERVENTION/METHODS:Culturally and linguistically tailored diabetes videos (two videos/week for 12 weeks) delivered via social media and support calls from community health workers. MAIN MEASURES/METHODS:Primary outcomes include feasibility (video watch rate, biweekly call completion rate, and retention rates), acceptability (patient satisfaction), and HbA1c. Secondary health-related outcomes include body weight, BMI, physical activity, and dietary intake. Video watch rate and biweekly call completion rate were assessed at baseline and 3 months, while others were measured at baseline, 3 months, and 6 months. RESULTS:We observed high feasibility and acceptability of the intervention, with retention rates over 87%, an 89% video watch rate, 80% biweekly phone call completion, and a satisfaction rating of 9 out of 10. The intervention group showed a significantly greater increase in fruit intake compared to the control group (0.15 cups vs. - 0.44 cups, adj_p = 0.023) at 3 months. While no significant differences in other outcomes were observed between the groups, the intervention group showed significant improvements in key outcomes, including reduced HbA1c levels (- 1.08%, adj_p < 0.001), weight loss (- 5.15 lbs, adj_p = 0.004), lower BMI (- 0.83, adj_p = 0.023), and reduced starchy food intake (- 0.33 cups, adj_p = 0.033) at 6 months. CONCLUSIONS:The observed high feasibility and acceptability suggest the intervention's feasibility. However, due to the limited sample size, a larger-scale RCT is warranted to test the efficacy of the intervention. TRIAL REGISTRATION/BACKGROUND:ClinicalTrials.gov NCT03557697; https://clinicaltrials.gov/ct2/show/NCT03557697.