FWFengxian Wang

Fengxian Wang

*This is an algorithmically generated author record based on publicly available information.

Current Organizations / Institutes

The organizations/institutes from the scholar's latest article


Fengxian Wang has a total of 114 co-authors. Here are the top co-authors by number of shared publications:


Latest Publications

  • Open Access

    Factors Associated With Telemedicine Use by Cardiologists for Medicare Beneficiaries in 2022 to 2023: An Observational Study

    by Samantha Harris, R. J. Waken, Fengxian Wang, Khavya C. Avula, Rishi K. Wadhera, Jose F. Figueroa, E. John Orav, Uchenna Ofoma, Karen E. Joynt Maddox

    Background: During the COVID‐19 pandemic, the Centers for Medicare and Medicaid Services created a waiver to reimburse telemedicine services. It is important to understand factors that facilitate incorporation of telemedicine into ongoing cardiovascular practice. Methods: This was a retrospective cohort study of telemedicine and office visits delivered by cardiologists between January 1, 2022, and December 31, 2023, for Medicare beneficiaries. We calculated the adjusted incidence rate ratio (aIRR) of telemedicine visits, representing the proportion of a physician's visits delivered by telemedicine, to identify factors associated with telemedicine use. Results: There were 23 334 physicians in our cohort; they were predominantly men (84.8%) and affiliated with a hospital (93.5%), and the majority were general cardiologists (66.1%). During 2022 and 2023, 3.4% of visits were delivered by telemedicine. In a regression model adjusted for beneficiary and provider characteristics, several physician‐level factors were associated with increased telemedicine: female sex (aIRR, 1.48 [95% CI, 1.41–1.57]), electrophysiology specialty (aIRR, 1.57 [95% CI, 1.47–1.67] compared with general cardiology), and caring for a high proportion of beneficiaries living in areas of social vulnerability (quartile 3 aIRR, 1.22 [95% CI, 1.12–1.32]; quartile 4 aIRR, 1.27 [95% CI, 1.16–1.39]). Caring for more beneficiaries residing in a rural area (aIRR, 0.71 [95% CI, 0.66–0.76]) or the South (aIRR, 0.61 [95% CI, 0.55–0.66]) and for beneficiaries aged >85 years (aIRR, 0.77 [95% CI, 0.73–0.81] were associated with lower use of telemedicine). Conclusions: Telemedicine is used relatively sparsely among cardiologists. Physician factors, including sex; specialty; and the vulnerability, rurality, and age of beneficiary panels, impact the degree to which telemedicine is a major part of clinical practice.

  • Open Access

    Characterizing early adopters of the rural emergency hospital designation

    by Jonah M Graves, R J Waken, Fengxian Wang, Paula Chatterjee, Karen E Joynt Maddox

    Jonah M Graves, MD, R J Waken, PhD, Fengxian Wang, PhD, Paula Chatterjee, MD MPH, Karen E Joynt Maddox, MD MPH; Characterizing early adopters of the Rural


  • Changes in Heart Transplant Listings by Insurance and Race During the COVID Public Health Emergency

    by Daniel K. Fox, R. J. Waken, Fengxian Wang, Khavya Avula, David S. Raymer, Justin M. Vader, Joel D. Schilling, Karen E. Joynt Maddox

    BACKGROUND: Socioeconomic inequities in access to heart transplantation, including those based on insurance coverage and race or ethnicity, are well documented. During the COVID pandemic, the federal government declared a Public Health Emergency (PHE), which prevented states from disenrolling people from Medicaid. We sought to determine whether this policy was associated with changes in the insurance or racial composition of the heart transplant waiting list. METHODS: Using the United Network for Organ Sharing thoracic data set, we conducted an observational cohort study that analyzed weekly counts of adults (aged >17 years) added to the heart transplant waitlist between January 1, 2018 and December 31, 2022. We obtained ratios of pre-PHE to PHE waitlisting rates by payer and race while accounting for demographics, trends over time, and overdispersion using latent lognormal Poisson dynamic generalized linear models. RESULTS: There were 8224 heart transplant listings before the PHE and 12 261 during. We found no change in listing among privately insured (rate ratio [RR], 1.06 [95% CI, 0.91–1.24]), Medicare (RR, 1.11 [95% CI, 0.94–1.30]), or other government coverage listings (RR, 1.13 [95% CI, 0.91–1.39]), but did find an increase in Medicaid-covered listings (RR, 1.19 [95% CI, 1.01–1.42]). We found no increase in transplant listings during the PHE for patients who were White (RR, 1.03 [95% CI, 0.90–1.21]), but did find increases among patients who were Black (rate ratio, 1.20 [95% CI, 1.04–1.43]), Hispanic (RR, 1.24 [95% CI, 1.06–1.49]), and Asian/American Indian or Alaska Native/Hawaiian or Pacific Islander/multiracial (RR, 1.36 [95% CI, 1.14–1.66]). CONCLUSIONS: The COVID-related PHE was associated with changes in the insurance and racial composition of the heart transplantation waitlist that, if maintained, could reduce socioeconomic inequities in access to heart transplantation.


  • Proteomics of brain, CSF, and plasma identifies molecular signatures for distinguishing sporadic and genetic Alzheimer’s disease

    by Yun Ju Sung, Chengran Yang, Joanne Norton, Matt Johnson, Anne Fagan, Randall J. Bateman, Richard J. Perrin, John C. Morris, Martin R. Farlow, Jasmeer P. Chhatwal, et al.

    Proteomic studies for Alzheimer’s disease (AD) are instrumental in identifying AD pathways but often focus on single tissues and sporadic AD cases. Here, we present a proteomic study analyzing 1305 proteins in brain tissue, cerebrospinal fluid (CSF), and plasma from patients with sporadic AD, TREM2 risk variant carriers, patients with autosomal dominant AD (ADAD), and healthy individuals. We identified 8 brain, 40 CSF, and 9 plasma proteins that were altered in individuals with sporadic AD, and we replicated these findings in several external datasets. We identified a proteomic signature that differentiated TREM2 variant carriers from both individuals with sporadic AD and healthy individuals. The proteins associated with sporadic AD were also altered in patients with ADAD, but with a greater effect size. Brain-derived proteins associated with ADAD were also replicated in additional CSF samples. Enrichment analyses highlighted several pathways, including those implicated in AD (calcineurin and Apo E), Parkinson’s disease (α-synuclein and LRRK2), and innate immune responses (SHC1, ERK-1, and SPP1). Our findings suggest that combined proteomics across brain tissue, CSF, and plasma can be used to identify markers for sporadic and genetically defined AD.


  • Open Access

    Identification of novel genetic variants and biomarkers for cognitive decline in Alzheimer’s disease

    by Muhammad Ali, Fengxian Wang, Samira Mafimoghaddam, Jack Euesden, Chloe Robins, David J. Pulford, Carlos Cruchaga

    Background: Alzheimer’s disease (AD) is a complex, multifactorial disease that is incurable in the aging population. The success of genome‐wide association studies (GWAS), as a cross sectional design, to identify novel genetic risk factors for AD in recent years is well known; however, case‐control studies are less likely to uncover genetic factors that influence other aspects of the disease, such as progression and onset.Method: In order to identify common and rare genetic variants associated with rates of cognitive decline in AD, we aim at systematically analyzing the largest collection of longitudinal and GWAS data (N = 7,241), across non‐Hispanic white individuals from multicenter cohorts (Knight‐ADRC, ADNI, NACC, and GSK) with at least 1.5 years of follow up after being diagnosed with AD. We also aim at exploring if genetic risk factors, alone or interacting with other genetic factors, are associated rate of cognitive decline in patients with AD when aggregated together into a single Polygenic Risk Score (PRS). Furthermore, external proteomics data will be combined with the genetics data for Mendelian Randomization (MR) analyses in order to discover causal biomarkers for cognitive decline and AD risk.Result: Preliminary results found a suggestive signal in chr21 (min P = 8.7e‐8) approaching close to the genome‐wide significance threshold. Moreover, some other suggestive signals were detected in chr2 (min P = 3.8e‐06) and chr1 (min P = 3.7e‐06) that warrant further investigation. Further results and key findings of the post‐GWAS analyses will be presented during the AAIC meeting.Conclusion: We have leveraged these datasets to generate prediction models for rate of cognitive decline in AD. The identification of AD‐specific genetic variants and biomarkers will enable the characterization of appropriate therapeutic targets for slowing or halting the AD progression.

  • Open Access

    Preclinical Alzheimer’s Disease accurate prediction using plasma cell‐free RNA sequences

    by Alejandro Cisterna‐García, Hsiang‐Han Chen, Joanne Norton, Jen Gentsch, Kristy Bergmann, Fengxian Wang, John P. Budde, Carlos Cruchaga, Juan A. Botía, Laura Ibanez

    Background: Cerebrospinal fluid (CSF) Alzheimer’s disease (AD) biomarkers are used to clinically diagnose AD with an accuracy of 70‐80%. However, they are invasive and expensive and need to be further studied on preclinical‐AD. Additionally, a blood‐based biomarker can be of great use. In order to obtain a blood‐based biomarker, we have sequenced plasma cell‐free RNA (cfRNA) from preclinical‐AD individuals to develop prediction models for AD using machine learning (Figure 1).Method: Plasma samples were obtained from the Knight‐Alzheimer‐Disease‐Research‐Center [preclinical‐AD (n = 67), clinical‐AD (n = 92), and controls (n = 48)] (Table 1) and sequenced in two batches. Sequences were processed following standard pipelines and normalized using DESeq2. Preclinical‐AD (n = 47) and controls (n = 26) from one sequencing batch were used to train the predictive model, then preclinical‐AD (n = 20) and controls (n = 22) from the other batch were used as testing population. We applied a Z‐score scaling to the normalized gene counts and used Kullback–Leibler (KL) divergence to select the genes that had similar expression distribution in both sequencing experiments (Figure 2). We used different KL thresholds to select genes as predictors and then Ridge regression to find the best predictive models. After we tested the models on preclinical‐AD, they were further tested on the clinical‐AD individuals. Finally, Parkinson’s disease (n = 96), dementia with Lewy bodies (n = 17), and frontotemporal dementia (n = 16) individuals were used to test for specificity analyses of the AD prediction models (Table 2).Result: Best predictors sizes include 40, 90, and 220 genes yielding models with 85.7%, 90.5%, and 95.2% testing accuracy respectively. Predicted AD risk consistently correlates with Aβ42 levels. The best model included genes such as MT‐ATP6, GAB2, or MAPK14, which have been associated with AD. We tested the specificity of these models for other neurodegenerative diseases, and we observed specificity for AD in comparison to PD, FTD, and DLB (Figure 3).Conclusion: Plasma cfRNA is a promising tool to screen for preclinical AD. It has the potential to be used as a minimally invasive biomarker for AD since it has an accuracy higher or comparable to CSF biomarkers and it is AD‐specific.

  • Open Access

    Use of the human reference genome 38 enhances integration of the multi‐tissue proteomics with genetics and disease

    by Heng Yi, Qijun Yang, Jigyasha Timsina, Fengxian Wang, Lihua Wang, Carlos Cruchaga, Yun Ju Sung

    Background: Most existing protein quantitative trait locus (pQTL) studies focus on a single tissue and use the reference Genome Reference Consortium Human Build 37 (GRCh37/hg19) published in 2009. GRCh38/hg38 was updated in 2013 with correcting thousands of sequencing artifacts that cause false SNPs and updating non‐nuclear genomic sequence. Many papers strongly recommend switching to GRCh38/hg38. The goal of this study is to use multiple tissues (brain, CSF and plasma) to explore the genetic architecture of protein levels in neurologically relevant tissues and to update the human reference genome as hg38 to compare the findings.Method: In this study, 805 CSF samples and 869 proteins, 529 plasma samples and 953 proteins and 378 brain samples and 1,300 proteins passed quality control process. We performed genome‐wide association analyses of over 8 million autosomal variants (MAF ≥ 0.01) imputed using the most updated hg38 TOPMed imputation panel to identify cis/trans pQTLs. We performed conditional analysis to identify independent pQTLs. After removing pleiotropic loci and including Alzheimer’s disease (AD) variants, Mendelian randomization (MR) was applied to detect the casual associations between proteins and AD risk. To decrease unmeasured pleiotropy effect, we used co‐localization analysis to get more supporting information among multiple traits. Finally, we replicated our findings in much larger studies.Result: In brain tissue, we identified 4269 pQTLs in hg38 compared to 2418 pQTLs in hg19 with a threshold of 5 × 10−8 for cis‐pQTLs and 5 × 10−8/(number of independent proteins) for trans‐pQTLs. We found the independent SNPs and highlight the complexity of regions with multiple independent local pQTLs. Analysis for CSF and plasma is underway.Conclusion: This study is to apply the hg38 as the human reference genome to detect much more pQTLs and correct possible false findings in the previous hg19 study. With the following analysis, we could get more valuable and confirmed information about the additional GWAS loci and identify the function of certain proteins on the disease risk. The multiple tissue samples and multiple traits could help us detect the complex genetic architecture of protein levels in neurologically relevant tissues.

  • Open Access

    Metabolomic and lipidomic signatures in autosomal dominant and late‐onset Alzheimer's disease brains

    by Brenna C. Novotny, Maria Victoria Fernandez, Ciyang Wang, John P. Budde, Kristy Bergmann, Abdallah M. Eteleeb, Joseph Bradley, Carol Webster, Curtis Ebl, Joanne Norton, et al.

    Introduction: The identification of multiple genetic risk factors for Alzheimer's disease (AD) suggests that many pathways contribute to AD onset and progression. However, the metabolomic and lipidomic profiles in carriers of distinct genetic risk factors are not fully understood. The metabolome can provide a direct image of dysregulated pathways in the brain.Methods: We interrogated metabolomic signatures in the AD brain, including carriers of pathogenic variants in APP, PSEN1, and PSEN2 (autosomal dominant AD; ADAD), APOE ɛ4, and TREM2 risk variant carriers, and sporadic AD (sAD).Results: We identified 133 unique and shared metabolites associated with ADAD, TREM2, and sAD. We identified a signature of 16 metabolites significantly altered between groups and associated with AD duration.Discussion: AD genetic variants show distinct metabolic perturbations. Investigation of these metabolites may provide greater insight into the etiology of AD and its impact on clinical presentation.HIGHLIGHTS: APP/PSEN1/PSEN2 and TREM2 variant carriers show distinct metabolic changes. A total of 133 metabolites were differentially abundant in AD genetic groups. β‐citrylglutamate is differentially abundant in autosomal dominant, TREM2, and sporadic AD. A 16‐metabolite profile shows differences between Alzheimer's disease (AD) genetic groups. The identified metabolic profile is associated with duration of disease.


  • Open Access

    Leveraging large multi-center cohorts of Alzheimer disease endophenotypes to understand the role of Klotho heterozygosity on disease risk

    by Muhammad Ali, Yun Ju Sung, Fengxian Wang, Maria V. Fernández, John C. Morris, Anne M. Fagan, Kaj Blennow, Henrik Zetterberg, Amanda Heslegrave, Per M. Johansson, et al.

    Two genetic variants in strong linkage disequilibrium (rs9536314 and rs9527025) in the Klotho (KL) gene, encoding a transmembrane protein, implicated in longevity and associated with brain resilience during normal aging, were recently shown to be associated with Alzheimer disease (AD) risk in cognitively normal participants who are APOE ε4 carriers. Specifically, the participants heterozygous for this variant (KL-SVHET+) showed lower risk of developing AD. Furthermore, a neuroprotective effect of KL-VSHET+ has been suggested against amyloid burden for cognitively normal participants, potentially mediated via the regulation of redox pathways. However, inconsistent associations and a smaller sample size of existing studies pose significant hurdles in drawing definitive conclusions. Here, we performed a well-powered association analysis between KL-VSHET+ and five different AD endophenotypes; brain amyloidosis measured by positron emission tomography (PET) scans (n = 5,541) or cerebrospinal fluid Aβ42 levels (CSF; n = 5,093), as well as biomarkers associated with tau pathology: the CSF Tau (n = 5,127), phosphorylated Tau (pTau181; n = 4,778) and inflammation: CSF soluble triggering receptor expressed on myeloid cells 2 (sTREM2; n = 2,123) levels. Our results found nominally significant associations of KL-VSHET+ status with biomarkers for brain amyloidosis (e.g., CSF Aβ positivity; odds ratio [OR] = 0.67 [95% CI, 0.55–0.78], β = 0.72, p = 0.007) and tau pathology (e.g., biomarker positivity for CSF Tau; OR = 0.39 [95% CI, 0.19–0.77], β = -0.94, p = 0.007, and pTau; OR = 0.50 [95% CI, 0.27–0.96], β = -0.68, p = 0.04) in cognitively normal participants, 60–80 years old, who are APOE e4-carriers. Our work supports previous findings, suggesting that the KL-VSHET+ on an APOE ε4 genotype background may modulate Aβ and tau pathology, thereby lowering the intensity of neurodegeneration and incidence of cognitive decline in older controls susceptible to AD.


  • Open Access

    Multi-tissue proteomics identifies molecular signatures for sporadic and genetically defined Alzheimer disease cases

    by Carlos Cruchaga, Yun Ju Sung, Chengran Yang, Fengxian Wang, Adam Suhy, Joanne Norton, Brenna Novotny, Abdallah Etelleb, Anne Fagan, Randall Bateman, et al.

    Alzheimer disease (AD) is a heterogeneous disease with many genes are associated with AD risk. Most proteomic studies, while instrumental in identifying AD pathways and genes, focus on single tissues and sporadic AD cases. Multi-tissue proteomic signatures for sporadic and genetically defined AD (e.g., pathogenic variant carriers in APP and PSEN1/2 and risk variant carriers in TREM2) will illuminate the biology of this heterogeneous disease.1,2 Here, we present one of the largest multi-tissue proteomic profiles, accessible through our web portal, based on 1,305 proteins in brain (n=360), cerebrospinal fluid (CSF; n=717), and plasma (n=490) from the Knight Alzheimer Disease Research Center (Knight ADRC) and Dominantly Inherited Alzheimer Network (DIAN) cohorts.3-5 We identified proteomic signatures in brain, CSF, and plasma for sporadic AD status and replicated these findings in multiple, independent datasets. The area under the curve (AUC) for CSF proteins was 0.89 in discovery and 0.90 in the replication dataset, which was significantly higher than the AUC for CSF p-tau181/Aβ42 (AUC = 0.81; P = 2.4×10-6). We also identified a specific proteomic signature for TREM2 variant carriers that differentiated TREM2 variant carriers from sporadic AD cases and controls with high sensitivity and specificity (AUC = 0.81 - 1). In addition, the proteins that showed differential levels in sporadic AD were also altered in autosomal dominant AD, but with greater effect size (1.4 times, P = 3.8×10-5), and proteins associated with autosomal dominant AD, in brain tissue also replicated on CSF (p=1.36×10-9). Enrichment analyses highlighted several pathways including AD (calcineurin, APOE, GRN), Parkinson disease (α-synuclein, LRRK2), and innate immune response (SHC1, MAPK3, SPP1) for the sporadic AD or TREM2 variant carriers. Our findings show the power of multi-tissue proteomics’ contribution to the understanding of AD biology and to the creation of tissue-specific prediction models for individuals with specific genetic profiles, ultimately supporting its utility in creating individualized disease risk evaluation and treatment.