Mentorship Meetings

Anish S. Shah, MD, MS

University of Utah

University of Illinois Chicago

Emory University

06/28/24 | DD

Agenda

  1. Review findings from ECG/TTN project
  2. Update on AFEQT/SDOH manuscript
  3. Timeline/priority for active projects
    1. ECG endo-phenotyping of AF genetic variants
    2. AF-CTRS
    3. Lab projects (data pulls, F-wave + BMP, future variant/data annotation)
  4. K23 timeline + publications

ECG association with TTN variants, with gradients highlighting heavily weighted features

O6/24/24 | RR

Agenda

  1. Introduction in person
  2. General timeline for progress over first 3-6 months (caveat being appropriate clinical progress)
  3. Lab integration and co-authorship/contribution to ongoing projects
  4. Specific aims for K23
  5. Logistics: IRB, cluster access, compute/statistical support, clinical data registry access
  6. Additional research thoughts/projects: a. Epidemiology & biostatistics background – communicated with Ben Steinberg about the research resources available a. Co-authorship with yourself, Jared Bunch, Rob MacLeod a. Software development for EGM analysis (ML approach) a. Atrial myopathy phenotyping by genetic variants (ML approach)

06/05/23 | AJS

Agenda

  1. AFEQT/SDOH
    • Manuscript
    • AHA Early Career Award (LOR)
  2. CARRS paper
  3. K23 application (aims following)
  4. Emory IRB as external collaborator
    • UK Biobank
    • GWAS from PPG data
  5. Zombie projects
    • Cosinor analysis

K23 aims

Aim 1: Measure the differential electrophysiologic effects of the ANS on the atrial substrate and risk for atrial arrhythmias. Hypothesis: EP characteristics with ANS modulation will show regional differences across the atria. Secondary hypothesis: Scar tissue will have increased heterogeneity of refractoriness in response to ANS modulation, with an exaggerated shortening of APD, relative to baseline. During clinically-indicated EP studies for atrial arrhythmias, we will measure EP characteristics under ANS modulation through both pacing (e.g. vagal stimulation) and pharmacological maneuvers (e.g. isoproterenol infusion).

Aim 2: Compare the efficacy of computational scar-based models of atrial arrhythmias with predictive models accounting for ANS. We have designed predictive models of atrial arrhythmias based off of scar patterns identified by clinical imaging. This aim will augment the models by incorporating ANS parameters. Hypothesis: Atrial-specific ANS parameters will improve prediction of clinically-relevant arrhythmias. Secondary hypothesis: ANS parameters obtained from Aim 1 will validate parameters utilized in computational models.

Aim 3: Explore the contribution of ANS genetic variants to the polygenic risk of atrial arrhythmias. Using a two-sample Mendelian randomization study design, we will identify SNPs that contribute to stress-related disorders, including potentially novel loci, and assess their impact on the polygenic risk of AF in a validation cohort. Hypothesis: Genetic variants associated with abnormalities in the ANS, including stress-related disorders, will overlap with and partially explain both risk and resilience to future development of AF.

05/22/24 | AJS

Agenda

  • HRV/CVD Mortality in JAMA Network Open
  • CARRS paper
  • UK Biobank / MR for AF genetics (validation cohort)
  • K23 approach
    1. EP-lab and pharmacological “mental stress cocktail” in PTSD
    2. Computational prediction of arrhythmias incorporating neurocardiac axis in PTSD
    3. Genetic variants in stress response and arrhythmia risk in PTSD
  • Software updates (modeling package published, EGM package + paper pending)
  • AHA early career investigator award (Epi? Clinical Cardiology?)
    • AFEQT/SDOH? → manuscript drafted but pending
    • HRV/CVD paper? → however abstract already published

05/07/24 | AJS

AF DAG

flowchart LR
    ecg["ECG"]
    scar["Scar"]
    comp["Computational Modeling"]
    arr["Arrhythmia Prediction"]
    eps["EPS"] 
    mri["MRI"]
    phen["Phenotypes"]
    ptsd["PTSD"] 
    ans["Autonomic Modulation"]
    geno["Genotypes"] 
    ans --> eps
    ptsd --> geno
    ecg --> geno
    scar --> ecg
    mri --> ecg
    ecg --> phen
    ecg --> geno
    geno <--> phen
    eps --> comp
    scar --> comp
    comp --> arr

04/17/24 | RR

Agenda

  1. Current/updates on research areas
    1. AFGen collaboration
    2. UIC/Emory mentors
  2. K23/KL2/CDA application
    1. Mentorship network at Utah
    2. Collaborative overlap
    3. Specific aims overview

Updates

  • Expect \(\ge\) 5 first-author publications prior to EP fellowship start (2 published, 2 in review, 1 being drafted)
  • Two software publications (harmonic regression statistical package, causality-based modeing package)
  • Current project on phenotyping AF based on ECG and neural networks to predict genetic variants (AFGen collaboration) \(\leftarrow\) manuscript (after validation)
  • Evaluating stress-associated genetic variants in WES/GWAS for future AF risk, with plans for validation via UK Biobank (Emory collaboration)

Utah-based research

  • MacLeod: work on ECG-I for arrhythmia prediction and atrial tissue characterization
  • Ranjan/Dosdall: work on MRI-based prediction of arrhythmia pathways
  • Tristani-Firouzi: genetics of arrhythmias, Utah Population Database

Draft of Aims

Aim 1: Identify ECG-based phenotypes of paroxysmal AF. We will evaluate the ECG manifestation of (A) patterns of scar based on cardiac MRI/EPS, and (B) contributions of genetic variants in sarcomeric/structural proteins and ion channelopathies that are associated with AF. Our working hypothesis is that scar-based atrial myopathy can be identified by ECG, and can distinguish between re-entry versus focal triggers for AF.

Aim 2: Determine the differences in APD heterogeneity of scar tissue in the context of modulated autonomic activity. Our working hypothesis is that the interaction between APD heterogeneity and autonomic modulation will predict differential risk for future AF, particularly in cases of abnormal baseline autonomic activity such as PTSD.

Aim 3: Identify the role of novel, rare variants for future AF risk in stress-related conditions. Our working hypothesis is that rare variants identified in a population with abnormal stress responses will be associated wiht increased future risk of AF in a large cohort study.

04/16/24 | SHC

Agenda

  1. Review slides from AFGen RIP presentation
  2. Extension of project
    1. Extension/continuation of project
    2. Timeline
    3. Validation

Project extension

  • Consider as part of K23 application focused on ECG-based phenotyping
  • Phenotyping but focusing on structural/sarcomeric and ion channel variants
  • Initial training data to be analyzed by mid-May, with intent for AHA submission
  • Similar to…
  • Validation?
    • All of Us
    • TOPMed
    • UK Biobank

ECG Prediction of Structural and Ion Channel Mutants in AF

This is an expansion of the ECG/TTN project, which found ECG could feasibly provide above-chance accuracy in identify TTN LOF variants. Now, expand to additional wild-types and mutants…

  • Sarcomeric/structural variants (e.g. TTN, PITX2, LMNA, NUP155, GJA1/5)
  • Ionic channel variants (e.g. KCNE1-5, KCNQ1, SCN5A)
  • No known genetic contributors (excluding any patient with non-benign VUS in any above genes \(\pm\) SNPs associated with polygenic risk for AF)

02/14/24 | EJB

Agenda

  1. Personal/professional updates (brief)
  2. “Big hairy audacious goal”
  3. AFGen fellowship/consortium, TOPMed
  4. K23 location and strategy
    1. Specific aims
    2. Mentorship network
    3. Importance of “ownership” of datasets

02/12/24 | RR

Agenda

  1. Personal research status
    1. “Big picture” goal - functional prediction of arrhythmia onset
    2. Current research projects
    3. Mentorship network
  2. Research overview at Utah
    1. Research groups with overlapping interest
    2. Strategic mentorship for K23 award
    3. Timeline and expectations

Big picture

Important

Why do large, underwater mammals, with respectively larger atria have low burden of atrial fibrillation? Why do arrhythmias only occur on ascent phase of dive?

  1. Plan to be with translational cardiac electrophysiologist, focusing on signal processing, molecular/genetic mechanisms, and the autonomic nervous system
  2. Underlying question is how stress can lead to arrhythmogenesis
  3. Study how arrhythmia events are mediated by potential inappropriate vagolytic mechanisms in atrial fibrillation.
  4. Long-term goal is to phenotype AF and develop targeted, individualized therapies (e.g. functional mapping, alternative boundary predictions in EP)

Current projects

  • ECG-based prediction of severity of TTN variants using deep learning (computational/genetics)
  • Machine learning for annotations of EGM (computational)
  • Phenotyping paroxysmal AF based on ECG and TTE criteria (epidemiology)
  • Mental stress reactivity and CVD mortality (epidemiology)
  • AFEQT and social determinants of health in minority population (epidemiology)
  • “Isolated” or “premature” AFL with family history (genetics)
  • Cosinor/harmonic regression software (computational)
  • GWAS of abnormal stress reactivity

Non-specific aims

  1. Epidemiology / ECG-based aim: risk prediction, machine learning (i.e. ECG-informed arrhythmia risk scores)
  2. EP-lab based aim: repolarization mapping, intra-EPS stress testing
  3. Genetics-focused aim: TTN impact on AF, genetics of stress physiology (e.g. HRV)

Research network

  • Amit J. Shah, MD, MSCR (Emory) = cardiologist \(\rightarrow\) Rob MacLeod, PhD
  • Alvaro Alonso, MD, PhD (Emory) = epidemiologist, ARIC, atrial fibrillation
  • Viola Vaccarino, MD, PhD (Emory) = epidemiologist, mental stress
  • Arshed Quyyumi, MD (Emory) = cardiologist, coronary disease
  • Rachel Lampert, MD (Yale) = electrophysiologist, mental stress
  • Dawood Darbar, MD (UIC) = electrophysiologist, genetics, atrial fibrillation \(\rightarrow\) Martin Tristani-Firouzi, MD
  • Emelia Benjamin, MD, ScM (BU) = cardiologist, epidemiologist, atrial fibrillation, genetics (AFGen Consortium)

Research at Utah

  • MacLeod = signal processing for ECG/EGM data
  • Ranjan = MRI and structural evaluation of atria (in AF)
  • Bunch = AF and cognitive decline (area of interest of Alvaro Alonso as well)
  • Tristani-Firouzi = genetics of arrhythmias, Utah Population Database
  • Palatinus = molecular mechanisms of current trafficking
  • Guo = genetics/VUS and CVD mechanisms
  • Dosdall = EGM/arrhythmia mapping, activation patterns

Wrap-up

  • Opinions and thoughts
  • Questions/concerns

References

Choi, Seung Hoan, Sean J. Jurgens, Christopher M. Haggerty, Amelia W. Hall, Jennifer L. Halford, Valerie N. Morrill, Lu-Chen Weng, et al. 2021. “Rare Coding Variants Associated With Electrocardiographic Intervals Identify Monogenic Arrhythmia Susceptibility Genes.” Circulation. Genomic and Precision Medicine 14 (4): e003300. https://doi.org/10.1161/CIRCGEN.120.003300.
Ntalla, Ioanna, Lu-Chen Weng, James H. Cartwright, Amelia Weber Hall, Gardar Sveinbjornsson, Nathan R. Tucker, Seung Hoan Choi, et al. 2020. “Multi-Ancestry GWAS of the Electrocardiographic PR Interval Identifies 202 Loci Underlying Cardiac Conduction.” Nature Communications 11 (1): 2542. https://doi.org/10.1038/s41467-020-15706-x.
Wang, Xin, Shaan Khurshid, Seung Hoan Choi, Samuel Friedman, Lu-Chen Weng, Christopher Reeder, James P. Pirruccello, et al. 2023. “Genetic Susceptibility to Atrial Fibrillation Identified via Deep Learning of 12-Lead Electrocardiograms.” Circulation: Genomic and Precision Medicine 0 (0): e003808. https://doi.org/10.1161/CIRCGEN.122.003808.
Weng, Lu-Chen, Amelia Weber Hall, Seung Hoan Choi, Sean J. Jurgens, Jeffrey Haessler, Nathan A. Bihlmeyer, Niels Grarup, et al. 2020. “Genetic Determinants of Electrocardiographic P-Wave Duration and Relation to Atrial Fibrillation.” Circulation. Genomic and Precision Medicine 13 (5): 387–95. https://doi.org/10.1161/CIRCGEN.119.002874.