Noura Serene Abul-Husn

Background: Electronic health records (EHR) are an important data source for genomic studies, but challenges exist in ascertaining cases and observation start time. We used data derived from the Electronic Medical Records and Genomics (eMERGE) IV study to examine how analytic assumptions regarding case ascertainment and EHR entry time influence estimation of monogenic and polygenic risks for coronary heart disease (CHD). Methods: We assessed agreement between CHD cases ascertained from EHR phenotyping and survey. Associations of monogenic variants and high (top 5%) PRS of CHD were evaluated using multivariate relative risk (RR) regression under three alternative case definitions: EHR-algorithm-defined, self-reported, and combined. Time-to-event analyses (Kaplan–Meier method and Cox proportional hazards models) were conducted under three entry time specifications: (1) entry at the first EHR record, (2) entry at the start of the latest consecutive observation period, and (3) no left truncation. Results: The agreement between CHD cases ascertained by the EHR-based algorithm versus self-report was 37.2% among individuals identified as cases by at least one source, with the EHR algorithm demonstrating higher accuracy. The adjusted RR [95% confidence interval (CI)] associated with high PRS was 2.05 [1.50, 2.81] for EHR-defined, 1.49 [1.04, 2.13] for self-reported, and 1.66 [1.27, 2.18] for combined CHD. Estimated cumulative incidence by age 75 was 0.188 using the first EHR code as left truncation and 0.225 using the most recent observation period. Hazard ratio (HR) estimates were similar across the three left truncation scenarios. Conclusion: The choice of case definition meaningfully influenced RR estimates, whereas alternative specifications of EHR entry time affected absolute cumulative incidence estimates but has minimal impact on HR. These findings highlight the impact of analytical choices in EHR and survey-data-based studies that are applicable beyond the context of CHD.

Objectives: To identify unique echocardiographic signatures associated with TTR + carrier status preceding onset of cardiac amyloidosis. Background: Carrier status for the most common pathogenic TTR variant in the United States, Val142Ile (V142I), found in 4% of African Americans (AA) and 1% of Hispanic/Latino (H/L) individuals, confers a 40-60% lifetime risk of developing variant transthyretin amyloidosis (ATTRv), including cardiac amyloidosis (CA) and heart failure (HF). Myocardial amyloid deposition is believed to progress over many years. Genomic screening programs and familial cascade genetic testing are increasingly uncovering pre-symptomatic TTR + carriers, yet no guidelines exist to pragmatically risk stratify these individuals for CA. Methods: V142I+ carriers (cases) without prior diagnoses of amyloidosis or HF were identified among Bio Me biobank participants with available exome sequencing data linked to electronic health records (EHRs) including at least one available echocardiogram. Controls were biobank participants with normal TTR sequencing who were age-, sex- and ancestry-matched to cases. Speckle-tracking echocardiography (STE) was applied to images and conventional and strain measurements were evaluated by univariate analyses. A random forest model was trained using a minimal redundancy maximal relevance (mRMR, applied to mitigate overfitting) feature set and evaluated by 5-fold cross-validation to minimize optimism bias. Discriminatory performance was assessed using the area under the receiver operating characteristic curve (AUC). Results: 49 TTR + (100% V142I, median age 61 years, 69.4% female) and 45 matched TTR -biobank participants were included in the model development cohort. STE generated approximately 200 features. Univariate analyses revealed no significant differences between carriers and controls on any individual strain or conventional echocardiographic measurements including global longitudinal, right ventricular and left atrial strain. mRMR feature selection resulted in a set of 15 features retained for all downstream modeling, integrating global amyloid signatures, regional inferolateral strain abnormalities, layer-specific deformation, and mechanical timing heterogeneity. Using this feature set, the model achieved good discrimination (AUC=0.76). Feature importance analysis highlighted relative apical sparing, inferolateral strain reduction, and basal-apical timing gradients as key contributors to model performance. External validation (n=115) confirmed good model discrimination (AUC=0.781, 95% CI: 0.688-0.869, sensitivity 0.983). Conclusions: Machine learning applied to routinely acquired echocardiographic data can identify subtle myocardial abnormalities associated with TTR V142I carrier status prior to development of CA. Key model features are physiologically relevant to known echocardiographic characteristics of overt CA. Genotype-guided echocardiographic surveillance may be a scalable strategy for early detection of CA risk.

Limited data are available on genomic risk factors for breast cancer in diverse populations. We evaluated breast cancer risk conferred by high and moderate penetrance genes, polygenic risk scores (PRS), and family history (FH) in an ancestrally diverse biobank. We identified expected pathogenic variants (EPVs) in 14 genes among 30,223 sequenced individuals, including 15,919 unrelated women. Using electronic health records (EHRs), we evaluated associations of EPVs and PRS with breast cancer, in the absence and presence of FH. EPVs in high penetrance genes were associated with increased risk of breast cancer in the absence or presence of FH (OR 6.8, p = 1.4 x 10-10; OR 11.8, p = 4.1 x 10-21; respectively). EPVs in moderate penetrance genes were associated with increased risk of breast cancer only in the presence of FH (OR 7.0, p = 4.7 x 10-10 with FH vs. OR 1.2, p = 0.6 without). High PRS (≥ 90th percentile) was associated with increased risk (OR 1.8, p = 3.3 x 10-8) and low PRS (≤ 10th percentile) with reduced risk (OR 0.6, p = 1.6 x 10-3) of breast cancer. FH modulated risk across all PRS strata. EHR review of 252 women with EPVs in moderate penetrance genes revealed that 75 (29.8%) met criteria for clinical genetic testing, and only 27 (36.0%) had undergone testing. High and moderate penetrance genes and PRS confer varying degrees of breast cancer risk, depending on FH. There is a need to better identify women with moderate penetrance gene variants.

Background: Phenotype Risk Scores (PheRS) leverage electronic health record (EHR) data to identify individuals at risk for Mendelian disorders, but their performance remains untested for diseases with common and/or non-specific features such as variant transthyretin amyloidosis (ATTRv), often presenting with heart failure (HF), atrial fibrillation, polyneuropathy, and other prevalent diagnoses. We optimized a PheRS for the most common form of ATTRv by integrating genomic and clinical data in Mount Sinai’s Bio Me biobank, focusing on expert-driven phenotype definitions for the TTR variant p.Val142Ile (V142I), which is prevalent in African American (AA) populations (4%). Methods: We developed and evaluated a customized PheRS for ATTRv that incorporated 21 expert-curated phenotypic features including 292 ICD-9 and ICD-10 diagnosis codes on a biobank cohort of V142I+ cases (n=383) and controls without any pathogenic/likely pathogenic TTR variants (n=30,642). We compared its performance with the standard automated PheRS approach using different metrics. To account for age-dependent penetrance and high lifelong risk of HF, we further tested the customized PheRS for V142I in a subset of individuals of age ≥ 60 with self-reported Black or AA race/ethnicity and at least one occurrence of HF in their EHRs. Results: The expert-curated PheRS outperformed the standard PheRS as measured by improved precision-at- k (0.05 vs. 0.00; k =100), a demonstrably, clinically relevant metric. In the subcohort enriched for anticipated penetrance (older, Black/AA HF patients), the expert-curated PheRS identified more V142I+ individuals (6.0%) among the top 100-scoring individuals than a strategy that randomly sampled from the population (3.6%). Conclusion: This work demonstrates that standard PheRS methods are insufficient for common, adult-onset cardiovascular genetic diseases such as V142I-related ATTRv, but when redesigned with disease biology, ancestry, age, and clinical context in mind, PheRS become clinically actionable tools for precision cardiology.

Background: Depression is a frequent focus of interest in genetic testing. Despite growing availability of polygenic risk scores (PRS) for depression, little is known about the psychological impact of receiving them in real-world settings. To quantify the impact of receiving an at-risk depression PRS result on depression and anxiety symptoms, we conducted a longitudinal, prospective cohort study of 23andMe, Inc. research participants. Methods: Surveys were conducted between October 19, 2022 and October 9, 2023. Eligible participants were U.S. residents ≥ 18 years old who completed two surveys assessing depression and anxiety symptoms and had an at-risk PRS result for depression (odds ratio ≥ 1.5). We compared individuals who viewed their result to individuals who did not. Primary outcomes were changes in depression (Patient Health Questionnaire-8) and anxiety (Depression Anxiety Stress Scale-21) symptom scores relative to baseline. We fitted linear regressions to model each outcome, adjusting for age, sex, genetic ancestry, income, prior depression and anxiety, and baseline scores. Using an equivalence testing framework, the smallest effect size of interest was defined as Cohen’s d = ±0.5. Findings: We analyzed data from 917 participants, including 361 who viewed the depression PRS and 556 who did not. Score changes from baseline to follow-up were statistically equivalent for individuals who viewed PRS results and those who did not. The adjusted between-group differences in score changes were −0.17 points for depression (90% CI, −0.59–0.24, two one-side tests p < 0.001) and −0.092 points for anxiety (90% CI, −0.35–0.17, two one-side tests p < 0.001), both equivalent within the predefined margin. Results were consistent in substrata defined by presence or absence of prior depression or anxiety. Interpretation: Among genetically at-risk individuals, exposure to a depression PRS result was well-tolerated in a real-world setting. Funding: 23andMe, Inc.