Survivors of childhood cancer treated with anthracycline chemotherapy or chest-directed radiation are at substantially increased risk for developing cardiomyopathy that progresses to heart failure (HF). Compared to the general population, survivors have a 15-fold greater risk of HF and 7-fold excess risk of death due to cardiovascular (CV) causes. Within this vulnerable clinical population, African American (AA) survivors face a disproportionate burden of cardiac morbidity and mortality, largely driven by the higher burden of potentiating CV risk factors and cardiomyopathy. In our recent work, we found the prevalence of cardiomyopathy was 2.5- times higher among AA survivors than Non-Hispanic Whites (NHWs). However, existing survivor-directed HF risk prediction models are not race-specific. Given effective strategies to identify/detect cardiomyopathy and prevent HF in survivors are likely to differ by race/ethnicity, we hypothesize that novel race-specific cardiomyopathy risk prediction models developed and validated exclusively in AA survivor cohorts will significantly enhance clinical risk stratification in this population. Using data from the St. Jude Lifetime Cohort (SJLIFE) with up to 1,200 AA survivors that is unprecedented in terms of its size and extent of clinical characterization, we propose to: (a) perform a first-of-its-kind analysis to systematically identify influential predictors and develop (“train”) new cardiomyopathy risk algorithms in 700 AA survivors with existing deep clinical annotation from direct, in-person assessment and whole-genome/-exome sequencing data; and (b) validate (“test”) these novel risk models in independent SJLIFE AA survivor data (N~481). Our specific aims are to: (1) develop and independently validate clinical risk prediction models for cardiomyopathy for AA survivors considering primary cancer treatments and conventional CV risk factors, followed by incremental model-building to include informative cardiac biomarker, lifestyle risk, and personal-/area-/policy-level social determinants of health (SDOH) predictors; (2) identify genetic predictors for cardiomyopathy and evaluate their added contributions to clinical risk prediction; and (3) contextualize the performance of these novel models, including a generalizability assessment in existing independent data from AA survivors in multi-institutional studies (N=349) and comparative analyses considering published/validated HF risk models. Along with a race-specific study design, key innovations of our approach include the comprehensive assessment of predictors that have not been evaluated previously in survivor HF risk prediction models (e.g., cardiac biomarkers; personal-/area-/policy-level SDOH) and novel utilization of genetic data that can potentially maximize cardiomyopathy risk prediction, including genetic analyses of echocardiographic parameters predictive of cardiomyopathy and the development of polygenic risk scores that capture genetic susceptibili...