Esophageal squamous cell carcinoma (ESCC) causes more than 509,000 deaths per year worldwide, and over 80% of these cases occur in low- and middle-income countries (LMICs). In the settings where it is available and affordable, ESCC diagnosis is made using the highly invasive and expensive procedure known as esophagogastroduodenoscopy (EGD, also referred to as upper endoscopy). However, citizens of underdeveloped countries often do not have widespread access to EGD; thus, there is a pressing need for easier, safer, cheaper, more accessible ESCC diagnostic strategies. We have developed a strategy based on our exciting preliminary data in ESCC patients: a swallowable, retrievable sponge-capsule device (EsophaCap) combined with a sensitive, specific DNA methylation biomarker panel (JHU patent pending) and machine learning classifiers. Capsulomics, Inc., a startup company from Johns Hopkins University, will commercialize this assay, termed “EsoCAN,” for ESCC diagnosis. The goal of this project is to explore the feasibility and development of a non-endoscopic diagnostic test as a more convenient, inexpensive, and more widely usable alternative to EGD in LMICs. Ultimately, this test, when applied to at-risk populations, is likely to significantly improve survival and quality of life. As an initial step toward this goal, we will pursue the following Specific Aims: Aim 1, analytical and clinical validation of quantitative methylation-specific PCR (qMSP) assays for 5 candidate loci: cg20655070, SLC35F1, TAC1, ZNF132, and ZNF542, including a pilot study of EsophaCap collected specimens from 50 ESCC patients and 70 controls, and Aim 2, creation of multivariate machine learning (ML) diagnostic biomarker models. In Phase II, we will further develop and evaluate our non-endoscopic diagnostic test in a larger patient cohort for accuracy of ESCC diagnosis in LMICs.