LAUSR

Simple Summary Gastric cancer (GC) is the fourth most common cancer that occurs worldwide, affecting specifically the Asian population. Currently, there are no available screening programs for GC in United States. Since saliva is a highly desirable body fluid for developing biomarkers of cancer screening, early detection, and monitoring, we previously reported that salivary extracellular RNAs could be developed to detect gastric cancer in a Korean cohort, and here, we validate them in a U.S. cohort. Our study emphasizes the importance of population-specific biomarker development and validation, and specifically, the noninvasive nature of salivary biomarkers for population-based screening in at-risk populations. Abstract Gastric cancer (GC) has the fifth highest incidence among cancers and is the fourth leading cause of cancer-related death GC has predominantly a higher number of cases in certain ethnic groups such as the Korean population. GC found at an early stage is more treatable and has a higher survival rate as compared with GC found at a late stage. However, a diagnosis of GC is often delayed due to the lack of early symptoms and available screening programs in United States. Extracellular RNA (exRNA) is an emerging paradigm; exRNAs have the potential to serve as biomarkers in panels aimed at early detection of cancer. We previously reported the successful use of a panel of salivary exRNA for detecting GC in a high-prevalence Korean cohort, and that genetic changes reflected cancer-associated salivary exRNA changes. The current study is a case-control study of salivary exRNA biomarkers for detecting GC in an ethnically distinct U.S. cohort. A model constructed for the U.S. cohort combined demographic characteristics and salivary miRNA and mRNA biomarkers for GC and yielded an area under the receiver operating characteristic (ROC) curve (AUC) of 0.78. However, the constituents of this model differed from that constructed for the Korean cohort, thus, emphasizing the importance of population-specific biomarker development and validation.