Salivary Extracellular Vesicles in Detection of Cancers Other than Head and Neck: A Systematic Review

Abstract

Cancer is one of the leading causes of death worldwide. Evidence indicates that extracellular vesicles are involved in cancer development and may be used as promising biomarkers in cancer detection. Concomitantly, saliva constitutes a non-invasive and inexpensive source of biomarkers. This systematic review investigates the use of salivary extracellular vesicles in detecting cancers located outside of the head and neck. PubMed, Web of Science, Scopus, and Embase were thoroughly searched from database inception to 16 July 2024. Data from sixteen eligible studies were analyzed, including glioblastoma, lung, esophageal, gastric, prostate, hepatocellular, breast, and pancreatobiliary tract cancers. The findings highlight strong diagnostic potential for lung and esophageal cancers, where specific exosomal RNAs and proteins demonstrated high accuracy in distinguishing cancer patients from healthy individuals. Additionally, biomarkers in glioblastoma showed prognostic value, while those in hepatocellular and pancreatobiliary cancers exhibited potential for early detection. However, gastric and prostate cancer biomarkers showed limited reliability, and breast cancer biomarkers require further validation. In conclusion, salivary extracellular vesicles present potential in non-invasive detection across multiple cancer types; however, their diagnostic power needs further research, including standardization and large-scale validation.

Keywords: biomarkers; cancer; diagnosis; extracellular vesicles; saliva.

Figure 1

Study design of the systematic review. Eligible studies describe saliva collection, extracellular vesicles isolation, analysis, and potential in cancer detection. This systematic review investigates the potential use of salivary extracellular vesicles in cancer detection and assesses their utility as cancer biomarkers.

Figure 2

PRISMA flow diagram presenting search strategy.

Figure 3

Quality assessment, including the main potential risk of bias (risk level: green—low, yellow—unspecified, and red—high and quality score: green—good, yellow—intermediate, and red—poor) [36,37,38,39,40,41,42,43,44,45,46,47,48,49,50,51].