Table 3.

Preclinical studies about the circulating tumor-derived EV-gDNA as a potential cancer biomarker (2020–2021).

Aims/Samples	Main Results	Reference
Proteomic profile of potential cancer biomarkers in 426 human cancer and noncancer samples derived from various cells, tissues, and body fluids.	Crude sEVs categorized into (EVPs) three subpopulations: small exosomes (Exo-S 50–70 nm), large exosomes (Exo-L 90–120 nm), and exomeres (non vesicular (NV) particles <50 nm). Analysis of 120 plasma-derived EVP proteomes from 77 cancer patients with 16 different cancer types and 43 healthy controls (HC) suggested that EVP proteins can be useful for cancer detection and determinization of cancer type.	[29]
Extensive review on the EV biogenesis, focusing mainly on EXOs and MVs.	Discussion about the current knowledge upon EV-uptake and cell–cell communication, as well as upon the cargo sorting into EVs. Possible EV bioengineering methodologies for therapy improvements.	[31]
Comparison of EV-mediated liquid biopsy with older liquid biopsies for lung adenocarcinoma diagnosis.	EVs are advocated for as ideal carriers of cancer biomarkers. Contrary to the passively released fragmented cfDNAs (about 200 bp), cEV DNAs consist of both large-sized ds-gDNAs (up to 10 kb) and fragmented, mutated DNAs. The membrane-protected EV-DNAs also have a high stability A higher sensitivity can be achieved by using EV-DNAs obtained from bronchoalveolar lavage fluid (BALF) than those from blood.	[32]
Nine small-cell lung cancer (SCLC) patients and twenty-two (SCLC) patients with known tumor EGFR mutation.	Platelet-poor plasma was fractionated by five sequential centrifugations and ExoQuick for preparing the exosomal fraction 5, which was then dominated by small (~160 bp) nucleosome-like DNAs.Improved detection of cell-free tumor DNAs (cf-tDNAs) is claimed in EV-depleted plasma (fraction 6), and higher mutation detection rates (14/22) are observed than in whole plasma (10/22).	[33]
Blood samples from healthy human donors.	This older study contradicts the previous one by showing the association of dsDNA inside the plasma exosomes and stating that “more than 93% of amplifiable cfDNA in plasma is located in plasma exosomes”.	[34]
Human osteosarcoma (OS) serum samples.	Copurification of OS-associated repetitive element DNAs with EVs in size exclusion chromatography but not in exosome immunoaffinity capture. Repetitive element DNAs showed a high sensitivity and specificity for sera of patients with an OS diagnosis but were not tightly bound to CD9+ or CD81+ exosomes, supporting that exosomes either do not contain DNA or are tightly associated with particles with DNA.	[35]
Comparison of cf-tDNA and EV-DNA in serial plasma samples of a metastatic breast cancer patient.	Of the 52 copy number variants (CNVs) (from 0.3 to 106.5 Mb) in tDNA, 36 were detected in at least one cf-tDNA and 13 in one EV-DNA sample and were distributed randomly throughout the genome. cf-tDNA, shed from apoptotic tumor cells, had a greater sensitivity for serial monitoring of breast cancer than EV-DNA actively secreted from viable neoplastic cells.	[36]
Summary of the biological and clinical aspects of EV-DNA and role of EV-DNA in cancer.	EV-DNA as a biomarker for liquid biopsy is a new but definitely promising area of study, but its study in the clinical context is still quite open for further validation.	[37]
Bronchoalveolar lavage fluid (BALF) of 20 (NSCLC) patients with EGFR-mutations and matched fixed-tissue samples.	Heterogeneous (100–300 nm) EVs from BALF contained mostly ~11kb DNAs from both vesicle surface and inside. The DNA yield from BALF-EVs was 100 times less than tissue DNA but had enough tumor-specific DNA for the identification of actionable genetic alterations with a high potential clinical utility.	[38]
54 plasma samples and 13 pleural fluids of (NSCLC) patients after tyrosine kinase inhibitor therapy.	By comparison of different technological tools to detect EGFR mutations, combined tumor nucleic acid analysis (exoTNA+cfTNA) in the plasma and exoTNa in the pleural fluid allowed for the detection of target EGFR mutations more sensitively than using cfDNA or total DNA alone.	[39]
Focus on the DNA inclusion in EVs, the techniques of EV-DNA detection and quantification, and the clinical use of EV-DNA.	Recapitulation of the cell-free DNA cell sources by active or passive mechanisms and summary of the tumor genome hallmarks reflected by EV-DNA as well as the results of the main clinical studies assessing the performance of EV-DNA biomarkers.Enumeration of the many challenging questions remaining to be solved before reaching the clinics.	[40]
Cell lines and glioblastoma stem-like (GS) cell cultures.Human glioma patients’ tissue and nontumoral tissue.	The vast majority of EVs carry DNA, which localizes more to the EV surface than inside EVs. Proof of principle that glioblastoma-derived EV-DNA reflects the genome-wide methylation, CNVs, and mutational status of glioblastoma cells with high accuracy and enables their molecular classification.	[41]
Plasma and matched primary tumor tissues of 21 patients with aggressive diffuse large B cell lymphoma (DLBCL).	First study to show the presence of DNA in plasma exosomes of DLBCL patients.CDKN2A and CDKN2B were methylated in both plasma exosomes and primary tumor tissue samples.Compared to 21 healthy individuals, exosome concentration was approximately 6 times higher in DLBCL patients, but the exosomal dsDNA content was extremely low compared to RNA contents.	[42]
First direct comparison on gene expression and DNA methylation markers in CTCs and paired plasma-derived exosomes.	Remarkable heterogeneity on gene expression and DNA methylation markers between EpCAM-positive CTCs and paired plasma-derived exosomes in metastatic castration-resistant prostate cancer (mCRPC) patients with significantly higher positivity in CTCs.	[43]
Extensive review of the characteristics and clinical applications of extracellular vesicle-derived DNA.	The presence of DNA in excreted exosomes will continue to be controversial until the development of a method for isolating pure exosomes or microvesicles. Nonetheless, the size of dsDNA found in EVs (from ~100 bp to ~20 kbp) can represent the entire genome and reflects the mutational status of tumor parental cells. With DNA extracted from all categories of EVs, EV-DNA is the latest and most promising biomarker for identifying tumor presence and complexity.	[44]