Cancer Extracellular Vesicles and Metastasis: Progress at a Crossroads

Anil Kumar Kalvala^1*Venkatesh Pooladanda²Nagavendra Kommineni³

• ¹Texas Tech University Health Sciences Center, Abilene, Abilene, United States
• ²Massachusetts General Hospital Center for Genomic Medicine, Boston, United States
• ³Thermo Fisher Scientific Inc., Cincinati, Ohio,, United States

stem cell-derived EVs promote tissue repair [4], while cancer-derived EVs facilitate tumor growth and metastasis [5]. A landmark study by Lyden and colleagues (Nature Medicine, 2012) provided the first evidence that EVs derived from highly metastatic melanoma cells can permanently "educate" bone marrow progenitor cells, thereby enhancing the metastatic potential of primary tumors [6]. One of the most intriguing aspects of these vesicles is their ability to promote organotrophic metastasis, the phenomenon where EV cargo dictates the preferred organ destination of metastatic spread. Evidence suggests that this organotropism is largely governed by specific integrin subtypes carried on cancer-derived EVs. The Hoshino et al. (Nature, 2015) study elegantly demonstrated that EVs from tumor cells expressing α6β4 and α6β1 integrins are associated with lung metastasis, while those expressing αvβ5 integrins preferentially promote liver metastasis [7].Targeting these integrins significantly reduced EV uptake and the corresponding metastatic colonization. Mechanistically, α6β4 and α6β1 bind to laminin, activating c-SRC-S100A4 signaling and promoting lung-tropic metastasis, while αvβ5 binds to fibronectin on liver Kupffer cells, promoting liver-tropic metastasis. Additionally, αvβ3 and αvβ8 integrins in breast cancer EVs facilitate brain metastasis, whereas α4β1 and α5β1 integrins are implicated in bone metastasis [8,9]. In our recent studies, we found that ITGB1 expression in EVs derived from TSC2-null cells directed the formation of lung pre-metastatic niches in NCG mice. plasma-derived EVs from lymphangioleiomyomatosis (LAM) patients displayed elevated expression of ITGB1, c-SRC, FAK, and other integrin adhesion complex components, consistent with our in vitro findings [10].A central question remains: how do these nanosized vesicles so effectively reprogram recipient cells? Do they merely deliver molecular cargo, or do they carry pre-assembled protein complexes capable of instant signaling? Understanding EV topology, the spatial orientation of proteins within the vesicle membrane is crucial to answer this. EV integrins exist as heterodimers, and their cytoplasmic tails often bind to adaptor proteins such as talin and kindlin, which determine the activation state of the integrin receptor [11]. These integrin-adhesion complexes enable immediate communication with recipient cells, initiating rapid downstream signaling. Several studies have also demonstrated that cancer-derived EVs contain matrix metalloproteinases (MMP3, MMP7, MMP9, etc.), which enhance metastatic potential either by remodeling the extracellular matrix or by activating integrin signaling pathways [10,12,13]. Furthermore, EVs are enriched in tetraspanins (CD9, CD63, CD81)-classical EV markers whose expression correlates with EV biogenesis. Interestingly, tetraspanins themselves can contribute to the formation of focal adhesion points, amplifying the migratory and invasive potential of cancer cells [14,15].Two key determinants govern the oncogenic impact of EVs: how they are secreted and how they are internalized by recipient cells. Although the precise mechanisms of these processes are still being elucidated, several drugs are under investigation for their ability to inhibit EV biogenesis or uptake. Tipifarnib (a farnesyltransferase inhibitor) has been identified through high-throughput screening as a potent inhibitor of both ESCRT-dependent (Alix) and ESCRT-independent (nSMase2) EV biogenesis in metastatic castration-resistant prostate cancer cells [16]. GW4869, an inhibitor of neutral sphingomyelinase 2 (nSMase2), suppresses EV secretion and has been shown to inhibit proliferation and promote apoptosis in colorectal cancer models [17]. Cannabidiol (CBD) also reduces EV release and exhibits anticancer potential in various tumor types, though its specific mechanisms remain under investigation [18]. Dyngo-4A, a dynamin inhibitor, and its analog dynasore block EV uptake by disrupting clathrin-mediated endocytosis, thereby reducing tumor-derived EV internalization and angiogenesis [19,20]. While these findings highlight the promise of EV-targeted therapies, practical challenges remain mainly in difficulty of selectively targeting tumor-derived EVs without affecting physiological EVs. Regarding EV isolation, current methodologies include ultracentrifugation, ultrafiltration, size-exclusion chromatography, polymer-based precipitation, and affinity-based capture using antibodies have each with its advantages and limitations. According to the MISEV2023 guidelines (ISEV 2023), combining multiple methods is recommended to enhance purity, reproducibility, and yield [21]. Before functional assays, it is essential to verify cargo reproducibility and EV integrity across methods used to isolate EVs. Despite significant progress, key limitations persist, including the lack of standardized dose optimization and precise definitions of EV purity. The future of EV research depends on a deeper understanding of EV biogenesis, uptake mechanisms, and protein topology, as these factors determine vesicle function and specificity. In conclusion, effective targeting EVs in cancer requires a comprehensive understanding of their molecular cargo, topology, and signaling mechanisms. Therapeutic strategies that selectively modulate EV communication rather than indiscriminately blocking biogenesis or uptake are more likely to yield safe and successful outcomes. Looking forward, microfluidic chip technologies hold promises for rapid, noninvasive tumor detection from a single drop of patient blood, marking an exciting frontier in cancer diagnostics and precision medicine using EVs.