Extracellular vesicles (EVs) that are secreted by cancer cells have increasingly emerged as important mediators that can reprogram various types of stromal cells, such as endothelial cells, fibroblasts and immune cells, to become permissive for tumor growth and metastasis. EVs encapsulate a variety of biomolecular cargo and have been largely thought to mediate intercellular communication though uptake and transfer of their luminal cargo into recipient cells. However, could cancer cell-derived EVs mediate stromal reprogramming via mechanisms that are independent of uptake? If so, what are the biological and clinical implications? In a study recently published in Communications Biology [1], we sought to address these fundamental questions.  

In our study, we found that small EVs (sEVs) that are secreted by several types of cancer cells (ovarian, colorectal, renal) can stimulate endothelial cells to migrate and form capillary-like tubes in the absence of uptake. These responses were found to be mediated by the 189 amino acid isoform of vascular endothelial growth factor (VEGF) that is present on the surface of sEVs and interacts with the extracellular domain of its cognate receptor on endothelial cells. We identified that VEGF₁₈₉, unlike two other common isoforms of VEGF (VEGF₁₂₁ and VEGF₁₆₅), preferentially localizes to sEVs through its high affinity for heparin (Figure 1).  Furthermore, we identified that interaction of VEGF₁₈₉ with the surface of sEVs profoundly increases ligand half-life. Whereas it has been thought that VEGF₁₈₉ acts locally, our findings that sEV-associated VEGF predominantly comprises VEGF₁₈₉, is signaling-competent, highly stable and present in the circulation of tumor-bearing mice and cancer patients implicate that VEGF₁₈₉, through being conveyed on secreted sEVs, might also mediate long-range signaling.

Another significant outcome of our findings is the impact on responsiveness of tumors to bevacizumab. Bevacizumab is a humanized monoclonal antibody to VEGF and has been used to treat a variety of solid tumors since 2004. However, clinical responses to bevacizumab greatly vary and often do not correlate with baseline levels of total VEGF. Although bevacizumab is thought to neutralize all isoforms of VEGF, prior studies have largely focused on characterizing its binding to soluble VEGF. In our study, we identified that heparin‐bound, sEV-associated VEGF is not neutralized by bevacizumab in vitro and in xenograft models (Figure 2). Furthermore, in a cohort of patients with newly diagnosed renal cell carcinoma who were treated with bevacizumab, we found that baseline levels of sEV-associated VEGF were higher in patients with progressing disease than in those with stable or regressing disease. Although validation in large independent cohorts is needed, our findings raise the possibilities that resistance to bevacizumab might stem in part from elevated levels of sEV-associated VEGF and that baseline levels of sEV-associated VEGF might be informative for assessing bevacizumab treatment benefit. 

[1] Ko SY, Lee W, Kenny HA, Dang LH, Ellis LM, Jonasch E, Lengyel E, Naora H. Cancer-derived small extracellular vesicles promote angiogenesis by heparin-bound, bevacizumab-insensitive VEGF, independent of vesicle uptake. Commun Biol. 2019 Oct 18;2:386. doi: 10.1038/s42003-019-0609-x.

Song Yi Ko & Honami Naora