Although TAMs are often M2-like, there is now evidence that such TAMs can be re-polarized towards an M1 phenotype that can inhibit tumor growth. Herein, we show that the STING agonists DMXAA and 2939-cGAMP are both able to repolarize M2-polarized marrow-derived macrophages in vitro, and we also provide evidence that DMXAA is able to shift M2-like macrophages towards an M1-like phenotype in vivo. The latter results were in agreement with a recent study reporting an M2-like to M1 shift in TAM populations in response to DMXAA treatment that was also were accompanied by an antitumor effect. Intra-tumoral TAM repolarization would provide a source of pro-inflammatory cytokines and chemokines, and reduce the levels of vascular endothelial growth factor, effects that could contribute towards either vascular disruption, or stabilization, respectively. TAM re-education with STING agonists may play a role in other important processes, including promotion of anti-tumor adaptive immune responses that are dependent on type I interferons and dendritic cell activation. The sensitivity of tumor vasculature to DMXAA is thought to be due to the immature and irregular vascular patterning within tumors. To produce subcutaneous tumors, large numbers of cancer cells are implanted and these divide rapidly, rendering them sensitive to chemotherapeutic agents. Growth of cells introduced in this manner may outstrip the angiogenic capacity of the host, leading to the development of large regions of ischemia and necrosis that can promote macrophage infiltration and activation. In both the 344SQ-ELuc metastases and the autochthonic NSCLC tumors there were no areas of necrosis, in contrast, subcutaneous 344SQ-ELuc tumors contained large necrotic regions and were densely populated with macrophages. It is possible that the ischemia and necrosis that typifies subcutaneous tumors renders their vessels more susceptible to DMXAA. In addition, it is plausible that compared to vessels derived from other vascular beds, dermal vasculature-derived tumor vessels are inherently unstable, and hence more vulnerable to DMXAA. Co-option of mature vessels may be a feature of both systemic metastases and primary lung adenocarcinomas, and such vessels, like other normal vascular beds in the animal, would be predicted to be refractory to the vascular disrupting effects of DMXAA. Our finding of structural differences in the vasculature between tumors at different anatomical sites lends some support the latter notion, an idea that was reinforced by the apparent inability of clodronate liposomes to cause macrophage deletion in 344SQ-ELuc metastases. Indeed, the latter observation provided a functional indication of structural differences between subcutaneous tumors and metastases. In summary, and as depicted in Figure S7, the features of the subcutaneous 344SQ-ELuc tumor vasculature may render them susceptible vascular disrupting effects of DMXAA.