Brain metastases, inflammation, and imaging
It is worth noting that CSCs constitute the main cell population mediating metastasis [57]. Brain metastases are more common than tumors originating primarily in the brain and confer grave prognosis to patients with various types of primary cancers originating in other sites, such as lung, breast, colorectal, and melanoma, given that available treatments show very limited efficacy. The tumor microenvironment is crucial to determine the establishment of brain metastases, and within this context, neuroinflammation plays a central role [58].
Brain metastases can produce a tissue lesion that induces a response comprising persistent astrocyte and microglia activation with cytokine and chemokine release, increased blood vessel permeability and recruitment of immune cells, resulting in neuroinflammation [58-61]. Also, non-neoplastic inflamed sites in the brain may facilitate the adhesion of circulating tumor cells from peripheral tumors to the activated endothelium in brain vessels, which is one of the possible mechanisms promoting metastasis initiation [62]. Inflammation at distant sites promotes adhesion of CTCs to the activated endothelium and then initiates the formation of metastases. Many different types of mediators and immune cells are involved in brain metastasis, depending among other factors on the type of primary tumor of origin, metastatic site in the brain, and differential astrocyte and microglial activation, resulting in high biological heterogeneity [58, 63-67].
In terms of imaging, differentiating brain tumors from brain metastases pose a challenge by itself, as both present imaging patterns with similar peritumoral hyperintensities and intratumoral texture on MRI. Improvements have been obtained with the use of relative cerebral blood flow and volume analyses, diffusion tensor imaging, neurite orientation dispersion, density imaging to examine intracellular volume fraction, isotropic volume fraction, and extracellular volume fraction, and metabolite analysis with MR spectroscopy. Also, intratumoral creatine suggests GBM, whereas its absence indicates metastasis when single-voxel proton MR spectroscopy is used for imaging [68, 69]. A study examining the use of the mean apparent diffusion coefficient and absolute standard deviation derived from apparent diffusion coefficient measurements based on cellularity levels could not find marked differences between GBM tumors from brain metastasis [70]. However, another study found that the apparent diffusion coefficient could differentiate GBM from metastasis, and that homogeneity and the inverse difference moment of GBM were significantly higher than those of metastases in the regions of interest placements examined [71, 72].
Assessing the heterogeneity of both the tumor masses and peritumoral edema with MRI texture analysis revealed that the heterogeneity of the GBMs peritumoral edema was significantly higher than the edema surrounding MET, differentiating them with a sensitivity of 80% and specificity of 90% [73]. Combining arterial spin labeling perfusion (ASL)- and diffusion tensor imaging (DTI)-derived metrics showed to be promising in differentiating GBM and solitary brain metastases [74]. The use of 2D texture features extracted from images obtained with MRI may be a useful alternative for discriminating between GBM and brain metastases [75]. Computational-aided quantitative analysis of MRI images may improve the accuracy in differentiating GBM from metastases, and texture features are more significant than fractal-based features for that purpose [76]. Increasingly, machine learning algorithms have been applied to imaging data to improve the differentiation between GBM and brain metastases [77-81]. Novel diagnostic support systems based on radiomic features extracted from post-contrast 3DT1 MR images may help improving the distinction between solitary brain metastases and GBM with high diagnosis performance and generalizability [77]. Machine learning and deep learning-based models applied to conventional MR images may support preoperative discrimination between GBM and solitary brain metastasis conventional MR images [78-80], and deep learning network models that allow automated, on-site analysis of resected tumor specimens based on confocal laser endoscopic techniques image datasets have been developed [81]. Other parameters such as the cerebral blood volume gradient in the peritumoral brain zone may enable the differentiation of GBMs from metastases [82]. Another approach is to evaluate peritumoral areas with color map of phase difference enhanced imaging (Color PADRE) [83].
As with tumors originating in the brain, metastases are treated with radiotherapy and stereotactic radiosurgery, making incidence of radiation necrosis an important issue, and the distinction between metastasis and inflamed and necrotic tissue by MRI can be challenging. One study examined the hypothesis that methionine levels could be increased in metastatic tissue, whereas the inflammation marker translocator protein (PBR-TSPO), which can be quantified with specific PET tracers, would be increased in necrosis. Thus, the use of the [11C]methionine and [11C]PBR28 tracers in PET was evaluated in 5 patients previously treated brain metastases showing regrowth. The use of [11C]methionine could accurately identify pathologically confirmed tumor regrowth in all 7 lesions examined, whereas [11C]PBR28 could only identify 3 of 7 lesions, indicating that the former, but not the later tracer can be used as a reliable marker [84].