Nov 17, 2014

Use of Computer Tumor Tissue Imaging Algorithms to Assess the Role of Microenvironment in Melanoma Brain Metastases

Dr. Stergios Moschos, Clinical Associate Professor of Medicine The University of North Carolina at Chapel Hill, Chapel Hill, North Carolina, USA

Background: High intratumoral hemorrhage and low degree of immune infiltrates in craniotomy specimens from melanoma brain metastases (MBM) are associated with shorter overall survival (OS). We hypothesize that immune checkpoint proteins (ICP), angiogenic factors, hypoxia, and the density of mature vs. immature blood vessels (MBV, IBV) are important prognostic factors.
Conclusions: Sophisticated computer tissue imaging algorithms can differentiate expression of molecules_structures within different tumor compartments, which may play a different role in tumor progression. Tumor response to hypoxia by upregulating HIF1α may actually be a favorable prognostic factor. High expression of PD-L1 in MBM is an adverse prognostic factor, and could be derived from its immunosuppressive effects in the brain microenvironment.