Introduction: The close interaction between cancer and immune cells in the tumor microenvironment (TME) affects the glucose metabolic profiles of tumors. As stomach cancer shows highly variable glucose metabolic patterns according to FDG PET, we comprehensively analyzed the association of glucose metabolic heterogeneity and TME profiles by integrating FDG PET and transcriptome data.
Methods: Twenty-one FDG PET data paired with RNA-seq data acquired from surgically removed human stomach cancer tissues were analyzed. The metabolic heterogeneity profiles were calculated by semi-automatically segmented primary tumor lesions on the baseline PET image. Texture features as well as Standardized uptake values (SUV) and tumor volumes were calculated. The cell-type enrichment scores including myeloid, lymphoid, and stromal cells were estimated based on cell-type specific marker genes and single-sample gene set enrichment analysis (ssGSEA).
Results: According to the immune cell profiles of TME, stomach cancer was divided into three different clusters, immune-desert, macrophage-enriched, and Tregs-enriched types. These clusters showed different immune profiles including cytolytic activity and PD-L1 expression. The metabolic profile estimated by FDG PET showed a trend of association with the Th2-Th1 response and the enrichment of M2 macrophages. In particular, texture features reflecting heterogeneity were positively associated with Th2 cells and negatively associated with M2 macrophages. Discussion: The variable glucose metabolic patterns of stomach cancer are associated with the heterogeneity of immune profiles of TME. The metabolic heterogeneous properties of FDG PET in stomach cancer according to the immune profiles of TME could contribute to stratification for personalized management including immunotherapy.
Oh, D., Choi, H., Bae, S. W., Chae, J., Paeng, J. C., Cheon, G. J., ... & Yang, H. K. (2020).