FABP6 Expression Correlates with Immune Infiltration and Immunogenicity in Colorectal Cancer Cells.

Background: Immune checkpoint inhibitors (ICIs) have rapidly revolutionized colorectal cancer (CRC) treatment, but resistance caused by the heterogeneous tumor microenvironment (TME) still presents a challenge. Therefore, it is necessary to characterize TME immune infiltration and explore new targets to improve immunotherapy. Methods: The compositions of 64 types of infiltrating immune cells and their relationships with CRC patient clinical characteristics were assessed. Differentially expressed genes (DEGs) between "hot" and "cold" tumors were used for functional analysis. A prediction model was constructed to explore the survival of CRC patients treated with and without immunotherapy. Finally, fatty acid-binding protein (FABP6) was selected for in vitro experiments, which revealed its roles in the proliferation, apoptosis, migration, and immunogenicity of CRC tissues and cell lines. Results: The infiltration levels of several immune cells were associated with CRC tumor stage and prognosis. Different cell types showed the synergistic or antagonism infiltration patterns. Enrichment analysis of DEGs revealed that immune-related signaling was significantly activated in hot tumors, while metabolic process pathways were altered in cold tumors. In addition, the constructed model effectively predicted the survival of CRC patients treated with and without immunotherapy. FABP6 knockdown did not significantly alter tumor cell proliferation, apoptosis, and migration. FABP6 was negatively correlated with immune infiltration, and knockdown of FABP6 increased major histocompatibility complex (MHC) class 1 expression and promoted immune-related chemokine secretion, indicating the immunogenicity enhancement of tumor cells. Finally, knockdown of FABP6 could promote the recruitment of CD8+ T cells. Conclusion: Collectively, we described the landscape of immune infiltration in CRC and identified FABP6 as a potential immunotherapeutic target for treatment.

Comprehensive Characterization of the Function of Metabolic Genes and Establishment of a Prediction Model in Breast Cancer.

Background: Breast cancer (BC) is a highly heterogeneous disease with high morbidity and mortality. Its subtypes may have distinctly different biological behaviors, clinical outcomes, and therapeutic responses. The metabolic status of BC tissue is closely related to its progress. Therefore, we comprehensively characterized the function of metabolic genes in BC and identified new biomarkers to predict BC patients' prognoses. Methods: Metabolic genes were identified by intersecting genes obtained from two published pieces of literature. The function of metabolic genes in BC was determined by extracting differentially expressed genes (DEGs), performing functional enrichment analyses, analyzing the infiltrating proportion of immune cells, and conducting metabolic subgroup analyses. A risk score model was constructed to assess the prognoses of BC patients by performing the univariate Cox regression, LASSO algorithm, multivariate Cox regression, Kaplan-Meier survival analyses, and ROC curve analyses in the training set. The prognostic model was then validated on the testing dataset, external dataset, the whole TCGA-BC database, and our clinical specimens. Finally, a nomogram was constructed for clinical prognostic prediction based on the risk score model and other clinicopathological parameters. Results: 955 metabolic genes were obtained. Among these, 157 metabolic DEGs were identified between BC and normal tissues for subsequent GO and KEGG pathway enrichment analyses. 5 metabolic genes were negatively correlated with CD8+ T cells, while 49 genes were positively correlated with CD8+ T cells. Furthermore, 5 metabolic subgroups with varying proportions of PAM50 subtypes, TNM classification, and immune cell infiltration were obtained. Finally, a risk score model was constructed to predict the prognoses of BC patients, and a nomogram incorporating the risk score model was established for clinical application. Conclusion: In this study, we elucidated tumor heterogeneity from metabolite profiling of BC. The roles of metabolic genes in the occurrence of BC were comprehensively characterized, clarifying the relationship between the tumor microenvironment (TME) and metabolic genes. Meanwhile, a concise prediction model was also constructed based on metabolic genes, providing a convenient and precise method for the individualized diagnosis and treatment of BC patients.