ABSTRACT
The evolutionary conserved DNA-sensing cGAS-STING innate immunity pathway represents one of the most important cytosolic DNA-sensing systems that is activated in response to viral invasion and/or damage to the integrity of the nuclear envelope. The key outcome of this pathway is the production of interferon, which subsequently stimulates the transcription of hundreds of genes. In oncology, the situation is complex because this pathway may serve either anti- or pro-oncogenic roles, depending on context. The prevailing understanding is that when the innate immune response is activated by sensing cytosolic DNA, such as DNA released from ruptured micronuclei, it results in the production of interferon, which attracts cytotoxic cells to destroy tumors. However, in tumor cells that have adjusted to significant chromosomal instability, particularly in relapsed, treatment-resistant cancers, the cGAS-STING pathway often supports cancer progression, fostering the epithelial-to-mesenchymal transition (EMT). Here, we review this intricate pathway in terms of its association with cancer progression, giving special attention to pancreatic ductal adenocarcinoma and gliomas. As the development of new cGAS-STING-modulating small molecules and immunotherapies such as oncolytic viruses involves serious challenges, we highlight several recent fundamental discoveries, such as the proton-channeling function of STING. These discoveries may serve as guiding lights for potential pharmacological advancements.
ABSTRACT
Metformin is an oral hypoglycemic agent widely used in clinical practice for treatment of patients with type 2 diabetes mellitus (T2DM). The wide interindividual variability of response to metformin therapy was shown, and recently the impact of several genetic variants was reported. To assess the independent and combined effect of the genetic polymorphism on glycemic response to metformin, we performed an association analysis of the variants in ATM, SLC22A1, SLC47A1, and SLC2A2 genes with metformin response in 299 patients with T2DM. Likewise, the distribution of allele and genotype frequencies of the studied gene variants was analyzed in an extended group of patients with T2DM (n = 464) and a population group (n = 129). According to our results, one variant, rs12208357 in the SLC22A1 gene, had a significant impact on response to metformin in T2DM patients. Carriers of TT genotype and T allele had a lower response to metformin compared to carriers of CC/CT genotypes and C allele (p-value = 0.0246, p-value = 0.0059, respectively). To identify the parameters that had the greatest importance for the prediction of the therapy response to metformin, we next built a set of machine learning models, based on the various combinations of genetic and phenotypic characteristics. The model based on a set of four parameters, including gender, rs12208357 genotype, familial T2DM background, and waist-hip ratio (WHR) showed the highest prediction accuracy for the response to metformin therapy in patients with T2DM (AUC = 0.62 in cross-validation). Further pharmacogenetic studies may aid in the discovery of the fundamental mechanisms of type 2 diabetes, the identification of new drug targets, and finally, it could advance the development of personalized treatment.
