Abstract
Endometrial cancer, a leading gynecological malignancy, is profoundly influenced by the uterine microbiota, a key factor in disease prognosis and treatment. Our study underscores the distinct microbial compositions in endometrial cancer compared to adjacent non-cancerous tissues, revealing a dominant presence of p_Actinobacteria in cancerous tissues as opposed to p_Firmicutes in surrounding areas. Through comprehensive analysis, we identified 485 unique microorganisms in cancer tissues, 26 of which correlate with patient prognosis. Employing univariate Cox regression and LASSO regression analyses, we devised a microbial risk scoring model, effectively stratifying patients into high and low-risk categories, thereby providing predictive insights into their overall survival. We further developed a nomogram that incorporates the microbial risk score along with age, grade, and clinical stage, significantly enhancing the accuracy of our clinical prediction model for endometrial cancer. Moreover, our study delves into the differential immune landscapes of high-risk and low-risk patients. The low-risk group displayed a higher prevalence of activated B cells and increased T cell co-stimulation, indicative of a robust immune response. Conversely, high-risk patients showed elevated tumor immune dysfunction and exclusion scores, suggesting less favorable outcomes in immunotherapy. Notably, the efficacy of IPS-CTLA4 and PD1/PD-L1/PD-L2 blockers was substantially higher in the low-risk group, pointing to a more responsive immunotherapeutic approach. In summary, our research elucidates the unique microbial patterns in endometrial cancer and adjacent tissues, and establishes both a microbial risk score model and a clinical prediction nomogram. These findings highlight the potential of uterine microbiota as a biomarker for customizing treatment strategies, enabling precise interventions for high-risk patients while preventing overtreatment in low-risk cases. This study emphasizes the microbiota's role in tailoring immunotherapy, offering a novel perspective in the treatment and prognosis of endometrial cancer. Significantly, our study's expansive sample analysis from the TCGA-UCEC cohort, employing linear discriminant analysis effect size methodology, not only validates but also enhances our understanding of the microbiota's role in endometrial cancer, paving the way for novel diagnostic and therapeutic approaches in its management.