Telomere maintenance-related genes are important for survival prediction and subtype identification in bladder cancer

Background: Bladder cancer ranks among the top three in the urology field for both morbidity and mortality. Telomere maintenance-related genes are closely related to the development and progression of bladder cancer, and approximately 60%–80% of mutated telomere maintenance genes can usually be found in patients with bladder cancer. Methods: Telomere maintenance-related gene expression profiles were obtained through limma R packages. Of the 359 differential genes screened, 17 prognostically relevant ones were obtained by univariate independent prognostic analysis, and then analysed by LASSO regression. The best result was selected to output the model formula, and 11 model-related genes were obtained. The TCGA cohort was used as the internal group and the GEO dataset as the external group, to externally validate the model. Then, the HPA database was used to query the immunohistochemistry of the 11 model genes. Integrating model scoring with clinical information, we drew a nomogram. Concomitantly, we conducted an in-depth analysis of the immune profile and drug sensitivity of the bladder cancer. Referring to the matrix heatmap, delta area plot, consistency cumulative distribution function plot, and tracking plot, we further divided the sample into two subtypes and delved into both. Results: Using bioinformatics, we obtained a prognostic model of telomere maintenance-related genes. Through verification with the internal and the external groups, we believe that the model can steadily predict the survival of patients with bladder cancer. Through the HPA database, we found that three genes, namely ABCC9, AHNAK, and DIP2C, had low expression in patients with tumours, and eight other genes—PLOD1, SLC3A2, RUNX2, RAD9A, CHMP4C, DARS2, CLIC3, and POU5F1—were highly expressed in patients with tumours. The model had accurate predictive power for populations with different clinicopathological features. Through the nomogram, we could easily assess the survival rate of patients. Clinicians can formulate targeted diagnosis and treatment plans for patients based on the prediction results of patient survival, immunoassays, and drug susceptibility analysis. Different subtypes help to further subdivide patients for better treatment purposes. Conclusion: According to the results obtained by the nomogram in this study, combined with the results of patient immune-analysis and drug susceptibility analysis, clinicians can formulate diagnosis and personalized treatment plans for patients. Different subtypes can be used to further subdivide the patient for a more precise treatment plan.