Construction and evaluation of a novel prognostic risk model of aging-related genes in bladder cancer

Abstract

Abstract. Background. Bladder cancer (BLCA) is the most common malignancy of the urinary system. Muscle-invasive bladder cancer (MIBC), which constitutes approximately 25% of all BLCA cases, is characterized by frequent recurrence and early onset of metastasis. Bladder cancer most commonly occurs in elderly patients and is significantly associated with aging. However, the prognostic value of age-related genes in BLCA, especially in MIBC, remains unclear. Materials and methods. Training and testing sets were obtained from The Cancer Genome Atlas BLCA project. Differentially expressed genes between BLCA and normal samples intersected with human aging-related genes. Univariate Cox regression and least absolute shrinkage and selection operator regression analyses were used to identify prognostic aging-related signatures, followed by the construction of a risk score model and nomogram. Kaplan-Meier and receiver operating characteristic analyses were conducted to assess the predictive power. An independent BLCA cohort of 165 samples was included for external validation. The CIBERSORT algorithm was used to explore the characteristics of the immune microenvironment. Results. Seven genes (IGF1, NGF, GCLM, PYCR1, EFEMP1, APOC3, and IFNB1) were identified by Cox and least absolute shrinkage and selection operator analyses. After combining the gene signature with the clinical parameters of patients with BLCA, a risk-prognosis model and nomogram were constructed and validated with the testing set. Bladder cancer cases with high 7-gene signature scores (high-risk group) and low scores (low-risk group) showed distinct prognoses. Furthermore, 7 types of immune cells were significantly altered between the low- and high-risk groups. Conclusions. Collectively, our data provide a 7-gene signature that serves as a potential biomarker for BLCA, especially MIBC. Moreover, this 7-gene signature highlights the role of the tumor immune microenvironment in prognosis and thus might be related to the response to anti-programmed cell death protein 1–based immunotherapy.