This study provides a combined 7-gene signature that could be a biomarker for predicting bladder cancer (BC) prognosis. The nomogram constructed by risk score and other clinical factors could serve as an effective tool for predicting the prognosis of patients with BC.

• Using gene expression profiles of BC samples and the corresponding clinicopathological data from Gene expression omnibus and The cancer genome atlas, researchers intended to identify and assess a prognostic gene signature in BC patients.

• Experts established a nomogram with the identified prognostic factors to predict the overall survival rates of BC patients.

• Via univariate and least absolute shrinkage and selection operator Cox regression analyses, a 7-gene signature, including killer cell lectin-like receptor subfamily B (KLRB1), placenta-specific 9 (PLAC9), SET binding protein 1 (SETBP1), nuclear receptor subfamily 2 group F member 1 (NR2F1), grainyhead-like 2 (GRHL2), annexin A1 (ANXA1) and apolipoprotein L1 (APOL1), was discovered from 285 differentially expressed genes.

• In univariate and multivariate Cox regression analyses, age, lymphovascular invasion, lymphatic metastasis, metastasis and the 7-gene signature risk score was shown to be an independent predictor of BC patient prognosis.

• Experts built a nomogram that integrated these independent prognostic factors.

• The nomogram’s good performance was evident based on Concordance index (0.73, CI 95%, 0.693–0.767) and calibration curve.

• In decision curve analysis of the nomogram, this model showed good net benefit.