Identification of neutrophil extracellular traps (NETs)-related molecular clusters in prostate cancer: Implications for predicting biochemical recurrence.

Objective: To identify neutrophil extracellular traps (NETs)-related molecular clusters and establish a novel gene signature for predicting biochemical recurrence in prostate cancer (PCa).

Methods: The transcriptome and clinicaldata of PCa sampleswere obtained from The TCGA and GEO databases. To identify NET-related molecular clusters, consensus clustering analyses were performed. Using univariate Cox and Lasso regression analysis, a novel NETs-related prognostic model was formulated. To evaluate the validity of the model, both internal and external validations were carried out. At last, preliminary experimental validations were performed to verify the biological functions of ANXA3 in PCa cells.

Results: After screening 75 NETs-related prognostic genes, two NET-related clusters with significantly different clinical features, immune cell infiltration, and biochemical recurrence were established. Next, a new NET-related model was constructed. In training, test, whole TCGA, and GEO cohorts, the biochemical recurrences free survival of the patients with high-risk scores was considerably lower. The AUCs for the four cohorts were 0.827, 0.696, 0.757, and 0.715, respectively. Subgroup analysis suggested that the novel NETs-related prognostic model has a strong clinical value in the identification of high-risk patients. Finally, we confirmed that chemotherapy might be more beneficial for patients at low risk. In preliminary experiments, the inhibition of ANXA3 could reduce the invasion, migration, and proliferation of PCa cells.

Conclusions: We have identified novel NETs-related clusters and developed a NETs-related model for PCa that has excellent predictive performance for predicting biochemical recurrences as well as chemotherapy efficacy.