Identification and validation of a pyroptosis-related prognostic signature for thyroid cancer

BACKGROUND: Pyroptosis is a form of programmed cell death triggered by inflammasomes. However, the roles of pyroptosis-related genes in thyroid cancer (THCA) remain still unclear. OBJECTIVE: This study aimed to construct a pyroptosis-related signature that could effectively predict THCA prognosis and survival. METHODS: A LASSO Cox regression analysis was performed to build a prognostic model based on the expression profile of each pyroptosis-related gene. The predictive value of the prognostic model was validated in the internal cohort. RESULTS: A pyroptosis-related signature consisting of four genes was constructed to predict THCA prognosis and all patients were classified into high- and low-risk groups. Patients with a high-risk score had a poorer overall survival (OS) than those in the low-risk group. The area under the curve (AUC) of the receiver operator characteristic (ROC) curves assessed and verified the predictive performance of this signature. Multivariate analysis showed the risk score was an independent prognostic factor. Tumor immune cell infiltration and immune status were significantly higher in low-risk groups, which indicated a better response to immune checkpoint inhibitors (ICIs). Of the four pyroptosis-related genes in the prognostic signature, qRT-PCR detected three of them with significantly differential expression in THCA tissues. CONCLUSION: In summary, our pyroptosis-related risk signature may have an effective predictive and prognostic capability in THCA. Our results provide a potential foundation for future studies of the relationship between pyroptosis and the immunotherapy response. SUPPLEMENTARY INFORMATION: The online version contains supplementary material available at 10.1186/s12935-021-02231-0.