Pyroptosis is an important factor affecting the proliferation,
invasion, and metastasis of tumor cells. However, in multiple myeloma
(MM), there are few studies on whether the occurrence of pyroptosis
is related to the occurrence and prognosis of the disease. Based on
the Gene Expression Omnibus and Cancer Genome Atlas database search
dataset, this study identified pyroptosis-related genes with a specific
prognosis, constructed and verified the prediction model by stepwise
Cox regression analysis and time receiver operating characteristic
curve analysis, and predicted specific functions by single-sample
gene set enrichment analysis and the Kyoto Encyclopedia of Genes and
Genomes. Dataset analysis identified key genes, which were used to
construct a risk scoring system for the prognosis of MM. The entire
test set and external verification set verified the results. The expression
levels of related genes in the clinical samples were detected using
fluorescence quantitative PCR. A prognostic gene model based on six
pyroptosis-related genes (CYCS, NLRP9, AIM2, NOD2, CHMP3, and GSDME)
was constructed. The model has an excellent prognostic ability and
can be popularized in the external validation set. The predictive
prognostic nomogram integrating clinical information can effectively
evaluate the risk score of each patient and predict their survival.
After sample validation, our study found three potential key pyroptosis-related
genes in multiple myeloma. GSDME, NOD2, and CHMP3 were significantly
different between MM and healthy subjects, suggesting that they are
pyroptosis-related protective genes. This study shows that the key
pyroptosis-related gene in the model can be used as a marker for predicting
the prognosis of myeloma, which may provide a basis for clinical individualized
stratification therapy.