CASP1 and SPTAN1 Identified as Potential PANoptosis-Related Biomarkers for Differentiating Active Tuberculosis from Latent Tuberculosis Infection Through Integrative Bioinformatics Analysis and Exploratory In Vitro Assessment

Distinguishing active tuberculosis (ATB) from latent tuberculosis infection (LTBI) remains a critical unmet need in tuberculosis control. PANoptosis, a newly recognized form of programmed cell death (PCD) integrating pyroptosis, apoptosis, and necroptosis, may provide novel biomarkers for this distinction. To explore this, transcriptomic datasets from Gene Expression Omnibus (GEO) were employed to identify PANoptosis-related differentially expressed genes (TPAN_DEGs) between ATB and LTBI. An ensemble machine-learning approach combining least absolute shrinkage and selection operator (LASSO), Support Vector Machine-Recursive Feature Elimination (SVM-RFE), and random forest (RF) was used to identify hub genes and construct a diagnostic model to differentiate between LTBI and ATB. The model was externally evaluated in independent transcriptomic datasets, with additional ssGSEA-based immune-related signature and regulatory network analyses. Peripheral blood samples were collected from clinical LTBI and ATB patients, and Western blotting was performed to assess the protein-level expression patterns of the hub genes. Consequently, a total of 26 TPAN_DEGs were identified, among which CASP1 and SPTAN1 were identified as hub genes by machine-learning algorithms. The diagnostic model showed good discriminatory performance in both the training and validation datasets. Moreover, Western blot analysis of peripheral blood samples from four ATB patients and four LTBI individuals showed higher levels of cleaved-CASP1 p20 and the cleaved-CASP1 p20/pro-CASP1 ratio, together with lower levels of full-length SPTAN1 and SPTAN1 (cleaved-Asp1185) in the ATB group. Correlation analysis showed that CASP1 expression was predominantly positively correlated, whereas SPTAN1 expression was predominantly negatively correlated, with ssGSEA-derived immune signature enrichment scores. YY1 and hsa-miR-34a-5p were predicted as potential shared regulators of both genes. Taken together, we identified a candidate two-gene expression signature through a PANoptosis-informed analytical approach, showing consistent transcriptomic alterations and preliminary protein-level differences between ATB and LTBI. The opposing expression and immune correlates of CASP1 and SPTAN1 suggest possible biological association with inflammatory activation and cellular structural alterations in ATB. These findings suggest that CASP1 and SPTAN1 may serve as potential biomarkers for distinguishing ATB from LTBI and provide a basis for further studies on PANoptosis-informed immune changes in tuberculosis.