Clinical landscape of S100A9 expressionGiven its remarkable sensitivity, S100A9 holds significant promise as a critical target and biomarker for cancer diagnosis. To evaluate S100A9 expression in tumors and adjacent normal tissues, we conducted a comprehensive analysis of S100A9 mRNA expression levels. Our results revealed significantly elevated S100A9 mRNA levels in cancer samples from BLCA, BRCA, CESC, COAD, ESCA, GBM, HNSC, KICH, KIRC, KIRP, LIHC, LUSC, PAAD, PRAD, THCA, and UCEC. These findings suggest a potential oncogenic role for S100A9 in the progression of these diverse cancers (Fig. 2a). Notably, S100A9 expression was particularly elevated in HNSC, CESC, LUSC, and ESCA, as highlighted in Fig. 2b. S100A9 activity was markedly enhanced in tumor categories including BLCA, CESC, CHOL, COAD, ESCA, GBM, KIRC, LUAD, LUSC, PAAD, READ, STAD, THCA, and UCEC, while it exhibited a significant reduction in BRCA, KICH, PCPG, and PRAD, as depicted in Fig. 2c. Furthermore, significantly higher activity levels were observed in HNSC, CESC, LUSC, and ESCA, as shown in Fig. 2d. Figure 3a illustrates distinct expression patterns of S100A9 in older patients with GBM, while lower expression levels were observed in BRCA, ESCA, LAML, SARC, STAD, and THYM. Gender disparities in S100A9 expression were significant in BRCA, HNSC, SARC, and SKCM, as depicted in Fig. 3b. Additionally, associations between S100A9 expression and cancer grade were observed in several cancer types, including ESCA, HNSC, and LGG, as illustrated in Fig. 3c. Moreover, S100A9 expression was linked to tumor stage in multiple cancers, including HNSC, KICH, LIHC, LUAD, PAAD, TGCT, and THCA, as shown in Fig. 3d. These findings underscore the multifaceted role of S100A9 in cancer biology, emphasizing its potential utility as a diagnostic biomarker and therapeutic target.Figure 2S100A9 activity. (a) Differential analysis of Tumor and normal. (b) Mean expression. (c) Activity analysis. (d) Mean activity.Figure 3Clinical information. (a) Age. (b) Gender. (c) Grade. (d) Stage.Prognostic expression of S100A9 in cancersForest plots were constructed to evaluate the prognostic significance of S100A9 across various cancer types. The analysis revealed a favorable association between S100A9 expression and overall survival (OS) in LAML, LGG, LIHC, THYM, and UVM, as depicted in Fig. 4. Notably, a clear positive correlation between S100A9 expression and disease-free survival (DFS) was observed in LIHC and READ, whereas a negative correlation was evident in lung squamous cell carcinoma (LUSC). Regarding disease-specific survival (DSS), elevated S100A9 expression emerged as a risk factor in BLCA, COAD, KIRC, LGG, LIHC, and UVM. The forest plot for progression-free survival (PFS) further substantiated the risk associated with S100A9 expression in BLCA, COAD, KIRC, LGG, LIHC, and UVM. Moreover, the graphical representation facilitated the identification of additional malignancies where S100A9 expression was determined to be a concomitant risk factor, notably in THYM and UVM. While not directly interfacing with clinical attributes, S100A9 expression demonstrated a robust association with survival outcomes across diverse neoplastic entities, particularly in LAML, LGG, and LIHC. These findings underscore the critical role of S100A9 as a prognostic biomarker, highlighting its potential impact on the management and therapeutic stratification of various cancers.Figure 4Univariate Cox regression analyses.S100A9 expression and immune infiltration levels in cancerTo evaluate the correlation between S100A9 expression and immune infiltration levels across various malignancies, we calculated the coefficients of S100A9 expression and immune infiltration. Figure 5 provides an overview of the stromal and immunological scores. S100A9 expression demonstrated significant associations with stromal scores in DLBC, GBM, KICH, LAML, LGG, PCPG, SARC, TGCT, THCA, and UVM. Additionally, S100A9 expression showed correlations with immune scores in COAD, GBM, KICH, KIRP, LAML, LGG, LIHC, PAAD, PCPG, PRAD, READ, SARC, THCA, and UVM (Table 1). Moreover, the analysis of immune cell infiltration revealed significant associations between S100A9 expression and specific immune cell subsets. Notably, S100A9 expression was negatively associated with CD4 memory resting T cells in ACC, monocytes in LAML, M1 macrophages in DLBC, activated natural killer (NK) cells in DLBC and KICH, naive B cells in LAML and TGCT, and neutrophils in ACC, CHOL, COAD, GBM, KICH, KIRC, PCPG, READ, and STAD, as depicted in Fig. 6. These findings highlight the intricate relationship between S100A9 expression and the tumor microenvironment, suggesting its potential role in modulating immune infiltration and influencing cancer progression.Figure 5ESTIMATE analyses. (a) StromalScore. (b) ImmuneScore.Table 1 The stromal and immunological ratings.Figure 6Immune infiltration analyses.Analysis of S100A9 expression and immune modulatorsTo investigate the intricate relationship between S100A9 expression and immune modulators, a comprehensive analysis was conducted. As shown in Fig. 7, a meticulous examination was carried out on 24 distinct types of immune inhibitors. Notably, S100A9 exhibited significant associations with specific immune modulators in particular cancer types. Specifically, S100A9 demonstrated a correlation with IL10 in GBM, HAVCR2 in THCA, and LGALS9 in SARC, while no significant association was observed with CD160 in CHOL. Moving forward, a thorough exploration was undertaken to assess the relationship between S100A9 expression and 45 immune stimulators, as illustrated in Fig. 8. The findings revealed intriguing patterns of association. S100A9 expression displayed a positive correlation with IL6 in GBM, CD86 in THCA, and IL2RA in SARC. Conversely, a negative association was observed with TNFSF13 in ESCA, suggesting a potential regulatory role of S100A9 in modulating immune responses. Furthermore, it is noteworthy that S100A9 expression exhibited distinct associations with HLA genes. Figure 9 highlights these associations, where S100A9 showed a positive correlation with HLA-DPA1 in KICH, HLA-DPB1 in THCA, and HLA-DRA in THCA. Conversely, a negative association was identified with HLA-A in CHOL, indicating potential intricate mechanisms underlying the interplay between S100A9 expression and HLA genes. These findings shed light on the complex interrelationships between S100A9 expression and immune modulators, providing valuable insights into the underlying mechanisms of immune regulation in various cancer types.Figure 7Figure 8Figure 9Immunotherapeutic markers and responseThe present study aimed to investigate the correlation between S100A9 and two novel dynamic markers associated with immune checkpoint blockade, namely TMB and MSI. The analysis revealed that S100A9 expression exhibited a positive association with TMB in BRCA, CESC, KIRC, and LGG. However, this association was not observed in ESCA, LAML, PAAD, PCPG, and PRAD. Conversely, MSI demonstrated a positive relationship with S100A9 expression in BRCA, CESC, KIRC, and LGG, while exhibiting a negative relationship in ESCA, LAML, PAAD, PCPG, and PRAD. The immune checkpoint pathway plays a pivotal role in cancer therapy, particularly the programmed cell death protein 1 (PD-1) pathway. Previous literature has reported the utilization of three commonly employed microarray datasets, namely GSE67501 (n = 11), GSE78220 (n = 28), and IMvigor210, to investigate the expression of individual genes in relation to PD-1. Accordingly, we investigated the association between S100A9 and immune checkpoints, specifically focusing on these three datasets. However, the analysis revealed no statistically significant differences in S100A9 expression between responder and non-responder groups in any of the three cohorts, implying that S100A9 may not be influenced by the microarray sets commonly employed in these immune checkpoint studies. Notably, patients exhibiting lower S100A9 expression displayed greater susceptibility to immunotherapy in the analyzed cohorts (Fig. 10). It is important to acknowledge that although the three GEO datasets used in this study are widely recognized and frequently employed in classic GEO microarray studies, they do possess certain limitations such as small sample sizes, which may introduce some inherent biases into the results. Consequently, future investigations should endeavor to employ larger datasets to validate these findings.Figure 10Immunotherapeutic markers and response.The association between changes in fatty acid metabolism regulating genes (CNV, SNP, and mutation) and clinicopathological characteristics in patients was investigated. Correlation study of S100A9 expression in the prognostic model and SNP revealed 6 SNP-driven cancers, including BLCA,STAD, BRCA,COAD, CSEC, BRCA (Fig. 11a–f). A correlation analysis of S100A9 expression in the prognostic model and CNV found numerous cancers driven by CNV. The expression of 6 cancers were upregulated in the single mutations group versus that of the non-mutations group. (P < 0.05), indicating that dysregulation of key genes might be driven by SNP in S100A9.Figure 11SNP and mutation analysis. (a–f) Prognostic signatures and SNP. (a) BLCA, (b) STAD, (c) BRCA, (d) COAD, (e) CSEC, (f) BRCA, (g) CNV analysis.Comprehensive gene regulatory networks and GSEATo unravel the fundamental mechanisms underlying S100A9, we constructed a comprehensive Comprehensive Gene Regulatory Networks (Fig. 12). Within this network, S100A9 exhibited a strong interaction with S100A8, S100A12, CD33, CSTA, MMP9, S100A7, which has been implicated in cancer metastasis (Table 2). These proteins exhibit diverse roles in tumor progression, inflammation, and immune response modulation, with their expression levels often being dysregulated in various malignancies. S100A8 and S100A9, members of the S100 calcium-binding protein family, are known to form heterodimers and are markedly upregulated in several types of cancer. They function as damage-associated molecular patterns, promoting pro-inflammatory responses and enhancing the recruitment of myeloid-derived suppressor cells, which are known to facilitate tumor progression by suppressing anti-tumor immunity24. These proteins also contribute to the remodeling of the extracellular matrix, thus aiding in metastasis. S100A12, another member of the S100 family, shares functional similarities with S100A8 and S100A925. It is often co-expressed with these proteins and plays a role in inflammatory processes and cancer. Its interaction with the receptor for advanced glycation end products is particularly notable, as it activates key signaling pathways involved in tumor growth and metastasis. CD33, a transmembrane receptor expressed on myeloid cells, plays a crucial role in modulating immune responses. In the context of cancer, CD33 is expressed on MDSCs and contributes to their immunosuppressive functions, thereby facilitating tumor immune evasion26. Targeting CD33-positive MDSCs has emerged as a potential therapeutic strategy to enhance anti-tumor immunity. CSTA, a cysteine protease inhibitor, is implicated in cancer progression through its regulatory effects on proteolytic activity. CSTA can influence tumor invasion and metastasis by modulating the activity of cathepsins, a family of proteases involved in extracellular matrix degradation27. Its expression levels have been correlated with tumor aggressiveness in various cancers. Collectively, these investigations provide compelling support for the credibility and plausibility of our findings, suggesting that S100A9 could serve as a novel diagnostic and prognostic biomarker in human cancers.Figure 12Comprehensive Gene Regulatory Networks of S100A9.Table 2 Comprehensive Gene Regulatory Networks.Subsequently, we employed GSEA to identify functional enrichments related to S100A9 (Fig. 13). The analysis revealed that elevated levels of S100A9 were significantly associated with metabolic-related activities, including olfactory transduction, autophagy regulation, the rig I like receptor signaling pathway, systemic lupus erythematosus, and taste transduction, as indicated by the KEGG pathway database. Moreover, based on the GO annotations, high levels of S100A9 were primarily associated with epidermis development, sensory perception of chemical stimulus, and sensory perception of smell.Figure 13GSEA. (a + c) Low expression. (b + d) High expression sample. (a + b) KEGG. (c + d): GO.Mendelian randomization analysisIn our exploration of the intrinsic connection between BLCA, CESC, COAD, ESCA, GBM, BRCA, HNSC, and S100A9, forest plots were meticulously employed to visually articulate the associations. Further dissecting the heterogeneity inherent in our analysis, the funnel plot tailored to KIRC revealed a deviation from the expected symmetrical distribution, albeit maintaining a general symmetry. This nuanced observation was further scrutinized through sensitivity analysis, employing a “leave-one-out” approach. Remarkably, the omission of any individual SNP from the analysis had a negligible effect on the results of the Inverse Variance Weighted (IVW) analysis, indicating that the remaining SNPs consistently mirrored the outcomes of the aggregate dataset. Substantiating the validity of our findings, the MR-Egger regression analysis was invoked, providing a solid foundation that bolsters both the robustness and authenticity of our results and the methodologies applied. This Mendelian randomization analysis unequivocally confirms the intimate association of BLCA, CESC, COAD, ESCA, GBM, BRCA, HNSC with S100A9. Hence, it delineates a potential pathway to modulate the incidence, evolution, and progression of S100A9 by intervening in the functions of BLCA, CESC, COAD, ESCA, GBM, BRCA, HNSC, presenting a promising avenue for therapeutic intervention and a deeper understanding of the disease mechanism (Table 3, Fig. 14).Table 3 Mendelian Randomization Analysis.Figure 14Mendelian randomization analysis. (a) BLCA, (b) CESC, (c) COAD, (d) ESCA, (e) GBM, (f) BRCA, (g) HNSC.
