Research ArticleExperimental Studies
Open Access

Mast Cell-related Prognosis Signature Characterizes Immune Landscape and Predicts Prognosis of Ovarian Cancer

QINGHUA LI, TIANYUN GUAN, JIANXIONG MAO, BENJAMIN P. KING, KESIYA JOHNSON, TRENTON G. MAYBERRY, BRAYDON C. COWAN, MARK R. WAKEFIELD, YUJIANG FANG and QINGYU ZHAO
Anticancer Research July 2024, 44 (7) 2775-2786; DOI: https://doi.org/10.21873/anticanres.17089

Abstract

Background/Aim: Ovarian cancer (OVC) is a common, aggressive, and heterogeneous malignancy, with a widely variable prognosis. With the advances of modern immunology, mast cells (MCs) have been shown to play a significant role in the prognosis of some malignant tumors. However, the role of mast cells in the prognosis of OVC is unknown. Materials and Methods: In this study, MC-associated prognostic genes (MRGs) were used to classify OVC from The Cancer Genome Atlas (TCGA)-OVC cohort. Genes were evaluated using univariate cox regression analysis. Twenty-nine prognostic gene signatures were identified using LASSO-COX analysis. COX regression models and principal component analysis (PCA) algorithms were used to construct MRG scores and individual MRGs patterns. External validation was performed in the TCGA-breast cancer (BRCA) and IMvigor210 cohorts. Immunity analysis based on MRGs was performed using CIBERSORT, and GSVA methods, and immunotherapy response was evaluated using the TIDE website. Results: Using TCGA-OVC data, we established a model for constructing MRG scores based on the twenty-nine identified prognostic gene signatures using the PCA algorithm. MRG scores were found to be strongly correlated with immune cell infiltration and were excellent predictors of prognosis in patients with OVC. Low MRG scores were associated with better prognosis and better response to immunotherapy and chemotherapy. Conclusion: MC-related prognosis signature characterizes the immune landscape and predicts the prognosis of OVC. Understanding the correlation between MC-related gene signatures and immunotherapy and chemotherapy may improve the development of personalized clinical treatment strategies.

Key Words:

Epithelial ovarian cancer, the most common type of ovarian cancer (OVC), has the highest mortality rate among gynecological cancers worldwide, with a 5-year survival rate ranging from 30% to 40%. Given its insidious onset, detection often transpires after the tumor has substantially advanced, necessitating rigorous clinical intervention thereafter (1). Survival rate for these patients has not significantly improved over the past few decades. Most patients are treated with surgery and chemotherapy, but micro metastases remain and become resistant to treatment, eventually leading to recurrence (2). The use of immune checkpoint inhibitors (ICIs) has been promising in improving survival in OVC patients. However, some clinical trials have shown that single dose ICIs is effective, and combination therapy has not achieved significant efficacy for most OVC patients (3). Early mast cell (MC) infiltration occurs in many tumors, particularly in melanoma, breast, gastric and colorectal cancers. On one hand, mast cells and their mediators are involved in promoting tumor growth. However, on the other hand, they may also play an anti-tumor role (4). Regarding tumor growth promotion, it has been reported that MCs promote angiogenesis and induce neovascularization through classical angiogenic factors including VEGF, FGF-2, PDGF and IL-6, and non-classical angiogenic factors, including proteases and chymases. MCs promote angiogenesis and induce neovascularization through the release of a broad spectrum of matrix metal proteases (MMPs) to support tumor invasiveness (5-13). When MCs infiltrates into the tumor stroma, the expansion and activation of Tregs can promote immune tolerance, leading to tumor progression (6). Regarding tumor growth inhibition, depending on the type of tumor, MCs can exert immunosuppressive effects by releasing IL-10, histamine and TNF-α, and adenosine (7). After activation and degranulation in the tumor microenvironment (TME), MCs actively recruit neutrophils, macrophages, eosinophils, B lymphocytes, and T cells to exert an anti-tumor immune response (8). MCs can influence the efficacy of immunotherapy, radiotherapy, and chemotherapy. MCs are involved in resistance to anti-PD-1 immunotherapy in melanoma patients (9). This is associated with lower expression of HLA class I, leading to tumor escape from cytotoxic T cells (10). MCs can enhance the resistance of tumors to chemotherapy and radiotherapy. For example, MCs promote docetaxel resistance in prostate cancer (11). Activation of cancer-associated fibroblasts (CAF) and TGF-β signaling by MCs can result in resistance to gemcitabine/nabucil paclitaxel (GEM/NAB) in pancreatic cancer (12). Poor response to neoadjuvant chemotherapy in inflammatory breast cancer is associated with MC infiltration (13).

However, the mechanism behind the interaction between MCs and OVC, as well as the crosstalk between MCs and immune cells, remains unclear. Tumor infiltrating MCs (TIMs) recognize the immune evasion subtype of advanced serous OVC, with poor prognosis and poor response to immunotherapy (14). As bioinformatics continues to evolve, biomarkers have been defined in a variety of ways. MC therapy in combination with conventional tumor therapy can be effective in improving the outcome of tumor therapy (15). In this study, twenty-nine MC-associated prognostic signature genes were identified using COX one-way regression and LASS0 regression analyses, and a scoring system of MC-associated prognostic genes (MRGs) was constructed. It can predict the effect of MCs on the prognosis and therapy more accurately at the molecular level.

Materials and Methods

Data sources. We obtained gene expression profiles and clinical data from the TCGA database for the TCGA-OVC cohort, comprising 420 tumor patients and 90 normal patients. A total of 354 OVC samples with complete clinical information were included in this study.

Screening mast cell-related prognostic genes using LASSO regression. The Molecular Signature Database (MSigDB) covered fifty-eight gene ontology (GO) pathways associated with MCs, and 601 MRGs were obtained. Differentially expressed genes (DEGs) were screened using the R package “limma” with |log2 fold change (FC)|>0 and adjusted for a p-value of <0.05. By performing univariate Cox regression analysis, we identified forty genes associated with survival. LASSO regression analyses were performed using “glmnet” in R, and the penalty regularization parameter λ was determined using ten-fold cross-validation. Multivariate Cox regression models were used to identify and calculate coefficients for the central genes in the following steps. Based on the optimal lambda values and corresponding coefficients, we constructed risk profiles based on 29 MRGs. The MRG’s risk score was calculated for each patient as follows: risk score = ExpressionmRNA1 × CoefmRNA1 + ExpressionmRNA2 × CoefmRNA2 + ExpressionmRNAn × CoefmRNAn. All OVC patients were given a risk score based on the output model equation, and the median value was calculated using the R package “survminer” to categorize patients with OVC into low-risk and high-risk subgroups and plot the survival curves for both subgroups. Using the “survivalROC” R package, risk scores were used to predict 1-, 3- and 5-year survival rates for patients in the TCGA-OVC cohort. Enrichment analysis and network analysis of prognostic genes were performed using Metascape.

MRGs model formulation. We distinguished the molecular characteristics of the 29 DEGs with prognostic values obtained from Lasso regression analysis using the PCA algorithm and developed an MC infiltration score (MCIS) using the following formula: MCIS score = ∑ (PC1 + PC2). In this formula, PC1 and PC2 represent the expression levels of DEGs with prognostic values in two different dimensions. The MRG scores were determined as a summary of PC1 and PC2, which can represent individual MRG levels.

Functional enrichment analysis. Gene Set Variation Analysis (GSVA) was performed using “c7.immunesigdb.v2022.1.Hs.symbols.gmt” from MsigDB (Molecular Signatures Database). GSVA enrichment analysis was performed using the “GSVA” R package. The “heatmap” R package was used to create the heatmap. According to the “limma” R package, an adjusted p-value of <0.05 indicates the statistical significance of subgroup differences. Functional annotation and enrichment pathways were explored through functional enrichment analysis of DEGs in OVC associated with MRG. Analysis of gene ontology (GO) and Kyoto Encyclopedia of Genes and Genomes (KEGG) pathways was performed using the “ClusterProfiler” R package, where a p-value of <0.05 represents a statistically significant difference.

MRG score validation and correlation between TME and MRG scores. Using the above formula, we calculated the MRGs for each sample, and the MRG scores in the TCGA_OVC cohort were categorized into high and low MRG groups with a cut-off of zero. Gene set enrichment analysis (GSEA) was performed to find differential functional enrichment of MRGs between high and low subgroups. We analyzed all patients’ survival rates to examine whether the MCIS score was a predictor of OVC prognosis. We used the single sample Gene Set Enrichment Analysis (ssGSEA)’s algorithm to investigate the correlation between TME and MCIS scores. As a result, we could obtain a comprehensive characterization of immune cell infiltration, immune-related pathways, and immune-related functions between the high and low groups. The correlation between MRG scores and immune cell infiltration was performed using linear regression tests.

Predicting response to chemotherapy/immunotherapy by MRG score. We analyzed the relationship between MRG scores and drug sensitivity to chemotherapy or immunotherapy. We used the TIDE algorithm to predict sensitivity to immunotherapy. OncoPredict was used to predict sensitivity to chemotherapy. All predictions of sensitivity to chemotherapy or immunotherapy were compared between high and low MCIS score groups using the Wilcoxon or χ2 test. Half-maximal inhibitory concentration (IC50) values, provided by the genomics of drug sensitivity in cancer (GDSC) database, were used per OVC patient.

Tumor Immunity Single Cell Hub (TISCH) database. TISCH is a large single-cell RNA-seg online database focusing on the tumor microenvironment (TME). In this database, the heterogeneity of TME across cell types and datasets was systematically examined.

Statistical analyses. All data processing, analyses, and graphics were performed using the R software Visual 4.2.3. The installation packages used include “tidyverse”, “glmnet”, “survival”, “ggpubr”, “caret”, “forestplot”, “org.Hs.eg.db”, “ISLR”, “XML”, “data.table”, “GSEABase”, “GSVA”, “pacman” and others. p-Values less than 0.05 were considered statistically significant.

Results

Construction and prediction of the MRG signature. We downloaded 607 MRGs from MSigDB, and the TCGA OVC cohort identified MRGs associated with 40 MRGs through univariate Cox regression analysis (Figure 1A). A risk scoring model based on MRGs was developed based on these forty genes to identify biomarkers that predict the prognosis of OVC patients. LASSO regression analysis of MRGs with prognostic value yielded LASSO regression plots (Figure 1B) and cross-validation plots (Figure 1C). In total, 29 genes involved in model construction were obtained from the cross-validation plot, which were SLC22A2, ELANE, CXCR2, DSE, ZC2HC1A, PPP3CA, NFKB1, TNFRSF8, CLEC5A, CITED2, PLEKHO2, IL27RA, CCDC80, RSPO4, FOS, SERPINA10, PLA2G2D, MSL3, SELENOS, CCR7, CALM1, HSP90AB1, SLC7A11, FZD3, EDNRB, ME1, FURIN, CXCL11, HPGD. prognostic index (PI)=(3.651* SLC22A2 exp.) + (0.550* ELANE exp.) + (0.391* CXCR2 exp.) + (0.024* DSE exp.) + (0.132* ZC2HC1A exp.) + (0.106*PPP3CA exp.) + (0.073* NFKB1 exp.) + (0.195* TNFRSF8 exp.) + (0.007* CLEC5A exp.) + (0.032* CITED2 exp.) + (0.168* PLEKHO2 exp.) + (0.159* IL27RA exp.) + (0.060* CCDC80 exp.) + (0.083* RSPO4 exp.) + (0.010* FOS exp.) + (−6.701 * SERPINA10 exp.) + (−0.270* PLA2G2D exp.) + (−0.135* MSL3 EDNRB exp) + (−0.090* ME1 exp) + (−0.016* FURIN exp) + (−0.067* CXCL11 exp) + (−0.038* HPGD exp).

Figure 1.
Figure 1.

Construction and prediction of the MRG signature: (A) Forest plot for univariate Cox regression analysis, (B) Ten-fold cross-validation of the choice of adjustment parameters in the LASSO model, (C) LASSO coefficient profile, (D) Comparison of risk scores between survivors and deaths in the TCGA cohort, (E) KM curves comparing overall OV patients between low and high-risk groups in the TCGA cohort (F) Model sensitivity and specificity assessed by ROC curves at 1, 3, and 5 years based on risk scores.

Based on the median scores, we also calculated prognostic risk scores for patients with OVC. The survival rate of the TCGA- OVC cohort shows that the risk score was even higher in the mortality population (Figure 1D). Kaplan-Meier (KM) survival analysis was performed by dividing the cohort into high- and low- risk groups. A better prognosis was found in the low-risk group (p<0.0001) (Figure 1E), and the sensitivity and specificity of the model were assessed using ROC curves (Figure 1F).

Twenty-nine prognostic genes screened using Lasso regression analysis. Twenty-nine prognostic genes were analyzed using the metascape online network analysis tool. These genes were mainly enriched in the regulation of apoptosis in myeloid cells, regulation of cytokines, regulation of inflammatory response to antigenic stimulation, cellular response to cytokine stimulation, transforming growth factor β-receptor signaling pathway, regulation of humoral immunity, positive regulation of calcium transmembrane translocation, cellular signaling, positive regulation of protein kinase activity, regulation of tumor necrosis factor production and regulation of endopeptidase activity Pathways (Figure 2A). Additionally, network analysis was conducted on the enriched signaling pathways (Figure 2B).

Figure 2.
Figure 2.

Enrichment and network analysis. (A) Enrichment analysis was performed on 29 prognostic genes; (B) Network analysis was conducted on the enriched signaling pathways.

Developing a scoring model for MRGs and immune infiltration analysis. Therefore, we constructed a scoring system to quantify an individual’s MCIS pattern. MCIS allows us to categorize patients into high or low MCIS score groups. We conducted gene set enrichment analysis using GSEA (Figure 3A) and GSVA (Figure 3B) methods to investigate differential activity pathways between the high and low groups. We found significant enrichment of B cells, monocytes, NK cells, CD11B+ dendritic cells, CD8 T cells, and bone marrow-derived macrophages in the high MCIS group. Subsequently, survival analysis was performed on the high and low MCIS groups, revealing a survival advantage for patients with low MRGs scores in the TCGA_OV (Figure 3C) and validation set TCGA_BRCA cohorts (Figure 3D). We conducted immune infiltration analysis on the high and low groups using the ssGSEA in the two cohorts (Figure 3E and F). We observed significantly higher levels of immune cell infiltration, including B cells, T cells (CD4+ and CD8+), macrophages (M1 and M2), myeloid dendritic cells, NK cells, mast cells, DC cells, TH2 cells, and TH17 cells, in the low MRGs score group. As the MRGs score decreased, the degree of immune infiltration increased.

Figure 3.
Figure 3.

Development of mast cell-associated prognostic genes (MRGs) scoring and functional annotation. (A) Gene set enrichment analysis (GSEA) of GO, KEGG, and HALLMARK gene sets in the low mast cell infiltration score (MCIS) group. (B) Gene set variation analysis (GSVA) enrichment analysis. (C) Survival analysis of the high and low MCIS groups in the TCGA_OV cohort. (D) Survival analysis of the high and low MCIS groups in the TCGA_BRCA cohort. (E) Immune infiltration analysis using the single sample Gene Set Enrichment Analysis (ssGSEA) method in the TCGA_OV cohort. (F) Immune infiltration analysis using the ssGSEA method in the TCGA_BRCA cohort.

MRGs score predicts chemotherapy sensitivity. By using the OncoPredict package to predict the sensitivity to 198 chemotherapeutic agents, we found that 5-Fluorouracil_1073, AZD2014_1441, AZD5438_1401, BMS-345541_1249, Dactolisib_1057, Dinaciclib_ 1180, GNE-317_1926, Leflunomide_1578, Pevonedistat_1529, Docetaxel_1819, and Rapamycin_1084 had higher IC50s in the high MCIS group than in the low MCIS group (Figure 4A-K). In addition, SB505124_1194 had higher IC50 in the low MCIs group for (Figure 4L). Based on the risk score, we can further investigate the sensitivity to chemotherapy of OVC patients and enhance the precision therapy.

Figure 4.
Figure 4.

Sensitivity to 5-Fluorouracil_1073 (A), AZD2014_1441 (B), AZD5438_1401 (C), BMS-345541_1249 (D), Dactolisib_1057 (E), Dinaciclib_1180 (F) GNE-317_1926 (G), Leflunomide_1578 (H), Pevonedistat_1529 (I), Docetaxel_1819 (J), Rapamycin_1084 (K), SB505124_1194 (L).

MRGs score predicts sensitivity to immune checkpoints. The TIDE algorithm predicted efficacy to immune checkpoint inhibitors and found that high subgroups had higher TIDE scores, MSI scores, and Dysfunction scores. Therefore, high subgroups had stronger immune escape and poorer sensitivity to immune checkpoint inhibitors (Figure 5A). We predicted that patients who responded to anti-PD-1 immunotherapy in the IMvigor210 cohort had lower MCIS (Figure 5B) (Note: patients who died within one month of using anti-PD-1 immunotherapy were removed).

Figure 5.
Figure 5.

Sensitivity to immune checkpoints. (A) TIDE scores, exclusion scores, dysfunction scores, and MSI score based on mast cell-associated prognostic genes (MRGs) score. (B) In the IMvigor210 cohort, mast cell infiltration score (MCIS) in the response group and non-response group.

Correlation analysis of MRGs scores and immune environment. The single-cell dataset OV_GSE151214 from the TISCH database was used to analyze the expression of the 29-MRGs in the immune microenvironment. In the GSE151214 dataset, there were 24 cell populations and nine immune cell types (Figure 6A and B), and the distribution and number of various cell types showing the expression levels of each MRG in immune cells are shown (Figure 6C and D). CLEC5A, CXCL11, CXCR2, ELANE, FURIN, FZD3, ME1, PLA2G2D, RSPO4, SLC22A2, and TNFRSF8 were barely expressed in the immune microenvironment; FOS and HSP90AB1 were highly expressed in the immune microenvironment. CALM1, NFKB1, and HPGD were highly expressed in MCs; CCR7, DSE, SLC7A11, and NFKB1 were highly expressed in mono/macro cells; CALM1 and ZC2HC1A were highly expressed in epithelial cells; CCDC80, CITED2, HPGD, and ZC2HC1A were highly expressed in fibroblasts; CALM1, CCR7, CITED2, and IL27RA were highly expressed in CD8+T cells; and CALM1 was highly expressed in endothelial cells. We analyzed the expression of transcription factors in each cell type, and we found that SMAD1, LMO2, LYL1, SPT1, ERG, RUNX1, RAD21, GATA2, TAL1, MED12, and RUNX1T1 were highly expressed in MCs, which promoted cell growth, development, and differentiation. MCs are strongly associated with signaling pathways, such as antigen presentation, cell-cell adhesion, and immune cell activation. We also found that MCs were specifically involved in the up regulation of CD4+ T, naïve CD4+ T, NKT, CD8+ T, and B cell cells. MCs were associated with epithelial-mesenchymal, IFN-r response, muscle differentiation and development. We also investigated the interaction between MCs and other immune cells (Figure 6F).

Figure 6.
Figure 6.

Single cell analysis in OV_GSE151214. (A) 24 cell populations. (B) Nine immune cell types. (C) The proportion of nine immune cell types. (D) Distribution of immune cells in each patient. (E) The interaction between mast cells and other immune cells. (F) The expression levels of each mast cell-associated prognostic gene in immune cells.

Discussion

MCs play a crucial role in the TME and immune surveillance. However, a comprehensive analysis of MC-related genes (MRGs) in OVC has not yet been reported. Therefore, we utilized mRNA expression data from the TCGA-OVC dataset to identify important prognostic genes and developed a multi-biomarker prognostic model based on MRGs.

To further investigate individual MRG patterns, we developed a scoring system called MCIS to quantify the MRG patterns in OVC patients. Using the MCIS, we divided all patients into two groups: high MCIS group and low MCIS group. We first performed GSEA to explore potential pathway enrichments between the two subgroups. We found significant enrichment in pathways related to purine nucleotide metabolism, ribose phosphate metabolism, proton transmembrane transport, response to corticosteroids, and corticosteroid metabolism. In terms of cellular components, pathways related to the proton transmembrane transport complex and ATPase complex were enriched in the low MCIS group. However, metabolism, cell growth, and differentiation pathways were enriched in the high MCIS group. This result suggests that the low MCIS group has greater anti-tumor potential. Subsequent survival analysis also demonstrated that patients in the low MCIS group had a significantly better prognosis than those in the high MCIS group. The low MCIS group showed higher overall immune cell infiltration levels, indicating a stronger immune response. Differential pathway analysis using GSEA and GSVA methods revealed significant enrichment in B cells, monocytes, NK cells, CD11B+ dendritic cells, CD8 T cells, and bone marrow-derived macrophages in the low MRG score group.

MCs can enhance anti-tumor response by recruiting natural killer (NK) cells, dendritic cells (DCs), and T cells and interacting with them (16). However, they can also promote tumor progression by interfering with peripheral antigen tolerance induced by Treg cells (17). SLC22A2 participates in histamine transport and stimulates histamine secretion. Histamine binds to H1R on TH1 cells, activates NK cells and CD8+CTLs, and increases the expression of MHC and co-stimulatory molecules, exhibiting strong antitumor activity (18). Histamine binding to H2R actively interferes with peripheral antigen tolerance induced by Treg cells (17). ELANE can liberate the death domain of CD95 and interact with histone H1 isoforms, eradicating cancer cells. It also generates CD8+ T cell-mediated radiation bystander effects to attack distant metastases (19). EDNRB mediates T cell homing to the tumor endothelial barrier, allowing the tumor to respond to ineffective immune therapy without altering systemic immunity (20). ZC2HC1A, also known as SART2, can activate CTLs to exhibit cytotoxicity against colon cancer cells upon stimulation with two immunogenic peptides of SART3 and SART2 (21, 22). MCs express IL-27R, which directly inhibits CD4+ T cell proliferation and the production of Th2 cell cytokines (23).

NFkB1, CLEC5A, SELENOS, and FOS inhibit the epithelial-mesenchymal transition (EMT) of tumor cells, while HSP90AB1 and HPGD promote the EMT of tumor cells. MCs activate the NF-kB-mediated signaling pathway and express inflammatory genes, such as NFkB1, IL6, IL1B, CXCL8, and CCL3 (24), promoting the recruitment of neutrophils, eosinophils, and B cells (25). NFKB1 can suppress the transcription of pro-tumor-related genes and inhibit the EMT of tumor cells (26, 27). CLEC5A and SELENOS are involved in MC apoptosis and inhibit tumor migration. Over-expression of CLEC5A significantly suppresses hepatocyte proliferation, migration, and invasion and reverses EMT (28). SELENOS inhibits apoptosis and suppresses EMT to inhibit tumor cell migration through the AKT/GSK3β/NF-Embedded ImageB signaling pathway in clear cell renal cell carcinoma (ccRCC) (29). The binding of c-Fos to IgE Fc receptors (FcεRI) is associated with the activation of immunoglobulin (Ig) MCs (30). In ovarian cancer cells, overexpression of c-FOS can reduce the adhesion between the extracellular matrix (type I and type IV collagen), endothelial cells, and mesothelial cells, and inhibit the metastasis of OVC (31). HSP90AB1 is a specific protein of MCs and promotes tumor proliferation, migration, glycolysis, and EMT and decreases the level of phosphorylated Akt (32, 33). The biliary exosome miR-182/183-5p promotes the proliferation, invasion, and EMT of cholangiocarcinoma cells by targeting HPGD in cholangiocarcinoma cells and MCs and increasing the release of PGE2 and VEGF-A (34).

CITED2 and CXCR2 are involved in angiogenesis. CITED2 is significantly up-regulated in MCs, and its down-regulation inhibits TGF-β-dependent VEGFA expression and suppresses angiogenesis (35). CXCR2 may promote the activation and recruitment of MCs and is highly expressed on recruited inflammatory cells in tumors, inducing angiogenesis and promoting tumor growth (36, 37). Interaction between CXCL11 and TWIST1 induces angiogenesis in epithelial ovarian cancer (38). In breast cancer cells, high expression of CITED2 promotes chemoresistance to doxorubicin and 5-fluorouracil (39). Hsp90B enhances cisplatin resistance mediated by MAST1 by protecting it from proteasomal degradation (40).

Currently, there is limited understanding of the phenotype and function of tumor-infiltrating MCs. Studies have shown that MCs are one of the first immune cells recruited to solid tumor sites, increasing in precancerous lesions. As cancer progresses, the content of MCs is higher (41). MCs can also enhance anti-tumor responses by recruiting and interacting with natural killer cells (NK), dendritic cells (DCs), and T cells. MCs infiltrate solid tumors and kill cancer cells, exerting anti-tumor effects, which suggests that inducing the biological activity of MCs can be utilized for anti-tumor purposes, making them an important target for immunotherapy (16). Treatment approaches include c-KIT inhibitors, mast cell stabilizers, FcεR1 signaling pathway activators/inhibitors, antibodies targeting inhibitory receptors and ligands, TLR agonists, and modulators of mast cell mediators. Recent research has shown synergistic and durable responses when combined with ICIs (42). The MCIS group has higher TIDE, MSI, and Dysfunction scores, indicating a stronger immune escape and lower sensitivity to ICIs. In the MCIS scoring system, we can identify individuals who are sensitive to immunotherapy. Our research also indicates that the low MCIS group is more sensitive to 5-Fluorouracil_1073, AZD2014_1441, AZD5438_1401, BMS-345541_1249, Dactolisib_1057, Dinaciclib_1180, GNE-317_1926, Leflunomide_1578, Pevonedistat_1529, Docetaxel_1819, and Rapamycin_1084. Therefore, the low MCIS group is not only more sensitive to immunotherapy and chemotherapy but also less likely to develop resistance.

It is necessary to point out that, although our study can have a greater clinical impact on the prognosis and treatment of OVC patients, some limitations remain. First, our study is a retrospective study that needs to be validated in future prospective studies. This study indirectly evaluated the potential of this signature to predict immunotherapy responses. Therefore, future validation should be combined with genetic data from OVC patients receiving immunotherapy. Nevertheless, our study strongly indicates that mast cell-related prognosis signature characterizes immune landscape and predicts prognosis of OVC. Thus, the correlation between mast cell related gene signatures and immunotherapy and chemotherapy may be able to help improve personalized clinical treatment strategies in the future.

Acknowledgements

This work was partially by a grant IOER 112-3119 from Des Moines University for Dr. Yujiang Fang.

Footnotes

  • Authors’ Contributions

    Qinghua Li, Yujiang Fang, and Qingyu Zhao concepted this study. Qinghua Li, Yujiang Fang, and Qingyu Zhao designed the study. Qinghua Li, Tianyun Guan, Jianxiong Mao, Benjamin P. King, Kesiya Johnson, Trenton G. Mayberry, Braydon C. Cowan, and Yujiang Fang collected, analyzed, and interpreted the data. Qinghua Li wrote the draft, Yujiang Fang and Mark R. Wakefield made critical revision.

  • Conflicts of Interest

    The Authors have no conflicts of interest to declare in relation to this study.

  • Received March 27, 2024.
  • Revision received May 8, 2024.
  • Accepted May 22, 2024.

This article is an open access article distributed under the terms and conditions of the Creative Commons Attribution (CC BY-NC-ND) 4.0 international license (https://creativecommons.org/licenses/by-nc-nd/4.0).

References

  1. ↵
    1. Nag S,
    2. Aggarwal S,
    3. Rauthan A,
    4. Warrier N
    : Maintenance therapy for newly diagnosed epithelial ovarian cancer- a review. J Ovarian Res 15(1): 88, 2022. DOI: 10.1186/s13048-022-01020-1
    OpenUrlCrossRefPubMed
  2. ↵
    1. Schoutrop E,
    2. Moyano-Galceran L,
    3. Lheureux S,
    4. Mattsson J,
    5. Lehti K,
    6. Dahlstrand H,
    7. Magalhaes I
    : Molecular, cellular and systemic aspects of epithelial ovarian cancer and its tumor microenvironment. Semin Cancer Biol 86(Pt 3): 207-223, 2022. DOI: 10.1016/j.semcancer.2022.03.027
    OpenUrlCrossRefPubMed
  3. ↵
    1. Porter R,
    2. Matulonis UA
    : Immunotherapy for ovarian cancer. Clin Adv Hematol Oncol 20(4): 240-253, 2022.
    OpenUrlPubMed
  4. ↵
    1. Ribatti D
    : A double-edged sword in tumor angiogenesis and progression. Dual roles of mast cells, macrophages, and neutrophils. Pathol Res Pract 240: 154167, 2022. DOI: 10.1016/j.prp.2022.154167
    OpenUrlCrossRefPubMed
  5. ↵
    1. Komi DEA,
    2. Redegeld FA
    : Role of mast cells in shaping the tumor microenvironment. Clin Rev Allergy Immunol 58(3): 313-325, 2020. DOI: 10.1007/s12016-019-08753-w
    OpenUrlCrossRefPubMed
  6. ↵
    1. Ribatti D,
    2. Crivellato E
    : Mast cells, angiogenesis, and tumour growth. Biochim Biophys Acta 1822(1): 2-8, 2012. DOI: 10.1016/j.bbadis.2010.11.010
    OpenUrlCrossRefPubMed
  7. ↵
    1. Riquelme-Neira R,
    2. Walker-Vergara R,
    3. Fernández-Blanco JA,
    4. Vergara P
    : IL-10 modulates the expression and activation of pattern recognition receptors in mast cells. Int J Mol Sci 24(12): 9875, 2023. DOI: 10.3390/ijms24129875
    OpenUrlCrossRefPubMed
  8. ↵
    1. Hempel HA,
    2. Cuka NS,
    3. Kulac I,
    4. Barber JR,
    5. Cornish TC,
    6. Platz EA,
    7. De Marzo AM,
    8. Sfanos KS
    : Low intratumoral mast cells are associated with a higher risk of prostate cancer recurrence. Prostate 77(4): 412-424, 2017. DOI: 10.1002/pros.23280
    OpenUrlCrossRefPubMed
  9. ↵
    1. Li J,
    2. Peng G,
    3. Zhu K,
    4. Jie X,
    5. Xu Y,
    6. Rao X,
    7. Xu Y,
    8. Chen Y,
    9. Xing B,
    10. Wu G,
    11. Shi L
    : PD-1(+) mast cell enhanced by PD-1 blocking therapy associated with resistance to immunotherapy. Cancer Immunol Immunother 72(3): 633-645, 2023. DOI: 10.1007/s00262-022-03282-6
    OpenUrlCrossRefPubMed
  10. ↵
    1. Somasundaram R,
    2. Connelly T,
    3. Choi R,
    4. Choi H,
    5. Samarkina A,
    6. Li L,
    7. Gregorio E,
    8. Chen Y,
    9. Thakur R,
    10. Abdel-Mohsen M,
    11. Beqiri M,
    12. Kiernan M,
    13. Perego M,
    14. Wang F,
    15. Xiao M,
    16. Brafford P,
    17. Yang X,
    18. Xu X,
    19. Secreto A,
    20. Danet-Desnoyers G,
    21. Traum D,
    22. Kaestner KH,
    23. Huang AC,
    24. Hristova D,
    25. Wang J,
    26. Fukunaga-Kalabis M,
    27. Krepler C,
    28. Ping-Chen F,
    29. Zhou X,
    30. Gutierrez A,
    31. Rebecca VW,
    32. Vonteddu P,
    33. Dotiwala F,
    34. Bala S,
    35. Majumdar S,
    36. Dweep H,
    37. Wickramasinghe J,
    38. Kossenkov AV,
    39. Reyes-Arbujas J,
    40. Santiago K,
    41. Nguyen T,
    42. Griss J,
    43. Keeney F,
    44. Hayden J,
    45. Gavin BJ,
    46. Weiner D,
    47. Montaner LJ,
    48. Liu Q,
    49. Peiffer L,
    50. Becker J,
    51. Burton EM,
    52. Davies MA,
    53. Tetzlaff MT,
    54. Muthumani K,
    55. Wargo JA,
    56. Gabrilovich D,
    57. Herlyn M
    : Tumor-infiltrating mast cells are associated with resistance to anti-PD-1 therapy. Nat Commun 12(1): 346, 2021. DOI: 10.1038/s41467-020-20600-7
    OpenUrlCrossRefPubMed
  11. ↵
    1. Xie H,
    2. Li C,
    3. Dang Q,
    4. Chang LS,
    5. Li L
    : Infiltrating mast cells increase prostate cancer chemotherapy and radiotherapy resistances via modulation of p38/p53/p21 and ATM signals. Oncotarget 7(2): 1341-1353, 2016. DOI: 10.18632/oncotarget.6372
    OpenUrlCrossRefPubMed
  12. ↵
    1. Porcelli L,
    2. Iacobazzi RM,
    3. Di Fonte R,
    4. Serratì S,
    5. Intini A,
    6. Solimando AG,
    7. Brunetti O,
    8. Calabrese A,
    9. Leonetti F,
    10. Azzariti A,
    11. Silvestris N
    : CAFs and TGF-β signaling activation by mast cells contribute to resistance to gemcitabine/nabpaclitaxel in pancreatic cancer. Cancers (Basel) 11(3): 330, 2019. DOI: 10.3390/cancers11030330
    OpenUrlCrossRefPubMed
  13. ↵
    1. Reddy SM,
    2. Reuben A,
    3. Barua S,
    4. Jiang H,
    5. Zhang S,
    6. Wang L,
    7. Gopalakrishnan V,
    8. Hudgens CW,
    9. Tetzlaff MT,
    10. Reuben JM,
    11. Tsujikawa T,
    12. Coussens LM,
    13. Wani K,
    14. He Y,
    15. Villareal L,
    16. Wood A,
    17. Rao A,
    18. Woodward WA,
    19. Ueno NT,
    20. Krishnamurthy S,
    21. Wargo JA,
    22. Mittendorf EA
    : Poor response to neoadjuvant chemotherapy correlates with mast cell infiltration in inflammatory breast cancer. Cancer Immunol Res 7(6): 1025-1035, 2019. DOI: 10.1158/2326-6066.CIR-18-0619
    OpenUrlAbstract/FREE Full Text
  14. ↵
    1. Zou R,
    2. Jiang Q,
    3. Jin T,
    4. Chen M,
    5. Yao L,
    6. Ding H
    : Pan-cancer analyses and molecular subtypes based on the cancer-associated fibroblast landscape and tumor microenvironment infiltration characterization reveal clinical outcome and immunotherapy response in epithelial ovarian cancer. Front Immunol 13: 956224, 2022. DOI: 10.3389/fimmu.2022.956224
    OpenUrlCrossRefPubMed
  15. ↵
    1. Ribatti D
    : Mast cells and resistance to immunotherapy in cancer. Arch Immunol Ther Exp (Warsz) 71(1): 11, 2023. DOI: 10.1007/s00005-023-00676-x
    OpenUrlCrossRefPubMed
  16. ↵
    1. Fereydouni M,
    2. Ahani E,
    3. Desai P,
    4. Motaghed M,
    5. Dellinger A,
    6. Metcalfe DD,
    7. Yin Y,
    8. Lee SH,
    9. Kafri T,
    10. Bhatt AP,
    11. Dellinger K,
    12. Kepley CL
    : Human tumor targeted cytotoxic mast cells for cancer immunotherapy. Front Oncol 12: 871390, 2022. DOI: 10.3389/fonc.2022.871390
    OpenUrlCrossRefPubMed
  17. ↵
    1. Jutel M,
    2. Blaser K,
    3. Akdis CA
    : Histamine in allergic inflammation and immune modulation. Int Arch Allergy Immunol 137(1): 82-92, 2005. DOI: 10.1159/000085108
    OpenUrlCrossRefPubMed
  18. ↵
    1. Okeke EB,
    2. Uzonna JE
    : The pivotal role of regulatory T cells in the regulation of innate immune cells. Front Immunol 10: 680, 2019. DOI: 10.3389/fimmu.2019.00680
    OpenUrlCrossRefPubMed
  19. ↵
    1. Cui C,
    2. Chakraborty K,
    3. Tang XA,
    4. Zhou G,
    5. Schoenfelt KQ,
    6. Becker KM,
    7. Hoffman A,
    8. Chang YF,
    9. Blank A,
    10. Reardon CA,
    11. Kenny HA,
    12. Vaisar T,
    13. Lengyel E,
    14. Greene G,
    15. Becker L
    : Neutrophil elastase selectively kills cancer cells and attenuates tumorigenesis. Cell 184(12): 3163-3177.e21, 2021. DOI: 10.1016/j.cell.2021.04.016
    OpenUrlCrossRefPubMed
  20. ↵
    1. Buckanovich RJ,
    2. Facciabene A,
    3. Kim S,
    4. Benencia F,
    5. Sasaroli D,
    6. Balint K,
    7. Katsaros D,
    8. O’Brien-Jenkins A,
    9. Gimotty PA,
    10. Coukos G
    : Endothelin B receptor mediates the endothelial barrier to T cell homing to tumors and disables immune therapy. Nat Med 14(1): 28-36, 2008. DOI: 10.1038/nm1699
    OpenUrlCrossRefPubMed
  21. ↵
    1. Sasatomi T,
    2. Suefuji Y,
    3. Matsunaga K,
    4. Yamana H,
    5. Miyagi Y,
    6. Araki Y,
    7. Ogata Y,
    8. Itoh K,
    9. Shirouzu K
    : Expression of tumor rejection antigens in colorectal carcinomas. Cancer 94(6): 1636-1641, 2002. DOI: 10.1002/cncr.10421
    OpenUrlCrossRefPubMed
  22. ↵
    1. Tanaka S,
    2. Tsuda N,
    3. Kawano K,
    4. Sakamoto M,
    5. Nishida T,
    6. Hashimoto T,
    7. Shichijo S,
    8. Kamura T,
    9. Itoh K
    : Expression of tumor-rejection antigens in gynecologic cancers. Jpn J Cancer Res 91(11): 1177-1184, 2000. DOI: 10.1111/j.1349-7006.2000.tb00902.x
    OpenUrlCrossRefPubMed
  23. ↵
    1. Artis D,
    2. Villarino A,
    3. Silverman M,
    4. He W,
    5. Thornton EM,
    6. Mu S,
    7. Summer S,
    8. Covey TM,
    9. Huang E,
    10. Yoshida H,
    11. Koretzky G,
    12. Goldschmidt M,
    13. Wu GD,
    14. De Sauvage F,
    15. Miller HRP,
    16. Saris CJM,
    17. Scott P,
    18. Hunter CA
    : The IL-27 receptor (WSX-1) is an inhibitor of innate and adaptive elements of type 2 immunity. J Immunol 173(9): 5626-5634, 2004. DOI: 10.4049/jimmunol.173.9.5626
    OpenUrlAbstract/FREE Full Text
  24. ↵
    1. Johnzon C,
    2. Rönnberg E,
    3. Pejler G
    : The role of mast cells in bacterial infection. Am J Pathol 186(1): 4-14, 2016. DOI: 10.1016/j.ajpath.2015.06.024
    OpenUrlCrossRefPubMed
  25. ↵
    1. Mayavannan A,
    2. Shantz E,
    3. Haidl ID,
    4. Wang J,
    5. Marshall JS
    : Mast cells selectively produce inflammatory mediators and impact the early response to Chlamydia reproductive tract infection. Front Immunol 14: 1166068, 2023. DOI: 10.3389/fimmu.2023.1166068
    OpenUrlCrossRefPubMed
  26. ↵
    1. Lin TC,
    2. Yang CH,
    3. Cheng LH,
    4. Chang WT,
    5. Lin YR,
    6. Cheng HC
    : Fibronectin in cancer: Friend or foe. Cells 9(1): 27, 2019. DOI: 10.3390/cells9010027
    OpenUrlCrossRefPubMed
  27. ↵
    1. Kravtsova-Ivantsiv Y,
    2. Shomer I,
    3. Cohen-Kaplan V,
    4. Snijder B,
    5. Superti-Furga G,
    6. Gonen H,
    7. Sommer T,
    8. Ziv T,
    9. Admon A,
    10. Naroditsky I,
    11. Jbara M,
    12. Brik A,
    13. Pikarsky E,
    14. Kwon YT,
    15. Doweck I,
    16. Ciechanover A
    : KPC1-mediated ubiquitination and proteasomal processing of NF-κB1 p105 to p50 restricts tumor growth. Cell 161(2): 333-347, 2015. DOI: 10.1016/j.cell.2015.03.001
    OpenUrlCrossRefPubMed
  28. ↵
    1. Lin J,
    2. Ou H,
    3. Wang W,
    4. Ma J,
    5. Zhang W,
    6. Liu Q
    : [Overexpression of CLEC5A inhibits cell proliferation and metastasis and reverses epithelial-mesenchymal transition in hepatocellular carcinoma]. Nan Fang Yi Ke Da Xue Xue Bao 43(1): 85-91, 2023. DOI: 10.12122/j.issn.1673-4254.2023.01.11
    OpenUrlCrossRefPubMed
  29. ↵
    1. Mao H,
    2. Zhao Y,
    3. Lei L,
    4. Hu Y,
    5. Zhu H,
    6. Wang R,
    7. Ni D,
    8. Liu J,
    9. Xu L,
    10. Xia H,
    11. Zhang Z,
    12. Ma M,
    13. Pan Z,
    14. Zhou Q,
    15. Xie Y
    : Selenoprotein S regulates tumorigenesis of clear cell renal cell carcinoma through AKT/GSK3β/NF-κB signaling pathway. Gene 832: 146559, 2022. DOI: 10.1016/j.gene.2022.146559
    OpenUrlCrossRefPubMed
  30. ↵
    1. Wang HN,
    2. Ji K,
    3. Zhang LN,
    4. Xie CC,
    5. Li WY,
    6. Zhao ZF,
    7. Chen JJ
    : Inhibition of c-Fos expression attenuates IgE-mediated mast cell activation and allergic inflammation by counteracting an inhibitory AP1/Egr1/IL-4 axis. J Transl Med 19(1): 261, 2021. DOI: 10.1186/s12967-021-02932-0
    OpenUrlCrossRefPubMed
  31. ↵
    1. Oliveira-Ferrer L,
    2. Rößler K,
    3. Haustein V,
    4. Schröder C,
    5. Wicklein D,
    6. Maltseva D,
    7. Khaustova N,
    8. Samatov T,
    9. Tonevitsky A,
    10. Mahner S,
    11. Jänicke F,
    12. Schumacher U,
    13. Milde-Langosch K
    : c-FOS suppresses ovarian cancer progression by changing adhesion. Br J Cancer 110(3): 753-763, 2014. DOI: 10.1038/bjc.2013.774
    OpenUrlCrossRefPubMed
  32. ↵
    1. Antoniali G,
    2. Serra F,
    3. Lirussi L,
    4. Tanaka M,
    5. D’Ambrosio C,
    6. Zhang S,
    7. Radovic S,
    8. Dalla E,
    9. Ciani Y,
    10. Scaloni A,
    11. Li M,
    12. Piazza S,
    13. Tell G
    : Mammalian APE1 controls miRNA processing and its interactome is linked to cancer RNA metabolism. Nat Commun 8(1): 797, 2017. DOI: 10.1038/s41467-017-00842-8
    OpenUrlCrossRefPubMed
  33. ↵
    1. Wang H,
    2. Deng G,
    3. Ai M,
    4. Xu Z,
    5. Mou T,
    6. Yu J,
    7. Liu H,
    8. Wang S,
    9. Li G
    : Hsp90ab1 stabilizes LRP5 to promote epithelial-mesenchymal transition via activating of AKT and Wnt/β-catenin signaling pathways in gastric cancer progression. Oncogene 38(9): 1489-1507, 2019. DOI: 10.1038/s41388-018-0532-5
    OpenUrlCrossRefPubMed
  34. ↵
    1. Shu L,
    2. Li X,
    3. Liu Z,
    4. Li K,
    5. Shi A,
    6. Tang Y,
    7. Zhao L,
    8. Huang L,
    9. Zhang Z,
    10. Zhang D,
    11. Huang S,
    12. Lian S,
    13. Sheng G,
    14. Yan Z,
    15. Zhang Z,
    16. Xu Y
    : Bile exosomal miR-182/183-5p increases cholangiocarcinoma stemness and progression by targeting HPGD and increasing PGE2 generation. Hepatology 79(2): 307-322, 2024. DOI: 10.1097/hep.0000000000000437
    OpenUrlCrossRefPubMed
  35. ↵
    1. Jayaraman S,
    2. Doucet M,
    3. Kominsky SL
    : Down-regulation of CITED2 attenuates breast tumor growth, vessel formation and TGF-β-induced expression of VEGFA. Oncotarget 8(4): 6169-6178, 2017. DOI: 10.18632/oncotarget.14048
    OpenUrlCrossRefPubMed
  36. ↵
    1. Yanagawa J,
    2. Walser TC,
    3. Zhu LX,
    4. Hong L,
    5. Fishbein MC,
    6. Mah V,
    7. Chia D,
    8. Goodglick L,
    9. Elashoff DA,
    10. Luo J,
    11. Magyar CE,
    12. Dohadwala M,
    13. Lee JM,
    14. St John MA,
    15. Strieter RM,
    16. Sharma S,
    17. Dubinett SM
    : Snail promotes CXCR2 ligand-dependent tumor progression in non-small cell lung carcinoma. Clin Cancer Res 15(22): 6820-6829, 2009. DOI: 10.1158/1078-0432.CCR-09-1558
    OpenUrlAbstract/FREE Full Text
  37. ↵
    1. Keane MP,
    2. Belperio JA,
    3. Xue YY,
    4. Burdick MD,
    5. Strieter RM
    : Depletion of CXCR2 inhibits tumor growth and angiogenesis in a murine model of lung cancer. J Immunol 172(5): 2853-2860, 2004. DOI: 10.4049/jimmunol.172.5.2853
    OpenUrlAbstract/FREE Full Text
  38. ↵
    1. Koo YJ,
    2. Kim TJ,
    3. Min KJ,
    4. So KA,
    5. Jung US,
    6. Hong JH
    : CXCL11 mediates TWIST1-induced angiogenesis in epithelial ovarian cancer. Tumour Biol 39(5): 101042831770622, 2017. DOI: 10.1177/1010428317706226
    OpenUrlCrossRef
  39. ↵
    1. Minemura H,
    2. Takagi K,
    3. Sato A,
    4. Takahashi H,
    5. Miki Y,
    6. Shibahara Y,
    7. Watanabe M,
    8. Ishida T,
    9. Sasano H,
    10. Suzuki T
    : CITED2 in breast carcinoma as a potent prognostic predictor associated with proliferation, migration and chemoresistance. Cancer Sci 107(12): 1898-1908, 2016. DOI: 10.1111/cas.13081
    OpenUrlCrossRefPubMed
  40. ↵
    1. Pan C,
    2. Chun J,
    3. Li D,
    4. Boese AC,
    5. Li J,
    6. Kang J,
    7. Umano A,
    8. Jiang Y,
    9. Song L,
    10. Magliocca KR,
    11. Chen ZG,
    12. Saba NF,
    13. Shin DM,
    14. Owonikoko TK,
    15. Lonial S,
    16. Jin L,
    17. Kang S
    : Hsp90B enhances MAST1-mediated cisplatin resistance by protecting MAST1 from proteosomal degradation. J Clin Invest 129(10): 4110-4123, 2019. DOI: 10.1172/JCI125963
    OpenUrlCrossRefPubMed
  41. ↵
    1. Ribatti D,
    2. Crivellato E
    : The controversial role of mast cells in tumor growth. Int Rev Cell Mol Biol 275: 89-131, 2009. DOI: 10.1016/s1937-6448(09)75004-x
    OpenUrlCrossRefPubMed
  42. ↵
    1. Lichterman JN,
    2. Reddy SM
    : Mast cells: a new frontier for cancer immunotherapy. Cells 10(6): 1270, 2021. DOI: 10.3390/cells10061270
    OpenUrlCrossRef
Back to top

In this issue

Print
Download PDF
Article Alerts
Email Article
Citation Tools
Reprints and Permissions
Mast Cell-related Prognosis Signature Characterizes Immune Landscape and Predicts Prognosis of Ovarian Cancer
QINGHUA LI, TIANYUN GUAN, JIANXIONG MAO, BENJAMIN P. KING, KESIYA JOHNSON, TRENTON G. MAYBERRY, BRAYDON C. COWAN, MARK R. WAKEFIELD, YUJIANG FANG, QINGYU ZHAO
Anticancer Research Jul 2024, 44 (7) 2775-2786; DOI: 10.21873/anticanres.17089
Twitter logo Facebook logo Mendeley logo

Jump to section

Keywords