Application of Biomarkers and Risk Models Based on Amino Acid Metabolism-Related Genes

Abstract

The present invention provides the application of the construction of biomarkers and risk models based on amino acid metabolism-related genes in predicting the diagnosis, treatment and prognosis of pheochromocytoma, belonging to the technical field of tumor diagnostic markers. We found through research that the amino acid metabolism-related genes DDC and / or SYT11 show high expression levels in pheochromocytoma, have an obvious impact on the diagnosis and prognosis of pheochromocytoma, can accurately evaluate the prognostic risk of patients, and can become a new target for the treatment of pheochromocytoma. At the same time, we constructed a risk model that can assist in judging the diagnosis and treatment of patients with the expression levels of DDC, SYT11, GCLM, PSMB7, TYRO3 and / or AGMAT as parameters. This risk model has strong predictive ability for the selection of immunotherapy methods, can be used as an important indicator for the selection of immunotherapy methods, can identify patients with high-risk factors at an early stage, and provides an important basis for the precise formulation of treatment strategies and individualized treatment decisions.

Description

Technical Field

[0001] The present invention relates to the technical field of tumor diagnostic markers, and particularly to the application of the construction of biomarkers and risk models based on amino acid metabolism-related genes in predicting the diagnosis, treatment and prognosis of pheochromocytoma. Background Art

[0002] Pheochromocytoma / Paraganglioma (PCPG) is a rare neuroendocrine tumor that originates from chromaffin cells in the adrenal medulla. Its typical clinical manifestations usually include hypertension, headache, palpitations and profuse sweating. All PCPGs are considered to have malignant potential, and early surgical resection is the way to prevent metastasis. Despite the progress made in the genetics and transcriptomics of PCPG, the treatment options for PCPG are still limited. In addition to surgery and radiotherapy, combined chemotherapy regimens usually include cyclophosphamide, vincristine and dacarbazine (CVD regimen). However, the CVD regimen has certain limitations in improving the quality of life and overall survival of PCPG patients. Due to the complex pathogenesis of pheochromocytoma, there is currently no effective targeted therapy to improve the prognosis of PCPG patients. Therefore, it is necessary to construct a well-performing PCPG prognosis model and new sensitive diagnostic biomarkers for the development of new treatment methods.

[0003] The metabolic ecosystem of tumors is extremely complex, in which amino acid metabolism plays a key role in multiple aspects, including energy production, nucleotide synthesis, cellular redox balance, and the interconnections between metabolic pathways. In many tumors, the changes in amino acids and their metabolites significantly affect the metabolic state of tumor cells. Taking leucine as an example, it regulates tumor metabolism and growth by binding to Sestrin2, which negatively regulates the mTORC1 pathway, and has a significant impact on the patient's condition. Tumors have extremely high demands for a large supply of amino acids to maintain the energy required for growth. The anabolic and catabolic metabolism of glutamine, serine, and glycine is considered an important metabolic regulatory mechanism to support the growth of tumor cells. In addition, abnormalities in tryptophan and arginine metabolism are also a significant feature in the tumor immune microenvironment, and increasing the content of serine and glycine can slow down the growth rate of tumors. Arginine metabolites are involved in processes related to RNA metabolism and regulate the function of specific T cells, thus promoting tumor growth and metastasis. However, there have been no relevant reports on the diagnostic and therapeutic potential of amino acid metabolism-related biomarkers in PCPG. Considering the interaction between amino acid metabolism and the tumor immune microenvironment, it is of great significance to explore the pathological mechanisms by which amino acid metabolism-related genes play a role in PCPG and their potential applications in treatment. Therefore, we studied the application of amino acid metabolism-related genes in the diagnosis, treatment, and prognosis of pheochromocytoma through the GEO and TCGA databases and combined with the clinical information we collected. Summary of the Invention

[0004] (1) Technical problems to be solved

[0005] In view of the deficiencies of the prior art, the present invention provides the application of the construction of biomarkers and risk models based on amino acid metabolism-related genes in predicting the diagnosis, treatment, and prognosis of pheochromocytoma.

[0006] (2) Technical solutions

[0007] To achieve the above invention purposes, the present invention is realized through the following technical solutions:

[0008] In the first aspect, the present invention provides a biomarker based on amino acid metabolism-related genes, and the biomarker comprises 6 gene tags, namely DDC, SYT11, GCLM, PSMB7, TYRO3, and / or AGMAT.

[0009] In the second aspect, the present invention provides the application of the above biomarker in the preparation of reagents for predicting the diagnosis, treatment, and / or prognosis of pheochromocytoma, and the biomarker includes DDC, SYT11, GCLM, PSMB7, TYRO3, and / or AGMAT.

[0010] Specifically, the diagnostic, therapeutic, and / or prognostic prediction reagent comprises a culture system expressing the DDC, SYT11, GCLM, PSMB7, TYRO3, and / or AGMAT gene or the protein encoded thereby.

[0011] Specifically, in the application of the above-mentioned biomarker in the preparation of a diagnostic prediction reagent for pheochromocytoma, the expressions of DDC and SYT11 are up-regulated in tumor tissues, while the expressions of GCLM, PSMB7, TYRO3, and AGMAT are down-regulated in tumor tissues.

[0012] The above-mentioned pheochromocytoma refers to a malignant tumor originating from the adrenal medulla.

[0013] In a third aspect, the present invention further provides a risk model for the selection of treatment methods for pheochromocytoma. The risk model is constructed using the expression levels of DDC, SYT11, GCLM, PSMB7, TYRO3, and / or AGMAT as parameters, and the coefficients of the above factors are used to calculate the risk score. The risk score = DDC * 0.051 + SYT11 * 0.586 + GCLM * (-0.758) + PSMB7 * (-0.658) + TYRO3 * 1.313 + AGMAT * 0.001.

[0014] Specifically, the level of the above-mentioned risk score is used as an indicator for the selection of immunotherapy methods. The higher the risk score value, the better the predicted immunotherapy effect, and the immunotherapy method can be selected for treatment.

[0015] (III) Beneficial effects

[0016] The present invention provides the application of the construction of biomarkers and risk models based on amino acid metabolism-related genes in predicting the diagnosis, treatment, and prognosis of pheochromocytoma. Our research found that the amino acid metabolism-related genes DDC and / or SYT11 showed high expression levels in pheochromocytoma, which had obvious effects on the diagnosis and prognosis of pheochromocytoma. Amino acid-related genes can become new targets for the treatment of pheochromocytoma. By combining the GEO database and the TCGA database with pheochromocytoma patient samples, a risk model that can be used for clinical auxiliary judgment of patient diagnosis and treatment was constructed.

[0017] We constructed a risk model using the expression levels of DDC, SYT11, GCLM, PSMB7, TYRO3, and / or AGMAT as parameters, and used the coefficients of the above factors to calculate the risk score. The risk score = DDC * 0.051 + SYT11 * 0.586 + GCLM * (-0.758) + PSMB7 * (-0.658) + TYRO3 * 1.313 + AGMAT * 0.001. The model provided by the present invention uses the Risk Score value to evaluate the treatment method of pheochromocytoma, and divides pheochromocytoma patients into a low-risk factor group and a high-risk factor group according to the median of the risk value. According to the model, different treatment methods are adopted for pheochromocytoma patients with different risks. We found that in pheochromocytoma, the proportion of immune cells was significantly negatively correlated with the risk score of PCPG patients. The immune infiltration level was higher in the low-risk factor group, and the high-risk factor group had a better response to immunotherapy. The risk model provided by the present invention has strong predictive ability for the selection of immunotherapy methods, can be used as an important indicator for the selection of immunotherapy methods, can accurately evaluate the treatment methods of patients, stratify patients, identify patients with high-risk factors at an early stage, and provide an important basis for the precise formulation of treatment strategies and individualized treatment decisions.

[0018] In addition, we also found that genes related to amino acid metabolism have important potential in diagnosis and prognosis, especially in the field of diagnosis and treatment of pheochromocytoma (PCPG). Among them, the SYT11 gene can accurately evaluate the prognostic risk of patients and support the formulation of more effective treatment plans based on the individual gene expression characteristics and survival information, which provides valuable support for the decision-making process of doctors and is expected to significantly improve the survival rate and quality of life of patients. The above research has broad clinical application prospects and will have a positive and profound impact on the health and well-being of PCPG patients. Brief Description of the Drawings

[0019] Figure 1 : Differential analysis of gene expression in pheochromocytoma and normal adrenal tissues was performed using the R package limma to identify differential genes related to amino acid metabolism differences (|logFC|>1, P<0.05); (A) Volcano plot showing differentially expressed genes between normal adrenal tissue and PCPG samples in the gene chip data of the GEO database; (B) Intersection genes of differentially expressed genes and genes related to amino acid metabolism.

[0020] Figure 2 : Genes closely related to pheochromocytoma were screened by weighted gene co-expression network analysis; (A) Dendrogram between all samples; (B) Clustering dendrogram, different colors represent different gene templates; (C) Heat map of the relationship between gene co-expression modules and pheochromocytoma phenotypes; (D) Scatter plot of the brown module.

[0021] Figure 3 : Construction and validation of the risk model; (A) Distribution of coefficients in the LASSO regression model; (B) Feature selection adjustment in the LASSO model; (C, D) Differential expression levels of genes in tumor and normal tissues in the GEO and TCGA cohorts; *P value < 0.05, **P value < 0.01, ***P value < 0.001.

[0022] Figure 4 : Expression patterns and potential biological functions of differential genes related to amino acid metabolism in PCPG; (A, B) GO (biological process, cellular component, molecular function) and KEGG enrichment analyses were performed on 292 differential genes related to amino acid metabolism; (C) Disease Ontology (DO) pathway enrichment analysis was performed on genes related to amino acid metabolism in PCPG.

[0023] Figure 5 : Immunohistochemical staining results of tissue samples from patients with pheochromocytoma who underwent adrenal or retroperitoneal mass resection; (A, B) The expression of the amino acid metabolism-related gene DDC in PCPG tissues was evaluated by immunohistochemical staining; (C, D) The expression of the amino acid metabolism-related gene SYT11 in PCPG tissues was evaluated by immunohistochemical staining.

[0024] Figure 6 : Immunocyte infiltration analysis of the low-risk factor group and the high-risk factor group; (A, B) The immunocyte infiltration of the low-risk factor group and the high-risk factor group in PCPG patients was evaluated by the ssGSEA algorithm; (C, D) The absolute abundances of 8 immunocytes and 2 stromal cells between the low-risk factor group and the high-risk factor group in the GEO and TCGA cohorts were evaluated by the MCPCOUNTER algorithm; *P value < 0.05, **P value < 0.01, ***P value < 0.001.

[0025] Figure 7 : Immunoscore analysis of the low-risk factor group and the high-risk factor group; (A, B) The relationship between the risk score and the immunoscore in the GEO and TCGA cohorts was analyzed by the Estimate algorithm; *P value < 0.05, **P value < 0.01, ***P value < 0.001.

[0026] Figure 8: Gene set enrichment analysis of the low-risk factor group and high-risk factor group; (A, B) Heatmaps respectively show the overall characteristics of 29 immune features in the low-risk factor group and high-risk factor group in the GEO and TCGA cohorts. (C, D) Violin plots respectively show the differences in 29 immune features between the low-risk factor group and high-risk factor group in the GEO and TCGA cohorts; *P value < 0.05, **P value < 0.01, ***P value < 0.001.

[0027] Figure 9 : The risk model predicts the treatment outcomes and potential active drugs for PCPG patients; (A, B) Prediction of the response to immunotherapy in the low-risk factor group and high-risk factor group of PCPG patients in the GEO and TCGA cohorts. (C) Prediction of potential drugs based on differentially expressed genes related to amino acid metabolism; *P value < 0.05, **P value < 0.01, ***P value < 0.001.

[0028] Figure 10 : Evaluate the performance of genes related to amino acid metabolism in the diagnosis and prognosis of PCPG; (A, B) Receiver operating characteristic curves of DDC and SYT11 genes in the test set GEO cohort. (C, D) Receiver operating characteristic curves of DDC and SYT11 genes in the validation set TCGA cohort. (E, F) Receiver operating characteristic curves of DDC and SYT11 genes in the validation set GEO cohort. (G, H) The expression of SYT11 and AGMAT in the validation set TCGA cohort is closely related to the prognosis of PCPG patients. Detailed implementation manners

[0029] To more clearly clarify the purpose, technical solution and advantages of the present invention, we will describe the technical solution of the present invention in detail in combination with actual embodiments. It should be clear that the described embodiments only represent some embodiments of the present invention, rather than all. Those of ordinary skill in the art can obtain other embodiments based on the embodiments of the present invention without creative efforts, and these embodiments are also included in the protection scope of the present invention.

[0030] 1 Analysis of data sources and differential gene expression analysis

[0031] The RNA sequencing data of patients with pheochromocytoma were sourced from the TCGA database platform (https: / / portal.gdc.cancer.gov / ) and the GEO database (https: / / www.ncbi.nlm.gov / geo). Genes related to amino acid metabolism were searched and downloaded in the Msigdb database (http: / / www.gsea-msigdb.org / gsea / index.jsp) using the keyword "amino acid metabolism".

[0032] Differential analysis of gene expression in pheochromocytoma and normal adrenal tissues was performed using the R package limma (|logFC| > 1, P < 0.05), and the results are as Figure 1 shown. The results showed that a total of 292 differentially expressed genes related to amino acid metabolism were dysregulated in cancer. Specifically, the differentially expressed genes related to amino acid metabolism included 131 upregulated genes and 161 downregulated genes.

[0033] 2 Screening genes closely related to pheochromocytoma through weighted gene co-expression network analysis

[0034] We performed weighted gene co-expression network (WGCNA) analysis using the R package WGCNA to explore key genes contributing to the biological differences among patient groups. The absolute median difference (MAD) was used as a robust statistic to screen genes, which is more adaptable to outliers in the dataset than the standard deviation. In addition, hierarchical clustering analysis was performed on the samples to remove outlier samples ( Figure 2 A). Then, dynamic tree cut analysis was used to construct the network and identify modules ( Figure 2 B). Correlation analysis was performed using the scale-free network module and the pheochromocytoma phenotype. The brown module feature with the highest correlation with PCPG was selected from the correlation heatmap (|cor| = 0.77, P-value = 1e-18) ( Figure 2 C). In the brown module, module membership and gene significance were highly correlated ( Figure 2 D). In summary, these results indicate a close association between PCPG and the brown module, and the genes contained in this module may play an important role in the development of PCPG.

[0035] 3 Construction and validation of the risk model

[0036] To further identify the optimal candidate genes, we used the Least Absolute Shrinkage and Selection Operator (LASSO) regression analysis algorithm to screen out six optimal variables from the differential gene variables related to amino acid metabolism in the above-mentioned brown module through ten-fold cross-validation. When the model reached the minimum value of λ (lambda), an optimal prediction model containing six genes with non-zero coefficients (DDC, SYT11, GCLM, PSMB7, TYRO3, and AGMAT) was constructed ( Figure 3 A-B). The coefficients of the above six optimal variables were used to calculate the risk score: Risk score = DDC * 0.051 + SYT11 * 0.586 + GCLM * (-0.758) + PSMB7 * (-0.658) + TYRO3 * 1.313 + AGMAT * 0.001. The risk score had a strong predictive ability for the choice of immunotherapy method, and this risk model could be used as an important indicator for the choice of immunotherapy method. Compared with the normal group, the expressions of DDC and SYT11 were significantly up-regulated in the tumor group, while the expressions of GCLM, PSMB7, TYRO3, and AGMAT were significantly down-regulated in the tumor group ( Figure 3 C). R packages (LASSO, ggpbur, rms, and ggplot) were used in this process. In addition, the expression levels of these six genes were verified in the TCGA dataset ( Figure 3 D). In summary, the six screened genes related to amino acid metabolism play a key role in the pathogenesis of PCPG disease.

[0037] 4 Expression patterns and potential biological functions of differential genes related to amino acid metabolism;

[0038] GO and KEGG enrichment analyses were performed on the differential genes related to amino acid metabolism using the R packages clusterProfiler and org.Hs.eg.db. The results are shown in Figure 4 A-B. It was found that these genes were significantly correlated with tyrosine modification and phosphorylation, cellular amino acid metabolic processes, amino acid transport, and glutathione metabolic processes. With glutathione transferase activity, amino acid transmembrane transporter activity, glutathione metabolism, cysteine and methionine metabolism, EGFR tyrosine kinase inhibitor resistance, tyrosine metabolism, and amino acid biosynthesis. DO enrichment analysis was performed using the R package DOSE, and it was found that these genes were mainly involved in renal failure, kidney disease, ischemia, and urinary tract diseases ( Figure 4 C).

[0039] 5 Verification of the expressions of amino acid metabolism-related genes DDC and SYT11 in PCPG

[0040] By performing immunohistochemical staining on PCPG samples and adjacent non-tumor tissues, our research results showed that compared with adjacent non-tumor tissues, the expression of DDC (dopa decarboxylase) in PCPG tissues was significantly upregulated ( Figure 5 A, 5B). In addition, the expression level of SYT11 (synaptotagmin 11) was also significantly upregulated in PCPG tissues compared with adjacent non-tumor tissues ( Figure 5 C, 5D). This indicates that DDC and SYT11 may play key roles in the development of PCPG, suggesting that they may serve as potential biomarkers or therapeutic targets.

[0041] 6 Enrichment differences in immune cell infiltration between the low-risk factor group and the high-risk factor group

[0042] Based on the ssGSEA algorithm, the overall infiltration of 28 immune cells in pheochromocytoma was determined. Differential analysis of the immune cell infiltration levels in the low-risk factor group and the high-risk factor group was performed, and it was observed that the infiltration levels of various immune cell types in the low-risk factor group were significantly higher, including activated B cells, activated CD8 T cells, central memory CD4 T cells, effector memory CD8 T cells, macrophages, mast cells, NK cells, neutrophils, and TFH cells ( Figure 6 A). Further verification by the MCPCOUNTER algorithm showed higher enrichment scores of T cells, cytotoxic lymphocytes, NK cells, monocytic lineage, myeloid dendritic cells, and neutrophils in the low-risk factor group. In addition, the ssGSEA algorithm ( Figure 6 B) and the MCPCOUNTER algorithm ( Figure 6 D) were also used in the validation set TCGA cohort to verify our findings. Overall, in pheochromocytoma, the dysregulation of the tumor immune microenvironment is closely related to the level of the risk score.

[0043] 7 Relationship between immune score and risk score in PCPG

[0044] Immune analysis based on the Estimate algorithm showed a negative correlation between the risk score and the immune level in the test set GEO cohort ( Figure 7 A). The Estimate algorithm was used in the validation set TCGA cohort to verify the negative correlation between the risk score and the immune level in PCPG ( Figure 7 B).

[0045] 8 Gene set enrichment analysis of the low-risk factor group and the high-risk factor group

[0046] We performed gene set variation analysis (GSVA) enrichment analysis using the R package GSVA to explore the immune-related mechanisms between the low-risk factor group and the high-risk factor group in PCPG patients. Compared with the high-risk factor group, the immune-related pathways were significantly more enriched in the low-risk factor group, including B cells, costimulatory receptors, effector cell trafficking, MHCII, and T cells ( Figure 8 A, 8C). In the validation set TCGA cohort, differential analysis of immune-related pathways between the two groups found that the enrichment in immune pathways was significantly reduced in the high-risk group ( Figure 8 B, 8D), indicating a weakened immune response of tumor cells in the high-risk factor group. Therefore, targeting amino acid metabolism-related genes may represent a potential treatment strategy.

[0047] 9 Using the risk model to predict the treatment methods and potential active drugs for patients with pheochromocytoma

[0048] To identify patients who may benefit from immunotherapy, we used the TIDE algorithm (TIDE, http: / / tide.dfci.harvard.edu / ) to predict the immunotherapy response of different risk groups. In the test set GEO cohort, the high-risk factor group had a significantly better response to immunotherapy than the low-risk group ( Figure 9 A). Consistent results were also observed in the validation set TCGA cohort, with the high-risk group benefiting more from immunotherapy ( Figure 9 B). Based on the Connectivity Map (cMAP, https: / / clue.io / ) database, we used the "PharmacoG" R package to identify potentially active drugs in PCPG to discover potential drugs. The prediction of drugs was performed using 292 DEGs related to amino acid metabolism as input genes. The results showed that the top five drugs with enrichment scores were clofibrate, butein, fasudil, imatinib, and irinotecan ( Figure 9 C). The results indicate that these drugs may have better clinical efficacy in PCPG patients. Using this risk model can accurately predict the treatment methods of patients, which can help doctors identify high-risk factor group patients early and take appropriate intervention measures.

[0049] 10 Evaluating the efficacy of amino acid metabolism-related differential genes in the diagnosis of patients with pheochromocytoma

[0050] The diagnostic performance of DDC and SYT11 was evaluated by calculating the area under the receiver operating characteristic curve (ROC) of DDC and SYT11 in the test set and validation set using the R package pROC. In the test set GEO dataset, the AUC values of DDC and SYT11 were 0.992 and 0.994, respectively ( Figure 10 A, 10B). In the validation set TCGA dataset, the AUC value of DDC was 0.991 and the AUC value of SYT11 was 0.996 ( Figure 10 C, 10D), showing excellent diagnostic accuracy. In the validation set GEO dataset 2, the AUC values of DDC and SYT11 were 0.84 and 0.67, respectively, indicating good diagnostic performance ( Figure 10 E, 10F). In addition, survival analysis performed using the R package survival showed that high expression of SYT11 in PCPG patients was significantly associated with poor prognosis ( Figure 10 G). The area under the ROC curve (AUC) for predicting the 1-year survival rate of pheochromocytoma patients by the amino acid metabolism-related gene SYT11 was 0.68, the AUC for the 3-year survival rate of pheochromocytoma patients was 0.71, and the AUC for the 5-year survival rate of pheochromocytoma patients was 0.78 ( Figure 10 H). In summary, DDC and SYT11 have shown excellent potential in the diagnosis and treatment of pheochromocytoma. Notably, the unique ability of SYT11 to accurately assess the prognostic risk of patients and tailor more effective treatment plans based on individual gene expression characteristics and survival information is expected to improve the survival rate and quality of life of patients.

[0051] The present invention constructs a pheochromocytoma risk model based on amino acid metabolism-related genes, which can more accurately evaluate the treatment methods of patients and provide an important basis for individualized treatment decisions. The importance of this risk model lies in its ability to stratify patients and early identify patients in the high-risk factor group, thus contributing to the formulation of precise treatment strategies. In addition, amino acid metabolism-related genes have important potential in diagnosis and prognosis, especially in the diagnosis and treatment of pheochromocytoma (PCPG), providing new targets for the treatment of pheochromocytoma. Notably, the unique ability of the gene SYT11 to accurately assess the prognostic risk of patients and support the formulation of more effective treatment plans based on individual gene expression characteristics and survival information provides valuable support for the decision-making process of doctors and is expected to significantly improve the survival rate and quality of life of patients. These important findings have broad clinical application prospects and will have a positive and profound impact on the health and well-being of PCPG patients.

[0052] The above embodiments are only used to illustrate the technical solutions of the present invention and are not intended to limit them. Although we have described these embodiments in detail, those of ordinary skill in the art should understand that it is still possible to modify the technical solutions described in these embodiments or replace some of the technical features therein, and such modifications or replacements will not change the essence of the corresponding technical solutions nor deviate from the spirit and scope of the technical solutions of the embodiments of the present invention.

Claims

1. Use of a reagent for detecting the amino acid metabolism-related gene SYT11 in the preparation of a reagent for diagnosing pheochromocytoma.

Images