CN113462776B - m 6 Application of A modification-related combined genome in prediction of immunotherapy efficacy of renal clear cell carcinoma patient - Google Patents

Abstract

The invention relates to the technical field of medical biological detection, and provides a new application of a combined genome of HNRNPA2B1 and ALKBH5, in particular to an application in preparation of a renal clear cell carcinoma immunotherapy curative effect prediction reagent or a kit. Also provides a kit for predicting the curative effect of the immunotherapy of the renal clear cell carcinoma and a system for predicting the curative effect of the immunotherapy of the renal clear cell carcinoma. The genome of the present invention is derived from renal clear cell carcinoma m ⁶ A modification pattern differential expression pattern, m ⁶ The discovery of the A modification related gene combination model provides a brand-new strategy for predicting the immunotherapy curative effect of the renal clear cell carcinoma patient, is beneficial to guiding a clinician to implement an individualized accurate treatment strategy, improves the survival rate of the patient, and has important guiding significance for the immunotherapy application of the renal clear cell carcinoma patient.

Description

m6A modification-associated combined genome in predicting immunotherapy in patients with clear cell renal cell carcinoma application in effect

technical field

The present invention belongs to the technical field of medical biological detection, and relates to the use of the m ⁶ A modification-related joint genome of HNRNPA2B1 and ALKBH5 as a marker, specifically the use of the joint genome in the preparation of a kit for predicting curative effect of renal clear cell carcinoma immunotherapy and a predictive system application.

Background technique

Renal cell carcinoma is one of the most common malignant tumors of the genitourinary system, accounting for about 5% of all adult male new cases and 3% of new female cases. According to statistics, in 2019, there were about 73,820 new cases of kidney cancer and 14,770 deaths in the United States. There are about 66,800 new kidney cancer patients in my country every year, ranking second in the incidence of urinary system tumors. Clear cell renal cell carcinoma (ccRCC) is the most common pathological type of renal cancer with a high degree of malignancy, accounting for about 70-85% of all renal cancer patients, and about 25-30% of ccRCC patients are diagnosed immediately metastases, and the 5-year survival rate for metastatic ccRCC is only 32%. For tumors with clinically limited stages, nephron-sparing surgery or radical nephrectomy is still the main treatment method. Further postoperative cytokine or individualized precision adjuvant therapy can reduce the rate of tumor recurrence and metastasis and improve the long-term survival of patients Rate. At present, the first-line treatment drugs for advanced RCC are mainly tyrosine kinase inhibitors (TKIs) targeting vascular endothelial growth factor receptor (VEGFR), such as pazopanib, Nitinib, cabozantinib, axitinib, etc. Although anti-angiogenic drugs can inhibit tumor proliferation to a certain extent and can significantly prolong the survival of low-risk ccRCC patients, the side effects of drugs are obvious and the overall curative effect is not good. The objective response rate of treatment is only less than 30%, and the prolonged median overall The survival time is also less than 12 months. In addition, even patients who initially respond to treatment will experience disease progression after a period of time, at which point most patients will lack subsequent effective treatment options.

In recent years, new immunotherapies represented by PD-1/PD-L1 and CTLA4 inhibitors have risen rapidly in the field of renal cancer treatment, and have shown encouraging curative effects on advanced refractory patients. Since the FDA approved nivolumab in 2015 based on the Checkmate025 study for patients with advanced renal cell carcinoma who had previously received anti-angiogenic drug therapy, ASCO GU published the 5-year follow-up results of the CheckMate025 study in 2020, showing that nivolumab The 5-year survival rate of monoclonal antibody second-line therapy is as high as 26%, highlighting the survival advantage of immunotherapy. Subsequently, immune checkpoint inhibitors gradually moved from second-line to first-line. At present, PD-1 monoclonal antibody combined with CTLA-4 monoclonal antibody and PD-1/PD-L1 monoclonal antibody combined with anti-angiogenic drugs have been approved by the FDA for the first-line treatment of advanced renal cancer. Approved, the treatment of advanced kidney cancer has ushered in a new chapter. Immune checkpoint inhibitors combined with TKIs play a role in inducing the normalization of anti-tumor immunity, inhibiting the main signaling pathways in the development of advanced renal cancer, and regulating the tumor microenvironment (Tumor microenvironment, TME). A deep understanding of interactions. With the deepening of research, more evidence shows that not only the efficacy of immunotherapy depends on the activation of the tumor immune microenvironment, but also the efficacy of traditional treatments such as targeted therapy depends on the strength of the body's anti-tumor immune response. Cells and molecules in the TME are in a process of dynamic change, reflecting the evolutionary nature of cancer, and jointly promoting immune escape, growth, and metastasis of tumors. Exploring the underlying mechanism of TME-driven tumorigenesis and development is of great significance for developing potential methods of cancer treatment, improving the efficiency of various existing treatments, and discovering new precise targets for RCC treatment.

The most common chemical modifications of RNA include N6-adenylate methylation ( ^m6A ), N1-adenylate methylation (m1A), cytosine hydroxylation (m5C), etc. m ⁶ A is the most abundant methylation modification of mRNAs in eukaryotic cells, and plays an important role in the chemical modification of mRNAs, miRNAs and lncRNAs. Recently, several studies have revealed a specific correlation between TME-infiltrated immune cells and ^m6A modification, which cannot be explained by RNA degradation mechanisms. Dali et al. reported that YTHDF1 binds to transcripts encoding lysosomal proteases modified by m ⁶ A methylation, increasing the translation efficiency of lysosomal cathepsins in dendritic cells (DCs), while inhibiting cathepsins in DCs Significantly enhanced its ability to cross-express tumor antigens, thereby enhancing the anti-tumor response of tumor-infiltrating CD8+ T cells. Inhibition of YTHDF1 also enhanced the efficacy of anti-PD-L1 blockade. The study by Huamin et al. revealed that m ⁶ A modification mediated by METTL3 promotes the activation and maturation of DCs. METTL3-specific depletion results in reduced expression of costimulatory molecules CD80 and CD40, reducing the ability to stimulate T cell activation. There is a close relationship between m ⁶ A and tumor immunity. Exploring the differences in m ⁶ A modification patterns may provide powerful help for precise immunotherapy of renal cancer.

In the last decade, high-throughput techniques combined with bioinformatics analysis have been widely used to detect comprehensive mRNA expression levels, which helps to identify differentially expressed genes (DEGs) and explore genes closely related to ccRCC immune microenvironment components. landmark.

Contents of the invention

The present invention is further developed on the basis of the above research, and the purpose is to provide a biomarker for predicting the efficacy of immunotherapy in patients with renal clear cell carcinoma. The purpose of the present invention is also the new use of the combined genome related to the m ⁶ A modification of HNRNPA2B1 and ALKBH5 , that is, the application in the immunotherapy efficacy prediction kit and prediction system for renal clear cell carcinoma immunotherapy.

The inventor downloaded the gene expression and survival data of patients with clear cell renal cell carcinoma from TCGA through complex bioinformatics screening in the early stage, and then the inventor extracted the expression of 21 m ⁶ A regulatory factors based on the expression matrix of clear cell renal cell carcinoma, and Three potential m ⁶ A modification patterns (Cluster1, 2, 3) were identified using consensus clustering, and it was found that the m ⁶ A modification pattern in Cluster 3 was associated with poor prognosis of patients and higher expression of immune checkpoints in tumor samples Significant correlation, suggesting that this type of m ⁶ A modification pattern may be used to predict the response to immunotherapy.

Further, the inventor constructed a transcriptome classifier based on this. Due to the lack of usable kidney cancer immunotherapy cohort data, the inventor used the classifier in the published bladder cancer IMvigor210 cohort and found that the classifier can effectively Predict patient response to immunotherapy. Applying binary logistic regression, the inventors simplified the classifier into the formula: prediction score = 1.889*HNRNPA2B1 expression level-0.451*ALKBH5 expression level. HNRNPA2B1 and ALKBH5 were discovered for the first time as biomarkers for predicting the efficacy of immunotherapy in patients with clear cell renal cell carcinoma.

The first aspect of the present invention provides the use of the combined genome of HNRNPA2B1 and ALKBH5 as a biomarker for predicting the efficacy of immunotherapy in patients with clear cell renal cell carcinoma.

The second aspect of the present invention provides the application of the above combined genome in the preparation of reagents or kits for predicting the curative effect of renal clear cell carcinoma immunotherapy, wherein the predictive reagent is a combination of reagents for detecting the relative expression levels of HNRNPA2B1 and ALKBH5 in biological samples; the predictive reagent The kit contains a combination of reagents for detecting the relative expression levels of HNRNPA2B1 and ALKBH5 in biological samples.

Preferably, the reagent combination is a reagent combination for detecting the mRNA expression level of the above-mentioned genes, and the reagent combination includes PCR primers with detection specificity for the above-mentioned genes, and the sequences of the primers are shown in the following table as SEQ ID NO.1-4.

Preferably, the biological sample is a section of a tumor specimen surgically resected from a patient with renal clear cell carcinoma.

The third aspect of the present invention provides a kit for predicting the curative effect of renal clear cell carcinoma immunotherapy, which is composed of a reverse transcription system, a primer system and an amplification system. The primer system includes PCR primers as shown in SEQ ID NO.1-4.

The fourth aspect of the present invention provides a system for predicting the curative effect of immunotherapy for clear cell renal cell carcinoma, including a prediction kit and an immune subgroup classification model installed on a terminal carrier. As mentioned above, the prediction kit detects the relative expression levels of the marker genes, and the immune subgroup classification model predicts and scores according to the following formula, and determines whether the immune typing of the current sample belongs to the immune rejection type or the immune desert according to the score Type, prediction score = 1.889*HNRNPA2B1 expression level-0.451*ALKBH5 expression level.

Preferably, the immune subgroup classification model is constructed according to binary logistic regression, and binary logistic analysis is used to predict the significant correlation between the prediction score and the patient's immunotherapy.

The fifth aspect of the present invention provides a method for predicting the curative effect of immunotherapy for clear cell renal cell carcinoma using the above immunotherapy curative effect prediction system, which specifically includes the following steps:

(a) using the reagents in the kit to perform reverse transcription and amplification on the tumor sample to obtain the mRNA expression level of each gene;

(b) The predictive score of immunotherapy for clear cell renal cell carcinoma is calculated according to the following formula: predictive score = 1.889*HNRNPA2B1 expression level-0.451*ALKBH5 expression level, and according to the score, determine whether the immune typing of the current sample belongs to the immune rejection type or the immune desert type.

The immune-exclusive Cluster3 regulatory subtype was significantly associated with poorer survival and was also significantly associated with higher T stage. Compared with cluster1/2, the gene expression profile of cluster3 is significantly enriched in biological processes such as steroid metabolism, synaptic membrane, and neuroactive ligand-receptor interactions, but such patients can benefit greatly from immunotherapy many.

Beneficial protection and effect of the present invention are as follows:

The gene combination of the present invention is derived from the gene combination involved in the formation of the m ⁶ A modified transcriptional subtype of renal clear cell carcinoma. The results of the IMvigor210 cohort and the FUSCC real world cohort verification are significant, and both can effectively predict the response of patients to immunotherapy . It shows that the expression levels of HNRNPA2B1 and ALKBH5 have a high value in predicting the efficacy of immunotherapy, which may help to apply immune checkpoint inhibitors more accurately in renal cancer patients.

Therefore, the discovery of this gene combination model provides a new strategy for predicting the efficacy of immunotherapy in patients with clear cell renal cell carcinoma. The postoperative survival rate of patients also has important guiding significance for postoperative follow-up monitoring and sequential treatment management of patients with clear cell renal cell carcinoma.

In terms of technology, the detection of the two gene expression levels is essentially a quantitative PCR detection of tissue samples, which has the characteristics of simple operation, sensitive detection, good specificity, and high repeatability, and has been increasingly used in clinical testing techniques. middle. This technology has been proved to be a highly sensitive and accurate detection method in modern experimental diagnostics, and the test technology is very mature. And we use the standard curve quantification method in this technology, which can accurately quantify the characteristic nucleic acid molecules in various samples.

Description of drawings

Figure 1 is the process of distinguishing the expression of m ⁶ A regulators in tumor tissue and normal tissue and identifying m ⁶ A transcriptional subtypes in renal clear cell carcinoma: (A) Differences in the expression of 21 m ⁶ A regulators in tumor tissue and adjacent normal tissues Distribution boxplot, the distribution of most m ⁶ A regulators is significantly different in tumor tissue and adjacent normal tissue; (B) Correlation heat map of the expression levels of 21 m ⁶ A regulators in tumor tissue, which can be It was found that the expressions of various m ⁶ A regulatory factors were positively correlated with each other, such as the correlation coefficient between CBLL1 and YTHDF3 was 0.58, the correlation coefficient between METTL14 and YTHDC1 was 0.66, and the correlation coefficient between METTL14 and LRPPRC was 0.6; (C) Consensus clustering shows , the effect of dividing TCGA-CCRCC into three m ⁶ A regulatory subclasses was the best; (D) Principal component analysis showed that the gene expression patterns of the three m ⁶ A regulatory subclasses were significantly different.

Figure 2 explores potential clinical phenotypic differences between m ⁶ A regulatory subclasses: (A) heat map of gene expression of m ⁶ A regulatory subclasses integrated with clinical information, which shows that there are significantly more death outcomes in cluster3 ; (B) Survival analysis showed that the overall survival of patients in cluster2 was similar to that of cluster1 and cluster2, while the overall survival of patients in cluster3 was significantly worse than that of patients in cluster1 and cluster2 (p=0.002); (C) Univariate regression analysis showed , HNRNPA2B1, ZC3H13, LRPPRC, METLL14, YTHDC1 and other m ⁶ A regulatory factors were significantly correlated with the overall survival of patients.

Figure 3 explores potential differences in biological function, immune checkpoint expression, and immune cell infiltration components between m ⁶ A regulatory subclasses (cluster3 compared to cluster1&2): (A) compared to cluster1 and cluster2, cluster3 Regulatory subtypes of RCC were significantly enriched in biological processes such as steroid metabolism, synaptic membrane function, and receptor ligand activity; (B) differential analysis showed that compared with cluster1 and cluster2, in cluster3, CTLA4, PDCD1, TNFSF14, and LAG3 all showed significant high expression (log ₂ FoldChange>1, p<0.05), suggesting that cluster3 may be associated with the immunosuppressive microenvironment; (C) CIBERSORT algorithm was used to predict the microenvironment of renal cancer The abundance of immune cells, the results showed that compared with cluster1 and cluster2, cluster3 renal carcinoma had significantly increased CD8 positive T cells (p<0.001) and Tfh cell infiltration (p<0.001), while M2 macrophages Infiltration decreased significantly in cluster3.

Figure 4 uses the Kaplan-Meier method to analyze the survival results of cluster1&2 and cluster3, showing that the overall survival of cluster3 patients is significantly worse than that of non-cluster3 patients.

Figure 5 Nomogram of correlation between predictive score and patient response to immunotherapy. The four immune checkpoint molecules, PDL1, PD1, CTLA4, and LAG3, were included in the nomogram construction, and the results showed that the prediction score could be effectively used to predict the patient's response to immunotherapy.

Figure 6 is the receiver operating characteristic curve (ROC) of predicting patient response to immunotherapy in the IMvigor210 cohort using the predictive score. The results show that the area under the curve (AUC) is 0.65, p<0.001, indicating that the predictive score constructed in this project can be compared with Good predictor of patient response to immunotherapy.

Figure 7 shows the immune microenvironment of the primary tumor specimens of clear cell renal cell carcinoma with high predictive scores in patients who have received immunotherapy;

Figure 8 shows the immune microenvironment of the primary tumor specimens of clear cell renal cell carcinoma with low predictive scores in patients who have received immunotherapy;

Figure 9 shows the ROC curves for evaluating response to immunotherapy in 98 patients in the FUSCC cohort using the predictive score.

Detailed ways

The implementation of the present invention will be described in detail below in conjunction with the accompanying drawings and embodiments of the present invention. The following embodiments are implemented under the premise of the technical solution of the present invention, and detailed implementation methods and specific operation processes are provided. However, the present invention The scope of protection is not limited to the examples described below.

The reagents and raw materials used in the present invention are commercially available or can be prepared according to literature methods. For the experimental methods without specific conditions indicated in the following examples, usually follow the conventional conditions or the conditions suggested by the manufacturer.

In the following examples, tumor tissue samples from patients with clear cell renal cell carcinoma were obtained from the Cancer Hospital Affiliated to Fudan University, and were diagnosed as clear cell renal cell carcinoma by pathologists.

The present invention aims to explore and investigate multi-cohort based gene expression profiles for predicting the efficacy of immunotherapy. The RNA-seq data of 602 clear cell renal cell carcinoma tumors and adjacent normal tissues (72 adjacent normal tissues and 530 tumor tissues) were downloaded from The Cancer Genome Atlas (TCGA). Then the expression data of 21 m ⁶ A regulators were extracted from RNA-seq, and the expression differences of m ⁶ A regulators in cancer and paracancerous cells and the interaction relationship between them were further explored. Three types of m ⁶ A transcriptional regulatory subtypes, cluster1\2\3, were identified using the consensus clustering method.

The prognostic value of 21 m ⁶ A regulatory factors was evaluated by univariate regression method. Univariate regression analysis showed that IGF2BP1, HNRNPA2B1, etc. were significantly correlated with poor prognosis, while CBLL1, FMR1, etc. were significantly correlated with good prognosis. The analysis found that the survival of cluster3 is far worse than that of cluster1\2. The expressions of all immune checkpoint-related genes were extracted, and it was found that PDCD1, CTLA4, LAG3, etc. were significantly highly expressed in cluster 3. It suggested that m ⁶ A regulatory subtypes of RCC may be related to the immune microenvironment. The CIBERSORT algorithm was further used to evaluate the composition of tumor-infiltrating lymphocytes, and it was found that the content of immune cells such as tumor-infiltrating CD8-positive T cells in cluster 3 and cluster 1/2 was significantly different.

Next, the reporter used binary logistic regression to construct an m ⁶ A regulatory subtype classifier, and obtained the formula: m6Ascore=1.889*HNRNPA2B1-0.451*ALKBH5. The prediction score of each patient in the IMvigor210 cohort was calculated using the formula. Binary logistic analysis found that the prediction score was significantly correlated with the patient's response to immunotherapy (p <0.0001), and the area under the ROC curve (AUC) of the model was 0.65. The relationship between the predictive score and patient response to immunotherapy was validated in the FUSCC cohort using RT-qPCR.

The present study consisted of three phases: first, we identified possible RCC m ⁶ A modification subtypes in the TCGA RCC cohort using consensus clustering; in the second phase, we explored RCC m ⁶ A modification subtypes. The differences in clinical phenotype and possible biological differences between genotypes were explored, and the differences in immune checkpoint expression and immune cell infiltration were explored; in the third stage, a classifier was constructed based on m ⁶ A modified subtypes, and then the IMvigor210 cohort was utilized And real-world data explored and verified the predictive role of the prediction score for patients' response to immunotherapy. A further detailed description will be given below in conjunction with the accompanying drawings.

Example 1: Construction of Immunotherapy Predictive Score

1. Identification of m ⁶ A transcriptional regulatory subtypes

The RNA-seq data of 602 clear cell renal cell carcinoma tumors and adjacent normal tissues (72 adjacent normal tissues and 530 tumor tissues) were downloaded from The Cancer Genome Atlas (TCGA). Then the applicant extracted the expression data of 21 m ⁶ A regulatory factors from RNA-seq, including 8 writers (METTL3, RBM15, METTL14, RBM15B, KIAA1429, WTAP, CBLL1, ZC3H13), 2 erasers (FTO, ALKBH5 ) and 11 readers (YTHDF1, YTHDF2, YTHDF3, YTHDC1, YTHDC2, HNRNPC, IGF2BP1, HNRNPA2B1, LRPPRC, FMR1, ELAVL1).

Further explored ^the difference in the expression of m ⁶ A regulatory factors between cancer and adjacent normal tissues. There were differences in the expression of 21 m ⁶ A regulatory factors between tumor tissue and adjacent normal tissues. There are significantly different distributions in normal tissues (Figure 1A); at the same time, the interaction relationship between them was explored, and the correlation heat map of the expression levels of 21 m ⁶ A regulators in tumor tissues showed that a variety of m ⁶ A The expressions of the regulatory factors were positively correlated with each other, for example, the correlation coefficient between CBLL1 and YTHDF3 was 0.58, the correlation coefficient between METTL14 and YTHDC1 was 0.66, and the correlation coefficient between METTL14 and LRPPRC was 0.6 (Figure 1B).

Consensus clustering is a method that provides quantitative evidence for determining the number and membership of likely clusters in a dataset. This method has been widely used in cancer genomics, and several authoritative omics studies have previously identified multiple new molecular subclasses of disease through this clustering method. In the present invention, three types of m ⁶ A transcriptional regulatory subtypes cluster1\2\3 were identified using the consensus clustering method, showing that the effect of dividing TCGA-CCRCC into these three types of m ⁶ A regulatory subtypes was the best (Fig. 1C ); principal component analysis showed that the gene expression patterns of the three m ⁶ A regulatory subclasses were significantly different (Fig. 1D).

2. Exploration of clinical phenotypic differences of m ⁶ A transcriptional regulatory subtypes

The patients' gender, age, tumor stage and grade were included together to explore the clinical differences among m ⁶ A regulatory subtypes, showing that there were significantly more death outcomes in cluster3 (Fig. 2A). Further analysis revealed that cluster3 regulatory subtypes were significantly associated with poorer survival and also significantly associated with higher T stage. And using the Kaplan-Meier method to compare the overall survival rates of the three types of m ⁶ A regulatory subtypes, it was found that the survival of cluster3 was far worse than that of cluster1\2; compared to the overall survival of patients in cluster1 and cluster2, the overall survival of patients in cluster3 Patient overall survival was significantly worse than patients in cluster1 and cluster2 (p=0.002) (Fig. 2B). The prognostic value of 21 m ⁶ A regulatory factors was evaluated by univariate regression method. The analysis showed that IGF2BP1, HNRNPA2B1, etc. were significantly associated with poor prognosis, while CBLL1, FMR1, etc. were significantly associated with good prognosis (Fig. 2C).

3. Exploration of potential biological differences of m ⁶ A transcriptional regulatory subtypes

Differential gene analysis was performed on cluster3 and cluster1/2 using the limma algorithm (difference multiple greater than 2, p less than 0.05 was considered significant). Then, functional enrichment analysis was performed on the differential genes. Compared with cluster1/2, the gene expression profile of cluster3 was significantly enriched in biological processes such as steroid metabolism, synaptic membrane, and neuroactive ligand-receptor interactions (Fig. 3A ); in order to explore the relationship between the m ⁶ A modification pattern and the tumor immune microenvironment, the expressions of all immune checkpoint-related genes were extracted, and it was found that compared with cluster1 and cluster2, in cluster3, CTLA4, PDCD1, TNFSF14 and LAG3 all showed significant High expression (log ₂ FoldChange>1, p<0.05), suggesting that cluster3 may be associated with the immunosuppressive microenvironment, suggesting that the m ⁶ A regulatory subtype of renal cancer may be related to the immune microenvironment (Figure 3B); further use CIBERSORT The algorithm predicted the abundance of immune cells in the kidney cancer microenvironment, and the results showed that compared with cluster1 and cluster2, cluster3 kidney cancer had significantly increased CD8-positive T cells (p<0.001) and Tfh cell infiltration (p<0.001) , while the infiltration of M2 macrophages was significantly decreased in cluster3 (Fig. 3C).

4. Construction of m ⁶ A regulatory subtype classifier

We grouped cluster1/2 into one category, and Kaplan-Meier analysis showed that cluster3 had a significantly worse prognosis (Fig. 4). A m ⁶ A regulatory subtype classifier was constructed using binary logistic regression, and the formula was obtained: prediction score = 1.889*HNRNPA2B1-0.451*ALKBH5. The classifier successfully divides the matrix into cluster1/2 and cluster3, which has a high degree of fit with the original classification, and the AUC value matching the original classification reaches 0.985.

5. Exploration of the correlation between the prediction score and the efficacy of immunotherapy

The expression of immune checkpoints in cluster3 and cluster1\2 is significantly different, and the published immunotherapy cohort (IMvigor210) data is used to predict the efficacy of immunotherapy. The predictive score of each patient in the IMvigor210 cohort was calculated using the formula, and then the nomogram was drawn (Fig. 5) to explore the application value of the predictive score. Binary logistic analysis found that the predictive score was significantly correlated with the patient's response to immunotherapy (p <0.0001), the area under the model ROC curve (AUC) was 0.65 (Figure 6).

Example 2 External Verification

From June 20018 to September 2020, 98 surgical specimens of ccRCC patients who had received immune checkpoint inhibitor therapy were selected from the Urology Department of Shanghai Cancer Center, Fudan University. Tissue samples including ccRCC and normal tissues were collected during surgery and are available from the FUSCC tissue bank.

1. Real-time quantitative PCR (RT-qPCR) analysis

Green qPCR预混液(AppliedBiosystems)制造商规程，对mRNA和GAPDH的特定操作循环条件进行了测定，并通过 2^-ΔΔCt计算靶mRNA的相对表达水平。Total RNA was isolated from harvested cells by Trizol (Invitrogen, Carlsbad, CA). It was reverse transcribed into cDNA using PrimeScript RT kit (Termo Fisher, USA). The primers were diluted and mixed in RNase-free dH2O using the SYBRGreen qPCR method (Takara Biotechnology Co.) with the primer sequences shown in Table 1. GAPDH RNA expression was measured for normalization. according to

Green qPCR Master Mix (AppliedBiosystems) manufacturer's protocol, specific operating cycling conditions for mRNA and GAPDH were determined, and relative expression levels of target mRNAs were calculated by 2 ^-ΔΔCt .

Table 1 Summary of primer sequences of three genes in qRT-PCR

2. Multi-marker immunohistochemical staining to identify differences in immune microenvironment between high and low prediction score groups

98 specimens of primary tumors of clear cell renal cell carcinoma from patients who had received immunotherapy were collected from the Cancer Hospital Affiliated to Fudan University (FUSCC), and the relative levels of HNRNPA2B1 and ALKBH5 in the samples were evaluated by RT-qPCR. The predictive score of each specimen was evaluated by using the formula, and then the specimens were divided into two groups with high and low scores according to the predicted score, and the CD3, CD4, CD8, CK, FOXP3, PD in the samples of the high and low score group were detected by multi-marker staining technology. -L1 was stained to assess the changes in the immune microenvironment of the high and low score groups. We initially found that in the high predictive score group, molecules such as PD-L1, CD8, and FOXP3 were stained significantly, and the expression of PD-L1 molecules was significantly increased, which was consistent with the trend of the TCGA clear cell renal cell carcinoma cohort in previous work (Figure 7) . In the low predictive score group, molecular staining such as PD-L1, CD8, and CD4 were weak (Figure 8). It suggested that renal cancer in the high predictive score group may be a functionally suppressive immune microenvironment.

3. Statistical analysis Artificial sequence in FUSCC cohort

The differential mRNA expression of 2 marker genes was analyzed in ccRCC samples and a predictive score was determined for each patient, which was determined as the sum of the weights of each significant oncogenic center gene. The correlation between predictive scores and response to immunotherapy was assessed. The receiver operating characteristic (ROC) curve was constructed to verify the specificity and sensitivity of the diagnosis, and the area under the curve (AUC) analysis was performed to determine the diagnostic ability.

4. Result analysis

After integrating RT-qPCR and clinical follow-up data of 98 ccRCC patients, we validated HNRNPA2B1 and ALKBH5 mRNA expression and derived prediction scores in the FUSCC cohort. An integrated genome was constructed using the selected 2 ^m6A modification-associated genes, which can serve as an independent method for predicting immunotherapy response in ccRCC patients. Generate ROC curves to identify the ability of the genetic model to predict the efficacy of immunotherapy. For the efficacy of immunotherapy, the AUC index of the integrated model was 0.752 (Figure 9), p<0.001, which verified the stability and validity of the prediction score for the prediction of immunotherapy efficacy.

In summary, the present invention utilizes well-characterized and complete ccRCC primary tumor system microarray data analysis to reveal unique gene expression subtypes related to tumor m ⁶ A modification. It was also found that one of the m ⁶ A transcriptional subtypes had significantly poorer prognosis, significantly increased expression of immune checkpoint molecules, and significantly different components of immune cell infiltration. From this, a classifier and a prediction score were constructed for the joint genome of HNRNPA2B1 and ALKBH5.

The results of the combination of the invention and the genome in the IMvigor210 cohort and the real world cohort of FUSCC are all significant, and both can effectively predict the patient's response to immunotherapy. Therefore, in this study, systematic analysis of ccRCC primary tissues could screen and identify promising biomarker expression profiles predictive of immunotherapy efficacy.

In conclusion, this study identifies subtypes of ^m6A modifications that may be involved in the formation of an inhibitory immune microenvironment in ccRCC. The expression levels of HNRNPA2B1 and ALKBH5 have a high value in predicting the efficacy of immunotherapy, which may help to more accurately apply immune checkpoint inhibitors to patients with renal cancer.

The preferred embodiments of the present invention have been specifically described above, but the present invention is not limited to the described embodiments, and those skilled in the art can also make various equivalents without violating the spirit of the present invention. Modifications or replacements, these equivalent modifications or replacements are all included within the scope defined by the claims of the present application.

sequence listing

<110> Cancer Hospital Affiliated to Fudan University

<120> Application of m6A modification-associated joint genome in predicting the efficacy of immunotherapy in patients with clear cell renal cell carcinoma

<130> Claims Specification

<160> 6

<170> SIPOSequenceListing 1.0

<210> 1

<211> 22

<212>DNA

<213> Artificial Sequence

<400> 1

agagttctag gagtggaaga gg 22

<210> 2

<211> 21

<212>DNA

<213> Artificial Sequence

<400> 2

cccattatag ccatccccaa a 21

<210> 3

<211> 19

<212>DNA

<213> Artificial Sequence

<400> 3

ccctgctctg aaacccaag 19

<210> 4

<211> 22

<212>DNA

<213> Artificial Sequence

<400> 4

gttctcttcc ttgtccatct cc 22

<210> 5

<211> 22

<212>DNA

<213> Artificial Sequence

<400> 5

gtcttctcca ccatggagaa gg 22

<210> 6

<211> 23

<212>DNA

<213> Artificial Sequence

<400> 6

catgccagtg agcttcccgt tca 23

Claims (7)

1. The application of the reagent for detecting the relative expression level of each gene of m ⁶ A modification-related joint genome in biological samples in the preparation of reagents or kits for predicting the curative effect of renal clear cell carcinoma immunotherapy, characterized in that the joint genome is HNRNPA2B1 and ALKBH5 joint. 2. The application according to claim 1, wherein the prediction kit comprises a reagent combination for detecting the relative expression levels of HNRNPA2B1 and ALKBH5 in biological samples. 3. The application according to claim 1 or 2, characterized in that the reagent combination contains PCR primers specific for the detection of the above-mentioned genes, and the sequences of the PCR primers are shown in SEQ ID NO.1-4. 4. The application according to claim 1 or 2, characterized in that the biological sample is a section of a tumor specimen surgically removed from a patient with clear cell renal cell carcinoma. 5. A kit for predicting the curative effect of immunotherapy for renal clear cell carcinoma, characterized in that: it consists of a reverse transcription system, a primer system and an amplification system, and the primer system includes PCR as shown in SEQ ID NO.1-4. primers. 6. A system for predicting the curative effect of immunotherapy for renal clear cell carcinoma, characterized in that it includes a prediction kit and an immune subgroup classification model installed on a terminal carrier, The prediction kit is the kit according to claim 5, and the immune subgroup classification model performs prediction scoring according to the following formula, and determines whether the immune typing of the current sample belongs to the immune rejection type or the immune desert type according to the score, Prediction score = 1.889*HNRNPA2B1 expression level-0.451*ALKBH5 expression level. 7. the renal clear cell carcinoma immunotherapy curative effect prediction system according to claim 6, is characterized in that, Wherein, the immune subgroup classification model is constructed according to binary logistic regression, and binary logistic analysis is used to predict the significant correlation between the prediction score and the patient's immunotherapy.

Images