Kit for method for predicting or assessing prognosis of glioma patients

Abstract

The invention provides a kit for a method for predicting or evaluating the prognosis of a glioma patient. In combination with immune-related genes, a group of characteristic genes capable of stably predicting the prognosis of the diffuse glioma is found; by adopting the system disclosed by the invention, the prognosis of the diffuse glioma can be well predicted; single-factor and multi-factor Cox proportional risk regression analysis is further carried out, and it is proved that the immune risk score can be really used as an independent prognosis factor of the diffuse glioma.

Description

[0001] This application is a divisional application of a Chinese patent application with an application date of June 18, 2021, application number 202110681350.9, and invention name “Application of immune-related genes in kits and systems for predicting the prognosis of diffuse glioma”. Technical Field

[0002] The invention belongs to the field of biomedicine, and in particular relates to the application of immune-related genes in a kit and system for predicting the prognosis of diffuse glioma. Background Art

[0003] Gliomas are the most common primary tumors of the brain and central nervous system, accounting for approximately 30% of all brain tumors and central nervous system tumors, and more than 80% of all malignant primary brain tumors. The annual incidence is 3.55 cases per 100,000 people. According to the 2007 World Health Organization (WHO) classification of central nervous system tumors, diffuse gliomas are divided into grade II and grade III low-grade gliomas (LGGs) and grade IV glioblastomas (GBMs) based on histological characteristics. GBM is the most lethal tumor of all grades. Despite significant advances in therapy, including chemotherapy, radiotherapy, and surgical resection, the median overall survival of GBM is only 15 months. At present, the histological grade of gliomas is the main clinical prognostic indicator of gliomas, but the treatment response and prognosis of patients at the same grade vary greatly, so molecular typing is an inevitable requirement for personalized treatment of gliomas.

[0004] Genetic molecular markers refer to models constructed based on the expression of a group of genes through statistical and machine learning methods for clinical prediction. Currently, commonly used gene expression detection methods include high-throughput RNA-seq technology, chip technology, and relatively low-throughput real-time quantitative polymerase chain reaction (RT-qPCR). Although there are many methods for gene expression detection, there is still a lack of relevant research on how to find a set of gene combinations for diffuse glioma prognosis prediction with good predictive performance.

[0005] The immune system has been shown to be a key factor in tumor development. Tumor-infiltrating immune cells are an important component of the tumor microenvironment and play a crucial role in tumor progression, metastasis, and immune escape. In recent years, immune checkpoint proteins such as cytotoxic T lymphocyte antigen 4 (CTLA-4) or programmed cell death ligand 1 / protein 1 (PD-L1 / PD-1) have been used as key targets for cancer immunotherapy. However, there has been no large-scale study on the use of immune-related genes to predict the prognosis of diffuse glioma.

[0006] The main disadvantages of existing technologies are that they do not organically combine the effects of immune-related genes on diffuse gliomas, and have not been validated on a large scale. Summary of the invention

[0007] The purpose of the present invention is to overcome the shortcomings of the prior art and provide an application of immune-related genes in a kit and system for the prognosis of diffuse glioma. By detecting the expression levels of the five immune-related genes screened by the present invention and substituting them into the immune characteristic gene model, the risk of diffuse glioma patients can be accurately predicted.

[0008] To achieve the above purpose, the technical scheme adopted is: the combined use of genes CDC42, PPP4C, NRG3, VIM and HDAC1 in the preparation of a kit for patients with diffuse glioma or a kit for predicting the prognosis of patients with diffuse glioma.

[0009] The present invention provides a method for evaluating the prognosis of a glioma patient. The method comprises: extracting tumor tissue of a glioma patient, detecting the expression levels of CDC42, PPP4C, NRG3, VIM and HDAC1 therein, performing linear transformation on the expression levels of the above five genes and the variable coefficients in the corresponding immune prognostic model, calculating a risk score, and dividing the glioma patients into different risk groups according to the value of the risk score: if the risk score is greater than or equal to 4.084, the glioma patient is in a high-risk group with a poor prognosis; if the risk score is less than 4.084, the glioma patient is in a low-risk group with a good prognosis.

[0010] In addition, the present invention also provides the use of a reagent for detecting the relative expression level of immune genes in preparing a kit for predicting diffuse glioma patients or a kit for predicting the prognosis of diffuse glioma patients, wherein the reagent is used to detect the expression levels of genes CDC42, PPP4C, NRG3, VIM and HDAC1.

[0011] In addition, the present invention also provides a kit for predicting the prognosis of diffuse glioma patients, which comprises reagents for detecting the expression levels of genes CDC42, PPP4C, NRG3, VIM and HDAC1.

[0012] In addition, the present invention also provides a prediction system for predicting the prognosis of diffuse glioma patients, which comprises:

[0013] A data input module is used to input the results of immune-related gene expression levels of diffuse glioma patients into a model calculation module, wherein the immune-related genes include CDC42, PPP4C, NRG3, VIM and HDAC1, and the gene expression level is the number of fragments per million mapped reads per kilobase of transcription (FPKM).

[0014] A model calculation module, including a LASSOCox regression model, is used to calculate the patient's immune risk score according to the expression level of immune-related genes in diffuse glioma patients and the LASSOCox regression model. The calculation formula of the risk score uses a weighted multiplication method: F(x)=Sum[Coeffcient (weight coefficient of each gene in the model)×expression level of each gene in the model], and the Coeffcients of the genes CDC42, PPP4C, NRG3, VIM and HDAC1 are: 0.1503546, 0.2003038, -0.0319212, 0.1307647, 0.1443997, respectively;

[0015] The result output module is used to predict the risk of diffuse glioma patients based on their immune risk scores. When the immune risk score of diffuse glioma patients is greater than the median risk score (4.084), diffuse glioma patients are at high risk, have a poor prognosis, and require more aggressive treatment. When the immune risk score of diffuse glioma patients is less than or equal to the median risk score (4.084), diffuse glioma patients are at low risk, have a better prognosis, and can use milder treatment options to avoid overtreatment.

[0016] The advantages of the present invention are as follows: the present invention provides a kit and system for the prognosis of patients with diffuse glioma using immune-related genes; the present invention combines immune-related genes to find a group of characteristic genes that can stably predict the prognosis of diffuse glioma; the system of the present invention can well predict the prognosis of diffuse glioma; the present invention also performs univariate and multivariate Cox proportional risk regression analysis, proving that the immune risk score can indeed be used as an independent prognostic factor for diffuse glioma. BRIEF DESCRIPTION OF THE DRAWINGS

[0017] Figure 1Figures A and B show the construction and validation of the immune prognostic feature (risk score) model. Figures A and B show that in the CGGA training data set, LASSOCox regression analysis identified the immune genes most relevant to the overall survival of diffuse glioma patients to construct the model. Figures C and F show the survival curves of the low-risk group and high-risk group divided by the 5 immune-related gene models in the CGGA training data set and the TCGA validation data set. The Log-rank rank test shows that the high-risk and low-risk groups divided by the 5 immune genes can effectively divide the overall survival of diffuse glioma patients (P<0.001). Figures D and G show the time-dependent ROC curves of the diffuse glioma prognostic model constructed by the 5 immune-related genes combined with the CGGA training set and the TCGA validation set for 1 year, 3 years, and 5 years of diffuse glioma patients. The AUC (area under the curve) shows that the 5 immune-related genes have a good prognostic prediction effect on diffuse glioma patients. Figures E and H show the distribution of risk scores in the CGGA training set and the TCGA validation set, the expression profiles of the 5 immune-related genes, and the survival status of each patient.

[0018] Figure 2 It showed that qPCR amplification of five pairs of IPS primers and internal control was specific.

[0019] Figure 3 qPCR detected the expression levels of 5 IPS genes in 12 diffuse glioma samples, which were consistent with the prognostic model. The qPCR results showed that there were significant differences in the expression of CDC42, VIM, PPP4C, HDAC1 and NRG3 in diffuse glioma and normal tissues. DETAILED DESCRIPTION

[0020] In order to better illustrate the purpose, advantages and technical solutions of the present invention, the present invention will be further described below in conjunction with specific embodiments and drawings.

[0021] Example 1

[0022] Construction of an immune prognostic model for diffuse glioma ( Figure 1 ):

[0023] The GSE4290 dataset was downloaded from the GEO database, which contains 153 cancer tissues and 27 adjacent normal tissues.

[0024] The limma R package was used to perform differential expression analysis on the cancer group and adjacent cancer group in the GSE4290 dataset. The screening criteria for differentially expressed genes were |log2foldchange|>0.5 and p.adjust<0.05. After differential expression analysis, a total of 3383 differentially expressed genes were obtained, including 1459 up-regulated genes and 1924 down-regulated genes.

[0025] Immune-related genes were obtained from the ImmPort database, totaling 1811 genes. These genes were intersected with the differentially expressed genes to obtain 236 differentially expressed immune-related genes (DE-IRGs).

[0026] We used these 236 DE-IRGs to perform weighted co-expression network analysis in the CGGA dataset to obtain the gene module most associated with glioma survival. We selected genes in the turquoise gene module that was significantly correlated with glioma survival (r=-0.6, P<0.001) to construct the model, which contained a total of 120 DE-IRGs.

[0027] These 120 DE-IRGs were subjected to univariate Cox regression and LASSO-Cox regression analysis, and finally, we obtained a 5-gene immune prognostic model for diffuse glioma.

[0028] The calculation formula of the model uses the weighted multiplication method: F(x)=Sum[Coeffcient(weight coefficient of each gene in the model)×expression level of each gene in the model (FPKM)], and the Coeffcients of the genes CDC42, PPP4C, NRG3, VIM and HDAC1 are 0.1503546, 0.2003038, -0.0319212, 0.1307647, and 0.1443997, respectively.

[0029] Example 2

[0030] This model was used to predict the prognosis of diffuse glioma;

[0031] The test was conducted on a training cohort (CGGA) and a validation cohort (TCGA) with a total of 902 case samples. Figure 1 As shown, the test effect in CGGA (the AUC values ​​for 1 year, 3 years and 5 years are 0.795, 0.855 and 0.896 respectively) and the test effect in glioma patients in the TCGA validation set (the AUC values ​​for 1 year, 3 years and 5 years are 0.815, 0.855 and 0.813 respectively)

[0032] As shown in Table 1 and Table 2, the present invention also performed univariate cox regression analysis and multivariate cox regression analysis, proving that the immune risk score model (IRGS) calculated by the model of the present invention can indeed independently predict the prognostic risk of glioma patients.

[0033] Finally, it should be noted that the above embodiments are only used to illustrate the technical solution of the present invention rather than to limit the scope of protection of the present invention. Although the present invention has been described in detail with reference to the preferred embodiments, those skilled in the art should understand that the technical solution of the present invention can be modified or replaced by equivalents without departing from the essence and scope of the technical solution of the present invention.

[0034] Table 1 Univariate and multivariate analysis of immune-related gene models combined with clinical and case factors

[0035]

[0036]

[0037] Table 2 Univariate and multivariate analysis of immune-related gene models combined with clinical and case factors

[0038]

[0039]

[0040]

[0041]

[0042]

[0043]

[0044]

[0045]

[0046]

[0047]

[0048]

[0049]

[0050]

[0051]

[0052]

[0053]

[0054]

[0055]

[0056]

[0057]

[0058]

[0059]

[0060]

[0061]

[0062] Example 3

[0063] Validation of expression levels of genes used in the immune prognostic model in clinical samples

[0064] Based on the GEO database and TCGA database, we constructed and preliminarily validated the diffuse glioma immune prognostic model. We collected 12 diffuse glioma tissue samples (728487G, 691253G, 689599G, 695948G, 754905G, 719303G, 779526G, 756384G, 786591G, 700690G, 694592G, 701763G) and 6 normal tissue samples (709872N, 531955N, 786009N, 605854N, 705306N, 474898N) from the hospital. Total RNA was extracted, and the expression levels of the five IPS genes used in the immune prognostic model: CDC42, PPP4C, NRG3, VIM and HDAC1 were detected by qPCR, and ACTB was used as an internal positive control. Validation of our model assessed significance at the qPCR level.

[0065] Table 3. qPCR primer list

[0066]

[0067]

[0068] “-F” indicates a forward primer, and “-R” indicates a reverse primer.

[0069] Main instruments and reagents used for qPCR validation

[0070] The main instruments are ABI's fluorescent quantitative PCR instrument Viia7 and Bio-Rad's gradient PCR instrument C1000 Touch. TM, Eppendorf's desktop high-speed refrigerated centrifuge 5424R. Key reagents include Takara's RNAiso Plus (Cat#9109), ThermoFisher's First Strand cDNA Synthesis Kit (Cat#K1612) and PowerTrack TM SYBR Green Master Mix (Cat#A46113).

[0071] Extraction of total RNA from tumor and normal tissues

[0072] Step 1: Sample processing. Take 50-100 mg of tissue blocks and put them into a mortar pre-cooled with liquid nitrogen, grind the sample into powder, and add 1 mL of RNAiso Plus. Then transfer the homogenate to a 1.5 mL centrifuge tube, shake and mix, let it stand at room temperature for 5 minutes, centrifuge at 12000g at 4℃ for 5 minutes, and carefully pipette the supernatant into a new centrifuge tube. Step 2: Phase separation. Add 0.2 mL of chloroform to each 1 mL of homogenate, cover the centrifuge tube tightly, shake and mix for 15 seconds, let it stand at room temperature for 5 minutes, and centrifuge at 4℃ and 12000g for 15 minutes. At this time, the homogenate is divided into three layers, namely: a colorless supernatant layer (containing RNA), a white protein layer in the middle, and a colored lower organic phase. Carefully pipette the supernatant and transfer it to a new centrifuge tube. Step 3: Precipitate RNA. Add 0.5 mL of isopropanol to the supernatant, turn the centrifuge tube upside down to mix thoroughly, and let it stand at room temperature for 10 minutes. Centrifuge at 4℃, 12000rpm, for 10 minutes. Step 4: Wash RNA. Discard the supernatant, add 1 mL of 75% ethanol, mix well, centrifuge at 4°C, 7500g for 5 minutes, and discard the supernatant. Step 5: Dissolve RNA. Dry the RNA precipitate at room temperature for 5 minutes, add an appropriate amount of RNase-free water to dissolve the precipitated RNA. Then determine the concentration and purity of the RNA and set aside.

[0073] Reverse transcription to prepare cDNA

[0074] Dissolve the reagents required for reverse transcription reaction, ThermoFisher's First Strand cDNA Synthesis Kit (Cat#K1612), mix by gently inverting upside down, centrifuge briefly, and place on ice until use.

[0075] Prepare RNA-Primer Mix. Add the following reagents to a pre-cooled RNase free reaction tube to a total volume of 11 μl.

[0076] Table 4. RNA-Primer Mix Reagents

[0077] Reagent components volume Final concentration total RNA 1 μg 250μM Random primer 1μl 10μM RNase-free Water To a total volume of 11 μl

[0078] Prepare the reverse transcription reaction solution. Add the following reagents to the RNA-Primer Mix reaction tube to a total volume of 20 μl.

[0079] Table 5. Reverse transcription reaction solution

[0080] Reagent components volume Final concentration RNA-Primer Mix 11μl 5×Reaction Buffer 4μl 1× 10mM dNTP Mix 2μl 1mM 20U / μl RNase Inhibitor 1 μl 1U / μl 20U / μl M-MLV RTase 2 μl 8U / μl Total volume 20μl

[0081] Reverse transcription reaction. Mix the reaction mixture, centrifuge briefly, and incubate at 37°C for 1 hour. Heat-treat at 85°C for 5 minutes to terminate the reverse transcription reaction, dilute the reverse transcription product 5 times with water, and store at -20°C for later use.

[0082] Quantitative PCR detection

[0083] Thaw SYBR Green Master Mix at 4°C, mix gently by inversion and centrifuge briefly, and prepare the reaction solutions in the following table on ice.

[0084] Table 6. Quantitative PCR

[0085] Element Addition amount Final concentration SYBR Green Master Mix 5μL 1× Forward Primer (2μM) 1μL 0.2μM Reverse Primer (2μM) 1μL 0.2μM cDNA 2μL RNase-free Water 1μL Total volume 10μL

[0086] Table 7. qPCR reaction program settings

[0087]

[0088]

[0089] After the PCR reaction, melting curve analysis was performed using the following procedure:

[0090] temperature Fluorescence collection 60℃～95℃，0.05℃ / S Yes

[0091] Result 1: qPCR amplification of 5 pairs of IPS primers and internal reference was specific ( Figure 2 );

[0092] Results 2: The expression levels of 5 IPS genes in 12 diffuse glioma samples detected by qPCR were consistent with those in the prognostic model ( Figure 3 ).

[0093] qPCR results showed that the expression of CDC42, VIM, PPP4C, HDAC1 and NRG3 in diffuse glioma and normal tissues was significantly different. The expression level of CDC42 in diffuse glioma is about 2.2 times that of normal tissue, the expression level of VIM in diffuse glioma is about 1.6 times that of normal tissue, the expression level of PPP4C in diffuse glioma is about 1.9 times that of normal tissue, and the expression level of HDAC1 in diffuse glioma is about 2.3 times that of normal tissue. NRG3 is lowly expressed in diffuse glioma, about 27% of normal tissue. qPCR results show that the expression basis of the five IPS genes used in our modeling is consistent with that in clinical samples.

[0094] Table 8. qPCR CT value data table

[0095]

[0096]

[0097]

Claims

1. A kit for predicting or evaluating the prognosis of a glioma patient, the kit comprising: A) a reagent for detecting the expression levels of CDC42, PPP4C, NRG3, VIM and HDAC1, the reagent comprising: CDC42-F:GGCGATGGTGCTGTTGGTAA CDC42-R:GTGGATAACTCAGCGGTCGT PPP4C-F:TCACGCAGGTCTATGGCTTC PPP4C-R:GCTGACAGGCTGAGGTAGTC VIM-F: GGACCAGCTAACCAACGACA VIM-R: AAGGTCAAGACGTGCCAGAG HDAC1-F: CTATCGCCCTCACAAAGCCA HDAC1-R: CTGCTTTGCTGTACTCCGACA NRG3-F:CCCGTTCTTCAGTAGCAGCA NRG3-R: AAGGTAGGAGGAGGTAGCGT ACTB-F:AGACCTGTACGCCAACACAG ACTB-R: CGCTCAGGAGGAGCAATGAT, and B) Instructions for use, The method for predicting or evaluating the prognosis of a glioma patient comprises: (a) Obtaining glioma tissue from glioma patients, (b) extracting total mRNA from it, (c) using reagents for detecting the expression levels of CDC42, PPP4C, NRG3, VIM and HDAC1 to detect the expression levels of CDC42, PPP4C, NRG3, VIM and HDAC1; (d) performing linear transformation on the expression levels of the above five genes and the corresponding variable coefficients in the immune prognostic model to calculate the risk score, wherein the calculation formula of the model uses the weighted multiplication method: F(x) = Sum[Coeffcient(weight coefficient of each gene in the model) × expression level of each gene in the model] The Coeffcients of genes CDC42, PPP4C, NRG3, VIM and HDAC1 were 0.1503546, 0.2003038, -0.0319212, 0.1307647 and 0.1443997, respectively. According to the risk score values, glioma patients were divided into different risk groups: if the risk score was greater than or equal to 4.084, it was a high-risk group with a poor prognosis; if the risk score was less than 4.084, it was a low-risk group with a good prognosis.

2. The kit for predicting or evaluating the prognosis of a glioma patient according to claim 1, wherein the glioma is a diffuse glioma.

Images