Serum protein marker, kit and detection method for early screening and diagnosis of gastric cancer
阅读说明：本技术 一种用于胃癌早期筛查和诊断的血清蛋白标志物、试剂盒及检测方法 (Serum protein marker, kit and detection method for early screening and diagnosis of gastric cancer ) 是由 张建营 王鹏 王晓 叶华 史健翔 代丽萍 杨倩 秦洁洁 王科妍 江秉华 于 2019-10-23 设计创作，主要内容包括：本发明公开一种用于胃癌早期筛查和诊断的血清蛋白标志物,属于生物医学技术领域,血清蛋白标志物为TP53、GNAS、PBRM1、SRSF2、SMARCB1、GNA11、ACVR1B、PIK3CA或COPB1编码的蛋白中的任意一种或两种以上的联合。同时提供上述血清蛋白标志物的试剂盒及检测方法。本发明基于癌症驱动基因在肿瘤发生和发展中所起的作用,定制138个癌症驱动基因编码的人类蛋白质芯片,共包含180个人源重组蛋白质,先通过蛋白质芯片初步筛选出胃癌的早期检测血清标志物,再经过ELISA间接法实验进行验证,筛选出可用于胃癌早期筛查和诊断的一组胃癌联合检测血清标志物,其为TP53、SRSF2、SMARCB1和GNA11基因编码的蛋白的联合,其联合诊断胃癌ROC曲线下面积达0.950,95％CI为0.925-0.976,保证特异度90.70％时,灵敏度为86.83％,一致率达到80.3％。(The invention discloses a serum protein marker for early screening and diagnosis of gastric cancer, which belongs to the technical field of biomedicine, and the serum protein marker is any one or combination of more than two of TP53, GNAS, PBRM1, SRSF2, SMARCB1, GNA11, ACVR1B, PIK3CA or COPB1 encoded proteins. Meanwhile, a kit and a detection method of the serum protein marker are provided. Based on the function of cancer driver genes in tumorigenesis and development, the invention customizes 138 human protein chips coded by the cancer driver genes, which contain 180 human source recombinant proteins, firstly preliminarily screens out early gastric cancer detection serum markers through the protein chips, and then screens out a group of gastric cancer combined detection serum markers for early gastric cancer screening and diagnosis through ELISA indirect method experiments, wherein the gastric cancer combined detection serum markers are the combination of proteins coded by TP53, SRSF2, SMARCB1 and GNA11 genes, the area under a ROC curve of the combined diagnosis gastric cancer reaches 0.950, 95% CI reaches 0.925-0.976, and the sensitivity reaches 86.83% and the consistency rate reaches 80.3% when the specificity is 90.70%.)
1. A serum protein marker for early screening and diagnosis of gastric cancer, characterized in that: the serum protein marker is any one or combination of more than two of TP53, GNAS, PBRM1, SRSF2, SMARCB1, GNA11, ACVR1B, PIK3CA or COPB1 encoded proteins.
2. The serum protein marker for early screening and diagnosis of gastric cancer according to claim 1, wherein: the serum protein marker is any one or combination of more than two of TP53, SRSF2, SMARCB1 or GNA11 gene coded protein.
3. The serum protein marker for early screening and diagnosis of gastric cancer according to claim 2, wherein: the serum protein markers are combinations of proteins encoded by the TP53, SRSF2, SMARCB1 and GNA11 genes.
4. The serum protein marker for early screening and diagnosis of gastric cancer according to claim 3, wherein: the protein coded by the TP53 gene has an amino acid sequence shown in SEQ ID NO. 1; the protein coded by the SRSF2 gene has an amino acid sequence shown in SEQ ID NO. 3; the protein coded by the GNA11 gene has an amino acid sequence shown as SEQ ID NO. 5; the protein coded by the SMARCB1 gene has an amino acid sequence shown in SEQ ID NO. 10.
5. A kit, characterized in that: comprising the serum protein marker for early screening and diagnosis of gastric cancer according to any one of claims 1 to 4.
6. The kit of claim 5, wherein: the serum protein marker is coated on a solid phase carrier.
7. The kit of claim 6, wherein: the solid phase carrier is made of polyvinyl chloride, polystyrene, polyacrylamide or cellulose.
8. The kit of any one of claims 5 to 7, wherein: the kit also comprises any one or the combination of more than two of positive control serum, negative control serum, confining liquid, serum diluent, a second antibody, second antibody diluent, washing liquid, developing liquid or stopping liquid.
9. The detection method using the serum protein marker for early screening and diagnosis of gastric cancer according to claim 3, characterized in that: the method comprises the following steps:
1) coating and sealing each serum protein marker, and then cleaning;
2) performing primary antibody incubation and cleaning with the diluted serum to be detected, and performing secondary antibody incubation and cleaning;
3) stopping the reaction after the color development of the color development system, and measuring the absorbance value;
4) by OD450-OD620The relative OD value is obtained, then a blank control is deducted, the absorbance value of the serum protein marker is substituted into the following formula, the prediction probability P value is calculated,
P＝1/(1+Exp(-(-7.118+9.545×ODTP53+6.183×ODSRSF2+8.237×ODSMARCB1+3.514×ODGNA11)))；
OD in the formulaTP53、ODSRSF2、ODSMARCB1、ODGNA11The relative OD value of each serum protein marker is subtracted by the absorbance value of a blank control;
when the P value is more than or equal to 0.5, the gastric cancer sample is preliminarily judged;
and when the P value is less than 0.5, the sample is preliminarily judged to be a normal sample.
10. The detection method using serum protein markers for early screening and diagnosis of gastric cancer according to claim 9, characterized in that: further comprising the step 5) of calculating the positive rate, the sensitivity specificity, the john index, the positive predictive value, the negative predictive value, the positive likelihood ratio and the negative likelihood ratio of each serum protein marker;
step 6), parallel joint detection: and (3) constructing a parallel joint detection model of the serum protein marker by using a logistic regression model, and calculating a joint diagnosis result.
Technical Field
The invention belongs to the technical field of biomedicine.
Background
Gastric cancer is a common high-lethality malignant tumor, the mortality rate of the gastric cancer is the second of the cancers, and the early diagnosis rate and the 5-year survival rate of the advanced gastric cancer are both less than 20%. The detection means widely used in the gastric cancer at present comprises noninvasive tests (such as B-ultrasonic and CT) and invasive tests (gastroscope and barium meal), but the detection means lacks of compliance and possibility of popularization and application. Therefore, a new biomarker is needed to find the gastric cancer in time, especially a serum molecular marker, so that a gastric cancer patient can be effectively and early checked, diagnosed and treated in time, and the biomarker is a key point for improving the survival rate and reducing the death rate of the gastric cancer patient. Although some tumor markers, such as CA125 (cancer antigen 125), CA19-9 (cancer antigen 19-9), CEA (carcinoembryonic antigen), CA724 (cancer antigen 724), and the like, can be used for auxiliary detection of gastric cancer at present, the sensitivity and specificity are not high, so that no ideal marker for early screening and diagnosis of gastric cancer for clinical use exists so far.
In recent years, in the field of human oncology, many studies have found that the sera of cancer patients contain a unique set of cellular proteins that induce autoantibody responses, called tumor-associated antigens (TAAs), and the antibodies that they induce are called anti-TAA antibodies (autoantibodies). The proposal of the concept guides a new direction for the research of early diagnosis of gastric cancer. Autoantibodies to tumor-associated antigens (TAAs) are absent or have low titers in the serum of normal human and non-tumor patients, and the level of autoantibodies in the serum of patients often increases earlier than the appearance of tumor symptoms. Furthermore, anti-TAA antibodies have the advantage that other tumor markers, on the one hand, are capable of sustained and stable presence in serum, while other markers, including TAA itself, are rapidly degraded after its release by tumor cells or cleared by the body shortly after it enters the blood circulation. Moreover, the popularity of methods and reagents for detecting autoantibodies has also contributed to the study of the production pattern and function of anti-TAA antibodies in cancer patients. Therefore, the detection of autoantibodies to anti-TAAs can be used as serum markers for early tumor-assisted screening and diagnosis. anti-TAAS autoantibodies are highly stable and durable in serum samples from cancer patients compared to other potential markers, and can be detected before the onset of disease symptoms. In recent years, anti-TAAS autoantibodies have been widely explored as early tumor biomarkers for detecting various types of tumors or as prognostic indicators for tumors. Overall, due to tumor heterogeneity, the sensitivity and specificity of monoclonal anti-TAA autoantibodies is low, and diagnosis of specific tumors lacks discriminatory power.
Subsequent studies over the years have attempted to find more sensitive and specific anti-TAA autoantibodies for the diagnosis of gastric cancer, optimizing the combination for the diagnosis of gastric cancer. There are two common methods for finding valuable TAA autoantibodies: the first is serological screening of recombinant cDNA expression library (serological analysis of recombinant cDNA expression libraries, SEREX); the other is proteomics technology. In contrast to SEREX, proteomics technology enables screening of multiple tumor sera and enables screening of TAAs with post-translational modifications. During the development of tumors, hundreds of thousands of mutations of genes are involved, but only some key genes, called cancer driver genes, are mutated to cause the development of tumors. Studies suggest that different types of tumorigenesis generally contain 2-8 driver genes, and that mutations in these genes lead to preferential tumor growth, and that these genes can be divided into 12 signaling pathways by regulating the cell cycle, cell survival and genome to maintain 3 cell core processes. 138 cancer driver genes (see Vogelstein B. science. (2013)339(6127):1546-1558), including 74 cancer suppressor genes and 64 cancer genes, are currently found in a variety of tumor whole genome sequencing studies. The protein coded based on the cancer driving gene can also induce the body to generate corresponding autoantibodies in circulating blood of the body, and the research on the protein coded by the cancer driving gene and the autoantibodies in serum induced by the protein can reveal the occurrence, development or prognosis of tumors to a certain extent.
Disclosure of Invention
The invention aims to provide a serum protein marker for early screening and diagnosis of gastric cancer, and a kit and a detection method for the serum protein marker.
Based on the purpose, the invention adopts the following technical scheme:
a serum protein marker for early screening and diagnosis of gastric cancer is any one or combination of more than two of TP53, GNAS, PBRM1, SRSF2, SMARCB1, GNA11, ACVR1B, PIK3CA or COPB1 encoded proteins.
The serum protein marker is any one or combination of more than two of TP53, SRSF2, SMARCB1 or GNA11 gene coded protein.
The serum protein markers are combinations of proteins encoded by the TP53, SRSF2, SMARCB1 and GNA11 genes.
The protein coded by the TP53 gene has an amino acid sequence shown in SEQ ID NO. 1; the protein coded by the SRSF2 gene has an amino acid sequence shown in SEQ ID NO. 3; the protein coded by the GNA11 gene has an amino acid sequence shown as SEQ ID NO. 5; the protein coded by the SMARCB1 gene has an amino acid sequence shown in SEQ ID NO. 10.
A kit comprising the serum protein marker for early screening and diagnosis of gastric cancer.
The serum protein marker is coated on a solid phase carrier.
The solid phase carrier is made of polyvinyl chloride, polystyrene, polyacrylamide or cellulose.
The kit also comprises any one or the combination of more than two of positive control serum, negative control serum, confining liquid, serum diluent, a second antibody, second antibody diluent, washing liquid, developing liquid or stopping liquid.
A detection method using serum protein markers for early screening and diagnosis of gastric cancer, comprising the steps of:
1) coating and sealing each serum protein marker, and then cleaning;
2) performing primary antibody incubation and cleaning with the diluted serum to be detected, and performing secondary antibody incubation and cleaning;
3) stopping the reaction after the color development of the color development system, and measuring the absorbance value;
4) by OD450-OD620The relative OD value is obtained, then a blank control is deducted, the absorbance value of the serum protein marker is substituted into the following formula, the prediction probability P value is calculated,
P＝1/(1+Exp(-(-7.118+9.545×ODTP53+6.183×ODSRSF2+8.237×ODSMARCB1+3.514×ODGNA11)))；
OD in the formulaTP53、ODSRSF2、ODSMARCB1、ODGNA11The relative OD value of each serum protein marker is subtracted by the absorbance value of a blank control;
when the P value is more than or equal to 0.5, the gastric cancer sample is preliminarily judged;
and when the P value is less than 0.5, the sample is preliminarily judged to be a normal sample.
Further comprising the step 5) of calculating the positive rate, the sensitivity specificity, the john index, the positive predictive value, the negative predictive value, the positive likelihood ratio and the negative likelihood ratio of each serum protein marker;
step 6), parallel joint detection: and (3) constructing a parallel joint detection model of the serum protein marker by using a logistic regression model, and calculating a joint diagnosis result.
Compared with the prior art, the invention has the following beneficial effects:
1) based on the functions of cancer driver genes in tumorigenesis and development, the invention prepares 138 human protein chips coded by the cancer driver genes, which contain 180 human recombinant proteins in total, firstly preliminarily screens out early detection serum markers of gastric cancer through the protein chips, and then carries out verification through an ELISA indirect method experiment, screens out a group of gastric cancer combined detection serum markers for early screening and diagnosis of gastric cancer, which is the combination of proteins coded by TP53, SRSF2, SMARCB1 and GNA11 genes, the area under a combined diagnosis gastric cancer ROC curve reaches 0.885, 95% CI is 0.852-0.919, the sensitivity is 70.8% when the specificity is 90.3%, the consistency rate reaches 80.3%, the invention can assist clinical diagnosis of gastric cancer, and has better reference value.
2) The detection method has the characteristics of high sensitivity, strong specificity, low cost and the like, is simple and quick to operate, and can provide a basis for early diagnosis of the gastric cancer.
Drawings
FIG. 1 is a schematic diagram of the detection of a focused array-based human protein chip in an experimental example;
FIG. 2 is ROC curve analysis chart of 9 TAAs screened by protein chip in experimental example for individual diagnosis of gastric cancer;
FIG. 3 is a scattergram of SNR values of 9 TAAs selected by the protein chip in the experimental example;
FIG. 4 is a schematic diagram of ELISA indirect detection in an experimental example;
FIG. 5 is ROC curve analysis chart of ELISA for verifying 9 TAAs for gastric cancer alone in experimental examples;
FIG. 6 is a graph showing the distribution of OD value scatter for 9 TAAs verified by ELISA in the experimental examples;
FIG. 7 is a ROC graph of the data in the training set of ELISA-verified 4 TAAs for combined diagnosis of gastric cancer in the experimental examples;
FIG. 8 is a ROC plot of the data in the validation set for the ELISA validation of 4 TAAs in the experimental examples in combination with the gastric cancer.
Detailed Description
Examples of the experiments
1 preparation of serum samples
1.1 serum samples for protein chip experiments
Primary gastric cancer patients (with pathological diagnosis of gastric cancer) were collected at the first subsidiary hospital of the university of zhengzhou and the beijing youan hospital, with patient consent and approval by the institutional review board and the hospital ethics committee. All samples are collected by a red blood collection tube for 5-10mL of whole blood of a research object, after the samples are placed for 2 hours at room temperature, 1000g of the whole blood is centrifuged for 15 minutes, supernatant is taken, each sample is subpackaged with a plurality of parts and labeled, and the samples are stored in a refrigerator at the low temperature of-80 ℃ to avoid repeated freezing and thawing.
According to epidemiological analysis, 100 primary gastric cancer sera and 50 normal control sera from the Yogan hospital for contemporaneous physical examination were finally collected for primary chip screening. The total of 74 (74%) cases and 26 (26%) cases of 100 patients with primary gastric cancer have an average age of 60.0 + -10.5 years and an age range of 27-86 years; in 50 cases of normal serum, there were 23 (46.0%) cases of males and 27 (54.0%) cases of females, with the average age of 40 + -13 years and the age range of 20-71 years. All gastric cancer patient sera were collected at the time the patient was initially diagnosed with gastric cancer and had not received any chemoradiotherapy or surgical treatment. The normal human serum is from the physical examination population participating in the annual health physical examination and free of any malignant tumor symptoms.
1.2 serum samples for experimental validation of ELISA Indirect methods
(1) Serum samples were collected from the Beijing Youran Hospital and the first subsidiary Hospital of Zhengzhou university (see section 1.1 above for details).
(2) From the clinical laboratory of the first subsidiary hospital of zhengzhou university (242 cases of primary gastric cancer) and the cardiovascular survey program of jinshui district of zhengzhou city (192 cases of normal persons), 111 cases (78.2%) of male patients, 31 cases (21.8%) of female patients, the average age of 58.4 ± 11.4 years old, and the age range of 23-94 years old were shared; 192 normal blood serums, 150 male (78.1%) and 42 female (21.9%) with average age of 59.4 + -11.7 years and age range of 23-89 years. All gastric cancer patient sera were collected at the time the patient was initially diagnosed with gastric cancer and had not received any chemoradiotherapy or surgical treatment.
2 protein chip customization for screening gastric cancer diagnosis marker
154 proteins encoded by 138 cancer driver genes (see Vogelstein B. science. (2013)339(6127):1546-1558) and 26 proteins provided by the laboratory (total 180 human recombinant proteins) were immobilized on protein chips for tumor marker screening. The protein chip for screening tumor markers was a HuProtTM human protein chip custom-made by Biotech, Inc. of Bo Chong, Guangzhou. The cancer driver gene is ABL1, ACVR1B, AKT1, ALK, APC, AR, ARID1A, ARID1B, ARID2, ASXL 2, ATM, ATRX, AXIN 2, B2 2, BAP 2, BCL2, BCOR, BRAF, BRCA2, CARD 2, CASP 2, CBL, CDC 2, CDH 2, CDKN 22, CEKN, CIC, CREBP, CRCP 2, CSF 12, CTNNB 2, CYLD, DAXX, DNMT 2, EGFR, DNMT3 2, EP300, ERBB2, EZH2, FAM123 2, XW 2, FGFR2, TFAS 2, NFR 2, NOTCNF 2, TFAS 2, NFS 2, NFR 2, TFS 2, NFS 2, TFAS 2, NFS 2, NFR 2, NFS 2, TFAS 2, TFS 2, TFN 2, TFS 2, TFN 2, TFS 2, TFN 2, TFS 2, TFN 2, TFS 2, TFN 2, TFS 2, WT1, CCND1, CDKN2C, IKZF1, LMO1, MAP2K4, MDM2, MDM4, MYC, MYCL1, MYCN, NCOA3, NKX2-1, SKP2, IMP1, P62, IMP3, CIP2A, RalA, c-Myc, Survivin, CyclinB1, 14-3-3 zeta, MDM2, P53, P16, NPM1 and CAPER alpha protein chip experiments
See figure 1 for experimental principles.
3.1 reagents required for the experiment:
1) sealing liquid: 3mL of 10% BSA, 7mL of 1 XPBS solution, mixed well and placed on ice.
2) Serum incubation liquid: 1mL of 10% BSA was added to 9mL of 1 XPBST solution, mixed well and placed on ice.
3) Cleaning solution: 1 XPBST solution, stored in a refrigerator at 4 ℃.
4) Secondary antibody incubation solution: including a fluorescently-labeled anti-human IgM secondary antibody (cy 5-labeled, appearing red) and a fluorescently-labeled anti-human IgG secondary antibody (cy 3-labeled, appearing green).
3.2 specific Experimental procedures for protein chips
(a) Rewarming: taking out the chip from a refrigerator at-80 deg.C, re-warming in a refrigerator at 4 deg.C for half an hour, and re-warming at room temperature for 15 min.
(b) And (3) sealing: and fixing the rewarming chip in 14blocks in a fence, adding sealing liquid into each block, placing the blocks on a side swing shaking bed, and sealing for 3 hours at room temperature.
(c) Incubation of serum samples: after the blocking was completed, the blocking solution was poured out, and then a serum incubation solution prepared in advance was quickly added thereto, 14 samples were incubated for each chip (the serum samples were frozen and thawed in a chromatography cabinet at 4 ℃ and diluted 1:50 by volume with 1 XPBST solution (serum dilution) containing 1% BSA), the loading volume of each serum sample was 200. mu.L, and the incubation was performed overnight at 4 ℃ with a shaker at 20 rpm.
(d) Cleaning: taking out the chip and the chip clamp together, sucking out the sample, then quickly adding an equal volume of 1 XPBST solution, and circulating for a plurality of times to ensure that no cross contamination exists among the serum samples when the chip clamp is detached. After the chip clamp was removed, the chip was placed in a chip washing cassette containing washing solution, and washed on a horizontal shaker at room temperature at 80rpm for 3 times, each time for 10 min.
(e) And (3) secondary antibody incubation: the chip was transferred to an incubation box containing 3mL of secondary antibody incubation solution, and the shaking table was shaken laterally at 40rpm, protected from light, and left at room temperature for 60 min.
(f) Cleaning: the chip was removed (note that the upper surface of the chip was not touched or scratched), placed in a chip washing cassette to which a washing solution was added, and washed 3 times at 80rpm for 10min each time on a horizontal shaker. After completion with ddH2O washing for 10min 2 times.
(g) And (3) drying: the chip is placed in a chip drier for centrifugal drying.
(h) Scanning: operating according to the operating specifications and instructions of the scanner.
(i) Data extraction: and aligning the chip image and each array of the result as a whole, pressing an automatic alignment button, and extracting and storing data.
(j) And carrying out data preprocessing.
(k) Data analysis was performed by first screening for AUC >0.5, P <0.05 and final gastric cancer serum markers by Logistic regression analysis, which protein chip experiments screened the following serum protein markers: the proteins encoded by the cancer driver genes TP53, GNAS, PBRM1, SRSF2, SMARCB1, GNA11, ACVR1B, PIK3CA and COPB1 (FIG. 2 is a ROC curve analysis chart of Gastric cancer individually diagnosed by 9 TAAs screened by the protein chip, in FIGS. 2 (1) - (9) are ROC curves of Gastric cancer individually diagnosed by TP53, GNAS, PBRM1, SRSF2, SMARCB1, GNA11, ACVR1B, PIK3CA and COPB1 in sequence; FIG. 3 is a SNR value Normal scatter chart of the 9 TAAs, in FIG. 3 NC is Normal control, namely healthy Normal serum, and GC is Gastric cancer, namely Gastric cancer case). Wherein, the TP53, GNAS, PBRM1, SRSF2, SMARCB1, GNA11, ACVR1B, PIK3CA and COPB1 genes encode proteins which sequentially have amino acid sequences shown in SEQ ID NO. 1-9.
4 ELISA Indirect method experimental verification
See figure 4 for experimental principles.
The specific experimental steps are as follows:
a) coating: coating according to the concentration in Table 1, 100. mu.L/well, 4 ℃ overnight;
b) and (3) sealing: 2% BSA (Solebao, Beijing, analytical pure) in1 XPBST (PBS, Tween20 Solebao, Beijing), 200. mu.L/well, overnight at 4 ℃;
c) cleaning: washing with 350 μ L/well 1 XPBST solution for 3 times;
d) primary antibody incubation: diluting serum and 1 XPBST solution containing 1% BSA at a volume ratio of 1:100, 100 mu L/hole, and carrying out half water bath at 37 ℃ for 1 h;
e) cleaning: washing with 350 μ L/well 1 × PBST solution for 5 times;
f) and (3) secondary antibody incubation: diluting HRP-labeled mouse anti-human IgG (Olympic, Wuhan) with 1% BSA-containing 1 XPBST solution at a volume ratio of 1:10000, then carrying out 100 mu L/hole treatment in a half water bath at 37 ℃ for 1 h;
g) cleaning: 350 μ L/well, 1 XPBST solution 5 washes.
h) Color development: TMB color development System, solution A (200 mgTMB.2HCl in 1L deionized water, Solebao, Beijing, analytical grade) and solution B (9.2g citric acid, 37g Na)2HPO4·12H2O and 8ml of 0.75% H2O2Dissolved in 1L of deionized water) and mixed according to the volume ratio of 1:1, and then the mixture is placed in a hole with the volume ratio of 100 mu L and is protected from light at room temperature to reach the expected color (about 5-15 min).
i) And (4) terminating: absorbance was measured in 10min after 50. mu.L/well of 10% concentrated sulfuric acid.
j) Measuring the absorbance: by OD450-OD620For relative OD values, the blank control was then subtracted, and IgG was normalized and then subjected to subsequent data processing (details of data processing are shown in "5 data processing" sections b) -d) described below).
The coating concentrations of the 9 TAAs screened by the protein chip experiment when the ELISA experiment is performed are shown in table 1 below, and the arrangement table of the 96-well plate of the ELISA experiment is shown in table 2 below. In table 2, the positive quality control refers to serum with a higher OD value of the ELISA experiment and positive corresponding antibody through Western Blot experiment verification, the negative quality control refers to serum with an OD value near the mean value of the ELISA experiment in normal control population and negative through Western Blot verification, the blank is serum diluent, human IgG-1-human IgG-8 are human IgG antibodies diluted in a gradient manner, and the concentrations are 10, 20, 50, 100, 150, 200, 250 and 300ng/ml in sequence.
TABLE 1 coating concentration of each of the 9 TAAs selected
Table 2 96-well plate arrangement for ELISA experiments
The experimental results are as follows: the results of 9 TAAs were detected by ELISA indirect methods are shown in fig. 5 and 6. FIG. 5 is a ROC curve analysis diagram of 9 TAAs individually diagnosed with gastric cancer in ELISA validation experiment, and in FIG. 5, (1) - (9) are ROC curves of TP53, GNAS, PBRM1, SRSF2, SMARCB1, COPB1, PIK3CA, ACVR1B and GNA11 encoded proteins individually diagnosed with gastric cancer; FIG. 6 is a graph showing the OD value scatter distribution of 9 TAAs in ELISA-verified experiments, in FIG. 6 NC indicates Normal Controls, i.e., healthy Normal serum, and GC indicates Gastric Cancers, i.e., Gastric cancer cases.
As can be seen from FIG. 5, the area under the ROC curve of the single index diagnosis gastric cancer is 0.560-0.759, and when the minimum specificity is ensured to be 90%, the sensitivity range is 17.86-47.27%. Wherein the area under the curve of SMARCB1 is the largest and is 0.759, the sensitivity reaches 47.27 percent, and the specificity is 90.4 percent; the area under the ROC curve of ACVR1B is 0.702, the sensitivity reaches 42.26%, and the specificity is 90.00%; the area under the ROC curve for GNAS was the smallest, 0.560, the sensitivity was 217.86%, and the specificity was 90.14%. As can be seen from FIG. 6, the OD values of the 9 indices were distributed between 0 and 1, the median OD values were substantially evenly distributed between 0.2 and 0.4, and the differences between the healthy control and the gastric cancer cases, except for GNAS and PIK3CA, were statistically significant.
5 data processing
Differential expression proteins are screened out by using a focused array human protein chip in a gastric cancer group and an NS normal control group through statistical data analysis, and the specific method is as follows:
(1) the initial screening result of the chip is obtained through Focused Array protein chip experiment.
(2) And (3) stability analysis: in the experimental process, the test samples test are repeated according to different time, different chips and different positions so as to evaluate the stability of different chips at different time.
(3) Data analysis and results: samples after high background and extreme sample interference were rejected and 180 proteins of each of the IgG and IgM response types were subjected to consistent statistical analysis with the following analysis logic:
a) in order to eliminate the situation of signal nonuniformity caused by inconsistent background values among different protein points in the same chip, the background normalization method is used for processing, the ratio of the foreground value to the background value of each protein, namely F/B, is realized, SNR (signal to noise ratio), namely the mean value of the F/B of two repeated proteins, is defined on the basis, and subsequent statistical analysis is carried out.
b) Assuming that samples needing to be aligned are respectively from two identical populations, and whether the two groups of variances needing to be aligned are homogeneous is determined through an F test, then the F test result is selected to correspond to a t test, and the t test result is characterized by P-value. By definition, when p-value <0.05, the original hypothesis is rejected, i.e. there is a significant difference between the two.
c) For any protein, fold change, which is the difference between the cancer group and the normal group, was calculated to indicate the difference between the two groups.
d) For any protein, according to the diagnostic significance of the two groups compared, firstly, defining cutoff-1.5 as a positive judgment threshold, namely, when the SNR of a sample on the protein is more than or equal to 1.5, the protein is a positive protein; then, based on the control group, setting a proper cutoff threshold, calculating the difference of the positive rate of the cancer group and the control group at the cutoff threshold, and taking the maximum difference as the positive rate of the protein in the compared cancer group to search the high response protein specific to the control group in the cancer group, and finally, defining that the positive rate is not lower than 15%.
e) Based on the logic, the gastric cancer group (100 primary gastric cancer patient serums collected from the first subsidiary hospital of the Beijing Youran hospital and Zheng State university) and the Youran control group (50 normal serums of Youran hospital) are compared, differential proteins which are obviously higher than those of the control group are screened out to be used as gastric cancer candidate markers, and finally 9 serum protein markers (TP53, GNAS, PBRM1, SRSF2, SMARCB1, GNA11, ACVR1B, PIK3CA and COPB1) are selected through a chip to evaluate the diagnostic value of gastric cancer.
Wherein, the protein coded by TP53 gene has an amino acid sequence shown as SEQ ID NO.1, the protein coded by GNAS gene has an amino acid sequence shown as SEQ ID NO.2, the protein coded by PBRM1 gene has an amino acid sequence shown as SEQ ID NO.3, the protein coded by SRSF2 gene has an amino acid sequence shown as SEQ ID NO.4, the protein coded by SMARCB1 gene has an amino acid sequence shown as SEQ ID NO.5, the protein coded by GNA11 gene has an amino acid sequence shown as SEQ ID NO.6, the protein coded by ACVR1B gene has an amino acid sequence shown as SEQ ID NO.7, the protein coded by PIK3CA gene has an amino acid sequence shown as SEQ ID NO.8, and the protein coded by COPB1 gene has an amino acid sequence shown as SEQ ID NO. 9. The information sources of the above 9 genes are shown in Table 3 below.
TABLE 3 information sources of the above 12 genes
(4) The ELISA experiment verification is carried out on the 9 serum protein markers screened by the protein chip: the method comprises the steps of verifying the samples of the submission chip and verifying the samples collected outside the submission chip again, thereby realizing the verification of the protein chip and ensuring the popularization.
(5) The experimental results are as follows: ELISA experimental verification is carried out on the 9 serum protein markers screened by the protein chip, 70% of the total population is extracted as a training set by using a random sampling method for all verified populations, a disease prediction model is constructed by using binary logistic regression, indexes are screened by using three methods of Forward (Forward: conditional), Backward (Backward: conditional) and direct input (Enter), 9, 4 and 4 proteins Enter the model respectively, and the corresponding area under the ROC curve (AUC), sensitivity (Se) and specificity (Sp) are shown in the following table 4.
TABLE 4 model indices screened by different screening methods
The diagnostic value and economic benefit analysis of the model constructed above shows that the model containing 4 indexes (TP53, SRSF2, SMARCB1 and GNA11) has the best effect, and is verified in the rest 30% of people (verification set), as shown in FIGS. 7 and 8, the area under the ROC curve of the combined diagnosis gastric cancer reaches 0.950, 95% CI is 0.925-0.976, and when the specificity is 90.70%, the sensitivity is 86.83%, and the consistency rate reaches 80.3%.