CN104713969B - Construction method for serum metabonomics analysis model for esophagus cancer primary screening - Google Patents

Construction method for serum metabonomics analysis model for esophagus cancer primary screening Download PDF

Info

Publication number
CN104713969B
CN104713969B CN201510149599.XA CN201510149599A CN104713969B CN 104713969 B CN104713969 B CN 104713969B CN 201510149599 A CN201510149599 A CN 201510149599A CN 104713969 B CN104713969 B CN 104713969B
Authority
CN
China
Prior art keywords
serum
sample
serum sample
analysis
construction method
Prior art date
Legal status (The legal status is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the status listed.)
Active
Application number
CN201510149599.XA
Other languages
Chinese (zh)
Other versions
CN104713969A (en
Inventor
王家林
张涛
薛付忠
赵德利
Current Assignee (The listed assignees may be inaccurate. Google has not performed a legal analysis and makes no representation or warranty as to the accuracy of the list.)
SHANDONG TUMOR HOSPITAL
Original Assignee
SHANDONG TUMOR HOSPITAL
Priority date (The priority date is an assumption and is not a legal conclusion. Google has not performed a legal analysis and makes no representation as to the accuracy of the date listed.)
Filing date
Publication date
Application filed by SHANDONG TUMOR HOSPITAL filed Critical SHANDONG TUMOR HOSPITAL
Priority to CN201510149599.XA priority Critical patent/CN104713969B/en
Publication of CN104713969A publication Critical patent/CN104713969A/en
Application granted granted Critical
Publication of CN104713969B publication Critical patent/CN104713969B/en
Active legal-status Critical Current
Anticipated expiration legal-status Critical

Links

Landscapes

  • Investigating Or Analysing Biological Materials (AREA)
  • Other Investigation Or Analysis Of Materials By Electrical Means (AREA)

Abstract

The invention discloses a serum metabonomics analysis model. A method for constructing the serum metabonomics analysis model comprises the following steps: collecting a healthy serum sample and a sick serum sample; carrying out LC-MS detection on the samples to obtain an original metabonomics fingerprint spectrum; preprocessing the spectrum, sequentially carrying out principal component analysis and partial least square discriminant analysis on the obtained two-dimensional matrix so as to obtain a PLS-DA model, and verifying the obtained PLS-DA model, and constructing the model if no overfitting risk exists. The serum metabonomics analysis model has the advantages that a serum metabonomics analysis technology is initially applied to early esophagus cancer screening, high-risk esophagus cancer population can be rapidly and conveniently screened out by utilizing the serum metabonomics analysis model, an esophagus cancer screening range is narrowed, the efficiency of screening high-risk esophagus cancer population is effectively improved, the screening cost is greatly reduced, pain, caused by a gastroscope, on part of population is effectively avoided, and important economic benefit and social benefit are realized, so that the serum metabonomics analysis model is convenient for popularization and application.

Description

A kind of esophageal carcinoma preliminary examination construction method of blood serum metabolic group analysis model
Technical field
The present invention relates to a kind of blood serum metabolic group analysis model, can be entered with convenient, fast by the model to the esophageal carcinoma The preliminary examination of row early stage, belongs to analysis technical field.
Background technology
The esophageal carcinoma(esophageal cancer)It is the evil formed by esophagus squameous epithelium or epithelioglandular paraplasm Sexually transmitted disease (STD) becomes, and is by the process for gradually developing of atypical hyperplasia to cancer.Show according to World Health Organization's latest data:The whole world is every There are about 400,000 people year and die from the esophageal carcinoma, China is that Incidence of esophageal cancer and mortality rate highest are national, and the tissue class of 90% patient Type is squamous cell carcinoma.Shandong Province river in Shangdong Province river valley is Esophageal Cancer in High Risk Areas, with Feicheng, Ningyang County and Wenshang County highest, morbidity Rate is respectively 93.95/10 ten thousand, 88.68/10 ten thousand and 62.26/10 ten thousand, far above national mean level(16.7/10 ten thousand), especially It is Feicheng, its sickness rate is 3.7 times of Shandong Province's average level, accounts for the 50% of whole county Death Causes of Tumor.The esophageal carcinoma it is occurred frequently Especially Shandong Province river in Shangdong Province river valley causes huge Disease Spectrum and financial burden to China for sick rate and high mortality, into For serious public health problem, it would be highly desirable to solve.
The cause of disease of the esophageal carcinoma is complicated, and most scholars are considered gene and the coefficient result of environment, Incidence of Esophageal Cancer Region aggregation points out it related to environmental factorss and bad life habits, and vitamin lacks with trace element such as in dietary ingredient Weary, feed containing the more food of nitrosaminess, such as like pickling sauerkraut or the food that goes mouldy, like for a long time into boiling hot food, smoking, drink not Good hobby etc..Moleculess research shows that the esophageal carcinoma is multifactor, multistage, the accumulation of polygenic variation and each factor The result interacted with environmental factorss, is related to the change of numerous proto-oncogenes, antioncogene and protein, and cell week Phase.Very long process is often experienced in generation from precancerous lesion to the esophageal carcinoma, and the important precancerous lesion of the esophageal carcinoma is esophageal squamous cell Columnar epithelium is light, in, severe atypical hyperplasia, several years even more than ten years are generally required by slight atypical hyperplasia to canceration.If During this period, precancerous lesion crowd changes bad life habits, carries out preventative medication, then will substantially reduce the several of its canceration Rate.Therefore, the EARLY RECOGNITION of precancerous lesion of cancer of esophagus and intervention is particularly important to reducing its canceration rate.
Be examination in slight, moderate and the precancerous lesion state or the case of the early stage esophageal carcinoma of severe atypical hyperplasia, Country is from 2008 in Feicheng, Shandong Province(National esophageal carcinoma early diagnosis early controls Demonstration Base), for all in Demonstration Base The crowd of 40 ~ 69 years old, by the indicative Biopsy of iodine staining under gastroscope, has carried out esophageal carcinoma early diagnosis and has early controlled project.In esophaguses To suspicious Mucosa Biopsy and histopathological examination is carried out after mirror iodine staining inspection, be current Esophageal Cancer in High Risk Areas generaI investigation, high-risk Goldstandard method in screening and early diagnosiss.Because it is directly perceived, Tu Xiang Cheongju are clear and specificity is high, there is presently no one kind Method can substitute its status.Although the method accuracy is high, but still with certain limitation:Gastroscope is invasive inspection, and receives Inspection person feels pain, there is non-compliance phenomenon in the process of implementation;The indicative Biopsy complex operation of iodine staining, cost under gastroscope It is higher, if the feasibility and efficiency promoted outside demonstration area occurred frequently still remain to be discussed;The iodine staining of Esophageal Cancer in High Risk Areas Less than 10%, up to 90% treats that examination crowd will bear the gastroscopic pain of invasive to positive rate, returns national early diagnosis and early controls item Mesh brings unnecessary financial burden.Therefore, it is badly in need of a kind of new easier method of exploitation, seeks and sent out in esophaguses carcinogenesis The change of exhibition stage organismic internal environment, is the goldstandard screening methods of the esophageal carcinoma(The indicative Biopsy of iodine staining i.e. under gastroscope) The scientific basis and translational medicine for providing primary dcreening operation is supported, to distinguish iodine staining feminine gender and iodine staining Positive Populations, so as to avoid iodine Negative staining crowd bears the pain of invasive gastroscope and reduces the expenditure that early diagnosis early controls project.Additionally, in esophageal carcinoma early diagnosis Early control project early stage once to try using Harvard risk of cancer forecast model(The content that questionnaire is related to includes living environment, diet side Five broad aspects of formula and custom, mood and emotion, medical history and Family history of cancer etc.)High-risk of the preliminary examination esophageal carcinoma Body, then implements again the indicative Biopsy of iodine staining under gastroscope for high-risk individuals.But find that it has in actually performing Higher false negative rate(Fail to pinpoint a disease in diagnosis), area AUC is 0.70 (95%CI, 0.66-0.74) under ROC curve(Literature reference: Thrift AP, Kendall BJ, Pandeya N, Vaughan TL, Whiteman DC. A clinical risk prediction model for Barrett esophagus. Cancer Prev Res (Phila), 2012,5(9): 1115-23.).Because traditional prescreening method effect is undesirable, current esophageal carcinoma examination scope only makees the age limit of 40-69 year It is fixed.
Metabolism group is to biological sample(Such as serum, urine, saliva)In all molecular weight be less than 1000Da small molecules Metabolite(Such as fatty acid, aminoacid, nucleoside and steroidal biological micromolecule)Qualitative and quantitative detection is carried out, is received so as to monitor body The metabolism response that endogenous material is made after the interference such as disease or risk factor accumulation.Internal bio information is transcribed by gene Jing Protein is passed to, small molecule metabolites are finally presented as.Different from genomics and protein science reflection organism in Difference, the research field of metabolism group extend to influencing each other between body and environment and act on.Small molecule metabolites are not Only it is the material base of body vital movement, biochemical metabolism, change of some foeign elements to internal metabolism environment is also presented, Thus the concentration of some unique metabolic things interindividual difference in fact reflect in disease performance and external disease Cause.Recent study discovery, such as metabolic disease and malignant tumor(Ovarian cancer)During disease development, body base Plinth biochemical metabolism there occurs significant change, the metabolic mechanism of human intelligible complex disease will be played a significant role, while being The screening of complex disease and early diagnosiss provide brand-new technical method.
The esophageal carcinoma is a kind of typical metabolic disease, is esophageal squamous cell carcinoma in the patient of China more than 90%(ESCC), It is usually expressed as the overall metabolism disorder of body, therefore metabonomic technology and method are very suitable for the research of the esophageal carcinoma. At present, someone is studied the esophageal carcinoma using metabolism group, such as Wu etc.(Wu H, Xue R, Lu C, et al. Metabolomic study for diagnostic model of oesophageal cancer using gas chromatography/mass spectrometry. J Chromatogr B Analyt Technol Biomed Life Sci, 2009,877(27):3111-7.), Zhang etc.(Zhang J, Bowers J, Liu L, et al. Esophageal cancer metabolite biomarkers detected by LC-MS and NMR methods. PLoS One, 2012,7(1):e30181.), Xu etc.(Xu J, Chen Y, Zhang R, et al. Global and targeted metabolomics of esophageal squamous cell carcinoma discovers potential diagnostic and therapeutic biomarkers. Mol Cell Proteomics, 2013,12 (5):1306-18.)All the esophageal carcinoma is studied using metabonomic technology, but on the whole these researchs focus mostly on greatly In the research of pathology and pathogenesis, the genesis mechanism for disease provides new clue, and the sample for being used is also based on The sample of the advanced esophageal carcinoma patient of hospital's collection, the result of study of gained is only capable of finding Incidence of Esophageal Cancer late period same normal healthy controls The metabolic profile difference compared, and the blood serum metabolic situation of cancer of late stage is huge with the blood serum metabolic difference of precancerous lesion, not Help can be brought for the early stage screening of the esophageal carcinoma and early diagnosiss.Also, at present the studies above method is chosen in normal healthy controls It is upper easily to be affected by selectivity bias, cause the generalization of evaluation result poor.
The content of the invention
For esophageal carcinoma methods for screening in prior art it is loaded down with trivial details, costly, the deficiency such as cause suffering to detection crowd, Also without a kind of simple and efficient, suitable large-scale defect, the invention provides a kind of blood serum metabolic group analysis model, this point Analysis model building method is simple, can replace the indicative Biopsy of iodine staining under gastroscope that the esophageal carcinoma is carried out to crowd in early days, just Step examination, it is both economical and practical, convenient and swift, the pain of crowd to be detected is avoided that again, it is easy to utilize.
The present invention carries out preliminary examination not for adopting the indicative Biopsy of iodine staining under gastroscope at present to the esophageal carcinoma Foot, proposes to replace the indicative Biopsy of iodine staining under gastroscope to carry out tentatively the esophageal carcinoma using blood serum metabolic omics technology first The thinking of examination." national esophageal carcinoma early diagnosis early controls Demonstration Base to present invention support(Feicheng, Shandong Province)" esophageal carcinoma screening with Follow-up crowd's queue, for the indicative biopsy object of iodine staining under all gastroscopes in project, the first upper digestive tract for finding of collection The serum specimen of pathological changes, precancerous lesion of cancer of esophagus and esophageal carcinoma early stage patient, and randomly select in screening without Lesions in Upper Gastrointestinal Tract Health objects as a control group, using sharp separation liquid chromatograph and mass spectrograph(LC/MS)High throughput testing obtain corresponding Metabolic fingerprinting, and built by further analysis to collection of illustrative plates and obtained model, the model can be analyzed and distinguish serum Metabolite situation, can tentatively be judged whether ill dangerous with the esophageal carcinoma by serum protein moteblites situation.Using mould of the present invention Type can be used for full crowd's esophageal carcinoma early stage screening of high region of disease, screen out without ill dangerous crowd, then again for height Danger individuality carries out conventional endoscopic inspection, judges whether with the esophageal carcinoma.The present invention reduces the scope of endoscopy, for esophaguses The enforcement and popularization of cancer screening has important economic and social benefit.
Concrete technical scheme of the present invention is as follows:
A kind of blood serum metabolic group analysis model, construction method is comprised the following steps:
(1)Healthy serum sample and ill serum sample are collected, as analysis sample;
(2)Each analysis sample is analyzed using LC-MS blood serum metabolic omics technologies, healthy serum sample and trouble is obtained The original Metabolic fingerprinting of sick serum sample;
(3)The original Metabolic fingerprinting of healthy serum sample and ill serum sample is carried out into collection of illustrative plates pretreatment, is obtained Per behavior analysiss sample, the two-dimensional matrix of metabolite information is often classified as, for further statistical analysiss;
(4)By step(3)Two-dimensional matrix carry out principal component analysiss successively(PCA)And partial least squares discriminant analysis (PLS-DA), PLS-DA models are obtained, the PLS-DA models to obtaining are verified, dangerous without over-fitting, then gained PLS-DA moulds Type is blood serum metabolic group analysis model.
In above-mentioned construction method, the LC-MS blood serum metabolics omics technology is examined using liquid chromatograph mass spectrography system The method for surveying the Metabolic fingerprinting of serum, in the preferred embodiment of the present invention, liquid chromatograph mass spectrography system used For Ultra Performance Liquid Chromatography-high resolution mass spectrum combined system(UPLC-QTOF/MS).Preferably, chromatographic column used by liquid chromatograph is Waters ACQUITY UPLC HSS T3(1.8 μm; 100 mm (length) × 2.1 mm)Chromatographic column, sample size For 6 μ L, injector temperature is 4 DEG C, and flow velocity is 0.5 ml/min.Chromatogram flow phase includes two kinds of solvents:A is that 0.1wt% formic acid is water-soluble Liquid(Cation ESI+)Or 0.5mmol/L ammonium fluoride aqueous solutions (anion ESI-), B for 0.1wt% formic acid acetonitrile solution (just Ion ESI+) or pure acetonitrile(Anion ESI-).Chromatograph condition of gradient elution is:0-1min is 1%B, and 1-8min is 1%B-100% B is gradually incremented by, and then 10-10.1min is kept to rapidly 1%B for 100%B, and then 1%B continues 1.9min.Preferably, Mass Spectrometer Method Using quadrupole rod time-of-flight mass spectrometry instrument Q-TOF, and using the positive ion mode (ESI+) and anion mould of electric spray ion source Formula (ESI-).Ion source temperature is set as 400 DEG C, and taper hole throughput is 12L/min.Meanwhile, desolventizing temperature is set as 250 DEG C, desolventizing gas flow 16L/min.Capillary voltage is respectively+3kV and -3kV, taper hole under cation and negative ion mode Voltage is 0V.Taper hole pressure is 20psi(Cation)And 40psi(Anion).Spectrum data collection mass charge ratio range be 50~1200 m/z, the rate of scanning of collection is 0.25s.
In above-mentioned construction method, the ill serum sample is selected from esophagitis patient, slight atypical hyperplasia patient, severe Atypical hyperplasia patient, moderate atypical hyperplasia patient, precancerous lesion of cancer of esophagus patient, esophageal carcinoma early stage patient and cancer in situ are suffered from The serum of person.Patients serum's sample can suffer from a kind of serum of above-mentioned patient of disease, or with above-mentioned The serum of the patient of two or more disease(Such as ill serum is esophagitis patients serum and slight atypical hyperplasia The serum of patient, or severe atypical hyperplasia patients serum, moderate atypical hyperplasia patients serum and esophageal carcinoma early stage trouble Person's serum), it is preferred that ill serum sample contains the serum of the patient with above-mentioned every kind of disease(In i.e. ill serum sample Contain esophagitis, slight atypical hyperplasia, severe atypical hyperplasia, moderate atypical hyperplasia, precancerous lesion of cancer of esophagus, food simultaneously Pipe cancer early stage and the serum of cancer in situ patient), prediction accuracy is higher.
In above-mentioned construction method, the healthy serum sample by without illness serum sample corresponding to disease or with it is right Answer the serum of the crowd of the similar disease of disease, it may also be said to be healthy serum sample be the crowd without Lesions in Upper Gastrointestinal Tract Serum.
In above-mentioned construction method, the R2X=0.231, R2Y=0.749, Q2cum=0.638 of gained PLS-DA models.Here In the case of model sensitivity it is high, specificity is good, with good extrapolation effect.
In above-mentioned construction method, the healthy serum sample and ill serum sample are all from 40-69 year crowd.
In above-mentioned construction method, selected serum sample quantity is big, the ill serum sample 453, healthy serum 187, sample, prediction effect is higher.
In above-mentioned construction method, quality control situation when building for timely monitor model adds Quality control samples Carry out quality control.Feed postition is:Sample is analyzed per 10 and adds a Quality control samples, the Quality control samples can be with For Real Time Monitoring sample from sample introduction pre-treatment to analysis during quality control situation.The Quality control samples group Into identical, same healthy serum sample and same ill serum sample are according to 1:The sample that 1 volume ratio is mixed to get.
In above-mentioned construction method, after analysis sample and Quality control samples are obtained, pretreatment is carried out to them, pre- place Add liquid chromatograph mass spectrography system after reason to be detected.Pretreatment before sample introduction is comprised the following steps:
(1)50 μ l analysis samples or Quality control samples are extracted with pipettor, the automatic sample processing systems of Bravo are placed in (Agilent, USA)96 orifice plates on;
(2)150 μ l methanol extraction, vortex 30s are added, and is hatched with protein precipitation at -20 DEG C.
(3)Then 20min are centrifuged with 4000 revs/min at 4 DEG C in high speed centrifuge;
(4)By step(3)Supernatant pour in LC-MS sample injection bottles, be stored at -80 DEG C in case LC-MS detection.
The automatic sample processing system of above-mentioned Bravo, is also Bravo automatic fluid processing platforms.
In above-mentioned analysis sample and Quality control samples preprocessing process, if analysis sample and Quality control samples sampling Time is longer, can place them in -80 DEG C of refrigerators and preserve, will be equipped with during pretreatment the 0.4ml analysis samples that freezed or The 0.5ml centrifuge tubes of Quality control samples are placed in water, and horizontal plane is put in cold compartment of refrigerator not have frozen surfaces to be advisable Row thaws;After thawing completely, vortex 30s.Pretreatment is carried out after above defrosting is processed with above-mentioned pre-treatment step again.
In above-mentioned construction method, collection of illustrative plates pretreatment is carried out to original Metabolic fingerprinting and is referred to:Use Masshunter softwares The original Metabolic fingerprinting for obtaining is converted to into MZdata data files, then Mzdata data files XCMS softwares is used into Bag carries out including the pretreatment operation of retention time correction, peak identification, peak match and peak alignment, obtains can be used for statistical analysiss Two-dimensional matrix, the every behavior analysiss sample or Quality control samples in matrix, is often classified as metabolite information.
In above-mentioned blood serum metabolic group analysis model, the metabolite information point of healthy serum sample and ill serum sample Not Wei Yu model diverse location, it is and each non-overlapping, healthy serum sample group region can be designated as negative areas, ill blood Clear sample group region is positive region.
Using above-mentioned blood serum metabolic group analysis model, can be used for the metabonomic analysis of serum sample.
The method of the analysis blood serum metabolic group of the present invention, comprises the following steps:Serum sample to be checked is carried out into pretreatment, Reach sample introduction requirement;The serum sample to be checked of pretreatment is analyzed using LC-MS blood serum metabolic omics technologies, this is obtained to be checked The original Metabolic fingerprinting of serum sample;The original Metabolic fingerprinting is carried out into collection of illustrative plates pretreatment, obtains can be used for system The metabolite information of meter analysis;The metabolite information is imported in blood serum metabolic group analysis model, for analyzing serum to be checked The metabolism group situation of sample.The prediction probability of esophageal carcinoma early screening risk can be obtained by analysis(Sun is more than 50% Property), so as to pass through body metabolism situation esophageal carcinoma examination risk is judged.In actual applications, if the generation of serum sample to be checked Thank to thing information positioned at the ill serum sample group region of model(I.e. positive region, i.e. prediction probability are more than 50%), then it represents that the people Illness with the esophageal carcinoma is dangerous, needs to carry out indicative biopsy under further scope;If the metabolism of serum sample to be checked Thing information is located at the healthy serum sample group region of model(That is negative areas, prediction probability is less than 50%), then it represents that the people does not have There is the illness of the esophageal carcinoma dangerous, it is not necessary to further to be checked.
In the method for above-mentioned analysis blood serum metabolic group, the step of carry out pretreatment to serum sample to be checked, by blood to be checked The step of original Metabolic fingerprinting of final proof sheet carries out collection of illustrative plates pretreatment all with above-mentioned structure blood serum metabolic group analysis model When pre-treatment step, collection of illustrative plates pre-treatment step it is identical.
" national esophageal carcinoma early diagnosis early controls Demonstration Base to present invention support(Feicheng)" Screening Platform, collection esophagitis, no Typical hypertrophy(It is called atypical hyperplasia, dysplasia), precancerous lesion and the serum specimen in esophageal carcinoma stage early stage, selected serum Specimen is up to 640, and scope is wide, and quantity is more, and verity is relatively reliable.By UPLC-QTOF/MS and statistical pattern recognition method Obtain serum protein moteblites analysis model, the model can as the model of Esophageal Cancer area crowd esophageal carcinoma early stage screening, Reach the purpose of the early screening to the esophageal carcinoma.
The method of blood serum metabolic group analysis model of the present invention and analysis blood serum metabolic group be based on the esophageal carcinoma in early days or The precancerous serum sample of person is obtained, compared to the mould that the serum sample of the esophageal carcinoma middle and advanced stage patient based on hospital is obtained Type, blood serum metabolic group situation of the present invention is higher with the blood serum metabolic group situation matching degree of the high-risk group for wanting examination, more suitable Close the early screening of the esophageal carcinoma.And by the optimization to modeling method, model sensitivity of the present invention is high, and specificity is good, can be very Resolution healthy population well and high-risk patient groups, are especially suitable for clinical practice.
The present invention is first used for blood serum metabolic group analytical technology in esophageal carcinoma early screening, can be right by the present invention The full crowd of Esophageal Cancer in High Risk Areas carries out preliminary examination, quickly, easily filters out Esophageal Cancer crowd, and accuracy is high, contracting The little scope of esophageal carcinoma examination.The present invention is by esophageal carcinoma screening method by the indicative biopsy of iodine staining under traditional direct gastroscope Screening method is changed into the method examination of first Jing present invention analysis blood serum metabolic group, and screening results are positive individuality row gastroscope again The indicative Biopsy of lower iodine staining carries out examination, and analysis only needs to gather serum blood serum metabolic group class hour, noinvasive, spends low, has Effect improves the efficiency of the full Mass screening in Esophageal Cancer in High Risk Areas, greatly reduces screening cost, and effectively prevent part population has The pain that wound gastroscope is caused, it is easy to utilize with important economic and social benefit.Additionally, the proposition of the inventive method The doubtful sufferer of the discovery esophageal carcinoma of simplicity is also helped, is conducive to early discovery, the early treatment of cancer, with very high scientific research and doctor Learn value.
Description of the drawings
Fig. 1 is the quality control chart of blood serum metabolic group detection, and transverse axis is retention time, and the longitudinal axis is metabolite in QC samples In RSD% values.
Fig. 2 is the PCA shot charts of metabolic profile preanalysis, and wherein NEG is feminine gender, represents healthy serum sample, and POS is sun Property, ill serum sample is represented, QC represents Quality control samples.
Fig. 3 be PLS-DA three-dimensional shot chart, R2X=0.231, R2Y=0.749, the Q2cum=0.638 of modeling, wherein Screening NEG are that examination is negative, represent healthy serum sample, and Screening POS are that examination is positive, represent ill serum Sample.
Fig. 4 is that the PLS-DA based on random permutation method models proof diagram.
Fig. 5 is the external certificate shot chart of the PLS-DA models for esophageal carcinoma early screening, wherein Screening NEG It is negative for examination, healthy serum sample is represented, Screening POS are that examination is positive, represent ill serum sample, Screening NEG-test represent that the examination of external testing sample is negative, and Screening POS-test represent external testing sample This examination is positive.
Fig. 6 is the ROC curve of PLS-DA model external testing samples.
Specific embodiment
Below, the present invention is further explained by detailed description below, and advantage of the present invention is carried out It is further to prove.
The construction method and compliance test result of serum protein moteblites analysis model of the present invention is as follows:
1st, object of study
" national esophageal carcinoma early diagnosis early controls Demonstration Base to this research support(Feicheng, Shandong Province)" esophageal carcinoma examination with Community-based population queue is visited, it is (true as goldstandard for the indicative biopsy object of iodine staining under the gastroscope of Feicheng, Shandong Province 40-69 year Recognize), collection is first to be found(Without treatment or medicine was not taken)Iodine staining the positive person under inspection(Including esophagitis, slightly Atypical hyperplasia, severe atypical hyperplasia, moderate atypical hyperplasia, precancerous lesion of cancer of esophagus, esophageal carcinoma early stage patient, cancer in situ Patient)Serum sample as ill sample;And randomly select the i.e. supreme digestion of iodine staining negative subject under gastroscope in screening The health objects of road pathological changes are used as healthy sample.
This research detects altogether 640 people, wherein iodine staining positive person under inspection(I.e. ill sample)Totally 453, iodine staining is negative Person under inspection(I.e. healthy sample)Totally 187;Age, sex are not statistically significant in two group differences, with comparability.
2nd, the blood serum metabolic group detection of LC-MS
The serum of clinical samples and healthy sample is gathered respectively, as patients serum's sample and healthy serum sample, is owned It is put in -80 DEG C of refrigerators after the serum sample centrifugation of collection and preserves, using Ultra Performance Liquid Chromatography-GC-MS(UPLC- QTOF/MS 6550, Agilent)With the automatic Pretreated systems of Bravo(Agilent, USA)Carry out metabolism group detection (Divide 3 big batches detections, and carry out quality control), obtain the original Metabolic Fingerprinting figure comprising chromatograph and Information in Mass Spectra of sample Spectrum.Concrete operations are as follows:
2.1 instrument and equipment
Experimental facilitiess include:The systems of UPLC-QTOF/MS 6550 (Agilent, USA), Bravo systems(Agilent, USA), high speed low temperature centrifugal machine, vibration scroll machine, nitrogen drying device, 4 DEG C of cold storage refrigerators(Haier), pure water meter(Siemens).
Experiment consumptive material includes:Waters ACQUITY UPLC HSS T3(particle size, 1.8 μm; 100 Mm (length) × 2.1 mm) chromatographic column, liquid nitrogen, High Purity Nitrogen;Cone bottom sample injection bottle, 2ml centrifugal rotors, 2ml centrifuge tubes(Circle Bottom), pipettor, 1000 μ l pipette tips, 200 μ l pipette tips, marking pen, altex glove, mask.
Experiment reagent includes:Methanol(Enlightening horse, HPLC levels are pure), acetonitrile(Enlightening horse, HPLC levels are pure), formic acid(Recovery precise treatment Learn institute, Tianjin), pure water(TOC<10ppb).
2.2 serum sample pretreatment
Before serum sample pretreatment, 60 parts of Quality control samples are prepared(QC), by all patients serum's samples, healthy serum Sample and Quality control samples carry out random number, using patients serum's sample and healthy serum sample as analysis sample, every 10 analysis samples add a Quality control samples.Quality control samples are by same patients serum's sample and same healthy blood Final proof is originally mixed, and is divided evenly into 60 parts.Patients serum's sample, healthy serum sample and Quality control samples are carried out Pretreatment, pretreatment includes following 4 steps:
(1)50 μ l analysis samples or Quality control samples are extracted with pipettor, the automatic sample processing systems of Bravo are placed in (Agilent, USA)96 orifice plates on;
(2)150 μ l methanol extraction, vortex 30s are added, and is hatched with protein precipitation at -20 DEG C.
(3)Then 20min are centrifuged with 4000 revs/min at 4 DEG C in high speed centrifuge;
(4)By step(3)Supernatant pour in LC-MS sample injection bottles, be stored at -80 DEG C in case LC-MS detection;
2.3 serum UPLC-QTOF/MS are detected
The pretreated sample of 6 μ L equal portions is injected ACQUITY by UPLC systems (1290 series, Agilent) UPLC HSS T3 (particle size, 1.8 μm;100 mm (length) × 2.1 mm) chromatographic column (Waters, Milford, USA).Loading sequence is completely random sample introduction, to exclude the bias that Loading sequence brings.Chromatograph Mobile phase includes two kinds of solvents:A is 0.1wt% formic acid(Water dilutes, cation ESI+)Or 0.5mM ammonium fluorides (water dilute, bear from Sub- ESI-), B is 0.1wt% formic acid (dilution in acetonitrile, cation ESI+) or 100% acetonitrile(Anion ESI-).Chromatograph gradient is: 0-1min is 1%B, and 1-8min is gradually incremented by for 1%B-100%B, and then 10-10.1min is kept to rapidly 1%B for 100%B, then 1% B continues 1.9min.Flow velocity is 0.5 ml/min.Whole sample detection process maintains 4 DEG C.Wherein, the percentage composition of A and B refers to Be volumn concentration.
Mass Spectrometer Method uses Agilent quadrupole rods time-of-flight mass spectrometry instrument Q-TOF (6550, Agilent), and using electricity The positive ion mode (ESI+) and negative ion mode (ESI-) of esi ion source.Ion source temperature is set as 400 DEG C, and taper hole gas Flow is 12L/min.Meanwhile, desolventizing temperature is set as 250 DEG C, and desolventizing gas flow 16L/min.Cation and bear from Capillary voltage is respectively+3kV and -3kV under subpattern, and taper hole voltage is 0V.Taper hole pressure is 20psi(Cation)With 40psi(Anion).The mass charge ratio range of spectrum data collection is 50~1200 m/z, and the rate of scanning of collection is 0.25s.
3rd, XCMS collection of illustrative plates pretreatment
UPLC-QTOF/MS serum cation ESI+ and anion ESI- detection obtains original Metabolic fingerprinting data and leads to The Masshunt softwares for crossing Agilent companies are converted into Mzdata data files, are then carried out using the XCMS software kits of R languageXCMSCollection of illustrative plates pretreatment, pretreatment is made an uproar including retention time correction, peak identification, peak match, peak alignment, filter, Overlapped peak resolution, threshold Value selection, standardization etc..XCMSThe relevant parameter of pretreatment is:The waist peak width of peak half is 10 (fwhm=10), and retention time window sets 10 (bw=10) are set to, and other specification is default value.XCMSObtain can be used for the Two-Dimensional Moment of statistical analysiss after collection of illustrative plates pretreatment Battle array, where each row is sample(Observation), often it is classified as metabolite(Variable), matrix intermediate value is corresponding metabolite concentration.And it is every Individual metabolite peak uses retention time(Retention time, RT)And mass-to-charge ratio(Mass-to-charge ratio,m/z)It is fixed Property.Then the two-dimensional matrix uses R software kitsCAMERACarry out metabolite peak mark(Including isotopic peak, adduct and fragment Ion).Sample is standardized before statistical analysiss, retention time range set to be analyzed is 0.5~10 min.Jing XCMS collection of illustrative plates pretreatment, in the data matrix of the UPLC-QTOF/MS spectrum generations of positive ion detection pattern 522 metabolism are included Thing peak, anionic textiles pattern is 212 metabolite peaks.
4th, LC-MS quality control of the experiment
When blood serum sample carries out metabolism group detection, the QC samples for preparing is pressed and analyze sample 1 QC of arrangement per 10 Uniformly in the order insertion analysis sample in, so as to real-time monitoring from Sample pretreatment to pattern detection during quality control Situation.The original Metabolic fingerprinting of gained calculates %RSD value of each metabolite in QC samples Jing after XCMS collection of illustrative plates pretreatment (The coefficient of variation), and with %RSD values as the longitudinal axis, draw by transverse axis of retention time(See Fig. 1), the round dot in figure represents metabolism Thing.The %RSD values of most metabolite can control below 30%, to illustrate Sample pretreatment to sample product examine as seen from the figure Quality control during survey is all right, and the metabolism group data for being obtained are genuine and believable.
5th, the metabolic profile preanalysis based on PCA
The dimensional matrix data for obtaining is randomly assigned into 2/3 as training sample training data(125 NEG: 303 POS), in addition 1/3 used as external testing sample test data(62NEG:151POS).Training sample is used unsupervised Analysis method is that principal component analysiss (principal component analysis, PCA) carry out classification preliminary between observation group Trend and outlier, are shown in Fig. 2.The repeatability of QC specimen can be shown that LC-MS quality control of the experiment is good in figure.Can be with from figure Find out, examination is positive(I.e. ill serum sample)It is negative with examination(I.e. healthy serum sample)Between have certain classification trend, But still have partial intersection, need to realize further classification using supervised learning method.
6th, the metabolic profiling analysis based on PLS-DA
For the deviation that gap in elimination group of trying one's best causes, obtain and be more significantly grouped trend, be further directed to train sample This using have supervision analysis method i.e. partial least squares discriminant analysis (partial least squares-discriminant Analysis, PLS-DA) show iodine staining the positive person under inspection and iodine staining feminine gender person under inspection metabolic profile difference and classification Trend.As shown in figure 3, there is trend of classifying between metabolic patterns difference and obvious group between iodine staining positive group and negative group, its The R2X=0.231 of modeling, R2Y=0.749, Q2cum=0.638.From figure 3, it can be seen that compared to the iodine staining of complete health Negative group, either esophagitis, atypical hyperplasia or even more serious cancer in situ and the early stage esophageal carcinoma(Iodine staining positive group) All occur in that the PLS-DA model external certificates of obvious metabolism 7, esophageal carcinoma early screening
By external testing sample dimensional matrix data(62NEG:151POS)In substituting into the PLS-DA models of above-mentioned foundation, obtain To factor 1-3 of prediction(That is t [1], t [2], t [3] value), external testing sample is drawn in PLS shot charts, see Fig. 5.And root According to the negative and positive of the external testing sample shown in PLS-DA models, with actual classification label(Examination is negative and positive)Do ROC curve is analyzed, and sees Fig. 6.Area AUC is 0.99 (95%CI wherein under ROC curve:0.978 ~ 1), sensitivity is 95.4%, specificity is 100%, the model external testing sample predictions strive for rate be 94.8%.This shows, according to present invention side The PLS-DA model predictions accuracy of the esophageal carcinoma early screening that method is set up is high, with good extrapolation effect.
The optimum prediction boundary value of the esophageal carcinoma early screening PLS-DA models of the inventive method(cutoff)For 0.50(See Fig. 6), the boundary value can be used as the standard for judging screening results in actual applications, sensitivity and specificity can be reached Optimization and balance, and other boundary values are not optimum, can lose sensitivity or specificity.
8th, contrast experiment
In model construction research process, 10 species diversity have selected according to the pretreated two-dimensional matrix of collection of illustrative plates the most notable Metabolite, and with each species diversity metabolite as criterion, esophageal carcinoma positive events are reacted by its content difference.Knot Fruit shows that the prediction level of each single metabolite is high, with the matching degree of practical situation less than 89%, not with extrapolation Effect.
9th, conclusion
The checking by more than is as can be seen that the model for only being built by the inventive method could have prediction effect well Really.Gained PLS-DA models of the invention are dangerous without over-fitting, and with trend of classifying between obvious group, by blood serum metabolic situation Can be good at making a distinction healthy serum and suspected lesion serum, sensitivity and specificity are high, can be used for esophageal carcinoma morning Phase examination.
When esophageal carcinoma early screening application is carried out using model of the present invention, step is as follows:
(1)Serum to be checked is gathered, the step in above-mentioned 2.2 is adopted after centrifugation(1)-(4)Pretreatment is carried out to serum, in case Sample detection;
(2)By pretreated serum sample to be checked according to LC-MS detections are carried out the step of above-mentioned 2.3, original metabolism is obtained Finger printing;
(3)Original Metabolic fingerprinting is carried out into collection of illustrative plates pretreatment according to the method for above-mentioned steps 3, the serum to be checked is obtained Metabolite information;
(4)The metabolite information is imported in PLS-DA models, t [1], t [2], t [3] value is calculated, it is possible to obtain food The prediction probability of pipe cancer early screening risk(The positive is more than 50%), so as to pass through body metabolism situation esophageal carcinoma sieve is judged Look into risk.In actual applications, if the metabolite information of serum sample to be checked is located at the ill serum sample group region of model (I.e. positive region, i.e. prediction probability are more than 50%), then it represents that the people has the illness of the esophageal carcinoma dangerous, needs to carry out further Scope under indicative biopsy;If the metabolite information of serum sample to be checked is located at the healthy serum sample group region of model (That is negative areas, prediction probability is less than 50%), then it represents that illness of the people not with the esophageal carcinoma is dangerous, it is not necessary to enter traveling one The inspection of step.
In addition, in order to accelerate screening efficiency, the serum sample of many people can be simultaneously gathered, and is numbered, will be many Individual sample disposably carries out LC-MS detections, collection of illustrative plates pretreatment and data and imports.
It is more than non-limiting to the description of patent of the present invention, based on the other embodiment of patent thought of the present invention, Among the scope of the present invention.

Claims (6)

1. a kind of esophageal carcinoma preliminary examination construction method of blood serum metabolic group analysis model, is characterized in that:The analysis model Construction method comprise the following steps:
(1)Healthy serum sample and ill serum sample are collected, as analysis sample;
(2)Each analysis sample is analyzed using LC-MS blood serum metabolic omics technologies, healthy serum sample and ill blood is obtained The original Metabolic fingerprinting of final proof sheet;
(3)The original Metabolic fingerprinting of healthy serum sample and ill serum sample is carried out into collection of illustrative plates pretreatment, every row is obtained To analyze sample, the two-dimensional matrix of metabolite information is often classified as, for further statistical analysiss;
(4)By step(3)Two-dimensional matrix carry out principal component analysiss and partial least squares discriminant analysis successively, obtain PLS-DA moulds Type, the PLS-DA models to obtaining are verified, dangerous without over-fitting, then gained PLS-DA models are the preliminary examination of the esophageal carcinoma With blood serum metabolic group analysis model;
Serum of the ill serum sample selected from following patient's:Esophagitis, precancerous lesion of cancer of esophagus, esophageal carcinoma early stage and esophaguses The serum of squamous epithelial cancer cancer in situ patient;The healthy serum sample is without disease or similar disease corresponding to illness serum Crowd serum;
Chromatographic column used by liquid chromatograph is Waters ACQUITY UPLC HSS T3 chromatographic columns, and specification is 100 mm × 2.1 Mm, 1.8 μm;Sample size is 6 L, and injector temperature is 4 DEG C, and flow velocity is 0.5 ml/min;Chromatogram flow phase includes two kinds of solvent orange 2 As And B:A under cation ESI+ patterns is 0.1wt% aqueous formic acids, and the A under anion ESI- models is 0.5mmol/L fluorinations Aqueous ammonium, the B under cation ESI+ patterns is the acetonitrile solution of 0.1wt% formic acid, and the B under anion ESI- models is pure second Nitrile;Chromatograph condition of gradient elution is:0-1min is 1%B, and 1-8min is gradually incremented by for 1%B-100%B, and 10-10.1min is 100%B 1%B is kept to rapidly, and then 1%B continues 1.9min;
Mass Spectrometer Method uses quadrupole rod time-of-flight mass spectrometry instrument Q-TOF, and using the positive ion mode ESI+ of electric spray ion source With negative ion mode ESI-, ion source temperature is 400 DEG C, and taper hole throughput is 12L/min, and desolventizing temperature is 250 DEG C, precipitation Agent throughput is 16L/min;Capillary voltage is respectively+3kV and -3kV under cation and negative ion mode, and taper hole voltage is equal For 0V;Positive ion mode lower cone hole pressure is 20psi, and negative ion mode lower cone hole pressure is 40psi;The matter of spectrum data collection Lotus is 50~1200 m/z than scope, and the rate of scanning of collection is 0.25s.
2. construction method according to claim 1, is characterized in that:The serum of the precancerous lesion of cancer of esophagus patient includes food The serum of pipe squamous epithelial cancer is slight atypical hyperplasia patient, the serum of esophagus squameous severe epithelial atypical hyperplasia patient and esophaguses The serum of squamous epithelial cancer moderate atypical hyperplasia patient.
3. construction method according to claim 1, is characterized in that:R2X=0.231, the R2Y=0.749 of PLS-DA models, Q2cum=0.638。
4. the construction method according to claim 1,2 or 3, is characterized in that:The healthy serum sample and ill serum sample Originally it is all from 40-69 year crowd.
5. the construction method according to claim 1,2 or 3, is characterized in that:The ill serum sample 453, healthy blood Final proof sheet 187.
6. the construction method according to claim 1,2 or 3, is characterized in that:Collection of illustrative plates is carried out to original Metabolic fingerprinting pre- Process is referred to:The original Metabolic fingerprinting for obtaining is converted to into MZdata data files with Masshunter softwares, then will Mzdata data files carry out including the pre- place of retention time correction, peak identification, peak match and peak alignment using XCMS software kits Reason operation, obtains can be used for the two-dimensional matrix of statistical analysiss, the every behavior analysiss sample or Quality control samples in matrix, each column For metabolite information.
CN201510149599.XA 2015-04-01 2015-04-01 Construction method for serum metabonomics analysis model for esophagus cancer primary screening Active CN104713969B (en)

Priority Applications (1)

Application Number Priority Date Filing Date Title
CN201510149599.XA CN104713969B (en) 2015-04-01 2015-04-01 Construction method for serum metabonomics analysis model for esophagus cancer primary screening

Applications Claiming Priority (1)

Application Number Priority Date Filing Date Title
CN201510149599.XA CN104713969B (en) 2015-04-01 2015-04-01 Construction method for serum metabonomics analysis model for esophagus cancer primary screening

Publications (2)

Publication Number Publication Date
CN104713969A CN104713969A (en) 2015-06-17
CN104713969B true CN104713969B (en) 2017-04-12

Family

ID=53413480

Family Applications (1)

Application Number Title Priority Date Filing Date
CN201510149599.XA Active CN104713969B (en) 2015-04-01 2015-04-01 Construction method for serum metabonomics analysis model for esophagus cancer primary screening

Country Status (1)

Country Link
CN (1) CN104713969B (en)

Families Citing this family (3)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
CN105044361B (en) * 2015-08-14 2017-07-28 山东省肿瘤防治研究院 A kind of diagnostic marker and its screening technique for being suitable for esophageal squamous cell carcinoma early diagnosis
CN108344761A (en) * 2018-02-10 2018-07-31 厦门大学 A kind of the effect of electroacupuncture treatment atrophic gastritis evaluation method
CN112763640B (en) * 2019-11-06 2024-04-09 中国检验检疫科学研究院 Novel method for identifying four serotypes of listeria monocytogenes differential metabolites

Family Cites Families (6)

* Cited by examiner, † Cited by third party
Publication number Priority date Publication date Assignee Title
US6548252B1 (en) * 2000-10-13 2003-04-15 Esa, Inc. Detection of DNA damage
CN1310033C (en) * 2004-03-23 2007-04-11 中国医学科学院肿瘤医院肿瘤研究所 New method for detecting esophageal cancer haemocyanin fingerprint
CN1873410B (en) * 2005-06-16 2011-11-30 中国医学科学院肿瘤研究所 Method for detecting id marks related to tumor in blood serum from patient of oesophagus cancer
WO2012024543A1 (en) * 2010-08-18 2012-02-23 Caris Life Sciences Luxembourg Holdings Circulating biomarkers for disease
CN103175935A (en) * 2011-12-26 2013-06-26 中国人民解放军第二军医大学 Manufacturing and diagnosis application of blood micro-molecular metabolin specific chromatogram
CN104007217A (en) * 2014-05-29 2014-08-27 深圳市第二人民医院 Serum protein fingerprint spectrum diagnosis model for screening thyroid cancer and establishment method for serum protein fingerprint spectrum diagnosis model

Also Published As

Publication number Publication date
CN104713969A (en) 2015-06-17

Similar Documents

Publication Publication Date Title
CN104713971B (en) The method that a kind of preliminary examination of the utilization esophageal carcinoma analyses model analysiss blood serum metabolic group with the credit of blood serum metabolic group
CN104713970B (en) A kind of construction method of blood serum metabolic group analysis model
CN109884302B (en) Lung cancer early diagnosis marker based on metabonomics and artificial intelligence technology and application thereof
CN105044361B (en) A kind of diagnostic marker and its screening technique for being suitable for esophageal squamous cell carcinoma early diagnosis
CN110646554B (en) Pancreatic cancer diagnosis marker based on metabonomics and screening method and application thereof
CN108414660B (en) Application of group of plasma metabolism small molecule markers related to early diagnosis of lung cancer
WO2004088309A2 (en) Methods for diagnosing urinary tract and prostatic disorders
CN102323246B (en) One group for detecting the characteristic protein of pulmonary carcinoma
CN111562338B (en) Application of transparent renal cell carcinoma metabolic marker in renal cell carcinoma early screening and diagnosis product
CN104713969B (en) Construction method for serum metabonomics analysis model for esophagus cancer primary screening
CN106716127A (en) Methods for detecting ovarian cancer
CN105738526A (en) Method for screening specific serum metabolism markers for triple-negative breast cancer
CN111710372A (en) Exhaled air detection device and method for establishing exhaled air marker thereof
CN110121647A (en) Method for detecting and treating ductal adenocarcinoma of pancreas
CN112183616B (en) Diagnostic marker and kit for diagnosis of glioma, screening method and construction method of glioma diagnostic model
Zeng et al. Image classification combined with faster R–CNN for the peak detection of complex components and their metabolites in untargeted LC-HRMS data
CN109946411B (en) Biomarker for diagnosis of ossification of yellow ligament of thoracic vertebra and screening method thereof
CN113567585A (en) Esophageal squamous carcinoma screening marker and kit based on peripheral blood
WO2016049828A1 (en) Obese population specific biomarker composition and use thereof
CN105044240A (en) Diagnosis marker suitable for early-stage esophageal squamous cell cancer diagnosis
Zhang et al. Advances in mass spectrometry for clinical analysis: Data acquisition, interpretation and information integration
CN105044343B (en) A kind of using method of the construction method, gained diagnostic model and model of esophageal squamous cell carcinoma diagnostic model
CN105044342B (en) A kind of diagnostic marker for being suitable for cancer of the esophagus early diagnosis
CN108548883B (en) Biomarker, method and application for early detection and early warning hepatic injury
CN109444277B (en) Application of metabolic marker in preparation of glioma diagnostic kit

Legal Events

Date Code Title Description
C06 Publication
PB01 Publication
C10 Entry into substantive examination
SE01 Entry into force of request for substantive examination
GR01 Patent grant
GR01 Patent grant