JP2018179962A - Method of diagnosing urothelial cancer using n-linked sugar chain - Google Patents

Abstract

PROBLEM TO BE SOLVED: To provide a urothelial cancer diagnosis method which is noninvasive and more accurate than urine cytology.SOLUTION: A urothelial cancer diagnosis method uses urothelial cancer identification marker including at least one N-linked sugar chain selected from m/z 1362, m/z 1525, m/z 1566, m/z 1591, m/z 1607, m/z 1648, m/z 1687, m/z 1709, m/z 1769, m/z 1794, m/z 1849, m/z 1871, m/z 1915, m/z 2011, m/z 2033, m/z 2058, m/z 2074, m/z 2118, m/z 2220, m/z 2337, m/z 2423, m/z 2525, m/z 2728, m/z 2744, m/z 3109, m/z 3341, and abundance of the urothelial cancer identification marker in a sample collected from a subject is used as an indicator.SELECTED DRAWING: None

Description

  The present invention relates to a method for determining urothelial cancer using a specific N-linked sugar chain (hereinafter also referred to as "N-glycan") as a marker for diagnosing urothelial cancer.

  Urothelial cancer includes bladder cancer and upper urothelial cancer (renal pelvis and ureteral cancer), and is caused by canceration of urothelial tissue.

  Bladder cancer is said to be more common in men, 70 years of age, and to smokers. Bladder cancer is classified into two types, superficial bladder cancer and invasive bladder cancer.

  The kidney is made up of the urine producing part of the renal parenchyma (the renal parenchyma is further divided into the cortex and the medulla) and the renal pelvis where the urine produced by the renal parenchyma collects. The ureter is a long tube connecting the kidney and the bladder, one on each side. Urine produced in the renal parenchyma collects in the renal pelvis and is sent to the bladder through the ureter for excretion. The renal pelvis and ureter are called the upper urinary tract, and the cancer that can be generated here is renal pelvic and ureteral cancer. The inside of the urinary tract leading from the renal pelvis to the ureter, bladder and part of the urethra is made of a mucous membrane called urothelium (transitional epithelium). Cancer that develops from these cells is called urothelial cancer and accounts for most of renal pelvis and ureteral cancer. Renal and ureteral cancer is a relatively rare disease, with a frequency only about 1/20 that of bladder cancer.

  The cystoscopy is the gold standard for the diagnosis of bladder cancer. However, it is difficult to identify a flat type tumor such as a small diameter tumor or a carcinoma in situ, or even the spread of a flat lesion associated with a raised lesion by observation under white sunlight in conventional cystoscopy. Therefore, random biopsy is recommended in cases where the presence of intraepithelial cancer is suspected, especially in cases where urinary cytology is positive, even if there are no obvious findings with cystoscopy. Recently, methods such as photodynamic diagnosis (PDD) and narrow-band imaging (NBI) using a fluorescent cystoscope have been developed, and improvement in cystoscopic diagnosis accuracy is expected. Photodynamic diagnosis (PDD) is based on fluorescence after administration of fluorescent precursors such as 5-aminolevulinic acid (5-ALA) and hexylaminolevulinic acid (HAL), which are selectively taken up by tumor cells. The inside of the bladder is observed using a cystoscope to detect lesions that fluoresce red. Narrow band light observation (NBI) is a special light limited to only a short wavelength, and highlights the difference between a cancerous mucous membrane and a normal mucous membrane by a fine pattern or color tone by blood vessels and detects a lesion.

  The only marker for which the efficacy has been sufficiently examined in screening screening for bladder cancer is the urine occult blood test strip method. Urinary cytology is a method of pathologically diagnosing the degree of variant of urothelial exfoliated cells excreted in urine. The sensitivity is reported to be 40 to 60% and the specificity to be 90 to 100%.

  In Japan, Nuclear Matrix Protein 22 (NMP 22), Bladder Tumor Antigen test (BTA test), etc. are covered by insurance as new bladder cancer related molecular markers. In recent years, in Europe, in addition to these, the utility of Microsatellite analysis for detecting microsatellite instability, which is a type of genomic instability, has been reported.

  On the other hand, abdominal ultrasound, venous urography, retrograde pyelography, renal pyeloscope, CT, MRI, urinary cytology, etc. are used for diagnosis of renal pelvis and ureteral cancer. Retrograde pelvic imaging is a method in which a thin catheter is inserted retrogradely into the ureter via the urethra and bladder, and a contrast agent is directly injected to visualize a lesion. Urine cytology is a method of examining the presence or absence of cancer cells by staining the cells detached in the urine with a dye.

  In recent years, sugar chains have been used as tumor markers for diagnosis of various cancers (Non-Patent Documents 1 to 5, Patent Documents 1 to 5).

  In order to analyze sugar chains, it is essential to separate only sugar chains and secure the necessary amount, but in the case of sugar chains, there is no method for directly amplifying a molecule. Therefore, in order to carry out structural analysis and quantification of sugar chains, it has been necessary to directly purify sugar chains in excess of the amount necessary for analysis from biological samples such as blood, urine, tissues and cells. The glycoblotting method developed in recent years can specifically recover only sugar chains, and enables comprehensive analysis of sugar chains. That is, carriers (beads) having an aminooxy group or a hydrazide group on the surface have the property of rapidly adsorbing an aldehyde group, and among many biomolecules, those having an aldehyde group in the molecule are almost free sugars. As it is only a chain, only sugar chains are specifically recovered by selectively capturing aldehyde compounds in a biological sample. The techniques disclosed in Non Patent Literatures 1 to 5 and Patent Literatures 1 to 5 are based on the glycoblotting method.

  In Non-Patent Document 1, five types of N-linked sugar chains are non-bladder cancer, with the task of providing a marker for identifying a specific change in bladder cancer using simple and recoverable blood or urine. It is disclosed that there is a significant increase in bladder cancer serum (n = 10) as compared to serum (n = 13). Urinary sugar chain analysis also discloses that 17 different types of N-linked sugar chains were increased in bladder cancer (n = 8) compared to non-bladder cancer (n = 11). In addition, it has been suggested that further large-scale analysis based on these results may establish useful serum sugar markers in bladder cancer.

  In addition, research based on glycoblotting has been reported in renal cell carcinoma (Non-patent document 2), prostate cancer (non-patent document 3), pancreatic cancer (non-patent document 4) and hepatocellular carcinoma (non-patent document 5) It is done.

  Patent Document 1 discloses a method of determining the presence or absence of renal cell carcinoma by analysis of N-linked sugar chains in serum. Specifically, comprehensive analysis of N-linked sugar chains is performed using patient's serum, and the presence or absence or malignancy of renal cell carcinoma can be evaluated by using the detected expression amount of sugar chains as an index. What can be done is disclosed. In addition, it is disclosed that the sugar chain is selected from the group consisting of RC1, RC5, RC15, RC19, RC22, RC23, RC25, RC32, RC33, and a sugar chain having a bisecting structure with a four-branch structure or a three-branch structure. It is done.

  Patent Document 2 discloses searching for sugar chains that can be cancer markers, particularly N-linked glycans, from sugar chain profiling for various cancers including bladder cancer and any diseases associated with glycosylation changes. Specific examples of cancer include, for example, prostate cancer, pancreatic cancer, breast cancer, bladder cancer, renal cancer, colon cancer, ovarian cancer, hepatocellular carcinoma, gastric cancer and lung cancer.

  Patent Document 3 discloses a hepatocellular carcinoma marker comprising a trisialyl sugar chain having a specific structure, and a method of examining hepatocellular carcinoma using the amount of the marker as an index.

  Patent Document 4 is a marker for diagnosing digestive organ cancer, which is used for identifying the presence or absence of digestive cancer and is an N-linked sugar chain having a specific structure contained in blood. And a method of testing for digestive tract cancer that can be used for early and convenient examination of digestive tract cancer using the marker.

  Patent Document 5 discloses a method for diagnosing pancreatic cancer using an N-linked sugar chain.

  As described above, sugar chains have been used as tumor markers for diagnosis of various cancers, but sugar chains have not been known as tumor markers for diagnosis of urothelial cancer.

Re-table 2011/370 045 issue Japanese Patent Application Publication No. 2008-539413 Patent No. 4752032 Unexamined-Japanese-Patent No. 2013-083490 Patent No. 5212893

Urological Surgery Vol. 27, No. 12, pp. 1977 (2014.12.15) Hatakeyama S et al., Serum N-glycan profiling predicts prognosis in patients undergoing hemodialysis. The Scientific World Journal. 2013; 2013: 268407 Ishibashi Y et al., Serum tri- and tetra-antennary N-glycan is a potential predictive biomarker for castration-resistant prostate cancer. The Prostate. 2014; 74: 1521-9 Nouso K et al., Clinical utility of high-throughput analysis analysis in patients with pancreatic cancer. Journal of gastroenterology. 2013; 48: 1171-9 Hepatology (Baltimore, Md). 2014; 59: 355-6 Miyahara K et al., Serum glycos as a prognostic marker in patients with advanced hepatocellular carcinoma treated with sorafenib.

  Although cystoscopy is essential for diagnosing bladder cancer, it is a test that is highly invasive to subjects such as pain and infections. Urine cytology is used as a non-invasive test, but its low sensitivity limits its usefulness.

  In addition, diagnosis of renal pelvis and ureteral cancer is also low in diagnostic accuracy, although it is accompanied by high invasiveness such as retrograde pelvic imaging and ureteroscopy. Urine cytology is used as a non-invasive test for urothelial carcinoma, but its low sensitivity limits its usefulness.

  Thus, although urinary cytology is used as a non-invasive test for diagnosis of urothelial cancer, the sensitivity is low and thus invasive tests including cystoscopy are required, and diagnostic accuracy and physical burden on the patient are required. The problem was not solved. On the other hand, blood tumor markers useful for early detection are not known for urothelial cancer, and no sugar chain biomarkers have been searched.

  Therefore, a non-invasive, highly accurate method for diagnosing urothelial cancer which surpasses urine cytology has been desired.

  As a result of earnest studies to solve the above problems, the present inventors have found a tumor marker specific to urothelial cancer by performing comprehensive mass analysis of serum N-glycan in urothelial cancer cases. . In particular, the combination of 26 urothelial carcinoma-related N-glycans (N-glycan score) can predict urothelial carcinoma with high sensitivity and high specificity, and can be predicted regardless of the urinary cytology classification. Therefore, they have found that they have high predictive value also for cytological negative subjects, and have completed the present invention. By using the tumor marker of the present invention, urothelial cancer can be determined with high accuracy.

The present invention includes the following.
[1] A marker for identifying urothelial cancer comprising at least one of the following N-linked sugar chains:

(Wherein ○ represents galactose (Gal), ◇ represents N-acetylneuraminic acid (NeuAc), ● represents mannose (Man), ■ represents N-acetylglucosamine (GlcNAc), and Δ represents deoxyhexose (dHex). Left bracket indicates binding to any N-terminal branched sugar chain)
[2] urothelial carcinoma is upper urothelial carcinoma (renal pelvis and ureteral carcinoma), and N-linked sugar chain is

(Wherein, 、, 、, 、, 、, は are the same as above; the left parenthesis indicates binding to a sugar chain at any of N-terminal branched ends) [1 ] The urothelial cancer identification marker as described in.
[3] A method for testing urothelial cancer, using the amount of the urothelial cancer identification marker according to [1] or [2] in a sample collected from a subject as an index.
[4] An amount of the urothelial cancer identification marker in a sample collected from a subject including the following steps (1) and (2) and the urothelial cancer identification in a sample collected from a healthy subject The test method for urothelial cancer according to [3], wherein a difference or ratio with the amount of marker is evaluated:
(1) measuring the amount of the urothelial cancer identification marker in the sample collected from the subject, and (2) collecting the amount of the urothelial cancer identification marker measured in the step and the healthy person Comparing the amount of the urothelial cancer identification marker in the sample to be evaluated and evaluating the difference or ratio between the two.
[5] The test method according to [3] or [4], wherein the urothelial cancer identification marker is detected using an antibody or lectin.
[6] The test method according to any one of [3] to [5], further including the step of testing hematuria from the sample collected from the subject.
[7] A determination method of urothelial cancer characterized in that the following steps (1) to (4) are sequentially provided:
(1) releasing N-linked sugar chains from glycoproteins in blood collected from a subject;
(2) a step of purifying the released sugar chain;
(3) mass spectrometry of the purified sugar chain;
(4) The mass-to-charge ratio (m / z) of the peak by mass spectrometry using a MALDI-TOF-MS analyzer is 1362, 1525, 1566, 1591, 1607, 1648, 1687, 1709, 1769, 1794, 1849 1871, 1915, 2011, 2033, 2058, 2074, 2118, 2220, 2337, 2423, 2525, 2728, 2744, 3109 and 3341. Among sugar chains corresponding to the above sugar chains, or Measuring the detected intensity of at least one sugar chain, wherein increasing or decreasing the detected intensity as compared to a healthy person indicates that the subject is a patient with urothelial cancer.
[8] The urothelial carcinoma is upper urothelial carcinoma (renal and ureteral carcinoma), and the peak mass-to-charge ratio (m / z) by mass spectrometry using MALDI-TOF-MS analyzer is 1362 The determination method according to [7], which is any one of 1525, 1566, 1648, 1709, 1769, 2011, 2033, 2074, 2118, 2423 and 2525.
[9] The determination method according to [7] or [8], further including the step of examining hematuria from the sample collected from the subject.
[10] Use of the following N-linked sugar chain as a biomarker for monitoring the onset of urothelial carcinoma.


(Wherein, 、, 、, 、, ■, は are the same as above; the left parenthesis indicates binding to a sugar chain at any of the N-terminal branched ends)
[11] urothelial carcinoma is upper urothelial carcinoma (renal pelvis and ureteral carcinoma), and N-linked sugar chain is

(Wherein, 、, 、, 、, 、, は are the same as above; the left parenthesis indicates binding to any of the N-terminal branched glycans) [10 Use described in].
[12] A kit for monitoring the onset of urothelial cancer, comprising means for measuring N-linked sugar chains in a blood sample obtained from a subject.
[13] A means for receiving a blood sample from a subject, a means for measuring an N-linked sugar chain contained in the blood sample, a means for comparing the measured N-linked sugar chain to a reference value, and a reference value [12] The kit for monitoring the onset of urothelial cancer onset according to [12], which comprises means for determining whether the subject has developed urothelial cancer from the comparison.
[14] A method for diagnosing urothelial cancer, comprising screening the following N-linked sugar chain in a blood sample collected from a subject.
(Wherein, 、, 、, 、, ■, は are the same as above; the left parenthesis indicates binding to a sugar chain at any of the N-terminal branched ends)
[15] The diagnostic method according to [14], wherein the urothelial cancer is upper urothelial cancer (renal pelvis and ureteral cancer) and the following N-linked sugar chain is screened.
(Wherein, 、, 、, 、, ■, は are the same as above; the left parenthesis indicates binding to a sugar chain at any of the N-terminal branched ends)

  The N-linked sugar chain of the present invention is a specific identification marker for urothelial carcinoma, and by analyzing a specific N-linked sugar chain in serum, the urothelium is not burdened on the subject. It is possible to determine cancer with high sensitivity and high specificity.

FIG. 1 shows urothelial carcinoma associated N-glycans selected by having an area under the curve (AUC) of 7.0 or more. A total of six N-glycans (glycans m / z 1687, 1871, 2011, 2337 elevated and glycans m / z 1566, 1769 elevated in urothelial cancer patients) were selected as biomarkers. FIG. 2A shows a comparison of N-glycan concentrations between healthy volunteers and patients with urothelial cancer. Error bars indicate 95% confidence intervals. FIG. 2B shows a comparison of N-glycan concentrations between healthy volunteers and patients with urothelial cancer. Error bars indicate 95% confidence intervals. FIG. 2C shows a comparison of N-glycan concentrations between healthy volunteers and patients with urothelial carcinoma. Error bars indicate 95% confidence intervals. FIG. 2D shows a comparison of N-glycan concentrations between healthy volunteers and patients with urothelial cancer. Error bars indicate 95% confidence intervals. FIG. 2E shows a comparison of N-glycan concentrations between healthy volunteers and patients with urothelial cancer. Error bars indicate 95% confidence intervals. FIG. 2F shows a comparison of N-glycan concentrations between healthy volunteers and patients with urothelial cancer. Error bars indicate 95% confidence intervals. FIG. 3A shows a comparison of N-glycan scores between healthy volunteers and patients with urothelial cancer. FIG. 3B shows a comparison of N-glycan score between bladder cancer and upper urothelial cancer. There were no significant differences. FIG. 3C shows a comparison of N-glycan score localized (Ta-1N0M0) and advanced (T2 or higher) urothelial cancer patients. There were no significant differences. FIG. 3D shows a comparison of urinary cytology classification of N-glycan scores. No correlation was found between N-glycan score and urinary cytology classification. FIG. 4 is a diagram showing the distribution of healthy volunteers and urothelial cancer patients in a Swimmers plot, with N-glycan scores on the vertical axis and subjects on the horizontal axis in ascending order of score. FIG. 5 shows the results of evaluation of diagnostic accuracy of N-glycan score by ROC analysis. FIG. 6 shows the results of ROC analysis for evaluating diagnostic accuracy when hematuria test is performed in addition to N-glycan score. FIG. 7 shows urothelial carcinoma associated N-glycans selected by the area under the curve (AUC) of 7.0 or more and ROC curves when hematuria test is added thereto. Figure 8 shows the distribution of healthy volunteers and patients with urothelial cancer as Swimmers plots, arranged on the vertical axis with the score with the hematuria test result added to the N-glycan score, with the horizontal axis representing the test subjects. FIG. FIG. 9 shows serum immunoglobulin-derived N-glycans identified by comprehensive analysis of serum immunoglobulins by glycoblotting. FIG. 10 shows the variation of N-glycans comparing upper urothelial carcinoma patients with healthy volunteers. FIG. 11 is a diagram showing N-glycan scores of upper urothelial cancer patients and healthy volunteers when the cutoff value is −0.615. In patients with upper urothelial carcinoma, the N-glycan score is lower than the cut-off value, and in healthy volunteers the N-glycan score is shown to be higher than the cut-off value. FIG. 12 shows the distribution of N-glycan scores in upper urothelial cancer patients and healthy volunteers. FIG. 13 is a diagram showing the predictability of upper urothelial carcinoma based on the obtained N-glycan score. FIG. 14 shows the variation in N-glycans of upper urothelial cancer patients compared with healthy volunteers for each of IgG1, IgG2, IgG3, IgG4, IgM, IgA, IgG (total) and Ig (total). is there. FIG. 15 shows the variation in N-glycans of upper urothelial carcinoma patients compared to healthy volunteers for lectin-binding immunoglobulin associated with upper urothelial carcinoma. FIG. 16 is a diagram showing a lectin array score of a patient with upper urothelial carcinoma and a healthy volunteer when the cutoff value is −1.275. In patients with upper urothelial carcinoma, the lectin array score is shown to be lower than the cut-off value, and in healthy volunteers the lectin array score is shown to be higher than the cut-off value. FIG. 17 shows the distribution of lectin array scores in upper urothelial cancer patients and healthy volunteers. FIG. 18 is a diagram showing the predictability of upper urothelial carcinoma based on the obtained lectin array score. FIG. 19 shows serum-derived N-glycans identified by comprehensive analysis by serum glycoblotting. FIG. 20 shows the variation of N-glycans comparing urothelial carcinoma patients with healthy volunteers. FIG. 21 shows N-glycan scores of urothelial cancer patients and healthy volunteers when the cutoff value is zero. In patients with urothelial carcinoma, the N-glycan score is lower than the cut-off value, and in healthy volunteers it is shown that the N-glycan score is higher than the cut-off value. FIG. 22 shows the distribution of N-glycan scores in patients with urothelial cancer and healthy volunteers. FIG. 23 is a diagram showing the predictability of urothelial cancer based on the obtained N-glycan score. FIG. 24 is a diagram showing an outline of a lectin array. FIG. 25 shows the levels of five urothelial carcinoma associated N-glycans in the immunoglobulin (Ig) fraction. (a) to (e) show complex biantennary-type N-glycans (m / z 1607, 1769 and Ig fractions of urothelial carcinoma (UC), healthy volunteers (HV) and prostate cancer (PC) 2074) Levels and fucosylated bisecting GlcNAc-type N-glycan (m / z 2118 and 2423) levels are shown. The results shown are representative of three independent experiments. The differences between groups were compared statistically using the Mann-Whitney U test for nonnormal distribution models. (f) shows the synthetic pathway of candidate N-glycans in the present invention. FIG. 25 shows the levels of five urothelial carcinoma associated N-glycans in the immunoglobulin (Ig) fraction. (a) to (e) show complex biantennary-type N-glycans (m / z 1607, 1769 and Ig fractions of urothelial carcinoma (UC), healthy volunteers (HV) and prostate cancer (PC) 2074) Levels and fucosylated bisecting GlcNAc-type N-glycan (m / z 2118 and 2423) levels are shown. The results shown are representative of three independent experiments. The differences between groups were compared statistically using the Mann-Whitney U test for nonnormal distribution models. (f) shows the synthetic pathway of candidate N-glycans in the present invention. FIG. 26 shows a general protocol of the glycoblotting method and high-throughput clinical glycan analysis based on the glycoblotting method. (a) Glyco SweetBlot ^TM to 10μl of serum Ig fraction samples for blotting (System Instruments, Hachioji, Japan) was added to. (b) Coomassie brilliant blue (CBB) stained bands of sodium dodecyl sulfate polyacrylamide gel electrophoresis (SDS-PAGE) on representative whole serum and purified Ig fraction samples. Lanes 1-4: whole serum, lanes 5-8: UIg fractions. (c) After enzymatic cleavage of N-glycans from Ig proteins, all N-glycans released in the digestion mixture are directly mixed with BlotGlyco H beads (Sumitomo Bakelite, Co., Tokyo) to obtain N-glycans I captured it. (d) Sialic acid was methyl esterified after separating the beads from other molecules by washing. (e) These processed N-glycans were then labeled with benzyloxyamine (BOA) and released from BlotGlyco H beads. (f) Mass spectra of BOA labeled N-glycans from Ig fractions were obtained using Ultraflex III instrument (Bruker Daltonics, Germany). A total of 32 N-glycans were identified in the Ig fraction. The structure of the identified N-glycan is shown by a monosaccharide code: ○: galactose (Gal), ●: mannose (Man), ■: N-acetylglucosamine (GlcNAc), ▲: fucose (Fuc) and ♦: sialic acid . Figure 27: Serum immunoglobulin (Ig) levels in patients with urothelial cancer (UC) (bladder cancer (UCB) and upper urothelial cancer (UTUC)), healthy volunteers (HV) and prostate cancer (PC) patients FIG. (a) Serum total Ig (total of IgG1-4, IgM and IgA) levels in urothelial carcinoma (UC) group, healthy volunteer (HV) group and prostate cancer (PC) group. (b) Serum total IgG (IgG1-4) levels in urothelial carcinoma (UC) group, healthy volunteer (HV) group and prostate cancer (PC) group. (c) Serum IgM levels in urothelial carcinoma (UC) group, healthy volunteer (HV) group and prostate cancer (PC) group. (d) Serum IgA levels in urothelial carcinoma (UC) group, healthy volunteer (HV) group and prostate cancer (PC) group. (e) Serum IgG1 levels in urothelial carcinoma (UC) group, healthy volunteer (HV) group and prostate cancer (PC) group. (f) Serum IgG2 levels in urothelial carcinoma (UC) group, healthy volunteer (HV) group and prostate cancer (PC) group. (g) Serum IgG3 levels in urothelial carcinoma (UC) group, healthy volunteer (HV) group and prostate cancer (PC) group. (h) Serum IgG4 levels in urothelial carcinoma (UC) group, healthy volunteer (HV) group and prostate cancer (PC) group. The results shown are representative of three independent experiments. The differences between groups were compared statistically using the Mann-Whitney U test for nonnormal distribution models. Figure 27: Serum immunoglobulin (Ig) levels in patients with urothelial cancer (UC) (bladder cancer (UCB) and upper urothelial cancer (UTUC)), healthy volunteers (HV) and prostate cancer (PC) patients FIG. (a) Serum total Ig (total of IgG1-4, IgM and IgA) levels in urothelial carcinoma (UC) group, healthy volunteer (HV) group and prostate cancer (PC) group. (b) Serum total IgG (IgG1-4) levels in urothelial carcinoma (UC) group, healthy volunteer (HV) group and prostate cancer (PC) group. (c) Serum IgM levels in urothelial carcinoma (UC) group, healthy volunteer (HV) group and prostate cancer (PC) group. (d) Serum IgA levels in urothelial carcinoma (UC) group, healthy volunteer (HV) group and prostate cancer (PC) group. (e) Serum IgG1 levels in urothelial carcinoma (UC) group, healthy volunteer (HV) group and prostate cancer (PC) group. (f) Serum IgG2 levels in urothelial carcinoma (UC) group, healthy volunteer (HV) group and prostate cancer (PC) group. (g) Serum IgG3 levels in urothelial carcinoma (UC) group, healthy volunteer (HV) group and prostate cancer (PC) group. (h) Serum IgG4 levels in urothelial carcinoma (UC) group, healthy volunteer (HV) group and prostate cancer (PC) group. The results shown are representative of three independent experiments. The differences between groups were compared statistically using the Mann-Whitney U test for nonnormal distribution models. FIG. 28 shows the clinical significance of diagnostic N-glycan scores. Diagnostic N-glycan score (dNGScore) is more common in urothelial cancer (UC) (bladder cancer (UCB) and upper urothelial cancer (UTUC)) than in healthy volunteers (HV) and prostate cancer (PC) Was significantly higher. Figure 29: Clinical significance of diagnostic N-glycan scores, N-glycans in healthy volunteers (HV), prostate cancer (PC), upper urothelial carcinoma (UTUC) and bladder cancer (BC) -Shows the distribution of scores (Waterfall plot of dNGScore). FIG. 30. Clinical significance of diagnostic N-glycan scores. The ROC curve analysis of dNGScore for the detection of urothelial cancer (UC) (bladder cancer (UCB) and upper urothelial cancer (UTUC)) shows the result of hematuria and urine cytology. Figure 31 shows the clinical significance of the diagnostic N-glycan score, showing the association of dNGScore levels with hematuria. The negative predicted value (Negative) and the positive predicted value (Positive) N-glycan score in hematuria when the cutoff value is -0.0955 are shown. Figure 32 shows the clinical significance of the diagnostic N-glycan score, showing the association between dNGScore levels and urinary cytology. The N-glycan score of class I-V in urine cytology at a cutoff value of -0.0955 is shown. FIG. 33 shows the clinical significance of diagnostic N-glycan scores and shows the association between dNGScore levels and tumor invasiveness. The N-glycan score of NMIBC, MIBC, non-muscle layer infiltrated upper urothelial carcinoma (UTUC) and muscle layer infiltrated upper urothelial carcinoma (UTUC) when the cutoff value is −0.0955 is shown.

  The present invention relates to a marker for identifying urothelial cancer comprising an N-linked sugar chain having a specific structure, wherein the amount of the marker for identifying urothelial cancer in a sample collected from a subject is used as an indicator. Provided is a kit for monitoring the onset of urothelial cancer, comprising a method for testing cancer and a means for measuring N-linked sugar chains of a blood sample obtained from a subject.

  Hereinafter, the present invention will be described in detail. The following embodiment is an example for explaining the present invention, and it is not the meaning which limits the present invention only to this embodiment. The present invention can be practiced in various forms without departing from the scope of the invention.

  So far, no specific serum tumor marker has been identified in urothelial carcinoma and no glycosylated biomarker has been searched. The present inventors performed comprehensive mass analysis of serum N-linked sugar chain (N-glycan) using glycoblotting in cases of urothelial carcinoma. That is, the present inventors focused on N-linked sugar chains (N-glycans), in particular, N-linked sugar chains (N-glycans) contained in the serum of patients with urothelial cancer, and diagnose urothelial cancer We tried to develop markers for urothelium and test methods for urothelial cancer.

  Specifically, 289 patients with urothelial cancer for which consent was obtained were used as subjects, blood was collected from each subject, and mass analysis was performed for N-linked sugar chains in serum. As a result, in the serum of a subject (a patient with urothelial cancer), N-linked sugar chains having a significantly changed abundance were successfully identified as compared to healthy subjects (348 patients), We came to complete the present invention

Here, the Glycoblotting method used in the analysis of the present invention selects a sugar chain, which is the only molecule having an aldehyde group (= hemiacetal group) in the presence of many biomolecules. It is based on the principle that only sugar chains can be specifically recovered by capturing them in a targeted manner.

Specifically, a bead having on the surface a functional group capable of nucleophilic addition rapidly to an aldehyde group such as an aminooxy group or a hydrazide group at a high density is prepared, and a sugar chain previously cleaved from a protein at the root by enzyme treatment is included. The reaction is performed with the sample solution, and only the sugar chain is covalently bonded to the carrier to completely wash away contaminants, thereby purifying only the sugar chain. The captured sugar chains are separated from the beads under selective conditions and subjected to mass spectrometry.

  In the present invention, "sugar chain" refers to monosaccharides such as glucose, galactose, mannose, fucose, xylose, N-acetylglucosamine, N-acetylgalactosamine, sialic acid, and derivatives thereof, which are linked in a chain by glycosidic linkage. Is a generic term for

  In the present invention, the “N-linked sugar chain” is a sugar chain bonded to the nitrogen atom of the amide group of the side chain possessed by the asparagine residue of the protein among the sugar chains of glycoproteins. , N-type sugar chain, asparagine-linked sugar chain. In the present invention, as a method for releasing N-linked sugar chains from glycoproteins in blood, for example, N-glycosidase F (also called glycopeptidase, PNGase, glycanase, glycoamidase etc.), glycopeptidase A, etc. Although the enzyme method which used these, and a hydrazine decomposition method can be illustrated concretely, the enzyme method by N- glycosidase F can be mentioned as a good example especially. At that time, protease such as trypsin can also be used in combination. Further, in the present invention, the method for purifying liberated sugar chains is not particularly limited as long as it is a method for selectively capturing and purifying sugar chains from a mixture in a sample, but a MALDI-TOF-MS type analyzer Specific examples of the method using BlotGlyco (registered trademark) for MALDI (manufactured by Sumitomo Bakelite Co., Ltd.), which is a sugar chain supplemented bead optimized for high sensitivity measurement in

In the present invention, “MALDI-TOF-MS” is an abbreviation of Matrix Assisted Laser Desorption Ionization-Time-of-Flight (Mass Spectrometer), and a method of measuring mass based on time of flight using MALDI method. It is. Specifically, in the MALDI method, a sample is spotted on a plate, and then a matrix solution (2,5-Dihydroxybenzoic acid) is added, dried, made into a crystalline state, and large energy is given to the matrix by pulse laser irradiation. MALDI-TOF-MS is a method of desorbing sample-derived ions such as (M + H) ⁺ and (M + Na) ⁺ and matrix-derived ions. To measure the When an ion is accelerated at a constant acceleration voltage V, the mass of the ion is m, the velocity of the ion v, the charge number of the ion z, the elementary charge e, and the flight time of the ion t m / z can be represented by m / z = ² eVt ² / L ² .

  In the present invention, analyzers other than the MALDI-TOF-MS analyzer can also be used, and as an ion source, for example, electron ionization method, chemical ionization method, electric field desorption method, fast atom collision method, electrospray ionization Atmospheric pressure chemical ionization method can be used, and as an analysis method, for example, methods such as magnetic field deflection type, quadrupole type, ion trap type, Fourier transform ion cyclotron resonance type can be used. . Furthermore, the above analytical method and high performance liquid chromatography (HPLC) can be used in combination.

  In the present invention, a sugar chain having a peak mass-to-charge ratio (m / z) of 1362 by mass spectrometry using a MALDI-TOF-MS analyzer is a result of mass spectrometry by a MALDI-TOF-MS analyzer It is a sugar chain exhibiting a 1362 m / z mass spectrum peak, and its structural formula is presumed to be (Hex) 2 + (Man) 3 (GlcNAc) 2.

  In the present invention, a sugar chain having a peak mass-to-charge ratio (m / z) of 1525 according to mass spectrometry using a MALDI-TOF-MS analyzer is a result of mass spectrometry using a MALDI-TOF-MS analyzer. It is a sugar chain exhibiting a 1525 m / z mass spectrum peak, and its structural formula is presumed to be (Hex) 3 + (Man) 3 (GlcNAc) 2.

  In the present invention, a sugar chain having a peak mass-to-charge ratio (m / z) of 1566 according to mass spectrometry using a MALDI-TOF-MS analyzer is a result of mass spectrometry using a MALDI-TOF-MS analyzer. It is a sugar chain exhibiting a 1566 m / z mass spectrum peak, and its structural formula is presumed to be (Hex) 5 (HexNAc) 3.

  In the present invention, a sugar chain having a peak mass-to-charge ratio (m / z) of 1591 determined by mass spectrometry using a MALDI-TOF-MS analyzer is a result of mass spectrometry using a MALDI-TOF-MS analyzer. It is a sugar chain exhibiting a 1591 m / z mass spectrum peak, and its structural formula is estimated to be (HexNAc) 2 (dHex) 1+ (Man) 3 (GlcNAc) 2.

  In the present invention, a sugar chain having a peak mass-to-charge ratio (m / z) of 1607 according to mass spectrometry using a MALDI-TOF-MS analyzer is a result of mass spectrometry using a MALDI-TOF-MS analyzer. It is a sugar chain exhibiting a 1607 m / z mass spectrum peak, and its structural formula is presumed to be (Hex) 1 (HexNAc) 2+ (Man) 3 (GlcNAc) 2.

  In the present invention, a sugar chain having a peak mass-to-charge ratio (m / z) of 1648 according to mass spectrometry using a MALDI-TOF-MS analyzer is a result of mass spectrometry using a MALDI-TOF-MS analyzer. It is a sugar chain exhibiting a 1648 m / z mass spectrum peak, and its structural formula is presumed to be (HexNAc) 3 + (Man) 3 (GlcNAc) 2.

  In the present invention, a sugar chain having a peak mass-to-charge ratio (m / z) of 1687 according to mass spectrometry using a MALDI-TOF-MS analyzer is a result of mass spectrometry using a MALDI-TOF-MS analyzer. It is a sugar chain exhibiting a 1687 m / z mass spectrum peak, and its structural formula is presumed to be (Hex) 4 + (Man) 3 (GlcNAc) 2.

  In the present invention, a sugar chain having a peak mass-to-charge ratio (m / z) of 1709 determined by mass spectrometry using a MALDI-TOF-MS analyzer is a result of mass spectrometry using a MALDI-TOF-MS analyzer. It is a sugar chain exhibiting a 1709 m / z mass spectrum peak, and its structural formula is presumed to be (Hex) 1 (HexNAc) 1 (NeuAc) 1+ (Man) 3 (GlcNAc) 2.

  In the present invention, a sugar chain having a peak mass-to-charge ratio (m / z) of 1769 according to mass spectrometry using a MALDI-TOF-MS analyzer is a result of mass spectrometry using a MALDI-TOF-MS analyzer. It is a sugar chain exhibiting a 1769 m / z mass spectrum peak, and its structural formula is estimated to be (Hex) 2 (HexNAc) 2+ (Man) 3 (GlcNAc) 2.

  In the present invention, a sugar chain having a peak mass-to-charge ratio (m / z) of 1794 by mass spectrometry using a MALDI-TOF-MS analyzer is a result of mass spectrometry by a MALDI-TOF-MS analyzer It is a sugar chain exhibiting a 1794 m / z mass spectrum peak, and its structural formula is presumed to be (HexNAc) 3 (dHex) 1+ (Man) 3 (GlcNAc) 2.

  In the present invention, a sugar chain having a peak mass-to-charge ratio (m / z) of 1849 according to mass spectrometry using a MALDI-TOF-MS analyzer is a result of mass spectrometry using a MALDI-TOF-MS analyzer. , 1849 m / z mass spectral peak, and its structural formula is presumed to be (Hex) 5 + (Man) 3 (GlcNAc) 2.

  In the present invention, a sugar chain having a peak mass-to-charge ratio (m / z) of 1871 by mass spectrometry using a MALDI-TOF-MS analyzer is a result of mass spectrometry by a MALDI-TOF-MS analyzer It is a sugar chain exhibiting a 1871 m / z mass spectrum peak, and its structural formula is presumed to be (Hex) 2 (HexNAc) 1 (NeuAc) 1+ (Man) 3 (GlcNAc) 2.

  In the present invention, a sugar chain having a peak mass-to-charge ratio (m / z) of 1915 according to mass spectrometry using a MALDI-TOF-MS analyzer is a result of mass spectrometry using a MALDI-TOF-MS analyzer It is a sugar chain exhibiting a 1915 m / z mass spectrum peak, and its structural formula is estimated to be (Hex) 2 (HexNAc) 2 (dHex) 1+ (Man) 3 (GlcNAc) 2.

  In the present invention, a sugar chain having a peak mass-to-charge ratio (m / z) of 2011 by mass spectrometry using a MALDI-TOF-MS analyzer is the result of mass spectrometry by a MALDI-TOF-MS analyzer It is a sugar chain exhibiting a 2011 m / z mass spectrum peak, and its structural formula is presumed to be (Hex) 6 + (Man) 3 (GlcNAc) 2.

  In the present invention, a sugar chain having a peak mass-to-charge ratio (m / z) of 2033 by mass spectrometry using a MALDI-TOF-MS analyzer is a result of mass spectrometry by a MALDI-TOF-MS analyzer It is a sugar chain exhibiting a 2033 m / z mass spectrum peak, and its structural formula is estimated to be (Hex) 3 (HexNac) 1 (NeuAc) 1+ (Man) 3 (GlcNAc) 2.

  In the present invention, a sugar chain having a mass-to-charge ratio (m / z) of 2058 of peaks determined by mass spectrometry using a MALDI-TOF-MS analyzer is a result of mass spectrometry using a MALDI-TOF-MS analyzer. It is a sugar chain exhibiting a 2058 m / z mass spectrum peak, and its structural formula is presumed to be (Hex) 1 (HexNAc) 2 (dHex) 1 (NeuAc) 1+ (Man) 3 (GlcNAc) 2.

  In the present invention, a sugar chain having a peak mass-to-charge ratio (m / z) of 2074 by mass spectrometry using a MALDI-TOF-MS analyzer is a result of mass spectrometry using a MALDI-TOF-MS analyzer. It is a sugar chain exhibiting a 2074 m / z mass spectrum peak, and its structural formula is estimated to be (Hex) 2 (HexNAc) 2 (NeuAc) 1+ (Man) 3 (GlcNAc) 2.

  In the present invention, a sugar chain having a peak mass-to-charge ratio (m / z) of 2118 determined by mass spectrometry using a MALDI-TOF-MS analyzer is a result of mass spectrometry using a MALDI-TOF-MS analyzer. It is a sugar chain exhibiting a 2118 m / z mass spectrum peak, and its structural formula is estimated to be (Hex) 2 (HexNAc) 3 (dHex) 1+ (Man) 3 (GlcNAc) 2.

  In the present invention, a sugar chain having a peak mass-to-charge ratio (m / z) of 2220 according to mass spectrometry using a MALDI-TOF-MS analyzer is a result of mass spectrometry using a MALDI-TOF-MS analyzer , 2219.8 m / z mass spectrum peak, and its structural formula is estimated to be (Hex) 2 (HexNAc) 2 (dHex) 1 (NeuAc) 1+ (Man) 3 (GlcNAc) 2.

  In the present invention, a sugar chain having a peak mass-to-charge ratio (m / z) of 2337 according to mass spectrometry using a MALDI-TOF-MS analyzer is a result of mass spectrometry using a MALDI-TOF-MS analyzer. It is a sugar chain exhibiting a 2337 m / z mass spectrum peak, and its structural formula is presumed to be (Hex) 3 (HexNAc) 4+ (Man) 3 (GlcNAc) 2.

  In the present invention, a sugar chain having a peak mass-to-charge ratio (m / z) of 2423 by mass spectrometry using a MALDI-TOF-MS analyzer is a result of mass spectrometry by a MALDI-TOF-MS analyzer It is a sugar chain exhibiting a 2423 m / z mass spectrum peak, and its structural formula is presumed to be (Hex) 2 (HexNAc) 3 (dHex) 1 (NeuAc) 1+ (Man) 3 (GlcNAc) 2.

  In the present invention, a sugar chain whose mass-to-charge ratio (m / z) of the peak by mass spectrometry using a MALDI-TOF-MS analyzer is 2525 is a result of mass spectrometry by a MALDI-TOF-MS analyzer It is a sugar chain exhibiting a 2525 m / z mass spectrum peak, and its structural formula is presumed to be (Hex) 2 (HexNAc) 2 (dHex) 1 (NeuAc) 2+ (Man) 3 (GlcNAc) 2.

  In the present invention, a sugar chain having a peak mass-to-charge ratio (m / z) of 2728 by mass spectrometry using a MALDI-TOF-MS analyzer is a result of mass spectrometry by a MALDI-TOF-MS analyzer It is a sugar chain exhibiting a 2728 m / z mass spectrum peak, and its structural formula is presumed to be (Hex) 2 (HexNAc) 3 (dHex) 1 (NeuAc) 2+ (Man) 3 (GlcNAc) 2.

  In the present invention, a sugar chain having a peak mass-to-charge ratio (m / z) of 2744 according to mass spectrometry using a MALDI-TOF-MS analyzer is a result of mass spectrometry using a MALDI-TOF-MS analyzer. It is a sugar chain exhibiting a 2744 m / z mass spectrum peak, and its structural formula is estimated to be (Hex) 3 (HexNAc) 3 (NeuAc) 2+ (Man) 3 (GlcNAc) 2.

  In the present invention, a sugar chain having a peak mass-to-charge ratio (m / z) of 3109 according to mass spectrometry using a MALDI-TOF-MS analyzer is a result of mass spectrometry using a MALDI-TOF-MS analyzer The sugar chain is a sugar chain exhibiting a 3109 m / z mass spectrum peak, and its structural formula is presumed to be (Hex) 4 (HexNAc) 4 (NeuAc) 2+ (Man) 3 (GlcNAc) 2.

  In the present invention, a sugar chain having a peak mass-to-charge ratio (m / z) of 3341 according to mass spectrometry using a MALDI-TOF-MS analyzer is a result of mass spectrometry using a MALDI-TOF-MS analyzer. It is a sugar chain exhibiting a 3341 m / z mass spectrum peak, and its structural formula is estimated to be (Hex) 3 (HexNAc) 3 (dHex) 2 (NeuAc) 3 (Man) 3 (GlcNAc) 2.

In the present invention, the following N-linked sugar chain can be used as a marker for identifying urothelial cancer.

(Wherein ○ represents galactose (Gal), ◇ represents N-acetylneuraminic acid (NeuAc), ● represents mannose (Man), ■ represents N-acetylglucosamine (GlcNAc), and Δ represents deoxyhexose (dHex). Left bracket indicates binding to any N-terminal branched sugar chain)

Further, among the above-mentioned N-linked sugar chains, the following N-linked sugar chains can be used particularly as a identification marker for upper urothelial cancer (renal pelvis and ureteral cancer).
(Wherein, 、, 、, 、, ■, は are the same as above; the left parenthesis indicates binding to a sugar chain at any of the N-terminal branched ends)

  In the present invention, the urothelial cancer identification marker can also be detected using an antibody or lectin.

  In the present invention, urothelial cancer is a cancer that originates from a mucous membrane called urothelium (transitional epithelium) that forms the inside of the urinary tract leading from the renal pelvis to the ureter, the bladder, and part of the urethra. And both upper urothelial carcinomas (renal and ureteral carcinomas). The marker for identifying urothelial cancer of the present invention can be used for examination / determination of any of bladder cancer and upper urothelial cancer (renal pelvis and ureteral cancer). Is preferred.

  Patients with urothelial cancer patients sera of 212 combined with age and sex (177 for bladder cancer, 35 for renal pelvic and ureteral cancer) (114 for early cancer, 98 for advanced cancer) and serum of 212 healthy volunteers Comprehensive analysis of N-linked sugar chains was performed by the glycoblotting method. The quantitative profile was obtained by mass spectrometry (MALDI-TOF MS) after capture and purification of N-linked sugar chains in serum. The amount of expression of each sugar chain was an absolute quantitative value of the sugar chain calculated from the area ratio on the spectrum with the known amount of oligosaccharide added as an internal standard, and the peak intensity observed by MALDI-TOF MS. Among the detected sugar chains, pay attention to the expression level of sugar chains m / z 1687, 1871, 2011, 2337 and sugar chains m / z 1566, 1769, which increase in patients with urothelial carcinoma, and receiver operation characteristics ( From the ROC curve, the cut off value of positive / negative judgment was calculated for each sugar chain. The selected sugar chains were as follows.

(Wherein, 、, 、, 、, ■, は are the same as above; the left parenthesis indicates binding to a sugar chain at any of the N-terminal branched ends)

  The number of positive sugar chains was defined as the N-glycan score based on the cut-off value, and the N-glycan score was significantly higher in patients with urothelial cancer (healthy volunteers: mean 1.406 vs. urine). Lung cancer patients: Average 4.425).

  In addition, when the N-glycan score is plotted on the vertical axis and the subject is plotted on the horizontal axis in ascending order of scores, it is shown that the number of patients with urothelial cancer increases clearly with the N-glycan score of 4 points or more. When the diagnostic accuracy of urothelial cancer by N-glycan score was examined by ROC curve, a significantly high value was obtained (AUC = 0.95).

  When the N-glycan score is 3 points or more, 4 points or more, or 5 points or more, the urothelial cancer diagnosis positive rate is 82.8% (197/238 persons) or 94.1% (160/170 persons). , 99.1% (115/116), suggesting that urothelial cancer can be differentiated by a minimally invasive method. By the way, urine cytology is used as a minimally invasive diagnostic marker in the clinic, but the positive rate of urine cytology collected at the same time in patients with urothelial cancer was 39% (83/212 patients). That is, the N-glycan score is a value that increases in the presence of urothelial cancer, and can be expected as a diagnostic marker for finding urothelial cancer by blood collection.

  Furthermore, the N-linked sugar chain to be detected and used in the method of the present invention also includes a sugar chain having a bisecting structure with a 4-branched structure or a 3-branched structure. The “four-branched structure” means that the sugar chain is branched into two from α1 → 6 linked mannose in the core structure of the sugar chain, and further by β1 → 6 and β1 → 2 linked N-acetylglucosamine, and Of the structures, α1 → 3 linked mannose means a structure in which a sugar chain is branched into two by β1 → 4 and β1 → 2 linked N-acetylglucosamine. In addition, “three-branch bisecting structure” refers to GlcNAc (ie bisecting GlcNAc) in which one of the four branched chains is missing and β1 → 4 is bonded to the end of the main chain of the sugar chain. Means a sugar chain having a structure in which

  In the present invention, by performing a hematuria test in addition to the above-described N-glycan score, it is possible to further enhance the determination accuracy of urothelial cancer. Hematuria is the most common symptom of urothelial cancer, and the main complaint is hematuria in 13% to 28% of urothelial cancer patients.

  The present invention also provides a kit for monitoring the onset of urothelial cancer, comprising means for measuring N-linked sugar chains of a blood sample obtained from a subject. Specifically, the kit for monitoring the onset of urothelial cancer of the present invention comprises a means for receiving a blood sample from a subject, a means for measuring an N-linked sugar chain contained in the blood sample, and an N-linked type measured. The method includes a means for comparing sugar chains with a reference value, and a means for determining whether the subject has developed urothelial cancer from comparison with the reference value.

The present invention also provides a method for diagnosing urothelial cancer, which comprises screening for N-linked sugar chains in a blood sample collected from a subject. Here, the N-linked sugar chain to be screened is at least one of the following N-linked sugar chains.

(Wherein, 、, 、, 、, ■, は are the same as above; the left parenthesis indicates binding to a sugar chain at any of the N-terminal branched ends)

  EXAMPLES Hereinafter, the present invention will be more specifically described by way of examples, but the present invention is not limited by these examples.

  Between March 2007 and September 2015, 1264 patients with urothelial cancer were treated at Hirosaki University Hospital and Tsugaru General Hospital. Among these, serum samples were collected from 278 patients (230 for bladder cancer and 48 for renal pelvic and ureteral cancer) before treatment and stored at -80 ° C. The stage of the tumor was determined according to the 2009 Tumor-Node-Metastasis classification. Control samples were obtained from healthy volunteers at Akita University Hospital and Hirosaki University Hospital. The serum bank consisted of a total of 358 volunteers, which were also stored at -80 ° C. Because of the retrospective nature of the study, patient background, especially age and gender, varied between groups. A propensity score matching strategy was used to adjust the patient's background and ensure the effectiveness of serum N-glycan comparisons between patients and healthy volunteers. Propensity scores were calculated by logistic analysis, and the data used for analysis included age and gender. Finally, a pair of 212 patients with urothelial cancer (the urothelial cancer patient group) and 212 healthy volunteers (the healthy volunteer group) participated in this study. Compare the N-glycan concentration (μM), detect the presence of urothelial carcinoma in the healthy volunteer group and the urothelial cancer patient group, and select the candidate N-glycans under the area under the curve (area-under-the-curve) Selected by AUC), and the optimal cutoff value was determined by receiver operating characteristic (ROC) curve. Candidate N-glycans were defined as positive when the concentration was above or below the cut-off value. Based on positive determinations, N-glycan scores were obtained by adding the positive number of candidate N-glycans. Diagnostic accuracy of the N-glycan score was assessed by ROC analysis.

Glyco blotting and mass spectrometry
^{SweetBlot TM (System Instruments, Hachijo,} Japan) using a serum N- glycan analysis was performed according to the methods described in the literature (Nishimura S et al High-throughput protein glycomics:. Combined use of chemoselective glycoblotting and MALDI-TOF / TOF mass spectroscopy. Angewandte Chemie (International ed in English). 2004; 44: 91-6). Briefly, serum samples (10 μL) containing 40 pmol of internal standard disialo-galactosylated bifurcated N-glycan (with amidated sialic acid (A2 amidoglycan)) were each reduced to DTT and iodoacetamide (sum It was alkylated using photo pure drug). The resulting mixture was then trypsinized and heat inactivated. After cooling to room temperature, peptide N-glycanase F (New England BioLabs, Ipswich, Mass., USA) was added to the mixture to release all N-glycans in the serum. After incubation for 360 minutes at 37 ° C., the resulting 20 μL mixture was equivalent to 2.5 μL serum. Aliquots of each pretreated sample are mixed with 500 μL of BlotGlyco H beads (Sumitomo Bakelite, Co., Tokyo, Japan) and glycans by stable hydrazone binding on MultiScreen Solvinert ^R filter plates (MerkMillipore, Billerica, MA, USA) Was captured. Subsequently, acetyl capping of unreacted hydrazide functional groups on the beads and methyl esterification of salicylic acid carboxyl groups (present at the end of captured glycans) were sequentially performed. Staged cleaning was performed before each step. The captured N-glycans were labeled with benzyloxyamine (BOA, Sigma-Aldrich, St. Louis, MO, USA) by transimination. Matrix-assisted laser desorption / ionization time-of-flight (MALDI-TOF) mass spectrometry (Ultraflex 3 TOF / TOF mass spectrometer, Bruker Daltonics, Bremen, Germany) It detected using. The composition and structure of glycans were estimated using GlycoMod Tool (http://br.expasy.org/tools/glcomod). A quantitative reproducibility test of Sweetblot was conducted.

Statistical analysis
Statistical analysis of clinical data was performed using SPSS v. 22.0 (IBM Corporation, Armonk, NY, USA) and GraphPad Prism v. 5.03 (GraphPad Software, San Diego, CA, USA). Categorical variables were reported as percentages and compared using Fisher's exact test. Quantitative data were expressed as median of quartiles 1 and 3 (Q1, Q3). Student's t test was used for data of normal distribution, and Mann-Whitney's U test was used for data showing nonnormal distribution, and differences between groups were compared statistically. The differences between the three groups were compared statistically using the ANOVA (Kruskal-Wallis) test. The optimal cutoff value of N-glycan intensity for detection of urothelial carcinoma was calculated from the receiver operating characteristic (ROC) curve using the following formula: (1- sensitivity) ² + (1-specificity Degree ² ). Multivariate analysis using a logistic regression model identified independent factors affecting urothelial cancer diagnosis. After simultaneously adjusting for potential confounders, the odds ratio (OR) of the 95% confidence interval (95% CI) was calculated.

From the result data, 212 patients with urothelial carcinoma and 212 healthy volunteers as controls were paired. The subject's background was as shown in Table 1 below.

  There were no significant differences between the healthy volunteer group and the urothelial cancer group in terms of age, gender and complications. Most (n = 177, 83%) patients in the urothelial carcinoma group suffer from bladder cancer (BC), and half (n = 114, 54%) of patients suffer from non-muscular infiltration disease The

  Seventy BOA-labeled N-glycans in serum samples were identified by serum glycomics using glycoblotting and mass spectrometry. After quantitative reproducibility tests, statistical analysis performed with 36 N-glycans showed good quantitative reproducibility. Among these, six N-glycans (m / z 1566, 1687, 1769, 1871, 2011, 2337) that exhibited an AUC higher than 0.70 by ROC curve analysis were selected (FIG. 1). Table 2 below shows the AU and optimal cutoff values for these 6 N-glycans as well as the AU of the non-selected N-glycans.

Note: * indicates what was identified in previous studies.

  N-glycan levels of high-mannose type (m / z 1687, 2011), terminal sialylated hybrid type (m / z 1871) and complex type with three- or four-branched glycans (m / z 2337) N-glycan levels of hybrid (m / z 1566) and complex (m / z 1769) with bifurcated glycans were significantly increased in patients with urothelial carcinoma, while they were significantly increased in patients with urothelial carcinoma. There was a significant reduction (FIGS. 2A-2F). Optimal cutoff values for these six N-glycans were calculated by ROC analysis (Table 2). N-glycans were identified as positive if the concentration of N-glycans was higher (m / z 1687, 1871, 2011, and 2337) or lower (m / z 1566 and 1769) than the respective cut-off value . The total number of positive N-glycans associated with urothelial cancer (N-glycan score) was compared between the healthy volunteer group and the urothelial cancer group. The median N-glycan score was significantly higher in the urothelial cancer group compared to the healthy volunteer group (5.0 vs. 1.0, P <0.001, respectively) (FIG. 3A). Between bladder cancer patients and upper urothelial cancer patients (Figure 3B), between localized (Ta-1N0M0) and advanced (T2 or higher) urothelial cancer patients (Figure 3C), or between urinary cytology classifications There were no significant differences in N-glycan scores in (Figure 3D). Also, the bar plot of the N-glycan score showed that the urothelial cancer patients had a higher N-glycan score as compared to the healthy volunteers (FIG. 4). The predictive value of N-glycan score for urothelial carcinoma was significant at AUC values of 0.95 (95% CI: 0.93-0.97, P <0.001) (Figure 5). The optimal cutoff value was defined as 2.5 at 93% sensitivity and 81% specificity (Table 3).

The following is a representative 36 N-glycans [m / z 2176 N-glycans that showed quantitative reproducibility in all the samples are internal standard disialo-galactosylated bi-branched N-glycans (amidated sialic acid A) (with A2 amidoglycan); composition and deduced structure are given in abbreviations].
Peak number m / z composition
1 1362.5 (Hex) 2 + (Man) 3 (GlcNAc) 2
2 1524.5 (Hex) 3 + (Man) 3 (GlcNAc) 2
3 1565.6 (Hex) 5 (HexNAc) 3
4 1590.6 (HexNAc) 2 (dHex) 1 + (Man) 3 (GlcNAc) 2
5 1606.6 (Hex) 1 (HexNAc) 2 + (Man) 3 (GlcNAc) 2
6 1647.6 (HexNAc) 3 + (Man) 3 (GlcNAc) 2
7 1686.6 (Hex) 4 + (Man) 3 (GlcNAc) 2
8 1708.6 (Hex) 1 (HexNAc) 1 (NeuAc) 1 + (Man) 3 (GlcNAc) 2
9 1752.6 (Hex) 1 (HexNAc) 2 (dHex) 1 + (Man) 3 (GlcNAc) 2
10 1768.6 (Hex) 2 (HexNAc) 2 + (Man) 3 (GlcNAc) 2
11 1793.7 (HexNAc) 3 (dHex) 1 + (Man) 3 (GlcNAc) 2
12 1809.7 (Hex) 1 (HexNAc) 3 + (Man) 3 (GlcNAc) 2
13 1848.6 (Hex) 5 + (Man) 3 (GlcNAc) 2
14 1854.7 (Hex) 1 (HexNAc) 1 (dHex) 1 (NeuAc) 1 + (Man) 3 (GlcNAc) 2
15 1870.7 (Hex) 2 (HexNAc) 1 (NeuAc) 1 + (Man) 3 (GlcNAc) 2
16 1914.7 (Hex) 2 (HexNAc) 2 (dHex) 1 + (Man) 3 (GlcNAc) 2
17 1955.7 (Hex) 1 (HexNAc) 3 (dHex) 1 + (Man) 3 (GlcNAc) 2
18 2010.7 (Hex) 6 + (Man) 3 (GlcNAc) 2
19 2032.7 (Hex) 3 (HexNac) 1 (NeuAc) 1 + (Man) 3 (GlcNAc) 2
20 2057.8 (Hex) 1 (HexNAc) 2 (dHex) 1 (NeuAc) 1 + (Man) 3 (GlcNAc) 2
21 2073.8 (Hex) 2 (HexNAc) 2 (NeuAc) 1 + (Man) 3 (GlcNAc) 2
2175.8 Internal standard (BOA labeled A2 amide)
22 2219.8 (Hex) 2 (HexNAc) 2 (dHex) 1 (NeuAc) 1 + (Man) 3 (GlcNAc) 2
23 2336.9 (Hex) 3 (Hex NAc) 4 + (Man) 3 (GlcNAc) 2
24 2378.9 (Hex) 2 (HexNAc) 2 (NeuAc) 2 + (Man) 3 (GlcNAc) 2
25 2524.9 (Hex) 2 (HexNAc) 2 (dHex) 1 (NeuAc) 2 + (Man) 3 (GlcNAc) 2
26 2728.0 (Hex) 2 (HexNAc) 3 (dHex) 1 (NeuAc) 2 + (Man) 3 (GlcNAc) 2
27 2744.0 (Hex) 3 (HexNAc) 3 (NeuAc) 2 + (Man) 3 (GlcNAc) 2
28 2890.1 (Hex) 3 (HexNAc) 3 (dHex) 1 (NeuAc) 2 + (Man) 3 (GlcNAc) 2
29 3049.1 (Hex) 3 (HexNAc) 3 (NeuAc) 3 + (Man) 3 (GlcNAc) 2
30 3109.1 (Hex) 4 (HexNAc) 4 (NeuAc) 2 + (Man) 3 (GlcNAc) 2
31 3195.2 (Hex) 3 (HexNAc) 3 (dHex) 1 (NeuAc) 3 + (Man) 3 (GlcNAc) 2
32 3341.2 (Hex) 3 (HexNAc) 3 (dHex) 2 (NeuAc) 3 + (Man) 3 (GlcNAc) 2
33 3414.2 (Hex) 4 (HexNAc) 4 (NeuAc) 3 + (Man) 3 (GlcNAc) 2
34 3560.3 (Hex) 4 (HexNAc) 4 (dHex) 1 (NeuAc) 3 + (Man) 3 (GlcNAc) 2
35 3719.3 (Hex) 4 (HexNAc) 4 (NeuAc) 4 + (Man) 3 (GlcNAc) 2
36 3865.4 (Hex) 4 (HexNAc) 4 (dHex) 1 (NeuAc) 4 + (Man) 3 (GlcNAc) 2
In the above, Man is mannose, GlcNAc is N-acetylglucosamine, Hex is hexose, dHex is deoxyhexose, HexNAc is N-acetylhexosamine, and NeuAc is N-acetylneuraminic acid.

  In the same manner as in Example 1, the number of positive sugar chains is scored (0 to 7) from the similarly calculated cut-off value for positive hematuria in addition to positive N-glycan score, and I drew a ROC curve. As a result, in addition to the positive N-glycan score, AUC 0.98 (P <0.001) was obtained for positive hematuria (FIG. 6 and FIG. 7).

  When the cutoff was> 2.5, the sensitivity was 93% and the specificity was 81%. In the above table, sensitivity is the ratio of positive among those who are actually ill, and specificity is the ratio of negative among those who are not ill. In addition, 83% of positive predictive value (PPV; percentage of people who actually got positive among those who got positive) was negative predictive value (NPV; negative). The proportion of people who are not actually afflicted is 92%.

  The result is as shown in FIG. 8 in the Swimmers plot. In FIG. 8, from left to right along the horizontal axis, score 0, urothelial cancer 2.5%; score 1, urothelial cancer 15%; score 2, urothelial cancer 32%; score 3, urothelium Score 65%, score 4, urothelial carcinoma 91%; and score 5, urothelial carcinoma 98%, respectively.

  In the same manner as in Example 1, N-glycans of serum secreted glycoproteins, immunoglobulins, were comprehensively analyzed, and sugar chain changes characteristic of upper urothelial carcinoma were searched. N-glycans of immunoglobulins (a mixture of IgG1, IgG2, IgG3, IgG4, IgM and IgA) purified from serum were comprehensively analyzed. For purification of immunoglobulins from serum, 100 μL of serum was added to the Zeba spin plate, centrifuged at 1000 g for 2 minutes, and 100 μL of desalted serum was collected, and then 100 μL of desalted serum was added to the Melon gel spin plate column. After standing at room temperature for 5 minutes, the mixture was centrifuged at 1000 g for 2 minutes, and the column-passing fraction was collected as a serum immunoglobulin fraction (100 μL). Melon Gel adsorbs albumin, transferrin, etc. contained in large amounts in a body fluid sample, and Ig is recovered in the flow-through fraction. Since it is possible to recover only the immunoglobulin without elution under acidic conditions or contamination with the sugar chain derived from the anti-Ig antibody-immobilized column, it is useful for the sugar chain analysis of Ig. Comprehensive analysis of N-glycans of serum immunoglobulin fractions was performed using a pretreatment robot "SweetBlot" that automatically performs all the protocols of the glycoblotting method. The subject's background was as shown in Table 5 below.

  Thirty types of serum Ig-derived N-glycans shown in FIG. 9 were identified. Among these serum Ig-derived N-glycans, N-glycans that were found to be altered in upper urothelial cancer patients compared to healthy volunteers were examined. If the concentration of N-glycans is higher (m / z 1362, 1525, 1566, 1648, 1709, 1769, 2011, 2033, 2074, 2118, and 2525) or lower (m / z 2423) than the respective cutoff value N-glycans were identified as positive.

Discriminant analysis is performed with the presence or absence of upper urothelial carcinoma as an explanatory variable and the obtained N-glycan numerical value as a target variable, and a sugar chain useful for differentiating upper urothelial carcinoma with an obtained F value of 2.0 or more Selected. Furthermore, discriminant analysis was performed using sex, age, and each selected sugar chain value as an explanatory variable, and the sum of numerical values obtained by integrating specific prediction functions for each item was calculated as an N-glycan score. The prediction function was as follows.
N-glycan score = (Gen * 0.5446) + (Age * 0.0050) + (m / z 1362 * 3.9720) + (m / z 1525 * -4.0053) + (m / z 1566 * -0.2758) + (m / z z 1648 * -1.7947) + (m / z 1709 * 4.5967) + (m / z 1769 * 1.4764) + (m / z 1794 * 0.4042) + (m / z 1915 * 0.4942) + (m / z 2011 *- 3.7266) + (m / z 2033 *-1.2859) + (m / z 2074 *-11.2450) + (m / z 2118 *-1.6229) + (m / z 2423 * 1.4247) + (m / z 2525 * 5.3805) + (-0.6815)

Based on the obtained N-glycan score, the predictability of upper urothelial carcinoma was examined. The results are shown in FIGS. 11 to 13 and Table 6.
In the table, NPV shows negative predictive ability and PPV shows positive predictive ability, respectively.

  97.8% of patients with upper urothelial carcinoma were detectable with cutoff <−0.615 pt. Thus, N-glycan analysis of serum Ig identified sugar chain changes characteristic of upper urothelial carcinoma, suggesting that N-glycan score may be useful for prediction of upper urothelial carcinoma It was done.

In the same manner as in Example 1, N-glycans of serum secreted glycoproteins were comprehensively analyzed, and serum immunoglobulins having N-type sugar chain mutations associated with upper urothelial carcinoma were searched by lectin array. The subject's background was as shown in Table 7 below.

Immunoglobulin N-glycans purified from serum were comprehensively analyzed. The lectin array is a method for detecting an immunoglobulin having a sugar chain useful for differentiation of upper urothelial carcinoma on a chip on which a lectin that binds to a specific sugar chain structure is immobilized (FIG. 24). 20 μL of serum was diluted with 100 μL of Probing Solution (20 mM Tris-HCl buffer, pH 7.5 containing 1 mM CaCl ₂ , 1 mM MgCl ₂ ), and 100 μL was added to the wells of the recombinant lectin immobilized lectin array chip (Rexxam). Incubate for 70 minutes at room temperature with horizontal shaking. Aspirate the serum sample on the array chip with an aspirator, completely remove it, gently add 100 μL of Probing Solution for washing, gently shake the array chip, aspirate the solution with an aspirator, and then completely remove the solution Removed. The washing operation was repeated a total of three times. After washing, 100 μL of Probing Solution containing 100 μL of 1 μg / mL of a biotin-labeled anti-human Ig light chain antibody (RM129, Abcam, ab193185) was added to the wells of the array chip. Incubate for 60 minutes at room temperature with horizontal shaking. Aspirate the antibody solution on the array chip with an aspirator, completely remove it, gently add 100 μl of Probing Solution for washing, gently shake the array chip, aspirate the solution with an aspirator, and complete the solution completely Removed. The washing operation was repeated a total of three times. 100 μL of Probing Solution containing 100 μL of 1 μg / mL of Cy3-labeled-streptavidin (GE, PA 43001) was added to the wells of the array chip. Incubate for 30 minutes at room temperature with horizontal shaking. Aspirate the antibody solution on the array chip with an aspirator, completely remove it, gently add 100 μl of Probing Solution for washing, gently shake the array chip, aspirate the solution with an aspirator, and complete the solution completely Removed. The washing operation was repeated a total of three times. 100 μL of Probing Solution was added to the wells of the array chip, and the fluorescence intensity value (Mean Fluoresence intensity, MFI) was measured with an exposure time of 300 msec using an Evanescent fluorescence scanner (BIO-REX SCAN 200, Rexxam). The results for each immunoglobulin class (IgG1, IgG2, IgG3, IgG4, IgM, IgA, IgG (total) and Ig (total)) in the same serum sample are shown in FIG. A box and whisker plot in which the fluorescence intensity values obtained by the lectin array are corrected with the immunoglobulin (IgG1, IgG2, IgG3, IgG4, IgM and IgA mixture) concentration values plotted is shown in FIG. A value obtained by correcting the obtained fluorescence intensity value with immunoglobulin (IgG1, IgG2, IgG3, IgG4, IgM and IgA mixture) concentration, sex and age as an explanatory variable, and useful for differentiation of upper urothelial cancer The following lectins were selected as lectins: lectins recognizing high mannose structure (rGRFT, rOrysata, rCalsepa, rBC2L-A, rPALa), lectins recognizing Bisect structure (rF17AG), lectins recognizing Fucose (rAAL, rPA- II L, RS-Fuc), Lectin recognizing the Gal / GalNAc structure (rCNL, rPA-IL, rABA, rMOA). A discriminant analysis is performed with the presence or absence of upper urothelial carcinoma as an explanatory variable, gender, age and each selected lectin / Ig value as objective variables, and the total value obtained by integrating the obtained specific prediction functions is a lectin array score Calculated as The prediction function was as follows.
Lectin array score = (Gen * 0.1476) + (Age * 2.8688) ++ (rF17AG / Ig * 0.0697) + (rGRFT / Ig * 0.0697) + (rOrysata / Ig * 0.2151) + (rCalsepa / Ig * -0.2643) + (rBC2L-A / Ig * 0.1813) + (rPALa / Ig * 0.3208) + (rAAL / Ig * 0.1085) + (rPA-II L / Ig * -0.1079) + (RS-Fuc / Ig * -0.0346) + (rCNL / Ig * -0.0986) + (rPA-IL / Ig * 0.1583) + (rABA / Ig *-0.6347) + (rMOA / Ig * 0.4423) + (-0.9912)

Based on the obtained lectin array score, the predictability of upper urothelial carcinoma was examined. The results are shown in FIGS. 15-18 and Table 8.
In the table, NPV shows negative predictive ability and PPV shows positive predictive ability, respectively. The upper urothelial cancer patients with 74.6% were detectable with cutoff <-1.275 pt.

  Thus, as a result of lectin array, significant differences were observed in lectins that respond to high mannose, bisect, fucose, Gal / GalNAc, and it is suggested that lectin array score may be useful for prediction of upper urothelial cancer It was done.

  In the same manner as in Example 1, N-glycans of serum secreted glycoproteins were comprehensively analyzed, and sugar chain changes characteristic of urothelial cancer (bladder cancer + upper urothelial cancer) were searched. The subject's background was similar to Table 1.

  Thirty-seven serum-derived N-glycans shown in FIG. 19 were identified. Among these serum-derived N-glycans, N-glycans that were found to be altered in urothelial cancer patients compared to healthy volunteers were examined. Whether the concentration of N-glycans is higher (m / z 1591, 1607, 1648, 1687, 1709, 1794, 2011, 2074, 2744, 3109, 3341) or lower than each cutoff value (m / z 1566, 1769) In some cases, N-glycans were identified as positive.

Discriminant analysis was performed with the presence or absence of urothelial carcinoma as an explanatory variable, and the obtained N-glycan numerical value as an objective variable, and a sugar chain useful for differentiating urothelial carcinoma with an obtained F value of 2.0 or more was selected. . Furthermore, discriminant analysis was performed using sex, age, and each selected sugar chain value as an explanatory variable, and the sum of numerical values obtained by integrating specific prediction functions for each item was calculated as an N-glycan score. The prediction function was as follows.
N-glycan score = (Gen * 0.1676) + (Age * -0.0075) + (m / z 1565 * 0.6488) + (m / z 1590 * 0.3244) + (m / z 1607 * -0.0544) + (m / z z 1648 * -0.3051) + (m / z 1687 * -0.7940) + (m / z 1709 * -1.0849) + (m / z 1769 * 0.1752) + (m / z 1794 * -0.2878) + (m / z 1849 * 0.3176) + (m / z 1915 * 0.4443) + (m / z 2011 * -0.2800) + (m / z 2058 * 0.1916) + (m / z 2074 * 1.0742) + (m / z 2220 * -0.8861 ) (m / 2728 * 0.2749) + (m / z 2744 * -1.2235) + (m / z 3109 * 0.4362) + (m / z 3341 * 0.1149) + (2.2792)

Based on the obtained N-glycan score, the predictability of urothelial carcinoma was examined. The results are shown in FIGS. 21-23 and Table 9.
In the table, NPV shows negative predictive ability and PPV shows positive predictive ability, respectively. 95.28% of urothelial cancer patients were detectable with cutoff <0 pt.

  Thus, the sugar chain changes characteristic of urothelial cancer (bladder cancer + upper urothelial cancer) are identified by serum N-type sugar chain analysis, and the N-glycan score is urothelial cancer (bladder It is suggested that it may be useful for prediction of cancer + upper urothelial cancer).

  Between 2009 and 2016, serum samples from 237 patients with urothelial carcinoma and serum samples from 96 prostate cancer patients were collected from our serum bank as a control for other cancers. In addition, serum samples were also collected from 339 healthy volunteers in the community of the Iwaki Health Promotion Project's health maintenance program. All serum samples were stored at -80 ° C. The stage of the tumor was determined according to 2017 Tumor-Node-Metastasis classification (8th edition). Histological classification of urothelial carcinoma was performed according to World Health Organization 1973 and 2004 grading systems18. Patient data is shown in Table 10.

Note a: dNGS = diagnostic N-glycan score b: UC = urothelial cancer c: UCB = bladder cancer d: UTUC = upper urothelial cancer

Was added purification and quantification each serum 100μL of Ig fractions from serum, Zeba equilibrated with phosphate-buffered saline ^{TM Spin Desalting resin Plate (Thermo Fisher} Scientific, Waltham, MA, USA) , the 2 1000 × g It was centrifuged for a minute. Column-passing fractions were collected as buffer-exchanged serum (100 μL). Purification of Ig fraction, using ^{Melon TM Gel Spin Purification Kit (Thermo} Fisher Scientific), were performed according to the manufacturer's instructions. Buffer-exchanged serum (100 μL) was added to Melon Gel resin equilibrated with purification buffer. After incubating for 5 minutes, Melon Gel resin was centrifuged at 1000 × g for 2 minutes, and the flow-through fraction was collected as a purified Ig fraction (100 μL). A 10 μL aliquot of the Ig fraction was subjected to N-glycan analysis. In order to confirm the purification of the Ig fraction, the Melon Gel resin column-passed Ig fraction was subjected to SDS-PAGE and stained with CBB. The Ig (IgG1, IgG2, IgG3, IgG4, IgM and IgA) levels of the purified Ig fraction were measured using Bio-plex Pro Human Isotyping 6-plex kit (Bio-Rad Laboratories, CA, USA) according to the instruction manual. .

Glycoblotting method and subjected to N- glycan analysis according to the method of serum N- glycan analysis literature of Ig was performed using mass spectrometry (Nishimura S et al High-throughput protein glycomics:. Combined use of chemoselective glycoblotting and MALDI- TOF / TOF mass spectrometry. Angewandte Chemie (International ed in English). 2004; 44: 91-6). The 10μL aliquot of the purified Ig fraction, using ^{SweetBlot TM (System Instruments, Hachijo,} Japan) were analyzed by glycoblotting method. Then, the obtained benzyloxyamine (BOA) -labeled glycan is subjected to matrix-assisted laser desorption / ionization time-of-flight (MALDI-TOF) mass spectrometry (Ultraflex 3 TOF / TOF mass) Detection was performed using a spectrometer; Bruker Daltonics, Bremen, Germany) (Fig. 25a-f). The composition and structure of glycans were estimated using GlycoMod Tool (http://br.expasy.org/tools/glcomod). Then, quantitative reproducibility tests of each N-glycan level were evaluated.

Statistical analysis
Statistical analysis of clinical data was performed using SPSS v. 22.0 (IBM Corporation, Armonk, NY, USA) and GraphPad Prism v. 6.03 (GraphPad Software, San Diego, CA, USA). Categorical variables were reported as percentages and compared using Fisher's exact test. Age data were expressed as median of quartiles 25 and 75 (Q1, Q3). Student's t test was used for data of normal distribution, and Mann-Whitney's U test was used for data showing nonnormal distribution, and differences between groups were compared statistically. The differences between the three groups were compared statistically using the ANOVA (Kruskal-Wallis) test. Multivariate discriminant analysis for detection of urothelial carcinoma was performed by inputting urothelial carcinoma event as the explanatory variable and N-glycan level as the objective variable. Diagnostic N-glycan scores were calculated by multiplying the candidate N-glycan levels by each function value. Diagnostic performance of N-glycan score using receiver operating characteristic (ROC) curve analysis developed with library 'rms' in R (http://www.r-project.org/) Statistical differences between the area under the curve (area-under-the-curve; AUC) were evaluated using the same program. Differences of p <0.05 were considered statistically significant.

result
The diagnostic N-glycan score for urothelial carcinoma based on abnormal N-glycosylation of Ig was found to be far superior to classical urinary cytology and hematuria tests. SDS-PAGE analysis of total serum (FIG. 26b, lanes 1 to 4) and UIg fractions (FIG. 26b, lanes 5 to 8) showed that non-Ig proteins were effectively removed from the whole serum by Melon Gel chromatography It was shown. N-glycan analysis of Ig fractions (Figures 26a-e) identified 32 BOA-labeled N-glycans on Ig as described below (Figure 26f).

Thirty-two types of N-glycans that shown good quantitative reproducibility in all Igs fraction samples and could be analyzed statistically
Peak number m / z composition
1 1362.5 (Hex) 2 + (Man) 3 (GlcNAc) 2
2 1524.5 (Hex) 3 + (Man) 3 (GlcNAc) 2
3 1565.5 (Hex) 5 + (HexNAc) 3
4 1590.6 (HexNAc) 2 (dHex) 1 + (Man) 3 (GlcNAc) 2
5 1606.6 (Hex) 1 (HexNAc) 2 + (Man) 3 (GlcNAc) 2
6 1647.6 (HexNAc) 3 + (Man) 3 (GlcNAc) 2
7 1686.6 (Hex) 4 + (Man) 3 (GlcNAc) 2
8 1708.6 (Hex) 1 (HexNAc) 1 (NeuAc) 1 + (Man) 3 (GlcNAc) 2
9 1752.6 (Hex) 1 (HexNAc) 2 (dHex) 1 + (Man) 3 (GlcNAc) 2
10 1768.6 (Hex) 2 (HexNAc) 2 + (Man) 3 (GlcNAc) 2
11 1793.7 (HexNAc) 3 (dHex) 1 + (Man) 3 (GlcNAc) 2
12 1809.7 (Hex) 1 (HexNAc) 3 + (Man) 3 (GlcNAc) 2
13 1848.6 (Hex) 5 + (Man) 3 (GlcNAc) 2
14 1870.7 (Hex) 2 (HexNAc) 1 (NeuAc) 1 + (Man) 3 (GlcNAc) 2
15 1914.7 (Hex) 2 (HexNAc) 2 (dHex) 1 + (Man) 3 (GlcNAc) 2
16 1955.7 (Hex) 1 (HexNAc) 3 (dHex) 1 + (Man) 3 (GlcNAc) 2
17 2010.7 (Hex) 6 + (Man) 3 (GlcNAc) 2
18 2032.7 (Hex) 3 (HexNac) 1 (NeuAc) 1 + (Man) 3 (GlcNAc) 2
19 2057.8 (Hex) 1 (HexNAc) 2 (dHex) 1 (NeuAc) 1 + (Man) 3 (GlcNAc) 2
20 2073.8 (Hex) 2 (HexNAc) 2 (NeuAc) 1 + (Man) 3 (GlcNAc) 2
21 2117.8 (Hex) 2 (HexNAc) 3 (dHex) 1 + (Man) 3 (GlcNAc) 2
22 2219.8 (Hex) 2 (HexNAc) 2 (dHex) 1 (NeuAc) 1 + (Man) 3 (GlcNAc) 2
23 2336.9 (Hex) 3 (Hex NAc) 4 + (Man) 3 (GlcNAc) 2
IS 2348.9 Internal Standard (BOA-Labeled A2 Amide)
24 2378.9 (Hex) 2 (HexNAc) 2 (NeuAc) 2 + (Man) 3 (GlcNAc) 2
25 2422.9 (Hex) 2 (HexNAc) 3 (dHex) 1 (NeuAc) 1 + (Man) 3 (GlcNAc) 2
26 2524.9 (Hex) 2 (HexNAc) 2 (dHex) 1 (NeuAc) 2 + (Man) 3 (GlcNAc) 2
27 2727.9 (Hex) 2 (HexNAc) 3 (dHex) 1 (NeuAc) 2 + (Man) 3 (GlcNAc) 2
28 2743.9 (Hex) 3 (HexNAc) 3 (NeuAc) 2 + (Man) 3 (GlcNAc) 2
29 2890.1 (Hex) 3 (HexNAc) 3 (dHex) 1 (NeuAc) 2 + (Man) 3 (GlcNAc) 2
30 3049.1 (Hex) 3 (HexNAc) 3 (NeuAc) 3 + (Man) 3 (GlcNAc) 2
31 3195.2 (Hex) 3 (HexNAc) 3 (dHex) 1 (NeuAc) 3 + (Man) 3 (GlcNAc) 2
32 3341.2 (Hex) 3 (HexNAc) 3 (Deoxyhexose) 2 (NeuAc) 3 + (Man) 3 (GlcNAc) 2
In the above, m / z 2348.9 is an internal standard and is disialo-galactosylated 2-branched N-glycan (A2 amidoglycan) containing amidated sialic acid residues. Man is mannose, GlcNAc is N-acetylglucosamine, Hex is hexose, dHex is deoxyhexose, HexNAc is N-acetylhexosamine, and NeuAc is N-acetylneuraminic acid.

Background of subjects in the non-urothelial carcinoma group (healthy volunteers and prostate cancer) and urothelial carcinoma group was as shown in Table 11 below. There were no statistically significant differences in age, smoking history, history of BPH (benign prostatic hyperplasia) and history of calculi between the non-urothelial carcinoma group and the urothelial carcinoma group.
note:
a: IQR (Interquartile range) = interquartile range b: BPH (benign prostatic hyperplasia) = benign prostatic hyperplasia c: HSPC (hormone sensitive prostate cancer) = hormone-sensitive prostate cancer

  To detect urothelial cancer, multivariate discriminant analysis is performed by inputting urothelial cancer event as an explanatory variable and N-glycan level as an objective variable, and the highest sensitivity for detecting urothelial cancer Five specific N-glycans (complex biantennary type: m / z 1607, m / z 1769, m / z 2074; core fucosylated bisecting GlcNAc type: m / z 2118, m / z 2423) Were selected (Figures 25a-f, Table 12).

note:
a: ODF (one degree of freedom) = 1 degree of freedom
b: TDF (two degree of freedom) = 2 degrees of freedom

The asialo 2 branched N-glycans (m / z 1607 and m / z 2074) on Ig were significantly downregulated in the urothelial carcinoma group compared to the healthy volunteer / prostate cancer group. Only asialo-2 branched N-glycan (m / z 1769) on Ig was significantly downregulated in prostate cancer group compared to urothelial cancer group and healthy volunteer group. Core fucosylated bisecting GlcNAc N-glycan-containing monosialyl bisecting GlcNAc (m / z 2423) was not significantly changed in the urothelial carcinoma group and in the healthy volunteer group but significantly reduced in the prostate cancer group Was. In particular, asialo bisecting GlcNAc (m / z 2118) with core fucosylated N-glycans on Ig was significantly upregulated in the urothelial carcinoma group but was not detected in the healthy volunteer group and the prostate cancer group . However, total Ig levels in urothelial cancer patients were significantly lower than in healthy volunteers and prostate cancer patients (Figures 27a-h). According to the synthesis pathway of N-glycans shown in FIG. 25 f, the above results indicate that downregulation of asialo 2-branched N-glycans on Ig and accumulation of asialo bisecting GlcNAc with core fucosylated N-glycans is associated with urothelial carcinoma Suggest that it occurs as aberrant N-glycosylation of Ig associated with In order to apply the candidate of abnormal N-glycosylation profile on these Ig to detection of urothelial cancer, we calculate urothelial cancer diagnostic N-glycan score (dNGScore) calculated by the following formula Established:
(m / z 1607 serum level x 0.1 925) + (m / z 1769 serum level x 0.4 932) + (m / z 2074 serum level x 0.4 941) + (m / z 2118 serum level x-3.2460) + ( Serum level of m / z 2423 x 0.6179) + (-0.4905)

  dNGScore is more likely in patients with urothelial carcinoma (UC) (bladder cancer (UCB) and upper urothelial carcinoma (UTUC)) compared to non-urothelial carcinomas including healthy volunteers (HV) and prostate cancer (PC) It was significantly higher (Figures 28 and 29). ANG of dNGScore for prediction of urothelial cancer (0.969 for urothelial cancer, 0.993 for bladder cancer, 0.907 for upper urothelial cancer) is worse than hematuria (0.892) and urine cytology (0.707) High (Figure 30, Table 10). With a cutoff dNGScore (-0.0955) to predict urothelial carcinoma, the negative predictive value (NPV) is 96.1%, which is the urine cytology NPV (75.8%) and hematuria NPV (89.5%) It was much higher than that. The diagnostic N-glycan score was not significantly associated with hematuria and urine cytology classes (Figures 31 and 32) but was significantly associated with bladder cancer invasiveness (Figure 33). .

  The 26 N-glycan profiles obtained by glycoblotting and matrix assisted laser desorption / ionization time-of-flight mass spectrometry according to the present invention can be used as a biomarker indicating the presence of urothelial cancer.

Claims (15)

A marker for identifying urothelial cancer comprising at least one of the following N-linked sugar chains:
(Wherein ○ represents galactose (Gal), ◇ represents N-acetylneuraminic acid (NeuAc), ● represents mannose (Man), ■ represents N-acetylglucosamine (GlcNAc), and Δ represents deoxyhexose (dHex). Left bracket indicates binding to any N-terminal branched sugar chain) The urothelial cancer is upper urothelial cancer (renal pelvis and ureteral cancer), N-linked sugar chain
(Wherein ○, 、, 、, 、, は are as defined above; the left parenthesis indicates binding to a sugar chain at any of N-terminal branched ends) The urothelial cancer identification marker according to 1.   A method of testing urothelial cancer using the amount of the marker for identifying urothelial cancer according to claim 1 or 2 as an index in a sample collected from a subject. An amount of the urothelial cancer identification marker in a sample collected from a subject, and an amount of the urothelial cancer identification marker in a sample collected from a healthy subject, including the following steps (1) and (2): The method for testing urothelial cancer according to claim 3, characterized in that a difference or a ratio thereof is evaluated.
(1) measuring the amount of the urothelial cancer identification marker in the sample collected from the subject, and (2) collecting the amount of the urothelial cancer identification marker measured in the step and the healthy person Comparing the amount of the urothelial cancer identification marker in the sample to be evaluated and evaluating the difference or ratio between the two.   The test method according to claim 3 or 4, wherein the urothelial cancer identification marker is detected using an antibody or lectin.   The test method according to any one of claims 3 to 5, further comprising the step of testing hematuria from the sample collected from the subject. A method of determining urothelial cancer characterized in that the following steps (1) to (4) are sequentially provided:
(1) releasing N-linked sugar chains from glycoproteins in blood collected from a subject;
(2) a step of purifying the released sugar chain;
(3) mass spectrometry of the purified sugar chain;
(4) The mass-to-charge ratio (m / z) of the peak by mass spectrometry using a MALDI-TOF-MS analyzer is 1362, 1525, 1566, 1591, 1607, 1648, 1687, 1709, 1769, 1794, 1849 1871, 1915, 2011, 2033, 2058, 2074, 2118, 2220, 2337, 2423, 2525, 2728, 2744, 3109 and 3341. Among sugar chains corresponding to the above sugar chains, or Measuring the detected intensity of at least one sugar chain, wherein increasing or decreasing the detected intensity as compared to a healthy person indicates that the subject is a patient with urothelial cancer.   The urothelial carcinoma is upper urothelial carcinoma (renal and ureteral carcinoma), and the mass-to-charge ratio (m / z) of the peak by mass spectrometry using a MALDI-TOF-MS analyzer is 1362, 1525, The determination method according to claim 7, which is any one of 1566, 1648, 1709, 1769, 1794, 1914, 2011, 2033, 2074, 2118, 2423 and 2525.   The determination method according to claim 7, further comprising the step of examining hematuria from the sample collected from the subject. Use of the following N-linked sugar chain as a biomarker for monitoring the onset of urothelial carcinoma.
(Wherein, 、, 、, 、, ■, は are the same as above; the left parenthesis indicates binding to a sugar chain at any of the N-terminal branched ends) The urothelial cancer is upper urothelial cancer (renal pelvis and ureteral cancer), and N-linked sugar chain is
(Wherein ○, 、, 、, 、, は are as defined above; the left parenthesis indicates binding to a sugar chain at any of N-terminal branched ends) Use according to 10.   A kit for monitoring the onset of urothelial cancer, comprising means for measuring N-linked sugar chains of a blood sample obtained from a subject.   From the means for receiving the blood sample from the subject, the means for measuring the N-linked sugar chain contained in the blood sample, the means for comparing the measured N-linked sugar chain to the reference value, and the comparison with the reference value The kit for monitoring the onset of urothelial cancer development according to claim 12, comprising means for determining whether the subject has developed urothelial cancer. A method for diagnosing urothelial cancer, comprising screening the following N-linked sugar chain in a blood sample collected from a subject.
(Wherein, 、, 、, 、, ■, は are the same as above; the left parenthesis indicates binding to a sugar chain at any of the N-terminal branched ends) The diagnostic method according to claim 14, wherein the urothelial cancer is upper urothelial cancer (renal pelvis and ureteral cancer), and the following N-linked sugar chain is screened.
(Wherein, 、, 、, 、, ■, は are the same as above; the left parenthesis indicates binding to a sugar chain at any of the N-terminal branched ends)