KR101479207B1 - Glycan markers for human ovarian cancer - Google Patents

Abstract

The present invention relates to a human serum glycoman map and a human ovarian cancer marker obtained by verifying serum samples of ovarian cancer patients using the same. The present inventors have found that the N-sugar chains of human serum proteins are theoretically possible in many kinds of sugar chains, but actually N-sugar chains have only the structure as shown in Fig. 3, and the appearance of N- N-glycans with low probability of occurrence can be used as markers of specific diseases by revealing percentages.

Description

{Glycan markers for human ovarian cancer}

The present invention relates to human ovarian cancer markers excavated using a human serum sugar chain map.

Glicomics involves extensive research on glycan structure and properties. Glycans are difficult to analyze because of their complex structure because of the diversity of the monosaccharide structure, the degree of branching, and the connection structure. Unlike the protein structure, which is directly related to the discrete DNA template of the genome, the glycan structure is determined by a series of glycosyltransferase enzymatic reactions.

N-Glycosylation is the addition of oligosaccharides to the asparagine of the peptide backbone. This is governed by the biological principles that determine its location and structural diversity. Changes in N-glycosylation are the result of errors in normal glycosylation systems in diseases such as tumors. For this reason, attempts have been made to use N-sugar chains as disease markers.

N-sugar chains are very diverse, but are distinguished from other glycoconjugates such as O-sugar chains, glycosaminoglycans, and glycolipids. N-glycans are found only in asparagine followed by various amino acids followed by serine, threonine or a less common cysteine amino acid. Except for proline, various amino acids can come after asparagine. Enzymes such as the peptide N-glycosides F (PNGase F) can be used to cleave N-linked sugar chains under physiological conditions. The N-linked core consists of three mannose sugars attached to chitobiose core (Man ₃ GlcNAc ₂ ). N-linked sugar chains often fall into three main types, depending on which monosaccharide is added to the N-linked core. The High Mannose Glycan type is a structure in which six mannose monosaccharides are added to the N-linked core. The complex type sugar chain is enzymatically bound to N-acetylglucosamine (GlcNAc), galactose, dioxyhexose (fucose), and cialic acid monosaccharides after the digestion precursor is degraded in the N-linked core. The hybrid type sugar chain includes both the hypertonic structure and the hybrid structure.

The cellular biologic changes that occur during carcinogenesis and metastasis show that many kinds of glycoproteins or glycolipids present on the surface of the cell membrane undergo a specific signal such as an oncogene leading to "aberrant glycosylation" (1983), J. Natl. Cancer Inst., 71: 371-357, 1985. However, it has been suggested that the changes in the sugar chain caused by the changes in cell adhesion and recognition may lead to cell carcinogenesis and metastasis (Hakomori and Kannagi, 231-251; Feizi, 1985, Nature, 314: 53-571). When stimulation occurs outside the cell, the expression of the glycosyltransferase GnT-V (Nacetylglucosaminyltransferase) is enhanced through the signaling of the cancer gene ras and the transcription factor ets-1. It is known that GnT-V is an enzyme that catalyzes the addition of N-acetylglucosamine to the basic sugar chain of the glycoprotein at the positions of [beta] 1,6, and is directly related to the invasion and metastasis of cancer ( Dennis et al., 1987, Science, 236: 582-5853).

In general, the glycoprotein is synthesized after protein synthesis in the endoplasmic reticulum (ER) and then transferred to the Golgi body. As a result of diverse life events of the cells, the sugar is added by several sugar transferases. -Acetylglucosamine glycosyltransferase (GnT-V), which is catalyzed by six kinds of enzymes (I-Ⅵ), is known to be involved in the development and metastasis of cancer. have. GnT-V enzymes are located in the Golgi, and target proteins are secreted to the cell surface or to the cell surface by changes in sugar chains. At this time, the glycoprotein causes cancer by recognizing and adhering to the surface protein of the target cell.

Glycomics, a research on glycans that play a very important role in the function and recognition of proteins as they exist on the surface of proteins following genomics and proteomics, which is a recent study on the functions of genes and proteins, . It has been found that among glycocytes, a significant change occurs in the sugar chains of specific proteins in various tumor cells. Many researchers have been actively studying biomarkers for detecting the change of sugar chains and diagnosing tumors at an early stage.

A method of profiling glycans extracted from biological samples such as blood using a mass spectrometer has been widely used in research on digestion of biomarkers based on glycomics. However, there is still no method to visually show and confirm the overall biological relationship between the obtained quality control method and the obtained party to confirm the accuracy of the sugar extraction process from the sample.

Korean Patent No. 475642 discloses a diagnostic kit using a method for diagnosing cancer by measuring a sugar chain change of a protein which is involved in the development and metastasis of cancer and showing a change in N-linked sugar chain and a method for measuring sugar chain change of the protein will be.

U.S. Patent No. 7,651,847 (Jan. 26, 2010) discloses a method of identifying a tumor-specific oligosaccharide, a method of determining an individual tumor strain, and detecting the presence or absence of a specific tumor marker Or a method for diagnosing a tumor progression stage.

It is an object of the present invention to provide a sugar chain marker for diagnosing human ovarian cancer or diagnosing ovarian cancer progress using human blood sugar chain map constructed from human serum sample.

In order to solve the above problems, the present inventors analyzed glycans (average of about 80 glycans) separated and purified from blood of 200 or more and found glycans (neutral glycans: 10, acidic sugar chains: 8, total 18; Table 1) were selected and used as an index for quality control in sugar chain extraction and analysis.

In addition, the present inventors have completed sugar chain mapping of human blood by visualizing the biological relationship between sugars based on the sugar profile.

In the present invention, sugars extracted from 200 blood samples of normal and cancer patients (lung cancer patients and ovarian cancer patients) were profiled using a mass spectrometer. One sample was subjected to mass spectrometry five times and was identified as "presence" only if a specific peak was detected more than 5 times. As a result, blood samples contained an average of about 80 different sugars on an individual basis, and 18 of them were common in all 200 (Table 1). In Table 1, there were 10 types of sugar chains obtained from 10% acetonitrile fraction, four types of sugar chains obtained from 20% acetonitrile fraction, and eight types of sugar chains obtained from 40% acetonitrile fraction. Since there are four common sugar chains in the 20% fraction, a total of 18 kinds of sugar chains were detected constantly regardless of the presence of disease, individual health and environment.

Therefore, serum samples can be verified by using 18 kinds of sugar chains which are constantly detected regardless of the presence or absence of such diseases. If there is a sugar chain in any serum sample that can not be detected in any of these 18 types of sugar chains, the serum sample may indicate a problem in sample collection and processing.

In addition, there has been no way to suggest biological information on the order in which sugar chains are formed by using the correlation between saccharides in relation to the structure of sugar chains thus obtained. In addition, it is in the present situation that it is inconvenient because there is no reference data that defines the structure according to the composition of the sugar obtained from the blood sample.

In order to overcome this problem, the present inventors produced a biological map of the sugar structures derived from the glycosylation process in a human blood sample, and confirmed that there is a specific correlation and order among the sugar chains using the biological map. Tandem mass spectrometry (Tandem MS) confirmed the structure and structure of the sugar one by one to analyze why the structure of each molecular weight had to have such a structure, thus completing the sugar structure mapping in human serum 3, and enlarged portions are shown in Figs. 4 to 13). The sugar structure list shown in this sugar structure map is shown in Tables 8 to 10.

The present invention can be applied to all sugar analysis platforms, and can be used in biotechnology or medical fields such as analysis of characteristics of biopharmaceuticals including biosimilars, or identification of cancer markers.

In addition, with respect to the bioinformatics that analyzes the processed data, the present invention provides data that is a standard, so that it is possible to provide a common standard related to sugar in human serum, It may be a more clear method and reference for the analysis based on the composition and structure of the sugar from the serum.

The present invention

(Hex) ₃ (HexNAc) ₃ Theoretical single isotope mass value 1136.4,

(Hex) ₃ (HexNAc) ₄ Theoretical Single Isotope Mass Value 1339.5,

(Hex) ₃ (HexNAc) ₅ (Fuc) ₁ Theoretical single isotope mass value 1688.6,

(Hex) ₄ (HexNAc) ₅ Theoretical single isotope mass value 1704.6,

(Hex) ₄ (HexNAc) ₃ (NeuAc) ₁ Theoretical single isotope mass value 1611.5,

(Hex) ₆ (HexNAc) ₃ (NeuAc) ₁ Theoretical single isotope mass values of 1935.6 and

(Hex) ₄ (HexNAc) ₅ (NeuAc) ₁ Theoretical single isotope mass value of 2017.7.

In addition, the present invention is characterized in that the marker is expressed at a frequency higher than 20% in a patient group sample as compared with a normal group sample.

In addition,

(Hex) ₃ (HexNAc) ₄ Theoretical Single Isotope Mass Value 1339.5,

(Hex) ₃ (HexNAc) ₄ (Fuc) ₁ Theoretical single isotope mass value 1485.5,

(Hex) ₃ (HexNAc) ₅ Theoretical Single Isotope Mass Value 1542.6,

(Hex) ₄ (HexNAc) ₅ Theoretical single isotope mass value 1704.6,

(Hex) ₅ (HexNAc) ₄ (Fuc) ₁ Theoretical single isotope mass value 1809.6,

(Hex) ₄ (HexNAc) ₃ Theoretical single isotope mass value 1298.45,

(Hex) ₆ (HexNAc) ₅ (NeuAc) ₁ Theoretical single isotope mass values of 2295.8 and

The present invention provides a human ovarian cancer intensity marker comprising one or more N-glycans of the Hex ₆ (HexNAc) ₅ (NeuAc) ₂ theoretical single isotope mass value of 2608.9.

In addition,

(Hex) ₃ (HexNAc) ₄ Theoretical Single Isotope Mass Value 1339.5,

(Hex) ₃ (HexNAc) ₄ (Fuc) ₁ Theoretical single isotope mass value 1485.5,

(Hex) ₃ (HexNAc) ₅ Theoretical Single Isotope Mass Value 1542.6,

(Hex) ₄ (HexNAc) ₅ Theoretical single isotope mass value 1704.6,

(Hex) ₄ (HexNAc) ₃ Theoretical single isotope mass values of 1298.45 and

(Hex) ₃ (HexNAc) ₄ (Fuc) ₁ The N-glycans of the theoretical single isotope mass value of 1485.5 were expressed at a strength of 40% or more in the sample group compared to the normal group samples. intensity marker.

In addition,

(Hex) ₅ (HexNAc) ₄ (Fuc) ₁ Theoretical single isotope mass value 1809.6,

(Hex) ₆ (HexNAc) ₅ (NeuAc) ₁ Theoretical single isotope mass values of 2295.8 and

(Hex) ₆ (HexNAc) ₅ (NeuAc) ₂ The N-glycans of the theoretical single isotope mass value of 2608.9 are expressed at a low intensity of less than 30% in the sample group compared to the normal group sample. (intensity marker).

According to the present invention, it has been found that there is a significant difference in the frequency of occurrence of a specific sugar chain between a normal group and a patient group through a sugar chain map, and this specific sugar chain can be used as a frequency marker of human ovarian cancer.

Further, according to the present invention, it has been found that there is a significant difference in the intensity of a specific sugar chain between a normal group and a patient group through a sugar chain map, and this specific sugar chain can be used as an intensity marker of human ovarian cancer.

In addition, according to the present invention, when there is a difference in expression frequency or expression intensity of a specific sugar chain depending on the progression stage of the tumor through the sugar chain map, the specific sugar chain can be used as a marker for identifying a specific tumor stage.

BRIEF DESCRIPTION OF THE DRAWINGS FIG. 1 is a schematic diagram of a method for preparing sugar chain maps from human serum samples.
Figure 2 shows MALDI-TOF results for sugar chains purified from human serum samples.
3 is a sugar chain map obtained from a human serum sample.
Figs. 4 to 13 are chain-link maps obtained by enlarging a part of Fig.
14 (A) is a representative MALDI-FTICR mass spectrum of N-sugar chains separated from ovarian cancer patients, and (B) is a sugar chain distribution and mass spectrum signal in the GCC fraction. 10, 20 and 40% represent the sugar chains from the corresponding fractions of acetonitrile in the positive and negative ion mode mass spectrometers. The structures are estimated based on exact mass and sugar biology.
Figure 15 (A) shows the peak alignment by calculating samples (96 persons) and control (48 persons) samples and placing them in a single m / z mass. (B) Box-Whiskers plots of abundant sugar chains found in the GCC 20% fraction.
Fig. 16 shows the relationship between the estimated structure and the structure of the ovarian cancer sugar chain marker.
Figure 17 shows the potential for disease stage determination according to the present invention. Five N-glycan structures are inserted. S, N, and L represent ovarian cancer patients, normal (control), and LMP ovarian tumor patients at different disease stages (S1 through S4).
Figure 18 is the ROC curve of tumor patients (n = 96) vs. healthy control (n = 48) glycosylated markers. (A) is frequency marker, and (B) is intensity marker.

DETAILED DESCRIPTION OF THE PREFERRED EMBODIMENTS Hereinafter, the present invention will be described in detail with reference to specific embodiments. However, it should be apparent to those skilled in the art that the scope of the present invention is not limited by the description of the embodiments.

Example 1: Human serum sample

1) The serum sample to be analyzed was stirred vigorously and sufficiently left.

2) Serum samples of 50 μl each were prepared.

3) To the prepared serum sample, 50 μl of 200 mM NH ₄ HCO ₃ (including 10 mM DTT) was added.

4) Stirred vigorously and allowed to stand still.

Example 2: N-sugar chain separation using an enzyme

PNGase F (peptide N-glycosidase F; 500,000 unit / ml) derived from Flavobacterium meningosepticum was purchased from New England BioLabs (Ipswich, Mass.). In order to separate glycans from the protein using the enzyme, 50 ㎕ of serum from Example 1 was dissolved in digestion buffer (pH 7.5, 100 mM ammonium bicarbonate, 5 mM DTT) and heated in boiling water for 2 minutes Denatured the protein. After cooling at room temperature, 2 μl of PNGase F (1,000 units) was added and the mixture was incubated for 16 hours in a water bath at 36 ° C.

400 [mu] l ethanol in a cold state was added to precipitate the peptide and the protein.

The resulting solution was frozen at -45 ° C for 60 minutes and centrifuged at 4 ° C for 14 minutes at 14,400 rpm for 20 minutes. Then, 400 μl of supernatant was taken for each sample and the ethanol contained in the supernatant was completely dried.

Thereafter, 1 ml of water was added to each sample, and the mixture was vigorously stirred to prepare a sugar chain-containing sample for purification.

Example 3: Sugar chain purification

The sugar chain-containing sample separated by PNGase F was purified by graphitized carbon cartridge SPE (GCC-SPE; packing amount 150 mg, cartridge volume 4 ml). GCC SPE cartridges were obtained from Alltech (Deerfield, IL). Prior to use, the cartridge was washed with 6 ml of ultrapure water, washed with 6 ml of 80% (v / v) acetonitrile (ACN) containing 0.1% trifluoroacetic acid (TFA) and then washed with 6 ml of ultrapure water. The sugar chain - containing samples were loaded into GCC cartridges and the deionized water was poured at a flow rate of 1 ml / min several times the volume of the cartridge. The sugar chain was eluted sequentially with 10% (v / v) acetonitrile, 20% (v / v) acetonitrile and 40% (v / v) acetonitrile and 0.05% (v / v) TFA. Each fraction was collected and dried with a centrifugal evaporator. Fractions were dissolved in ultrapure water prior to mass spectrometry.

Example 4: Mass spectrometry

The mass spectrum was recorded on a Fourier transform-ion cyclotron resonance mass spectrometer (FT-ICR MS) with an external source of HiResMALDI (IonSped Corporation) equipped with a magnet of 7.0 Tesla. HiResMALDI was equipped with a pulsed Nd: YAG laser (355 nm). DHB (2,5-Dihycrosy-benzoic acid) was used as a matrix of all sugar chains (0.05 mg / l in 50% ACN: H ₂ O). For negative mode analysis, 0.6 μl of oligosaccharide solution was applied to the MALDI probe followed by 1 μl of the appropriate matrix solution. The samples were vacuum dried and applied to MS analysis. For positive mode analysis, the same sample preparation procedure was applied except that 0.3 μl of 0.01 M NaCl was added to the matrix-analyte mixture.

Three fractions eluted during GCC-SPE, namely, 10% ACN fraction, 20% ACN fraction and 40% ACN fraction were analyzed by MALDI-FT-ICR respectively and the spectra were recorded. The sugar chains were eluted with solvent polarity, resulting in neutral sugar chains detected in 10% and 20% ACN fractions, and acidic sugar chains detected in 20% and 40% ACN fractions. The charge polarity of the mass spectrometer was changed between positive and negative modes to improve the ionic signal and minimize ionic inhibition of neutral or acidic sugar chains, respectively. Four spectra were collected for each sample. That is, 10% and 20% ACN fractions were obtained in the positive mode, and 20% and 40% ACN fractions were obtained in the negative mode.

Example 5: Spectral data preprocessing

A 6 S / N threshold was used to remove noise from the signal. The spectra were calculated internally using the six known glycocalin masses present in all spectra with Omega 8 software (IonSpec). Each spectrum was normalized to total ion abundance.

Example 6: ovarian cancer sugar chain marker test

For the ovarian cancer patients (21 patients, 2 patients: 14 patients, 3 patients: 52 patients, 4 patients: 13 patients), healthy female control patients (48 patients) and LMP patients (Low malignancy Patient: 50 patients) Was prepared as in Example 1 above.

As in Examples 2 to 4, enzymes were used to separate N-sugar chains of each sample, purify them, and conduct mass spectrometry. As in Example 5, spectral data was pre-processed.

The results of Example 6 are as follows.

Result 1: Glycoprotein profile of ovarian cancer patients serum

An experimental step including the separation of sugar chains is shown in Fig. This method was developed to minimize sample manipulation and increase speed. We have chosen N-sugar chains because they are much more abundant than O-sugar chains in human serum. N-sugar chains can also be separated more easily and selectively by enzymes. The present inventors have profiled N-sugar chains without the need for protein identification, isolation and purification. The separated sugar chains were first concentrated and then fractionated into solvents with different polarity (10, 20 and 40% aqueous acetonitrile solution containing 0.05% TFA). The sugar chain mixtures eluted from the three fractions were mass analyzed using MALDI-FTICR MS. In this way, the sugar chain signal was prevented from being suppressed by other sugar chains. The exact sugar chain structure was deduced from the exact mass. The abundant sugar chains were identified by tandem mass spectrometry using collision-induced dissociation (CID) or infrared multiphoton dissociation (IRMPD).

Serum of 100 non-mucinous epithelial ovarian cancer patients (1-21 patients, 2 -14 patients, 3 -52 patients, 4 patients -13 patients), 48 healthy female control serum and 50 low malignant potential GOG samples containing serum from ovarian cancer patients were used for glycoconjugate marker studies. Table 2 summarizes tumor stage, age, and serum sample structure. All samples were confirmed by pathology following surgery.

Representative mass spectra of serum-derived N-sugar chains of ovarian cancer patients of 10, 20 and 40% SPE fraction in positive and negative ion detection mode are shown in Figure 14A. The N - sugar chains were well separated from the trannanose core (Man ₃ GlcNAc ₂ ) and the exact mass was obtained by FTICR MS (Fourier transform ion cyclotron resonance mass spectroscopy). This figure shows only the predicted structure of abundant N-sugar chains. As we zoomed in on the less rich signals, we could reveal more sugar chains. In the 10% ACN spectrum, most of the sugar chains were Gorman nos and Hybrid. Acid oligosaccharides containing sialic acid (NeuAc) in the higher mass region are found in the 40% ACN fraction of negative mode. The 20% ACN fraction contains neutral and acidic oligosaccharides in the middle mass region. Thus, positive and negative scans were performed on this fraction. Neutral sugar chains are inhibited in acid, and vice versa. Therefore, not only the sugar chain size increases but also the sugar chain acidity is increased from 10% to 40% fraction. Four mass spectra were measured for each sample to maximize the number of sugar chains. The sugar chain separation in the SPE-GCG fraction is summarized in Figure 14B. With this approach, more than 80 sugar chain structures per sample were identified.

Result 2: Mass spectrum quality control

All mass spectra were calibrated internally (<5 ppm) with the N-sugar chains present in the spectrum. The peak selection of the experimental calibrant ion is very important for biomarker discovery, since the wrong choice can lead to a miscalculation of the entire spectrum. To ensure mass calculations and to confirm sample preparation, quality control of mass spectra was performed using a group of peaks of N-sugar chains known to exist in both patient and control groups. For example, biantennary oligosaccharides such as biantennary N-glycans (NB1) and biantennary N-glycans (NB2) can be used for both control and cancer patients (Morelle, W. et al., Glycobiology 2006 (16), 281-293). Without such a peak it indicates that calibration and / or sample preparation has not been performed well. Table 6 shows the sugar chains used for quality control and internal calibration. Individual scans for patients (96 cases) and controls (48cases) were corrected. For example, a single sugar chain (3Hex: 4HexNAc: 1Fuc) peak at m / z 1485.529 was overlaid to show peak alignment and correction accuracy (Fig. 15A). Initially, we started with 198 GOG samples (100 patient samples, 50 low malignant potential ovarian cancer patient samples and 48 normal control samples), but 194 (96 patient samples, 50 low malignant potential ovarian cancer Patient samples and 48 normal control samples). Biomarker studies were performed on the samples only. The peaks from the four samples were excluded by quality control because the main peak was a peptide rather than a saccharide.

Result 3: Changes in sugar chain in ovarian cancer patients

All mass spectral signals observed in this example were standardized for a more direct comparison between patient and control by total ionic strength. The overall difference in mass spectra between patient and control can be explained by Box-Whiskers plots of sugar chains. As an example, the distribution of abundant sugar chains found in the GCC 20% fraction is shown in Figure 15B. With respect to the biomarker, it can be determined with two types of markers - a frequency marker and an intensity marker, respectively. In frequency markers, the presence of each sugar chain peak in the control and patient samples was first calculated as%. Seven glycans containing four neutral and three acid glycols in Table 3 were identified as frequency markers. These are mostly biantennary or triantennary hybrid / hybrid N-glycans consisting of [Hex] _3-6 [HexNAc] _3-5 [NeuAc] _0-1 . Interestingly, potential glycoprotein markers were also structurally related to each other as shown in FIG.

Peaks that were present in greater than 90% of both patient and control groups were tested for their potential as intensity markers. Possibilities as intensity markers were identified for ten N-glycans containing eight neutral (present in 10% and 20% fractions) and two acidic glycans (present in 40% fractions) (Table 4). (NT-1, m / z 2295.812) and a double sialylated N-glycine (NT-2, m / z 2608.890) of two sialylated sugar chains, single sialylated N-glycans Mass spectral signal intensity was reduced in patient samples through all disease stages (stages 1 to 4), respectively. The biantennary N - glycans (NBF1, m / z 1809.640) in neutral glycans increased peak intensities of the other glycans corresponding to the hybrid / hybrid type, while decreased in the patient group. The estimated structure of the potential sugar chain markers is shown in Figure 16 to illustrate the relationship between them. All sugar chain markers are closely related to each other regardless of whether they are frequency markers or intensity markers. Generally, an increase in neutral glycans, whether it is a strength marker or a frequency marker, is generally observed in the patient group, whereas a decrease in acid (sialylated) sugar chains is observed in the patient group.

In addition, the number of intensity markers increased as the clinical stage increased from the initial stage S1 to the terminal stage S4 (Fig. 17). These are typical conjugated N-sugar chains (NB) consisting of [Hex] _3-4 [HexNAc] _3-5 [Fuc] _0-1 . These results show that the changes of glycosylation in ovarian cancer patients continue through disease progression. Thus, the sugar chain as a step marker can be used to investigate the progression of the disease or to predict the prognosis.

Result 4: Change in sugar chain in ovarian cancer patients with low malignant potential

The glycans profiled in 50 serum samples from patients with low malignant potential ovarian cancer were compared with patients and controls, respectively. Interestingly, the frequency and intensity patterns of low-malignant potential tumor-derived glycans show intermediate stages in patients and controls. Some sugar chains can be used to distinguish between patient groups and LMP tumor groups, while other sugar chains can be used to distinguish between control and LMP tumor groups. Potential sugar chain markers of LMP ovarian cancer are four frequency markers and five intensity markers, as shown in Table 5. The incidence (% presence) of each sample group (patient group, control group and LMP group) was compared to test for changes in sugar chain.

Four hybrid / hybrid N-glycans consisting of [Hex] _3-5 [HexNAc] _3-5 [Fuc] _0-1 were identified as frequency markers. They were present in a large number of patients with ovarian cancer (82.8% <f (p) <94.3%). On the other hand, the control group was less than 50% (27.9% <f (N) <48.8%). Interestingly, the frequency of low-malignancy potential tumors (51.1% <f (L) <80.0%) was intermediate between the patient and control groups. Borderline tumors generally exhibit lower marker levels than penetrating tumors. Five glycans were determined as intensity markers of low malignant potential tumors. These are three tri-antenna complexes (NT: m / z 2028.7, NT1: m / z 2295.8, NT1F: m / z 2441.9), one tetra-antennary complex : m / z 2660.0), and one conjugated sugar chain ( m / z 1542.6). Three of these glycans ( m / z 1542.6, 2028.7 and 2295.8) can distinguish the control group from the patient group and the LMP group, whereas NT1F and NTe1 can distinguish the LMP group from the patient group and the control group. In clinical practice, borderline tumors are often misdiagnosed as ovarian cancer because of the high levels of CA125. In addition, many LMP women have high pelvic masses. Thus, combined CA125, ultrasound, Doppler, and biopsy are used to diagnose benign tumors, suggesting the need for additional markers to complement the use of CA125 for the diagnosis of this fatal tumor. Changes in sugar chains in LMP ovarian cancer clearly demonstrate the nature of LMP tumors and also suggest that certain sugar chains can be identified in clinical settings corresponding to disease initiation.

Result 5: Evaluation of sugar chain markers

A strategic tool for evaluating the performance of biomarkers in disease detection is the receiver operating characteristic (ROC) curve. In general, ROC curves with area under the curve (AUC) values of 0.50-0.75 are considered to have fair discrimination. We calculated the AUC in the ROC curve for the frequency and intensity of the sugar chain marker, respectively. For frequency markers, each frequency marker could be combined, and one ROC curve was created by the same cutoff value. ROC curves are optimized by weighting each frequency marker (Table 3). The ROC curve for frequency sugar chain markers is shown in Figure 18A. The AUC was 0.93 (95% Cl: sensitivity 0.92, specificity 0.91) on the ROC curve for all samples. Unlike frequency markers, each intensity marker must be considered independently since the cutoff value is different. The ROC curve for each intensity glycan marker is shown in Figure 18B. The AUC for the other intensity markers was added in Table 4. M / z 1542.556 corresponding to (HexNAc) _3- (Man) ₃ (GlcNAc) ₂ in intensity markers showed the highest AUC value of 0.94.

The AUC produced by the sugar chain in the present invention was not directly compared with the AUC determined by the CA125 assay. Since the cut-off value of CA125 35 U / ml is already established for the sample used in the present invention, the training set for the CA125 value is a fully biased sample (Table 7). Therefore, the CA125 value can not be used to generate the ROC curve compared to the sugar chain analysis because of the high AUC value. GOG samples showing unusually high AUC values for CA125 have also been reported in other studies.

Ovarian cancer is a collection of diverse diseases with more than 30 malignant subtypes each with distinctive geopathology and clinical behavior. Confirmation is also possible only through pathologic examination of surgically excised tissue. The diversity and low incidence of ovarian cancer interfere with biomarker research. The availability of CA125 as a screen marker is particularly limited by positive and malignant conditions due to high false positive rates. Therefore, a new test method with improved diagnostic performance is required compared to CA125. A sugar chain study involves obtaining a sugar chain from a biological fluid without considering the related protein. Glycosylation involves glycosyltransferases that compete with each other to add to the initial sugar chain. Glycosylation first occurs in two parts of the cell, the endoplasmic reticulum and Golgi. Thus, if the glycosylation device malfunctions, almost all of the glycoproteins derived from the cell may potentially fail to glycosylation. Thus, gathering sugar chains provides a way to test changes in glycosylation in an individual. Furthermore, unlike proteins whose content increases or decreases with disease stages, the sugar chain change is more pronounced. Certain glycoproteins can be produced at various rates depending on the cell type. Nonetheless, it is known that glycosylation varies greatly in tumor cells compared to normal cells. Thus, monitoring changes in glycosylation can be a more specific and sensitive method of detecting tumors than indirect methods using monoclonal antibodies that recognize a limited number of protein antigens.

In the present invention, a mass spectrometry-based approach has been invented that determines the N-sugar chain of an ovarian cancer biomarker in human serum. Seventeen frequency markers and seventeen glycans, including ten intensity markers, were identified as potential biomarkers, most of which were hybrid / hybrid N-glycans. Some intensity markers increase with clinical progression, indicating the stage of the disease. In addition, although LMP tumors differ in terms of how LMP tumors are categorized, they exhibit intermediate characteristics between patient and control groups in glycocalyx profiling. A glycosmic approach to disease marker discovery is not dependent on the relevant protein separation or analysis. Instead, N-sugar chains are separated from human serum and used as a sample. Although these markers require confirmation and confirmation, these results illustrate the potential for use of glycoconjugate biomarkers to detect ovarian cancer with high sensitivity and specificity.

Claims (5)

A frequency marker composition for diagnosing ovarian cancer comprising N-sugar chains,
(Hex) ₃ (HexNAc) ₃ Theoretical single isotope mass value 1136.4,
(Hex) ₃ (HexNAc) ₄ Theoretical Single Isotope Mass Value 1339.5,
(Hex) ₃ (HexNAc) ₅ (Fuc) ₁ Theoretical single isotope mass value 1688.6,
(Hex) ₄ (HexNAc) ₅ Theoretical single isotope mass value 1704.6,
(Hex) ₄ (HexNAc) ₃ (NeuAc) ₁ Theoretical single isotope mass value 1611.5,
(Hex) ₆ (HexNAc) ₃ (NeuAc) ₁ Theoretical single isotope mass values of 1935.6 and
(Hex) ₄ (HexNAc) ₅ (NeuAc) ₁ Includes at least one N-sugar chain among the theoretical single isotope mass values of 2017.7. If the mass is expressed more than 20% A frequency marker composition for the diagnosis of ovarian cancer, wherein HexNAc, N-acetylhexosamine; Hex, Hexose; Fuc, Fucose; NeuAc, N-acetylneuraminic acid}.
delete delete 1. An intensity marker composition for diagnosing ovarian cancer comprising N-sugar chains,
(Hex) ₃ (HexNAc) ₄ Theoretical Single Isotope Mass Value 1339.5,
(Hex) ₃ (HexNAc) ₄ (Fuc) ₁ Theoretical single isotope mass value 1485.5,
(Hex) ₃ (HexNAc) ₅ Theoretical Single Isotope Mass Value 1542.6,
(Hex) ₄ (HexNAc) ₅ Theoretical single isotope mass value 1704.6,
(Hex) ₄ (HexNAc) ₃ Theoretical single isotope mass values of 1298.45 and
(Hex) ₃ (HexNAc) ₄ (Fuc) _{1 It} contains at least one N-sugar chain among the theoretical single isotope mass value of 1485.5. When mass spectrometric analysis shows more than 40% An intensity marker composition for diagnosing ovarian cancer, wherein HexNAc, N-acetylhexosamine; Hex, Hexose; Fuc, Fucose; NeuAc, N-acetylneuraminic acid}.
1. An intensity marker composition for diagnosing ovarian cancer comprising N-sugar chains,
(Hex) ₅ (HexNAc) ₄ (Fuc) ₁ Theoretical single isotope mass value 1809.6,
(Hex) ₆ (HexNAc) ₅ (NeuAc) ₁ Theoretical single isotope mass values of 2295.8 and
(Hex) ₆ (HexNAc) ₅ (NeuAc) _{2 It} contains at least one N-sugar chain of 2608.9 of theoretical single isotope mass value. When mass spectrometry is expressed at a low intensity of 30% (HexNAc, N-acetylhexosamine, or N-acetylhexosamine) for diagnosing ovarian cancer. Hex, Hexose; Fuc, Fucose; NeuAc, N-acetylneuraminic acid}.

Images