CN108459157B - A composition for predicting irinotecan chemotherapy toxicity biomarkers - Google Patents

Abstract

The present invention relates to a biomarker panel for predicting the severity of delayed diarrhea and myelosuppression after administration. Compared with non-sensitive individuals with mild delayed diarrhea after administration, sensitive individuals with severe delayed diarrhea had significantly higher serum levels of cholic acid, deoxycholic acid and glycocholic acid before administration, while Phenylalanine levels were significantly lower. Compared with non-sensitive individuals with less severe myelosuppression after administration, sensitive individuals with severe myelosuppression had significantly higher serum glycocholic acid levels before administration, while phenylalanine, lysine and Tryptophan content was significantly lower.

Description

A composition for predicting irinotecan chemotherapy toxicity biomarkers

technical field

The invention relates to the field of predictive biomarkers, in particular to a composition of a group of biomarkers for predicting the toxicity of irinotecan chemotherapy, the composition is derived from endogenous small molecule metabolites in serum, and the composition is useful for rapidly predicting an individual's response to Irinotecan. The sensitivity of rinotecan and clinical individualized treatment are of great significance.

Background technique

Irinotecan (CPT-11), a DNA topoisomerase I inhibitor, is mainly used in the first-line treatment of metastatic colorectal cancer. Studies have shown that irinotecan has serious chemotherapy toxicity in clinical use, mainly including delayed diarrhea and bone marrow suppression, which has become the main factor limiting its clinical application.

There are obvious individual differences in the chemotherapy toxicity of irinotecan. Therefore, predicting the sensitivity of patients to irinotecan before administration and adjusting the drug regimen is the basis for reducing chemotherapy toxicity and realizing individualized treatment. At present, the commonly used clinical method is to predict the chemotherapy sensitivity of patients by detecting UGT1A1 gene polymorphisms. However, this method is expensive, not widely available, and due to the complex reasons for individual differences, single genotype detection is often prone to misjudgment.

Therefore, it is particularly important to develop a multifactorial and convenient method for predicting the toxicity of irinotecan chemotherapy. The metabolome is located at the end of biological information flow, and differences at various levels such as genes, proteases, and environmental factors are reflected at the level of endogenous small molecule metabolites. Serum analysis is a commonly used research method in metabolomics research and clinical practice, and is widely used because of its simplicity and economy. Serum metabolite levels have not been used to predict individual differences in irinotecan chemotherapy toxicity.

Therefore, the application of serum metabolomics to find biomarkers to predict the toxicity of irinotecan chemotherapy is of great significance for the clinical application and individualized treatment of irinotecan.

SUMMARY OF THE INVENTION

The present invention relates to a composition for predicting irinotecan chemotherapy toxicity biomarkers:

Serum samples before administration of irinotecan chemotherapy-sensitive group and non-sensitive group were detected;

Detection of pre-dose serum samples from mammals in irinotecan chemotherapy-sensitive and non-sensitive groups.

Pre-dose serum samples of subjects in the irinotecan chemotherapy-sensitive and non-sensitive groups were tested.

The invention relates to a composition for predicting irinotecan chemotherapy toxicity biomarkers: the application of glycocholic acid and phenylalanine in preparing a composition for predicting irinotecan chemotherapy toxicity biomarkers.

The invention relates to a composition for predicting the toxicity biomarkers of irinotecan chemotherapy: the application of glycocholic acid, phenylalanine, cholic acid and deoxycholic acid in the preparation of markers for predicting delayed diarrhea of irinotecan.

Compared with the non-sensitive group with delayed diarrhea, the sensitive group had up-regulated glycocholic acid, cholic acid and deoxycholic acid, and down-regulated phenylalanine before administration.

The present invention relates to a composition for predicting sensitivity to irinotecan chemotherapy toxicity:

Glycocholic acid: 2.04～2.86;

Phenylalanine: 55.23～69.07;

Cholic acid: 24.44～34.12;

Deoxycholic acid: 1.72～2.52;

Sum, or, unit: any one of μmol and μmol/L.

The present invention relates to a non-sensitive composition for predicting the toxicity of irinotecan chemotherapy:

Glycocholic acid: 1.47～2.25;

Phenylalanine: 62.66～75.78;

Cholic acid: 16.56～22.68;

Deoxycholic acid: 1.43 to 2.09; and, or,

Sum, or, unit: any one of μmol and μmol/L.

The invention relates to a composition for predicting the toxicity biomarkers of irinotecan chemotherapy: the application of glycocholic acid, phenylalanine, lysine and tryptophan in the preparation of markers for predicting the myelosuppression of irinotecan.

The present invention relates to a composition for predicting irinotecan chemotherapy toxicity biomarkers:

Serum samples before administration of irinotecan chemotherapy-sensitive group and non-sensitive group were detected;

Detection of pre-dose serum samples from mammals in irinotecan chemotherapy-sensitive and non-sensitive groups.

Pre-dose serum samples of subjects in the irinotecan chemotherapy-sensitive and non-sensitive groups were tested.

Compared with the myelosuppressive non-sensitive group, the sensitive group had up-regulated glycocholic acid and down-regulated phenylalanine, lysine and tryptophan before administration.

The present invention relates to a sensitive composition for predicting the toxicity of irinotecan chemotherapy:

Glycocholic acid: 2.04～2.86;

Phenylalanine: 55.23～69.07;

Lysine: 264.31～312.01;

Tryptophan: 53.76～65.62;

Sum, or, unit: any one of μmol and μmol/L.

The present invention relates to a composition of insensitive biomarkers for predicting the toxicity of irinotecan chemotherapy:

Glycocholic acid: 1.47～2.25;

Phenylalanine: 62.66～72.78;

Lysine: 303.39～349.95;

Tryptophan: 60.43～71.93;

Sum, or, unit: any one of μmol and μmol/L.

The technical scheme adopted in the present invention is as follows: applying non-targeted metabolomics analysis based on liquid chromatography-mass spectrometry and gas chromatography-mass spectrometry combined with multivariate data analysis and processing methods to carry out serum metabolism of individuals with different irinotecan chemotherapy toxicity Omics research, looking for a combination of endogenous small molecule biomarkers in serum that can predict the toxicity of irinotecan chemotherapy, and using quantitative methods based on liquid chromatography mass spectrometry to accurately quantify small molecule biomarkers and establish Logit equations prediction model.

The biological marker determination process for predicting the toxicity of irinotecan chemotherapy of the present invention is as follows:

Materials: methanol, acetonitrile and formic acid (chromatographic grade) were purchased from Merck, Germany; methoxyamine chloride and N-methyl-N-(trimethylsilane) trifluoroacetamide (containing 1% trimethyl methacrylate) Chlorosilane) was purchased from Sigma-Aldrich Company in the United States;

Deionized water was prepared by the Milli-Q ultrapure water system of Millipore Company in the United States;

Standard compounds include: cholic acid, deoxycholic acid, glycocholic acid, phenylalanine, lysine, tryptophan, all purchased from Sigma-Aldrich, USA;

The internal standard compounds cortisone acetate and fodosteine were purchased from China National Institute for Food and Drug Control.

Serum sample collection: 40 healthy male SD rats (SPF grade, 200±10g, purchased from Shanghai Sipple-Bikai Laboratory Animal Co., Ltd.) were injected with irinotecan (60mg/kg, purchased from Jiangsu Hengrui Co., Ltd.) into the tail vein ), and the blood collection time was in the early morning on an empty stomach.

Preparation and analysis of samples for untargeted metabolomics research:

GC/MS

Sample processing method: Precisely draw 10 μL of serum sample, place it in a 1.5 mL high-speed centrifuge tube, add internal standard heptanoic acid (100 μg/mL methanol solution 100 μL), vortex and mix for 3 min, and then centrifuge at high speed (16000 rpm, 4 ℃, After 10 min), 80 μL of the supernatant was accurately drawn into a brown reaction tube, 25 μL of MOX reagent (MOX pyridine solution, 10 mg/mL) was added, and the reaction was shaken at 37 ° C and 1200 rpm for 1.5 h, and the reaction solution was vacuumed at 50 ° C Dry for 2 hours; then add 120 μL of MSTFA reagent (MSTFA:ethyl acetate=1:1) to the reaction tube, vortex and mix, shake and react at 37°C and 1200rpm for 2 hours, and the solution after the reaction is immediately injected for analysis.

GC/MS conditions: Shimadzu GCMS-QP2010 gas chromatography-mass spectrometer. The chromatographic column is Rtx-5MS quartz capillary column (30m×0.25mm×0.25μm); gradient heating method: 0-2min is 70℃, 2-27min is 70-320℃, 27-29min is 320℃; the carrier gas flow rate is 57cm/s (99.99% helium); inlet temperature is 250℃; injection volume is 1μL (split ratio 50:1); ionization energy is 70eV; ion source temperature is 200℃; interface temperature is 250℃; solvent The delay time was 5 min; the mass scan range was m/z: 45-600 Daltons.

UPLC-IT-TOF/MS

Sample processing method: Precisely aspirate 20 μL of serum, place it in a 1.5 mL centrifuge tube, add 140 μL of acetonitrile solution (5 μg/mL) containing internal standard glyburide, vortex for 5 min, and then centrifuge at high speed (4°C, 16000 rpm, 10 min). , transfer the supernatant to the injection vial for UPLC-IT-TOF/MS analysis.

UPLC-IT-TOF/MS conditions: Japan Shimadzu liquid chromatography-mass spectrometer. The chromatographic column is Phenomenex Kinetex C18column (100×2.1mm, 2.6μm, Phenomenex, USA); the mobile phase contains two solvents: A is 0.1% formic acid, B is acetonitrile; the flow rate is 0.4mL/min; the column temperature is 40°C; The temperature of the injector is 16°C; the volume of injection is 5 μL; the chromatographic gradient elution conditions are: 0-20min for 5%-95%B, 20-23min for 95%-5%B, 23-30min for 5%B ; Electrospray ion source (ESI) uses positive and negative ion mode to simultaneously collect data, the atomizing gas rate is 1.5L/min, the drying gas pressure is 0.1MPa, the detection voltage is 1.70KV, the ion source interface voltage is positive 4.5kV, negative - 3.5kV; mass scan range m/z: 100-1000 Daltons.

Data processing and analysis

The data obtained by GC/MS and UPLC-IT-TOF/MS were imported into SIMCA-p software (version 11.0, Umetrics) for multivariate statistical analysis. Differential metabolites were screened by establishing an OPLS-DA (Orthogonal Partial Least Squares-Discriminant Analysis) model.

The results showed that the chemotherapy toxicity sensitive group (23 animals, accounting for 57.5% of the total) had more severe delayed diarrhea and bone marrow suppression compared with the non-sensitive group (17 animals, accounting for 42.5% of the total) after administration. There was a clear distinction in the OPLS-DA graph between individuals in the chemotoxicity-sensitive group and the non-sensitive group before administration. As shown in Figure 1.

The material structure was searched through databases such as HMDB (http://www.hmdb.ca/) and Metline (http://metlin.scripps.edu/), using the exact molecular weight provided in the database and the MS/MS obtained above Differential metabolites were identified by the profiles and confirmed by comparison with standards.

To further differentiate biomarkers that characterize irinotecan for two common chemotherapy toxicities: delayed diarrhea and myelosuppression, we used Lasso regression analysis ((http://www.r-project.org) to screen for Biomarkers associated with delayed diarrhea and myelosuppression.

The results showed that glycocholic acid, cholic acid, deoxycholic acid and phenylalanine could be used to predict delayed diarrhea, and glycocholic acid, phenylalanine, lysine and tryptophan could be used to predict myelosuppression;

Six compounds identified and characterized, namely glycocholic acid, cholic acid, deoxycholic acid, phenylalanine, tryptophan, and lysine, their chromatographic retention times were consistent with those of the standard, and their chromatographic retention times were multilevel. The mass spectral fragmentation information was consistent with the structural features of the standard.

Process for accurate quantification of differential metabolites:

Bile acids:

Sample processing method: Precisely aspirate 50 μL of serum, add 10 μL internal standard (cortisone acetate, 100 μg/mL) and 200 μL methanol, vortex for 5 min, centrifuge at high speed (4°C, 14000 rpm, 10 min), transfer the supernatant to injection Vials for LC/MS analysis.

LC/MS conditions: Thermo Fisher triple quadrupole mass spectrometer, the chromatographic column is ZORBAXEclipse XDB-C18 (2.1×150mm, 3.5μm, Agilent, USA), the mobile phase contains two solvents: A is 0.1 % formic acid, B is acetonitrile; the flow rate is 0.45mL/min; the column temperature is 45°C; the injector temperature is 15°C; A, 21-28min is 73%A, 28-31min is 73%-60%A, 31-56min is 60%-68%A; Electrospray ion source (ESI) uses negative ion mode to collect data, source spray voltage is 3.8 kV, and the ion transport capillary temperature was 360 °C. The sample spectrum is shown in Figure 2.

Amino acids:

Sample processing method: Precisely aspirate 40 μL of serum, add 10 μL of internal standard (fodosteine, 80 μg/mL) and 350 μL of acetonitrile, vortex for 5 min, centrifuge at high speed (4°C, 14000 rpm, 10 min), transfer the supernatant to injection Vials for LC/MS analysis.

LC/MS conditions: American Thermo Fisher triple quadrupole mass spectrometer, the chromatographic column is BEH HILIC (2.1×100mm, 1.7μm, Waters, Ireland), and the mobile phase contains two solvents: A is 0.01% formic acid , B is acetonitrile; the flow rate is 0.30 mL/min; the column temperature is 30 °C; the injector temperature is 15 °C; 3-6min is 15%-60%A, 6-10min is 15%A; Electrospray ion source (ESI) uses positive ion mode to collect data, source spray voltage is 4.5kV, ion transmission capillary temperature is 270℃. The sample spectrum is shown in Figure 2.

Based on the accurate quantitative results of the landmark compounds and the differences in delayed diarrhea and myelosuppression among individuals after administration, a Logit equation was established for the prediction of chemotherapy toxicity.

the result shows,

The prediction equation for delayed diarrhea is Logit(P)=26.190+1.403CA+3.652DCA+5.717GCA-1.196Phe;

The prediction equation for myelosuppression is Logit(P)=26.381-0.205Phe-0.056Lys-0.059Trp+4.002GCA;

When the Logit value of the prediction equation is greater than 0, it is a toxicity-sensitive individual, and when it is less than 0, it is a non-sensitive individual.

The invention relates to a composition for predicting irinotecan chemotherapy toxicity biomarkers: the application of glycocholic acid and phenylalanine in preparing a composition for predicting irinotecan chemotherapy toxicity biomarkers.

The present invention relates to using cholic acid, deoxycholic acid, glycocholic acid and phenylalanine to prepare a composition for predicting irinotecan delayed diarrhea biomarkers; the present invention relates to combining lysine, tryptophan, Phenylalanine and glycocholic acid are used to prepare a composition for predicting biomarkers of irinotecan myelosuppression. The results are shown in Table 1.

Table 1. Summary of biomarkers predicting irinotecan delayed diarrhea and myelosuppression

Validation: The receiver operating curve (ROC) method was further used to test the prediction of irinotecan delayed diarrhea by the logit equation fitted by the four biomarkers of glycocholic acid, phenylalanine, cholic acid and deoxycholic acid. Accuracy.

The results show that in the training set, the area under the curve is 0.987, and the prediction accuracy of the equation is 95%. A new batch of serum samples from animal models (25 animals) was selected for verification. The results showed that after grouping by the prediction equation, individuals in the sensitive group (15 animals, accounting for 60% of the total) and the non-sensitive group (10 animals, accounting for 40% of the total) were There were significant differences in the degree of delayed diarrhea, and the sensitive group developed more severe delayed diarrhea. The results are shown in Figure 3.

The receiver operating curve (ROC) method was further used to test the prediction accuracy of irinotecan myelosuppression by the logit equation fitted by the four biomarkers of phenylalanine, lysine and tryptophan.

The results show that in the training set, the area under the curve is 0.967, and the prediction accuracy of the equation is 87.5%. A new batch of serum samples from animal models (25 animals) were selected for verification. The results showed that after grouping by the prediction equation, the sensitive group (14 animals, accounting for 56% of the total) and the non-sensitive group (11 animals, accounting for 44% of the total) were There were obvious differences in the degree of myelosuppression, and the sensitive group developed more severe myelosuppression. The results are shown in Figure 3.

To sum up, the present invention provides a new biomarker group for clinical prediction of irinotecan delayed diarrhea and myelosuppression through the accurate quantification of serum serum metabolomics and endogenous small molecular substances. Quick, easy, economical and non-invasive advantages.

The invention has important significance for predicting the toxicity of irinotecan chemotherapy, and provides an important basis for irinotecan clinical medication, dose adjustment and individualized treatment.

Description of drawings

Figure 1: Individual difference diagram of chemotherapy toxicity sensitive group vs chemotherapy toxicity non-sensitive group in Example.

Figure 2: Chromatograms for the determination of the contents of glycocholic acid, cholic acid, deoxycholic acid, phenylalanine, lysine, and tryptophan in the embodiment.

Figure 3: Validation graph of the prediction accuracy of the predictive biomarker panel in the example for irinotecan chemotherapy toxicity in training set and validation set.

Detailed explanation of Figure 1-3

A in Figure 1 is: the diarrhea score curve of chemotherapy toxicity sensitive group and non-sensitive group;

B of Figure 1 is: the cell count diagram of chemotherapy toxicity sensitive group and non-sensitive group;

C in Figure 1 is: OPLS-DA diagram of chemotherapy toxicity sensitive group and non-sensitive group under GC/MS before administration;

D in Figure 1 is: OPLD-DA diagram of chemotherapy toxicity-sensitive group and non-sensitive group in UPLC-IT-TOF/MS positive ion mode before administration;

E in Figure 1 is: OPLD-DA diagram of chemotherapy toxicity-sensitive group and non-sensitive group in UPLC-IT-TOF/MS negative ion mode before administration;

A of Fig. 2 is: the chromatogram of glycocholic acid, cholic acid, deoxycholic acid content determination;

B of Fig. 2 is: the chromatogram of phenylalanine, lysine, tryptophan content determination;

A of Figure 3 is: the receiver operating curve of the delayed diarrhea sensitive group and the non-sensitive group in the training set;

B of Figure 3 is: the diarrhea score curve of the delayed diarrhea sensitive group and the non-sensitive group in the validation set;

C of Figure 3 is: the receiver operating curve of the myelosuppression sensitive group and the non-sensitive group in the training set;

D of FIG. 3 is: the cell count diagram of the myelosuppression sensitive group and the non-sensitive group in the validation set.

specific embodiment

Example 1

The combination of biomarkers used to predict delayed diarrhea of irinotecan, its composition and accurate content are shown in Table 2:

Table 2 Predicted differential metabolites of late-onset diarrhea sensitive vs insensitive

Example 2

The combination of biomarkers used to predict myelosuppression with irinotecan, its composition and accurate content are shown in Table 3: Table 3 predicts the differential metabolites of myelosuppression sensitive vs insensitive

Example 3:

A composition for predicting sensitivity to irinotecan chemotherapy toxicity, the proportions of the components are:

Glycocholic acid: 2.04～2.86;

Phenylalanine: 55.23～69.07;

Cholic acid: 24.44～34.12;

Deoxycholic acid: 1.72～2.52;

Sum, or, unit: any one of μmol and μmol/L.

The purpose of the composition is to take the composition as a reference substance, and detect the contents of the above four substances in the sample, so as to predict the sensitivity of the organism to the toxicity of irinotecan chemotherapy.

Example 4:

A non-sensitive composition for predicting the toxicity of irinotecan chemotherapy, the proportions of the components are:

Glycocholic acid: 1.47～2.25;

Phenylalanine: 62.66～75.78;

Cholic acid: 16.56～22.68;

Deoxycholic acid: 1.43～2.09;

Sum, or, unit: any one of μmol and μmol/L.

The purpose of the composition is to take the composition as a reference substance, and detect the contents of the above four substances in the sample, so as to predict the insensitivity of the organism to the toxicity of irinotecan chemotherapy.

Example 5:

A sensitive composition for predicting the toxicity of irinotecan chemotherapy, the proportions of the components are:

Glycocholic acid: 2.04～2.86;

Phenylalanine: 55.23～69.07;

Lysine: 264.31～312.01;

Tryptophan: 53.76～65.62;

Sum, or, unit: any one of μmol and μmol/L.

The purpose of the composition is to take the composition as a reference substance, and detect the contents of the above four substances in the sample, so as to predict the sensitivity of the organism to the toxicity of irinotecan chemotherapy.

Example 6:

A non-sensitive composition for predicting the toxicity of irinotecan chemotherapy, the proportions of the components are:

Glycocholic acid: 1.47～2.25;

Phenylalanine: 62.66～72.78;

Lysine: 303.39～349.95;

Tryptophan: 60.43～71.93;

Sum, or, unit: any one of μmol and μmol/L.

The purpose of the composition is to take the composition as a reference substance, and detect the contents of the above four substances in the sample, so as to predict the insensitivity of the organism to the toxicity of irinotecan chemotherapy.

Claims (7)

1. a composition that predicts the sensitivity to irinotecan chemotherapy toxicity is characterized in that: the ratio of each component is Glycocholic acid: 2.04～2.86; Phenylalanine: 55.23～69.07; Cholic acid: 24.44～34.12; Deoxycholic acid: 1.72～2.52; Unit: either μmol or μmol/L. 2. a non-sensitive composition for predicting the toxicity of irinotecan chemotherapy is characterized in that: the ratio of each component is Glycocholic acid: 1.47～2.25; Phenylalanine: 62.66～75.78; Cholic acid: 16.56～22.68; Deoxycholic acid: non-sensitive group 1.43-2.09; Unit: either μmol or μmol/L. 3. A composition sensitive to the toxicity of irinotecan chemotherapy is predicted, characterized in that: the ratio of each component is Glycocholic acid: 2.04～2.86; Phenylalanine: 55.23～69.07; Lysine: 264.31～312.01; Tryptophan: sensitive group 53.76-65.62; Unit: either μmol or μmol/L. 4. A composition for predicting sensitivity to irinotecan chemotherapy toxicity, characterized in that: the ratio of each component is Glycocholic acid: 1.47～2.25; Phenylalanine: 62.66～72.78; Lysine: 303.39～349.95; Tryptophan: 60.43～71.93; Unit: either μmol or μmol/L. 5. The preparation and application of a composition according to claims 1-4, characterized in that: the preparation of biomarkers for predicting the sensitivity of delayed diarrhea and myelosuppression after administration of irinotecan. 6. The preparation and application of a composition according to claims 1-4, characterized in that: application of a biomarker for predicting delayed diarrhea after administration of irinotecan. 7. The preparation and use of a composition according to claims 1-4, characterized in that: the use of biomarkers for predicting myelosuppression after administration of irinotecan.

Images