Blood metabolite markers for estimating glomerular filtration rate (GFR) and uses thereof
By identifying metabolic biomarkers through non-targeted metabolomics and constructing equations to estimate GFR, the problem of large estimation errors in existing technologies is solved, and accurate GFR estimation independent of creatinine is achieved.
Patent Information
- Authority / Receiving Office
- CN · China
- Patent Type
- Patents(China)
- Current Assignee / Owner
- 彭洪泉
- Filing Date
- 2021-11-05
- Publication Date
- 2026-05-15
AI Technical Summary
Existing methods for estimating glomerular filtration rate (GFR) are greatly affected by factors other than creatinine, resulting in large estimation errors. They also require demographic and clinical characteristics as alternative indicators and lack more accurate biomarkers.
Using a non-targeted metabolomics approach, we identified metabolic markers such as hydroxyasparagine, S-adenosine homocysteine, gluconate, and N6-succinyladenosine, and constructed equations to estimate GFR, reducing systematic bias and demographic effects.
It provides a reliable, fast, and convenient method for estimating GFR, reduces estimation errors, and does not rely on creatinine and other demographic information, thus improving the accuracy and precision of the estimation.
Smart Images
Figure 211105121709 
Figure 97738DEST_PATH_IMAGE003 
Figure 352001DEST_PATH_IMAGE004
Abstract
Description
Technical Field
[0001] This invention relates to the field of disease diagnostic reagent technology, and in particular to a blood metabolic marker for estimating glomerular filtration rate (GFR) and its application. Background Technology
[0002] Chronic kidney disease (CKD) is a global public health problem. Estimating glomerular filtration rate (GFR) is crucial for the identification and diagnosis of CKD. Currently, the best estimation method requires measuring serum creatinine levels, which can be influenced not only by abnormal kidney function but also by many other factors (e.g., drugs that inhibit renal tubular secretion, muscle mass, age, sex, ethnicity, etc.). Therefore, there is a need for other biomarkers better than creatinine to accurately estimate GFR. In this study, non-targeted metabolomics was applied to identify all metabolite biomarkers in the human body, discovering GFR-related metabolic markers to replace creatinine in estimating GFR. This approach reduces estimation errors caused by variability in non-GFR determinants and reduces the need to use demographic (e.g., ethnicity) and clinical characteristics (e.g., height, weight) as surrogate indicators for non-GFR.
[0003] Biomarkers discovered using high-throughput omics methods are highly susceptible to systematic bias. This bias can arise from specific confounding factors or from differing sample processing procedures within a hospital. Ideally, blood biomarkers should be detectable in both serum and plasma samples without significant bias. We know that small amounts of metabolites differ between serum and plasma samples. Some biomarkers, whose metabolites are found in serum samples, may not be good markers in plasma samples. Therefore, ideally, blood biomarkers should be detectable in both serum and plasma samples without significant bias. Summary of the Invention
[0004] For the reasons mentioned above, in this study, we applied non-targeted metabolomics to identify all metabolic biomarkers in the human body and discover metabolic biomarkers related to GFR to replace creatinine in estimating GFR. This approach reduces estimation errors caused by variations in non-GFR determinants and reduces the need to use demographic (e.g., race) and clinical characteristics (e.g., height, weight) as surrogate indicators for non-GFR.
[0005] This invention proposes a blood metabolic marker for estimating glomerular filtration rate (GFR) and its application. Specifically, to achieve the objectives of this invention, the following technical solution is proposed:
[0006] This invention relates to a blood metabolic marker for estimating glomerular filtration rate (GFR), comprising hydroxyasparagine but excluding creatinine. It should be noted that hydroxyasparagine is a newly discovered blood metabolic marker for estimating glomerular filtration rate (GFR).
[0007] In a preferred embodiment of the present invention, the blood metabolic markers further include S-adenosyl homocysteine, gluconate, and N6-succinyl adenosine. The equations constructed using these four metabolic markers provide sufficient information for estimating GFR.
[0008] In another preferred embodiment of the invention, the blood metabolic markers further include S-adenosine homocysteine, gulonate, and N6-succinyladenosine. The equation constructed using these four metabolic markers provides sufficient information for estimating GFR. Furthermore, by using gulonate instead of gluconate, the potential influence of gluconate contained in food additives and supplements can be eliminated.
[0009] In another aspect of the invention, the invention also relates to the use of the above-mentioned blood metabolic markers in the preparation of diagnostic reagents for estimating glomerular filtration rate (GFR).
[0010] In a preferred embodiment of the invention, the glomerular filtration rate (GFR) is estimated using the CKD-msMET4a equation.
[0011] In another preferred embodiment of the invention, the glomerular filtration rate (GFR) is estimated using the CKD-msMET4b equation.
[0012] Beneficial effects
[0013] This invention, through the use of a rigorous two-center study design, successfully identified new GFR-related metabolites and further determined two new metabolite marker combinations for constructing equations that allow for reliable estimation of GFR without the need for creatinine, cystatin C, and demographic and clinical characteristics (such as race, sex, weight, etc.). These results can be used to develop rapid, convenient, and accurate diagnostic kits. Attached Figure Description
[0014] Figure 1: The experimental process and results of a rigorous two-center study design;
[0015] Part 1: Biomarker discovery and construction of GFR equations;
[0016] Part 2: Independent validation of the identified biomarkers and GFR equations.
[0017] The study design involved (i) two types of blood samples from two centers, namely serum and plasma, and (ii) plasma samples from healthy volunteers to minimize false findings of biomarkers associated with systematic and / or systemic bias. Serum samples were obtained from 52 CKD patients at the Third Affiliated Hospital of Sun Yat-sen University in Guangzhou; plasma samples were obtained from 135 CKD patients and 10 healthy volunteers at Kiang Wu Hospital in Macau, totaling 197 cases. 79 CKD cases (19 from Guangzhou and 70 from Macau, representing 40% of all cases) were randomly selected to form an independent validation group. The remainder were assigned to biomarker discovery and GFR equation construction. The statistical methods used at different steps and the corresponding summaries of results are also shown in the figure. Logarithmic transformations were performed on mGFR, normalized abundance of metabolite biomarkers, serum / plasma creatinine levels, and serum / plasma CysC levels prior to partial correlation analysis and equation construction. Detailed Implementation
[0018] To further understand the present invention, the technical solutions in the embodiments of the present invention will be clearly and completely described below in conjunction with the embodiments of the present invention. Obviously, the described embodiments are only some embodiments of the present invention, and not all embodiments. Based on the embodiments of the present invention, all other embodiments obtained by those of ordinary skill in the art without creative effort are within the scope of protection of the present invention.
[0019] Unless otherwise specified, all reagents involved in the embodiments of this invention are commercially available products and can be purchased through commercial channels.
[0020] Example 1:
[0021] To minimize the impact of systematic bias, a rigorous two-center study design was employed to discover and independently validate blood metabolic biomarkers for estimating GFR. Serum samples were collected from 52 CKD patients at the Third Affiliated Hospital of Sun Yat-sen University in Guangzhou; plasma samples were collected from 135 CKD patients and 10 healthy volunteers (Group 1C) at Kiang Wu Hospital, totaling 197 participants. 79 CKD cases (19 from Guangzhou and 60 from Macau, representing 40% of all participants) were randomly selected to form the independent validation group. The remaining 118 participants were assigned to biomarker discovery and equation construction to assess GFR. The demographic and clinical characteristics of the study participants are summarized in Table 1; there were no substantial differences in age, sex, etc., between the groups.
[0022] Table 1: Summary of Clinical Characteristics of Groups 1A-1C
[0023]
[0024] DiscoveryHD4 using Metabolon TM The platform performed untargeted metabolomics analysis on plasma and serum samples based on UPLC-MS / MS. The normalized abundance of 669 metabolites was significantly correlated with the GFR (mGFR, the gold standard for glomerular filtration rate measured by iohexol clearance) tested in Discovery Group 1A (serum samples from 33 CKD patients in Guangzhou) or Discovery Group 1B (plasma samples from 75 CKD patients in Macau), respectively (Spearman rank correlation, Benjamini-Hochberg (BH) adjusted P < 0.05). Among these, 389 metabolites (58%) were significantly correlated with sex, height, weight, and / or age in both groups (BH adjusted P < 0.05). After controlling for the confounding effects of age, sex, height, and weight, the log-transformed normalized abundance of 296 metabolites (279 negative; 17 positive) was significantly correlated with the log-transformed mGFR in both discovery groups (partial correlation, BH corrected P < 0.05).
[0025] To avoid identifying false biomarkers, the normalized abundance of 296 metabolites was further compared between 12 CKD patients with severe kidney disease (from discovery group 1B) and 10 healthy volunteers (discovery group 1C). Only those significantly different (Mann-Whitney test, BH-adjusted p-value < 0.05) and >2-fold changes in the consistent direction were retained as definitive GFR-related biomarkers. A total of 215 GFR metabolite markers were ultimately identified.
[0026] Independent validation was performed using 79 CKD cases. Figure 1 Of these, 212 cases (97.8%) were successfully verified. Figure 1 Of the 207 biomarkers, 207 were negatively correlated with mGFR and 5 were positively correlated with mGFR. This validation dataset has not yet been used for biomarker discovery or equation construction. Among the top 20 biomarkers, hydroxyasparagine and N,N-dimethylproline-proline were found to be closely related to renal function for the first time (Table 2). Furthermore, creatinine is one of the commonly used biomarkers among the top 20, indicating the success of the study design.
[0027] Table 2: Metabolic markers and their relationship with renal function
[0028]
[0029] For the proof-of-concept, to demonstrate its feasibility, as described in the Chronic Kidney Disease-Epidemiology Collaboration (CKD-EPI), a multiple linear regression equation was constructed to correlate the log-transformed mGFR with the log-transformed normalized abundance of the top 20 identified metabolic biomarkers in the discovery dataset. S-adenosyl homocysteine (P=0.010), gluconate (P=0.00006), N6-succinyl adenosine (P=3.5x10⁻¹⁰) -7 S-adenosyl homocysteine (P=0.011), gulonate (P=0.001), N6-succinyl adenosine (P=3.8x10⁻⁷), and hydroxyasparagine (P=0.002) were retained in the final equation and named "CKD-msMET4a" (Table 2). The main difference from the first equation is the substitution of gluconate for gulonate. Gulouses and gluconate are stereoisomers, indicating that the two equations are nearly identical in properties. Adding age, height, weight, and even log-transformed cystatin C (CysC, another commonly used renal function marker) to the equations did not significantly improve either equation (P > 0.05). This indicates that the four metabolites provide sufficient information for estimating GFR.
[0030] CKD-msMET4a equation: eGFR = exp(3.999 - 0.137ln(normalized abundance of S-adenosylhomocysteine) - 0.162ln(normalized abundance of gluconic acid) - 0.209ln(normalized abundance of N6-succinyladenosine) - 0.319ln(normalized abundance of S-adenosylhomocysteine) - 0.319ln(normalized abundance of hydroxyasparagine).
[0031] CKD-msMET4b equation: eGFR = exp(3.991 - 0.141ln(normalized abundance of S-adenosylhomocysteine) - 0.146ln(normalized abundance of gulonic acid) - 0.214ln(normalized abundance of N6-succinyladenosine) - 0.270ln(normalized abundance of hydroxyasparagine).
[0032] In terms of bias, precision, and accuracy (as described in CKD-EPI), an independent validation dataset (i.e., 79 CKD cases) was used. Figure 1This dataset was not involved in equation construction. The performance of the two constructed equations was evaluated and compared with five commonly used equations in clinical practice (Table 3).
[0033] Table 3: Independent verification of the two equations in this invention and comparison with commonly used clinical equations.
[0034]
[0035]
[0036] a The bias, precision, and accuracy of each equation in estimating GFR were evaluated using the methods described by Levey et al. (2009) and Inker et al. (2012) [8,9].
[0037] b The best values for the two equations are shown in italics and bold.
[0038] c CKD-msMET4a equation: eGFR = exp(3.999 - 0.137ln(normalized abundance of S-adenosylhomocysteine) - 0.162ln(normalized abundance of gluconic acid) - 0.209ln(normalized abundance of N6-succinyladenosine) -0.319ln(normalized abundance of S-adenosylhomocysteine) - 0.319ln(normalized abundance of hydroxyasparagine).
[0039] d The CKD-msMET4b equation is: eGFR = exp(3.991 - 0.141ln(normalized abundance of S-adenosylhomocysteine) - 0.146ln(normalized abundance of gulonic acid) - 0.214ln(normalized abundance of N6-succinyladenosine) -0.270ln(normalized abundance of hydroxyasparagine).
[0040] e MDRD is a four-variable 175 MDRD equation that has been revised using creatinine standardization.
[0041] f Accuracy is calculated as the root mean square error relative to m GFR, the percentage of estimates that differ from m GFR by less than 30% (P30), and the percentage that differ by less than 20% (P20).
[0042] Experimental results show that, on the one hand, CKD-msMET4a and CKD-msMET4b outperform three equations based on serum creatinine, age, and sex (CKD-EPI creatinine, MDRD, and Cockcroft-Gault). On the other hand, the performance of CKD-msMET4a and CKD-msMET4b is similar to that of CKD-EPI creatinine-cystatin C, which is currently considered the best equation used globally. It is worth noting that CKD-EPI creatinine-cystatin C was developed based on demographic and hematological data from 5352 participants, while this study used LC-MS data from 118 cases to construct the equation. Considering the small sample size of blood metabolites detected by LC-MS and the potential for larger errors, GFR estimation using s-adenosylhomocysteine, gluconate, gulonate, N6-succinyladenosine, and hydroxyasparagine can be further optimized.
[0043] The preferred embodiments of the present invention have been described above, but are not intended to limit the invention. Those skilled in the art can make modifications and variations to the embodiments disclosed herein without departing from the scope and spirit of the invention.
Claims
1. A blood metabolic marker for estimating glomerular filtration rate, said blood metabolic marker comprising hydroxyasparagine, S-adenosine homocysteine, gluconate and N6-succinyladenosine.
2. A blood metabolic marker for estimating glomerular filtration rate, said blood metabolic marker comprising hydroxyasparagine, S-adenosine homocysteine, gulonate and N6-succinyladenosine.
3. The use of the reagent for detecting the blood metabolic markers of claim 1 in the preparation of diagnostic reagents for estimating glomerular filtration rate.
4. The use of the reagent for detecting the blood metabolic markers of claim 2 in the preparation of diagnostic reagents for estimating glomerular filtration rate.
5. In the application according to claim 3, the glomerular filtration rate is estimated using the CKD-msMET4a equation, which is: eGFR = exp(3.999 - 0.137ln(normalized abundance of S-adenosylhomocysteine) - 0.162ln(normalized abundance of gluconic acid) - 0.209ln(normalized abundance of N6-succinyladenosine) - 0.319ln(normalized abundance of S-adenosylhomocysteine) - 0.319ln(normalized abundance of hydroxyasparagine).
6. In the application according to claim 4, the glomerular filtration rate is estimated using the CKD-msMET4b equation, which is: eGFR = exp(3.991 - 0.141ln(normalized abundance of S-adenosylhomocysteine) - 0.146ln(normalized abundance of gulonic acid) - 0.214ln(normalized abundance of N6-succinyladenosine) - 0.270ln(normalized abundance of hydroxyasparagine).