A test formulation and kit for diagnosing the risk of atrial fibrillation

Abstract

This invention belongs to the field of biomedical technology, specifically relating to a diagnostic agent for the risk of new-onset atrial fibrillation in the elderly and its application. This invention discovers that plasma QUIN expression levels can be used as a quantitative screening standard for the risk of atrial fibrillation in the elderly population. When the QUIN expression level in a subject's biological sample is ≥88.02 ng / mL, the subject has a risk of developing atrial fibrillation within the next 10 years; the higher the QUIN level, the higher the risk. When the QUIN content in a subject's biological sample is ≥215.5 ng / mL and ≥280.4 ng / mL, respectively, the predicted cumulative risk of atrial fibrillation within 3 years is over 90% and over 95%, respectively. This technique of clearly stratifying subjects and quantifying risk assists in individualized intervention and clinical management, providing a reference for precision treatment.

Description

Technical Field

[0001] This invention belongs to the field of biomedical technology, specifically relating to a diagnostic agent, reagent kit, and application for diagnosing the risk of new-onset atrial fibrillation in the elderly. Background Technology

[0002] Atrial fibrillation (AF) is one of the most common sustained arrhythmias, which can cause serious diseases such as stroke and heart failure. Aging is one of its strongest risk factors. During an AF episode, due to the loss of effective atrial contraction function, some blood is retained in the atria. The slow blood flow easily induces thrombus formation in the atria, which can lead to complications such as heart failure and systemic embolism, increasing the risk of stroke fivefold. It is characterized by high disability and high mortality rates (Gómez-Outes A, Lagunar-Ruíz J, Terleira-Fernández AI, et al. Causes of Death in Anticoagulated Patients With Atrial Fibrillation[J]. J Am Coll Cardiol, 2016,68(23):2508-2521.). Furthermore, the hospitalization rate for atrial fibrillation (AF) patients is extremely high, reaching 43.7 times per 100 person-years (Meyre P, Blum S, Berger S, et al. Risk of Hospital Admissions in Patients With Atrial Fibrillation: A Systematic Review and Meta-analysis[J]. Can J Cardiol, 2019,35(10):1332-1343.), and the lifetime risk of major adverse cardiovascular events exceeds 30% (Pistoia F, Sacco S, Tiseo C, et al. The Epidemiology of Atrial Fibrillation and Stroke[J]. Cardiol Clin, 2016, 34(2):255-268.), severely impacting patients' quality of life and placing a significant burden on public health. Therefore, atrial fibrillation risk assessment can promptly screen for risks and provide patients with early warning information.

[0003] Current atrial fibrillation risk assessments largely rely on traditional clinical indicators such as age, hypertension, diabetes, obesity, heart failure, chronic inflammation, and other comorbidities. However, their ability to identify new-onset atrial fibrillation in older adults is limited. Existing methods for understanding and assessing atrial fibrillation risk primarily include: 1. Traditional cardiovascular risk factor assessment Age, sex, hypertension, diabetes, obesity, and smoking are widely considered risk factors for atrial fibrillation and can be used for population-level risk description and statistical analysis, but they are insufficient to provide reliable predictions of new-onset atrial fibrillation for individuals.

[0004] Scoring systems based on age, sex, and clinical risk factors (such as the Framingham score) can be used to describe or stratify the risk of cardiovascular events in a population, but they cannot achieve quantitative prediction of the risk of atrial fibrillation based on molecular markers (70-year legacy of the Framingham Heart Study. Nat Rev Cardiol. 2019 May 7. doi:10.1038 / s41569-019-0202-5.; Review of: "Framingham Heart Study CardiovascularDisease 10-Year Risk Score Clinical Classification"). Especially in the elderly population, due to the presence of high-competition mortality events, traditional scores cannot accurately reflect the individual's actual risk of atrial fibrillation.

[0005] 2. Myocardial enzyme indicators Indicators such as myocardial enzymes and troponin are used to assess myocardial damage or necrosis, and are mainly used for the diagnosis of cardiovascular events such as acute myocardial infarction. However, their predictive ability for individuals who have not yet developed atrial fibrillation is limited, and they cannot be used as specific predictive indicators for atrial fibrillation.

[0006] 3. Inflammatory markers Multiple cohort studies have assessed the association between inflammatory markers and the risk of atrial fibrillation. However, existing non-specific inflammatory markers, such as hsCRP and myocardial enzyme levels, are easily affected by various chronic diseases and inflammatory states, resulting in limitations in specificity and sensitivity as independent predictive indicators, making it difficult to accurately quantify individual risk. For example, in the Ansan-Ansung cohort in South Korea, a single measurement of high-sensitivity C-reactive protein (hsCRP) was not significantly associated with the risk of atrial fibrillation, while persistently elevated hsCRP levels could independently predict the occurrence of atrial fibrillation. This suggests that inflammatory markers have certain value and limitations in predicting atrial fibrillation (Lee, Yonggu, Hwan-Cheol Park, Jeong hun Shin, Young-Hyo Lim, Jinho Shin and Jinkyu Park. “Single and persistent elevation of C-reactive protein levels and the risk of atrial fibrillation in a general population: The Ansan-Ansung Cohort of the Korean Genome and Epidemiology Study.” International journal of cardiology 277 (2019): 240-246.).

[0007] 4. Electrocardiogram and Monitoring Methods A 12-lead electrocardiogram and Holter monitoring can definitively diagnose atrial fibrillation that has already occurred, but cannot provide early prediction for individuals who have not yet developed atrial fibrillation.

[0008] In summary, current technologies primarily rely on traditional risk factors, non-specific inflammatory markers, and routine electrocardiogram monitoring. Among these, blood biomarkers and traditional risk factor scores mainly depend on statistical association or population risk stratification. Risk stratification is often based solely on quantiles or statistical associations, lacking operability and individualized predictive capabilities, and therefore cannot provide a specific, quantifiable, and operable method for predicting the risk of new-onset atrial fibrillation. Furthermore, existing studies have not established a clear risk threshold corresponding to a specific cumulative incidence rate (e.g., new-onset atrial fibrillation within 3 years). This fails to meet the need for early, specific, and quantifiable prediction of new-onset atrial fibrillation in the elderly population. Summary of the Invention

[0009] This invention discovers that plasma QUIN expression levels can serve as a quantitative screening standard for the risk of atrial fibrillation in the elderly population, and can be directly applied in clinical practice. Based on this, this invention was completed.

[0010] Firstly, this invention provides the use of a reagent for detecting the expression level of QUIN in biological samples in the preparation of a kit for diagnosing the risk of atrial fibrillation in subjects. When the QUIN expression level in a subject's biological sample is detected to be ≥88.02 ng / mL, the subject has a risk of developing atrial fibrillation within the next 10 years. The higher the QUIN level, the higher the risk of developing atrial fibrillation within the next 10 years. When the QUIN expression level in a subject's biological sample is detected to be ≥215.5 ng / mL, the probability of the subject developing atrial fibrillation within the next 3 years is 90%. Thus, it can be seen that QUIN can quantitatively predict the early occurrence of atrial fibrillation in subjects.

[0011] Furthermore, the biological sample is a blood sample.

[0012] Furthermore, the blood sample is at least one of peripheral blood, plasma, and / or serum.

[0013] Preferably, the blood sample is plasma.

[0014] Furthermore, the subjects were mammals.

[0015] Furthermore, the subjects were elderly people.

[0016] Preferably, the elderly person is selected from those aged 65-80 years.

[0017] Furthermore, the atrial fibrillation includes paroxysmal atrial fibrillation, persistent atrial fibrillation, long-term persistent atrial fibrillation, permanent atrial fibrillation, and new-onset atrial fibrillation.

[0018] Preferably, the atrial fibrillation is new-onset atrial fibrillation.

[0019] Secondly, this invention provides a kit for predicting the risk of atrial fibrillation. The kit contains reagents for detecting the expression level of QUIN in a subject's biological sample. When the QUIN expression level in a subject's biological sample is ≥88.02 ng / mL, the subject has a risk of developing atrial fibrillation within the next 10 years. The higher the QUIN level, the higher the risk of developing atrial fibrillation within the next 10 years. When the QUIN expression level in a subject's biological sample is ≥215.5 ng / mL, the subject has a 90% probability of developing atrial fibrillation within the next 3 years. This kit can provide early quantitative prediction of the risk of developing atrial fibrillation in a subject.

[0020] Furthermore, the biological sample is a blood sample.

[0021] Furthermore, the blood sample is at least one of peripheral blood, plasma, and / or serum.

[0022] Preferably, the blood sample is plasma.

[0023] Furthermore, the subjects were mammals.

[0024] Furthermore, the subjects were elderly people.

[0025] Preferably, the elderly person is selected from those aged 65-80 years.

[0026] Furthermore, the atrial fibrillation includes paroxysmal atrial fibrillation, persistent atrial fibrillation, long-term persistent atrial fibrillation, permanent atrial fibrillation, and new-onset atrial fibrillation.

[0027] Preferably, the atrial fibrillation is new-onset atrial fibrillation.

[0028] Thirdly, this invention provides a method for predicting the cumulative risk of atrial fibrillation in a subject over a certain period of time. The model includes the expression level of QUIN in the subject's biological sample, the time limit for the risk of atrial fibrillation, and the proportion of cumulative risk of atrial fibrillation. The three factors of QUIN expression level, time limit, and proportion of cumulative risk satisfy the following formula:

[0029] in, For the Fine-Gray subdistribution risk model The regression coefficients; The baseline cumulative subdistribution risk function at 3 years; ; The target is the cumulative incidence of new-onset atrial fibrillation; t is the follow-up time. When the time frame is limited to 3 years and the cumulative risk of atrial fibrillation is 90%, the threshold for QUIN in biological samples is 215.5 ng / mL; that is, when the QUIN content in the subject's biological sample is ≥215.5 ng / mL, the predicted cumulative risk of atrial fibrillation in the subject within 3 years is greater than 90%.

[0030] In one implementation, when the time period is limited to 3 years and the cumulative risk of atrial fibrillation is 95%, the content of QUIN in the biological sample is 280.4 ng / mL. That is, when the content of QUIN in the subject's biological sample is ≥280.4 ng / mL, the predicted cumulative risk of atrial fibrillation in the subject within 3 years is greater than 95%.

[0031] Furthermore, the time period is 0.1-10 years.

[0032] Preferably, the time period is 3 years.

[0033] Furthermore, the biological sample is a blood sample.

[0034] Furthermore, the blood sample is at least one of peripheral blood, plasma, and / or serum.

[0035] Preferably, the blood sample is plasma.

[0036] Furthermore, the subjects were mammals.

[0037] Furthermore, the subjects were elderly people.

[0038] Preferably, the elderly person is selected from those aged 65-80 years.

[0039] Furthermore, the atrial fibrillation includes paroxysmal atrial fibrillation, persistent atrial fibrillation, long-term persistent atrial fibrillation, permanent atrial fibrillation, and new-onset atrial fibrillation.

[0040] Preferably, the atrial fibrillation is new-onset atrial fibrillation.

[0041] Beneficial effects 1. This invention measures the plasma QUIN expression level threshold of 88.02 ng / mL in elderly individuals aged 65-80 years. Specifically, when the QUIN expression level in a subject's biological sample is detected to be ≥88.02 ng / mL, the subject is at risk of developing atrial fibrillation within the next 10 years. The higher the QUIN level, the greater the risk of atrial fibrillation within the next 10 years. The accuracy of predicting the risk of new-onset atrial fibrillation in the elderly population using plasma QUIN expression level is high (AUC = 0.745, 95% CI 0.717-0.773); specificity is 0.81, and sensitivity is 0.59. Compared to existing methods using non-specific inflammatory markers (such as hsCRP) or traditional cardiovascular risk factors, this invention significantly improves the specificity and accuracy of predicting the risk of new-onset atrial fibrillation in the elderly.

[0042] 2. This invention provides a method for predicting the cumulative risk of atrial fibrillation in a subject over a certain period of time. The model quantifies the risk of atrial fibrillation in the elderly at an early stage. When the level of QUIN in the subject's biological sample is ≥215.5 ng / mL, the predicted cumulative risk of atrial fibrillation within 3 years is over 90%; when the level of QUIN in the subject's biological sample is ≥280.4 ng / mL, the predicted cumulative risk of atrial fibrillation within 3 years is over 95%. This technique of clearly stratifying subjects and quantifying risks assists in individualized intervention and clinical management, providing a reference for precision treatment.

[0043] 3. Existing studies have shown that factors with a trend associated with atrial fibrillation (AF) such as kynurenine acid and kynurenine have a higher trend than QUIN. However, this invention found that in the model, compared to kynurenine and kynurenine acid, QUIN significantly increased susceptibility to AF, and the duration of AF was significantly prolonged in the QUIN group, while this was not significant in the kynurenine and kynurenine acid groups. Among the trend-related factors, only plasma QUIN expression level can serve as a biomarker for predicting the risk of new-onset AF in the elderly population and can be used for clinical diagnosis. Attached Figure Description

[0044] Figure 1 The effects of QUIN, KYNA, and L-KYN on the induction rate and duration of atrial fibrillation in isolated rat hearts.

[0045] Note: A represents the process of collecting electrocardiograms; B represents the electrocardiogram; C represents the results of atrial fibrillation induction rate and duration.

[0046] Figure 2 Add hiPSC-AMs arrhythmia-like events and ectopic pacing activity to QUIN.

[0047] Note: A represents the arrhythmia-like event induction rate; B represents the ectopic pacing activity induction rate.

[0048] Figure 3 This study investigated changes in plasma concentrations and atrial fibrillation susceptibility in rats after QUIN exposure.

[0049] Note: A represents intravenous injection of QUIN; B represents changes in plasma QUIN over time; C represents the risk of atrial fibrillation.

[0050] Figure 4 QUIN was used to increase susceptibility to atrial fibrillation in Langendorff isolated perfused rat hearts.

[0051] Figure 5 To enhance atrial fibrillation susceptibility in QUIN in vivo, isolated pig heart, and isolated human heart models.

[0052] Note: A is the in vivo model; B is the isolated perfused pig heart model; C is the isolated human heart perfused model.

[0053] Figure 6 The study found elevated levels of QUIN in the circulation and atrial regions of aging rats, accompanied by increased susceptibility to atrial fibrillation.

[0054] Note: A represents the level of QUIN in the atrial tissue of young and old rats; B represents the level of circulating QUIN in young and old rats; C represents the susceptibility to atrial fibrillation.

[0055] Figure 7The accumulation of quinine in aging atria mainly originates from CD68-positive macrophages.

[0056] Note: AE indicates that the key enzyme HAAO for QUIN synthesis is significantly upregulated in senescent atrial macrophages; F indicates flow cytometry results. G represents the QUIN expression level in macrophages, fibroblasts, and cardiomyocytes; H represents the QUIN expression level after macrophage depletion.

[0057] Figure 8 Macrophage depletion and macrophage HAAO-targeted silencing reduce susceptibility to atrial fibrillation in aged rats.

[0058] Note: A is the electrocardiogram; B is the atrial fibrillation induction rate after macrophage depletion; C is the decrease in HAAO protein expression after silencing macrophages' QUIN synthesis of the key enzyme HAAO; D is the electrocardiogram after silencing macrophages' QUIN synthesis of the key enzyme HAAO; E is the atrial fibrillation induction rate after silencing macrophages' QUIN synthesis of the key enzyme HAAO.

[0059] Figure 9 Minocycline and caroverine reduce susceptibility to atrial fibrillation after QUIN exposure and promote termination of established atrial fibrillation.

[0060] Note: A represents the intervention model; B represents the treatment model.

[0061] Figure 10 Caroverine was used to terminate established atrial fibrillation in pig and isolated human heart models.

[0062] Note: A is a pig model; B is an isolated human heart model.

[0063] Figure 11 To investigate the association between baseline plasma QUIN and new-onset atrial fibrillation in the elderly population.

[0064] Note: a represents the cohort study procedure; b represents the hazard ratio (HR) for new-onset atrial fibrillation grouped by baseline plasma quartile concentration; c represents the cumulative incidence of new-onset atrial fibrillation grouped by high and low baseline plasma qUIN levels; the cumulative incidence of new-onset atrial fibrillation grouped by high and low baseline plasma qUIN levels, with mortality treated as a competing event, is shown in shaded areas representing 95% CI, and differences between groups were assessed using the Gray test. The number of at-risk individuals and the cumulative number of events are shown below the curve; d represents the subdistributed hazard ratio (SHR) for new-onset atrial fibrillation grouped by high and low baseline plasma qUIN levels; estimated using a Fine-Gray competing risk model, with all-cause mortality treated as a competing event. Model 1 is uncorrected; Model 2 is corrected for age and sex; Model 3 is a fully corrected model (correction variables are the same as...). Figure 11(b) High QUIN is defined as baseline plasma QUIN ≥ 88.02 ng / mL; Low QUIN is defined as < 88.02 ng / mL.

[0065] Figure 12 Association of plasma quinolinic acid with new-onset atrial fibrillation (ad) and all-cause mortality (e and f) in an elderly cohort.

[0066] Note: a is the ROC curve for baseline plasma QUIN prediction of atrial fibrillation during follow-up; b is the Kaplan-Meier estimate of atrial fibrillation-free survival in the high-QUIN and low-QUIN groups (log-rank P < 0.001); the inset plot shows the upper segment of the y-axis. c is the quartile range for each QUIN; d is the estimated cumulative probability of atrial fibrillation by QUIN group (1 - Kaplan-Meier), with the number of at-risk individuals and cumulative events shown below the curve, and the shaded band representing the 95% CI; e is the estimated cumulative incidence of all-cause mortality in the high-quinolinate and low-quinolinate groups, using the Fine-Gray method, with new-onset atrial fibrillation as a competing event; the number of at-risk individuals and events is shown below the curve, and the shaded band represents the 95% CI; f is the subdistributed hazard ratio (SHR) of all-cause mortality and its 95% CI (high-quinolinate group vs. low-quinolinate group), from the Fine-Gray method with new-onset atrial fibrillation as a competing event. Competing risk models: Model 1 is unadjusted; Model 2 is adjusted for age and sex; Model 3 is fully adjusted for age, sex, body mass index (BMI), systolic blood pressure (SBP), diabetes, history of myocardial infarction, history of stroke, smoking, alcohol consumption, high-sensitivity C-reactive protein (hs-CRP), serum creatinine (Scr), low-density lipoprotein cholesterol (LDL-C), and thyroid-stimulating hormone (TSH); vertical dashed lines indicate SHR = 1 (no association). In bd, death prior to atrial fibrillation is considered a censored event; because this method does not consider competing risk of death, these figures are for exploratory visualization only. Conversely, e and f are Fine-Gray competing risk analyses for all-cause mortality, with new-onset atrial fibrillation as a competing event.

[0067] Figure 13 Forest plot for subgroup analysis based on the Cox proportional hazards model.

[0068] Note: Within each predefined subgroup, the HR compared the risk of new-onset atrial fibrillation during a median follow-up of 8.58 years between participants with high and low baseline plasma QUIN levels (divided into two groups based on the ROC-derived cutoff of 88.02 ng / mL, with the low QUIN group as the reference). HR and 95% CI were estimated using a multivariate Cox proportional hazards model. Squares represent point estimates, horizontal lines represent 95% CIs, and vertical dashed lines represent the null line (HR = 1). The p-values ​​shown for each layer represent the association between high baseline QUIN levels and new-onset atrial fibrillation within that subgroup. Except for the previous myocardial infarction (MI) subgroup (P = 0.468, whose wider CI reflects the smaller number of participants in this subgroup [n = 49]), this association was statistically significant in all subgroups (P < 0.001 in all subgroups except the previous stroke subgroup, P = 0.003). Two-sided P < 0.05 was considered statistically significant.

[0069] Above: HR: Hazard ratio; QUIN: Quinolinic acid; AF: Atrial fibrillation; ROC: Receiver operating characteristic; AUC: Area under the curve; CI: Confidence interval; MI: Myocardial infarction; BMI: Body mass index; SBP: Systolic blood pressure; hs-CRP: High-sensitivity C-reactive protein; Scr: Serum creatinine; TC: Total cholesterol; LDL-ch: Low-density lipoprotein cholesterol; TSH: Thyroid-stimulating hormone; Y: Yes; N: No. Detailed Implementation

[0070] The specific embodiments of the present invention will be further described below. It should be noted that these descriptions are for the purpose of aiding understanding the present invention, but do not constitute a limitation thereof. Furthermore, the technical features involved in the embodiments described below can be combined with each other as long as they do not conflict with each other.

[0071] Unless otherwise specified, the experimental methods used in the following embodiments are conventional methods, and the experimental materials used in the following embodiments are all available through conventional commercial channels.

[0072] The results in the following examples are based on a cohort analysis of community populations aged ≥65 years with baseline plasma quinolinic acid (QUIN) measurements. Unless otherwise stated, hazard ratios (HR) and subdistribution hazard ratios (SHR) are reported with 95% confidence intervals (CI), and all tests are two-tailed.

[0073] Terminology Explanation QUIN (3-hydroxyanthralinic acid): a metabolite of the quinolinic acid pathway, and a circulating metabolite associated with aging.

[0074] AF (Atrial Fibrillation): A common type of arrhythmia in the elderly, and the primary outcome event in this study.

[0075] The Fine-Gray model, proposed by Fine and Gray in 1999, is a competing risk model based on the Cumulative Incidence Function (CIF). This model is specifically designed for survival analysis scenarios with multiple endpoint events. The Fine-Gray model analyzes the impact of covariates on the probability of event occurrence by estimating the subdistribution hazard function.

[0076] The Cox Proportional Hazards Model is a semi-parametric regression model proposed by British statistician David Cox in 1972. This model is used for survival analysis to study the relationship between the timing of an event (such as death, failure, etc.) and a set of covariates (such as age, gender, treatment method, etc.).

[0077] Baseline Survival Function (S0(t)): In the Cox proportional hazards model, the survival probability at QUIN=0 or the reference covariate level is used to infer the QUIN threshold for achieving the target cumulative risk.

[0078] Cumulative Incidence Function (CIF): The cumulative probability of events in the Fine-Gray competitive risk model after considering competing events such as death.

[0079] Youden-optimal cutoff: In time-dependent ROC analysis, the threshold that maximizes the sum of sensitivity and specificity is used to initially determine the QUIN high and low groups.

[0080] Langendorff perfusion is a classic technique for studying the electrophysiological activity of isolated hearts. Compared to in vivo heart studies, this technique eliminates interference from neural and humoral regulation, offering significant advantages such as fewer influencing factors and greater controllability—allowing for precise control of key experimental conditions such as preload, afterload, nutrient solution composition, and temperature. The principle of Langendorff isolated heart perfusion is as follows: Under isothermal and pressure-controlled conditions, a cannula is inserted into the aorta of the isolated animal heart, and oxygenated perfusion fluid is administered retrogradely. The myocardium is perfused through the coronary arteries, and the perfusion fluid flows out through the coronary sinus from the right atrium, right ventricle, and pulmonary artery, maintaining the isolated heart's normal physiological activity in the external environment.

[0081] Lead: In electrocardiogram (ECG) terminology, a lead is the position of the electrodes on the human body surface and the connection method between the electrodes and the amplifier when recording an ECG.

[0082] A 12-lead electrocardiogram (ECG) detects the electrical activity of cells (such as the electrical vectors generated by cells) through electrodes and converts it into waveforms using an ECG machine. In terms of lead location, it includes 6 limb leads and 6 chest leads. The limb leads include standard bipolar leads and compression leads.

[0083] Human-induced pluripotent stem cells (hiPSCs) are a type of pluripotent stem cell obtained by reprogramming somatic cells. They are "artificial embryonic stem cells" obtained by reprogramming adult cells (such as skin fibroblasts, peripheral blood B / T lymphocytes, and renal tubular epithelial cells) by introducing specific transcription factors.

[0084] Kynurenic acid: Kynurenic acid is an endogenous tryptophan metabolite and a broad-spectrum antagonist targeting NMDA, glutamate, and α7 nicotinic acetylcholine receptor. Kynurenic acid is also an agonist of GPR35 / CXCR8.

[0085] Caroverine is a non-selective NMDA and AMPA glutamate receptor antagonist and also a class B calcium channel blocker. Pharmacologically, it is described as a non-specific calcium channel blocker and an antagonist of neither NMDA nor NMDA glutamate receptors. Carvacine (Spasmium, Tinnitin, Tinnex) is a muscle relaxant used in Austria and Switzerland to relieve spasms of smooth muscle (including the intestines, arteries, and other organs). Its use has expanded to help treat cerebrovascular diseases and ultimately tinnitus. Carvacine is also a potential chemotherapeutic agent for HNSCC cell lines.

[0086] Boltz2 is a fully open-source large-scale model that can simultaneously perform structural conformation prediction and affinity scoring, applicable to various interaction scenarios such as proteins, small molecules, and nucleic acids. It does not belong to traditional QSAR or docking-based methods, but rather, through end-to-end training, it performs energy modeling while predicting structural conformation, possessing the key capability to achieve molecular-level "generation-screening-optimization".

[0087] GROMACS (GROningen MAchine for Chemical Simulations) is a general-purpose software used for molecular dynamics simulations of systems with millions of particles based on Newton's equations of motion. GROMACS is primarily used for the analysis of biochemical molecules, such as proteins, lipids, and nucleic acids with complex bonding interactions. Typical simulation applications of GROMACS include the efficient calculation of non-bonded interactions, making it widely used by researchers to study polymers in non-biological systems. GROMACS supports common molecular dynamics algorithms and can utilize GPUs to accelerate the core computational processes.

[0088] Example 1: QUIN increases the risk of atrial fibrillation and prolongs the time to onset of atrial fibrillation. Test methods This embodiment is used to verify whether QUIN has a direct atrial fibrillation-promoting effect and to compare its atrial fibrillation-promoting effect with that of kynurenic acid (KYNA) and L-kynurenine (L-KYN), which belong to the same tryptophan-kynurenic acid metabolic pathway.

[0089] (1) Langendorff isolated heart perfusion experiment in rats: Twenty-eight adult male Sprague-Dawley rats were randomly divided into four groups of seven each: Vehicle control group, QUIN group, KYNA group, and L-KYN group. After anesthesia, the heart was harvested to establish a Langendorff isolated perfusion model. The heart was perfused at a constant temperature of 37°C with oxygenated Tyrode perfusion fluid and kept at a stable equilibrium. Subsequently, the rats were treated with Vehicle, 1 μM QUIN, 1 μM KYNA, or 1 μM L-KYN perfusion, respectively. Atrial fibrillation was induced by atrial programmed stimulation after treatment, and atrial electrocardiograms were continuously recorded. Atrial fibrillation was defined as rapid, irregular atrial electrical activity or cardiac electrical activity with a duration reaching a preset standard. The atrial fibrillation induction rate and duration of atrial fibrillation in each group were statistically analyzed.

[0090] (2) Detection of electrical activity of hiPSC-derived atrial myocytes: Human induced pluripotent stem cell-derived atrial myocytes (hiPSC-AMs) were seeded on a multi-electrode array (MEA) plate. After they developed stable spontaneous electrical activity, they were treated with a vehicle or 1 μMQUIN. The field potential signal was continuously recorded using the MEA system to analyze the occurrence of arrhythmia-like events and ectopic pacing activity.

[0091] (3) In vivo QUIN exposure experiment in rats: Adult male Sprague-Dawley rats were administered different doses of QUIN or Vehicle intravenously. Plasma samples were collected at different time points after administration, and changes in plasma QUIN concentration were detected by liquid chromatography-tandem mass spectrometry (LC-MS / MS). Subsequently, the atrial fibrillation induction rate and duration were detected by programmed atrial stimulation in vivo to evaluate the effect of elevated circulating QUIN on susceptibility to atrial fibrillation.

[0092] (4) Validation experiments of isolated rat heart, in vivo and isolated pig heart, and isolated human heart: To eliminate interference from systemic neurohumoral factors, different concentrations of QUIN were perfused into isolated Langendorff rat hearts, and atrial fibrillation was induced by atrial pacing. QUIN or Vehicle was then administered to in vivo pig models, isolated perfused pig heart models, and isolated perfused human heart models. Atrial electrocardiograms were recorded according to the same principles, and the atrial fibrillation induction rate and duration were statistically analyzed. All electrocardiogram recordings and atrial fibrillation interpretations were performed by researchers unaware of the grouping information.

[0093] Test results In a survey of high-risk cardiovascular populations, it was found that individual indicators such as kynurenine and kynurenic acid showed a more significant trend in the occurrence of atrial fibrillation than QUIN in individuals with atrial fibrillation. However, this invention found that in a disease model, QUIN was more significantly associated with the occurrence of atrial fibrillation than kynurenine and kynurenic acid. Compared with the control group, treatment with kynurenine and kynurenic acid was not significantly associated with an increased risk of atrial fibrillation, nor was it significantly associated with a prolonged duration of atrial fibrillation.

[0094] Specific results are as follows Figure 1 As shown, compared with kynurenine and kynurenine quinolinic acid, QUIN significantly increased susceptibility to atrial fibrillation (AF). In the rat Langendorff ex vivo perfusion model, AF was not induced in the Vehicle group (0 / 7), the AF induction rate was significantly increased in the QUIN group (6 / 7), while the AF induction rates in the KYNA and L-KYN groups were both low (1 / 7). The duration of AF was significantly prolonged in the QUIN group, while KYNA and L-KYN treatments did not significantly prolong the duration of AF. These results indicate that among the tryptophan-kynurenine pathway-related metabolites tested, QUIN has the most definite pro-AF effect; although KYNA and L-KYN belong to the same related metabolic pathway, they did not show a pro-AF effect comparable to QUIN under the experimental conditions, therefore they cannot be used as clinical predictors of AF risk; while QUIN has higher specificity in AF risk prediction and mechanism intervention. Therefore, plasma QUIN expression level can serve as a biomarker for predicting the risk of new-onset AF in the elderly and has clinical predictive value.

[0095] like Figure 2 As shown, in hiPSC-AMs, 1 μM QUIN treatment significantly increased arrhythmia-like events and ectopic pacing activity. No significant arrhythmia-like events were observed in the Vehicle group, but abnormal spontaneous electrical activity and ectopic pacing signals appeared after QUIN treatment, indicating that QUIN can directly act on atrial myoblasts and induce abnormal electrical activity. These cellular-level results suggest that QUIN is not merely a disease marker, but a functional metabolite that directly affects the excitability of atrial myoblasts.

[0096] like Figure 3As shown, in a rat in vivo model, plasma QUIN levels rapidly increased after intravenous injection and then gradually decreased over time, suggesting that this administration route can create detectable and quantifiable QUIN exposure in vivo. QUIN increased the induction rate and duration of atrial fibrillation in a dose-dependent manner. For example, low-dose QUIN increased susceptibility to atrial fibrillation in some animals, while treatment with 0.1 mg / kg QUIN increased the induction rate to 7 / 8 and significantly prolonged the duration of atrial fibrillation. These results further demonstrate that elevated QUIN levels in vivo can enhance atrial electrical instability and increase the risk of atrial fibrillation.

[0097] like Figure 4 As shown, in Langendorff's isolated perfused rat heart, QUIN still increased the induction rate of atrial fibrillation and prolonged its duration. Since the isolated perfused model excludes the influence of systemic neural regulation, humoral factors, and circulatory system factors, this result indicates that QUIN has a direct myocardial proarrhythmic effect, rather than depending on peripheral neurohumoral responses or systemic metabolic changes.

[0098] like Figure 5 As shown in Figure A, in the porcine in vivo model, atrial fibrillation was not induced in the Vehicle group, while the induction rate of atrial fibrillation increased and the duration of atrial fibrillation was prolonged after QUIN treatment. Figure 5 As shown in Figure B, in the isolated perfused porcine heart model, QUIN was also observed to increase the induction rate of atrial fibrillation and prolong the duration of atrial fibrillation episodes. Furthermore, as... Figure 5 As shown in Figure C, in the isolated human heart perfusion model, QUIN perfusion also increased the atrial fibrillation induction rate and prolonged atrial fibrillation episodes. These results from cross-cellular, cross-animal species, and isolated human heart models collectively demonstrate that QUIN can directly enhance atrial electrical instability and is an important metabolic signal promoting an increased risk of new-onset atrial fibrillation.

[0099] Example 2: QUIN expression level was significantly associated with the risk of atrial fibrillation in aged rats. Test methods This study aimed to verify whether circulating and atrial QUIN levels are elevated in aging individuals and whether they are associated with susceptibility to atrial fibrillation. Young adult male Sprague-Dawley rats were selected as the control group, and aged male Sprague-Dawley rats were selected as the experimental group. Young rats weighed approximately 250-300 g, and aged rats weighed approximately 550-750 g. All animals were housed under identical conditions, and no evidence of baseline atrial fibrillation was observed prior to the experiment.

[0100] Peripheral blood samples were collected from rats in each group. Plasma was separated and used to detect circulating QUIN levels. Simultaneously, atrial tissue was collected, washed with pre-cooled PBS, weighed, homogenized, and then the QUIN content in the atrial tissue was detected using a commercially available QUIN ELISA kit. To evaluate susceptibility to atrial fibrillation (AF), programmed atrial stimulation was used to induce AF, and the AF induction and duration were recorded for each animal. The relationship between circulating QUIN levels, atrial tissue QUIN levels, and AF induction rate and duration was then analyzed.

[0101] Test results like Figure 6 As shown in Figure A, compared with young rats, the level of QUIN in the atrial tissue of aged rats was significantly increased. Figure 6 As shown in Figure B, the circulating QUIN level in aged rats was significantly higher than that in young rats, and aged rats were more susceptible to atrial fibrillation under programmed atrial stimulation, with atrial fibrillation lasting longer. These results indicate that aging is accompanied by an increase in circulating and local atrial QUIN, and that this increase in QUIN is consistent with increased susceptibility to atrial fibrillation. Figure 6 (C)

[0102] In this embodiment, ELISA results of atrial tissue showed that the content of QUIN in the atria of aged rats was significantly higher than that in young controls, suggesting that QUIN is not only a marker of changes in circulating metabolism, but can also accumulate locally in the aging atria. Since the origin and maintenance of atrial fibrillation are closely related to the local electrophysiological environment of the atria, this result provides tissue-level support for QUIN as a biomarker for predicting the risk of new-onset atrial fibrillation in the elderly.

[0103] Example 3: QUIN accumulates locally in aging atria, mainly originating from macrophages. Test methods This embodiment was used to determine the main cellular source of QUIN accumulation in aging atria. Atrial tissues from young and aged rats were collected, and cell morphology localization analysis was performed using these atrial tissue samples.

[0104] (1) Single-nuclear RNA sequencing analysis: Single nuclei were isolated, constructed, and sequenced from human left atrial tissue. Single-nuclear transcriptome data were obtained using the 10xGenomics platform. The obtained cell nuclei were subjected to quality control, dimensionality reduction clustering, and cell type annotation. The activity of the tryptophan-kynurenine metabolic pathway in different cell types was evaluated using AUCell, UCell, singscore, and GSVA.

[0105] (2) Immunofluorescence localization: Human and rat atrial tissue sections were subjected to immunofluorescence staining for QUIN, HAAO, and CD68. HAAO is a key enzyme in QUIN synthesis, and CD68 is used to label macrophages. Colocalization analysis was used to determine whether HAAO or QUIN signals were mainly located in CD68-positive macrophages.

[0106] (3) Flow cytometry analysis and cell sorting: Rat atrial tissue was digested into a single-cell suspension, and flow cytometry analysis was performed using CD45 and CD68 antibodies to count the proportion of atrial macrophages. The main cell populations, such as atrial macrophages, fibroblasts and cardiomyocytes, were further sorted, and the QUIN content in each cell population was detected.

[0107] (4) Macrophage depletion verification: Macrophages were depleted by administering clophosphamide liposomes to aged rats. After administration, the level of QUIN in atrial tissue was measured to verify the functional contribution of macrophages to atrial QUIN accumulation.

[0108] Test results like Figure 7 China A- Figure 7 As shown in Figure E, single-nuclear RNA sequencing revealed that tryptophan-kynurenine metabolism-related signals in the atria were selectively enriched in myeloid cells / macrophages, while their enrichment was lower in structural cells such as cardiomyocytes, fibroblasts, and endothelial cells. Furthermore, immunofluorescence showed that the key enzyme HAAO in kynurenine synthesis was mainly located in CD68-positive macrophages.

[0109] Flow cytometry results showed ( Figure 7 In middle F, the proportion of CD68-positive macrophages in the atria of aged rats was significantly increased. Cell sorting analysis showed that the increased QUIN was mainly observed in macrophages, rather than fibroblasts or cardiomyocytes. The QUIN content in atrial macrophages of aged rats was significantly higher than that in macrophages of young rats, while the changes in QUIN in fibroblasts and cardiomyocytes were smaller. Figure 7 (G).

[0110] After macrophage depletion using clophosphamide liposomes, the atrial QUIN level in aged rats decreased significantly, for example, from approximately 0.025 nmol / mg tissue to approximately 0.005 nmol / mg tissue. Figure 7 (H). This result indicates that macrophages not only spatially co-localize with the QUIN signaling pathway but also make a substantial contribution to QUIN accumulation in aging atria. These results collectively demonstrate that macrophages are the primary cellular source of QUIN accumulation in aging atria.

[0111] Example 4: Evidence of Intervention Reversibility Test methods This embodiment is used to verify whether the QUIN-related pathway is operable, and to further illustrate that targeting QUIN-derived or downstream receptor signals can reduce susceptibility to atrial fibrillation or promote the termination of established atrial fibrillation.

[0112] (1) Macrophage depletion experiment: Aged male rats were randomly divided into Vehicle liposome group and clodronate liposome group. Clodronate liposome was administered intravenously at a dose of 10 mg / kg once a week for two consecutive weeks. After administration, the atrial QUIN level was measured, and atrial fibrillation was induced by atrial programmed stimulation. The atrial fibrillation induction rate and duration were recorded.

[0113] (2) Macrophage HAAO targeted silencing experiment: AAV9-CD68-shHAAO vector was used to target and silence HAAO expression in macrophages, while the control group was given AAV9-CD68-scramble. Four weeks after administration, HAAO protein expression, atrial QUIN level, atrial fibrillation induction rate and duration of atrial fibrillation were detected.

[0114] (3) Downstream drug prevention experiment: In the rat QUIN exposure model, minocycline (10 mg / kg, intraperitoneal injection), a QUIN synthesis inhibitor, or caroverine (10 mg / kg, intravenous injection), an iGluR blocker, was administered, followed by QUIN exposure and atrial programmed stimulation to evaluate the effects of each intervention on the atrial fibrillation induction rate and duration of atrial fibrillation.

[0115] (4) Atrial fibrillation termination experiments have been established: In elderly rat atrial fibrillation models, in vivo porcine atrial fibrillation models, and isolated human heart perfusion atrial fibrillation models, after establishing persistent atrial fibrillation, caroverine or the positive control drug propafenone were administered, and the termination of atrial fibrillation, the sinus rhythm recovery rate, and the conversion time were recorded. All induction, recording, and interpretation were performed by blinded researchers.

[0116] Test results This invention demonstrates through two types of experiments—upstream source intervention and downstream receptor blockade—that the QUIN pathway is clearly operable, and that intervention in this pathway can reduce susceptibility to atrial fibrillation or promote the termination of established atrial fibrillation.

[0117] like Figure 8 In the middle, A represents the electrocardiogram (ECG). For example... Figure 8 As shown in Figure B, in aged rats, after macrophage depletion using clophosphamide liposomes, the atrial fibrillation induction rate decreased from 7 / 8 to 2 / 8, and the duration of atrial fibrillation was significantly shortened. Further targeting and silencing the macrophage's QUIN enzyme HAAO with AAV9-CD68-shHAAO resulted in a decrease in HAAO protein expression. Figure 8In the middle C), the atrial fibrillation induction rate also decreased from 7 / 8 to 2 / 8, and the duration of atrial fibrillation was significantly shortened ( Figure 8 China D and Figure 8 (E). This result demonstrates that macrophage HAAO-dependent QUIN production is an important upstream factor in susceptibility to age-related atrial fibrillation.

[0118] like Figure 9 As shown in Figure A, regarding downstream drug intervention, both minocycline, a QUIN synthesis inhibitor, and caroverine, an iGluR blocker, can reduce susceptibility to atrial fibrillation after QUIN exposure. In a rat QUIN exposure model, the atrial fibrillation induction rate was 6 / 7 in the Vehicle group, while it decreased to 1 / 7 in both the minocycline and caroverine groups, suggesting that inhibiting QUIN production or blocking its downstream iGluR signaling has an anti-atrial fibrillation effect.

[0119] like Figure 9 As shown in Figure B, in the established treatment model of atrial fibrillation, caroverine not only reduces susceptibility to atrial fibrillation but also promotes its termination. In an aged rat model of atrial fibrillation, the sinus rhythm recovery rate was 88.9% in the caroverine group and 37.5% in the propafenone group, with caroverine significantly shortening the conversion time.

[0120] like Figure 10 As shown in Figure A, in a porcine atrial fibrillation model, caroverine terminated atrial fibrillation in 5 out of 5 animals, while propafenone terminated it in only 1 out of 5. Figure 10 As shown in Figure B, in an isolated human heart model, caroverine terminated atrial fibrillation within 2 minutes in all tested specimens, while propafenone had a conversion rate of approximately 25%. These results indicate that targeting the downstream iGluR signaling of QUIN has cross-species therapeutic potential for atrial fibrillation.

[0121] The combined results of experiments involving macrophage exhaustion, macrophage HAAO-targeted silencing, inhibition of QUIN synthesis, and iGluR blockade indicate that the QUIN-related pathway is clearly modifiable. This pathway can serve as a mechanistic basis for predicting the risk of new-onset atrial fibrillation and also as an important direction for subsequent interventions and drug development.

[0122] Example 5: Clinical validation of the association between QUIN and the risk of atrial fibrillation. Inclusion criteria The study included elderly community residents aged ≥65 years who had no history of atrial fibrillation at baseline as confirmed by electrocardiogram or medical records (for longitudinal analysis of new-onset atrial fibrillation), had available frozen plasma samples, and had signed extensive informed consent.

[0123] If the participants agree to be followed up, we can obtain accurate and detailed annual physical examination data, 12-lead electrocardiogram data, 24-hour Holter monitoring data, outpatient and emergency medical records, inpatient medical records, and death registration data. Exclusion criteria Age <65 years old; The baseline atrial fibrillation status could not be determined.

[0124] The baseline plasma sample volume is insufficient or the quality is substandard, making it impossible to complete the QUIN test.

[0125] Baseline information on key covariates is severely lacking (such as age, sex, or history of hypertension).

[0126] Sample collection Data from the Shanghai Community Elderly Cohort included 4,220 participants with no atrial fibrillation at baseline, aged ≥65 years (1,890 males and 2,330 females). Baseline assessment and biosample collection took place from January 1, 2017 to May 31, 2017, with administrative follow-up ending on October 31, 2025. The median follow-up time was 8.58 years (IQR 8.49–8.69 years), with a total of 32,747 follow-ups. 242 new cases of atrial fibrillation occurred, 675 died from all causes, and 55 were lost to follow-up due to interruptions in subsequent recording.

[0127] Judgment criteria The outcome of atrial fibrillation was determined independently by two cardiologists under unknown QUIN concentration conditions, with any discrepancies decided by a third senior reviewer.

[0128] New-onset atrial fibrillation event is defined as the first recorded 12-lead electrocardiogram, 24-hour Holter monitoring, or medical record conforming to ICD-10 code I48.x.

[0129] Analytical methods The method's repeatability and robustness are ensured through standardized QUIN assays (uniform reagent batches, internal standard calibration, LC-MS / MS), rigorous sample processing, statistical model robustness analysis, and sensitivity analysis (subgroup analysis).

[0130] Test methods Venous blood samples were collected, plasma was separated by centrifugation and stored at -80°C to ensure sample stability. Plasma QUIN levels were quantitatively determined using liquid chromatography-tandem mass spectrometry (LC-MS / MS). A deuterium-labeled internal standard (d3QUIN) was used for calibration to ensure accuracy and batch-to-batch consistency. The calibration curve covered the physiological concentration range, and the measurement repeatability was good (intra / inter-assay CV < 8%).

[0131] Demographic, clinical risk factors, and laboratory indicators were collected from the sample: age, sex, body mass index (BMI), systolic blood pressure (SBP), diabetes, hsCRP, serum creatinine (Scr), serum total cholesterol (TC), low-density lipoprotein (LDL), and thyroid-stimulating hormone (TSH), etc.

[0132] Record follow-up time and outcome events (new-onset atrial fibrillation, death, loss to follow-up) to ensure the integrity of model fitting.

[0133] The association between plasma QUIN and the risk of new-onset atrial fibrillation was quantified using a Cox proportional hazards model and a Fine-Gray competing hazard model. Multifactor correction was used to eliminate confounding effects and ensure the independence and reliability of the QUIN signal.

[0134] like Figure 11 As shown in Figure a, 4,367 community-dwelling individuals aged ≥65 years with baseline plasma QUIN test results were included. After excluding 147 individuals with pre-existing atrial fibrillation, a final total of 4,220 participants were included for follow-up analysis. The follow-up period was from January 1, 2017 to October 31, 2025, with a median follow-up time of 8.58 years (IQR 8.49–8.69). During the follow-up period, 242 participants developed new-onset atrial fibrillation, accounting for 5.73% of the included population; 675 participants died before the onset of atrial fibrillation, accounting for 15.99%; 3,248 participants did not develop atrial fibrillation at the end of the follow-up period; and 55 participants were lost to follow-up or withdrew before the end of the follow-up period, resulting in a total of 3,303 participants being censored. The baseline plasma QUIN level of subjects who developed new-onset atrial fibrillation (115.7 ± 37.1 ng / mL) was significantly higher than that of subjects who did not have atrial fibrillation during follow-up (89.3 ± 88.9 ng / mL), and the difference was statistically significant (P=0.001).

[0135] like Figure 11As shown in Figure b, stratification was performed according to the quartiles of baseline plasma QUIN concentration, with the lowest quartile (Q1) as the reference. A multivariate Cox proportional hazards model was used to analyze the relationship between QUIN levels and the risk of new-onset atrial fibrillation. After adjusting for multiple factors including age, sex, body mass index, systolic blood pressure, diabetes, history of myocardial infarction, history of stroke, smoking, alcohol consumption, high-sensitivity C-reactive protein, serum creatinine, low-density lipoprotein cholesterol, and thyroid-stimulating hormone, the risk of new-onset atrial fibrillation increased progressively with increasing QUIN levels: the HR for Q2 was 3.80, with a 95% CI of 1.55–9.33 (P = 0.003); the HR for Q3 was 7.67, with a 95% CI of 3.26–18.00 (P = 0.001); and the HR for Q4 was 21.91, with a 95% CI of 9.56–50.28 (P = 0.001). The above results indicate that higher baseline QUIN concentrations are significantly associated with an increased risk of new-onset atrial fibrillation, and show a clear dose-response relationship.

[0136] like Figure 11 As shown in Figure c, based on ROC curve analysis, subjects were divided into a high-QUIN group (≥88.02 ng / mL, n=1,838) and a low-QUIN group (<88.02 ng / mL, n=2,382) using 88.02 ng / mL, corresponding to the maximum Yoden index, as the cutoff value. Fine-Gray analysis of the cumulative incidence of atrial fibrillation, considering death as a competing event, showed that the cumulative incidence of new-onset atrial fibrillation was significantly higher in the high-QUIN group than in the low-QUIN group (Gray test, P<0.001). At follow-up of 0, 2, 4, 6, and 8 years, the number of patients at risk in the low-QUIN group was 2,382, 2,346, 2,280, 2,120, and 1,992, respectively, with corresponding cumulative atrial fibrillation events of 0, 0, 3, 24, and 41. The number of patients at risk in the high-QUIN group was 1,838, 1,772, 1,629, 1,475, and 1,325, respectively, with corresponding cumulative atrial fibrillation events of 0, 22, 94, 140, and 180.

[0137] like Figure 11As shown in Figure d, the summative risk hrs (SHRs) of atrial fibrillation (AF) estimated based on the Fine-Gray competing risk model (with death as a competing event) further confirm that high QUIN levels are significantly associated with an increased risk of new-onset AF. Unadjusted, the SHR in the high QUIN group was 5.81, with a 95% CI of 4.21–8.02 (P<0.001); after adjusting for age and sex, the SHR was 5.30, with a 95% CI of 3.83–7.34 (P<0.001); even after adequate adjustment for the above multivariates, the SHR remained at 5.84, with a 95% CI of 4.02–8.49 (P<0.001). These results indicate that, even after considering the competing risk of death in the elderly, baseline high QUIN levels remain independently associated with the risk of new-onset AF.

[0138] Figure 12 The association between plasma quinolinic acid and new-onset atrial fibrillation (ad) and all-cause mortality (e and f) in an elderly cohort.

[0139] like Figure 12 As shown in Figure a, the ROC curve for baseline plasma QUIN predicting atrial fibrillation during follow-up was presented; the AUC was 0.745, with a 95% CI (0.717–0.773). This suggests that baseline plasma QUIN has good predictive value for new-onset atrial fibrillation in the elderly population. The red cross marks the optimal cutoff value for the Youden index (88.02 ng / mL), at which the sensitivity was 0.81 and the specificity was 0.59. Based on this, a high QUIN group (≥ 88.02 ng / mL) and a low QUIN group (< 88.02 ng / mL) were defined.

[0140] Figure 12 In the middle b, the Kaplan-Meier estimates of atrial fibrillation-free survival were found in the high-QUIN group and the low-QUIN group. The atrial fibrillation-free survival probability of the high-QUIN group was significantly lower than that of the low-QUIN group, and the curves of the two groups began to separate from the early follow-up period, and the difference continued to widen during the follow-up period (log-rank test, P<0.001).

[0141] like Figure 12 As shown in Figure c, after stratification according to baseline plasma QUIN quartiles, the probability of atrial fibrillation-free survival gradually decreased with increasing QUIN quartiles, with the lowest probability of atrial fibrillation-free survival in group Q4 and the highest in group Q1. The difference between groups was statistically significant (log-rank test, P<0.001).

[0142] like Figure 12As shown in Figure d, when death is considered a censored event, the cumulative incidence of atrial fibrillation (AF) estimated using the 1-Kaplan-Meier method was significantly higher in the high-QUIN group than in the low-QUIN group. The high-QUIN group had 22, 94, 140, and 180 cumulative AF events at 2, 4, 6, and 8 years of follow-up, respectively, while the low-QUIN group had 0, 3, 24, and 41 events at the same times. Because the 1-Kaplan-Meier method treats death as a common censoring event, it may overestimate the actual cumulative risk of AF. Therefore... Figure 11 c and Figure 11 The Fine-Gray competitive risk analysis (death as a competitive event) shown in d is the main inference result.

[0143] Figure 12 In the figure, e represents the higher, unadjusted, cumulative all-cause mortality rate in the high-QUIN group after atrial fibrillation was considered a competing event. The cumulative number of deaths in the low-QUIN group at 2, 4, 6, and 8 years of follow-up were 36, 91, 216, and 314, respectively; while the cumulative number of deaths in the high-QUIN group at the same times were 44, 114, 214, and 312, respectively.

[0144] Figure 12 The subdivided hazard ratio (SHR) of baseline plasma QUIN level to all-cause mortality risk was estimated using a Fine-Gray competing risk model with atrial fibrillation as the competing event. Results showed that, without adjusting for any covariates, the high QUIN group (relative to the low QUIN group) had a significantly higher risk of all-cause mortality, with an SHR of 1.33 (95% CI: 1.14–1.55, P < 0.001). However, after adjusting for age and sex, this association was no longer statistically significant (SHR = 0.99, 95% CI: 0.85–1.16, P = 0.924); after further adjusting for several potential confounding factors, the SHR was 0.87 (95% CI: 0.73–1.03, P = 0.109). These results indicate that the crude association between QUIN and all-cause mortality is mainly attributed to confounding factors such as age and sex, and that QUIN is not an independent risk factor for all-cause mortality after adequate adjustment. Combined with the finding in this study that QUIN is independently associated with new-onset atrial fibrillation ( Figure 12 a- Figure 12 (e) This result supports the view that baseline plasma QUIN has relatively specific predictive value for the occurrence of atrial fibrillation, rather than being a generalized marker of mortality risk.

[0145] like Figure 13As shown, a multivariate Cox proportional hazards model (same as the multivariate model described above) was used to assess the association between high QUIN levels (defined as baseline plasma QUIN ≥ 88.02 ng / mL, with < 88.02 ng / mL as the reference) and the risk of new-onset atrial fibrillation in pre-specified clinical subgroups. Specific results for each subgroup are as follows: Age stratification: In the <75-year-old subgroup (n=3,027), the high QUIN group had a significantly increased risk (HR=4.46, 95%CI: 2.73-7.29, P<0.001); the same was true in the ≥75-year-old subgroup (n=1,193) (HR=6.05, 95%CI: 3.95-9.27, P<0.001).

[0146] Gender stratification: Male (n=1,888) HR=6.70 (95% CI: 4.31-10.44, P<0.001); Female (n=2,330) HR=5.42 (95% CI: 3.39-8.65, P<0.001).

[0147] Hypertension stratification: HR=4.82 (95%CI: 3.26-7.15, P<0.001) in patients with hypertension (n=2,385); HR=8.36 (95%CI: 4.69-14.89, P<0.001) in patients without hypertension (n=1,758).

[0148] Stratification of diabetes: HR=7.57 (95%CI: 3.39-16.95, P<0.001) in patients with diabetes (n=798); HR=5.74 (95%CI: 4.02-8.19, P<0.001) in patients without diabetes (n=3,327).

[0149] Stratification of previous myocardial infarction: HR=6.00 (95%CI: 4.30-8.38, P<0.001) for those without previous myocardial infarction (n=4,072); HR=1.83 (95%CI: 0.35-9.46, P=0.468) for those with previous myocardial infarction (n=49). Due to the small sample size in this subgroup, the estimation accuracy is insufficient, and the results need to be interpreted with caution.

[0150] Stratification of prior stroke: HR=3.86 (95%CI: 1.57-9.53, P=0.003) for those with a history of stroke (n=3,848); HR=6.30 (95%CI: 4.47-8.89, P<0.001) for those without a history of stroke.

[0151] As shown in Table 1, the multivariate model adjusted for age, sex, body mass index (BMI), systolic blood pressure (SBP), diabetes, history of myocardial infarction and stroke, smoking status, alcohol consumption, high-sensitivity C-reactive protein (hs-CRP), serum creatinine (Scr), low-density lipoprotein cholesterol (LDL-C), and thyroid-stimulating hormone (TSH). The binary cutoff value for QUIN (88.02 ng / mL) was derived from receiver operating characteristic (ROC) analysis maximizing the Youden index. P-values ​​were two-tailed tests. Participants were censored at death, loss to follow-up, or at the end of follow-up.

[0152] Table 1. Association between baseline plasma QUIN and new-onset atrial fibrillation in the Cox proportional hazards model.

[0153] In summary, the association between high baseline plasma QUIN levels and an increased risk of new-onset atrial fibrillation remained consistent across clinical subgroups of different ages, sexes, hypertension status, diabetes status, and previous stroke status. Except for the subgroup with a history of myocardial infarction, where the association was not statistically significant due to the small sample size, high QUIN levels were significantly associated with an increased risk of new-onset atrial fibrillation in all other subgroups. These results suggest that plasma QUIN levels could serve as a potential circulating biomarker for assessing susceptibility to atrial fibrillation in older adults.

[0154] A cohort of 3540 elderly individuals aged 65-80 years was selected for a 3-year follow-up, and data were recorded. A risk model was used for data fitting. Based on the model fitting results and the cumulative incidence rate (CIF), the plasma QUIN threshold corresponding to reaching the preset cumulative risk (e.g., a 3-year cumulative incidence rate of atrial fibrillation of 90%) was calculated. All participants in this study were individuals without atrial fibrillation at baseline; therefore, at the start of follow-up... At that time, the cumulative incidence of new-onset atrial fibrillation was 0. It should be noted that in the following formula... This does not refer to the incidence of atrial fibrillation at the time of enrollment, nor to the baseline prevalence of atrial fibrillation, but rather to the follow-up time under reference conditions in the Fine-Gray model. The baseline cumulative subdistribution hazard function was estimated at that time. Therefore, although none of the applicants in the cohort had atrial fibrillation at enrollment, the cumulative incidence of new-onset atrial fibrillation at enrollment was 0; however, at 3 years of follow-up, the model could estimate the cumulative incidence of new-onset atrial fibrillation events and competing death events that actually occurred during the follow-up period. And based on this, the plasma QUIN threshold is deduced. "No atrial fibrillation at baseline" and " The statement that it can be used to reverse-engineer the 3-year target risk threshold is not contradictory.

[0155] The plasma QUIN threshold was not empirically set, but rather derived by back-calculation based on a Fine-Gray competing hazard model established from the applicant's clinical cohort. Since death during follow-up affects the observation of new-onset atrial fibrillation, this application treats death as a competing event and uses the model to estimate the relationship between QUIN levels and the cumulative incidence of new-onset atrial fibrillation over 3 years. Specifically, the regression coefficients of the QUIN variable and the cumulative subdistribution hazard function of the 3-year baseline were first obtained using the Fine-Gray model. Then, the target cumulative incidence of new-onset atrial fibrillation over 3 years was set to 90% or 95%, substituted into the model's cumulative incidence expression, and the QUIN concentration was inversely solved to obtain the corresponding thresholds of 215.5 ng / mL and 280.4 ng / mL.

[0156] set up: This indicates the measured concentration of QUIN in plasma, in ng / mL. : Represents continuous variables included in the Fine-Gray model; : indicates that in the Fine-Gray subdistribution risk model The regression coefficients; : Indicates the Fine-Gray model followed up to time under reference conditions. The baseline cumulative subdistribution risk function; : Indicates the cumulative incidence of new-onset atrial fibrillation; In the Fine-Gray competing risk model, the sub-distribution risk function for new-onset atrial fibrillation can be expressed as:

[0157] Further, the plasma QUIN threshold was determined. The calculation formula is:

[0158] This formula is the plasma QUIN threshold back-calculation formula used in this application.

[0159] Test results 1. Classification threshold: 88.02 ng / mL Patients with higher baseline plasma QUIN had a significantly increased risk of developing new-onset atrial fibrillation (AF) (HR Q4 vs Q1 = 21.91; Fine-Gray SHR ~5.84), and this risk was dose-responsive. Follow-up results showed that patients who subsequently developed AF had significantly higher baseline plasma QUIN than those who did not. The AF was new-onset.

[0160] ROC analysis showed that the AUC of plasma QUIN for predicting atrial fibrillation (AF) was 0.745 (95% CI 0.717-0.773), with a Youden-optimal cutoff of 88.02 ng / mL; the specificity was 0.81 and the sensitivity was 0.59. Compared with existing prediction methods that use non-specific inflammatory markers (such as hsCRP) or traditional cardiovascular risk factors, this invention significantly improves the specificity and accuracy of predicting the risk of new-onset atrial fibrillation in the elderly.

[0161] The threshold of 88.02 ng / mL is derived from the optimal cutoff value corresponding to the maximization of the Youden exponent in the ROC curve analysis, and is used to distinguish between the high-QUIN group and the low-QUIN group.

[0162] 2. Back-calculation of target risk level based on competitive risk model The plasma QUIN threshold corresponding to a 90% cumulative incidence of atrial fibrillation over 3 years was calculated, providing a decision-making basis for early clinical screening and intervention in high-risk populations. In this embodiment, the follow-up period is set to 3 years, i.e.: After fitting the applicant's clinical cohort data to a Fine-Gray sub-distribution risk model using R v4.3.0, the following results were obtained: , which is the baseline cumulative subdistribution risk function at 3 years.

[0163] The corresponding reference rate for the cumulative incidence of new-onset atrial fibrillation over 3 years is approximately: ; When the cumulative incidence of new-onset atrial fibrillation over the target 3-year period is 90%: ; When the cumulative incidence of new-onset atrial fibrillation over the target 3-year period is 95%: .

[0164] When the time frame is limited to 3 years and the cumulative risk of atrial fibrillation is 90%, the threshold for QUIN in biological samples is 215.5 ng / mL; that is, when the QUIN content in the subject's biological sample is ≥215.5 ng / mL, the predicted cumulative risk of atrial fibrillation in the subject within 3 years is greater than 90%.

[0165] When the time frame is limited to 3 years, and the cumulative risk of atrial fibrillation is 95%, the concentration of QUIN in the biological sample is 280.4 ng / mL. That is, when the concentration of QUIN in the subject's biological sample is ≥280.4 ng / mL, the predicted cumulative risk of atrial fibrillation within 3 years is greater than 95%. This refers to new-onset atrial fibrillation.

[0166] The aforementioned thresholds have clear statistical model sources, parameter sources, mathematical derivation processes, and repeatable calculation paths. The thresholds calculated using the formulas provided in this invention can be directly applied to clinical practice as quantitative screening standards for high-risk populations.

Claims

1. Use of reagents for detecting QUIN expression levels in biological samples in the preparation of kits for diagnosing the risk of atrial fibrillation in subjects.

2. The use as described in claim 1, wherein the biological sample is a blood sample; the subject is a mammal; and the atrial fibrillation includes paroxysmal atrial fibrillation, persistent atrial fibrillation, long-term persistent atrial fibrillation, permanent atrial fibrillation, and new-onset atrial fibrillation.

3. The use as described in claim 1, wherein the biological sample is plasma; the subject is an elderly person aged 65-80 years; and the atrial fibrillation is new-onset atrial fibrillation.

4. A kit for predicting the risk of atrial fibrillation, the kit containing a reagent for detecting the expression level of QUIN in a subject's biological sample.

5. The kit according to claim 4, wherein the biological sample is a blood sample; the subject is a mammal; and the atrial fibrillation includes paroxysmal atrial fibrillation, persistent atrial fibrillation, long-term persistent atrial fibrillation, permanent atrial fibrillation, and new-onset atrial fibrillation.

6. The kit according to claim 4, wherein the biological sample is plasma; the subject is an elderly person aged 65-80 years; and the atrial fibrillation is new-onset atrial fibrillation.

Images