Methods for establishing a predictive model for renal histopathological assessment of ANCA-associated nephritis in children

Abstract

This application provides a method for establishing a renal histopathological assessment and prediction model for ANCA-associated nephritis in children, relating to the field of medical technology. This method establishes a GSCIT classification model for ANCA renal histopathology in children using five indicators: glomerular sclerosis, segmental sclerosis, crescent formation, interstitial inflammatory cell infiltration / tubular necrosis, and tubular atrophy / interstitial fibrosis. This model is applicable to the pathological classification of AAGN in children, aiding in treatment and prognosis assessment. Compared with existing classification systems, this invention has higher reference value for the clinical pathological diagnosis, clinical medication, and prognostic assessment of renal injury in children with AAGN, and is suitable for widespread application.

Description

Technical Field

[0001] This application relates to the field of medical technology, and in particular to a method for establishing a renal tissue pathological assessment and prediction model for ANCA-associated nephritis in children. Background Technology

[0002] AAV (anti-vasculitis) is a dangerous condition that can simultaneously damage multiple organs, including the kidneys, lungs, nervous system, musculoskeletal system, gastrointestinal tract, and skin. The kidneys are among the most commonly affected organs, leading to ANCA-associated glomerulonephritis (AAGN), a critical factor threatening patients' lives and determining their prognosis. Clinical studies show that 70-80% of patients will exhibit clinical symptoms of kidney involvement. Unlike adult patients, kidney involvement is more common and severe in children with AAV; 29-40% of children with MPA and approximately 10% of children with GPA progress to end-stage renal disease (ESRD). Some children may only have kidney involvement without other extrarenal manifestations, with an insidious onset, presenting with ESRD already present.

[0003] Once a child is diagnosed with AAGN, timely renal prognostic risk assessment is crucial to guide precise clinical treatment. The renal pathology of AAV patients presents with multiple different lesions and / or different stages of lesions, such as cellular, fibrous crescentic, segmental fibrinoid necrosis of the glomeruli, and glomerular sclerosis coexisting. Some patients may also have fibrinoid necrosis of the renal arterioles outside the glomerulus. Interstitial renal lesions manifest as varying degrees and extents of lymphocyte, monocyte, and plasma cell infiltration, with late-stage interstitial fibrosis and tubular atrophy. In recent years, organ damage assessment and prognostic indicators related to renal prognosis have gradually gained attention in clinical research. The Berden score for vasculitis, the AAV organ damage assessment—Vasculitis Injury Index (VDI), the Disease Extent Index (DEI), and the 5-Factor FFS have played a certain predictive role in the disease activity and overall prognosis of AAV, but they have not provided guidance for the assessment and treatment of renal histopathology and are not suitable for children. In 2010, the European Vasculitis Study Group (EUVAS) proposed the Berden pathological classification of anti-neutrophil cytoplasmic antibody (ANCA) based on the characteristics of renal tissue structure and pathological changes, classifying ANCA into focal, crescent, sclerotic, and mixed types. Currently, it is generally believed that the focal type has the best prognosis, while the sclerotic type has the worst prognosis. However, a meta-analysis involving more than 1,500 patients showed no significant difference in prognosis between the crescent and mixed types (Huang S, Shen Q, Yang R, Lai H, Zhang J. An evaluation of the 2010 histoathological classification of anti-neutrophil cytoplasmic antibody (ANCA)-associated glomerulonephritis: a Bayesian network meta-analysis. Int Urol Nephrol. Oct 2018; 50(10):1853-1861.), suggesting that this pathological classification has certain limitations in predicting renal prognosis. Research from the Department of Nephrology at Peking University First Hospital also indicates that pathological classification is not an independent predictor of end-stage renal disease (ESRD). In 2017, Brix et al., through a two-way study, proposed a new scoring system—the ANCA renal risk score (ARRS)—to predict renal outcomes in AAGN patients at the early stage of diagnosis. This clinical score includes three parameters: the proportion of normal glomeruli, the estimated glomerular filtration rate (eGFR) at diagnosis, and tubular atrophy and interstitial fibrosis, classifying the risk of ESRD into low, intermediate, and high-risk groups.However, these studies did not offer further suggestions for optimizing treatment strategies. How to stratify treatment based on the risk of poor renal outcomes remains a pain point and challenge in the precise diagnosis and treatment of AAGN.

[0004] The applicant has established a pediatric ANCA renal histopathology GSCIT classification (including glomerular sclerosis, segmental sclerosis, total crescent, interstitial inflammatory cell infiltration / tubular necrosis, and tubular atrophy / interstitial fibrosis). Unlike the adult Berden classification, we found that interstitial inflammatory cell infiltration / tubular necrosis is closely related to prognosis. Clinical validation is currently underway in several hospitals, including Chongqing Children's Hospital and the Children's Hospital Affiliated to Zhejiang University, preliminarily confirming the clinical value of this pathological classification.

[0005] Therefore, analyzing independent risk factors affecting the prognosis of AAGN in children and establishing a suitable renal pathological classification and prediction system for AAGN in children to assist in treatment has become a technical challenge that urgently needs to be solved in this field. Summary of the Invention

[0006] To address the aforementioned issues, this application provides a method for establishing a renal histopathological assessment and prediction model for ANCA-associated nephritis in children for auxiliary medical purposes, which predicts prognosis and assists in treatment.

[0007] Specifically, this application is implemented through the following technical solution:

[0008] First, univariate analysis was performed. Based on the current consensus on pathological classification, pathological changes were graded and scored according to the proportion of affected individuals. Cox survival curve analysis was used to optimize the proportion grading and related pathological grading indicators. Further, a multivariate Cox regression model was conducted using univariate analysis with p-values ​​<0.1 and relevant indicators selected based on clinical experience. Ultimately, five indicators were established to create a GSCIT classification model for pediatric ANCA renal tissue pathology: glomerular sclerosis, segmental sclerosis, crescent formation, renal interstitial inflammatory cell infiltration / tubular necrosis, and renal tubular atrophy / interstitial fibrosis.

[0009] The above model was established based on retrospective and prospective cohort studies. In the retrospective study, based on different renal outcomes, a unified standard for interpreting renal biopsy pathology slides was developed using the existing AAGN renal biopsy pathology database and referring to the Berden classification, based on the renal biopsy pathology characteristics of pediatric AAGN. Three renal pathologists independently reviewed the slides, and the results were then compiled. Each pathology slide was re-read, and univariate and multivariate Cox regression analyses were performed to analyze pathological evaluation indicators related to renal prognosis and progression to ESRD. Simultaneously, univariate and multivariate Cox regression analyses were performed again, combining pathological indicators with clinical manifestations and laboratory tests, such as demographic characteristics, baseline eGFR before renal biopsy, serum creatinine, blood urea nitrogen, serum albumin, complement C3, ANCA antibody type, ANCA titer, AAV clinical type, and treatment regimen, to analyze independent risk factors affecting the prognosis of AAGN. The Kaplan-Meier method was used to analyze renal survival time.

[0010] The above model is applicable to the pathological classification of AAGN in children to guide treatment and predict prognosis. Compared with existing classification systems, this invention has higher reference value for the clinical and pathological diagnosis, clinical medication, and prognostic assessment of AAGN-related kidney injury in children, thus assisting medical treatment, improving prognosis, and is suitable for widespread application. Attached Figure Description

[0011] Figure 1 This is a schematic diagram illustrating the process of establishing a prediction model for an example.

[0012] Figure 2 The correlation between renal biopsy pathological lesions and ESRD is analyzed using Cox survival curves; where A and B are schematic diagrams of the correlation between normal glomeruli, glomerular endothelial cell proliferation and ESRD, respectively.

[0013] Figure 3 For based on Figure 2 Schematic diagram before and after optimization of glomerular segmental sclerosis in proportional grading based on correlation analysis;

[0014] A and B are the relevant pathological grading indicators before and after optimization, respectively.

[0015] Figure 4 For based on Figure 2 Schematic diagram before and after optimization of spherical hardening by correlation analysis and proportional gradation;

[0016] A and B are the relevant pathological grading indicators before and after optimization, respectively.

[0017] Figure 5 For based on Figure 2 Schematic diagram of glomerular crescent before and after optimization of proportional grading based on correlation analysis;

[0018] A and B are the relevant pathological grading indicators before and after optimization, respectively.

[0019] Figure 6 For based on Figure 2 A schematic diagram of the cell / cell fibrous crescent before and after optimization of the proportional grading of the correlation analysis; where A and B are the relevant pathological grading indicators before and after optimization, respectively.

[0020] Figure 7 For based on Figure 2 A schematic diagram of the fibrous crescent body before and after optimization of the proportional grading based on correlation analysis; where A and B are the relevant pathological grading indicators before and after optimization, respectively.

[0021] Figure 8 The results represent the grading of relevant pathological indicators; where A and B are the grading indicators for renal tubular atrophy / interstitial fibrosis and renal tubular inflammatory cell infiltration / renal tubular necrosis, respectively.

[0022] Figure 9 To verify the statistical results of the distribution of pathological changes in renal biopsies in the cohort; where AF represents the statistical results of Berden classification, segmental glomerular sclerosis, glomerular crescent, glomerular sclerosis, tubular atrophy / interstitial fibrosis, and tubular inflammatory cell infiltration / tubular necrosis grading.

[0023] Figure 10 To analyze and validate the relationship between Berden classification and renal outcome ESRD in cohort children using the Kaplan-Meier method.

[0024] Figure 11 To analyze and validate the relationship between the pathological classification of glomerular segmental sclerosis and the renal outcome of ESRD in cohort of children using the Kaplan-Meier method.

[0025] Figure 12 To analyze and validate the relationship between the pathological classification of glomerular sclerosis and the renal outcome ESRD in cohort of children using the Kaplan-Meier method.

[0026] Figure 13 To analyze and validate the relationship between glomerular crescent pathological type and renal outcome ESRD in cohort children using the Kaplan-Meier method.

[0027] Figure 14 To analyze and validate the relationship between the pathological subtype of renal tubular necrosis / interstitial inflammatory cell infiltration and the renal outcome ESRD in cohort children using the Kaplan-Meier method.

[0028] Figure 15To analyze and validate the relationship between the pathological classification of renal tubular atrophy / interstitial fibrosis and the occurrence of ESRD in cohort of children using the Kaplan-Meier method. Detailed Implementation

[0029] The following examples re-examine the renal biopsies of 124 AAGN children who underwent renal biopsies at the applicant's institution and its collaborating institutions from January 2012 to December 2022. Of these, 23 cases were lost to follow-up or had missing outcome data. Among the 102 children with ≥7 glomeruli, 22 were lost to follow-up or had missing outcome data; ultimately, 80 cases were included in a retrospective cohort study. Demographic characteristics, clinical characteristics, renal pathological images, and prognoses of these children were retrospectively collected.

[0030] From January 2023 to the present, a validation cohort study was conducted on 12 newly diagnosed patients. The Kaplan-Meier method was used to analyze the relationship between pathology and renal prognosis. Results showed that the Berden classification did not accurately predict ESRD events, but this may be associated with a relatively positive correlation with treatment of the crescent type. Globular sclerosis, segmental sclerosis, crescent type, interstitial inflammatory cell infiltration / tubular necrosis, and tubular atrophy / interstitial fibrosis showed good association with ESRD. However, due to the small sample size, the proportional grading of globular sclerosis and its prognosis still require validation with a larger sample size.

[0031] The following experiments have been approved by the hospital's ethics committee.

[0032] Example 1: Establishing a predictive model for renal histopathological assessment of ANCA-associated nephritis in children.

[0033] The specific steps are as follows:

[0034] First, a univariate analysis was performed (Table 1). Based on the current consensus on pathological classification, pathological changes were graded and scored according to the proportion of affected individuals. The proportion grading was optimized through Cox survival curve analysis. The relevant pathological grading indicators were statistically optimized (univariate regression analysis) based on the proportion of pathological damage in order to screen the optimal grading proportion.

[0035] Table 1. Univariate analysis of the correlation between renal biopsy pathology and ESRD

[0036]

[0037] Further univariate analysis with p-values ​​<0.1 and relevant indicators selected based on clinical experience were used to conduct a multivariate Cox regression model study. Finally, five indicators were established to form the ANCA renal histopathology GSCIT classification for children (Table 2): glomerular sclerosis, segmental sclerosis, crescent, renal interstitial inflammatory cell infiltration / tubular necrosis, and renal tubular atrophy / renal interstitial fibrosis.

[0038] Table 2. Preliminary establishment of GSCIT classification for ANCA renal histopathology in children.

[0039]

[0040] Correlation analysis of renal biopsy pathological lesions with ESRD, optimization of pre- and post-treatment proportional grading, and optimization of relevant pathological grading indicators are as follows: Figures 2-8 As shown.

[0041] From January 2023 to the present, a total of 12 newly diagnosed patients were included in the validation cohort study. Basic clinical information is shown in Table 3, and the statistical results of the basic pathological subtype distribution are shown in [Table 3]. Figure 9 The modeling process is as follows Figure 1 As shown.

[0042] Table 3 Basic clinical information of the validation cohort

[0043]

[0044] The relationship between pathology and renal outcome prognosis was analyzed using the Kaplan-Meier method ((1). Berden AE, Ferrario F, Hagen EC, et al. Histopathologic classification of ANCA-associated glomerulonephritis. J AmSoc Nephrol. Oct 2010; 21(10):1628-36.). The results showed that the Berden classification could not predict ESRD events well, possibly due to a relatively positive correlation with the treatment of crescentic glomerulonephritis. Globular sclerosis, segmental sclerosis, crescentic glomerulonephritis, renal interstitial inflammatory cell infiltration / tubular necrosis, and renal tubular atrophy / interstitial fibrosis were well associated with ESRD, but due to the small sample size, the proportional grading of globular sclerosis and its prognosis still need to be verified with a larger sample size. Figure 10-15 To validate the relationship between pathological classification and renal outcome ESRD in cohort children using the Kaplan-Meier method.

Claims

1. A method for establishing a renal histopathological assessment and prediction model for ANCA-associated nephritis in children, characterized in that, The specific steps are as follows: 1) A GSCIT classification model for pediatric ANCA renal histopathology was established using five indicators: glomerular sclerosis, segmental sclerosis, crescent, interstitial inflammatory cell infiltration / tubular necrosis, and tubular atrophy / interstitial fibrosis.

Images