Repeat renal biopsy improves the Oxford classification-based prediction of immunoglobulin A nephropathy outcome

Interpretable Machine Learning Model Based on Superb Microvascular Imaging for Non-Invasive Determination of Crescent Status of IgAN

Purpose

To assess the crescentic status of IgA nephropathy (IgAN) non-invasively using a superb microvascular imaging (SMI)-based radiomics machine learning (ML) model.

Patients and Methods

IgAN patients who underwent renal biopsy from June 2022 to October 2023, with two-dimensional ultrasound (US) and SMI examinations conducted one day prior to the renal biopsy. The patients selected were divided randomly into a training group and a test group in a 7:3 ratio. Radiomic features were extracted from US and SMI images, then radiomic features were constructed and ML models were further established using logistic regression (LR) and extreme gradient boosting (XGBoost)XGBoost to determine the crescentic status. The utility of the proposed model was evaluated using receiver operating characteristics, calibration, and decision curve analysis. The SHapley Additive exPlanations (SHAP) was utilized to explain the best-performing ML model.

Results

A total of 147 IgAN patients were included in the study, with 103 in the training group and 44 in the test group .Among them, the US-SMI based XGBoost model achieved the best results, with an the area under the curve (AUC) of 0.839 (95% CI,0.756-0.910) and an accuracy of 78.6% in the training group.In the test group, the AUC was 0.859 (95% CI,0.721-0.964), and the accuracy was 81.8%, significantly surpassing the ML model of a single modality and the clinical model established based on occult blood. Additionally, the decision curve analysis (DCA) demonstrated that the XGBoost model provided a higher overall net benefit in the both groups.

Conclusion

The SMI radiomics ML model has the capability to accurately predict the crescentic status of IgAN patients, providing effective assistance for clinical treatment decisions.

Development and External Validation of a Multidimensional Deep Learning Model to Dynamically Predict Kidney Outcomes in IgA Nephropathy

Key Points

A dynamic model predicts IgA nephropathy prognosis based on deep learning. Longitudinal clinical data and deep learning improve predictive accuracy and interpretability in GN.

Background

Accurately predicting kidney outcomes in IgA nephropathy is crucial for clinical decision making. Insufficient use of longitudinal data in previous studies has limited the accuracy and interpretability of prediction models for failing to reflect the chronic nature of IgA nephropathy. The aim of this study was to establish a multivariable dynamic deep learning model using comprehensive longitudinal data for the prediction of kidney outcomes in IgA nephropathy.

Methods

In this retrospective cohort study of 2056 patients with IgA nephropathy from 18 kidney centers, a total of 28,317 data points were collected by the sliding window method. Among them, 15,462 windows in a single center were randomly assigned to training (80%) and validation (20%) sets and 8797 windows in 18 kidney centers were assigned to an independent test set. Interpretable multivariable long short-term memory, a deep learning model, was implemented to predict kidney outcomes (kidney failure or 50% decline in kidney function) based on time-invariant variables measured at biopsy and time-variant variables measured during follow-up. Risk performance was evaluated using the Kaplan–Meier analysis and C-statistic. Trajectory analysis was performed to assess the various trends of clinical variables during follow-up.

Results

The model achieved a higher C-statistic (0.93; 95% confidence interval, 0.92 to 0.95) on the test set than the machine learning prediction model that we developed in a previous study using only baseline information (C-statistic, 0.84; 95% confidence interval, 0.80 to 0.88). The Kaplan–Meier analysis showed that groups with lower predicted risks from the full model survived longer than groups with higher risks. Time-variant variables demonstrated higher importance scores than time-invariant variables. Within time-variant variables, more recent measurements showed higher importance scores. Further interpretation showed that certain trajectory groups of time-variant variables such as serum creatinine and urine protein were associated with elevated risks of adverse outcomes.

Conclusions

In IgA nephropathy, a deep learning model can be used to accurately and dynamically predict kidney prognosis based on longitudinal data, and time-variant variables show strong ability to predict kidney outcomes.

A lack of KDIGO guidelines for adolescents and young adults with IgA nephropathy

BACKGROUND

IgA nephropathy (IgAN) is one of the most prevalent primary glomerular diseases in children and adolescents. In 2021, The Kidney Disease: Improving Global Outcomes (KDIGO) released Clinical Practice Guidelines for the Management of Glomerular Diseases as an update to the 2012 guidelines. However, the lack of available evidence for the treatment of IgAN in children has led to an absence of treatment recommendations.

CLINICAL CASE

We present the case of a 19-year-old male with IgAN who was lost to follow-up after an appointment at a children's hospital 3 years prior. He presents for care at an adult hospital after running out of his medications for many months. He is found to have an elevated blood pressure of 152/97 and an elevated creatinine at 0.8 mg/dL.

DISCUSSION

There is not only a need for treatment guidelines for IgAN in pediatric patients, but also a need for guidelines for adolescent patients with IgAN as they transition from pediatric to adult care. Therefore, we review the KDIGO treatment guidelines for adults with IgAN and the treatment evidence for children with IgAN and discuss the management dilemma that exists for adolescents and young adults (AYA) with IgAN. Specifically, we propose renin-angiotensin-aldosterone blockade (RASB) treatment, irrespective of blood pressure, for AYA with proteinuria >0.5 g/day. We also propose treatment with corticosteroids for patients with proteinuria >1 g/day and/or mesangial hypercellularity.

CONCLUSION

The formation of treatment guidelines for patients transitioning from pediatric to adult nephrology care is paramount.

A novel clinical-radiomic nomogram for the crescent status in IgA nephropathy

OBJECTIVE

We used machine-learning (ML) models based on ultrasound radiomics to construct a nomogram for noninvasive evaluation of the crescent status in immunoglobulin A (IgA) nephropathy.

METHODS

Patients with IgA nephropathy diagnosed by renal biopsy (n=567) were divided into training (n=398) and test cohorts (n=169). Ultrasound radiomic features were extracted from ultrasound images. After selecting the most significant features using univariate analysis and the least absolute shrinkage and selection operator algorithm, three ML algorithms were assessed for final radiomic model establishment. Next, clinical, ultrasound radiomic, and combined clinical-radiomic models were compared for their ability to detect IgA crescents. The diagnostic performance of the three models was evaluated using receiver operating characteristic curve analysis.

RESULTS

The average area under the curve (AUC) of the three ML radiomic models was 0.762. The logistic regression model performed best, with AUC values in the training and test cohorts of 0.838 and 0.81, respectively. Among the final models, the combined model based on clinical characteristics and the Rad score showed good discrimination, with AUC values in the training and test cohorts of 0.883 and 0.862, respectively. The decision curve analysis verified the clinical practicability of the combined nomogram.

CONCLUSION

ML classifier based on ultrasound radiomics has a potential value for noninvasive diagnosis of IgA nephropathy with or without crescents. The nomogram constructed by combining ultrasound radiomic and clinical features can provide clinicians with more comprehensive and personalized image information, which is of great significance for selecting treatment strategies.

Comparison between outcomes of IgA nephropathy with nephrotic-range proteinuria and nephrotic syndrome: do podocytes play a role?

Background

Nephrotic syndrome (NS) and nephrotic-range proteinuria (NRP) are uncommon in IgA nephropathy (IgAN), and their clinicopathology and prognosis have not been discussed. Podocytes may play an important role in both clinical phenotypes.

Methods

We investigated 119 biopsy-proven IgAN patients with proteinuria over 2 g/d. The patients were divided into three groups according to proteinuria level: the overt proteinuria (OP) group, NS group, and NRP group. In addition, according to the severity of foot process effacement (FPE), the patients were divided into three groups: the segmental FPE (SFPE) group, moderate FPE (MFPE) group, and diffuse FPE (DFPE) group. The outcome was survival from a combined event defined by a doubling of the baseline serum creatinine and a 50% reduction in eGFR or ESRD.

Results

Compared with the NRP group, patients in the NS group had more severe microscopic hematuria, presented with more severe endocapillary hypercellularity and had a higher percentage of DFPE. The Kaplan–Meier curve showed that MFPE patients had a better outcome in the NRP group <50% of tubular atrophy/interstitial fibrosis. In the multivariate model, the NRP group (HR = 17.098, 95% CI = 3.835–76.224) was associated with an increased risk of the combined event, while MFPE (HR = 0.260, 95% CI = 0.078–0.864; p = 0.028) was associated with a reduced risk of the combined event. After the addition of renin-angiotensin system inhibitors (RASi), the incidence of the combined event in the MFPE group (HR = 0.179, 95% CI = 0.047–0.689; p = 0.012) was further reduced.

Conclusions

NS presented more active lesions and more severe FPE in IgAN. NRP was an independent risk factor for progression to the renal endpoint, while MFPE indicated a better prognosis in NRP without obvious chronic renal lesions, which may benefit from RASi.

Validation of two IgA nephropathy risk-prediction tools using a cohort with a long follow-up

BACKGROUND

Recently, two immunoglobulin A nephropathy prediction tools were developed that combine clinical and histopathological parameters. The International IgAN Prediction Tool predicts the risk for 50% declines in the estimated glomerular filtration rate or end-stage renal disease up to 80 months after diagnosis. The IgA Nephropathy Clinical Decision Support System uses artificial neural networks to estimate the risk for end-stage renal disease. We aimed to externally validate both prediction tools using a Norwegian cohort with a long-term follow-up.

METHODS

We included 306 patients with biopsy-proven primary immunoglobulin A nephropathy in this study. Histopathologic samples were retrieved from the Norwegian Kidney Biopsy Registry and reclassified according to the Oxford classification. We used discrimination and calibration as principles for externally validating the prognostic models.

RESULTS

The median patient follow-up was 17.1 years. A cumulative dynamic time-dependent receiver operating characteristic analysis showed area under the curve values of ranging from 0.90 at 5 years to 0.83 at 20 years for the International IgAN Prediction Tool, while time-naive analysis showed an area under the curve value at 0.83 for the IgA Nephropathy Clinical Decision Support System. The International IgAN Prediction Tool was well calibrated, while the IgA Nephropathy Clinical Decision Support System tends to underestimate risk for patients with higher risk, and overestimates risk in the lower risk categories.

CONCLUSIONS

We have externally validated two prediction tools for IgA nephropathy. The International IgAN Prediction Tool performed well, while the IgA Nephropathy Clinical Decision Support System has some limitations.

Crescents and IgA Nephropathy: A Delicate Marriage

IgA nephropathy (IgAN) is a progressive disease with great variability in the clinical course. Among the clinical and pathologic features contributing to variable outcomes, the presence of crescents has attracted particular interest as a distinct pathological feature associated with severity. Several uncontrolled observations have led to the general thought that the presence and extent of crescents was a prognostic indicator associated with poor outcomes. However, KDIGO 2021 guidelines concluded that either the presence or the relative number of crescents should not be used to determine the progression of IgAN nor should they suggest the choice of immunosuppression. Our aim is to report and discuss recent data on the debated issue of the value of active (cellular and fibrocellular) crescents in the pathogenesis and clinical progression of IgAN, their predictive value, and the impact of immunosuppression on renal function. We conclude that the value of crescents should not be disregarded, although this feature does not have an independent predictive value for progression in IgAN, particularly when considering immunosuppressed patients. An integrated overall evaluation of crescents with other active MEST scores, clinical data, and novel biomarkers must be considered in achieving a personalized therapeutic approach to IgAN patients.

Utilizing the MEST score for prognostic staging in IgA nephropathy

Background

The Oxford classification/MEST score is an established histopathologic scoring system for patients with IgA nephropathy (IgAN). The objective of this study was to derive a prognostic model for IgAN based on the MEST score and histopathologic features.

Methods

A total of 306 patients with biopsy-proven primary IgAN were included. Histopathologic samples were retrieved from the Norwegian Kidney Biopsy Registry and reclassified according to the Oxford classification. The study endpoint was end-stage renal disease (ESRD). Patients were subclassified into three risk models based on histologic features (Model A), a composite score calculated from the adjusted hazard ratio values (Model B), and on quartiles (Model C).

Results

The mean follow-up time was 16.5 years (range 0.2–28.1). In total, 61 (20%) patients reached ESRD during the study period. Univariate analysis of M, E, S, T and C lesions demonstrated that all types were associated with an increased risk of ESRD; however, a multivariate analysis revealed that only S, T and C lesions were associated with poor outcomes. Statistical analysis of 15-year data demonstrated that Models A and B were as predictive as the MEST score, with an area-under-the-curve at 0.85. The Harrel c index values were 0.81 and 0.80 for the MEST score and Models A and B, respectively. In the present cohort, adding C lesions to the MEST score did not improve the models prognostic value.

Conclusions

Patients can be divided into risk classes based on their MEST scores. Histopathologic data provide valuable prognostic information at the time of diagnosis. Model B was the most suitable for clinical practice because it was the most user-friendly.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12882-021-02653-y.

Factors associated with the progression of mesangial lesions in IgA nephropathy: A comparative analysis of renal re-biopsies

OBJECTIVES

IgA nephropathy (IgAN) is the most common primary glomerular disease, and is the leading cause of chronic renal failure. Because mesangial lesions are the main pathological changes seen in IgAN, we investigated factors associated with the progression of mesangial lesions in IgAN.

METHODS

We enrolled participants with IgAN who underwent repeat renal biopsies. Based on the progression of mesangial proliferative lesions, the participants were divided into progressive and stable groups. The progression group included participants with a ratio of mesangial cell proliferation score ≥ 1.1 (i.e., proliferation of > 10%) in the second biopsy specimen compared to the first biopsy specimen. The stable group included participants who did not fulfill the aforementioned criteria. We recorded the laboratory parameters, expression of renin-angiotensin system (RAS) receptors (angiotensin II type 1 receptor [AT1R], angiotensin II type 2 receptor [AT2R], Mas receptor [MasR], and the Mas-related G protein-coupled receptor, member D [MrgD]) and mesangial matrix proteins (collagen [Col] IV, fibronectin [FN] and laminin) at the first and second renal biopsies, and the use of immunosuppressive therapy and/or RAS blockers after the first biopsy.

RESULTS

We enrolled 24 patients with IgAN who underwent repeat renal biopsies. Half of patients showed progression of mesangial lesions on repeat renal biopsy after a median of 4.3 (1-6) years. The progression group had significantly higher expression levels of AT1R and mesangial matrix proteins (Col IV and FN), and significantly lower expression of AT2R and MasR, compared to the stable group. Multivariate analysis showed that the use of RAS blockers (hazard ratio [HR], 0.27; 95% CI, 0.08-0.97; p < 0.05) and the level of proteinuria (HR, 1.8; 95% CI, 1.04-3.12; p < 0.05) were associated with progression of mesangial lesions. Additionally, the progression group exhibited a more rapid decline of renal function compared to the stable group (0.38 and 0.012 ml/min/1.73 m²/month, respectively; p = 0.004).

CONCLUSIONS

Continuous activation of the intrarenal RAS and massive proteinuria correlate with histological progression of mesangial lesions in IgAN patients, which may further accelerate the deterioration of renal function.

A mixed-method evaluation of the relationship between Oxford classification scores and longitudinal changes in proteinuria in patients with immunoglobulin A nephropathy

INTRODUCTION

This study aimed to investigate the relationship between Oxford Classification scores and longitudinal changes in proteinuria in patients with immunoglobulin A nephropathy (IgAN).

METHODS

The study was a single-center retrospective cohort study involving 358 patients with primary IgAN who were treated at the Shenzhen Second People's Hospital, China, between January 2011 and May 2021. Multivariate linear regression and generalized additive mixed models (GAMMs), adjusted for traditional risk confounders, were used to evaluate the correlation between scores for mesangial hypercellularity (M), endocapillary hypercellularity (E), segmental glomerulosclerosis (S), tubular atrophy/interstitial fibrosis (T), and crescents (C) (known as the Oxford Classification MEST-C score system), with proteinuria/creatinine ratio (PCR) at the time of renal biopsy and longitudinal changes in PCR, respectively.

RESULTS

The median PCR was 1061 mg/g, and it increased on average by 68.82 mg/g per year in these patients. Among patients with renal insufficiency, compared with patients without relative lesions, those with E present (E1) (1153.44; 95% confidence interval [CI], 188.99-2117.89 mg/g) and C > 0 (C1/2) (1063.58; 95% CI, 185.25-1941.90 mg/g) were associated with increased PCR levels at the time of renal biopsy. What's more, S present (S1) (194.96; 95% CI, 54.50-335.43 mg/g per year) was associated with the fastest PCR increase; C > 0 (C1/2) (147.59; 95% CI, 8.32-286.86 mg/g per year) and T >25% (T1/2) (77.04; 95% CI, 7.18-146.89 mg/g per year), were also correlated with a faster PCR increase. In patients with normal kidney function, associations between S1 (55.46; 95% CI, 8.93-101.99 mg/g per year) and E1 (94.02; 95% CI, 21.47-166.58 mg/g per year) and PCR change could be observed. Additionally, in patients with overweight/obesity, S1 (156.09; 95% CI, 52.41-259.77 mg/g per year), E1 (143.34; 95% CI, 35.30-251.38 mg/g per year), T1/2 (116.04; 95% CI, 22.58-209.51 mg/g per year), as well as C1/2 (134.03; 95% CI, 41.73-226.32 mg/g per year) were associated with noticeably quicker PCR increase.

CONCLUSIONS

Overall, E1 and C1/2 were independently associated with raised proteinuria levels at the time of renal biopsy, and S1, E1, T1/2, C1/2 were independently associated with a longitudinal increase in proteinuria in the patients with IgAN, especially in those with renal insufficiency or overweight/obesity, suggesting that currently available treatments might not be satisfactory, and weight control might be beneficial. Individual therapy development might benefit from the use of the Oxford Classification system.

The Association of the Oxford Classification Score with Longitudinal Estimated Glomerular Filtration Rate Decline in Patients with Immunoglobulin A Nephropathy: A Mixed-Method Study

Introduction

The Oxford Classification score, which predicts renal outcomes for immunoglobulin A nephropathy (IgAN), is widely used in clinical practice. Nevertheless, the relationship between these markers and longitudinal changes in renal function are poorly understood.

Methods

This was a population-based retrospective cohort study of 280 adults with biopsy-proven primary IgAN from 2011 to 2018. We used generalized additive mixed models to control for traditional kidney disease risk factors to analyze the associations between Oxford Classification MEST-C scores (mesangial hypercellularity, M; endocapillary hypercellularity, E; segmental glomerulosclerosis, S; tubular atrophy/interstitial fibrosis, T; crescents, C) and longitudinal changes in the estimated glomerular filtration rate (eGFR) after renal biopsy.

Results

The median eGFR was 78.2 mL/min/1.73 m² at baseline, and then it decreased on average by 1.3 mL/min/1.73 m² per year in the entire cohort. In adjusted models, compared with patients without relative lesions, the presence of T > 50% (T2) (−5.7; 95% confidence interval [CI], −9.5 to −2.0 mL/min/1.73m² per year) was associated with the fastest eGFR decline. S present (S1) (−2.9; 95% CI, −4.6 to −1.1 mL/min/1.73m² per year) and C > 25% glomeruli (C2) (−3.4; 95% CI, −6.4 to −0.5 mL/min/1.73m² per year) also demonstrated steeper eGFR declines. However, we found no association between M > 0.5 (M1), E present (E1), T 26%–50% (T1), and C present ≥ 1 glomerulus (C1), and progressive eGFR decline (p > 0.05).

Conclusion

The Oxford Classification scores, S1, T2, and C2, were independently associated with the longitudinal decreases in renal function in patients with IgAN. These findings suggested therapies targeted at improving early damage to these lesions might be essential to delay renal progression.

Clinical and histological differences between adults and children in new onset IgA nephropathy

BACKGROUND

Previous reports suggest initial presentation of IgA nephropathy (IgAN) in children is different from adults. No systematic comparison of clinical, biological, and histological childhood- and adult-onset IgAN is currently available.

METHODS

We compared pediatric and adult clinical and histological characteristics at IgAN diagnosis. Data on 211 consecutive patients from two different centers in Paris (82 children, 129 adults) were reviewed. Kidney biopsies were scored for Oxford classification and podocytopathic (P1) features.

RESULTS

We report higher eGFR at diagnosis in children compared to adults (89.5 vs. 64 ml/min/1.73 m²; p = 0.0001) but no difference in proteinuria. Histological analysis of kidney biopsy found higher proportions of mesangial (M1) and endocapillary (E1) hypercellularity in children compared with adults (M1 [80.7% vs. 27.9%, p = 0.0001]; E1 [71.3% vs. 30%, p = 0.0001]). Focal glomerulosclerosis (S1), tubular atrophy/interstitial fibrosis ≥ 25% (T1), and P1 were more frequent in adults (S1 [81.5% vs. 61.3%, p = 0.0012], T1 [49.5% vs. 1.35%, p = 0.0001], P1 [33.8% vs. 16.4%, p = 0.008). Proteinuria associated with M1, E1, and C1 in children (M1, p = 0.0001; E1, p = 0.0005; C1, p = 0.0014) but S1, P1, and T1 in adults (S1, p = 0.0001; P1, p = 0.0001; T1, p = 0.001). After steroid treatment (41 children and 28 adults), proteinuria decreased in children (p < 0.001, follow-up 38 months) and adults (p < 0.001, follow-up 76.9 months), whereas eGFR remained stable in adults but increased significantly in children (90.6 to 110 ml/min/1.73m²).

CONCLUSION

Proteinuria in children with IgAN is a marker of glomerular proliferative lesions whereas its presence in adults often reflects the presence of chronic lesions. This suggests the need for histological assessment.

Has The Time Arrived to Refine The Indications of Immunosuppressive Therapy and Prognosis in IgA Nephropathy?

Immunoglobulin A nephropathy (IgAN) is the most frequent glomerular disease worldwide and a leading cause of end-stage renal disease. Particularly challenging to the clinician is the early identification of patients at high risk of progression, an estimation of the decline in renal function, and the selection of only those that would benefit from additional immunosuppressive therapies. Nevertheless, the pathway to a better prognostication and to the development of targeted therapies in IgAN has been paved by recent understanding of the genetic and molecular basis of this disease. Merging the data from the Oxford Classification validation studies and prospective treatment studies has suggested that a disease-stratifying algorithm would be appropriate for disease management, although it awaits validation in a prospective setting. The emergence of potential noninvasive biomarkers may assist traditional markers (proteinuria, hematuria) in monitoring disease activity and treatment response. The recent landmark trials of IgAN treatment (STOP-IgAN and TESTING trials) have suggested that the risks associated with immunosuppressive therapy outweigh the benefits, which may shift the treatment paradigm of this disease. While awaiting the approval of the first therapies for IgAN, more targeted and less toxic immunotherapies are warranted. Accordingly, the targeting of complement activation, the modulation of mucosal immunity, the antagonism of B-cell activating factors, and proteasomal inhibition are currently being evaluated in pilot studies for IgAN treatment.