Clinical Outcomes in the Nephrotic Syndrome Study Network: Disease Burden and Treatment Patterns over Time Open Access

Primary glomerular diseases, including minimal change disease (MCD), focal segmental glomerulosclerosis (FSGS), and membranous nephropathy (MN), are individually rare but together account for up to 15% of prevalent kidney failure in the USA and are associated with significant morbidity and mortality [1‒3]. The terms MCD, FSGS, and MN are histological descriptors for diseases that have heterogeneous etiopathogeneses, clinical presentations, treatment responses, and long-term outcomes [4, 5]. Knowledge of clinical outcomes in these diseases is largely based on a few registry-based or single-center studies conducted at tertiary care referral centers [4, 6, 7]. Due to a paucity of randomized controlled trials, clinical practice guidelines are generally based on medium- to low-quality evidence and there is substantial variation in practice patterns among nephrologists [8].

There is a critical need to better characterize the clinical presentations, treatment practice patterns, and outcomes of these glomerular diseases in a diverse, well-phenotyped patient population. Addressing this knowledge gap can help clinicians and the research community better leverage prognostic information, identify opportunities for standardization of care, and facilitate optimal clinical trial design to ultimately improve patient outcomes. The Nephrotic Syndrome Study Network (NEPTUNE) is a multicenter, prospective observational cohort study of patients with proteinuric glomerular diseases [9]. The NEPTUNE study has enrolled more than 600 patients from 23 clinical centers in the USA and Canada. Here, we describe the clinical presentation, treatment patterns, and outcomes of patients with biopsy-proven MCD, FSGS, and MN enrolled in the NEPTUNE study.

The NEPTUNE study (NCT01209000) is a multicenter, prospective, observational cohort study of adult and pediatric patients who undergo unblinded standardized evaluation of prespecified study outcomes in 2 parallel cohorts. Cohort A includes adults and children with incident, biopsy-proven FSGS, MCD, or MN who are recruited at the time of their first clinically indicated kidney biopsy. Cohort B includes children aged <19 years with treatment-naïve incident nephrotic syndrome (defined as urinalysis with >2+ protein and edema OR urinalysis with >2+ protein and serum albumin <3 g/dL OR random urine protein-to-creatinine ratio [UPCR] >2 g/g and serum albumin <3 g/dL). Participants are enrolled at clinical centers in the USA and Canada at the time of their first clinically indicated kidney biopsy (cohort A) or at disease onset (cohort B). Enrollment began on December 11, 2009, and is ongoing. Following Institutional Review Board approval, informed consent/assent was obtained from all participants enrolled in NEPTUNE. Participants undergo an initial research visit at the time of biopsy (cohort A) or at the time of confirming eligibility (cohort B), followed by a subsequent baseline visit within 45 days. Thereafter, follow-up visits occur every 4 months in the first year and every 6 months in year 2 and beyond. Participants in cohort B may undergo a biopsy due to a clinical indication later in their follow-up period. For the current analysis, we included patients with histopathological confirmation of one of the three relevant diagnoses (FSGS, MCD, or MN), including patients from cohort A as well as those cohort B participants who underwent a kidney biopsy after enrollment.

Procedures performed during study visits include laboratory studies, blood and urine collection for inclusion in the biorepository, clinical examination, and administration of questionnaires. Demographic and clinical characteristics are described at biopsy and study baseline. Race and ethnicity are self-reported or reported by parents of children. Detailed light and electron microscopic histopathological characteristics obtained by study pathologists are described using the NEPTUNE Digital Pathology Scoring System [10, 11]. The frequency of APOL1 risk alleles is described for Black participants, and the frequency of antibodies to the phospholipase A2 receptor (PLA2R) as measured by enzyme-linked immunosorbent assay (ELISA; positive defined as >14 U/mL) is described for participants with MN. Additionally, we describe the characteristics of the FSGS cohort according to the presence or absence of full nephrotic syndrome (UPCR >3.5 g/g and serum albumin <3.0 g/dL with diffuse foot process effacement >75% on kidney biopsy), including the frequency of APOL1 risk alleles, obesity, and potential viral causes.

Immunosuppressive therapy (IST) was categorized into the following types: corticosteroid monotherapy, calcineurin inhibitors (CNIs), mycophenolate mofetil/sodium (MPA), cyclophosphamide (CYC), rituximab, and “other” IST. Any combination of two or more non-corticosteroid IST started within 4 weeks of each other (e.g., CNI and MPA) was defined as “multiple” IST. The concurrent use of corticosteroids and non-corticosteroid IST was categorized under the specific non-corticosteroid IST group. For example, the concurrent use of a CNI and corticosteroids in an FSGS patient was categorized as CNI-based therapy if the CNI and corticosteroids were started within 4 weeks of each other. The concurrent use of CYC and corticosteroids in an MN patient was categorized as CYC-based therapy if the CYC and corticosteroids were started within 3 months of each other to reflect the use of the modified Ponticelli regimen [12, 13]. Corticosteroid monotherapy was defined as the use of corticosteroids for ≥4 weeks without other IST use during that period; corticosteroid use for <4 weeks was not included in the analysis. For each type of IST, the percentage of patients receiving this therapy was calculated. For patients receiving “multiple” IST, the start date for the initial therapy was used. Among those receiving each type of IST, the percentage of study follow-up time on therapy was calculated based on medication start and stop dates.

We also separately analyzed treatment patterns for participants with versus without IST use in the 6 months prior to biopsy. Among those with IST use prior to biopsy, their IST use within the first month after biopsy is shown descriptively to understand post-biopsy treatment patterns. Among those without IST use prior to biopsy, i.e., “treatment-naïve” patients, we conducted competing-risk analyses to model the time from biopsy to first IST initiation, showing the cumulative incidence plot of each type of IST over time while accounting for other types of IST use. A similar analysis was done to examine the time from first IST to the initiation of second-line therapy.

Clinical outcomes included (1) time from biopsy to any instance of complete remission (CR), defined as UPCR ≤0.3 g/g or negative (or trace) urine protein on dipstick if UPCR was not available, (2) time from biopsy to the composite disease progression outcome defined as kidney failure (initiation of dialysis or receipt of a kidney transplant) or 40% decline in eGFR (based on 2 or more consecutive eGFR measurements) [14] to an eGFR <90 mL/min/1.73 m², (3) time from biopsy to an alternate composite disease progression outcome defined as kidney failure or 40% decline in eGFR to an eGFR <60 mL/min/1.73 m², and (4) time from biopsy to kidney failure, defined as two consecutive eGFR measurements <15 mL/min/1.73 m² or initiation of dialysis or receipt of a kidney transplant. eGFR was calculated using the U25 formula [15] including age and sex for participants aged 0–25 years and the CKD-EPI formula without the race coefficient [16] for participants aged ≥25 years. Outcomes were described as rates per 100 person-years (p-y) with 95% confidence intervals estimated using Poisson regression models. Cumulative incidence and Kaplan-Meier curves were plotted for first post-biopsy CR and the composite disease progression outcome by disease diagnosis, respectively. Subsequent analyses were performed to describe proteinuria severity by IST status (on/off IST) at biopsy and at 4, 12, 24, and 36 months of follow-up. A Sankey diagram was used to describe patient trajectories according to combined remission status (CR vs. no CR) and therapy status (on vs. off IST) at biopsy and at 12 and 24 months. All analyses were conducted using SAS software, version 9.4 (SAS Institute, Cary, NC, USA) except for the Sankey diagram, which was done using R software, version 4.2.0 (R Development Core Team, Vienna).

A total of 507 NEPTUNE participants had a kidney biopsy and were followed for a median of 3.2 years (interquartile range [IQR] 1.6, 4.5 years). Of these, 203 were classified as having FSGS, 193 as having MCD, and 111 as having MN (Table 1). Among participants aged <18 years, 61% had MCD, 38% had FSGS, and 1% had MN. Among adults, 41.5% had FSGS, 34% had MN, and 24% had MCD. Patients with MN were more likely to be male and White compared to patients with FSGS or MCD, whereas patients with FSGS were more likely to be Black. Sex and ethnicity were similar across subgroups.

At the time of biopsy, patients with FSGS had the lowest mean eGFR (71.2 ± 34.0), followed by MN (83.2 ± 30.0) and MCD (98.4 ± 42.7 mL/min/1.73 m², Table 1). Forty-four percent of FSGS patients had an eGFR ≤60, compared with 19% of MN and 14% of MCD patients. Twenty percent of patients with MCD had an eGFR >120, compared with 9% of MN and 10% of FSGS patients. Among patients with MCD, 28% had UPCR ≤1.0 g/g at the time of biopsy (Table 1; Fig. 1); 58% of those had received IST prior to biopsy. Similarly, among n = 85 MCD patients with complete data, 42% of those who did not have the nephrotic syndrome (UPCR >3.5 g/g and serum albumin <3.0 g/dL with diffuse foot process effacement >75%) had received IST prior to biopsy. Nearly half of all patients with MCD (44%) were treated with corticosteroids prior to biopsy, in contrast to 5% of MN and 19% of FSGS patients. The majority of patients with MN (73%) had proteinuria in the nephrotic range (Table 1; Fig. 1). Serum albumin was lowest among patients with MN. Among Black patients (n = 129), the frequency of monoallelic APOL1 risk variants was 37% overall and 28% among those with FSGS; the frequency of biallelic APOL1 risk variants was 38% overall and 53% among those with FSGS. Among MN patients with anti-PLA2R antibody data available at baseline (n = 64), 55% had a positive anti-PLA2R titer. Characteristics of the FSGS cohort according to the presence or absence of full nephrotic syndrome (UPCR >3.5 g/g and serum albumin <3.0 g/dL with diffuse foot process effacement >75% on kidney biopsy) are presented in Table 2.

The degree of interstitial fibrosis and tubular atrophy was lowest among MCD and highest among FSGS patients (Table 1). The prevalence of global glomerulosclerosis abnormal for age was highest among FSGS patients (64%) and lower among MCD and MN patients (31% and 25%, respectively).

Corticosteroid exposure was high overall and highest for MCD; 76% of patients with MCD, 53% of patients with FSGS, and 48% of patients with MN were treated with corticosteroids at some point during the follow-up (Table 3). A CNI was the most frequently used non-corticosteroid IST among patients with MCD and FSGS, while rituximab was the most frequently used IST among patients with MN. Thirty-seven percent of MCD patients received a CNI during follow-up; those that were treated with a CNI remained on it for a median of 57% (IQR: 33–86%) of their follow-up time, and their median time on treatment was 18.9 (IQR: 10.4–32.4) months. Similarly, 31% of FSGS patients received a CNI during follow-up; those that were treated with a CNI remained on it for a median of 51% (IQR: 15–86%) of their follow-up time, and the median time on treatment was 15.4 (IQR: 3.9–29.6) months. Rituximab and CYC were used less frequently, with the exception of MN, where rituximab was used in 29% and a CYC-based regimen was used in 22%.

Next, we evaluated treatment patterns over time. Among MN participants who were not on IST prior to biopsy (n = 102), nearly 60% initiated IST within 1 year after biopsy and only few participants initiated IST more than 1 year after biopsy (Fig. 2a). The median time to IST initiation was 5.6 months. A CNI or rituximab was the most frequently used, followed by CYC, corticosteroid monotherapy, MPA, and “other” IST. Among MCD participants who were not on IST prior to biopsy (n = 95), approximately 60% started IST, consisting predominantly of corticosteroids, within 1 year after biopsy. Among FSGS participants who were not on IST prior to biopsy (n = 151), approximately 40% initiated IST within 1 year after biopsy. Specifically, 64% of participants with UPCR >3.5 g/g initiated IST within 1 year after biopsy, compared to 26% of participants with UPCR ≤3.5 g/g. Similarly, 88% of participants with the nephrotic syndrome (UPCR >3.5 g/g and serum albumin <3.0 g/dL with diffuse foot process effacement >75%) initiated IST, compared to 34% of participants without those features. Corticosteroids were the most commonly used, followed by a CNI (Fig. 2a).

Among patients who received at least one course of IST after biopsy, we evaluated the cumulative incidence of a second IST course by disease category (Fig. 2b). Approximately 70% of MN and MCD participants initiated a second course of IST within 3 years after the start of first IST. The most frequently used second IST among MN participants was rituximab followed by CYC or a CNI. The most frequently used second IST among MCD participants consisted of corticosteroids followed by MPA or a CNI. For FSGS participants, half initiated a second IST within 3 years of their first IST, consisting primarily of a CNI followed by corticosteroids.

IST use changed frequently in the first month after biopsy (online suppl. Table 1; for all online suppl. material, see https://doi.org/10.1159/000546655). Among 133 participants who received IST in the 6 months prior to biopsy, 40 patients (30%) stopped IST before biopsy and 93 patients (70%) remained on IST at the time of biopsy. Of the 40 patients who stopped IST before biopsy, 58% restarted some form of IST within 1 month after biopsy (mainly corticosteroids and/or CNI). Of the 93 patients who remained on IST at the time of biopsy, 57% continued the same IST, 10% stopped IST, and 33% started a new form of IST within 1 month after biopsy (mainly CNI or MPA).

Over time, the proportion of patients with UPCR ≤0.3 g/g increased, and half of all patients were in CR at the 36-month visit (Fig. 3). While the proportion of patients with UPCR 1.5–3.5 and >3.5 g/g decreased over time, 20.2% of patients at the 36-month visit continued to have UPCR >1.5 g/g; 16.6% of patients who were off IST and 26.3% of patients who were on IST at their 36-month visit had UPCR >1.5 g/g.

The rates of kidney failure and 40% eGFR decline were highest among patients with FSGS (3.31 [95% CI, 2.13–5.12] per 100 p-y and 10.17 [95% CI, 7.81–13.25] per 100 p-y, respectively) and lowest among patients with MCD (0.35 [95% CI, 0.09–1.40] per 100 p-y and 2.35 [95% CI, 1.36–4.04] per 100 p-y, respectively) (Table 4; Fig. 4a, b). CR rates were lowest among patients with FSGS (21.03 [95% CI, 17.69–25.01] per 100 p-y) and highest among patients with MCD (46.90 [95% CI, 41.60–52.86] per 100 p-y). The highest probability and shortest time to CR were seen in patients with MCD, followed by MN, with the poorest outcomes among those with FSGS, who experienced ∼50% cumulative probability of first remission at 3 years after biopsy (Fig. 5).

Lastly, we evaluated trends in remission status and IST use at biopsy, 12 months, and 24 months among those with data at all three time points (n = 257, Fig. 6). At biopsy, 91.4% were not in CR: 73% were not in CR and off IST and 18% were not in CR while on IST. At 12 months, 40% had achieved CR, whereas 60% were not in CR (including 33% not in CR off IST and 27% not in CR while on IST). At 24 months, the remission rate was similar to that at 12 months, with 56% not in CR (including 35% not in CR off IST and 21% not in CR while on IST).

Utilizing a multicenter, prospective observational cohort study of over 500 racially and ethnically diverse children and adults with primary glomerular diseases, we describe the clinical presentation, treatment patterns, and outcomes reflective of clinical care over the past decade among individuals with biopsy-proven MCD, FSGS, and MN in the NEPTUNE study.

Several observational studies in contemporary cohorts have found FSGS to be the most common subtype of nephrotic syndrome overall and across all racial groups [7, 17, 18]. MCD is considered the most common cause of primary nephrotic syndrome in children and the third most common cause in adults [7, 18, 19]. However, existing studies are often limited to a single tertiary care center or health system, and few studies have examined the demographic characteristics of a diverse population of patients with primary glomerular diseases in North America. While we did find FSGS to be the most common disease category overall and among Black participants (with more than half carrying two APOL1 risk alleles and more than 80% carrying one or two APOL1 risk alleles), the distribution of disease subtypes was more balanced among White patients, in whom FSGS and MCD were seen at a similar frequency. Extending findings from prior cohorts limited to a specific geographic region and a prior NEPTUNE analysis [7, 17, 18, 20], we found that half of all FSGS patients were White, possibly reflecting the burden of comorbidities such as obesity leading to secondary FSGS lesions (45% of all FSGS patients and 49% of those without the full nephrotic syndrome had obesity) and accounting for the lower-than-expected rates of IST use in this population.

We identified substantial heterogeneity in the degree of proteinuria at the time of biopsy, a finding that may reflect several different factors, including the use of corticosteroid therapy prior to biopsy among patients suspected of having MCD or FSGS and/or the presence of secondary FSGS lesions in patients with risk factors such as obesity. In particular, among patients with MCD, nearly 60% of those patients who had less than 1 g of proteinuria at the time of biopsy, and greater than 40% of patients without the full nephrotic syndrome, had received prior corticosteroid therapy. Among patients with FSGS, despite lower rates of corticosteroid use prior to biopsy, 49% did not have nephrotic-range proteinuria and 31% had ≤50% foot process effacement, supporting that many of these cases were secondary or unknown cause FSGS. In contrast, the MN cohort demonstrated the highest degree of proteinuria, the highest proportion of patients with nephrotic-range proteinuria, and the lowest serum albumin. Our findings are consistent with existing data from natural history studies that one-quarter of patients with MN present with subnephrotic proteinuria [6, 21].

Remission of proteinuria is an important therapeutic target and a surrogate marker of long-term risk of progression to kidney failure [6, 22‒25]. The severity and persistence of proteinuria are strongly correlated with long-term kidney survival, particularly among patients with MN and FSGS. While CR is associated with the best long-term kidney survival and the lowest risk of relapse, a novel definition of partial remission (>40% proteinuria reduction to <1.5 g/g) also strongly correlates with long-term outcomes [22‒24, 26, 27]. In our analysis, we found that a substantial proportion of patients had persistent proteinuria during the long-term follow-up. At 1 year and 3 years, one-third and one-fifth of all patients in the cohort had proteinuria >1.5 g/g, respectively. Notably, only half of patients were in CR at 3 years, and the majority of patients on IST at their 3-year follow-up were not in CR despite being on therapy. Among those who were on IST at 3 years, one-quarter had persistent proteinuria (>1.5 g/g). The presence of persistent proteinuria and inability to achieve remission in more than half of the study cohort, despite ongoing exposure to IST in many cases, underscores the need for better and more precise therapeutic options in patients with primary glomerular diseases.

Therapeutic decision-making remains challenging, in part due to the complexity of glomerular diseases, scarcity of high-quality evidence, and concerns about the relative efficacy and toxicity of existing treatment options, creating a barrier to the implementation of existing guidelines [28]. In this analysis, we found substantial variability in the choice of IST, particularly among patients with MN and among those starting second-line therapy. In the MCD cohort, a CNI or MPA was the most frequently used non-corticosteroid immunosuppressant. The use of a CNI as the most common non-corticosteroid immunosuppressant is consistent with published single-center reports [4, 29]. In contrast, we found higher rates of MPA use and lower rates of CYC use compared to other cohorts that examined adult patients with MCD [4], which may reflect changes in practice patterns over time or the broader age distribution of MCD patients in our cohort. The overall rates of CYC use were low, with the exception of the MN cohort, wherein a CYC-based regimen was used with similar frequency compared to a CNI or rituximab. Among those individuals initiating second-line therapy, marked heterogeneity in treatment choices was evident, again reflecting the paucity of high-quality data to guide the management of patients who do not respond to a first-line agent or relapse after initial remission. Finally, our analysis highlights the burden of IST exposure among patients with glomerular disease, most notably the findings that nearly two-thirds of all patients were exposed to corticosteroids during the follow-up period, and that MCD and FSGS patients who were treated with a CNI remained on it for a majority of their follow-up time.

There are limited data comparing remission rates, eGFR decline, and kidney failure among patients with FSGS, MCD, and MN [25, 30, 31]. In our study, patients with FSGS had the worst kidney outcomes, including the highest rates of kidney failure and 40% eGFR decline and the lowest rates of remission, consistent with existing retrospective observational and healthcare system-based studies as well as data from the US Renal Data System [25, 30, 32]. While MCD patients had the highest kidney survival probability as well as the highest probability and shortest time to CR, a subset of patients did progress during the long-term follow-up. MN had an intermediate risk of adverse kidney outcomes. Despite similar rates of CR, the rate of kidney failure was ∼3.5 times lower and the rate of 40% eGFR decline was two-fold lower among patients with MN compared to patients with FSGS. These findings build on the results from a prior NEPTUNE analysis demonstrating lower rates of CR for MN than for MCD or FSGS that was limited by a shorter follow-up of 1.6 years (compared with 3.2 years in the current study). We found that CR rates were in fact higher in MN compared to FSGS starting at 2 years from the time of biopsy, in line with the longer time from diagnosis to CR of proteinuria seen among patients with MN [27].

There are several limitations of this study. It is important to note that this report includes a pediatric population undergoing biopsy and thus may not be representative of all pediatric nephrotic syndrome patients. We also acknowledge several atypical features of our MCD cohort, namely the subset of patients who did not have nephrotic-range proteinuria and did not receive IST therapy at any point, perhaps reflecting patients with transient proteinuria and/or mild glomerular pathology who were misclassified as having MCD. There is also a possibility of misclassification bias with regard to the IST therapy received. For example, while treatment with corticosteroids for less than 4 weeks was excluded from the analysis (due to the possibility that a short duration of corticosteroid therapy may have reflected use for another indication), some participants may have received a short course of corticosteroids but discontinued early due to side effects. This could explain the lack of apparent treatment among some MCD patients who experienced CR. Additionally, it is important to highlight that in the NEPTUNE study, the inclusion criteria for proteinuria were relatively broad, and kidney biopsies demonstrating FSGS were not further characterized based on suspected etiology (i.e., primary, secondary, or genetic) [10]. While outcomes in the FSGS cohort should be interpreted with caution given the heterogeneity of the cohort, this report nonetheless provides valuable context regarding the characteristics and treatment patterns in the FSGS cohort according to the presence or absence of full nephrotic syndrome. Lastly, our cohort includes patients who may have received IST at different time points prior to biopsy, primarily consisting of the empiric use of corticosteroids for nephrotic syndrome. While prior treatment can affect histological findings and clinical features at the time of biopsy, our analysis provides important data on post-biopsy treatment patterns and treatment burden in a cohort with broad inclusion criteria representative of patients undergoing their first kidney biopsy in a real-world setting.

In summary, in a prospective observational cohort study of diverse children and adults with biopsy-proven MCD, FSGS, and MN, our study illustrates the substantial heterogeneity of clinical presentation, treatment patterns, and disease burden of these primary glomerular diseases in North America. These findings emphasize that a substantial proportion of patients do not achieve a CR despite treatment with immunosuppression several years after their initial diagnosis, and that persistence of proteinuria is common during the long-term follow-up despite considerable exposure to ISTs. This underscores the large unmet need for more effective treatments to achieve long-term remission and highlights opportunities to utilize nonimmunosuppressive treatments such as renin-angiotensin system and sodium-glucose cotransporter-2 inhibitors to reduce long-term risk in patients with glomerular diseases.

A full list of members of the NEPTUNE Clinical Outcomes Working Group and NEPTUNE Investigators are provided as an online supplementary material.

This study was performed in accordance with the Declaration of Helsinki. Ethical approval was not received for this human study because this study was considered not human subjects research and did not require formal IRB review. This study consisted of secondary data analysis of existing, de-identified data from NEPTUNE. The data presented in this study were provided by the NEPTUNE Data and Analysis Coordinating Center. This study did not require additional local ethical approval in accordance with local and national guidelines. All parents, guardians, or next of kin provided written informed consent for minors, and all adult participants provided written informed consent, to participate in the NEPTUNE study.

Dr. Yelena Drexler has received consulting funds from Travere Therapeutics, Calliditas Therapeutics, and Genentech. Dr. Ambarish Athavale has received consulting funds from Calliditas, Novartis, Travere, and Idorsia. Dr. Abigail R. Smith has received research grant funding from Hi-Bio (Biogen). Dr. Jarcy Zee serves on a Data and Safety Monitoring Board for US Renal Care. Dr. Laura H. Mariani has received consulting funds from Novartis, Calliditas, Dimerix, Vera, and Travere, and research grant funding from Boehringer Ingelheim, Travere, Hi-Bio (Biogen), Takeda, Calliditas, and Reliant Glycosciences. Dr. Richard A. Lafayette has received consulting funds from Alebund, Calliditas, Omeros, Otsuka, Alexion, Travere, Vera, Novartia, Aurinia, Visterra, Vertex, BeiGene, and Biogen, and research grant funding from NIH, Roche, University of Michigan, Alexion, Otsuka, Omeros, Calliditas, Travere, Novartis, Vera, and BeiGene. Dr. Jarcy Zee and Dr. Laura H. Mariani were members of the journal’s Editorial Board at the time of submission. Other authors do not have anything to disclose.

The NEPTUNE study is part of the Rare Diseases Clinical Research Network (RDCRN), which is funded by the National Institutes of Health (NIH) and led by the National Center for Advancing Translational Sciences (NCATS) through its Division of Rare Diseases Research Innovation (DRDRI). NEPTUNE is funded under Grant No. U54DK083912 as a collaboration between NCATS and the National Institute of Diabetes and Digestive and Kidney Diseases (NIDDK). Additional funding or programmatic support is provided by the University of Michigan, NephCure Kidney International, Alport Syndrome Foundation, and the Halpin Foundation. RDCRN consortia are supported by the RDCRN Data Management and Coordinating Center (DMCC), funded by NCATS and the National Institute of Neurological Disorders and Stroke (NINDS) under U2CTR002818. Dr. Ambarish Athavale was supported by NIH Grant R01 DK138395.

Y.D., A.A., A.R.S., Q.L., J.Z., L.H.M., and R.A.L. conceptualized and designed the study. A.R.S., Q.L., and J.Z. conducted data analysis. Y.D., A.A., A.R.S., Q.L., J.Z., L.H.M., and R.A.L. interpreted analysis results. Y.D., A.A., A.R.S., Q.L., and J.Z. drafted the manuscript. L.H.M. and R.A.L. provided supervision. All authors reviewed the manuscript and approved the final version.

The data in this study were obtained from the Nephrotic Syndrome Study Network (NEPTUNE) where data sharing requires ancillary study approval and a data use agreement. The dataset may be requested from the NEPTUNE Data and Analysis Coordinating Center (DACC), www.neptune-study.org.