Abstract
Introduction
The aim of this study was to evaluate the prognostic value of the Mayo Clinic Chronicity Score (MCCS) compared with the National Institutes of Health (NIH) Chronicity Index (NIH-CI) in lupus nephritis (LN).
Methods
We conducted a retrospective cohort study of 307 patients with biopsy-proven LN evaluated at Mayo Clinic (1992–2023). Chronic histologic injury was graded using the NIH-CI and MCCS. Outcomes were proteinuria remission < 500 mg/d (PR500), complete renal response (CRR), end-stage kidney disease (ESKD), and all-cause mortality. Sex-stratified, age-adjusted Cox models were used. Prognostic performance was evaluated using change in Harrell’s C-index (ΔC over a clinical model) and decision-curve analysis was used to assess clinical utility for 5-year ESKD. Subgroup analyses were used to test for effect modification by baseline estimated glomerular filtration rate (eGFR) (< 60 vs. ≥ 60 ml/min per 1.73 m2) and histologic class (proliferative vs. nonproliferative).
Results
Higher NIH-CI and MCCS scores were associated with lower likelihood of PR500 and CRR (hazard ratio [HR]: 0.75 for both) and greater risk of ESKD (HR: 1.40 for NIH-CI; 1.33 for MCCS). Adding either score to a clinical model improved discrimination for PR500 (C-index: 0.65 to 0.71), CRR (0.64 to 0.71), and ESKD (0.81 to 0.85), but not mortality (ΔC = 0.00). Decision-curve analysis showed similar net benefits for NIH-CI and MCCS. Interstitial fibrosis (IF) and tubular atrophy (TA) (IF/TA) were the only components independently predictive of renal outcomes. Our findings support applicability of both scores in nonproliferative LN and in patients with reduced baseline kidney function.
Conclusion
The MCCS and NIH-CI provide comparable and additive prognostic information in LN. MCCS captured the chronic lesions most strongly associated with renal outcomes, with IF/TA as the principal determinant of prognosis.
Keywords: blood vessels/pathology, end-stage kidney disease, kidney/pathology, lupus nephritis, lupus nephritis/pathology, proteinuria
Graphical abstract
LN is among the most severe manifestations of systemic lupus erythematosus. Patients with LN experience a 6-fold increase in mortality compared with people without systemic lupus erythematosus, and approximately 1 in 10 will develop ESKD within 10 years of diagnosis.1 Kidney biopsy remains the gold standard for diagnosing LN, providing essential information for histologic classification, treatment decisions, and prognosis.2,3
The histopathologic evaluation of kidney biopsy in LN enables the assessment of both activity and chronicity changes.4,5 Chronic changes, such as glomerulosclerosis (GS), TA, and IF are considered largely irreversible and are associated with poor long-term outcomes. The most widely used scoring system to evaluate chronic injury in LN is the NIH-CI.6 This index has been modified more recently by the IISN and the Renal Pathology Society.5
Recognizing the prognostic importance of chronic changes, a group of nephrologists and pathologists developed the MCCS as a standardized semiquantitative approach to assess chronic lesions in glomerular diseases.7,8 The MCCS differs from the NIH-CI by including arteriosclerosis (AE) (based on intimal thickening) and excluding fibrous crescents (FCs). GS, IF, and TA are each scored from 0 to 3 based on the percentage of involvement, whereas AE is scored from 0 to 1. The sum of these scores yields an overall chronicity grade: minimal (0–1), mild (2–4), moderate (5–7), or severe (8–10). The MCCS has been previously evaluated in several glomerular diseases, including those associated with systemic autoimmune diseases, such as antineutrophil cytoplasmic antibody–associated vasculitis.9,10 However, this score has never been applied to patients with LN.
In this study, we applied the MCCS in a cohort of patients with LN to quantify chronic histologic injury, evaluate its prognostic value for key renal outcomes, and compare its performance against the widely used ISN/Renal Pathology Society–modified NIH-CI. In addition, we analyzed whether individual biopsy components provide incremental prognostic value beyond clinical parameters typically available to clinicians, and whether the prognostic value of chronicity scores remains consistent across different clinical subgroups and LN classes.
Methods
We conducted a single-center, retrospective cohort study of patients with biopsy-proven LN, whose renal specimens were evaluated at the Mayo Clinic (Rochester, MN) between July 8, 1992, and January 10, 2023. For patients with > 1 biopsy, only the first sample was analyzed and its date taken as the index date.
We included patients who met the 2019 European Alliance of Associations for Rheumatology/ American College of Rheumatology Classification Criteria for systemic lupus erythematosus11 and had active LN, defined by ≥1 of the following: 24-hour proteinuria ≥ 500 mg (or a urine protein-to-creatinine ratio ≥ 0.5 g/g), hematuria ≥ 5 cells/high power field, or reduction of ≥ 15% in eGFR from baseline, attributed to LN.12 A minimum of 6 months postbiopsy follow-up was required.
We excluded patients who, within 3 months before biopsy, had ESKD (eGFR ≤ 15 ml/min per 1.73 m2), were receiving chronic dialysis, or had a kidney allograft. Additional exclusions included biopsies samples with only medulla, patients who did not have proteinuria and serum creatinine values (≥ 2 measurements) within the first year after the index date, and concomitant renal disorders such as antineutrophil cytoplasmic antibody–associated vasculitis or diabetic glomerulopathy; thrombotic microangiopathy was the only cooccurring lesion permitted.
Data Collection
Demographic data, comorbidities, histopathology, laboratory results, and treatments were abstracted at the index date and entered into a REDCap database. All serum creatinine, proteinuria and treatment records from the first year were collected; thereafter, patients were followed up with for ESKD or death until July 21, 2023, death, or lost to follow-up.
Histopathological Analyses
All biopsies were reviewed by a renal pathologist (SS). Chronic lesions were scored according to the NIH-CI and the MCCS. For the NIH-CI, GS, FCs, IF, and TA were each scored 0 to 3 (0%, 1%–24%, 25%–50%, >50%), producing a 0 to 12 total.5 For the MCCS, GS, IF, and TA were scored 0 (< 10 %), 1 (10%–25%), 2 (26%–50%), or 3 (> 50%), and AE was scored 0 or 1 according to intimal-to-media thickness (AE assessment available in the Supplementary Material); totals were categorized as minimal (0–1), mild (2–4), moderate (5–7) or severe (8–10).7 IF and TA were identical in every case (Spearman’s ρ = 1.0) and were therefore analyzed as a single composite (IF/TA) in lesion-level models; the published NIH-CI and MCCS totals were left unaltered.
Outcomes
Our study assessed 4 outcomes as follows: (i) PR500; (ii) CRR defined as PR500 plus eGFR within ± 15 % of baseline; (iii) development of ESKD defined as sustained eGFR ≤ 15 ml/min per 1.73 m2 for ≥ 3 months, requirement for renal replacement therapy > 3 months, or receiving a kidney transplant; and (iv) all-cause mortality.
For PR500 and CRR, observations were censored at ESKD or death; for ESKD, death was treated as a competing risk. Patients who had proteinuria < 500 mg/d at baseline were not included in the PR500 and CRR analyses.
Statistical Analysis
Continuous variables were summarized as median (interquartile range [IQR]) and categorical variables as counts (%). Baseline characteristics were stratified by MCCS categories, and correlation between NIH-CI and MCCS was assessed using Spearman’s ρ. For each outcome, we fitted sex-stratified, age-adjusted Cox proportional-hazards models—cause-specific for ESKD—with NIH-CI and MCCS entered separately as continuous predictors (per 1-point increase) and, for MCCS, also as ordinal grades. Sex effects were estimated using age-adjusted univariable models. Model discrimination was quantified using Harrell’s C-index. Clinical utility was explored by adding each score to a baseline clinical model (age, sex, baseline eGFR, and proteinuria) and estimating the ΔC. For 5-year ESKD, we additionally performed decision-curve analysis to evaluate the net benefit of using predicted 5-year ESKD risk to flag patients as high risk across threshold probabilities (i.e., the predicted risk cut-off used to flag high risk; for example, 10% threshold flags patients with predicted risk ≥ 10%). Because death may preclude ESKD, predicted 5-year ESKD risks were obtained from competing-risk models. Net benefit was calculated across thresholds of 5% to 30% (primary range) and curves were displayed up to 50% to illustrate performance at more conservative thresholds; models were compared against flag-all and flag-none strategies.
Lesion-level analyses comprised sex-stratified, age-adjusted univariable Cox models for GS, IF/TA, AE, and FC (each 0–3) followed by a multivariable model including all 4 lesions; collinearity was assessed using variance-inflation factors and continuous as well as categorical (0, 1, ≥ 2) parameterizations were examined.
Effect-modification was tested by adding interaction terms between each score and the following: (i) baseline eGFR (< 60 vs. ≥ 60 ml/min per 1.73 m2) or (ii) histologic phenotype (proliferative LN: classes III or IV ± V vs. nonproliferative: pure class V or class II). Stratum-specific HRs were reported. All analyses were performed using R 4.2.2 (R Foundation for Statistical Computing, Vienna, Austria) with the survival and tidyverse packages, and the study was approved by the Mayo Clinic Institutional Review Board (23-002317). Patients who declined use of their medical records for research purposes were removed from the study per Minnesota statute.
Results
We identified 659 patients with a diagnosis of LN whose kidney biopsy was reviewed at the Mayo Clinic. Of these, 307 patients met inclusion criteria (Figure 1).
Figure 1.
Flowchart of patient selection and exclusions. Initially, 659 patients with LN were identified. After applying the inclusion criteria, 352 patients had > 6 months of follow-up. Of these, 325 had appropriate renal biopsies. Eighteen patients were excluded because of lack of research authorization, leaving 307 patients with biopsy-proven active LN for analysis. LN, lupus, nephritis; SLE, systemic lupus erythematosus.
The mean age of patients was 34 (SD: 14) years, 75% were female, and 63.5% were White. The median follow-up time was 11 (IQR: 5–16) years. The primary indication for kidney biopsy was the diagnosis of LN, and the mean duration of systemic lupus erythematosus disease was 1 (IQR: 0.0–6.3) year. Hypertension was the most common comorbidity (36.6%), followed by diabetes (6.5%), and coronary artery disease (3.3%). At the time of biopsy, 234 (79.6%) of 294 tested were serologically active (increased double-stranded DNA), 186 (63.7%) had decreased C3, and 205 (70.7%) had decreased C4. The median serum albumin was 3 g/dl (IQR: 2.5–3.5), median creatinine was 1.1 mg/dl (IQR: 0.8–1.5), median eGFR was 73 ml/min per 1.73 m2 (IQR: 49–103) and the median proteinuria was 2995 mg/d (IQR: 1254–6012). Eighty-two percent (n = 252) of the patients received high dose glucocorticoids with or without additional immunosuppressants. Of the patients, 68.9% received mycophenolate mofetil and 25% received cyclophosphamide. There were 7.5% of patients with class II LN, 15.3% with class III LN, 26.1% with class IV LN, 25.4% with class V LN, 24% with class III/IV + V LN, and 1.6% with class VI LN. The median NIH-CI was 0 (IQR: 0–3) and the median MCCS was 0 (IQR: 0–3). GS was present in 36.4% of the biopsies, IF/TA in 37.8%, FCs in 4.9%, and AE in 12.1%. MCCS grading were categorized as 188 (61.2%) for minimal chronicity, 78 (25.4%) mild, 24 (7.8%) moderate, and 17 (5.5%) severe. Further baseline characteristics stratified by MCCS categories are detailed in (Table 1). The NIH-CI and MCCS were highly correlated (Spearman’s ρ = 0.98; Supplementary Figure S1).
Table 1.
Baseline demographic and clinical characteristics of a cohort of patients with biopsy-proven lupus nephritis, including laboratory values and induction therapies, stratified by the MCCS
| Variable | MCCS minimal (0–1,), n = 188 | MCCS mild (2–4), n = 78 | MCCS moderate (5–7), n = 24 | MCCS severe (8–10), n = 17 | Total, N = 307 |
|---|---|---|---|---|---|
| Age at index (yrs), mean (SD) | 31 (14) | 37 (13) | 39 (16) | 46 (9) | 34 (14) |
| Sex, n (%) | |||||
| Male | 49 (26.1) | 20 (25.6) | 4 (16.7) | 3 (17.6) | 76 (24.8) |
| Female | 139 (79.3) | 58 (74.4) | 20 (83.3) | 14 (82.4) | 231 (75.2) |
| Race / Ethnicity, n (%) | |||||
| Non-Hispanic White | 127 (67.6) | 48 (61.5) | 7 (29.2) | 13 (76.5) | 195 (63.5) |
| Hispanic or Latino | 18 (9.6) | 5 (6.4) | 4 (16.7) | 2 (11.8) | 29 (9.4) |
| Asian | 17 (9.0) | 6 (7.7) | 5 (20.8) | 0 (0.0) | 28 (9.1) |
| Black | 10 (5.3) | 8 (10.3) | 5 (20.8) | 1 (5.9) | 24 (7.8) |
| American Indian | 3 (1.6) | 1 (1.3) | 0 (0.0) | 1 (5.9) | 5 (1.6) |
| Other/ Unknown | 13 (6.9) | 10 (12.8) | 3 (12.5) | 0 (0.0) | 26 (8.5) |
| Laboratories at index date, n (%) | |||||
| Increased anti-dsDNA | 148/182 (81.3) | 55/73 (75.3) | 18/23 (78.3) | 13/16 (81.2) | 234/294 (79.6) |
| Decreased C3 | 115/176 (65.3) | 50/75 (66.7) | 15/24 (62.5) | 6/17 (35.3) | 186/292 (63.7) |
| Decreased C4 | 125/175 (71.4) | 56/75 (74.7) | 15/24 (62.5) | 9/16 (56.2) | 205/290 (70.7) |
| Positive anti-phospholipid antibodies | 35/102 (34.3) | 9/38 (23.7) | 5/15 (33.3) | 3/7 (42.9%) | 52/162 (32.1) |
| Serum albumin g/dl, n = 247, median (IQR) | 3 (2.5–3.5) | 3 (2.7–3.5) | 3.1 (2.5–3.4) | 3 (2.7–3.3) | 3.0 (2.5–3.5) |
| Proteinuria mg/d, median (IQR) | 2368 (1072–5130) | 3354 (2009–6507) | 4296 (1776–7792) | 3500 (1200–6000) | 2995 (1254–6012) |
| eGFR at diagnosis ml/min, median (IQR) | 86 (63–109) | 60 (43–88) | 42 (21–60) | 24 (17–33) | 73 (49–103) |
| Creatinine mg/dl, median (IQR) | 0.9 (0.8–1.2) | 1.2 (0.9–1.7) | 1.6 (1.3–2.9) | 2.6 (2.1–3.1) | 1.1 (0.8–1.5) |
| Hydroxychloroquine, n (%)a | 82/187 (43.9) | 44/78 (56.4) | 8/24 (33.3) | 8/17 (47.1) | 142/306 (46.4) |
| ARBs or ACEi, n (%)b | 76/186 (40.9) | 52/78 (66.7) | 9/24 (37.5) | 8/17 (47.1) | 145/305 (47.5) |
| Treatment Induction, n (%)c | |||||
| High dose glucocorticoids | 155 (82.9) | 67 (87.5) | 21 (87.5) | 9 (52.9) | 252 (82.4) |
| Glucocorticoids in pulses | 72/155 (46.5) | 27/67 (40.3) | 11/21 (52.4) | 2/9 (22.2) | 112/252 (44.4) |
| Oral glucocorticoids at induction phase | 152/155 (98.1) | 67/67 (100) | 21/21 (100) | 9/9 (100) | 249/252 (98.8) |
| Induction with immunosuppressants | 143/155 (92.3) | 63/67 (94) | 21/21 (100) | 8/9 (88.9) | 235/252 (93.3) |
| Lupus nephritis class, n (%) | |||||
| Mesangial (II) | 17 (9.0) | 4 (5.1) | 1 (4.2) | 1 (5.9) | 23 (7.5) |
| Proliferative (III, IV) | 84 (44.7) | 25 (32.1) | 15 (62.5) | 3 (17.6) | 127 (41.4) |
| Membranous (V) | 46 (24.5) | 25 (32.1) | 2 (8.3) | 5 (29.4) | 78 (25.4) |
| Proliferative + Membranous (III, IV + V) | 41 (21.8) | 24 (30.8) | 5 (20.8) | 4 (23.5) | 74 (24.1) |
| Sclerosing (VI) | 0 (0.0) | 0 (0.0) | 1 (4.2) | 4 (23.5) | 5 (1.6) |
| NIH chronicity score, median (IQR) | 0 (0,0) | 3 (2,3) | 6 (6,6) | 9 (8,9) | 0 (0,3) |
| MCCS chronicity score, median (IQR) | 0 (0,0) | 3 (2,3) | 6 (6,7) | 10 (8, 10) | 0 (0,3) |
| Glomerulosclerosis score, n (%) | |||||
| 0 | 159 (84.6) | 36 (46.2) | 0 (0.0) | 0 (0.0%) | 195 (63.5) |
| 1 | 29 (15.4) | 31 (39.7) | 3 (12.5) | 1 (5.9) | 64 (20.8) |
| 2–3 | 0 (0.0) | 11 (14.1) | 21 (87.5) | 16 (94.1) | 48 (15.6) |
| Interstitial fibrosis score, n (%) | |||||
| 0 | 188 (100) | 3 (3.8) | 0 (0.0) | 0 (0.0) | 191 (62.2) |
| 1 | 0 (0.0) | 72 (92.3) | 5 (20.8) | 0 (0.0) | 77 (25.1) |
| 2–3 | 0 (0.0) | 3 (3.8) | 19 (79.2) | 17 (100) | 39 (12.7) |
| Tubular atrophy, n (%) | |||||
| 0 | 188 (100) | 3 (3.8) | 0 (0.0) | 0 (0.0) | 191 (62.2) |
| 1 | 0 (0.0) | 72 (92.3) | 5 (20.8) | 0 (0.0) | 77 (25.1) |
| 2–3 | 0 (0.0) | 3 (3.8) | 19 (79.2) | 17 (100) | 39 (12.7) |
| Fibrous crescent score, n (%) | |||||
| 0 | 184 (97.9) | 69 (88.5) | 22 (91.7) | 17 (100) | 292 (95.1) |
| 1–2 | 4 (2.1) | 9 (11.5) | 2 (8.3) | 0 (0.0) | 15 (4.9) |
| Arteriosclerosis score, n (%) | |||||
| 0 | 184 (97.9) | 72 (92.3) | 13 (54.2) | 1 (5.9) | 270 (87.9) |
| 1 | 4 (2.1) | 6 (7.7) | 11 (45.8) | 16 (94.1) | 37 (12.1) |
ACEI, angiotensin-converting enzyme inhibitors; AE, arteriosclerosis; ARBs, angiotensin II receptor blockers; dsDNA, double-stranded DNA; eGFR, estimated glomerular filtration rate; FC, fibrous crescents; GS, glomerulosclerosis; IF, interstitial fibrosis; IQR, interquartile range; RPS, Renal Pathology society; MCCS, Mayo Clinic Chronicity Score; NIH, National Institutes of Health; TA, tubular atrophy.
The NIH chronicity score comprised GS, FC, IF, and TA, whereas the MCCS comprised GS, IF, TA, and AE. GS, FC, IF, and TA were each graded on a 0–3 scale according to the percentage involvement of the respective compartment; AE was graded on a 0–1 scale according to the degree of intimal thickening.
The table also presents the distribution of histopathologic findings according to ISN/RPS class and the frequency of individual components of the chronicity indices defined by both the NIH and MCCS.
To apply the MCCS; GS, IF, and TA are first scored from 0 to 3 according to the percentage of involvement of each compartment (< 10, 10–25, 26–50, and > 50%). Subsequently, the scores are added to grade the overall severity of the chronic lesions into 4 MCCS grades: minimal (0–1), mild (2–4), moderate (5–7), and severe (8–10). Activity Index was not considered during biopsy analysis.
Continuous variables are summarized as median (25th–75th percentile) and categorical variables are summarized as frequency (percent).
55 out of 307 patients did not receive immunosuppressive treatment, because of having pure mesangial, pure membranous, and sclerosing nephritis classes.
Number of patients with hydroxychloroquine at index date.
Number of patients with ARB or ACEi at index date.
Immunosuppressants: cyclophosphamide, mycophenolate mofetil, rituximab, belimumab, tacrolimus, azathioprine, cyclosporine.
Association of Chronicity Scores With Clinical Outcomes
Of the 307 patients, 129 of 277 (46.6%) achieved PR500 and 118 of 277 (42.5%) achieved CRR within 1 year. During the entire follow-up, 33 patients (10.7%)died, and 60 patients (19.5%) progressed to ESKD.
Older age was significantly associated with lower likelihood of achieving PR500 (HR per 10-year increase: 0.86, 95% confidence interval [CI]: 0.75–0.98) and CRR (HR: 0.83, 95% CI: 0.72–0.96), and with higher mortality risk (HR: 1.74, 95% CI: 1.39–2.17). Sex was not associated with any of the outcomes (Table 2).
Table 2.
Proteinuria and clinical outcomes defined by proteinuria < 500 mg/d, complete renal response (CRR), End-stage kidney disease (ESKD) and death, applying NIH chronicity index and Mayo Clinic chronicity score
| Element | PR500 HR (95% CI) | CRR HR (95% CI) | ESKD HR (95% CI) | Death HR (95% CI) |
|---|---|---|---|---|
| Age, per 10-yr increase | 0.86 (0.75–0.98) | 0.83 (0.72–0.96) | 0.99 (0.82–1.18) | 1.74 (1.39–2.17) |
| Sex female | 1.03 (0.68–1.55) | 1.07 (0.69–1.65) | 0.87 (0.48–1.56) | 0.77 (0.37–1.60) |
| NIH chronicity index (1-unit increase) | 0.76 (0.68–0.85) | 0.76 (0.68–0.86) | 1.53 (1.38–1.68) | 1.11 (0.98–1.25) |
| MCCS grading (1-unit increase) | 0.77 (0.69–0.86) | 0.77 (0.68–0.86) | 1.44 (1.32–1.57) | 1.10 (0.99–1.22) |
| MCCS minimal (0–1) | Reference | Reference | Reference | Reference |
| MCCS mild (2–4) | 0.47 (0.30–0.73) | 0.46 (0.29–0.74) | 3.59 (1.81–7.11) | 1.13 (0.46–2.82) |
| MCCS moderate (5-7) | 0.31 (0.12–0.76) | 0.28 (0.10–0.75) | 18.46 (8.41–40.53) | 2.13 (0.68–6.64) |
| MCCS severe (8-10) | 0.10 (0.01–0.71) | 0.11 (0.02–0.83) | 23.02 (9.15–57.90) | 2.36 (0.84–6.60) |
AE, arteriosclerosis; CI, confidence interval; CRR, complete renal response; ESKD, end-stage kidney disease; FC, fibrous crescent; GS, glomerulosclerosis; HR, hazard ratio; IF, interstitial fibrosis; MCCS, Mayo Clinic Chronicity Score; NIH, National Institutes of Health; PR500, proteinuria < 500 mg/d; TA, tubular atrophy.
Estimates for age and sex are from univariable models. All other estimates are from models adjusted for age and stratified by sex. Estimates for ages are per 10-yr increase.
Higher NIH-CI scores predicted lower likelihood of remission (HR: 0.76 per 1-point increase, 95% CI: 0.68–0.85 for PR500 and HR: 0.76, 95% CI: 0.68–0.86 for CRR; Table 2) and higher risk of progression to ESKD (HR: 1.53, 95% CI: 1.38–1.68; Table 2). MCCS grading similarly demonstrated a clear gradient of risk. Compared with patients with minimal chronicity (MCCS: 0–1), severe chronicity (MCCS: 8–10) was associated with approximately 90% reduction in remission likelihood (HR: 0.10, 95% CI: 0.01–0.71 for PR500; HR: 0.11, 95% CI: 0.02–0.83 for CRR; Table 2), and dramatically elevated risk of progression to ESKD (HR: 23.02, 95% CI: 9.15–57.90; Table 2). Neither the NIH nor MCCS chronicity scores significantly predicted all-cause mortality (Table 2).
Clinical Added Value of Chronicity Scores
We evaluated whether adding biopsy chronicity scores provided incremental prognostic information beyond standard clinical predictors (age, sex, proteinuria, and eGFR). Incorporating the NIH-CI and MCCS modestly improved discrimination for renal outcomes (Table 3). Specifically, the addition of chronicity scores increased the Harrell’s C-index from 0.65 (95% CI: 0.60–0.70) to 0.71 (0.66–0.75; ΔC = + 0.06) for predicting PR500, and from 0.64 (0.59–0.69) to 0.71 (0.66–0.75; ΔC = + 0.07) for predicting CRR. For progression to ESKD, discrimination improved from 0.81 (0.74–0.88) to 0.85 (0.80–0.90; ΔC = + 0.04). In contrast, neither score improved the prediction of mortality, with the C-index remaining unchanged at 0.67 (0.52–0.82; ΔC = 0.00). In decision-curve analysis for 5-year ESKD, models incorporating either NIH-CI or MCCS provided net benefit over the clinical model and the flag-all/flag-none strategies. Net benefit for the clinical model and score-augmented models was similar across threshold probabilities up to 30%, with separation emerging at higher thresholds (> 30%), where the addition of either chronicity score yielded greater net benefit than clinical variables alone. The NIH-CI and MCCS decision curves were nearly overlapping across the full range of thresholds, indicating similar clinical utility for flagging patients as high risk for 5-year ESKD (Figure 2).
Table 3.
Incremental prognostic value of biopsy chronicity scores over clinical variablesa
| End point | Clinical model C-index (95% CI) |
Clinical model plus score C-Index(95% CI) |
ΔC | NIH-CI HR per 1 -point increase (95% CI) |
MCCS HR per 1 -point increase (95% CI) |
|---|---|---|---|---|---|
| PR500 | 0.65 (0.60–0.70) | 0.71 (0.66–0.75) | + 0.06 | 0.75 (0.66–0.85) | 0.76 (0.67–0.86) |
| CRR | 0.64 (0.59–0.69) | 0.71 (0.66–0.75) | + 0.07 | 0.75 (0.66–0.85) | 0.75 (0.66–0.86) |
| ESKD | 0.81 (0.74–0.88) | 0.85 (0.79–0.90) | + 0.04 | 1.40 (1.26–1.56) | 1.33 (1.21–1.47) |
| Death | 0.67 (0.53–0.82) | 0.67 (0.52–0.82) | 0.00 | 1.05 (0.91–1.22) | 1.05 (0.92–1.20) |
CI, confidence interval; CRR, complete renal response; eGFR, estimated glomerular filtration rate; ESKD, end-stage kidney disease; HR, hazard ratio; MCCS, Mayo Clinic Chronicity Score; NIH-CI, National Institutes of Health Chronicity Index; PR500, proteinuria < 500 mg/d; ΔC, change in Harrell’s C-index.
Adjusted for age, sex, and baseline proteinuria, and eGFR.
Figure 2.
Decision-curve analysis for 5-year ESKD. Net benefit is plotted against the threshold probability of 5-year ESKD (i.e., the predicted 5-year ESKD risk cut-off used to flag a patient as high risk). Curves compare the clinical model (age, sex, baseline estimated glomerular filtration rate, and baseline proteinuria) with the clinical model plus NIH-CI or MCCS. “Flag all” and “Flag none” represent the extreme strategies of classifying all patients or no patients, respectively, as high risk for 5-year ESKD. Predicted 5-year ESKD risks were derived from competing-risk models accounting for death as a competing event. ESKD, end-stage kidney disease; MCCS, Mayo Clinic Chronicity Score; NIH-CI, National Institutes of Health Chronicity Index.
Each 1-point increase in the NIH-CI after stratifying by sex and adjusting for age, baseline proteinuria, and eGFR, significantly decreased the likelihood of achieving remission outcomes (HR: 0.75, 95% CI: 0.66–0.85) for both proteinuria remission and CRR; and increased the risk of progression to ESKD (HR: 1.40; 95% CI: 1.26–1.56). The MCCS showed very similar HRs, underscoring that both chronicity scores provided comparable prognostic information beyond clinical predictors alone. Neither chronicity score provided additional prognostic value for mortality (HR: 1.05; 0.91–1.22; not significant).
Lesion-Level Analyses
In univariable analysis adjusted for age and stratified by sex, several individual biopsy lesion components demonstrated significant associations with clinical outcomes (Table 4). GS, IF/TA, and AE each significantly reduced the probability of achieving PR500 and CRR. Specifically, higher GS scores lowered the likelihood of PR500 (HR: 0.58; 95% CI: 0.43–0.77) and CRR (HR: 0.57; 95% CI: 0.42–0.78). Similarly, IF/TA strongly reduced the probability of both PR500 (HR: 0.48; 95% CI: 0.35–0.66) and CRR (HR: 0.48; 95% CI: 0.34–0.67). AE significantly predicted lower remission rates for both PR500 (HR: 0.47; 95% CI: 0.23–0.97) and CRR (HR: 0.39; 95% CI: 0.17–0.90).
Table 4.
Analysis by each component of the chronicity scores and its contribution to proteinuria and clinical outcomes after univariable and multivariable adjustments (modeled per 1-point increase)
| End point | Biopsy component | Univariable HR (95% CI) | Multivariable HR (95% CI) |
|---|---|---|---|
| PR500 | GS | 0.58 (0.43–0.77) | 0.81 (0.56–1.15) |
| IF (or TA) | 0.48 (0.35–0.66) | 0.54 (0.37–0.79) | |
| AE | 0.47 (0.23–0.97) | 1.06 (0.49–2.28) | |
| FC | 0.88 (0.44–1.74) | 0.93 (0.47–1.85) | |
| CRR | GS | 0.57 (0.42–0.78) | 0.81 (0.55–1.18) |
| IF (or TA) | 0.48 (0.34–0.67) | 0.55 (0.37–0.82) | |
| AE | 0.39 (0.17–0.90) | 0.86 (0.36–2.06) | |
| FC | 1.00 (0.50–1.99) | 1.06 (0.53–2.11) | |
| ESKD | GS | 2.57 (1.97–3.36) | 1.32 (0.91–1.91) |
| IF (or TA) | 3.39 (2.57–4.47) | 3.70 (2.40–5.70) | |
| AE | 3.56 (1.87–6.78) | 0.33 (0.14–0.81) | |
| FC | 0.92 (0.26–3.26) | 1.07 (0.29–3.97) | |
| Death | GS | 1.40 (1.01–1.94) | 1.48 (0.85–2.57) |
| IF (or TA) | 1.33 (0.94–1.88) | 1.35 (0.73–2.47) | |
| AE | 1.09 (0.41-2.90) | 0.31 (0.08-1.23) | |
| FC | 0.00 (0.00- inf) | 0.00 (0.00- inf) |
AE, arteriosclerosis; CI, confidence interval; CRR, complete renal response; ESKD, end-stage kidney disease; FC, fibrous crescent; GS, glomerulosclerosis; HR, hazard ratios; IF, interstitial fibrosis; PR500, proteinuria < 500 mg/d; TA, tubular atrophy.
Univariable, stratified by sex and adjusted for age.
Multivariable, stratified by sex, adjusted for age and all 5 lesions together (GS, IF, TA, AE, and FC).
Higher GS scores (HR: 2.57; 95% CI: 1.97–3.36), IF/TA scores (HR: 3.39; 95% CI: 2.57–4.47), and AE scores (HR: 3.56; 95% CI: 1.87–6.78) were significantly associated with progression to ESKD. GS showed a borderline association with increased mortality (HR: 1.40; 95% CI: 1.01–1.94); however, other lesions were not significantly predictive of mortality in the univariable models. FC did not significantly impact any of the outcomes.
Multivariable analysis, including all lesion components simultaneously along with adjustment for age and stratification by sex, identified IF/TA as the only consistently independent predictor across renal outcomes. IF/TA remained significantly predictive of a lower probability of achieving PR500 (HR: 0.54; 95% CI: 0.37–0.79) and CRR (HR: 0.55; 95% CI: 0.37–0.82) and strongly increased risk of ESKD (HR: 3.70; 95% CI: 2.40–5.70). Notably, AE displayed a reversed association in the multivariable ESKD model (HR: 0.33; 95% CI: 0.14–0.81). Examining age-adjusted and sex-stratified models, including AE and other combinations of individual biopsy lesion components revealed collinearity between AE and both GS and IF, which explains the reverse association of AE in the multivariable model as well as the reduced HR for GS. None of the biopsy components independently predicted mortality in multivariable analyses. The predictive associations of individual kidney biopsy lesion components with clinical outcomes were consistent whether lesions were analyzed as continuous variables (per 1-point increase; Table 4) or categorically (Supplementary Table S1).
Subgroup Analyses by Clinical and Histological Features
In interaction analyses adjusted for age and stratified by sex, we assessed whether baseline kidney function (eGFR < 60 vs. ≥ 60 ml/min per 1.73 m2) or histologic classification (proliferative LN [class III/IV ± V] vs. nonproliferative LN [class II or V]) modified the association of NIH-CI and MCCS with clinical outcomes. No meaningful interactions were identified (all interaction P-values > 0.10), indicating that the prognostic performance of the chronicity scores was consistent across these clinically important patient subgroups (Table 5).
Table 5.
Subgroup analyses to assess the association between NIH-CI and MCCS with outcomes according to eGFR < 60 ml/min and LN proliferative classes
| Effect | Outcome | Among eGFR < 60 ml/min HR (95% CI), n = 112 | Among eGFR ≥ 60 ml/min HR (95% CI) n = 194 | Among nonproliferative HR (95% CI) n = 101 | Among proliferative HR (95% CI) n = 201 |
|---|---|---|---|---|---|
| NIH chronicity, per 1-unit increase | PR500 | 0.77 (0.66–0.90) | 0.73 (0.61–0.88) | 0.68 (0.52–0.90) | 0.79 (0.70–0.90) |
| CRR | 0.76 (0.65–0.89) | 0.73 (0.60–0.89) | 0.64 (0.46–0.89) | 0.80 (0.70–0.91) | |
| ESKD | 1.44 (1.27–1.62) | 1.47 (1.24–1.74) | 1.43 (1.22–1.66) | 1.60 (1.40–1.83) | |
| Death | 1.09 (0.93–1.28) | 0.98 (0.72–1.33) | 0.97 (0.75–1.26) | 1.15 (0.98–1.35) | |
| MCCS chronicity, per 1-unit increase | PR500 | 0.79 (0.69–0.91) | 0.71 (0.57–0.87) | 0.68 (0.52–0.90) | 0.80 (0.71–0.91) |
| CRR | 0.78 (0.67–0.91) | 0.69 (0.55–0.86) | 0.64 (0.46–0.89) | 0.80 (0.70–0.91) | |
| ESKD | 1.36 (1.22–1.51) | 1.40 (1.20–1.63) | 1.35 (1.17–1.55) | 1.52 (1.35–1.71) | |
| Death | 1.08 (0.93–1.24) | 1.00 (0.74–1.33) | 0.96 (0.75–1.23) | 1.13 (0.98–1.31) |
CRR, complete renal response; eGFR, estimated glomerular filtration rate; ESKD, end-stage kidney disease; LN, lupus nephritis; MCCS, Mayo Clinic Chronicity Score; NIH-CI, National Institute of health Chronicity Index; PR500, proteinuria < 500 mg/d.
Variables adjusted for age and sex.
Discussion
In this retrospective cohort of patients with LN, we found that both the NIH-CI and the MCCS were predictive of remission and progression to ESKD and their addition to clinical data improves risk stratification for renal outcomes. Among individual histologic components, GS, IF/TA, and AE were predictors of adverse renal outcomes, whereas FCs were not. However, IF/TA emerged as the most robust and consistently significant lesion. Importantly, the prognostic value of both chronicity scores remained consistent across key clinical and histologic subgroups, including patients with reduced baseline kidney function and those with nonproliferative LN histologic classes.
Despite differences in structure, the NIH-CI and MCCS demonstrated near-identical predictive performance. The NIH-CI includes FCs and omits AE,5 whereas the MCCS incorporates AE but not FCs.5,7 The NIH-CI and MCCS demonstrated nearly perfect correlation, which is expected because both instruments capture a shared core of chronic histologic damage in LN. Both scores were strongly associated with lower rates of proteinuria remission and higher risk of progression to ESKD, supporting their validity as markers of chronic injury.13 Notably, neither score was associated with all-cause mortality. Their high concordance and similar predictive performance suggest they are interchangeable in clinical practice. However, a distinguishing feature of the MCCS is that it has been validated across a broad range of glomerular diseases.9,10 Thus, its incorporation into LN assessment may therefore offer practical advantages, including greater harmonization of pathology workflows across glomerular diseases.
One of the aims of this study was to determine whether these histologic scores offer incremental value beyond clinical data available to treating physicians. Our findings support this added value: incorporating either the NIH-CI or MCCS into a baseline model containing age, eGFR, proteinuria, and stratified by sex improved model discrimination for key renal outcomes. For example, the C-index for predicting proteinuria remission increased from 0.65 to 0.71 with either score; and the C-index for predicting ESKD increased from 0.81 to 0.85. In addition, our decision-curve analysis showed that models incorporating either NIH-CI or MCCS provided net benefit over the baseline model. These findings highlight that chronic histologic damage, captured at the time of biopsy, provides clinically actionable information that cannot be fully inferred from standard laboratory measures alone. In fully adjusted models, each 1-point increase in the NIH-CI or MCCS produced nearly identical HRs in the reduction in the likelihood of achieving PR500 and CRR, and increase in the risk of progression to ESKD, reinforcing that both scores offer comparable and independent prognostic value beyond conventional clinical variables. Neither score was significantly associated with mortality, consistent with the notion that chronic kidney injury predicts renal progression but not overall survival in LN at least within the follow-up interval of this study.
At the lesion level, IF/TA emerged as the most powerful and consistent predictors of all renal outcomes. Even mild IF/TA (score 1) was significantly associated with a reduced likelihood of achieving PR500 and CRR, and with a higher risk of progression to ESKD. The prognostic impact was magnified in patients with moderate-to-severe IF/TA (scores 2–3), where the risk of ESKD increased > 20-fold compared with patients with no fibrosis. This strong dose-response relationship underscores the clinical importance of identifying and quantifying IF/TA on kidney biopsy, because these lesions reflect irreversible parenchymal scarring and reduced nephron reserve—features that may limit the potential for recovery, even with effective immunosuppression. Previous studies have demonstrated that IF/TA is an independent predictor of progression to ESKD, irrespective of the class of LN, highlighting the prognostic value of chronic lesions beyond the conventional classification14; others support that it is an independent predictor even in the presence of hypertension and diabetes.15 Furthermore, a recent publication found that patients with proliferative LN are more likely to show progression of IF/TA on follow-up biopsy.16
Although GS and AE also demonstrated associations with renal outcomes in univariable models, their predictive value was not independent in multivariable analysis. FCs, despite their inclusion in the NIH-CI, were not associated with any outcome. This finding raises questions about their long-term prognostic relevance. One plausible explanation is that FC lesions represent an intermediate stage in the chronicity continuum. In a study of serial kidney biopsies by Malvar et al.,17 fibrocellular crescents were observed to evolve into FCs at second biopsy, with a subsequent decline in their proportion over time. This supports the concept that FC lesions transition into GS, which represents the final scarring phase of crescentic injury. Another possible explanation is that FCs simply represent “burned” lesions; in fact, in IgA vasculitis nephritis, FC were found to be associated with stable renal function over time and did not correlate with the development of ESKD in a cohort of 361 patients diagnosed with this particular glomerulopathy.18
Our results are consistent with findings from a Mexican LN cohort that reported no association between FC and complete response.19 However, that study did report an increased ESKD risk associated with FCs, a discrepancy likely attributable to differences in model adjustment. Unlike their study, we accounted for age and sex, which may have rendered FCs nonsignificant in our analyses. A Dutch cohort similarly found that FC were associated with ESKD even after adjustment, but included biopsies from the 1980s and earlier, raising the possibility that differences in treatment era, patient demographics, or standard-of-care influenced those findings.20
Taken together, our data suggest that the MCCS captures all chronic histologic lesions independently associated with renal outcomes in LN. These findings are in line with previous multiinstitutional validations of the MCCS across glomerular diseases.9 Once IF/TA is included in outcome models, the predictive contribution of other chronic lesions diminishes; likely because of shared fibrotic pathways.21 This reinforces the centrality of IF/TA in chronic kidney disease progression and supports its weighting in future scoring systems. Given that IF/TA was the most predictive lesion in large validation cohorts such as that by Srivastava et al.,9 these findings suggest that IF/TA may be a universal histologic driver of renal prognosis across glomerular diseases.
Given that the NIH-CI was developed from patients with proliferative lesions and few studies have evaluated its performance in nonproliferative classes, we examined whether the prognostic value of chronicity scores differed by histologic subtype.6 No significant interactions were observed, indicating that both the NIH-CI and MCCS performed consistently across patients with proliferative (class III/IV ± V) and nonproliferative (class II or V) LN. In additional subgroup analyses, the scores performed similarly in patients with reduced (eGFR < 60 ml/min per 1.73 m2) versus preserved kidney function. These findings support the broad applicability of chronicity scoring systems across the full spectrum of LN presentations, including cases traditionally considered lower risk, such as class V disease. This consistency reinforces the clinical utility of the NIH-CI and MCCS as generalizable tools for risk stratification in diverse LN populations.
This study has limitations. It was conducted at a single tertiary care center, and the retrospective design limits causal inference. Treatment data were collected but not included in outcome models, as immunosuppression likely lies on the causal pathway between biopsy findings and outcomes. As the standard of care changes, studies looking for associations of the biopsy with outcomes need to be repeated with some periodicity, perhaps every 10 years. Although biopsy interpretation followed standard protocols and was performed by an expert nephropathologist, interobserver variability was not assessed. This is important because published studies demonstrate only moderate interobserver reproducibility for NIH indices and systematic reviews report similar “moderate” concordance for activity and chronicity scoring.22,23 Ongoing standardization efforts (including the 2018 ISN/ Renal Pathology Society revision with clarified definitions and modified NIH indices) may improve reproducibility.5
In conclusion, both the NIH-CI and MCCS are strong, independent predictors of renal outcomes in LN that provides additional prognostic value than routine clinical data available to the clinician, with IF/TA lesions emerging as the key histologic determinant of long-term prognosis. Importantly, the predictive performance of both chronicity scores was consistent across key clinical subgroups, including patients with nonproliferative LN and those with reduced kidney function, supporting their broad applicability across the clinical spectrum of LN. The MCCS includes all lesions associated with outcomes and has the added advantage of cross-disease validation, making it a practical and generalizable tool for nephropathology. Its broader adoption may help streamline and harmonize chronicity assessment across glomerular diseases.
Disclosure
In the past 3 years, AD-G has received unrelated grant funding from the Centers for Disease Control and Prevention (U01DP06491), the Rheumatology Research Foundation Investigator Award, the Lupus Research Alliance Diversity in Lupus Research Career Development Award, and the Mayo Clinic Margaret Harvey Schering Clinician Career Development Award for Arthritis research. All the other authors declared no competing interests.
Data Availability Statement
Data is available upon reasonable request from the corresponding author.
Author Contributions
MCC-G contributed to conceptualization, methodology, data curation, writing - original draft, and writing - review and editing. EN-M contributed to conceptualization, methodology, data curation, writing - original draft, and writing - review and editing. JF-G contributed to data curation and writing - review and editing. GF-P contributed to conceptualization, methodology, and writing - review and editing. KAW contributed to conceptualization, methodology, and writing - review and editing. MCM contributed to conceptualization and writing - review and editing. FCF contributed to conceptualization, methodology, and writing - review and editing. ACH contributed to formal analysis and writing - review and editing. CSC contributed to formal analysis and writing - review and editing. SS contributed to conceptualization, methodology, data curation, writing - original draft, and writing - review and editing. ADG contributed to conceptualization, methodology, data curation, writing - original draft, writing - review and editing, and supervision. All the authors had full access to all the data in the study and had a final responsibility for the decision to submit for publication.
Footnotes
Arteriosclerosis Assessment.
Figure S1. Relationship between NIH chronicity score and MCCS grade.
Table S1. Analysis by each component of the chronicity scores and its contribution to proteinuria and clinical outcomes after univariable and multivariable adjustments using categorical variables.
Supplementary Material
Arteriosclerosis Assessment. Figure S1. Relationship between NIH chronicity score and MCCS grade. Table S1. Analysis by each component of the chronicity scores and its contribution to proteinuria and clinical outcomes after univariable and multivariable adjustments using categorical variables.
References
- 1.Hocaoǧlu M., Valenzuela-Almada M.O., Dabit J.Y., et al. Incidence, prevalence, and mortality of lupus nephritis: a population-based study over four decades using the lupus midwest network. Arthritis Rheumatol. 2023;75:567–573. doi: 10.1002/art.42375. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2.Sammaritano L.R., Askanase A., Bermas B.L., et al. 2024 American College of Rheumatology (ACR) guideline for the screening, treatment, and management of lupus nephritis. Arthritis Rheumatol. 2025;77:1115–1135. doi: 10.1002/art.43212. [DOI] [PubMed] [Google Scholar]
- 3.Sammaritano L.R., Askanase A., Bermas B.L., et al. 2024 American College of Rheumatology (ACR) guideline for the screening, treatment, and management of lupus nephritis. Arthritis Care Res (Hoboken) 2025;77:1045–1065. doi: 10.1002/acr.25528. [DOI] [PubMed] [Google Scholar]
- 4.Schwartz M.M., Lan S.P., Bernstein J., Hill G.S., Holley K., Lewis E.J. Role of pathology indices in the management of severe lupus glomerulonephritis. Lupus Nephritis Collaborative study group. Kidney Int. 1992;42:743–748. doi: 10.1038/ki.1992.342. [DOI] [PubMed] [Google Scholar]
- 5.Bajema I.M., Wilhelmus S., Alpers C.E., et al. Revision of the International Society of Nephrology/Renal Pathology Society classification for lupus nephritis: clarification of definitions, and modified National Institutes of Health activity and chronicity indices. Kidney Int. 2018;93:789–796. doi: 10.1016/j.kint.2017.11.023. [DOI] [PubMed] [Google Scholar]
- 6.Austin H.A., 3rd, Muenz L.R., Joyce K.M., Antonovych T.T., Balow J.E. Diffuse proliferative lupus nephritis: identification of specific pathologic features affecting renal outcome. Kidney Int. 1984;25:689–695. doi: 10.1038/ki.1984.75. [DOI] [PubMed] [Google Scholar]
- 7.Sethi S., D’Agati V.D., Nast C.C., et al. A proposal for standardized grading of chronic changes in native kidney biopsy specimens. Kidney Int. 2017;91:787–789. doi: 10.1016/j.kint.2017.01.002. [DOI] [PubMed] [Google Scholar]
- 8.Wu L.H., Yu F., Tan Y., et al. Inclusion of renal vascular lesions in the 2003 ISN/RPS system for classifying lupus nephritis improves renal outcome predictions. Kidney Int. 2013;83:715–723. doi: 10.1038/ki.2012.409. [DOI] [PubMed] [Google Scholar]
- 9.Srivastava A., Palsson R., Kaze A.D., et al. The prognostic value of histopathologic lesions in native kidney biopsy specimens: results from the Boston Kidney biopsy cohort study. J Am Soc Nephrol. 2018;29:2213–2224. doi: 10.1681/ASN.2017121260. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10.Casal Moura M., Fervenza F.C., Specks U., Sethi S. Kidney biopsy chronicity grading in antineutrophil cytoplasmic antibody-associated vasculitis. Nephrol Dial Transplant. 2022;37:1710–1721. doi: 10.1093/ndt/gfab250. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Aringer M., Costenbader K., Daikh D., et al. 2019 European League Against Rheumatism/American College of Rheumatology classification criteria for systemic lupus erythematosus. Arthritis Rheumatol. 2019;71:1400–1412. doi: 10.1002/art.40930. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.Fanouriakis A., Kostopoulou M., Cheema K., et al. 2019 Update of the Joint European League Against Rheumatism and European Renal Association-European Dialysis and Transplant Association (EULAR/ERA-EDTA) recommendations for the management of lupus nephritis. Ann Rheum Dis. 2020;79:713–723. doi: 10.1136/annrheumdis-2020-216924. [DOI] [PubMed] [Google Scholar]
- 13.Nakagawa S., Toyama T., Iwata Y., et al. The relationship between the modified National Institute of Health activity and chronicity scoring system, and the long-term prognosis for lupus nephritis: a retrospective single-center study. Lupus. 2021;30:1739–1746. doi: 10.1177/09612033211034234. [DOI] [PubMed] [Google Scholar]
- 14.Leatherwood C., Speyer C.B., Feldman C.H., et al. Clinical characteristics and renal prognosis associated with interstitial fibrosis and tubular atrophy (IFTA) and vascular injury in lupus nephritis biopsies. Semin Arthritis Rheum. 2019;49:396–404. doi: 10.1016/j.semarthrit.2019.06.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 15.Gomes M.F., Mardones C., Xipell M., et al. The extent of tubulointerstitial inflammation is an independent predictor of renal survival in lupus nephritis. J Nephrol. 2021;34:1897–1905. doi: 10.1007/s40620-021-01007-z. [DOI] [PubMed] [Google Scholar]
- 16.Shao D., Jimenez A.L., Guerrero M.S., Wang S., Broder A. Factors associated with worsening interstitial fibrosis/tubular atrophy in lupus nephritis patients undergoing clinically indicated repeat kidney biopsy. BMC Nephrol. 2025;26:189. doi: 10.1186/s12882-025-04108-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Malvar A., Alberton V., Lococo B., et al. The evolution of crescents in lupus nephritis. Kidney Int Rep. 2025;10:1579–1581. doi: 10.1016/j.ekir.2025.02.009. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Barbour S.J., Coppo R., Er L., et al. Histologic and clinical factors associated with kidney outcomes in IgA vasculitis nephritis. Clin J Am Soc Nephrol. 2024;19:438–451. doi: 10.2215/CJN.0000000000000398. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Tinajero-Sánchez D.N., Zúñiga-González E.Y., Zavala-Miranda M.F., et al. Histologic predictors of kidney outcomes in lupus nephritis: reevaluating the role of segmental glomerulosclerosis in the chronicity index. Rheumatology (Oxford) 2025;64:4787–4795. doi: 10.1093/rheumatology/keaf194. [DOI] [PubMed] [Google Scholar]
- 20.Rijnink E.C., Teng Y.K.O., Wilhelmus S., et al. Clinical and histopathologic characteristics associated with renal outcomes in lupus nephritis. Clin J Am Soc Nephrol. 2017;12:734–743. doi: 10.2215/CJN.10601016. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Ramasamy A., Mohan C. Molecular and cellular mediators of renal fibrosis in lupus nephritis. Int J Mol Sci. 2025;26:2621. doi: 10.3390/ijms26062621. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Dasari S., Chakraborty A., Truong L., Mohan C. A systematic review of interpathologist agreement in histologic classification of lupus nephritis. Kidney Int Rep. 2019;4:1420–1425. doi: 10.1016/j.ekir.2019.06.011. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23.Grootscholten C., Bajema I.M., Florquin S., et al. Interobserver agreement of scoring of histopathological characteristics and classification of lupus nephritis. Nephrol Dial Transplant. 2008;23:223–230. doi: 10.1093/ndt/gfm555. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Arteriosclerosis Assessment. Figure S1. Relationship between NIH chronicity score and MCCS grade. Table S1. Analysis by each component of the chronicity scores and its contribution to proteinuria and clinical outcomes after univariable and multivariable adjustments using categorical variables.
Data Availability Statement
Data is available upon reasonable request from the corresponding author.