ABSTRACT
Background
End-stage kidney disease (ESKD) from lupus nephritis (LN) is a major cause of morbidity and mortality in patients with systemic lupus erythematosus (SLE). Kidney biopsy is the gold standard for diagnosis and prognostication of LN. While interstitial fibrosis and tubular atrophy (IFTA) predict progression to ESKD, the National Institutes of Health (NIH) classification of interstitial inflammation in unscarred cortical parenchyma is not predictive of chronic kidney disease (CKD) progression. The objective of this study was to determine whether total cortical interstitial inflammation that accounts for inflammation in the entire cortical parenchyma could predict CKD progression in patients with LN. Early identification of at-risk patients may improve outcomes.
Methods
This retrospective cohort study included 125 SLE patients with LN class III, IV, V or mixed (III/V, IV/V) on the index biopsy (2005–2018). Kidney biopsies were reviewed and assigned based on the 2018 NIH Activity Index (AI) and tubulointerstitial lesion categories. Total interstitial inflammation in the entire cortical parenchyma was graded as 0, 1, 2 or 3, corresponding to <10%, 10–25%, 26–50% and >50%, respectively, of the total cortical parenchyma containing an inflammatory infiltrate (similar to the definition used in the Banff total inflammation score). CKD progression was defined as an estimated glomerular filtration rate decrease of ≥30% within 5 years after the index biopsy. Kaplan–Meier survival curves and Cox proportional hazards models were performed to compare the two scoring systems, the total cortical intestinal inflammation score and the NIH interstitial inflammation score as predictors of CKD progression.
Results
Of 125 patients, 46 experienced CKD progression; 21 of 46 subsequently developed ESKD, 28 (22.4%) had moderate–severe total cortical interstitial inflammation and 8 (6.4%) had moderate–severe NIH interstitial inflammation. There were no differences in baseline characteristics between progressors and nonprogressors. Total cortical interstitial inflammation was associated with CKD progression in time-dependent analyses [hazard ratio 2.45 (95% confidence interval 1.2–4.97)] adjusted for age at biopsy, race, sex, LN class and hypertensive vascular change on kidney biopsy. The NIH interstitial inflammation was not associated with CKD progression.
Conclusions
In contrast to the current NIH interstitial inflammation classification, accounting for interstitial inflammation in the entire cortical parenchyma allows identification of patients at risk for CKD progression in LN.
Keywords: chronic kidney disease, CKD progression, interstitial inflammation, lupus nephritis, SLE
Graphical Abstract
Graphical Abstract.
KEY LEARNING POINTS.
What is already known about this subject?
Current National Institutes of Health (NIH) interstitial inflammation score in unscarred cortical parenchyma does not predict progression to CKD.
What this study adds?
Total cortical interstitial inflammation that accounts for inflammation in the entire cortical parenchyma with and without scarring (similar to the Banff total inflammation score) predicts CKD progression in LN.
What impact this may have on practice or policy?
This study underscored the need to modify the existing NIH definition of interstitial inflammation to account for inflammation in areas with and without fibrosis. This modification may help identify and treat at-risk patients before irreversible chronic changes develop.
INTRODUCTION
Lupus nephritis (LN) leads to end-stage kidney disease (ESKD) in up to 30% of pediatric and adult SLE patients [1–3]. Interstitial fibrosis and tubular atrophy (IFTA) have been shown to predict progression to chronic kidney disease (CKD) independent of the glomerular lesions [4–10]. Although IFTA cannot be reversed with the current available therapies [4–10], the presence of interstitial inflammation in LN is associated with a favorable response to immunosuppressive therapy [11]. Because persistent interstitial inflammation has been shown to lead to irreversible fibrosis, it is important to identify interstitial inflammation early to implement interventions to prevent progression to CKD and ESKD [4, 11–13].
The need for better classification of interstitial inflammation in LN is highlighted in the consensus report on the revisions to the current International Society of Nephrology/Renal Pathology Society (ISN/RPS) classification [14]. Interstitial inflammation defined by the National Institutes of Health (NIH) activity index (AI) [14, 15] does not predict progression to CKD or ESKD in patients with LN [6, 8, 16, 17]. The limitation of this score is that it only focuses on assessing interstitial inflammation in areas without fibrosis and therefore misses interstitial inflammation in areas with IFTA [14]. In contrast, total cortical interstitial inflammation [defined as the total inflammation (ti) score in the current Banff classification for kidney allograft pathology], evaluates the extent of total interstitial inflammation in cortical parenchyma with and without fibrosis [18]. The Banff total interstitial inflammation score predicts CKD progression of kidney allografts, while the Banff interstitial inflammation score (similar to the NIH interstitial inflammation score) is not predictive of CKD progression [18, 19].
The objective of this study was to address the ISN/RPS consensus recommendation to assess interstitial lesions in the entire cortical area for prognostic significance by using the total interstitial inflammation score, similarly to the ti score used in Banff scoring system [14]. We hypothesized that total interstitial inflammation may predict CKD progression in LN. Using the total interstitial inflammation score may identify patients at risk for progression and thus lead to more aggressive therapy and better control of the disease outcome.
MATERIALS AND METHODS
Study participants
This was a retrospective cohort study conducted at Montefiore Medical Center. Adult and pediatric patients who fulfilled either the American College of Rheumatology criteria or the Systemic Lupus International Collaborating Clinics classification criteria [20–22] and had undergone a percutaneous renal biopsy between 1 January 2005 and 31 December 2018 with LN class III, IV, V or mixed (III/V, IV/V) were included. Kidney biopsies were evaluated in accordance with the 2003 ISN/RPS classification [23]. The study was approved by the Einstein College of Medicine Institutional Review Board. For SLE patients with multiple biopsies during this period, only the index biopsy was included. The index biopsy was defined as the initial kidney biopsy verifying the first episode of LN class III, IV, V or mixed (III/V, IV/V). We excluded LN patients who did not have sufficient postbiopsy follow-up clinical data and those patients who had kidney biopsies with poor tissue quality or samples containing <10 glomeruli.
Renal pathology
All hematoxylin and eosin (H&E) and/or periodic acid–Schiff (PAS)-stained kidney biopsy slides were reviewed by an experienced renal pathologist (D.S.) who assigned the overall modified 2018 NIH AI tubulointerstitial lesion categories to all kidney biopsies included in the study. The NIH AI (total score of 24) included glomerular lesions (endocapillary hypercellularity, neutrophils or karyorrhexis, fibrinoid necrosis, hyaline deposits, cellular or fibrocellular crescents) and interstitial inflammation. Scoring was based on the percentages of glomeruli or cortex lesions: 0, none; 1, <25%; 2, 25–50%; 3, >50%. The NIH AI was analyzed as <11 versus ≥11 [14]. The NIH interstitial inflammation, defined as inflammation with interstitial leukocytes in unscarred cortical parenchyma, was graded as none/mild, moderate and severe, corresponding to <25%, 25–50% and >50% inflammation in unscarred cortical parenchyma and was analyzed as none–mild versus moderate–severe) [14]. The total cortical interstitial inflammation in the entire cortical parenchyma (analogous to the Banff ti score in the Banff classification of kidney allograft pathology) was graded as 0, 1, 2 and 3 corresponding to <10%, 10–25%, 26–50% and >50% of the total cortical parenchyma containing an inflammatory infiltrate and analyzed as none–mild (0/1) versus moderate–severe (2/3) [18].
In addition, inflammation in the area of IFTA (i-IFTA) was graded as 0, 1, 2 and 3, corresponding to <10%, 10–25%, 26–50% and >50% inflammation in scarred cortical parenchyma, and analyzed as none–mild (0/1) versus moderate–severe (2/3) [18]. Tubulitis was defined by the number of mononuclear leukocytes per tubular cross section (or 10 tubular epithelial cells) in nonatrophic tubules [18]. Additional results of kidney biopsy, including thrombotic microangiopathy (TMA), antiphospholipid syndrome (ASP) nephropathy, hypertensive vascular changes and diabetic nephropathy were collected.
Data collection
A retrospective chart review was performed to collect the following demographic, clinical and laboratory parameters: age, sex, self-reported race/ethnicity, total and renal SLE disease activity index (SLEDAI) scores [24], history of hypertension and diabetes, medications, estimated glomerular filtration rate (eGFR), serum complement levels (C3 and C4), anti-dsDNA antibodies, urinalysis and urine protein:creatinine ratio (UPCR) at the time of biopsy or within 30 days of biopsy. The eGFR calculation was based on the Modification of Diet in Renal Disease (MDRD) equation for patients ≥18 years of age [25] and the bedside Schwartz equation for patients <18 years of age [26]. The baseline eGFR was calculated using serum creatinine at the index biopsy or within 1 month of the kidney biopsy date if serum creatinine was not available at the time of biopsy. The time from clinical LN onset to the index biopsy was defined as the time of the first documented abnormal proteinuria (UPCR >0.2 mg/mg with or without hematuria) to the index biopsy.
Outcome and exposure definitions
The primary study outcome, CKD progression, was defined as a decrease in eGFR of 30% within up to 5 years after the index biopsy in LN patients. This outcome was chosen because a decrease in eGFR of ≥30% is strongly associated with progression to ESKD and mortality [27, 28]. It is widely used as an outcome in clinical trials [27, 28]. Furthermore, using CKD progression as an outcome provides a window of opportunity for treatment and prevention of progression to ESKD. LN patients were defined as progressors if they developed the outcome within 5 years from the index biopsy. Patients whose eGFR remained stable during the 5-year study period were defined as nonprogressors. We chose the maximum time to outcome to be 5 years, as it would be difficult to attribute the outcome to biopsy findings beyond 5 years due to potential confounding factors.
The modified NIH AI, NIH interstitial inflammation, total cortical interstitial inflammation and i-IFTA were evaluated as predictors of an eGFR decrease of ≥30% within 5 years.
Statistical analyses
Baseline characteristics were compared between CKD progressors and nonprogressors. Wilcoxon–Mann–Whitney tests were used to compare continuous variables. Pearson's chi-squared test (or Fisher's exact test when appropriate) was used to compare categorical variables.
In time-dependent analyses, time to the outcome was defined as the time from the index biopsy date to the first recorded 30% decrease in eGFR. Nonprogressors were censored at the time of the last documented visit or at the end of 5 years.
Survival analyses, including Kaplan–Meier survival curves and Cox proportional hazards models, were used to evaluate estimated hazard ratios (HRs) and to adjust for potential confounders. Multiple models were adjusted for age at biopsy, sex and race. Additional variables were considered if there were statistically significant differences in the univariable comparisons. Sensitivity analyses were performed including the LN class [proliferative or mixed glomerulonephritis (GN) versus nonproliferative GN] and hypertensive vascular changes on kidney biopsy as additional variables in the multivariable models. Additional sensitivity analyses were performed with ESKD as the outcome.
RESULTS
A total of 147 renal biopsies with LN class III, IV, V or mixed (III/V or IV/V) performed between 1 January 2005 and 31 December 2018 were available for review. We excluded 15 repeat biopsies and 7 patients with missing clinical follow-up data, leaving 125 first kidney biopsies for analysis.
Of the 125 patients, 89 (71.2%) had proliferative or mixed GN (LN class III/V or IV/V) and 48 (38.4%) had pure proliferative GN (LN class III or IV). The median follow-up time from the index biopsy to the end of the follow-up period was 35.7 months [interquartile range (IQR) 13.4–55.7]. The median eGFR at baseline was 91.1 ml/min/1.73 m2 (IQR 53–117). Ninety-two (73.6%) patients had a baseline eGFR ≥60 ml/min/1.73 m2. At the time of renal biopsy, 108 (86.4%) patients were taking corticosteroids, 30 (24%) were taking mycophenolate mofetil and 87 (69.6%) were taking hydroxychloroquine (Table 1).
Table 1:
Baseline characteristics of LN patients with and without CKD progression.
| Characteristics | Total (N = 125) | Progressorsa (n = 46) | Nonprogressors (n = 79) | P-value |
|---|---|---|---|---|
| Demographic data | ||||
| Female, n (%) | 105 (84.0) | 36 (78.3) | 69 (87.3) | .18 |
| Age at SLE diagnosis (years), median (IQR) | 23 (16–34) | 22 (16–34) | 25 (16–38) | .35 |
| Age at SLE <18 years, n (%) | 39 (31.2) | 16 (34.8) | 23 (29.1) | .51 |
| Age at renal biopsy (years), median (IQR) | 28 (20–43) | 26 (20–41) | 29 (21–43) | .53 |
| Age at renal biopsy <18 years, n (%) | 20 (16.0) | 7 (15.2) | 13 (16.5) | .86 |
| Time from SLE diagnosis to biopsy (years), median (IQR) | 2 (0–5) | 3 (1–7) | 1 (0–5) | .15 |
| Time from diagnosis of LN to index biopsy (months), median (IQR) | 1.9 (0.7–4.8) | 2.5 (0.7–4.9) | 1.7 (0.6–4.8) | .49 |
| Follow-up time (months), median (IQR) | 35.7 (13.4–55.7) | 13.1 (4.8–35.7) | 49.2 (24.4–58.9) | <.001 |
| Race, n (%) | .82 | |||
| White | 8 (6.4) | 4 (8.7) | 4 (5.1) | .43 |
| Black | 53 (42.4) | 18 (39.1) | 35 (44.3) | .54 |
| Asian | 2 (1.6) | 1 (2.2) | 1 (1.2) | .7 |
| Other/unknown | 62 (49.6) | 23 (50.0) | 39 (49.4) | .92 |
| Ethnicity, n (%) | .73 | |||
| Hispanic | 51 (40.8) | 19 (41.3) | 32 (40.5) | .93 |
| Non-Hispanic | 61 (48.8) | 21 (45.7) | 40 (50.6) | .59 |
| Unknown | 13 (10.4) | 6 (13.0) | 7 (8.9) | .56 |
| Clinical data | ||||
| Diabetes, n (%) | 8 (6.4) | 5 (10.9) | 3 (3.8) | .12 |
| Hypertension, n (%) | 88 (70.4) | 33 (71.7) | 55 (69.6) | .8 |
| Baseline eGFR (ml/min/1.73 m2), median (IQR) | 91.1 (53–117) | 96.1 (56.6–117.8) | 88.9 (47–117) | .51 |
| C3 (mg/dl), median (IQR)b | 68 (45–92) | 71 (52–88) | 67.5 (43–94) | .58 |
| C4 (mg/dl), median (IQR)b | 12 (8–21) | 12.65 (9–23) | 12 (7–20) | .48 |
| Anti-dsDNA titer (IU), median (IQR)c | 143.4 (38.5–200) | 108.7 (36.6–194.9) | 151.3 (39.2–200) | .57 |
| Total SLEDAI score, median (IQR) | 12 (8–16) | 11 (8–16) | 12 (8–16) | .47 |
| Renal SLEDAI score, median (IQR) | 8 (4–12) | 8 (4–12) | 8 (4–12) | .9 |
| Urine cast, n (%) | 7 (5.6) | 5 (10.9) | 2 (2.5) | .05 |
| Hematuria, n (%) | 72 (57.6) | 25 (54.3) | 47 (59.5) | .58 |
| Pyuria, n (%) | 70 (56.0) | 26 (56.5) | 44 (55.7) | .9 |
| UPCR (mg/mg), median (IQR) | 1.6 (0.9–4.3) | 1.5 (0.8–4.9) | 1.6 (0.96–4.3) | .86 |
| LN class, n (%) | ||||
| Proliferative or mixed GN (LN class III/V or IV/V) | 89 (71.2) | 37 (80.4) | 52 (65.8) | .08 |
| Nonproliferative GN (LN class V) | 36 (28.8) | 9 (19.6) | 27 (34.2) | |
| Hypertensive vascular change on kidney biopsy, n (%) | 41 (32.8) | 16 (34.8) | 25 (31.7) | .72 |
| Treatment at biopsy, n (%) | ||||
| Acute hemodialysis | 4 (3.2) | 3 (6.5) | 1 (1.3) | .11 |
| Corticosteroids | 108 (86.4) | 42 (91.3) | 66 (83.5) | .22 |
| Mycophenolate mofetil | 30 (24.0) | 15 (32.6) | 15 (19.0) | .09 |
| Azathioprine | 8 (6.4) | 3 (6.5) | 5 (6.3) | .9 |
| Hydroxychloroquine | 87 (69.6) | 31 (67.4) | 56 (70.9) | .68 |
| ACE inhibitors/ARBs | 51 (40.8) | 21 (45.7) | 30 (38.0) | .4 |
| Additional antihypertensive medications | 56 (44.8) | 21 (45.6) | 35 (44.3) | .88 |
CKD progressors were defined as LN patients with an eGFR decrease ≥30% within 5 years after the index biopsy.
bLow serum C3 and C4 are <80 and 20 mg/dl, respectively.
cElevated anti-dsDNA antibody is >70 IU.
ACE, angiotensin-converting enzyme; ARB, angiotensin receptor blockers.
Comparison of baseline characteristics of LN patients with and without CKD progression
Of the 125 patients, 46 (36.8%) progressed to CKD and 21 of 46 (45.65%) subsequently developed ESKD within 5 years. Comparisons of baseline characteristics between CKD progressors and nonprogressors is summarized in Table 1. There were no significant differences between CKD progressors and nonprogressors in terms of age, sex, race, ethnicity, hypertension, comorbidities, total or renal SLEDAI scores, baseline eGFR, serum C3 and C4 levels, anti-dsDNA titers, UPCR, LN class with proliferative or mixed GN versus nonproliferative GN or hypertensive vascular change on biopsy. Forty-one (32.8%) patients had hypertensive vascular changes on kidney biopsies. The time from SLE diagnosis to the index biopsy was comparable between CKD progressors and nonprogressors [3 years (IQR 1–7) versus 1 year (IQR 0–5); P = .15]. The time from LN onset to the index biopsy was similar between the two groups [median 2.5 months (IQR 0.7–4.9) for progressors versus 1.7 months (IQR 0.6–4.8) nonprogressors; P = .49] (Table 1). The median follow-up was 13.1 months (IQR 4.8–35.7) for progressors and 49.2 months (IQR 24.4–58.9) for nonprogressors (P < .001) (Table 1).
At the time of the index kidney biopsy, 108 (86.4%) patients with LN were treated with immunosuppressive medications including mycophenolate mofetil or azathioprine and/or steroids (Table 1). There were no differences in the frequency of these medications between progressors and nonprogressors.
Association of interstitial inflammation scores and the modified NIH AI with CKD progression
Of the 125 patients, 28 (22.4%) had moderate–severe total cortical interstitial inflammation. Moderate–severe total cortical interstitial inflammation was found in 16 (34.8%) progressors versus 12 (15.2%) nonprogressors (P = .01) (Table 2). In the Kaplan–Meier analysis (Fig. 1) and in the Cox proportional hazards model, the total cortical interstitial inflammation score was associated with an increased risk of CKD progression adjusted for age at biopsy, sex and race. The conclusions remained the same when the models were further adjusted for LN class and hypertensive vascular change on kidney biopsy {HR 2.45 [95% confidence interval (CI) 1.2–4.97]} (Table 3).
Table 2:
Modified NIH AI and interstitial inflammation categories in LN patients with and without CKD progression.
| Categories | Total (n = 125) | Progressorsa (n = 46) | Nonprogressors (n = 79) | P-value |
|---|---|---|---|---|
| Overall NIH AI, median (IQR) | 1 (0–3) | 1 (0–4) | 1 (0–3) | .61 |
| NIH AI score ≥11, n (%) | 4 (3.2) | 1 (2.2) | 3 (3.8) | .62 |
| NIH interstitial inflammation, n (%) | .9b | |||
| None–mild | 117 (93.6) | 43 (93.5) | 74 (93.7) | |
| Moderate–severe | 8 (6.4) | 3 (6.5) | 5 (6.3) | |
| Total cortical interstitial inflammation (%) | .01b | |||
| None–mild | 97 (77.6) | 30 (65.2) | 67 (84.8) | |
| Moderate–severe | 28 (22.4) | 16 (34.8) | 12 (15.2) | |
| i-IFTA | N = 77c | n = 26 | n = 51 | .09b |
| None–mild | 34 (44.2) | 8 (30.8) | 26 (51.0) | |
| Moderate–severe | 43 (55.8) | 18 (69.2) | 25 (49.0) |
CKD progressors were defined as LN patients with an eGFR decrease ≥30% within 5 years after the index biopsy.
b P-values for none–mild versus moderate–severe scores between progressors and nonprogressors.
cTotal number of biopsies is smaller due to an inability to apply scores to biopsies without areas of IFTA.
Figure 1:
Kaplan–Meier analysis of kidney survival versus total cortical interstitial inflammation score (moderate–severe versus none–mild). Kidney survival was defined as not reaching the CKD progression end point as a decrease in eGFR ≥30% within 5 years after the index biopsy. ti score, total cortical interstitial inflammation score.
Table 3:
HRs for developing CKD progression in LN patients with NIH AI and interstitial inflammation categories.
| Categories | Unadjusted HR(95% CI) | P-value | Adjusted HRa (95% CI) | P-value |
|---|---|---|---|---|
| NIH AI (cut-off ≥11 versus <11) | 0.59 (0.08–4.29) | .6 | 0.64 (0.08–4.84) | .67 |
| NIH interstitial inflammation (moderate-to-severe versus none–mild) | 1.30 (0.40–4.20) | .67 | 0.79 (0.18–3.48) | .76 |
| Total cortical interstitial inflammation (moderate–severe versus none–mild) | 2.96 (1.61–5.48) | .001 | 2.45 (1.2–4.97) | .01 |
| Inflammation in areas with IFTA (moderate–severe versus none–mild) | 2.11 (0.91–4.86) | .08 | 2.04 (0.81–5.16) | .13 |
CKD progression was defined as LN patients with an eGFR decrease ≥30% within 5 years after the index biopsy.
aAdjusted for age at kidney biopsy, sex, race, LN class (proliferative or mixed GN versus nonproliferative GN) and hypertensive vascular change on kidney biopsy.
In contrast, NIH interstitial inflammation was not associated with CKD progression. In these patients, 3 (6.5%) progressors versus 5 (6.3%) nonprogressors had moderate–severe NIH interstitial inflammation. In the time-dependent analysis, NIH interstitial inflammation did not predict CKD progression in the Cox models (Table 3 and Fig. 2).
Figure 2:
Kaplan–Meier analysis of kidney survival versus NIH interstitial inflammation score (moderate–severe versus none–mild). Kidney survival was defined as not reaching the CKD progression endpoint as a decrease in eGFR ≥30% within 5 years after the index biopsy. NIH i score, NIH interstitial inflammation score.
The association between moderate–severe i-IFTA and CKD progression was shown in the univariate analysis [18 (69.2%) progressors versus 25 (49%) nonprogressors; P = .09] (Table 2). The association was also borderline statistically significant in the multivariable analyses. The unadjusted HR in the Cox model was 2.11 (95% CI 0.91–4.86; P = .08) (Table 3). There was no evidence of tubulitis on any of the biopsies.
The median NIH AI score was 1 (IQR 0–4) in progressors and 1 (IQR 0–3) in nonprogressors (P = .61). The number of patients with an NIH AI score ≥11 was small among progressors and nonprogressors: 1 (2.2%) and 3 (3.8%), respectively (Table 2). An NIH AI score ≥11 was not associated with CKD progression (Table 3 and Fig. 3).
Figure 3:
Kaplan–Meier analysis of kidney survival versus NIH AI (≥11 versus <11). Kidney survival was defined as not reaching the CKD progression endpoint as a decrease in eGFR ≥30% within 5 years after the index biopsy.
The total cortical interstitial inflammation score and i-IFTA were associated with progression to ESKD in unadjusted and adjusted analyses (Supplementary Table 1). NIH interstitial inflammation was not associated with ESKD progression. In the subset of patients with a baseline eGFR ≥60 ml/min/1.73 m2, total cortical inflammation was also associated with the development of ESKD.
DISCUSSION
This study underscores that identifying interstitial inflammation in the cortical areas with and without scars (similar to the Banff score for kidney allograft rejection) in LN is predictive of CKD progression. Consistent with prior studies, neither the AI score nor the interstitial inflammation score as defined in the current NIH classification were predictive of CKD progression [4, 6–8, 16, 17, 29]. This study addresses the need to revise the current NIH classification of interstitial inflammation in LN as outlined in the ISN/RPS consensus document [14]. To the best of our knowledge, this is the first study to demonstrate the association of total cortical interstitial inflammation in LN with CKD progression and ESKD development. The study also shows that i-IFTA may be associated with a higher risk of ESKD development in patients with LN. Untreated interstitial inflammation in any part of the tubular and interstitial areas leads to tubular atrophy and interstitial fibrosis, an irreversible process that results in the loss of kidney function and development of ESKD in patients with LN [11, 30, 31]. Based on studies assessing the prognosis of kidney allograft function, the total cortical interstitial inflammation score (including inflammation in areas with IFTA) is superior to interstitial inflammation in areas without scars in predicting kidney allograft survival [18, 19]. Inflammation in IFTAs still responds to immunosuppressive therapy, thus delaying the progression of fibrosis [32].
Prior studies showed tubulointerstitial inflammation in repeat biopsies predicts poor long-term kidney function outcome [33]. However, most LN patients had end-stage kidney fibrosis by the time of repeat biopsy. The strength of our study is that we evaluated the first index biopsy, which provides a window for early intervention to improve and prevent renal outcome prior to progression to ESKD.
This study has several limitations due to the retrospective nature of its design and a relatively small number of outcomes in some analyses. The study may have been underpowered to detect i-IFTA differences between progressors and nonprogressors. However, i-IFTA was a predictor of ESKD in adjusted and unadjusted analyses, suggesting that inflammation in the areas of fibrosis is a potential predictor of CKD/ESKD progression.
Kidney outcomes in LN patients depend on multiple variables, including access to care, treatment adherence, acute kidney injury episodes and LN flares [3, 34–37]. We could not account for the effects of these potential confounders in this retrospective study. There were three kidney biopsies showing TMA and one biopsy showing diabetic nephropathy. None of the biopsies showed APS nephropathy. Therefore we could not adjust for these findings in our analyses.
We could not determine the effect of medication use on the outcome due to a lack of data related to medication duration and adherence. However, the main objective of our study was to compare two inflammation scores—total cortical inflammation and NIH interstitial inflammation—which are unlikely to be differentially affected by the background variables, including medication use and pre-LN eGFR.
This study has several important strengths. It includes a diverse patient population with well clinically phenotyped SLE and with extensive clinical follow-up. The biopsies were reviewed and scored by an experienced nephropathologist (D.S.) who was blinded to the study outcome.
In conclusion, the assessment of interstitial inflammation in the entire cortex may be a valuable tool to predict CKD progression in LN.
Supplementary Material
Contributor Information
Minh Dien Duong, Department of Pediatrics, Nephrology, Children's Hospital at Montefiore, Albert Einstein College of Medicine, Bronx, NY, USA.
Shudan Wang, Department of Medicine, Rheumatology, Montefiore Medical Center, Albert Einstein College of Medicine, Bronx, NY, USA.
Daniel Schwartz, Department of Surgical Pathology, Montefiore Medical Center, Albert Einstein College of Medicine, Bronx, NY, USA.
Wenzhu B Mowrey, Department of Epidemiology and Biostatistics, Albert Einstein College of Medicine, Bronx, NY, USA.
Anna Broder, Department of Medicine, Rheumatology, Hackensack University Medical Center, Hackensack, NJ, USA.
Beatrice Goilav, Department of Pediatrics, Nephrology, Children's Hospital at Montefiore, Albert Einstein College of Medicine, Bronx, NY, USA.
ACKNOWLEDGEMENTS
B.G. is grateful for the invaluable input from Prof. P.M. Saumon.
FUNDING
This research was supported by NIH/National Center for Advancing Translational Science Einstein–Montefiore CTSA grant KL2 TR002558 (to S.W.) and NIH/NIAMS K23 AR068441 (to A.B.). The NIH provides K grants to support S.W. and A.B. in their career development and was not involved in the study design, data analysis or writing of the manuscript.
AUTHORS’ CONTRIBUTIONS
M.D.D., D.S. and B.G. created the research idea and study design. M.D.D., D.S. and A.B. were responsible for data acquisition. A.B., W.B.M. and S.W. were responsible for data analysis and statistical analysis. All authors reviewed and approved the final manuscript. A.B. and B.G. are co-senior authors. M.D.D. and S.W. are co-first authors.
DATA AVAILABILITY STATEMENT
The data underlying this article cannot be shared publicly to protect the study participants’ privacy. The datasets generated and analyzed during the current study are available from the corresponding author upon request.
CONFLICT OF INTEREST STATEMENT
The authors declare that they have no competing interests.
REFERENCES
- 1. Mina R, Brunner HI.. Pediatric lupus—are there differences in presentation, genetics, response to therapy, and damage accrual compared with adult lupus? Rheum Dis Clin North Am 2010;36:53–80. 10.1016/j.rdc.2009.12.012. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 2. Brunner HI, Gladman DD, Ibanez Det al. Difference in disease features between childhood-onset and adult-onset systemic lupus erythematosus. Arthritis Rheum 2008;58:556–62. 10.1002/art.23204. [DOI] [PubMed] [Google Scholar]
- 3. Maroz N, Segal MS.. Lupus nephritis and end-stage kidney disease. Am J Med Sci 2013;346:319–23. 10.1097/MAJ.0b013e31827f4ee3. [DOI] [PubMed] [Google Scholar]
- 4. Hsieh C, Chang A, Brandt Det al. Predicting outcomes of lupus nephritis with tubulointerstitial inflammation and scarring. Arthritis Care Res 2011;63:865–74. 10.1002/acr.20441. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5. Yu F, Wu LH, Tan Yet al. Tubulointerstitial lesions of patients with lupus nephritis classified by the 2003 International Society of Nephrology and Renal Pathology Society system. Kidney Int 2010;77:820–9. 10.1038/ki.2010.13. [DOI] [PubMed] [Google Scholar]
- 6. Wu CY, Chien HP, Yang HYet al. Role of tubulointerstitial lesions on predicting renal outcome among pediatric onset lupus nephritis – a retrospective cohort study. J Microbiol Immunol Infect 2020;53:33–41. 10.1016/j.jmii.2017.11.003. [DOI] [PubMed] [Google Scholar]
- 7. Londoño Jimenez A, Mowrey WB, Putterman Cet al. Brief report: tubulointerstitial damage on lupus nephritis: a comparison of the factors associated with tubulointerstitial inflammation and renal scarring. Arthritis Rheumatol 2018;70:1801–6. 10.1002/art.40575. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 8. Broder A, Mowrey WB, Khan HNet al. Tubulointerstitial damage predicts end stage renal disease in lupus nephritis with preserved to moderately impaired renal function: a retrospective cohort study. Semin Arthritis Rheum 2018;47:545–51. 10.1016/j.semarthrit.2017.07.007. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 9. Wilson PC, Kashgarian M, Moeckel G.. Interstitial inflammation and interstitial fibrosis and tubular atrophy predict renal survival in lupus nephritis. Clin Kidney J 2018;11:207–18. 10.1093/ckj/sfx093. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 10. Leatherwood C, Speyer CB, Feldman CHet 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. 10.1016/j.semarthrit.2019.06.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11. Lan-Ting H, You-Ming C, Li-Xin Wet al. Clinicopathological factors for tubulointerstitial injury in lupus nephritis. Clin Rheumatol 2020;39:1617–26. 10.1007/s10067-019-04909-3. [DOI] [PubMed] [Google Scholar]
- 12. Chang A, Henderson SG, Brandt Det al. In situ B cell-mediated immune responses and tubulointerstitial inflammation in human lupus nephritis. J Immunol 2011;186:1849–60. 10.4049/jimmunol.1001983. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13. Trotter K, Clark MR, Liarski VM.. Overview of pathophysiology and treatment of human lupus nephritis. Curr Opin Rheumatol 2016;28:460–7. 10.1097/BOR.0000000000000319. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 14. Bajema IM, Wilhelmus S, Alpers CEet 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;93:789–96. 10.1016/j.kint.2017.11.023. [DOI] [PubMed] [Google Scholar]
- 15. Austin HA 3rd, Muenz LR, Joyce KMet al. Prognostic factors in lupus nephritis. Contribution of renal histologic data. Am J Med 1983;75:382–91. 10.1016/0002-9343(83)90338-8. [DOI] [PubMed] [Google Scholar]
- 16. Alsuwaida AO. Interstitial inflammation and long-term renal outcomes in lupus nephritis. Lupus 2013;22:1446–54. 10.1177/0961203313507986. [DOI] [PubMed] [Google Scholar]
- 17. Park DJ, Kang JH, Lee JWet al. Risk factors to predict the development of chronic kidney disease in patients with lupus nephritis. Lupus 2017;26:1139–48. 10.1177/0961203317694257. [DOI] [PubMed] [Google Scholar]
- 18. Roufosse C, Simmonds N, Clahsen-van Groningen Met al. A 2018 reference guide to the Banff classification of renal allograft pathology. Transplantation 2018;102:1795–814. 10.1097/TP.0000000000002366. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19. Mengel M, Reeve J, Bunnag Set al. Scoring total inflammation is superior to the current Banff inflammation score in prediction outcome and the degree of molecular disturbance in renal allografts. Am J Transplant 2009;9:1859–67. 10.1111/j.1600-6143.2009.02727.x. [DOI] [PubMed] [Google Scholar]
- 20. Tan EM, Cohen AS, Fries JFet al. The 1982 revised criteria for the classification of systemic lupus erythematosus. Arthritis Rheum 1982;25:1271–7. 10.1002/art.1780251101. [DOI] [PubMed] [Google Scholar]
- 21. Hochberg MC. Updating the American College of Rheumatology revised criteria for classification of systemic lupus erythematosus. Arthritis Rheum 1997;40:1725. 10.1002/art.1780400928. [DOI] [PubMed] [Google Scholar]
- 22. Petri M, Orbai AM, Alarcon GSet al. Derivation and validation of the systemic lupus international collaborating clinics classification criteria for systemic lupus erythematosus. Arthritis Rheum 2012;64:2677–86. 10.1002/art.34473. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 23. Weening JJ, D'Agati VD, Schwartz MMet al. The classification of glomerulonephritis in systemic erythematosus lupus revisited. J Am Soc Nephrol 2004;15:241–50. 10.1097/01.ASN.0000108969.21691.5D. [DOI] [PubMed] [Google Scholar]
- 24. Bombardier C, Gladman DD, Urowitz MBet al. Derivation of the SLEDAI. A disease activity index for lupus patients. Arthritis Rheum 1992;35:630–40. 10.1002/art.1780350606. [DOI] [PubMed] [Google Scholar]
- 25. Levey AS, Bosch JP, Lewis JBet al. A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Ann Intern Med 1999;130:461–70. 10.7326/0003-4819-130-6-199903160-00002. [DOI] [PubMed] [Google Scholar]
- 26. Schwartz GJ, Work DF.. Measurement and estimation of GFR in children and adolescents. Clin J Am Soc Nephrol 2009;4:1832–43. 10.2215/CJN.01640309. [DOI] [PubMed] [Google Scholar]
- 27. Levey AS, Inker LA, Matsushita Ket al. GFR decline as an end point for clinical trials in CKD: a scientific workshop sponsored by the National Kidney Foundation and the US Food and Drug Administration. Am J Kidney Dis 2014;64:821–35. 10.1053/j.ajkd.2014.07.030. [DOI] [PubMed] [Google Scholar]
- 28. Coresh J, Turin TC, Matsushita Ket al. Decline in estimated glomerular filtration rate and subsequent risk of end-stage renal disease and mortality. JAMA 2014;311:2518–31. 10.1001/jama.2014.6634. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 29. Nakagawa S, Toyama T, Iwata Yet 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;30:1739–46. 10.1177/09612033211034234. [DOI] [PubMed] [Google Scholar]
- 30. Mejia-Vilet JM, Malvar A, Arazi Aet al. The lupus nephritis management renaissance. Kidney Int 2022;101:242–55. 10.1016/j.kint.2021.09.012. [DOI] [PubMed] [Google Scholar]
- 31. Arora S, Rovin BH.. Expert perspective: an approach to refractory lupus nephritis. Arthritis Rheumatol 2022;74:915–26. 10.1002/art.42092. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 32. Nankivell BJ, Shingde M, Keung KLet al. The causes, significance and consequences of inflammatory fibrosis in kidney transplantation: the Banff i-IFTA lesion. Am J Transplant 2018;18:364–76. 10.1111/ajt.14609. [DOI] [PubMed] [Google Scholar]
- 33. Parodis I, Adamichou C, Aydin Set al. Per-protocol repeat kidney biopsy portends relapse and long-term outcome in incident cases of proliferative lupus nephritis. Rheumatology (Oxford) 2020;59:3424–34. 10.1093/rheumatology/keaa129. [DOI] [PubMed] [Google Scholar]
- 34. Anders HJ, Rovin B.. A pathophysiology-based approach to the diagnosis and treatment of lupus nephritis. Kidney Int 2016;90:493–501. 10.1016/j.kint.2016.05.017. [DOI] [PubMed] [Google Scholar]
- 35. Parikh SV, Nagaraja HN, Hebert Let al. Renal flare as a predictor of incident and progressive CKD in patients with lupus nephritis. Clin J Am Soc Nephrol 2014;9:279–84. 10.2215/CJN.05040513. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 36. Almaani S, Meara A, Rovin BH.. Update on lupus nephritis. Clin J Am Soc Nephrol 2017;12:825–35. 10.2215/CJN.05780616. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 37. Anders HJ, Saxena R, Zhao MHet al. Lupus nephritis. Nat Rev Dis Primers 2020;6:7. 10.1038/s41572-019-0141-9. [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Supplementary Materials
Data Availability Statement
The data underlying this article cannot be shared publicly to protect the study participants’ privacy. The datasets generated and analyzed during the current study are available from the corresponding author upon request.