Visual Abstract
Keywords: CKD, chronic nephropathy, clinical immunology, clinical nephrology, interstitial fibrosis
Abstract
Background
IgG4-related kidney disease is a major manifestation of IgG4-related disease, a systemic fibroinflammatory disorder. However, the clinical and prognostic kidney-related factors in patients with IgG4-related kidney disease are insufficiently defined.
Methods
We conducted an observational cohort study using data from 35 sites in two European countries. Clinical, biologic, imaging, and histopathologic data; treatment modalities; and outcomes were collected from medical records. Logistic regression was performed to identify the possible factors related to an eGFR ≤30 ml/min per 1.73 m2 at the last follow-up. Cox proportional hazards model was performed to assess the factors associated with the risk of relapse.
Results
We studied 101 adult patients with IgG4-related disease with a median follow-up of 24 (11–58) months. Of these, 87 (86%) patients were male, and the median age was 68 (57–76) years. Eighty-three (82%) patients had IgG4-related kidney disease confirmed by kidney biopsy, with all biopsies showing tubulointerstitial involvement and 16 showing glomerular lesions. Ninety (89%) patients were treated with corticosteroids, and 18 (18%) patients received rituximab as first-line therapy. At the last follow-up, the eGFR was below 30 ml/min per 1.73 m2 in 32% of patients; 34 (34%) patients experienced a relapse, while 12 (13%) patients had died. By Cox survival analysis, the number of organs involved (hazard ratio [HR], 1.26; 95% confidence interval [CI], 1.01 to 1.55) and low C3 and C4 concentrations (HR, 2.31; 95% CI, 1.10 to 4.85) were independently associated with a higher risk of relapse, whereas first-line therapy with rituximab was protective (HR, 0.22; 95% CI, 0.06 to 0.78). At their last follow-up, 19 (19%) patients had an eGFR ≤30 ml/min per 1.73 m2. Age (odd ratio [OR], 1.11; 95% CI, 1.03 to 1.20), peak serum creatinine (OR, 2.74; 95% CI, 1.71 to 5.47), and serum IgG4 level ≥5 g/L (OR, 4.46; 95% CI, 1.23 to 19.40) were independently predictive for severe CKD.
Conclusions
IgG4-related kidney disease predominantly affected middle-aged men and manifested as tubulointerstitial nephritis with potential glomerular involvement. Complement consumption and the number of organs involved were associated with a higher relapse rate, whereas first-line therapy with rituximab was associated with lower relapse rate. Patients with high serum IgG4 concentrations (≥5 g/L) had more severe kidney disease.
Introduction
IgG4-related disease is a systemic fibroinflammatory disorder that may affect any organ.1–3 First descriptions of IgG4-related disease were reported more than a century ago, yet a unified concept of the disease emerged in the early 2000s.4,5 Of particular note, this organ involvement is characterized by pseudotumoral lesions combining lymphoplasmacytic IgG4-positive rich cell infiltrates and a dense tissue fibrosis, usually described as “storiform.” The most typical laboratory findings are the total serum IgG and IgG4 levels increase, although normal IgG4 levels should not rule out the diagnosis.4,6,7 Recently, an international classification criterion was established to harmonize the IgG4-related disease diagnosis.8
Kidney involvement in IgG4-related disease began to be reported in 2002–20049–11 and affects approximately 30% of patients with IgG4-related disease.12,13 Features of kidney impairment in IgG4-related kidney disease include tubulointerstitial nephritis and glomerular lesions, notably membranous nephropathy14–16 and macroscopic kidney abnormalities observed by dedicated imaging. So far, few studies from Asia and North America have described the effect of IgG4-related disease on the kidney.12,14,17 Furthermore, data on long-term prognosis, relapses, and treatment options are also lacking, with little known about patients in Western Europe.6 Accordingly, we performed a retrospective observational study to describe clinical, biologic, imaging, and histopathologic data, as well as treatment modalities and outcomes.
Methods
Study Design and Ethical Statement
We conducted a multicentric retrospective observational study of patients with IgG4-related kidney disease in 33 French and two Belgian Nephrology and Internal Medicine Centers. Data were collected under relevant French guidelines and regulations, with patient nonopposition a prerequisite for data use. The local institutional review board approved this study as minimal risk research, and according to the French law, a declaration of conformity was performed by the French data protection authority (CNIL: 2224949).
Study Criteria
IgG4-related disease was defined according to the 2019 American College of Rheumatology/European League Against Rheumatism Classification Criteria (ACR/EULAR).8 However, patients with specific ANCA autoantibodies were included if they met the criteria of IgG4-related disease after having ruled out the diagnosis of ANCA-associated vasculitis according to the 2022 ACR/EULAR classification criteria.18–20 We categorized patients with IgG4-related kidney disease into two groups: (1) patients with IgG4-related disease with biopsy-proven kidney involvement and no alternative diagnosis and (2) patients with established IgG4-related disease who displayed kidney failure of no alternative cause and/or proteinuria and/or kidney lesions by computerized tomography scan (CT scan), 18-FDG positron emission tomography (18-FDG-PET CT scan), or magnetic resonance imaging, but who did not undergo a kidney biopsy. We excluded patients with kidney failure due to isolated retroperitoneal fibrosis and patients with missing follow-up data.
Data Collection
Clinical, biologic, imaging, and histopathologic data were retrieved from medical records. Demographic data and comorbidities included age, sex, hypertension, diabetes mellitus, body mass index, and other coexisting comorbidities, including CKD, cardiovascular disease, and other relevant medical history. Laboratory findings included serum levels of electrolytes; albumin; C-reactive protein; C3 and C4 complement components; IgG, IgA, IgM, and IgG subclasses; and positivity for antinuclear and ANCA autoantibodies.
Kidney function assessment was established on serial serum creatinine measurements from the diagnosis to the last follow-up (when available) and on analysis of urine protein-to-creatinine ratio (UPCR) and urine red and white blood cell counts. AKI was scored according to the 2012 Kidney Disease Improving Global Outcomes Clinical Practice Guideline for Acute Kidney Injury.21 Proteinuria and nephrotic range proteinuria were defined by urine protein ≥0.3 g/d (UPCR ≥0.3 g/g) and ≥3 g/d (UPCR ≥3 g/g), respectively. The GFR was estimated using the 2008 Chronic Kidney Disease Epidemiology Collaboration equation.22
Radiologic findings included CT scan and 18-FDG-PET CT scan results. Histologic data were retrieved from initial pathologic reports and had principal diagnosis, glomerular lesions, tubulointerstitial fibrosis and tubular atrophy, and vascular lesions. When available, the degree of fibrosis and tubulointerstitial inflammatory infiltration by lymphoplasmacytic cells was also recorded with the rate of IgG4(+)/IgG(+) cells and the number of IgG4(+) plasma cells per high-power field.
Treatment and outcomes were also collected, and follow-up data were censored at disease relapse and death. Relapse was defined as a progressive disease or recurrence of clinical symptoms, biologic abnormalities, or imaging findings after remission, with or without elevation of the serum IgG4 level.23–25
Statistical Analysis
Descriptive statistics were used to summarize the data. The results were reported as median and interquartile range (IQR) for continuous variables and as counts and percentages for categorical variables. Multivariable models were built using a conditional backward stepwise variable selection process on the basis of variable influence in unadjusted analysis. Critical entry and exit P values were 0.2 and 0.1, respectively. Logistic regression was performed to identify the possible factors involved in severe CKD (as defined by an eGFR ≤30 ml/min per 1.73 m2) at the last follow-up, while Kaplan–Meier survival curves and Cox proportional hazards models were performed to assess the factors associated with the risk of relapse. Nevertheless, it was preplanned to force clinically relevant variables (age, sex, and corticosteroid treatment) into the final model, determining the risk factors for severe CKD. In addition, correlation and interaction were checked within the last models, as well as the assumption for log-linearity of continuous variables and proportional hazard assumptions for survival models. Data are given as odds ratios (ORs), 95% confidence interval (CI), or hazard ratios (HRs), as appropriate. Models were built assuming 10% of missing data for the relapse status and eGFR at the last follow-up. If we encountered more missing values, a multiple imputations method was planned.
To assess the robustness of our findings regarding the rituximab effect, we also performed a propensity score weighting as sensitivity analysis.26–28 The propensity score was built using logistic regression according to rituximab-associated variable. Covariates included in the model were age, sex, serum IgG4 concentration, corticosteroid therapy, and serum peak creatinine. The influence of rituximab on relapse was then evaluated using overlap propensity score-weighted logistic regression models. According to the reviewer's suggestion, we performed additional sensitivity analyses. First, we assessed the time effect on the relapse outcome by forcing the year of diagnosis in the Cox model. Then, we built a conditional logistic regression model stratified by the year of diagnosis, and finally, we used a generalized mixed model with the year of diagnosis as a random effect. Statistical analyses were performed using R version 4.2.2 (R Foundation for Statistical Computing) using “survival,” “survey,” “ggstatsplot,”29 “lme4,” and “WeightIt” packages.
Results
Baseline Characteristics
We identified 125 patients with a differential diagnosis of IgG4-related kidney disease between January 1, 1997, and December 31, 2019, from 35 sites. Among them, 18 did not fulfill the ACR/EULAR 2019 criteria, four had isolated retroperitoneal disease, and two had missing data regarding the treatment and follow-up. Finally, 101 patients with IgG4-related kidney disease were included in our cohort (Figure 1). Patients' characteristics are presented in Table 1. The median (IQR) age at diagnosis was 68 (57–76) years. Most (86%) patients were men. Kidney involvement was one of the presenting features at the diagnosis in 61 (60%) patients and a part of systemic organ involvement in 87 (86%) patients. Extrarenal features included lymphadenopathies (57%), type 1 autoimmune pancreatitis (42%), sialadenitis (36%), lung involvement (28%), and cholangitis (25%). The main striking laboratory findings included polyclonal hypergammaglobulinemia and increased serum IgG4 levels in 94% and 90% of cases. On the other hand, complement levels were decreased in 45% of cases when tested, with a significant proportion of patients also displaying autoimmunity features, including nonspecific antinuclear antibody or nonspecific ANCA positivity in 36% and 22%, respectively.
Figure 1.
Flow chart. ACR, American College of Rheumatology; EULAR, European League Against Rheumatism Classification Criteria.
Table 1.
Demographic, clinical, and laboratory findingsa
| Variable | Entire Cohort N=101 |
|---|---|
| Demographic data | |
| Age, yr, median (IQR) | 68 (57–76) |
| Male, n (%) | 87 (86) |
| Diagnostic delay, mo, median (IQR) | 9 (2–20) |
| Extrarenal involvement, n (%) | 87 (86) |
| Pancreatitis | 42 (42) |
| Retroperitoneal fibrosis | 15 (15) |
| Sialadenitis | 36 (36) |
| Cholangitis | 25 (25) |
| Lung | 28 (28) |
| Adenopathy | 58 (58) |
| Aortitis/periaortis | 8 (8) |
| Orbital | 5 (5) |
| Thyroiditis | 3 (3) |
| Breast | 2 (2) |
| Hypophysitis | 1 (1) |
| Laboratory data | |
| Serum IgG4 dosage, n (%) | 94 (93) |
| IgG4 concentration, g/L, median (IQR) | 5.1 (2.3–9) |
| IgG4 concentration >1.35 g/L | 87 (93) |
| Serum γ-globulins dosage, n (%) | 95 (94) |
| Concentration, g/L, median (IQR) | 23.1 (18.7–29.0) |
| Hyper-γ-globulinemia, n, (%) | 81 (85) |
| Serum IgG dosage, n (%) | 84 (83) |
| ↗ IgG, n (%) | 73 (86) |
| Other serum immunoglobulin dosage, n (%) | 71 (70) |
| ↗ IgA, n (%) | 13 (18) |
| ↗ IgM, n (%) | 4 (6) |
| Other serum IgG subclasses dosage, n (%) | 66 (65) |
| ↗ IgG1, n (%) | 29 (44) |
| ↗ IgG2, n (%) | 19 (28) |
| ↗ IgG3, n (%) | 25 (38) |
| Serum complement dosage, n (%) | 82 (81) |
| ↘ C3 or ↘ C4, n (%) | 37 (45) |
| ↘ C3, n (%) | 33 (40) |
| ↘ C4, n (%) | 32 (39) |
| Antinuclear antibodies dosage, n (%) | 97 (96) |
| Positive, n (%) | 35 (36) |
| ANCA dosage, n (%) | 81 (80) |
| Positive, n (%) | 18 (22) |
| C-reactive protein, mg/L, median (IQR) | 9 (4.5–37) |
IQR, interquartile range.
Percentages were calculated out of the number of tested patients.
Kidney Findings
The main kidney findings are listed in Table 2. Patients presented AKI, AKI-on-CKD, and isolated CKD in 51%, 23%, and 14% of cases, respectively. At diagnosis, the median (IQR) serum creatinine level was 2.4 mg/dl (1.6–3.6), corresponding to a median (IQR) eGFR of 25 ml/min per 1.73 m2 (17–43). Urinary sediment was most often bland. The median (IQR) UPCR was 0.6 g/g (0.2–1.1), but 31% of patients had more than 1 g/g, almost exclusively those with glomerular involvement. Kidney lesions by CT scan abnormalities were identified in 54 (61%) patients. The imaging findings included bilateral kidney hypertrophy, pseudotumor, and low-density areas in both renal cortices in 19%, 27%, and 25% of cases, respectively. The 18-FDG-PET CT scan was performed in 64 (63%) patients and revealed hypermetabolic kidney lesions in 24 (38%) patients and extrarenal lesions in 46 (74%) patients.
Table 2.
Kidney findingsa
| Variable | Entire Cohort N=101 | Tubulointerstitial Alone n=85 | Tubulointerstitial and Glomerular n=16 | |
|---|---|---|---|---|
| Kidney parameters at presentation | ||||
| AKI alone, n (%) | 51 (51) | 41 (48) | 10 (63) | |
| AKI-on-CKD, n (%) | 23 (23) | 19 (22) | 4 (25) | |
| CKD alone, n (%) | 14 (14) | 14 (17) | 0 (0) | |
| Preserved kidney function, n (%) | 13 (13) | 11 (13) | 2 (13) | |
| Baseline serum creatinine, mg/dl, median (IQR) | 1.1 (0.9–1.3) | 1.1 (0.9–1.3) | 1.1 (0.9–1.3) | |
| Baseline eGFR, ml/min per 1.73 m2, median (IQR) | 69 (52–87) | 69 (51–82) | 68 (54–95) | |
| Peak serum creatinine at diagnosis, mg/dl, median (IQR) | 2.4 (1.6–3.6) | 2.4 (1.6–3.6) | 1.9 (1.5–3.6) | |
| eGFR at diagnosis, ml/min per 1.73 m2, median (IQR) | 25 (17–43) | 25 (17–43) | 29 (17–47) | |
| UPCR, g/g, median (IQR) | 0.6 (0.2–1.1) | 0.39 (0.12–0.9) | 4 (2.0–4.7) | |
| UPCR >1 g/g, n (%) | 31 (31) | 18 (21) | 13 (81) | |
| Hematuria, n (%) | 27 (27) | 16 (19) | 11 (69) | |
| Leukocyturia, n (%) | 16 (16) | 12 (14) | 4 (25) | |
| Imaging data | ||||
| Kidney CT scan, n (%) | 89 (88) | 74 (73) | 15 (94) | |
| Kidney lesions, n (%) | 54 (61) | 45 (61) | 9 (89) | |
| Low-density areas in both renal cortices, n (%) | 22 (25) | 19 (26) | 3 (20) | |
| Enlarged kidneys, n (%) | 17 (19) | 13 (18) | 4 (27) | |
| Pseudotumor lesions, n (%) | 24 (27) | 19 (27) | 2 (13) | |
| 18FDG-PET CT scan, n (%) | 64 (63) | 53 (62) | 11 (69) | |
| Positive PET CT scan, n (%) | 55 (86) | 46 (87) | 9 (82) | |
| Kidney hypermetabolic lesion, n (%) | 24 (38) | 19 (36) | 5 (46) | |
| Extrarenal hypermetabolic lesion, n (%) | 46 (74) | 40 (76) | 6 (55) | |
| Histologic findings, n (%) | 83 (82) | 67 (79) | 16 (100) | |
| Glomerulosclerosis, %, mean (SD) | 23 (43) | 22 (42) | 33 (58) | |
| Plasmacytic infiltrate, n (%) | 76 (92) | 62 (93) | 14 (88) | |
| IgG4(+) plasmacytic infiltrate, n (%)b | 72 (87) | 58 (87) | 14 (88) | |
| Lymphocytic infiltration, n (%) | 77 (93) | 63 (94) | 14 (88) | |
| Lymphoid follicle formation, n (%) | 15 (18) | 12 (18) | 3 (19) | |
| Eosinophilic infiltration, n (%) | 10 (12) | 9 (13) | 1 (6) | |
| Interstitial fibrosis and tubular atrophy, n (%) | 72 (87) | 57 (85) | 15 (94) | |
| Interstitial fibrosis and tubular atrophy grade (%) | ||||
| <10% | 7 (8) | 7 (10) | 0 (0) | |
| 10%–25% | 7 (8) | 5 (8) | 2 (13) | |
| 25%–50% | 12 (15) | 6 (9) | 6 (38) | |
| ≥50% | 35 (42) | 29 (43) | 6 (38) | |
| Tubulitis, n (%) | 32 (39) | 26 (39) | 6 (38) | |
| Tubular deposits by immunofluorescence, n (%) | 22 (27) | 17 (25) | 5 (31) | |
| Glomerular deposition by immunofluorescence, n (%) | 11 (13) | 2 (3) | 9 (56) | |
IQR, interquartile range; UPCR, urine protein-to-creatinine ratio; CT, computerized tomography scan; PET CT, positron emission tomography.
Percentages were calculated out of the number of tested patients.
The IgG4 stain was positive in all kidney biopsies when performed. Four patients with plasmacytic infiltrate did not have IgG4 staining.
Overall, 83 (82%) patients underwent a kidney biopsy showing tubulointerstitial involvement in all cases. Additional glomerular lesions were reported in 16 (16%) patients, with membranous nephropathy being the predominant pattern. The presence of tubulointerstitial lymphoplasmacytic infiltrates has been observed in all biopsied patients. IgG4 staining was performed in 72 patients and revealed predominant IgG4(+) plasma cells in all patients. The presence of a dense fibrosis, affecting >50% of the kidney tissue, was also described in 42% of cases, but the so-called “storiform” pattern was rarely addressed (Figure 2, Table 2).
Figure 2.
Typical kidney pathology in IgG4-TIN. (A) Light microscopic kidney biopsy findings in an IgG4-TIN patient with extensive interstitial fibrosis, tubular atrophy, and interstitial inflammation (trichrome stain, magnification ×55, scale bar=200 μm). (B) High-power view of lymphoplasmacytic infiltrate within the kidney interstitium with interlacing fibrils of storiform fibrosis (Periodic acid–Schiff stain, magnification ×400, scale bar=20 μm). (C) Typical storiform fibrosis in IgG4-TIN (periodic acid-methenamine silver stain, magnification ×200, scale bar=50 μm). Lymphocytes and plasma cells are encircled by collagenous tissue and diffuse fibrosis. (D) Immunohistochemistry for IgG4 showing numerous interstitial IgG4-positive plasma cells (magnification ×400, scale bar=20 μm). IgG4-TIN, IgG4-related tubulointerstitial nephritis. Images courtesy of S. Ferlicot, Bicêtre University Hospital, France. Figure 2 can be viewed in color online at www.cjasn.org.
Treatment and Outcomes
Treatment and outcomes are summarized in Table 3. Almost all patients (90%) were treated with corticosteroid therapy with a mean initial dose of 0.8±0.3 mg/kg per day. In addition, 18 patients received rituximab as first-line therapy after a schema of 1 g at day 1 and day 15 in 77% of cases, with around two cycles realized (Supplemental Table 1). After a median (IQR) follow-up of 24 (11–58) months, 35 (35%) patients relapsed within a median of 12 (8.5–24) months for the first relapse. Relapse-free survival analysis is presented in Figure 3 and Table 4. By multivariable analysis, the number of organs involved (HR, 1.26; 95% CI, 1.01 to 1.55) and low C3 and C4 concentrations (HR, 2.31; 95% CI, 1.10 to 4.85) were associated with a higher risk of relapse, while the use of rituximab as first-line therapy was characterized by a lower risk of relapse (HR, 0.22; 95% CI, 0.06 to 0.78) (Figure 3, A–C, Table 4). Therefore, if the patients who received rituximab as first-line therapy seemed to have less relapse (22% versus 37%) and obtain similar kidney outcomes, the proportion of complications, such as death (6% versus 15%) and infections (17% versus 25%), seemed lower compared with the patient without rituximab as first-line therapy (Supplemental Tables 1 and 2). After weighting both groups and propensity score, the rituximab as first-line therapy was still associated with a lower risk of relapse (HR, 0.27; 95% CI, 0.10 to 0.77) (Supplemental Figures 1–3). This effect also remains despite considering the year of diagnosis (Supplemental Table 3).
Table 3.
Treatment and outcomes
| Variable | Entire Cohort N=101 |
|---|---|
| First-line treatment | |
| Corticosteroids, n (%) | 90 (90) |
| Initial dose, mg/kg per day, mean (SD) | 0.8 (0.3) |
| Duration ≥12 mo, n (%) | 38 (39) |
| Rituximab+corticosteroids, n (%) | 13 (13) |
| Rituximab alone, n (%) | 5 (5) |
| Overall treatment | |
| Corticosteroids, n (%) | 90 (90) |
| Rituximab, n (%) | 36 (37) |
| Azathioprine, n (%) | 11 (12) |
| Mycophenolate mofetil, n (%) | 6 (6) |
| Outcomes | |
| Follow-up, mo, median, IQR | 24 (11–58) |
| Kidney function | |
| Serum creatinine at the last follow-up, mg/dl, median, IQR | 1.5 (1.1–2.2) |
| eGFR at the last follow-up, ml/min per 1.73 m2, median, IQR | 45 (25–68) |
| eGFR at the last follow-up ≤30 ml/min per 1.73 m2, n (%) | 29 (32) |
| End-stage kidney disease, n (%) | 11 (12) |
| Relapses | |
| Number of relapses, n (%) | 35 (35) |
| Time to relapse, mo, median, IQR | 12 (9–24) |
| Complications | 48 (55) |
| Death, n (%) | 12 (13) |
| Diabetes, n (%) | 29 (35) |
| Infection, n (%) | 20 (23) |
IQR, interquartile range.
Figure 3.
Relapse risk. (A–C) Kaplan–Meier survival estimates of relapse-free survival according to the presence of (A) ANCA; (B) complement concentration; and (C) first-line rituximab therapy. (D) Violin plot of the number of organs involved according to relapse occurrence. Figure 3 can be viewed in color online at www.cjasn.org.
Table 4.
Relapse-free survival (Cox survival analysis)
| Variable | Unadjusted Analysis | Multivariable Analysis | ||
|---|---|---|---|---|
| HR | 95% CI | HR | 95% CI | |
| Number of organs involved | 1.29 | (1.03 to 1.60) | 1.26 | (1.01 to 1.55) |
| ANCA positivity | 3.18 | (1.39 to 7.31) | ||
| ↘ C3 and C4 levels | 1.83 | (0.90 to 3.72) | 2.31 | (1.10 to 4.85) |
| Kidney imaging lesions | 1.85 | (0.89 to 3.85) | ||
| First-line rituximab | 0.37 | (0.12 to 0.96) | 0.22 | (0.06 to 0.78) |
HR, hazard ratio; CI, confidence interval.
At the last follow-up, the eGFR was missing for nine (9%) patients. Of the overall cohort, 72 (71%) had CKD characterized by a median eGFR of 45 ml/min per 1.73 m2 (26–68), including 32% with an eGFR ≤30 ml/min per 1.73 m2. Eleven (12%) patients progressed to end-stage kidney disease, and 12 (13%) patients died. Factors associated with severe CKD at the last follow-up (i.e., an eGFR ≤30 ml/min per 1.73 m2) were age (OR, 1.07; 95% CI, 1.03 to 1.13) and peak serum creatinine (OR, 2.53; 95% CI, 1.67 to 4.24), while prolonged corticosteroids duration >12 months (OR, 0.30; 95% CI, 0.10 to 0.80) and cholangitis (OR, 0.26; 95% CI, 0.06 to 0.84) were associated with a lower risk. Using multivariable logistic regression, only age (OR, 1.11; 95% CI, 1.03 to 1.20), peak serum creatinine at diagnosis (OR, 2.74; 95% CI, 1.71 to 5.47), and serum IgG4 concentration ≥5 g/L (OR, 4.46; 95% CI, 1.23 to 19.40) remained independently associated with the risk of severe CKD at the last follow-up (Table 5). Of note, the serum IgG4 level at diagnosis and the interstitial fibrosis and tubular atrophy (IFTA) state reported on the kidney biopsy also seemed related to the eGFR at the last follow-up (Supplemental Figure 4).
Table 5.
Covariates associated with eGFR at the last follow-up ≤30 ml/min per 1.73 m2 (logistic regression)
| Variable | Unadjusted Analysis | Multivariable Analysis | ||
|---|---|---|---|---|
| OR | 95% CI | OR | 95% CI | |
| Age | 1.07 | (1.03 to 1.13) | 1.11 | (1.03 to 1.20) |
| Cholangitis | 0.26 | (0.06 to 0.84) | ||
| Peak serum creatinine | 2.53 | (1.67 to 4.24) | 2.74 | (1.71 to 5.47) |
| Serum IgG4 concentration ≥5 g/L | 2.34 | (0.93 to 6.22) | 4.46 | (1.23 to 19.40) |
| Corticotherapy >12 mo | 0.30 | (0.10 to 0.80) | 0.37 | (0.08 to 1.43) |
| Number of organs involved | 0.81 | (0.58 to 1.10) | ||
| ↘ C3 and C4 levels | 0.89 | (0.34 to 2.30) | ||
| First-line rituximab | 1.19 | (0.37 to 3.53) | ||
OR, odd ratio; CI, confidence interval.
Discussion
This study represents a large series depicting the spectrum and outcome of patients with intrinsic kidney involvement in the course of IgG4-related disease. Patients were included from multiple centers in France and Belgium, and all had a definite diagnosis of IgG4-related disease according to the 2019 ACR/EULAR.8 Most had a histologic assessment of IgG4-related kidney disease, and most displayed a follow-up that allowed the analysis of treatment response, disease relapse(s), and long-term kidney damage due to IgG4-related kidney disease. Collectively, our findings provide further insights into the description and management of IgG4-related kidney disease in patients from Western Europe.
Similar to IgG4-related disease, IgG4-related kidney disease affects middle-aged men and is associated with a remarkable increase in serum IgG4 levels and typical kidney infiltration by polytypic IgG4-expressing plasma cells along with tissue fibrosis.16 The pathologic spectrum of IgG4-related kidney disease includes two significant patterns. IgG4-related tubulointerstitial nephritis is by far the most prevalent, representing all cases in our study, while IgG4-related glomerulonephritis is less common, representing <20% of cases, with membranous nephropathy being the most common.30,31 The spectrum of IgG4-related kidney disease also includes macroscopic kidney lesions that may be identified by different imaging techniques in up to two thirds of patients.12,14,32 Various patterns have been described, including nodules, kidney hypertrophy, cortical low-density areas, and pseudotumor lesions. Notably, such lesions may be observed without kidney dysfunction, proteinuria, and abnormal urinary sediment.
Interestingly, although IgG4-related disease may be either systemic or restricted to a single organ, IgG4-related kidney disease is almost exclusively associated with extrarenal features. Previous reports have underscored the correlation between IgG4 titers and systemic involvement, which is consistent with the high proportion of patients with IgG4-related kidney disease who have an increased serum IgG4 level at diagnosis. Kidney involvement thus represents a part of a systemic disease, and patients with IgG4-related kidney disease should be systematically assessed for extrarenal lesions. 18-FDG-PET CT scan may be helpful in these cases to screen for hypermetabolic and sometimes asymptomatic organ involvement. In our series, more than 85% of patients displayed extrarenal hypermetabolic lesions, some of which were not clinically obvious.
Beyond serum IgG4 increase, IgG4-related kidney disease is commonly associated with unspecific immunological findings. Low complement levels are observed in one third of patients with IgG4-related kidney disease across various studies.12,14,16 Surprisingly, such results are not reported in patients without kidney involvement. Because IgG4 alone is unlikely to activate the classical complement pathway,33 multiple hypotheses have been raised to account for complement activation in the course of IgG4-related kidney disease, including the concomitant increase of other IgG subclasses leading to the formation of immune complexes and the activation of complement cascade through the lectin pathway.34
One third of patients also display unspecific antinuclear antibodies positivity,12,14 and 18% have ANCA positivity. Although the ACR/EULAR guidelines consider the presence of anti-proteinase-3 (PR3) and anti-myeloperoxidase (MPO) antibodies as an exclusion criterion for IgG4-related disease,8 most ANCA-positive patients in our series, including those with anti-MPO/PR3 specificity, had a definite diagnosis of IgG4-related kidney disease. Danlos et al. reported 18 patients with a concomitant diagnosis of ANCA-associated vasculitis and IgG4-related disease, including 22% with kidney involvement.35 Noteworthy, none of the patients with IgG4-related disease had pauci-immune glomerulonephritis, and none with IgG4-related kidney disease had concomitant extracapillary glomerulonephritis. Whether patients with ANCA positivity represent a peculiar subset among IgG4-related disease remains to be determined.
Almost all patients were treated, with most who received first-line corticosteroid therapy. Most had a partial-to-complete response, which was consistent with previous reports. Such remarkable corticosteroid therapy efficiency is now considered as a retrospective criterion for the classification of IgG4-related disease.8 Unfortunately, the disease, particularly IgG4-related kidney disease, is burdened by a high risk of relapse, approximately one third of cases, and a significant risk of long-term organ damage. In our study, 70% and 7% of patients developed CKD and end-stage kidney disease, pointing to the need for an optimal management strategy. In our study, prolonged corticosteroid therapy for at least 12 months was associated with better kidney function at the last follow-up by unadjusted analysis. Such association was not apparent by multivariable analysis, with the peak serum creatinine, age at diagnosis, and serum IgG4 ≥5 g/L as the three independent variables associated with worse kidney outcomes. Our data underscored the need for a prompt and accurate diagnosis and initiation of efficient treatment in IgG4-related kidney disease. Alternative therapies to corticosteroids include rituximab and other immunosuppressants, such as azathioprine or mycophenolate mofetil. Some authors have also suggested the potential benefit of bortezomib and abatacept in some cases.36 In our cohort, 37% of patients were administered rituximab, including 18% as first-line therapy, most often associated with corticosteroids. This may allow faster corticosteroid tapering without affecting kidney outcomes. In addition, this strategy was associated with a significantly lower risk of overall relapse. If the retrospective design cannot assure the control of all confounders, the number of death and infections also seemed reduced in the first-line rituximab group. Our findings underline the results of the prospective open-label trial conducted by Carruthers et al., which showed that rituximab could be used as first-line therapy with or without concomitant glucocorticoids with a robust clinical response.37 However, our results can only be hypothesis-generating, and the potential benefit of rituximab in IgG4-related kidney disease remains to be established by further prospective investigation.
Despite major strengths, our study had several limitations, mainly because of its retrospective design leading to potential confounding factors, missing data, and recall bias. Moreover, using ACR/EULAR 2019 criteria for assessing IgG4-related disease diagnosis made us exclude patients with possible concomitant IgG4-related kidney disease and ANCA-associated vasculitis or systemic lupus. In addition, different policies in kidney biopsy indication and proteinuria and extrarenal features between the various participating centers have also accounted for potential biases in our study. Most importantly, the low number of patients treated with rituximab should moderate the potential benefit of its use as a first-line option and underscores the need for further collaborative randomized controlled trials.
In conclusion, IgG4-related kidney disease is a well-recognized entity that affects middle-aged men and predominantly presents as tubulointerstitial nephritis with glomerular involvement in approximately a quarter of patients. The response to corticosteroids is usually favorable, but the disease remains characterized by a high risk of relapse in about one third of patients and more frequently in patients with CKD. Therefore, tighter monitoring should be maintained for patients with many organs involved and high-serum IgG4 levels because of poorer outcomes. Rituximab as first-line therapy seems to reduce the risk of relapse without a signal for harmfulness. Owing to the retrospective design of this study, our findings remain hypothesis-generating and should be confirmed in future interventional studies.
Supplementary Material
Acknowledgments
We sincerely thank the patients and the clinicians for their contribution to this study, as well as Dr. Glenn Eastwood for his expertise in writing the manuscript.
Footnotes
See related editorial, “IgG4-Related Kidney Disease,” on pages 994–996.
Disclosures
V. Audard reports consultancy for, honoraria from, and advisory or leadership roles on Advisory boards for Addmedica, Alnylam, AstraZeneca, Bayer, and Vifor pharma. J.J. Boffa reports consultancy for AstraZeneca, Bayer, BMS, Boehringer Ingelheim, GlaxoSmithKline, Novartis, Otsuka, Travere, and Vifor pharma; honoraria from AstraZeneca, Boehringer Ingelheim, GlaxoSmithKline, Otsuka, and Vifor pharma; and advisory or leadership roles for AstraZeneca, Bayer, BMS, Boehringer Ingelheim, GlaxoSmithKline, Novartis, Otsuka, Travere, and Vifor pharma. J. Chemouny reports consultancy for AstraZeneca. A. Devresse reports consultancy for and advisory or leadership roles for Alnylam Pharmaceuticals and Merck. S. Duquennoy reports employment with Fondation AUB SANTE. J.M. Halimi reports consultancy for Alexion, AstraZeneca, Bayer, Boehringer Ingelheim France, Servier, and Vifor Fresenius; research funding from AstraZeneca; and honoraria from Alexion, AstraZeneca, Bayer, Boehringer Ingelheim France, MSD, Sanofi, Servier, and Vifor. D. Joly reports honoraria from AstraZeneca and Bayer. A. Karras reports consultancy for Alnylam, GlaxoSmithKline, Novartis, Otsuka, Vifor; honoraria from AstraZeneca, Bohringer-Ingelheim, GlaxoSmithKline, Novartis, Otsuka, Pfizer, and Vifor; advisory or leadership roles for Novartis, Otsuka, and Vifor; and speakers bureau for AstraZeneca, Boehringer-Ingelheim, Otsuka, Pfizer, and Vifor. A. Mathian reports consultancy for AstraZeneca, GlaxoSmithKline, Novartis, and Otsuka; honoraria from AstraZeneca, GlaxoSmithKline, Novartis, and Otsuka; advisory or leadership role for AstraZeneca; and speakers bureau for AstraZeneca, GlaxoSmithKline, and Otsuka. L. Mercadal reports honoraria for Congress travel and hotel in France and reports advisory or leadership roles for nephrologie thérapeutique review board and Société francophone de néphrologie dialyse transplantation. J. Olagne reports serving on speakers bureau for GlaxoSmithKline. S. Palat reports employment with and advisory or leadership role for Janssen Cilag. E. Plaisier reports employment with AURA PARIS. O. Thaunat reports consultancy for AstraZeneca, Biotest, and Novartis; research funding from bioMérieux, BMS, and Immucor; honoraria from Astellas, Biotest, and Novartis; and advisory or leadership role for ESOT. E. Vilaine reports consultancy for and honoraria from AstraZeneca. All remaining authors have nothing to disclose.
Funding
None.
Author Contributions
Conceptualization: Anis Chaba.
Data curation: Anis Chaba.
Investigation: Julie Belliere, Claire Cartery, Anis Chaba, Jonathan Chemouny, Claudine Contamin, Cecile Courivaud, Clément Deltombe, Simon Duquennoy, Mikael Ebbo, Anne Laure Faller, Sophie Ferlicot, Hugo Garcia, Nabila Goumri, Jean Michel Halimi, Mohamed Hamidou, Guillaume Hanouna, Dominique Joly, Cédric Landron, David Launay, Celine Lebas, Mathieu Legendre, Agathe Masseau, Alexis Mathian, Lucile Mercadal, Rafik Mesbah, Nathalie Morel, Charlotte Mussini, Prisca Mutinelli-Szymanski, Mathilde Nouvier, Jerome Olagne, Sylvain Palat, Jean-Loup Pennaforte, Marie Noelle Peraldi, Emmanuelle Plaisier, Agnieszka Pozdzik, Eric Prinz, Sarah Richter, Nicolas Schleinitz, Renaud Snanoudj, Olivier Thaunat, Dimitri Titeca-Beauport, Sonia Touati, Eve Vilaine, Clarissa Von-Kotze, Mohamad Zaidan.
Methodology: Anis Chaba.
Validation: Alexandre Karras.
Writing – original draft: Vincent Audard, Jean Jacques Boffa, Anis Chaba, Arnaud Devresse, Mohamad Zaidan.
Data Sharing Statement
All data are included in the manuscript and/or supporting information.
Supplemental Material
This article contains the following supplemental material online at http://links.lww.com/CJN/B778.
Supplemental Table 1. Outcomes according to rituximab use.
Supplemental Table 2. Relapses characteristics.
Supplemental Table 3. Sensitivity analysis of relapse risk according to the year of diagnosis.
Supplemental Figure 1. Distribution balance of propensity score.
Supplemental Figure 2. Proportional balance of propensity score and covariate balance.
Supplemental Figure 3. Kaplan–Meier survival estimates of relapse-free survival according to the use of rituximab as first-line therapy after propensity score weighting.
Supplemental Figure 4. ANOVA analysis of eGFR at the last follow-up according to the IFTA grade on the diagnostic kidney biopsy.
References
- 1.Stone JH, Deshpande V. IgG4-Related disease. N Engl J Med. 2012;366(6):539–551. doi: 10.1056/NEJMra1104650 [DOI] [PubMed] [Google Scholar]
- 2.Kamisawa T, Zen Y, Pillai S, Stone JH. IgG4-related disease. Lancet. 2015;385(9976):1460–1471. doi: 10.1016/s0140-6736(14)60720-0 [DOI] [PubMed] [Google Scholar]
- 3.Umehara H, Okazaki K, Masaki Y, Kawano M, Yamamoto M, Saeki T. A novel clinical entity, IgG4-related disease (IgG4RD): general concept and details. Mod Rheumatol. 2012;22(1):1–14. doi: 10.3109/s10165-011-0508-6 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 4.Hideaki H Shigeyuki K Akira H, et al. High serum IgG4 concentrations in patients with sclerosing pancreatitis. N Engl J Med. 2001;344(10):732–738. doi: 10.1056/NEJM200103083441005 [DOI] [PubMed] [Google Scholar]
- 5.Kamisawa T, Funata N, Hayashi Y, Eishi Y, Koike M, Tsuruta K. A new clinicopathological entity of IgG4-related autoimmune disease. J Gastroenterol. 2003;38(10):982–984. doi: 10.1007/s00535-003-1175-y [DOI] [PubMed] [Google Scholar]
- 6.Deshpande V, Zen Y, Chan JK, Yi EE, Sato Y, Yoshino T. Consensus statement on the pathology of IgG4-related disease. Mod Pathol. 2012;25(9):1181–1192. doi: 10.1038/modpathol.2012.72 [DOI] [PubMed] [Google Scholar]
- 7.Bledsoe JR, Della-Torre E, Rovati L, Deshpande V. IgG4-related disease: review of the histopathologic features, differential diagnosis, and therapeutic approach. APMIS. 2018;126(6):459–476. doi: 10.1111/apm.12845 [DOI] [PubMed] [Google Scholar]
- 8.Wallace ZS, Naden RP, Chari S, Choi H, Della-Torre E, Dicaire J. The 2019 American College of Rheumatology/European League against rheumatism classification criteria for IgG4-related disease. Arthritis Rheumatol. 2020;72(1):7–19. doi: 10.1002/art.41120 [DOI] [PubMed] [Google Scholar]
- 9.Uchiyama-Tanaka Y, Mori Y, Kimura T, Sonomura K, Umemura S, Kishimoto N. Acute tubulointerstitial nephritis associated with autoimmune-related pancreatitis. Am J Kidney Dis. 2004;43(3):e13.1–e13.8. doi: 10.1053/j.ajkd.2003.12.006 [DOI] [PubMed] [Google Scholar]
- 10.Takeda S, Haratake J, Kasai T, Takaeda C, Takazakura E. IgG4-associated idiopathic tubulointerstitial nephritis complicating autoimmune pancreatitis. Nephrol Dial Transplant. 2004;19(2):474–476. doi: 10.1093/ndt/gfg477 [DOI] [PubMed] [Google Scholar]
- 11.Watson SJW, Jenkins DAS, Bellamy COS. Nephropathy in IgG4-related systemic disease. Am J Surg Pathol. 2006;30(11):1472–1477. doi: 10.1097/01.pas.0000213308.43929.97 [DOI] [PubMed] [Google Scholar]
- 12.Evans RDR, Cargill T, Goodchild G, Oliveira B, Rodriguez-Justo M, Pepper R. Clinical manifestations and long-term outcomes of IgG4-related kidney and retroperitoneal involvement in a United Kingdom IgG4-related disease cohort. Kidney Int Rep. 2019;4(1):48–58. doi: 10.1016/j.ekir.2018.08.011 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 13.Kuruma S, Kamisawa T, Kikuyama M, Chiba K, Shimizuguchi R, Koizumi S. Clinical characteristics of autoimmune pancreatitis with IgG4 related kidney disease. Adv Med Sci. 2019;64(2):246–251. doi: 10.1016/j.advms.2018.12.005 [DOI] [PubMed] [Google Scholar]
- 14.Saeki T, Nishi S, Imai N, Ito T, Yamazaki H, Kawano M. Clinicopathological characteristics of patients with IgG4-related tubulointerstitial nephritis. Kidney Int. 2010;78(10):1016–1023. doi: 10.1038/ki.2010.271 [DOI] [PubMed] [Google Scholar]
- 15.Saeki T, Kawano M. IgG4-related kidney disease. Kidney Int. 2014;85(2):251–257. doi: 10.1038/ki.2013.393 [DOI] [PubMed] [Google Scholar]
- 16.Raissian Y, Nasr SH, Larsen CP, Colvin RB, Smyrk TC, Takahashi N. Diagnosis of IgG4-related tubulointerstitial nephritis. J Am Soc Nephrol. 2011;22(7):1343–1352. doi: 10.1681/ASN.2011010062 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 17.Alexander MP, Larsen CP, Gibson IW, Nasr SH, Sethi S, Fidler ME. Membranous glomerulonephritis is a manifestation of IgG4-related disease. Kidney Int. 2013;83(3):455–462. doi: 10.1038/ki.2012.382 [DOI] [PubMed] [Google Scholar]
- 18.Robson JC, Grayson PC, Ponte C, Suppiah R, Craven A, Judge A. 2022 American College of Rheumatology/European alliance of associations for rheumatology classification criteria for granulomatosis with polyangiitis. Arthritis Rheumatol. 2022;74(3):393–399. doi: 10.1002/art.41986 [DOI] [PubMed] [Google Scholar]
- 19.Grayson PC, Ponte C, Suppiah R, Robson JC, Craven A, Judge A. 2022 American College of Rheumatology/European alliance of associations for rheumatology classification criteria for eosinophilic granulomatosis with polyangiitis. Arthritis Rheumatol. 2022;74(3):386–392. doi: 10.1002/art.41982 [DOI] [PubMed] [Google Scholar]
- 20.Suppiah R, Robson JC, Grayson PC, Ponte C, Craven A, Khalid S. 2022 American College of Rheumatology/European alliance of associations for rheumatology classification criteria for microscopic polyangiitis. Arthritis Rheumatol. 2022;74(3):400–406. doi: 10.1002/art.41983 [DOI] [PubMed] [Google Scholar]
- 21.Khwaja A. KDIGO Clinical practice guidelines for acute kidney injury. Nephron Clin Pract. 2012;120(4):c179–c184. doi: 10.1159/000339789 [DOI] [PubMed] [Google Scholar]
- 22.Levey AS. A more accurate method to estimate glomerular filtration rate from serum creatinine: a new prediction equation. Ann Intern Med. 1999;130(6):461. doi: 10.7326/0003-4819-130-6-199903160-00002 [DOI] [PubMed] [Google Scholar]
- 23.Zongfei J, Lingli C, Ying S, Lingying M, Lijuan Z, Dongmei L. Clinical and pathological predictors of relapse in IgG4-related disease. Arthritis Res Ther. 2022;24(1):106. doi: 10.1186/s13075-022-02792-z [DOI] [PMC free article] [PubMed] [Google Scholar]
- 24.Yunyun F, Yu P, Panpan Z, Xia Z, Linyi P, Jiaxin Z. Efficacy and safety of low dose Mycophenolate mofetil treatment for immunoglobulin G4-related disease: a randomized clinical trial. Rheumatology. 2019;58(1):52–60. doi: 10.1093/rheumatology/key227 [DOI] [PubMed] [Google Scholar]
- 25.Lee HW, Moon S-H, Kim M-H, Cho DH, Jun JH, Nam K. Relapse rate and predictors of relapse in a large single center cohort of type 1 autoimmune pancreatitis: long-term follow-up results after steroid therapy with short-duration maintenance treatment. J Gastroenterol. 2018;53(8):967–977. doi: 10.1007/s00535-018-1434-6 [DOI] [PubMed] [Google Scholar]
- 26.Thomas LE, Li F, Pencina MJ. Overlap weighting: a propensity score method that mimics attributes of a randomized clinical trial. JAMA. 2020;323(23):2417. doi: 10.1001/jama.2020.7819 [DOI] [PubMed] [Google Scholar]
- 27.Mehta N, Kalra A, Nowacki AS, Anjewierden S, Han Z, Bhat P. Association of use of angiotensin-converting enzyme inhibitors and angiotensin II receptor blockers with testing positive for coronavirus disease 2019 (COVID-19). JAMA Cardiol. 2020;5(9):1020. doi: 10.1001/jamacardio.2020.1855 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 28.Li F, Thomas LE, Li F. Addressing extreme propensity scores via the overlap weights. Am J Epidemiol. 2019;188(1):250–257. doi: 10.1093/aje/kwy201 [DOI] [PubMed] [Google Scholar]
- 29.Patil I. Visualizations with statistical details: the “ggstatsplot” approach. J Open Source Softw. 2021;6(61):3167. doi: 10.21105/joss.03167 [DOI] [Google Scholar]
- 30.Cornell LD. IgG4-related kidney disease. Semin Diagn Pathol. 2012;29(4):245–250. doi: 10.1053/j.semdp.2012.07.004 [DOI] [PubMed] [Google Scholar]
- 31.Nishi S, Imai N, Yoshida K, Ito Y, Saeki T. Clinicopathological findings of immunoglobulin G4-related kidney disease. Clin Exp Nephrol. 2011;15(6):810–819. doi: 10.1007/s10157-011-0526-x [DOI] [PubMed] [Google Scholar]
- 32.Ling J, Wang H, Pan W, Li S, Guan J, Zhang X. Clinical and imaging features of IgG4-related kidney disease. Abdom Radiol N Y. 2020;45(6):1915–1921. doi: 10.1007/s00261-020-02477-8 [DOI] [PubMed] [Google Scholar]
- 33.Davies AM, Sutton BJ. Human IgG4: a structural perspective. Immunol Rev. 2015;268(1):139–159. doi: 10.1111/imr.12349 [DOI] [PMC free article] [PubMed] [Google Scholar]
- 34.Hu C, Zhang P, Li L, Liu C, Li J, Zhang W. Assessing serum IgG4 glycosylation profiles of IgG4-related disease using lectin microarray. Clin Exp Rheumatol. 2021;39(2):393–402. doi: 10.55563/clinexprheumatol/2i3uvr [DOI] [PubMed] [Google Scholar]
- 35.Danlos F-X, Rossi GM, Blockmans D, Emmi G, Kronbichler A, Durupt S. Antineutrophil cytoplasmic antibody-associated vasculitides and IgG4-related disease: a new overlap syndrome. Autoimmun Rev. 2017;16(10):1036–1043. doi: 10.1016/j.autrev.2017.07.020 [DOI] [PubMed] [Google Scholar]
- 36.Khan ML, Colby TV, Viggiano RW, Fonseca R. Treatment with bortezomib of a patient having hyper IgG4 disease. Clin Lymphoma Myeloma Leuk. 2010;10(3):217–219. doi: 10.3816/clml.2010.n.034 [DOI] [PubMed] [Google Scholar]
- 37.Carruthers MN, Topazian MD, Khosroshahi A, Witzig TE, Wallace ZS, Hart PA. Rituximab for IgG4-related disease: a prospective, open-label trial. Ann Rheum Dis. 2015;74(6):1171–1177. doi: 10.1136/annrheumdis-2014-206605 [DOI] [PubMed] [Google Scholar]
Associated Data
This section collects any data citations, data availability statements, or supplementary materials included in this article.
Data Availability Statement
All data are included in the manuscript and/or supporting information.
