Comparative Histological Subtyping of Immune Cell Infiltrates in MPO-ANCA and PR3-ANCA Glomerulonephritis

Background Acute kidney injury (AKI) is a common and severe complication of anti-neutrophil cytoplasmic antibodies (ANCA)-associated vasculitis (AAV), potentially leading to chronic kidney disease (CKD), end-stage renal disease (ESRD), or death. Pathogenic ANCAs, in particular proteinase 3 (PR3) and myeloperoxidase (MPO), trigger a deleterious immune response with intrarenal immune cell infiltration resulting in a pauci-immune necrotizing and crescentic glomerulonephritis (GN). However, a systematic analysis of intrarenal immune cell subtypes concerning neutrophils, eosinophils, plasma cells, and mononuclear cell infiltrates (macrophages, lymphocytes) in ANCA GN remains elusive. Therefore, we aimed to compare distinct immune cell infiltrates in association with clinicopathological findings in ANCA GN. Methods A total of 53 kidney biopsies with ANCA GN at the University Medical Center Göttingen were retrospectively analyzed. Histological infiltrates of neutrophils, eosinophils, plasma cells, and mononucleated cells (macrophages, lymphocytes) were quantified as a fraction of the total area of inflammation. Results Neutrophilic infiltrates were associated with glomerular necrosis and severe kidney injury in ANCA GN. Among tubulointerstitial lesions, intrarenal neutrophils correlated with interstitial inflammation, tubulitis, and inflammation in areas of interstitial fibrosis/tubular atrophy (IFTA), representing active inflammatory lesions. Concerning eosinophils, infiltrates were associated with severe kidney injury, interstitial inflammation, and cellular casts independent of glomerular lesions, implicating a distinct role in inflammation and damage in ANCA GN. Plasma cell infiltrates correlated with tubulitis and interstitial fibrosis and were associated with renal replacement therapy during the short-term disease course. Finally, mononuclear cell infiltrates correlated with severe kidney injury and active histopathological lesions (glomerular crescents, interstitial inflammation, tubulitis, inflammation, and tubulitis in areas of IFTA) besides chronic lesions (interstitial fibrosis and tubular atrophy) in ANCA GN. Interestingly, intrarenal subtypes of immune cell infiltrates differed in MPO-ANCA versus PR3-ANCA GN and were associated with distinct glomerular and tubulointerstitial lesions, implicating different pathogenic mechanisms of kidney injury in ANCA subtypes. Conclusion Our observations imply distinct pathomechanisms contributing to inflammation and renal injury in MPO vs. PR3-associated ANCA GN and potentially contribute to new therapeutic targets in specific ANCA subtypes.


INTRODUCTION
According to the 2012 revised Chapel Hill Consensus Conference Nomenclature of Vasculitides, anti-neutrophil cytoplasmic antibody (ANCA)-associated vasculitis (AAV) is a small vessel vasculitis, most frequently presenting as microscopic polyangiitis (MPA) or granulomatosis with polyangiitis (GPA) (1,2). Acute kidney injury (AKI) is a common and severe complication of AAV as it can cause progressive chronic kidney disease (CKD), end-stage renal disease (ESRD), or death (3,4). Pathogenic ANCAs, in particular proteinase 3 (PR3) and myeloperoxidase (MPO), trigger a deleterious immune response resulting in a pauci-immune necrotizing and crescentic glomerulonephritis (GN), a common manifestation of glomerular injury in AAV (5). Several studies have investigated determinants of renal outcomes in ANCA GN, including baseline kidney function, acute and chronic histopathological lesions (6)(7)(8)(9). On a mechanistic level, neutrophils are activated by pathogenic ANCAs, causing a release of inflammatory cytokines, reactive oxygen species, and lytic enzymes, resulting in an excessive formation of neutrophil extracellular traps (NETs) (10)(11)(12). Besides neutrophils, plasma cell infiltration is also commonly observed in ANCA GN and is associated with active and early disease (13). While neutrophil and plasma cell infiltrations predominate the early phase of inflammation in AAV, the neutrophil-rich necrotizing lesions are converted into a monocyte/macrophage-rich granuloma. This is accompanied by an infiltration of lymphocytes, such as T cells (14,15). We previously established that MPO-ANCA GN more frequently shows active tubulointerstitial lesions, including total inflammation (16). However, a systematic analysis of intrarenal immune cell subtypes with regard to neutrophils, eosinophils, plasma cells, and mononuclear cell infiltrates (macrophages, lymphocytes) in ANCA GN remains elusive. Therefore, we aimed to compare distinct immune cell infiltrates in association with clinicopathological findings in a cohort of patients with biopsy-proven ANCA GN.

Study Population
A total number of 53 kidney biopsies with ANCA GN at the University Medical Center Göttingen were retrospectively included between 2015 till 2020; the patient cohort has previously been described (16)(17)(18)(19)(20)(21). While no formal approval was required for the use of routine clinical data, a favorable ethical opinion was granted by the local Ethics committee (no. 22/2/14 and 28/09/17). At admission, the Birmingham Vasculitis Activity Score (BVAS) version 3 was assessed (22). Medical records were used to obtain data on age, sex, duration of disease onset before admission, diagnosis (MPA or GPA), and laboratory results. The estimated glomerular filtration rate (eGFR) was calculated using the Chronic Kidney Disease Epidemiology Collaboration (CKD-EPI) equation (23). When required, renal replacement therapy (RRT) was performed intermittently in all cases. Indications for RRT included severe electrolyte and acid-base abnormalities, volume overload, or encephalopathy. Glucocorticoids (GCs) were administered either as intravenous pulse therapy or orally with a tapering schedule. At time of kidney biopsy, all patients received GCs and further remission induction therapy was initiated thereafter based on histopathological confirmation of ANCA GN.

Statistical Methods
Variables were tested for normal distribution using the Shapiro-Wilk test. Statistical comparisons were not formally powered or prespecified. Non-normally distributed continuous variables are shown as the median and interquartile range (IQR), categorical variables are presented as frequency and percentage. Spearman's correlation was performed to assess the correlation between clinical, laboratory, and histopathological parameters, and heatmaps reflecting the mean values of Spearman's r are shown, the asterisks indicating statistical significance correlations. Data analyses were performed with GraphPad Prism (version 8.4.3 for macOS, GraphPad Software, San Diego, California, USA). Multiple regression analyses were performed using IBM SPSS Statistics (version 27 for MacOS, IBM Corporation, Armonk, NY, USA). A probability (p) value of <0.05 was considered statistically significant.  Table 1. In this cohort, the median age at diagnosis was 64 (54.25-73.75) years. All patients were Caucasian, and 20/48 (41.7%) were female. The median disease onset before admission was 17.5 (7-40) days; a kidney biopsy was performed within 6 (3-9) days RRT within 30 days after admission. Systematic histological scoring of intrarenal immune cell subtypes in ANCA GN revealed that a neutrophilic infiltrate was detectable in 16/48 (33.3%), an eosinophilic infiltrate in 10/48 (20.8%), a plasma cell infiltrate in 32/48 (66.7%), and a mononuclear (macrophages, lymphocytes) cell infiltrate in 43/48 (89.6%, Figure 2 and Table 2) of evaluable biopsies. To elucidate the association between infiltrates of immune cell subtypes, we next quantified the presence of neutrophils, eosinophils, plasma cells, and mononucleated cells as a fraction of the total area. There was a strong correlation of immune cell subtypes with each other; in particular, the presence of neutrophils correlated with eosinophils, plasma cells and mononuclear cells, and eosinophils with plasma cells ( Figure 3A), respectively. Because medication with a non-steroidal anti-inflammatory drug (NSAID), antibiotics, or proton-pump inhibitors (PPI) potentially triggers an interstitial inflammation in the kidney, we next analyzed a correlation between distinct intrarenal immune cell subtypes and these medications in addition to GC dose at time of kidney biopsy. We observed an association between NSAID use and neutrophilic (r=0.3074, p=0.0336) as well as eosinophilic cell infiltrates (r=0.4682, p=0.0008, Figure 3B). In summary, subtypes of immune cells in kidney biopsies were observed in a majority of ANCA GN cases. In addition, we observed an association between distinct immune cell infiltrates and prior NSAID use.

Distinct Clinical Parameters and Laboratory Markers in Association With Intrarenal Immune Cell Subtypes
We next analyzed intrarenal subtypes of immune cell infiltrates in association with clinical parameters, laboratory, and urinary markers in AAV. The only association with clinical parameters was observed between plasma cell infiltrates and less frequent sinus involvement in AAV (r=-0.2954, p=0.0415, Figure 4A). Neutrophilic, plasma cellular, and mononuclear infiltrates correlated with plasma creatinine (r=0.3217, p=0.0258, r=0.3745, p=0.0087, r=0.4608, p=0.0010, respectively) and eGFR loss (r=-0.3141, p=0.0297, r=-0.3330, p=0.0207, r=-0.4278, p=0.0024, Figure 4B). Interestingly, intrarenal infiltrates of plasma (r=0.3936, p=0.0056) and mononuclear cells (r=0.3345, p=0.0201) were also associated with severe kidney injury requiring RRT during the initial course of the disease ( Figure 4B). Multiple regression analysis comparing all immune cell infiltrates confirmed that presence of mononuclear cells was associated with severe kidney injury reflected by RRT requirement during the initial disease course ( Table 3), confirming an important of mononuclear cell infiltrates in the early phase of ANCA GN. In contrast, intrarenal immune cell subtypes did not correlate with urinary findings despite an association between plasma cellular infiltrates and urinary IgG levels ( Figure 4B). Furthermore, intrarenal eosinophils did not correlate with any clinical parameter, laboratory, or urinary marker in AAV. In summary, systematic scoring of intrarenal immune cell subtypes revealed that distinct immune cell types were specifically associated with a deterioration of kidney  function independent of clinical parameters, laboratory or urinary markers in AAV.

DISCUSSION
We have previously shown that MPO-ANCA GN is associated with more severe deterioration of kidney function independent of systemic markers of AAV disease activity, associated with more interstitial vasculitis and total inflammation, as well as interstitial fibrosis and tubular atrophy (16). In addition, we previously observed tubulointerstitial inflammation was associated with presence of Bowman's capsule ruptures, independently described to correlate with poor outcome in ANCA GN (20,28,29). Based on these previous observations that interstitial inflammation is frequently observed in the majority of ANCA GN cases, we here aimed to provide a comprehensive analysis of histological subtyping of immune cell infiltrates in association with clinicopathological findings in ANCA GN. At disease onset, neutrophils are activated by pathogenic ANCAs, causing a release of inflammatory cytokines, reactive oxygen species, and lytic enzymes, thus resulting in an excessive formation of neutrophil extracellular traps (NETs) (10)(11)(12). Besides neutrophils, plasma cell infiltration is also commonly observed in ANCA GN and has been associated with active and early disease (13). Intrarenal infiltrates of primed neutrophils are activated by MPO-ANCA and PR3-ANCA, leading to degranulation and the release of cytoplasmic granules into the glomerular and interstitial space in ANCA GN. As a result, reactive oxygen radicals (RORs) accumulate and cause vascular damage (11). In addition, stimulation of neutrophils causes the release of factors that activate the alternative complement system, aggravating vascular damage (30). Small vessel leakage of serum proteins and the formation of fibrin gives rise to fibrinoid necrosis (31). While neutrophil and plasma cell infiltrates predominate the early phase of inflammation in AAV, the neutrophil-rich necrotizing lesions are converted into a monocyte/macrophage-rich granuloma, and this is accompanied by infiltration of lymphocytes, such as T cells (14,15). This concept is in line with our observation that neutrophilic infiltrates were associated with glomerular necrosis and severe kidney injury in ANCA GN. Among tubulointerstitial lesions, intrarenal neutrophils correlated with interstitial inflammation, tubulitis, and inflammation in areas of IFTA representing active inflammatory lesions, again confirming the concept that activated neutrophils mediate inflammation and injury in the early phase of ANCA GN. Eosinophils do not contain MPO or PR3 but can be activated by ANCA-activated neutrophils. With regard to eosinophils, infiltrates were associated with cellular casts independent of glomerular lesions, implicating a distinct role in inflammation and tubular injury in ANCA GN. The pathogenic role of plasma cells in ANCA GN remains unclear, but local antibody production in the kidney may also contribute to interstitial inflammation in ANCA GN. Based on previous reports, B cell autoimmune response in AAV is facilitated by impaired T and B cell regulation and by B cell-stimulating factors released by activated neutrophils (32). Plasma cell-rich ANCA GN is frequently observed and has previously been associated with tubulointerstitial inflammation, implicating that plasma cell infiltrates may contribute to tubulitis (13,33). This is confirmed by our observation that plasma cell infiltrates correlated with tubulitis and interstitial fibrosis. Finally, mononuclear cell infiltrates, including macrophages and lymphocytes, such as T cells, are found in ANCA GN (14,15). Macrophages are a predominant immune cell subtype in ANCA GN infiltrating normal glomeruli and present in developing glomerular lesions (34). The presence of macrophages suggests an important role in renal injury and the recruitment of T cells, which dominate later stages in ANCA GN (35). Previous studies have shown that monocytes in ANCA GN localize to sites of active glomerular lesions, including fibrinoid necrosis, cellular crescents, and periglomerular inflammation (36). In addition, intrarenal macrophages display positivity for MPO and produce extracellular traps containing MPO, further contributing to MPO antigen presentation and aggravating renal injury (37). This concept concurs with our observation that mononuclear cell infiltrates not only correlate with chronic lesions, including interstitial fibrosis and tubular atrophy (suggesting a role in clearing injured tissue), but also correlated with active histopathological lesions including glomerular crescents, interstitial inflammation, and tubulitis (thereby probably contributing to active tubulointerstitial damage). Furthermore, presence of mononuclear cells was associated with severe kidney injury and RRT requirement during the initial disease course in multivariate comparison of immune cell infiltrates, confirming an important role also in the early phase of ANCA GN. Interestingly, intrarenal subtypes of immune cell infiltrates were associated with distinct glomerular and tubulointerstitial lesions and differed in MPO-ANCA versus PR3-ANCA GN, implicating that distinct signatures of immune cell infiltrates contribute to kidney injury in these two subtypes of the disease. The only association between medication and immune cell infiltration was observed for prior NSAID use and intrarenal eosinophils, further supporting that distinct immune cell infiltrates are predominantly attributed to ah, arteriolar hyalinosis; ANCA, anti-neutrophil cytoplasmic antibodies; ATI, acute tubular injury; BVAS, Birmingham Vasculitis Activity Score; ci, interstitial fibrosis; CRP, C-reactive protein; ct, tubular atrophy; eGFR, estimated glomerular filtration rate; g, glomerulitis; GN, glomerulonephritis; i, interstitial inflammation; IgG, immunoglobulin G; i-IFTA, inflammation in IFTA; MPA, microscopic polyangiitis; MPO, myeloperoxidase; RBC, red blood cell; RRT, renal replacement therapy; t, tubulitis; ptc, peritubular capillaritis; ti, total inflammation; t-IFTA, tubulitis in IFTA; uACR, urinary albumin-to-creatinine ratio; uPCR, urinary protein-to-creatinine ratio; v, intimal arteritis.
ANCA GN. These important observations are of relevance and require further investigation with regard to distinct pathomechanisms contributing to inflammation and renal injury in ANCA GN, potentially generating knowledge on new therapeutic targets in specific ANCA subtypes.
The main limitations of our study are its retrospective design, the small patient number, and no data on efficacy of remission induction therapy or long-term renal outcomes. We here aimed to systematically analyze histological subtyping of immune cell infiltrates in ANCA GN and specifically chose univariate correlations of distinct immune cell infiltrates with clinicopathological findings. However, multivariate comparisons and analysis of immune cell infiltrates adjacent to distinct histopathological lesions including previously described Bowman's capsule rupture would also be of great interest and requires further investigation (20). Nevertheless, we here provide a novel systematic analysis of distinct subtypes of intrarenal immune cell infiltrates in association with clinical and histopathological findings, including glomerular and tubulointerstitial lesions in ANCA GN.

DATA AVAILABILITY STATEMENT
The original contributions presented in the study are included in the article/supplementary material. Further inquiries can be directed to the corresponding author.

ETHICS STATEMENT
The studies involving human participants were reviewed and approved by the University Medical Center Göttingen, Germany.
The patients/participants provided their written informed consent to participate in this study. Informed written consent was obtained from all subjects involved in the study for the use of routinely collected data for research purposes as part of their regular medical care in the contract of the University Medical Center Göttingen.

AUTHOR CONTRIBUTIONS
BT conceived the study, collected and analyzed data, and wrote the first draft. SH and DT collected and analyzed data. SH and PS evaluated histopathological findings. PK analyzed data and edited the manuscript. All authors contributed to the article and approved the submitted version.