Inflammation in areas of fibrosis precedes loss of kidney function in lupus nephritis

Abstract

Background: Interstitial fibrosis in lupus nephritis (LN) is often infiltrated by immune cells but typically regarded as non-specific 'scar reaction'. This study aimed to investigate the relationship between inflammatory fibrosis and kidney disease progression in LN.

Methods: Interstitial fibrosis and tubular atrophy (IFTA) were scored in 124 LN kidney biopsies. Inflammation in areas of IFTA (i-IFTA) was graded 0-3 according to the Banff Classification of Allograft Pathology. Significant glomerular filtration rate (GFR) loss was defined as a decline of >15 mL/min at 3 years from biopsy. Immune cell phenotype was defined by serial immunohistochemistry (13-plex).

Results: IFTA was observed in 88/124 (71%) biopsies, and i-IFTA was identified in 76/88 (86%) cases. The distribution of i-IFTA grades was heterogeneous across all IFTA grades. In patients with moderate-to-severe IFTA (>25%), the degree of i-IFTA was associated with a higher risk of significant GFR loss: 0/1 (0%), 0/3 (%), 3/4 (75%) and 7/9 (78%) for i-IFTA grades 0, 1, 2 and 3, respectively (p=0.015). Multiplexed histology revealed that i-IFTA was mostly composed of CD163+ macrophages and CD4 T cells, followed by CD8 T cells and granulocytes.

Conclusion: I-IFTA is frequently observed in LN and is dominated by macrophages and T cells. For patients with baseline IFTA >25%, the degree of i-IFTA emerged as a predictor of GFR loss. These data support the routine scoring of i-IFTA in LN due to its prognostic implications and nominate i-IFTA as a potential therapeutic target.

Keywords: Inflammation; Lupus Nephritis; Risk Factors.

Figure 1. Inflammation in areas of interstitial fibrosis and tubular atrophy (i-IFTA) is frequently observed in lupus nephritis (LN). (A) Representative kidney biopsy showing extensive interstitial fibrosis (light blue stain by Trichrome) with interspersed tubular atrophy (iFTA). (B) Representative histological image showing an inflammatory infiltrated in areas of IFTA (i-IFTA). (C) Grading of IFTA and i-IFTA according to the percentage of fibrotic cortex for IFTA and the percentage of fibrotic cortex with inflammatory infiltrate for i-IFTA as per Banff criteria. (D) Distribution of i-IFTA grades (0–3) according to IFTA grades in 124 LN biopsies showing high frequency and heterogeneity in the degree of i-IFTA in patients with IFTA.

Figure 2. Risk of future GFR loss according to i-IFTA. Distribution of patients who developed significant loss of GFR (>15 mL or ESKD) within 3 years of lupus nephritis (LN) biopsy according to i-IFTA in patients with mild (A) or moderate-severe (B) IFTA. The results in (A and B) are reported as percentages in (C and D). P for trend was confirmed by an exact test by permutation (A, p=ns; B, p=0.0179). eGFR, estimated glomerular filtration rate; ESKD, end-stage kidney disease; i-IFTA, inflammation in areas of interstitial fibrosis and tubular atrophy.

Figure 3. Immune cell composition of i-IFTA. Using 13-plex serial IHC, seven immune cell clusters were identified and annotated based on their marker expression. (A) UMAP of 8875 kidney-infiltrating immune cells (including all kidney regions) from four LN biopsies (2 ISN Class III, 1 pure V and 1 VI). (B) Pie chart showing the distribution of the immune cell infiltrate in i-IFTA. (C) Density of immune cell populations infiltrating areas of IFTA and tubulointerstitial regions without IFTA in four LN kidney biopsies. (D) Representative multichannel image showing the difference in immune infiltrate density in IFTA. Pseudocolours from five selected markers are shown. Significance: *p=0.05; **p<0.05; ***p<0.01. IHC, immunohistochemistry; i-IFTA, inflammation in areas of interstitial fibrosis and tubular atrophy; LN, lupus nephritis; MPO, myeloperoxidase; ns, not significant; UMAP, Uniform Manifold Approximation and Projection.