Clinical and histological significance of urinary CD11c+ macrophages in lupus nephritis

Abstract

Background: Infiltration of immune cells into the kidney is one of the key features of lupus nephritis (LN). The presence of immune cells in the urine may be used as a non-invasive biomarker of LN. Here, we aimed to analyze the clinicopathologic significance of urinary CD11c⁺ macrophages in patients with LN.

Methods: The numbers and proportions of CD11c⁺ macrophages in the urine samples of patients with LN at the time of kidney biopsy were examined using flow cytometry. We also examined the association between the levels of urinary CD11c⁺ macrophages and the clinical and pathologic features of patients with LN.

Results: Compared with patients without LN or those with non-proliferative LN, patients with proliferative LN had significantly higher numbers and proportions of urinary CD11c⁺ macrophages, which were strongly correlated with the serum anti-dsDNA antibody titer. The numbers and proportions of urinary CD11c⁺ macrophages were significantly associated with the values of chronicity indices such as tubular atrophy and interstitial fibrosis. No significant relationships were found between the levels of urinary CD11c⁺ macrophages and the activity scores, degree of proteinuria, or lupus disease activity. Urinary CD11c⁺ macrophages were more abundant in patients who did not achieve renal response to induction treatment with immunosuppressants than in those who achieved complete or partial response. The receiver operating characteristic (ROC) curve analysis showed that the number of urinary CD11c⁺ macrophages was the most powerful predictor of renal response at 6 months (ROC-AUC = 1.00, p = 0.0004).

Conclusion: The urinary levels of CD11c⁺ macrophages were closely associated with the chronic pathologic changes of LN and renal response and may thus be used as a novel biomarker in LN.

Keywords: Lupus nephritis; Macrophage; Tubulointerstitial change; Urine.

Fig. 1

CD11c⁺ macrophages in the urine of patients with lupus nephritis (LN). a Numbers and proportions of urinary CD11c⁺ macrophages among CD45⁺ cells in patients with inactive systemic lupus erythematosus (SLE), active SLE, active LN (proliferative and non-proliferative), inactive LN, or other diseases with nephropathy (i.e., IgA nephropathy, ANCA-associated vasculitis) (n = 10, 5, 14, 27, 5, and 6, respectively). b The numbers and proportions of urinary CD19⁺ B cells and CD3⁺ T cells in SLE patients with non-proliferative LN or proliferative LN (n = 25). c Correlation between the numbers and proportions of CD11c⁺ macrophages in CD45⁺ cells and serum anti-dsDNA antibody titers in active LN patients (n = 41). d Correlation between serum anti-dsDNA antibody titers, and numbers and proportions of urinary CD19⁺ B or CD3⁺ T cells in LN patients (n = 25). *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001

Fig. 2

The numbers of urinary CD11c⁺ macrophages and clinicopathologic features of proliferative lupus nephritis (LN). a The numbers of urinary CD11c⁺ macrophages in patients with proliferative LN (n = 27) according to the presence of tubular atrophy, interstitial fibrosis, glomerular crescents, and glomerular sclerosis. b Correlation between the numbers of urinary CD11c⁺ macrophages, and chronicity, activity scores, the amounts of proteinuria, and disease activity in patients with proliferative LN (n = 27). c The numbers of urinary CD11c⁺ macrophages in patients with proliferative LN according to renal response to immunosuppressants (no [−], partial/complete [+]) (n = 23)

Fig. 3

The proportions of urinary CD11c⁺ macrophages and clinicopathologic features of proliferative lupus nephritis (LN). a The proportions of urinary CD11c⁺ macrophages among CD45⁺ cells in patients with proliferative LN (n = 27) according to the presence of tubular atrophy, interstitial fibrosis, glomerular crescents, and glomerular sclerosis. b Correlation between the proportions of urinary CD11c⁺ macrophages, and chronicity, activity scores, the degrees of proteinuria, and disease activity in patients with proliferative LN (n = 27). c Difference in the proportions of urinary CD11c⁺ macrophages between patients with and without renal response to immunosuppressants (no [−], partial/complete [+]) (n = 23)

Fig. 4

Receiver operating characteristic (ROC) curve of the factors to predictive renal response in proliferative LN. ROC curves of the predictive values of urinary CD11c⁺ macrophages, T cells, chronicity and activity scores, eGFR, C3, C4, and anti-dsDNA antibody titer values for no renal response at 6 months after treatment (n = 23)