Background Lupus nephritis (LN) is a major organ manifestation of Systemic Lupus Erythematosus (SLE). Renal biopsy is the current gold standard for assessment of renal pathology in LN, as non-invasive reliable markers are lacking. Ferritin is found in most tissues as a cytosolic protein that stores iron. Small amounts are secreted into the serum, where ferritin functions as an iron carrier. Ferritin is considered an acute phase protein, as it increases in response to stresses. Studies have reported elevated serum concentrations in patients with active compared to patients with inactive SLE, and also associations with LN.
Objectives We assessed serum ferritin levels in patients with LN, in order to clarify its role in this SLE subset.
Methods Serum levels of ferritin were determined by ELISA (Raybiotech Inc.) in 64 patients with biopsy-proven active LN (52 proliferative LN, PLN; 12 membranous LN, MLN), before and after completion of induction treatment. Post-treatment renal biopsies were performed after a median time of 7.7 months. Clinical responders (CR) were defined by ≥50% reduced proteinuria, normal or ≥25% improved estimated glomerular filtration rate, and inactive urinary sediment. Histopathological responders (HR) were defined by ≥50% improvement of renal Activity Index in post-treatment biopsies.
Results Serum levels of ferritin decreased following induction therapy for LN in the entire cohort (p<0.001) and in patients with PLN (p<0.001), but not in patients with MLN (p=0.35). In the entire cohort, ferritin levels decreased in both CR (p<0.001) and clinical non-responders (CNR; p=0.030). They also decreased in HR (p<0.001), but remained stable in histopathological non-responders (HNR; p=0.06). In PLN, ferritin levels decreased in CR (p<0.001) and in CNR (p=0.003), as well as in HR (p<0.001) and HNR (p=0.011). In MLN, ferritin levels decreased in CR (p=0.03) and HR (p=0.046), but neither in CNR (p=0.69) nor in HNR (p=0.89). In patients with MLN, both baseline and post-treatment ferritin levels were higher in CNR (baseline median: 172.9 ng/mL, range: 15.9–486.3; post-treatment median: 309.7 ng/mL, range: 8.4–714.8) than in CR (baseline median: 59.8 ng/mL, range: 24.4–200.9; post-treatment median: 32.3 ng/mL, range: 4.9–43.2; p=0.03 and p=0.005, respectively); however, they did not differ between HR and HNR (p=0.79 and p=0.43, respectively).
Conclusions Our data suggest ferritin as a marker of histopathological response to induction treatment in patients with LN. In patients with MLN in particular, high baseline ferritin levels predicted unfavourable clinical response to induction treatment, and, despite being initially lower, baseline ferritin levels decreased in CR. This implies a role of ferritin in the pathogenesis of this LN subset and warrants further investigation.
Disclosure of Interest None declared