Antiphospholipid Antibodies in Lupus Nephritis

Abstract

Lupus nephritis (LN) is a major manifestation of systemic lupus erythematosus (SLE). It remains unclear whether antiphospholipid antibodies (aPL) alter the course of LN. We thus investigated the impact of aPL on short-term and long-term renal outcomes in patients with LN. We assessed levels of aPL cross-sectionally in SLE patients diagnosed with (n = 204) or without (n = 294) LN, and prospectively in 64 patients with active biopsy-proven LN (52 proliferative, 12 membranous), before and after induction treatment (short-term outcomes). Long-term renal outcome in the prospective LN cohort was determined by the estimated glomerular filtration rate (eGFR) and the Chronic Kidney Disease (CKD) stage, after a median follow-up of 11.3 years (range: 3.3-18.8). Cross-sectional analysis revealed no association between LN and IgG/IgM anticardiolipin or anti-β2-glycoprotein I antibodies, or lupus anticoagulant. Both aPL positivity and levels were similar in patients with active LN and non-renal SLE. Following induction treatment for LN, serum IgG/IgM aPL levels decreased in responders (p<0.005 for all), but not in non-responders. Both at active LN and post-treatment, patients with IgG, but not IgM, aPL had higher creatinine levels compared with patients without IgG aPL. Neither aPL positivity nor levels were associated with changes in eGFR from either baseline or post-treatment through long-term follow-up. Moreover, aPL positivity and levels both at baseline and post-treatment were similar in patients with a CKD stage ≥3 versus 1-2 at the last follow-up. In conclusion, neither aPL positivity nor levels were found to be associated with the occurrence of LN in SLE patients. However, IgG aPL positivity in LN patients was associated with a short-term impairment of the renal function while no effect on long-term renal outcome was observed. Furthermore, IgG and IgM aPL levels decreased following induction treatment only in responders, indicating that aPL levels are affected by immunosuppressive drugs in a response-dependent manner.

Fig 1. Creatinine levels (μmol/L) in LN patients with and without IgG aPL.

At baseline, creatinine levels were higher in LN patients with (n = 8) versus without (n = 56) IgG aCL (A; median: 104.5 μmol/L, range: 64–185, versus 77.0 μmol/L, range: 46–284; p = 0.03). Consistently, creatinine levels were higher in LN patients with (n = 9) versus without (n = 55) IgG anti-β₂-GPI (B; median: 94.0 μmol/L, range: 64–18, versus 75.0 μmol/L, range: 46–284; p = 0.02). Similar findings were observed post-treatment, with higher creatinine levels in LN patients with (n = 6) versus without (n = 58) IgG aCL (C; median: 86.5 μmol/L, range: 72–128, versus 75.5 μmol/L, range: 40–306; p = 0.04), as well as with (n = 6) versus without (n = 58) IgG anti-β₂-GPI (D; median: 86.5 μmol/L, range: 72–128, versus 75.5 μmol/L, range: 40–306; p = 0.04). Bounds of the boxes denote the 25^th and 75^th percentiles (IQR). Lines in the boxes denote the 50^th percentile (median). Whiskers denote the range. Circles (out values, 1.5–3 IQRs further from the closest box bound) and stars (far out or extreme values, ≥3 IQRs further from the closest box bound) denote outliers. Some extreme values do not appear in the figure due to scaling. LN: lupus nephritis; aCL: anticardiolipin antibodies; anti-β₂-GPI: anti-β₂-glycoprotein I antibodies.