#4570 RELATION OF KIDNEY BIOPSY FINDINGS TO CLINICAL OUTCOMES IN PATIENTS WITH LUPUS NEPHRITIS TREATED WITH AN ANTAGONISTIC ANTI-CD40 ANTIBODY (BI 655064)

Abstract

Listen

Translate article

Abstract Background and Aims Lupus nephritis (LN) is the most serious complication of systemic lupus erythematosus with a high risk of end-stage kidney disease and a high need of better treatment. A recent randomized, double-blind, placebo-controlled, phase II trial tested doses of 120, 180 and 240 mg of the anti-CD40 monoclonal antibody BI 655064 as add-on to standard of care (MMF + glucocorticoids), in patients with active LN (proteinuria > 1 g/day) for 52 weeks. Although the trial did not demonstrate a dose-response relationship for the primary endpoint determined by complete renal response, a potential efficacy benefit of BI 655064 180 mg and 240 mg was found in a post-hoc analysis. We investigated whether a treatment effect was related to findings in the pre-treatment renal biopsy. Method Biopsies from 101 patients with LN class III or IV, confirmed in a central review conducted according to a previously established protocol, were included. Protein/creatinine ratio (UPCR, based on spot urine) and eGFR were determined at every visit. For each, the last available UPCR and the last available eGFR values, linear regressions (based on the respective baselines) of all patients irrespective of therapy were calculated, and subjects were classified according to their individual performance, “Better” or “Worse” than the average. The modified activity and chronicity indices as well as all biopsy parameters, which are not already contained in the indices, were compared for differences between the “Better” and “Worse”. Parameters with p< 0.1 in the univariate analyses were entered into a multivariate logistic regression model, and a routine model selection procedure (with p< 0.2) was used to identify parameters predictive for clinical outcome. As monocytes express CD40, which is the target of BI 655064, a further question was whether high dose treatment with BI 655064 (180/240 mg) was beneficial when monocytes were present in the biopsy, using a model which adjusted for all predictors identified in the multivariate model. Results The results of the univariate and the multivariate analysis of biopsy parameters are shown in Table 1. Improvement in UPCR was better if the modified chronicity index was low. In addition, LN class IV subjects had a marginally better prognosis than class III. The effect of high-dose treatment with BI 655064 was better when monocytes were present. eGFR outcome was better if the modified chronicity index was low and, conversely, the modified activity index was high. Further, the presence of synechia/adhesions was a negative predictor for positive outcome. There was no indication for an effect of high-dose treatment with 655064 on eGFR when monocytes were present. Conclusion This post-hoc analysis of the BI 655064 trial used a novel approach which investigated whether specific histological lesions in renal biopsies were related to treatment outcome. Although the number of patients was limited, our results suggest that treatment with higher dose anti-CD40 may improve the reduction of proteinuria when monocytes are present in the biopsy. This indicates that specific information from renal biopsies could ultimately improve the choice of treatment for individual LN patients.