Prognostic value of endocapillary hypercellularity in IgA nephropathy patients with no immunosuppression

© 2015, Italian Society of Nephrology. Aim: Interpretation of retrospective clinicopathological studies of IgA nephropathy (IgAN) has been confounded by immunosuppression bias. In published validation studies of the Oxford Classification of IgAN, an average of 33 % of patients received non-randomised steroid and/or cytotoxic therapy. In order to determine the true impact of proliferative lesions on the natural history of IgAN, analysis of patient cohorts that have received no immunosuppression is required. Methods: We performed a retrospective single centre study of patients with IgAN managed without immunosuppressive therapy. Biopsies were scored according to the Oxford Classification. The primary outcomes were renal survival or a rapid loss of renal function defined as a decline in eGFR of >5 ml/min/year. Results: 237 patients with IgAN were identified with a mean follow-up of 82 months. 200 had biopsies available for review, of which 156 were adequate for scoring using the Oxford Classification. 9/156 patients (5.8 %) received some immunosuppressive therapy, mostly for unrelated conditions: these were excluded. In multivariate COX regression, including histological and clinical data, the only independent predictors of time to ESRD were baseline eGFR (HR 0.96 per ml/min increase, p = 0.018), baseline proteinuria (HR 1.36 per doubling, p = 0.004) and endocapillary hypercellularity (HR 4.75 for E1 compared to E0, p <0.001). Independent predictors of a rapid decline in eGFR were proteinuria (OR 1.45 per doubling, p = 0.006), endocapillary hypercellularity (OR 3.41 for E1 compared to E0, p = 0.025) and tubular atrophy/interstitial fibrosis (OR 8.77 for T2 compared to T0, p = 0.006). Conclusions: In a cohort of IgAN patients receiving no immunosuppression, endocapillary proliferation and tubular atrophy/interstitial fibrosis are independent predictors of rate of loss of renal function. The lack of predictive value of E score in other clinicopathological studies is most likely a result of immunosuppression-associated bias. Our findings provide evidence to support immunosuppressive treatment of endocapillary-pattern IgAN.