ANCA-associated glomerulonephritis. Prognostic factors and outcome in a Norwegian cohort

Abstract

Objective. The objective of this dissertation is to (I) explore the temporal survival and identify prognostic factors in patients with antineutrophil cytoplasmic antibody (ANCA)-associated vasculitis with renal affection, (II) validate a recently proposed histopathological classification model, and (III) quantify the incidence of malignancy in this patient group. Methods. Patients with biopsy-proven pauci-immune crescentic glomerulonephritis and positive ANCA serology were included from the Norwegian Kidney Biopsy Registry from (I/III) 1988 through 2012 and (II) 1991 through 2012. Using the unique 11-digit personal number, end-stage renal failure and death were identified by record-linkage with the Norwegian Renal Registry and the Norwegian Population Registry. The cause of death was obtained from the Cause of Death Registry, Statistics Norway. (II) All biopsies were scored according to the histopathological classification model (focal, mixed, crescentic, and sclerotic class) by an experienced nephropathologist. A receiver operator characteristic curve was calculated to test the performance of the classification model. (III) The incidence of cancer was identified by cross-linkage with the Norwegian Cancer Registry. Results. (I) Four hundred and fifty-five patients were included. One hundred and twenty-four (27%) developed end-stage renal disease, and 165 (36%) deaths occurred. Independent negative predictors were low baseline estimated-glomerular filtration rate < 15 ml/min per 1.73 m2, older age (> 60 years) and baseline serum albumin < 30 g/L. Patients diagnosed between 2003 and 2012 had better baseline renal function than those diagnosed between 1988 and 2002 (27 versus 37 ml/min per 1.73 m2) and more favourable end-stage renal disease-free survival (1-year risk, 19% versus 13%; 10-year risk, 37% versus 26%). Compared to the general population, the risk of mortality was 2.8-fold higher (95% confidence interval, 2.4 to 3.3). (II) The histopathologic classification model was validated in 250 patients. The model could successfully predict renal survival, with the focal and sclerotic classes having the best and worst outcome (hazard ratio, 9.65; 95% confidence interval, 2.4 to 39.2), respectively. The area under the curve in the receiver operator characteristic curve was calculated to be 0.72 (95% confidence interval, 0.7 to 0.8). (III) Four hundred and nineteen patients were included in the study cohort. The standardised incidence ratio for the development of at least one cancer in this patient cohort compared to the general population was 1.09 (95% confidence interval, 0.8 to 1.5). There were large number of cases of non-melanoma skin cancers and bone marrow malignancies, with a standardised incidence ratio of 3.40 (95% confidence interval, 1.6 to 7.1) and 3.52 (95% confidence interval, 1.3 to 9.4), respectively. Conclusion. (I) Although the prognosis improved in patients with ANCAassociated glomerulonephritis, mortality remains high compared to the general population. Improvement in prognosis might be due to earlier diagnosis before irreversible organ damage occurs. (II) The histopathological classification model is of average quality as a predictor of end-stage renal disease. Patients with kidney biopsies classified as sclerotic class are particularly at risk of developing end-stage renal disease. (III) The risk of malignancy was not significantly increased in this patient cohort compared to the general population. Non-melanoma skin cancer was the most frequent malignancy, and a major contributor to the observed overall cancer occurrence in this patient population.