Diagnostic accuracy of anti-phospholipase A2 receptor (PLA2R) antibodies in idiopathic membranous nephropathy: an Italian experience
Brunetta Porcelli, Andrea Guarnieri, Fabio Ferretti, Guido Garosi, Lucia Terzuoli, Francesca Cinci, Antonella Tabucchi, Marilina Tampoia, Letizia Abbracciavento, Chiara Villani, Gaia Deleonardi, Ana Gabriela Grondona, Marcello Mazzolini, Gaetano La Manna, Marisa Santostefano, María Infantino, Mariangela Manfredi, Giuseppe Spatoliatore, Alberto Rosati, Chiara Somma, Nicola Bizzaro
Abstract
BACKGROUND: Autoantibodies against-phospholipase A2 receptor (PLA2R) are specific markers of idiopathic membranous nephropathy (iMN). Enzyme-linked immunosorbent assay (ELISA) is becoming the preferred method in many laboratories for the determination of anti-PLA2R antibodies, because it provides quantitative results, and is not prone to subjective interpretation, as is the case with indirect immunofluorescence assay. METHODS: The purpose of our study was to determine the diagnostic performance of serum PLA2R antibodies detected by commercially available ELISA in a large Italian multicenter cohort of patients with biopsy-proven iMN and in patients with other renal diseases, with special focus on evaluating the optimal cut-off value to discriminate positive and negative results. A total of 495 consecutive patients were recruited. Renal biopsies were performed in all patients, and blood samples were taken before the initiation of immunosuppressive treatment. RESULTS: According to the clinical diagnosis and to kidney biopsy, 126 patients were diagnosed with iMN and 369 had other non-membranous nephropathies. Anti-PLA2R autoantibodies were detected using a commercial anti-PLA2R ELISA. At a cut-off value of 20 relative units (RU)/ml indicated by the manufacturer for positive classification, sensitivity was 61.1% and specificity 99.7%. At a cut-off value of 14 RU/ml indicated by the manufacturer for borderline results, sensitivity was 63.5% and specificity remained the same (99.7%). At a cut-off of 2.7 RU/ml, selected as the optimal cut-off on the basis of ROC curve analysis, sensitivity was 83.3% and specificity 95.1%. The best overall efficiency of the test was observed at 2.7 RU/ml; however, the highest positive likelihood ratio and diagnostic odds ratio were achieved at 14 RU/ml. A cut-off threshold higher than 14 RU/ml or lower than 2.7 RU/ml entailed worse test performance. CONCLUSION: Depending on the clinical use (early diagnosis or as a support to confirm clinical diagnosis), nephrologists may take advantage of this evidence by choosing the most convenient cut-off. However, renal biopsy remains mandatory for the definitive diagnosis of iMN and for the assessment of disease severity.