Is it possible to predict parameters of the Oxford classification of primary IgA Nephropathy from clinical laboratory data? Focus on the role of segmental glomerulosclerosis subtypes

Introduction: IgA nephropathy (IgAN) is the most common primary glomerulonephritis in the world and has a broad range of histological and clinical manifestations, ranging from morphologically normal to globally sclerotic glomeruli with clinical manifestations varying from isolated hematuria to end stage renal disease. This study aims to assess sensitivity, specificity and accuracy of clinical data at the time of biopsy in predicting 2017 updated Oxford classification parameters and to investigate if subtypes of segmental sclerosis (FSGS) influence clinical presentation.

Material and methods: Renal biopsies from 103 patients with IgAN were analyzed. Oxford classification was updated and FSGS lesions were subclassified. ROC curves, univariate and multivariate logistic regression were used.

Results: In Oxford classification, the majority of patients had mesangial hypercellularity in less than a half of glomeruli (M0), did not have endocapillary hypercellularity (E0), had segmental glomerulosclerosis (S1), had interstitial fibrosis and tubular atrophy in more than a half of the sample (T2) and had no crescents (C0). Hypertension increases the chance of M1 in 2.54x (p = 0.02). For each unit of increased creatinine, 2.6x more chances of E1 (p = 0.001). S1 is predicted by proteinuria with 75% sensitivity and 90.9% specificity (p < 0.0001). For each unit of increase in GFR, there is a reduction of 6% in the chance of T2 in relation to T0 (p = 0.0001). If hypertension, there is 5x more chances of T2 than T0 (p = 0.01). For each unit of increase in creatinine, there are 2.8x more chances of crescents- C (p = 0.003). Creatinine also showed 75.8% sensitivity and 75% specificity for prediction of C (p = 0.002). Inversely, for each unit of GFR, the chance of C is reduced by 4% (p = 0.007). Other clinical data related with C are hypertension (p = 0.03) and proteinuria (p = 0.02). To determine the role of FSGS subtypes in clinical presentation, we divided patients in S0 and S1 groups. Proteinuria was the only clinical parameter with significative difference, respectively, 0.3 (0-2.1) and 1.6 (0.02-16.2) g/24 h (p < 0.0001). FSGS subtypes related to proteinuria were cellular (p = 0.03) and peri-hilar (p = 0.02). Subtypes classically related to podocytopathies showed no correlation with clinical data.

Conclusion: In the future, with noninvasive methods for diagnosis of IgAN, it will be essential to predict Oxford classification parameters using clinical laboratory data for establishment of prognosis and therapeutics. We showed that Oxford classification parameters correspond to some clinical laboratory data, making this approach possible. FSGS lesions not specifically related to podocytopathies may also influence clinical parameters that affect renal disease progression.