#2683 Characterizing the NIH activity and chronicity indices in two independent lupus nephritis cohorts: a critical analysis

Abstract Background and Aims The inclusion of National Institutes of Health (NIH) activity and chronicity indices (AI and CI) in the International Society of Nephrology/Renal Pathology Society (ISN/RPS) classification of lupus nephritis (LN) aims to provide a more precise characterization of the amount of active and chronic lesions next to the lupus histological class. These indices were based on expert opinion over 40 years ago and their clinical usefulness today is uncertain. The NIH AI is based on six lesions (range 0–24) and the NIH CI on four lesions (range 0–12). The aim of this study is to investigate how the NIH AI and CI scores are represented in two large international LN cohorts; which lesions mostly determine the value of the indices; and what their relation is to LN histological class. Method We collected renal biopsies of patients diagnosed with LN, confirmed by light microscopy and immunofluorescence. Cohort 1 included 90 patients from Italy and The Netherlands, while cohort 2 comprised 104 patients from the UK (Table). We excluded biopsies that had fewer than 10 scorable glomeruli, those from transplanted kidneys, or with a concomitant glomerular disease. We calculated the NIH AI and CI according to the 2018 ISN/RPS classification revision. We used the Spearman coefficient to explore correlations among the NIH indices and the active and chronic lesions. Mann Whitney U test was used to investigate differences in NIH AI and CI distribution between classes III and IV and other classes (I, II and V), and between adult and pediatric populations. Results The NIH AI score was usually medium-low, reaching a maximum value of 16 in both cohorts, while the NIH CI had a more even distribution and reached a maximum value of 10. The NIH AI score was significantly higher in classes III and IV compared to the other classes (P < .001); the latter mostly had a NIH AI score of 0. The NIH CI showed higher scores in classes III and IV only in cohort 2 (P < .001). Despite the lack of significant difference in the distribution of histological classes, adults presented a higher NIH CI compared to pediatric patients in both cohorts (P < .001). Endocapillary hypercellularity was present in more than 70% of biopsies and showed a strong correlation with neutrophils/karyorrhexis (rs = 0.842, P < .001); both categories were also strongly correlated, especially in cohort 2, with cellular/fibrocellular crescents (rs = 0.610 and 0.635, P < .001). Fibrinoid necrosis was quite rare to encounter (13.3% and 8.7% of the biopsies, in cohorts 1 and 2, respectively) and showed a correlation with cellular/fibrocellular crescents (rs = 0.302, P = .004 in cohort 1). Among chronic lesions, total glomerulosclerosis, tubular atrophy and interstitial fibrosis were present in about half of the renal biopsies and had a strong correlation with each other (P < .001); fibrous crescents were less frequently encountered and had the strongest correlations with NIH CI (rs = 0.390, P < .001), NIH AI (rs = 0.256, P = .009) and cellular/fibrocellular crescents (rs = 260, P = .008) in cohort 2. Conclusion This study revealed key limitations of the NIH indices, such as the restricted range of attainable NIH AI scores, the strong relationship between the NIH AI and histological classes, the major role of the endocapillary hypercellularity in driving the NIH AI, and the likely independent value of the NIH CI across different classes. All these aspects raise doubts about their effectiveness in accurately reflecting the extent of activity and chronicity in LN biopsies and predicting patient outcomes. It is crucial to further investigate whether potential modifications to these indices could lead to a more balanced and reliable scoring system.