Histologic Predictors of Response and Prognosis in Lupus Nephritis: Why Segmental Sclerosis Should Not Be Part of the Chronicity Index

Background: The current histological chronicity index (CI) is calculated by summing the percentages of glomeruli with global sclerosis (GS) and segmental sclerosis (SS), fibrous crescents (FC), and the percentages of interstitial fibrosis (IF) and tubular atrophy (TA). The study aimed to evaluate the prognostic value of individual histological items in predicting complete response to treatment (CR) and progression to kidney failure (KF) Methods: This retrospective cohort study included all biopsy-proven LN patients diagnosed between 2008 and 2021 with follow-up ≥36 months. Demographic, laboratory, and histopathological variables were recorded at presentation. The study outcomes were evaluated through time-to-event analyses, including time to CR and time to KF. The best predictive models for CR and KF were constructed using Cox regression analysis and evaluated through Harrell’s C statistic Results: We studied 492 patients with a median follow-up of 99 months (IQR 66-129). The median age was 28 years (IQR 22-37), and 94% were female. Creatinine at presentation was 0.96 mg/dL (IQR 0.67-1.49), proteinuria was 3.4 g/g (IQR 2.0-5.7). Median activity and CI were 4 points (IQR 2-8) and 4 points (IQR 2-6), respectively. Individual items of the CI, except SS, were associated with the time to CR in the univariate analysis, while only interstitial inflammation was associated with time to CR (Figure 1). All activity and chronicity items were associated with time to KF. The best predictive model for KF included GS, FC, IF, and TA (C-statistic 0.74 [0.69-0.78]). The model did not improve with the addition of SS (C-statistic 0.73 [0.69-0.78]). The calculated CI did not improve with the addition of SS (C-statistic 0.72 [0.67-0.76] and 0.71 [0.68-0.76], respectively) Conclusion: Individual histological items of the activity and CI are variably associated with time to CR and KF. SS is not associated with time to KF, and its inclusion in the calculation of the CI does not improve its predictive value for this outcome