A Prediction Model of Disease Progression in X-Linked Alport syndrome Based on Clinical Characteristics and Genetic Variants

Introduction

Alport syndrome (AS) is an inherited kidney disease with significant clinical heterogeneity. Prognosis prediction and risk assessment are important to assist patient care. However, a predictive tool of disease progression is still lacking.

Methods

The prediction model was developed in 363 patients (124 kidney failure events) with X-linked AS (XLAS) from a single-center retrospective cohort study and validated in 2 external cohorts, including 193 (27 events) and 125 patients (33 events) with XLAS from 6 centers and the literature database, respectively. Cox proportional hazards regression analysis with stepwise selection was used to select the important variables related to the progression to kidney failure, by using the baseline demographic, clinical, and genetic data. The performance of the prediction model was evaluated and compared using receiver-operating characteristic (ROC) curve and calibration plot.

Results

There were 4 variables identified that were significantly associated with the progression to kidney failure in the final model, namely sex, proteinuria, estimated glomerular filtration rate (eGFR), and pathogenic variants in COL4A5. Based on the model risk stratification, the median age at kidney failure was 23, 30, and 61 years in the low-, intermediate-, and high-risk groups, respectively. This model shows the best discrimination in predicting the progression to kidney failure before the age of 30 years, with areas under the curve (AUCs) > 0.80 in both development and external cohorts.

Conclusion

A prediction model of progression to kidney failure based on clinical characteristics and genetic variants was developed and validated in patients with XLAS.