New Prediction Tool Projects Pediatric IgA Nephropathy Progression Post-Biopsy

Sean Barbour, MD

Credit: Centre for Advancing Health Outcomes

Researchers have developed an IgA nephropathy (IgAN) prediction tool for pediatric patients to determine their risk of disease progression 1 to 2 years post-biopsy, a resource previously only available for adult patients.¹

The updated pediatric model was developed based on existing, externally validated adult post-biopsy prediction tool models to facilitate the identification of children at risk of a 30% decline in estimated glomerular filtration rate (eGFR) or end-stage kidney disease (ESKD) up to 2 years after biopsy.¹

The 2021 Kidney Disease Improving Global Outcomes (KDIGO) guidelines recommend risk stratification with the International IgAN Prediction Tool to determine patients’ risk of a 50% decline in eGFR or progression to kidney failure.² A similar model was developed for use in children at the time of kidney biopsy, but while adult patients now have additional post-biopsy prediction tool models to reassess individual risk of kidney function decline 1 or 2 years after biopsy, no such tool exists for children.¹

“We previously derived and externally validated the adult post-biopsy Prediction Tool models which can be used 1 or 2 years after biopsy to reassess individual risk of kidney function decline,” Sean Barbour, MD, a clinical associate professor in the division of nephrology at the University of British Columbia, and colleagues wrote.¹ “The ability to repeat risk stratification in children at time points post-biopsy is as important as in adults, however, there are currently no established risk prediction models for this purpose."

To develop a pediatric version of the adult post-biopsy prediction tool, investigators examined an international multi-ethnic cohort of children with biopsy-proven IgAN and frequent follow-up into adulthood. For inclusion, patients were required to be <18 years of age 1 year after kidney biopsy. Investigators excluded those who progressed to ESKD within the first year, had < 1 year of follow-up, or did not have ≥ 1 eGFR measurement before and after the new landmark time 1 year after biopsy.¹

The primary outcome was a composite of the first occurrence of ESKD, defined as eGFR < 15 ml/min/1.73m2, dialysis or transplantation, or a permanent reduction in eGFR below 30% of the value at the 1-year landmark time.¹

In total, the analytic cohort included 947 pediatric patients, the majority of whom were male (65%) and Chinese (37%). At the time of biopsy, 9% of participants were on a renin-angiotensin system blocker and 12% had received prior immunosuppression. By 1 year after biopsy, these figures increased to 54% and 51%, respectively. Additionally, investigators noted eGFR, proteinuria, and MAP had all improved by 1 year post-biopsy.¹

During a median 3.3 (Interquartile range [IQR], 1.6-5.8) years of follow-up after the baseline time point at 1-year post-biopsy, 361 (38%) participants were followed into adulthood. The primary outcome occurred in 94 children, with a 4-year risk of 7.2% (95% CI, 5.1-9.2).¹

Investigators noted the existing adult post-biopsy prediction tool and the original pediatric Prediction Tool designed for use at the time of biopsy both exhibited inadequate prediction performance to be used without modification in children 1 year after biopsy. Thus, investigators updated the adult post-biopsy Prediction Tool model using an overall recalibration, selective reestimation of beta-coefficients, and a new baseline survival to generate new models designed to be used in children after biopsy.¹

Compared to the original pediatric prediction tool, with the updated post-biopsy prediction tool, investigators observed better model fit with lower AIC and higher R2D values and better discrimination for the 4-year risk given by higher C-statistics (ΔC-statistic=0.09, 95%CI: 0.07, 0.10; and 0.14, 95%CI: 0.12, 0.15). Calibration curves showed improved agreement between predicted and observed risk, with better calibration than either the original pediatric or adult post-biopsy models demonstrated by lower integrated calibration indices (0.74 vs 2.45 vs 3.52 and 0.64 vs 1.01 vs 3.05). Of note, results were similar after internal validation and when the models were applied 2 years after biopsy.¹

Investigators pointed out the eGFR trajectories after a baseline 1-year post-biopsy were non-linear and varied based on predicted risk from the updated pediatric post-biopsy prediction tool, with those at higher predicted risk starting with a lower eGFR and experiencing a more rapid decline over time. Across all risk groups, eGFR had a variable rate of increase until 15-18 years of age, at which point it decreased linearly with a more rapid decline in higher-risk groups.¹

Investigators acknowledged multiple limitations to these findings, including uncertainty regarding the models’ ability to predict outcomes over longer time periods; the prevalence of White, Chinese, and Japanese ethnicities in the cohort; the lack of external validation; and the availability of new supportive therapies for IgAN not considered in the cohort and the subsequent prediction tool.¹

“These results now provide the first method that can be applied up to 1- or 2- years after biopsy to accurately identify children with IgA nephropathy who are more likely to experience disease progression,” investigators concluded.¹

References

1. ^{Barbour SJ, Coppo R, Er L, et al. Application of the updated International IgA Nephropathy Prediction Tool in children one- or two-years post-biopsy. Kidney International. https://doi.org/10.1016/j.kint.2024.07.012}
2. ^{Kidney Disease Improving Global Outcomes. Top 10 Takeaways for Clinicians from the KDIGO 2021 Clinical Practice Guideline for the Management of IgA Nephropathy. Accessed August 12, 2024. https://kdigo.org/wp-content/uploads/2017/02/KDIGO-GD-Guideline-Key-Takeaways-for-Clinicians-IgAN.pdf}