EBioMedicine · 2026
Individualised treatment effects of corticosteroids in IgA nephropathy.
СтатьяНазвание и аннотация: Перевод в очереди
Обработка статьи · этапы, ошибки и управление
Загрузка состояния…
Ошибки и ограничения
История · было → стало
Состояние обновляется каждые 5 секунд, пока блок открыт. Процент показывается только там, где известен объём работы. Пауза останавливает продолжение этапа; текущий запрос к источнику может завершиться.
Hölscher David L, Schmitz Nikolas E J, Niggemeier Leon, Pilva Pourya, Strauch Martin, Tesar Vladimir, Barratt Jonathan, Roberts Ian S D, Coppo Rosanna, Barisoni Laura, Yanagita Motoko, Alabalik Ulas, Rule Andrew D, Jagtap Jaidip M, Abreu Eliabe S, Seikrit Claudia, von Stillfried Saskia, Taal Maarten W, Kalra Philip A, Floege Juergen, Kramann Rafael, Boor Peter, Bülow Roman D, VALIGA investigators, CureGN investigators, NURTuRE academic steering group, AI4IgAN study group
Аннотация
BACKGROUND: IgA nephropathy (IgAN) has diverse clinical presentations and responses to treatment. For systemic corticosteroids in particular, randomised controlled trials have reported conflicting effects, highlighting the need for individualised treatment strategies.
METHODS: In this retrospective cohort study, we derived and validated a causal machine learning (ML) framework to estimate individualised corticosteroid treatment effects in IgAN. Eight international cohorts, including the VALIGA, CureGN, and NURTuRE-CKD repositories, comprising 1022 patients, were analysed (derivation, n = 464; validation, n = 558). We integrated baseline clinical data, histopathological classification scores (MEST-C), and deep learning-based histomorphological biomarkers (pathomics) from digitised kidney biopsies. The framework estimated the effect of systemic corticosteroids on the composite endpoint of a ≥50% decline in estimated glomerular filtration rate or kidney failure within five years of biopsy.
FINDINGS: Across the overall study population, systemic corticosteroid therapy was not associated with a significant improvement in the composite outcome (p = 0·27). However, the causal ML framework revealed substantial treatment heterogeneity, identifying patients with high predicted benefit who achieved longer progression-free survival with corticosteroids (0·43 years, 95% CI 0·18-0·73, p < 0·01), while no benefit was observed in those with low predicted benefit (-0·005 years, 95% CI -0·3 to 0·22, p > 0·05). An individualised framework-guided treatment assignment was estimated to reduce systemic corticosteroid use by 60·7%. Pathomics facilitated the identification of interstitial inflammation and tubulitis as key features of corticosteroid response.
INTERPRETATION: This study demonstrates that a causal ML framework integrating clinical, histopathological, and pathomics predictors can individualise treatment assignments for systemic corticosteroids in IgAN. This approach provides a blueprint for precision therapy in IgAN, supporting AI-enhanced clinical decision-making in the era of emerging targeted treatments.
FUNDING: German Research Foundation; European Research Council; German Federal Ministry of Education and Research; German Innovation Fund of the Federal Joint Committee; Clinician Scientist Program of the Faculty of Medicine RWTH Aachen University.
METHODS: In this retrospective cohort study, we derived and validated a causal machine learning (ML) framework to estimate individualised corticosteroid treatment effects in IgAN. Eight international cohorts, including the VALIGA, CureGN, and NURTuRE-CKD repositories, comprising 1022 patients, were analysed (derivation, n = 464; validation, n = 558). We integrated baseline clinical data, histopathological classification scores (MEST-C), and deep learning-based histomorphological biomarkers (pathomics) from digitised kidney biopsies. The framework estimated the effect of systemic corticosteroids on the composite endpoint of a ≥50% decline in estimated glomerular filtration rate or kidney failure within five years of biopsy.
FINDINGS: Across the overall study population, systemic corticosteroid therapy was not associated with a significant improvement in the composite outcome (p = 0·27). However, the causal ML framework revealed substantial treatment heterogeneity, identifying patients with high predicted benefit who achieved longer progression-free survival with corticosteroids (0·43 years, 95% CI 0·18-0·73, p < 0·01), while no benefit was observed in those with low predicted benefit (-0·005 years, 95% CI -0·3 to 0·22, p > 0·05). An individualised framework-guided treatment assignment was estimated to reduce systemic corticosteroid use by 60·7%. Pathomics facilitated the identification of interstitial inflammation and tubulitis as key features of corticosteroid response.
INTERPRETATION: This study demonstrates that a causal ML framework integrating clinical, histopathological, and pathomics predictors can individualise treatment assignments for systemic corticosteroids in IgAN. This approach provides a blueprint for precision therapy in IgAN, supporting AI-enhanced clinical decision-making in the era of emerging targeted treatments.
FUNDING: German Research Foundation; European Research Council; German Federal Ministry of Education and Research; German Innovation Fund of the Federal Joint Committee; Clinician Scientist Program of the Faculty of Medicine RWTH Aachen University.