Machine learning in predicting T-score in the Oxford classification system of IgA nephropathy

Lin-Lin Xu; Di Zhang; Hao-Yi Weng; Li-Zhong Wang; Ruo-Yan Chen; Gang Chen; Su-Fang Shi; Li-Jun Liu; Xu-Hui Zhong; Shen-Da Hong; Li-Xin Duan; Ji-Cheng Lv; Xu-Jie Zhou; Hong Zhang

doi:10.3389/fimmu.2023.1224631

Machine learning in predicting T-score in the Oxford classification system of IgA nephropathy

Front Immunol. 2023 Aug 4:14:1224631. doi: 10.3389/fimmu.2023.1224631. eCollection 2023.

Authors

Lin-Lin Xu¹, Di Zhang^{2

3

4}, Hao-Yi Weng^{2

3

4}, Li-Zhong Wang^{2

3

4}, Ruo-Yan Chen^{2

3

4}, Gang Chen^{2

3

4}, Su-Fang Shi¹, Li-Jun Liu¹, Xu-Hui Zhong⁵, Shen-Da Hong⁶, Li-Xin Duan⁷, Ji-Cheng Lv¹, Xu-Jie Zhou¹, Hong Zhang¹

Affiliations

¹ Renal Division, Peking University First Hospital, Kidney Genetics Center, Peking University Institute of Nephrology, Key Laboratory of Renal Disease, Ministry of Health of China, Key Laboratory of Chronic Kidney Disease Prevention and Treatment, Peking University, Ministry of Education, Beijing, China.
² Hunan Provincial Key Lab on Bioinformatics, School of Computer Science and Engineering, Central South University, Changsha, China.
³ WeGene, Shenzhen Zaozhidao Technology, Shenzhen, China.
⁴ Shenzhen WeGene Clinical Laboratory, Shenzhen, China.
⁵ Department of Pediatrics, Peking University First Hospital, Beijing, China.
⁶ Institute of Medical Technology, Health Science Center of Peking University, Beijing, China.
⁷ The Sichuan Provincial Key Laboratory for Human Disease Gene Study, Research Unit for Blindness Prevention of Chinese Academy of Medical Sciences (2019RU026), Sichuan Academy of Medical Sciences and Sichuan Provincial People's Hospital, University of Electronic Science and Technology of China, Chengdu, China.

Abstract

Background: Immunoglobulin A nephropathy (IgAN) is one of the leading causes of end-stage kidney disease (ESKD). Many studies have shown the significance of pathological manifestations in predicting the outcome of patients with IgAN, especially T-score of Oxford classification. Evaluating prognosis may be hampered in patients without renal biopsy.

Methods: A baseline dataset of 690 patients with IgAN and an independent follow-up dataset of 1,168 patients were used as training and testing sets to develop the pathology T-score prediction (T _pre) model based on the stacking algorithm, respectively. The 5-year ESKD prediction models using clinical variables (base model), clinical variables and real pathological T-score (base model plus T _bio), and clinical variables and T _pre (base model plus T _pre) were developed separately in 1,168 patients with regular follow-up to evaluate whether T _pre could assist in predicting ESKD. In addition, an external validation set consisting of 355 patients was used to evaluate the performance of the 5-year ESKD prediction model using T _pre.

Results: The features selected by AUCRF for the T _pre model included age, systolic arterial pressure, diastolic arterial pressure, proteinuria, eGFR, serum IgA, and uric acid. The AUC of the T _pre was 0.82 (95% CI: 0.80-0.85) in an independent testing set. For the 5-year ESKD prediction model, the AUC of the base model was 0.86 (95% CI: 0.75-0.97). When the T _bio was added to the base model, there was an increase in AUC [from 0.86 (95% CI: 0.75-0.97) to 0.92 (95% CI: 0.85-0.98); P = 0.03]. There was no difference in AUC between the base model plus T _pre and the base model plus T _bio [0.90 (95% CI: 0.82-0.99) vs. 0.92 (95% CI: 0.85-0.98), P = 0.52]. The AUC of the 5-year ESKD prediction model using T _pre was 0.93 (95% CI: 0.87-0.99) in the external validation set.

Conclusion: A pathology T-score prediction (T _pre) model using routine clinical characteristics was constructed, which could predict the pathological severity and assist clinicians to predict the prognosis of IgAN patients lacking kidney pathology scores.

Keywords: IgA nephropathy; Oxford classification system; end-stage kidney disease; machine learning; prediction model.

Publication types

Research Support, Non-U.S. Gov't

MeSH terms

Algorithms
Glomerulonephritis, IGA* / diagnosis
Humans
Kidney
Kidney Failure, Chronic* / etiology
Machine Learning

Grants and funding

This work was supported by the National Science Foundation of China (82022010, 82131430172, 81970613, 82070733), Beijing Natural Science Foundation (Z190023), Academy of Medical Sciences—Newton Advanced Fellowship (NAFR13\1033), King’s College London—Peking University Health Science Center Joint Institute for Medical Research (BMU2021KCL004), Fok Ying Tung Education Foundation (171030), Chinese Academy of Medical Sciences (CAMS) Innovation Fund for Medical Sciences (2019-I2M-5-046, 2020-JKCS-009), and National High Level Hospital Clinical Research Funding (Interdisciplinary Clinical Research Project of Peking University First Hospital, 2022CR41, 2022CR40). The funders had no role in study design, data collection and analysis, decision to publish, or preparation of the manuscript.