Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

A PLA2R-IgG4 Antibody-Based Predictive Model for Assessing Risk Stratification of Idiopathic Membranous Nephropathy

A PLA2R-IgG4 Antibody-Based Predictive Model for Assessing Risk Stratification of Idiopathic... Hindawi Journal of Healthcare Engineering Volume 2021, Article ID 1521013, 6 pages https://doi.org/10.1155/2021/1521013 Research Article A PLA2R-IgG4 Antibody-Based Predictive Model for Assessing Risk Stratification of Idiopathic Membranous Nephropathy 1 2 2 3 4 2 Xiaobin Liu, Jing Xue, Xiaoyi Guo, Yijie Ding , Yi Zhang, Xiran Zhang, 2 5 6,7 1 2 Yiqing Huang, Biao Huang, Zhigang Hu , Guoyuan Lu , and Liang Wang Department of Nephrology, e First Affiliated Hospital of Soochow University, Suzhou 215006, China Department of Nephrology, e Affiliated Wuxi People’s Hospital of Nanjing Medical University, Wuxi 214023, China Yangtze Delta Region Institute (Quzhou), University of Electronic Science and Technology of China, Quzhou 324000, China NHC Key Laboratory of Nuclear Medicine, Jiangsu Key Laboratory of Molecular Nuclear Medicine, Jiangsu Institute of Nuclear Medicine, Wuxi 214063, China College of Life Sciences and Medicine, Zhejiang Sci-Tech University, Hangzhou 310018, China Medical Laboratory, e Affiliated Wuxi Children’s Hospital of Nanjing Medical University, Wuxi 214023, China Medical Laboratory, e Affiliated Wuxi People’s Hospital of Nanjing Medical University, Wuxi 214023, China Correspondence should be addressed to Guoyuan Lu; snklgy@126.com and Liang Wang; wangliang_wuxi@126.com Received 6 August 2021; Revised 19 August 2021; Accepted 20 August 2021; Published 31 August 2021 Academic Editor: Shengrong Gong Copyright © 2021 Xiaobin Liu et al. +is is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Background. Known as an autoimmune glomerular disease, idiopathic membranous nephropathy (IMN) is considered to be associated with phospholipase A2 receptor (PLA2R) in terms of the main pathogenesis. +e quantitative detection of serum PLA2R-IgG and PLA2R-IgG4 antibodies by time-resolved fluoroimmunoassay (TRFIA) was determined, and the value of them, both in the clinical prediction of risk stratification in IMN, was observed in this study. Methods. 95 patients with IMN proved by renal biopsy were enrolled, who had tested positive for serum PLA2R antibodies by ELISA, and the quantitative detection of serum PLA2R-IgG and PLA2R-IgG4 antibodies was achieved by TRFIA. All the patients were divided into low-, medium-, and high-risk groups, respectively, which were set as dependent variables, according to proteinuria and renal function. Random forest (RF) was used to estimate the value of serum PLA2R-IgG and PLA2R-IgG4 in predicting the risk stratification of progression in IMN. Results. Out-of-bag estimates of variable importance in RF were employed to evaluate the impact of each input variable on the final classification accuracy. +e variable of albumin, PLA2R-IgG, and PLA2R-IgG4 had high values (>0.3) of 0.3156, 0.3981, and 0.7682, respectively, which meant that these three were more important for the risk stratification of progression in IMN. In order to further assess the contribution of PLA2R-IgG and PLA2R-IgG4 to the model, we built four different models and found that PLA2R-IgG4 played an important role in improving the predictive ability of the model. Conclusions. In this study, we established a random forest model to evaluate the value of serum PLA2R-IgG4 antibodies in predicting risk stratification of IMN. Compared with PLA2R-IgG, PLA2R-IgG4 is a more efficient biomarker in predicting the risk of progression in IMN. who received only symptomatic treatment had a relatively 1. Introduction benign course, end-stage renal disease developed in 16% of Idiopathic membranous nephropathy (IMN), known as the patients during a 5-year follow-up [5]. +erefore, it is primary membrane nephropathy, is the most common cause critical to identify the degree of disease activity and pro- of primary nephropathy syndrome in adults [1, 2], which is gression risk in patients with IMN. approximately 20%∼30% of all the renal pathological biopsy In 2009, Beck et al. [6] found 70% of the patients with reports [3]. IMN is usually manifested as nephropathy IMN had antibodies against a conformation-dependent syndrome and depends on renal biopsy in diagnosis [4]. epitope in M-type phospholipase A2 receptor (PLA2R) Although the study demonstrated that patients with IMN which was present in normal podocytes and colocalized with 2 Journal of Healthcare Engineering IgG4 in immune deposits in glomeruli. Currently, PLA2R (China). An AutoDELFIA was purchased from Perkin antibodies have been confirmed to be major pathogenic Elmer (USA). antibodies in IMN [7], which are mainly IgG4, and the titer levels of anti-PLA2R antibodies are related to the activity of the disease [8, 9]. In 2019, Kidney Disease Improving Global 2.2. Anti-PLA2R-IgG and PLA2R-IgG4 Detection Procedure. Firstly, 100 μL of standards or diluted sera were pipetted to Outcomes (KDIGO) recommended that quantitative de- tection and regular follow-up of anti-PLA2R antibodies the microtiter plates fixed with 5 μg/mL of rPLA2R. +e working dilutions of serum samples were 1 : 200 and 1 : 20 in would contribute to differential diagnosis and assessment of the anti-PLA2R-IgG and anti-PLA2R -IgG4 assays, re- activity in IMN [10]. spectively. +e mixture was reacted with continuously At present, the detection of sera anti-PLA2R antibodies shaking at 25 C for 1 h. After the unreacted substances were normally applies indirect immunofluorescence assay removed by washing for 3 times, the plates were pipetted (IIFA) and enzyme-linked immunosorbent assay (ELISA) with europium-labelled goat anti-human IgG or mouse anti- [11]. IIFA is not a quantitative detection measure, while human IgG4 antibodies, shaken for 1 h at 25 C, and then ELISA is characterized by low detection sensitivity [12]. In rinsed for 6 times. Finally, 96-well plates were added with 2017, Huang et al. [13] developed an ultrasensitive quan- titative assay, using time-resolved fluoroimmunoassay 200 μL of enhancement solution, agitated for 5 min, and measured in AutoDELFIA . +e concentrations of serum (TRFIA), for the detection of anti-PLA2R-IgG testing. Establishing the cutoff value for anti-PLA2R-IgG of samples of anti-PLA2R-IgG and anti-PLA2R-IgG4 were automatically calculated from the fluorescence of wells by 1990 ng/mL, the diagnostic sensitivity and specificity in AutoDELFIA . According to the previous work [13, 14], IMN were 74% and 100%, respectively. Huang et al. [14] the cutoff values were 1990 ng/mL and 161.2 ng/mL for anti- further tested anti-PLA2R-IgG4 and found that the diag- PLA2R-IgG and anti-PLA2R-IgG4, respectively. nostic sensitivity and specificity in IMN were 90% and 100%, respectively when established the cutoff value for anti-PLA2R-IgG4 of 161.2 ng/mL. 2.3. Statistical Analyses. +is work employed 38 features In this study, we determined the quantitative detection (variables) including pathological and clinical features to of anti-PLA2R-IgG and -IgG4 antibodies by TRFIA and describe the patients’ characters, which contained mean, observed the value of them both in the clinical prediction of standard deviation, and correlation coefficient of samples as different risk-stratified IMN. shown in Table 1. +e three groups (low-, medium-, and high-risk groups), which were divided according to pro- 2. Materials and Methods teinuria and renal function, were set as dependent variables. +e results demonstrated that PLA2R-IgG (0.394) and 2.1. Subjects Selection. A total of 95 patients with IMN PLA2R-IgG4 (0.524) had great correlation coefficients with proved by renal biopsy, who had tested positive for serum the dependent variables. In our study, the number of patients PLA2R antibodies by ELISA, from the Affiliated Wuxi in class 1, 2, and 3 (three types of risk stratification) was 45, People’s Hospital of Nanjing Medical University, were 41, and 9, respectively. enrolled from January 2016 to December 2017. According to proteinuria <4 g/d, 4–8 g/d, and >8 g/d, with renal function taken into consideration, all the patients were 2.4. Random Forest. In machine learning, the random forest divided into low-, medium-, and high-risk groups, re- (RF) [17] is a classifier that contains multiple decision trees, spectively [15]. and the output category is determined by the mode of the Blood samples were collected before renal biopsy and category output by the individual trees. In this work, random before the immunosuppressive therapy, which were left forest was used as an analysis and classification tool to es- standing to clot thoroughly before centrifuging at 3000 rpm/ timate the value of serum PLA2R-IgG and PLA2R-IgG4 in min for 4 min to obtain serum, and sera were then stored at ° predicting the risk stratification of progression in IMN. −80 C for pending analysis. All renal tissue specimens were Compared with k-nearest neighbor (KNN) and support examined using light microscope, immunofluorescence, and vector machines (SVM) classifiers, it has the following ad- electron microscope. Pathological grading was performed by vantages: (1) it can assess the importance of variables when Ehrenreich and Churg standards [16]. determining categories; (2) when building a forest, it can Goat anti-human IgG antibodies were obtained from produce an unbiased estimate of the generalized error in- Jackson ImmunoResearch (USA), and mouse anti-human ternally; (3) for unbalanced classification data sets, it can IgG4 antibodies were offered by Hytest (Finland). Europium balance errors; (4) the learning process is fast. labeling kits (1244-302) were purchased from Perkin Elmer (USA). +e polystyrene microtiter plates were obtained from Nunc International (Denmark). +e recombinant PLA2R 3. Results antigen, series of standards of anti-human PLA2R-IgG and PLA2R-IgG4 were prepared in our laboratory as previously 3.1. Evaluation Measurements. +e accuracy (ACC) is uti- reported [13, 14]. All the buffer solutions were supplied from lized to evaluate the performance of the RF model under 5- Jiangyuan Co. (China). +e other reagents were of analytical fold cross-validation (5-CV). +e calculation method of grade and obtained from Sinopharm Chemical Reagent ACC is as follows: Journal of Healthcare Engineering 3 Table 1: +e information of data set. No. Feature (variable) Value r 1 Age (years) 55.110± 15.110 0.175 2 Gender (males/females) 56/39 0.105 3 Renal tubular atrophy score 1.057± 0.721 0.080 4 Renal interstitial fibrosis score 1.100± 0.731 0.113 5 Renal interstitial lymphoplasmacytic infiltrate score 1.068± 0.861 0.187 6 Total score of renal tubular and interstitium 3.236± 2.138 0.136 7 Pathological stage of IMN 1.452± 0.495 0.009 8 IF IgA 0.220± 0.477 0.089 9 IF IgM 0.252± 0.635 0.034 10 IF IgG 2.789± 0.697 0.009 11 IF C1q 0.242± 0.488 0.098 12 IF C3 1.205± 0.738 −0.035 13 Renal tissue PLA2R antigen 0.952± 0.357 −0.031 14 SBP (mmHg) 133.389± 14.973 0.017 15 DBP (mmHg) 79.989± 9.079 −0.109 16 Serum C3 (mg/L) 894.305± 253.312 −0.008 17 Serum C4 (mg/L) 241.147± 87.192 0.058 18 Serum IgA (g/L) 5.206± 28.002 0.061 19 Serum IgG (g/L) 7.084± 2.891 −0.077 20 Serum IgM (g/L) 1.257± 0.637 −0.013 21 Serum albumin (g/L) 23.335± 8.233 −0.449 22 Hematuria (/uL) 102.377± 138.301 −0.014 23 Serum creatinine (umol/L) 83.664± 34.042 0.168 24 eGFR-EPI (ml/min) 86.535± 24.060 −0.189 25 BUN (mmol/L) 4.986± 1.929 0.172 26 Serum glucose (mmol/L) 5.142± 0.810 0.321 27 Serum lithic acid (umol/L) 344.676± 89.678 0.048 28 TG (mmol/L) 2.406± 1.471 0.415 29 TC (mmol/L) 7.119± 2.289 0.141 30 LDL-C (mmol/L) 3.902± 1.379 0.077 31 HDL-C (mmol/L) 1.323± 0.432 −0.110 32 WBC (×10 /L) 8.577± 13.978 0.088 33 Hemoglobin (g/L) 122.204± 24.962 0.058 34 PLT (×10 /L) 219± 65.409 −0.104 35 C-reactive protein (mg/L) 2.807± 3.167 0.057 36 ESR (mm/H) 53.589± 34.176 0.233 37 Serum PLA2R-IgG (ng/mL) 5243.957± 9282.902 0.394 38 Serum PLA2R-IgG4 (ng/mL) 1762.615± 2662.328 0.524 denotes that each feature correlated with risk stratification of progression in IMN using Pearson correlation coefficient (r). c i increase in mean square error (MSE) averaged over all 􏽐 TP i�1 whole ACC � × 100%, trees in the ensemble and divided by the standard devi- ation taken over the trees, for each variable. So, we could get the importance scores of all input variables. In general, TP ACC � × 100%, (1) the larger the score, the more important it is for the prediction model. Figure 1 shows the results of relative c importance for inputs. It could be seen from the figure M � 􏽘 M , that the 21-th (albumin), 37-th (PLA2R-IgG), and 38-th i�1 (PLA2R-IgG4) variables had high values (>0.3) of 0.3156, 0.3981, and 0.7682, respectively. where c denotes the number of classes. TP denotes the number of true positive (TP) in subclass i. M and M denote the number of whole test samples and subclass test samples. 3.3. Comparison of PLA2R-IgG and PLA2R-IgG4. In order to ACC denotes the accuracy in subclass i. further evaluate the contribution of PLA2R-IgG and PLA2R- IgG4 to the model, we constructed four different models, 3.2. Relative Importance of Inputs in Estimating IMN. To which contain both PLA2R-IgG and PLA2R-IgG4 variables, evaluate the impact of each input variable on the final PLA2R-IgG variable, PLA2R-IgG4 variable, and no PLA2R- classification accuracy. We employed out-of-bag esti- IgG and PLA2R-IgG4 variables. +e relevant information of mates of variable importance in RF. +e RF stored the the models is shown in Table 2. 4 Journal of Healthcare Engineering 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 -0.1 -0.2 0 5 10152025303540 Feature Index Figure 1: Relative importance for inputs. Table 2: +e information of models. Model Number of features (variables) Model 1 38 With PLA2R-IgG and PLA2R-IgG4 Model 2 37 With PLA2R-IgG and without PLA2R-IgG4 Model 3 37 Without PLA2R-IgG and with PLA2R-IgG4 Model 4 36 Without PLA2R-IgG and PLA2R-IgG4 +e classification performance of the 4 models was Table 3: Comparison on four models via 5-fold cross-validation. verified by 5-fold cross-validation. And, the results are listed 1 2 3 Model Overall ACC ACC ACC ACC in Table 3. Obviously, when PLA2R-IgG and PLA2R-IgG4 were included, the prediction performance of the model Model 1 0.6743 0.8 0.6583 0.1 Model 2 0.6422 0.7778 0.5833 0.2 (model 1) was the best, and the overall classification accuracy Model 3 0.6639 0.7778 0.6806 0 was 0.6743. In the overall ACC, the performance of model 2 Model 4 0.6290 0.7556 0.6306 0 (0.6422) was not better than model 3 (0.6639). +is further 1 2 3 ACC : the accuracy of class 1; ACC : the accuracy of class 2; ACC : the verified that PLA2R-IgG4 is more important than PLA2R- accuracy of class 3. IgG for classification accuracy. In addition, the performance of the model (model 4) was the worst (0.6290), when PLA2R-IgG4 and PLA2R-IgG were not contained at the same time. From the above test, it could be found that Table 4: Analysis of statistical significance for different methods via PLA2R-IgG and PLA2R-IgG4 were very helpful for classi- 5-fold cross validation (10 times). fication. However, PLA2R-IgG4 could achieve better results than PLA2R-IgG. P value In this study, we employed t-test to evaluate the sig- Between model 1 and model 2 6.355e − 4 nificant differences of average ACC between different Between model 1 and model 3 0.0085 Between model 1 and model 4 1.8e − 5 models. +e results were list in Table 4, which show that the Between model 2 and model 3 0.1142 differences between model 1 and other three models were Between model 2 and model 4 0.1510 significant. Furthermore, the P value of model 3 and model 4 Between model 3 and model 4 0.0022 was 0.0022. +is means that the PLA2R-IgG4 feature had a more significant performance improvement compared to the ordinary model (without PLA2R-IgG and PLA2R-IgG4). major target antigen, progress was made in understanding the pathogenesis of IMN [6]. Circulating anti-PLA2R an- tibodies not only contribute to distinguish primary mem- 4. Discussion branous nephropathy from secondary membranous It is widely recognized that IMN is an autoimmune glo- nephropathy in diagnosis but also conduce to monitor the merular disease in which autoantibodies combine with immunological activity degree during the treatment period antigens on glomerular podocytes and deposit in glomerular [19]. Accordingly, the quantitative detection of circulating capillary walls [18]. With the discovery of M-type phos- anti-PLA2R antibodies is particularly important in diagnosis pholipase A2 receptor (PLA2R) which was identified as the and treatment of IMN. Out-of-Bag Feature Importance Journal of Healthcare Engineering 5 Among all the PLA2R-IgG antibodies, PLA2R-IgG4 Ethical Approval antibodies are predominant [6, 18]. Lacking of mature +e experimental protocol was established, according to the commercial testing means TRFIA was employed in this ethical guidelines of the Helsinki Declaration and was ap- study for the quantitative detection of anti-PLA2R-IgG4 proved by the Human Ethics Committee (Wuxi People’s antibodies. As a novel nonisotopic labeling technology, −18 Hospital Ethics Committee). +is study had been approved TRFIA has the advantages of high sensitivity (10 mol/L), by the ethics committee of the hospital (ethical approval no. wide monitoring range, and less susceptibility to matrix kyl2016001). interference [14]. Our previous work discovered that using the cutoff value of 161.2 ng/mL, anti-PLA2R-IgG4 had higher sensitivity in diagnosis than anti-PLA2R-IgG by the Consent cutoff value of 1990 ng/mL (90% versus 74%) [14]. +erefore, Written informed consent for publication was obtained we speculate that, in addition to anti-PLA2R-IgG, anti- from all participants. PLA2R-IgG4 may be an efficient biomarker in the assess- ment of the severity and prognosis of IMN too. In recent years, machine learning methods have been Conflicts of Interest widely used in medicine [20–22] and biology [23–25] to solve +e authors declare that they have no conflicts of interest. difficult data analysis problems for researchers. In our study, RF was employed to evaluate the importance of all the features (input variables). And, we found that the 21th (albumin), 37th Authors’ Contributions (PLA2R-IgG), and 38th (PLA2R-IgG4) variables had high +ey are joint first authors: Xiaobin Liu, Jing Xue, Xiaoyi values (>0.3) of 0.3156, 0.3981, and 0.7682, respectively, which Guo. meant these three features were more important to the risk stratification of progression in IMN. At the same time, as shown in Figure 1, PLA2R-IgG4 manifested a better predictive Acknowledgments value compared with PLA2R-IgG (0.7682 versus 0.3981). In +anks to the Department of Nephrology of Wuxi People’s addition, it could be seen from Table 3 that the prediction Hospital for collecting data in this study. +is study was effect of the PLA2R-IgG4 feature (0.6639) was better than that funded by the Top Talent Support Program for Young and of the PLA2R-IgG feature (0.6422). When both PLA2R-IgG4 Middle-Aged People of Wuxi Health Committee and PLA2R-IgG features were input into the model, its pre- (HB2020008), Medical and Public Health Project of Wuxi diction performance was the best (0.6743). +e above test Sci-Tech Development Fund (WX18 II AN047), the Sci- results further validated the importance of PLA2R-IgG4 and entific research project of Wuxi Health Committee PLA2R-IgG in assessing the risk level model. In Table 4, we (MS201927), the Scientific research project of Wuxi Health evaluated the significant differences between different models. Committee (Z201914), Major projects of precision medicine Obviously, the difference between model 1 and other models of Wuxi health Committee (J202001), Maternal and Child was significant (P value< 0.05). Compared with model 4, Health Research Project of Jiangsu Province (F202033), the model 3 also had significant difference (P value � 0.0022). It Scientific Research Projects of Jiangsu Provincial Health was obvious that PLA2R-IgG4 played an important role in Commission (LGY201801), and Jiangsu Province “333” improving the predictive ability of the model. Project (BRA2020142). 5. Conclusions References In this study, we evaluated the value of serum PLA2R-IgG4 [1] J. M. Hofstra and J. F. M. Wetzels, “Management of patients antibodies in predicting risk stratification of IMN by with membranous nephropathy,” Nephrology Dialysis establishing a random forest model. Compared with PLA2R- Transplantation, vol. 27, no. 1, pp. 6–9, 2012. IgG, PLA2R-IgG4 is a more efficient biomarker in predicting [2] F. C. Fervenza, S. Sethi, and U. Specks, “Idiopathic mem- the risk of progression in IMN. +e study results are sat- branous nephropathy: diagnosis and treatment,” Clinical isfactory, while disadvantages remain (1) there is lack of Journal of the American Society of Nephrology, vol. 3, analysis about the value of serum PLA2R-IgG4 in the pp. 905–919, 2008. [3] X. Pan, J. Xu, H. Ren et al., “Changing spectrum of biopsy- prediction of treatment effect; (2) the sample size needs to be proven primary glomerular diseases over the past 15 Years: a further expanded to minimize the prediction bias. In the single-center study in China,” Contributions to Nephrology, next work, we will expand the sample size and survey the vol. 181, pp. 22–30, 2013. predictive effect of serum PLA2R-IgG4 in the therapeutic [4] K. Zuo, Y. Wu, S.-J. Li, F. Xu, C.-H. Zeng, and Z.-H. Liu, regimen and prognosis of IMN. “Long-term outcome and prognostic factors of idiopathic membranous nephropathy in the Chinese population,” Data Availability Clinical Nephrology, vol. 79, pp. 445–453, 2013. [5] A. Schieppati, L. Mosconi, A. Perna et al., “Prognosis of +e data used to support the research can be obtained from untreated patients with idiopathic membranous nephropa- the corresponding authors according to the requirements of thy,” New England Journal of Medicine, vol. 329, no. 2, the institution. pp. 85–89, 1993. 6 Journal of Healthcare Engineering [6] L. H. Beck, R. G. Bonegio, G. Lambeau et al., “M-type transfer TSK fuzzy system,” IEEE Transactions on Intelligent phospholipase A2 receptor as target antigen in idiopathic Transportation Systems, vol. 22, no. 3, pp. 1752–1764, 2021. membranous nephropathy,” New England Journal of Medi- [23] H. Wang, J. Tang, Y. Ding, and F. Guo, “Exploring associa- tions of non-coding RNAs in human diseases via three-matrix cine, vol. 361, pp. 11–21, 2009. [7] J. N. Chi, T. S. Lai, C. F. Wu et al., “+e relationship of anti- factorization with hypergraph-regular terms on center kernel phospholipase A2 receptor antibody and C5a complement alignment,” Briefings in Bioinformatics, 2021. with disease activity and short-term outcome in idiopathic [24] Y. Jiang, X. Gu, D. Wu et al., “A novel negative-transfer- membranous nephropathy,” Journal of the Formosan Medical resistant fuzzy clustering model with a shared cross-domain Association, vol. 118, no. 5, pp. 898–906, 2019. transfer latent space and its application to brain CT image [8] S. Provatopoulou, D. Kalavrizioti, M. Stangou et al., “Cir- segmentation,” IEEE/ACM Transactions on Computational culating anti-phospholipase A2 receptor antibodies as a di- Biology and Bioinformatics, vol. 18, no. 1, pp. 40–52, 2021. agnostic and prognostic marker in Greek patients with [25] Y. Jiang, K. Zhao, K. Xia et al., “A novel distributed multitask idiopathic membranous nephropathy-a retrospective cohort fuzzy clustering algorithm for automatic MR brain image study,” Romanian Journal of Internal Medicine, vol. 57, no. 2, segmentation,” Journal of Medical Systems, vol. 43, no. 5, pp. 141–150, 2019. 118 pages, 2019. [9] A. S. Bomback, “Management of membranous nephropathy in the PLA2R era,” Clinical Journal of the American Society of Nephrology, vol. 13, no. 5, pp. 784–786, 2018. [10] J. Floege, S. J. Barbour, D. C. Cattran et al., “Management and Treatment of Glomerular diseases (part1):Conclusions from a Kidney Disease: Improving Global Outcomes (KDIGO),” Nephrology (Saint-Petersburg), vol. 24, no. 2, pp. 22–41, 2020. [11] W. Schlumberger, N. Hornig, S. Lange et al., “Differential diagnosis of membranous nephropathy with autoantibodies to phospholipase A2 receptor 1,” Autoimmunity Reviews, vol. 13, no. 2, pp. 108–113, 2014. [12] G. Cheng, J. Liu, A. Gilbert et al., “Serum phospholipase A2 receptor antibodies and immunoglobulin G subtypes in adult idiopathic membranous nephropathy: clinical value assess- ment,” Clinica Chimica Acta, vol. 490, pp. 135–141, 2019. [13] B. Huang, L. Wang, Y. N. Cao et al., “Improvement of idi- opathic membranous nephropathy diagnosis with ultrasen- sitive quantitative detection of anti-phospholipase A2 receptor,” e Journal of Allergy and Clinical Immunology, vol. 139, no. 6, pp. 1988–1990 e2, 2017. [14] B. Huang, Y. Zhang, L. Wang et al., “Phospholipase a2 re- ceptor antibody IgG4 subclass improves sensitivity and specificity in the diagnosis of idiopathic membranous ne- phropathy,” Kidney Blood Press Res.vol. 44, no. 4, pp. 848–857, [15] C. Daniel, “Cattran.Idiopathic membranous glomerulone- phritis,” Kidney International, vol. 59, pp. 1983–1994, 2001. [16] P. H. Nachman, J. C. Jennette, and R. J. Falk, “Membranous glomerulopathy,” Brenner BM. Brenner and Rector’s the Kidneypp. 1121–1131, Saunders, Boston, MA, USA, 9th edi- tion, 2012. [17] B. Leo, “Random forests,” Machine Learning, vol. 45, no. 1, pp. 5–32, 2001. [18] L. H. Beck and D. J. Salant, “Membranous nephropathy: from models to man,” Journal of Clinical Investigation, vol. 124, no. 6, pp. 2307–2314, 2014. [19] R. Ayalon and L. H. Beck, “Membranous nephropathy: not just a disease for adults,” Pediatric Nephrology, vol. 30, no. 1, pp. 31–39, 2015. [20] X. Y. Guo, W. Zhou, B. Shi et al., “An efficient multiple kernel support vector regression model for assessing dry weight of hemodialysis patients,” Current Bioinformatics, vol. 16, pp. 284–293, 2021. [21] J. Gao, L. Zhang, G. Yu, G. Qu, Y. Li, and X. Yang, “Model with the GBDT for colorectal adenoma risk diagnosis,” Current Bioinformatics, vol. 15, pp. 971–979, 2020. [22] Y. Jiang, Y. Zhang, C. Lin, D. Wu, and C. T. Lin, “EEG-based driver drowsiness estimation using an online multi-view and http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png Journal of Healthcare Engineering Hindawi Publishing Corporation

A PLA2R-IgG4 Antibody-Based Predictive Model for Assessing Risk Stratification of Idiopathic Membranous Nephropathy

Download PDF
6 pages

Loading next page...
 
/lp/hindawi-publishing-corporation/a-pla2r-igg4-antibody-based-predictive-model-for-assessing-risk-03d77ncls7

References (30)

Publisher
Hindawi Publishing Corporation
Copyright
Copyright © 2021 Xiaobin Liu et al. This is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited.
ISSN
2040-2295
eISSN
2040-2309
DOI
10.1155/2021/1521013
Publisher site
See Article on Publisher Site

Abstract

Hindawi Journal of Healthcare Engineering Volume 2021, Article ID 1521013, 6 pages https://doi.org/10.1155/2021/1521013 Research Article A PLA2R-IgG4 Antibody-Based Predictive Model for Assessing Risk Stratification of Idiopathic Membranous Nephropathy 1 2 2 3 4 2 Xiaobin Liu, Jing Xue, Xiaoyi Guo, Yijie Ding , Yi Zhang, Xiran Zhang, 2 5 6,7 1 2 Yiqing Huang, Biao Huang, Zhigang Hu , Guoyuan Lu , and Liang Wang Department of Nephrology, e First Affiliated Hospital of Soochow University, Suzhou 215006, China Department of Nephrology, e Affiliated Wuxi People’s Hospital of Nanjing Medical University, Wuxi 214023, China Yangtze Delta Region Institute (Quzhou), University of Electronic Science and Technology of China, Quzhou 324000, China NHC Key Laboratory of Nuclear Medicine, Jiangsu Key Laboratory of Molecular Nuclear Medicine, Jiangsu Institute of Nuclear Medicine, Wuxi 214063, China College of Life Sciences and Medicine, Zhejiang Sci-Tech University, Hangzhou 310018, China Medical Laboratory, e Affiliated Wuxi Children’s Hospital of Nanjing Medical University, Wuxi 214023, China Medical Laboratory, e Affiliated Wuxi People’s Hospital of Nanjing Medical University, Wuxi 214023, China Correspondence should be addressed to Guoyuan Lu; snklgy@126.com and Liang Wang; wangliang_wuxi@126.com Received 6 August 2021; Revised 19 August 2021; Accepted 20 August 2021; Published 31 August 2021 Academic Editor: Shengrong Gong Copyright © 2021 Xiaobin Liu et al. +is is an open access article distributed under the Creative Commons Attribution License, which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly cited. Background. Known as an autoimmune glomerular disease, idiopathic membranous nephropathy (IMN) is considered to be associated with phospholipase A2 receptor (PLA2R) in terms of the main pathogenesis. +e quantitative detection of serum PLA2R-IgG and PLA2R-IgG4 antibodies by time-resolved fluoroimmunoassay (TRFIA) was determined, and the value of them, both in the clinical prediction of risk stratification in IMN, was observed in this study. Methods. 95 patients with IMN proved by renal biopsy were enrolled, who had tested positive for serum PLA2R antibodies by ELISA, and the quantitative detection of serum PLA2R-IgG and PLA2R-IgG4 antibodies was achieved by TRFIA. All the patients were divided into low-, medium-, and high-risk groups, respectively, which were set as dependent variables, according to proteinuria and renal function. Random forest (RF) was used to estimate the value of serum PLA2R-IgG and PLA2R-IgG4 in predicting the risk stratification of progression in IMN. Results. Out-of-bag estimates of variable importance in RF were employed to evaluate the impact of each input variable on the final classification accuracy. +e variable of albumin, PLA2R-IgG, and PLA2R-IgG4 had high values (>0.3) of 0.3156, 0.3981, and 0.7682, respectively, which meant that these three were more important for the risk stratification of progression in IMN. In order to further assess the contribution of PLA2R-IgG and PLA2R-IgG4 to the model, we built four different models and found that PLA2R-IgG4 played an important role in improving the predictive ability of the model. Conclusions. In this study, we established a random forest model to evaluate the value of serum PLA2R-IgG4 antibodies in predicting risk stratification of IMN. Compared with PLA2R-IgG, PLA2R-IgG4 is a more efficient biomarker in predicting the risk of progression in IMN. who received only symptomatic treatment had a relatively 1. Introduction benign course, end-stage renal disease developed in 16% of Idiopathic membranous nephropathy (IMN), known as the patients during a 5-year follow-up [5]. +erefore, it is primary membrane nephropathy, is the most common cause critical to identify the degree of disease activity and pro- of primary nephropathy syndrome in adults [1, 2], which is gression risk in patients with IMN. approximately 20%∼30% of all the renal pathological biopsy In 2009, Beck et al. [6] found 70% of the patients with reports [3]. IMN is usually manifested as nephropathy IMN had antibodies against a conformation-dependent syndrome and depends on renal biopsy in diagnosis [4]. epitope in M-type phospholipase A2 receptor (PLA2R) Although the study demonstrated that patients with IMN which was present in normal podocytes and colocalized with 2 Journal of Healthcare Engineering IgG4 in immune deposits in glomeruli. Currently, PLA2R (China). An AutoDELFIA was purchased from Perkin antibodies have been confirmed to be major pathogenic Elmer (USA). antibodies in IMN [7], which are mainly IgG4, and the titer levels of anti-PLA2R antibodies are related to the activity of the disease [8, 9]. In 2019, Kidney Disease Improving Global 2.2. Anti-PLA2R-IgG and PLA2R-IgG4 Detection Procedure. Firstly, 100 μL of standards or diluted sera were pipetted to Outcomes (KDIGO) recommended that quantitative de- tection and regular follow-up of anti-PLA2R antibodies the microtiter plates fixed with 5 μg/mL of rPLA2R. +e working dilutions of serum samples were 1 : 200 and 1 : 20 in would contribute to differential diagnosis and assessment of the anti-PLA2R-IgG and anti-PLA2R -IgG4 assays, re- activity in IMN [10]. spectively. +e mixture was reacted with continuously At present, the detection of sera anti-PLA2R antibodies shaking at 25 C for 1 h. After the unreacted substances were normally applies indirect immunofluorescence assay removed by washing for 3 times, the plates were pipetted (IIFA) and enzyme-linked immunosorbent assay (ELISA) with europium-labelled goat anti-human IgG or mouse anti- [11]. IIFA is not a quantitative detection measure, while human IgG4 antibodies, shaken for 1 h at 25 C, and then ELISA is characterized by low detection sensitivity [12]. In rinsed for 6 times. Finally, 96-well plates were added with 2017, Huang et al. [13] developed an ultrasensitive quan- titative assay, using time-resolved fluoroimmunoassay 200 μL of enhancement solution, agitated for 5 min, and measured in AutoDELFIA . +e concentrations of serum (TRFIA), for the detection of anti-PLA2R-IgG testing. Establishing the cutoff value for anti-PLA2R-IgG of samples of anti-PLA2R-IgG and anti-PLA2R-IgG4 were automatically calculated from the fluorescence of wells by 1990 ng/mL, the diagnostic sensitivity and specificity in AutoDELFIA . According to the previous work [13, 14], IMN were 74% and 100%, respectively. Huang et al. [14] the cutoff values were 1990 ng/mL and 161.2 ng/mL for anti- further tested anti-PLA2R-IgG4 and found that the diag- PLA2R-IgG and anti-PLA2R-IgG4, respectively. nostic sensitivity and specificity in IMN were 90% and 100%, respectively when established the cutoff value for anti-PLA2R-IgG4 of 161.2 ng/mL. 2.3. Statistical Analyses. +is work employed 38 features In this study, we determined the quantitative detection (variables) including pathological and clinical features to of anti-PLA2R-IgG and -IgG4 antibodies by TRFIA and describe the patients’ characters, which contained mean, observed the value of them both in the clinical prediction of standard deviation, and correlation coefficient of samples as different risk-stratified IMN. shown in Table 1. +e three groups (low-, medium-, and high-risk groups), which were divided according to pro- 2. Materials and Methods teinuria and renal function, were set as dependent variables. +e results demonstrated that PLA2R-IgG (0.394) and 2.1. Subjects Selection. A total of 95 patients with IMN PLA2R-IgG4 (0.524) had great correlation coefficients with proved by renal biopsy, who had tested positive for serum the dependent variables. In our study, the number of patients PLA2R antibodies by ELISA, from the Affiliated Wuxi in class 1, 2, and 3 (three types of risk stratification) was 45, People’s Hospital of Nanjing Medical University, were 41, and 9, respectively. enrolled from January 2016 to December 2017. According to proteinuria <4 g/d, 4–8 g/d, and >8 g/d, with renal function taken into consideration, all the patients were 2.4. Random Forest. In machine learning, the random forest divided into low-, medium-, and high-risk groups, re- (RF) [17] is a classifier that contains multiple decision trees, spectively [15]. and the output category is determined by the mode of the Blood samples were collected before renal biopsy and category output by the individual trees. In this work, random before the immunosuppressive therapy, which were left forest was used as an analysis and classification tool to es- standing to clot thoroughly before centrifuging at 3000 rpm/ timate the value of serum PLA2R-IgG and PLA2R-IgG4 in min for 4 min to obtain serum, and sera were then stored at ° predicting the risk stratification of progression in IMN. −80 C for pending analysis. All renal tissue specimens were Compared with k-nearest neighbor (KNN) and support examined using light microscope, immunofluorescence, and vector machines (SVM) classifiers, it has the following ad- electron microscope. Pathological grading was performed by vantages: (1) it can assess the importance of variables when Ehrenreich and Churg standards [16]. determining categories; (2) when building a forest, it can Goat anti-human IgG antibodies were obtained from produce an unbiased estimate of the generalized error in- Jackson ImmunoResearch (USA), and mouse anti-human ternally; (3) for unbalanced classification data sets, it can IgG4 antibodies were offered by Hytest (Finland). Europium balance errors; (4) the learning process is fast. labeling kits (1244-302) were purchased from Perkin Elmer (USA). +e polystyrene microtiter plates were obtained from Nunc International (Denmark). +e recombinant PLA2R 3. Results antigen, series of standards of anti-human PLA2R-IgG and PLA2R-IgG4 were prepared in our laboratory as previously 3.1. Evaluation Measurements. +e accuracy (ACC) is uti- reported [13, 14]. All the buffer solutions were supplied from lized to evaluate the performance of the RF model under 5- Jiangyuan Co. (China). +e other reagents were of analytical fold cross-validation (5-CV). +e calculation method of grade and obtained from Sinopharm Chemical Reagent ACC is as follows: Journal of Healthcare Engineering 3 Table 1: +e information of data set. No. Feature (variable) Value r 1 Age (years) 55.110± 15.110 0.175 2 Gender (males/females) 56/39 0.105 3 Renal tubular atrophy score 1.057± 0.721 0.080 4 Renal interstitial fibrosis score 1.100± 0.731 0.113 5 Renal interstitial lymphoplasmacytic infiltrate score 1.068± 0.861 0.187 6 Total score of renal tubular and interstitium 3.236± 2.138 0.136 7 Pathological stage of IMN 1.452± 0.495 0.009 8 IF IgA 0.220± 0.477 0.089 9 IF IgM 0.252± 0.635 0.034 10 IF IgG 2.789± 0.697 0.009 11 IF C1q 0.242± 0.488 0.098 12 IF C3 1.205± 0.738 −0.035 13 Renal tissue PLA2R antigen 0.952± 0.357 −0.031 14 SBP (mmHg) 133.389± 14.973 0.017 15 DBP (mmHg) 79.989± 9.079 −0.109 16 Serum C3 (mg/L) 894.305± 253.312 −0.008 17 Serum C4 (mg/L) 241.147± 87.192 0.058 18 Serum IgA (g/L) 5.206± 28.002 0.061 19 Serum IgG (g/L) 7.084± 2.891 −0.077 20 Serum IgM (g/L) 1.257± 0.637 −0.013 21 Serum albumin (g/L) 23.335± 8.233 −0.449 22 Hematuria (/uL) 102.377± 138.301 −0.014 23 Serum creatinine (umol/L) 83.664± 34.042 0.168 24 eGFR-EPI (ml/min) 86.535± 24.060 −0.189 25 BUN (mmol/L) 4.986± 1.929 0.172 26 Serum glucose (mmol/L) 5.142± 0.810 0.321 27 Serum lithic acid (umol/L) 344.676± 89.678 0.048 28 TG (mmol/L) 2.406± 1.471 0.415 29 TC (mmol/L) 7.119± 2.289 0.141 30 LDL-C (mmol/L) 3.902± 1.379 0.077 31 HDL-C (mmol/L) 1.323± 0.432 −0.110 32 WBC (×10 /L) 8.577± 13.978 0.088 33 Hemoglobin (g/L) 122.204± 24.962 0.058 34 PLT (×10 /L) 219± 65.409 −0.104 35 C-reactive protein (mg/L) 2.807± 3.167 0.057 36 ESR (mm/H) 53.589± 34.176 0.233 37 Serum PLA2R-IgG (ng/mL) 5243.957± 9282.902 0.394 38 Serum PLA2R-IgG4 (ng/mL) 1762.615± 2662.328 0.524 denotes that each feature correlated with risk stratification of progression in IMN using Pearson correlation coefficient (r). c i increase in mean square error (MSE) averaged over all 􏽐 TP i�1 whole ACC � × 100%, trees in the ensemble and divided by the standard devi- ation taken over the trees, for each variable. So, we could get the importance scores of all input variables. In general, TP ACC � × 100%, (1) the larger the score, the more important it is for the prediction model. Figure 1 shows the results of relative c importance for inputs. It could be seen from the figure M � 􏽘 M , that the 21-th (albumin), 37-th (PLA2R-IgG), and 38-th i�1 (PLA2R-IgG4) variables had high values (>0.3) of 0.3156, 0.3981, and 0.7682, respectively. where c denotes the number of classes. TP denotes the number of true positive (TP) in subclass i. M and M denote the number of whole test samples and subclass test samples. 3.3. Comparison of PLA2R-IgG and PLA2R-IgG4. In order to ACC denotes the accuracy in subclass i. further evaluate the contribution of PLA2R-IgG and PLA2R- IgG4 to the model, we constructed four different models, 3.2. Relative Importance of Inputs in Estimating IMN. To which contain both PLA2R-IgG and PLA2R-IgG4 variables, evaluate the impact of each input variable on the final PLA2R-IgG variable, PLA2R-IgG4 variable, and no PLA2R- classification accuracy. We employed out-of-bag esti- IgG and PLA2R-IgG4 variables. +e relevant information of mates of variable importance in RF. +e RF stored the the models is shown in Table 2. 4 Journal of Healthcare Engineering 0.8 0.7 0.6 0.5 0.4 0.3 0.2 0.1 -0.1 -0.2 0 5 10152025303540 Feature Index Figure 1: Relative importance for inputs. Table 2: +e information of models. Model Number of features (variables) Model 1 38 With PLA2R-IgG and PLA2R-IgG4 Model 2 37 With PLA2R-IgG and without PLA2R-IgG4 Model 3 37 Without PLA2R-IgG and with PLA2R-IgG4 Model 4 36 Without PLA2R-IgG and PLA2R-IgG4 +e classification performance of the 4 models was Table 3: Comparison on four models via 5-fold cross-validation. verified by 5-fold cross-validation. And, the results are listed 1 2 3 Model Overall ACC ACC ACC ACC in Table 3. Obviously, when PLA2R-IgG and PLA2R-IgG4 were included, the prediction performance of the model Model 1 0.6743 0.8 0.6583 0.1 Model 2 0.6422 0.7778 0.5833 0.2 (model 1) was the best, and the overall classification accuracy Model 3 0.6639 0.7778 0.6806 0 was 0.6743. In the overall ACC, the performance of model 2 Model 4 0.6290 0.7556 0.6306 0 (0.6422) was not better than model 3 (0.6639). +is further 1 2 3 ACC : the accuracy of class 1; ACC : the accuracy of class 2; ACC : the verified that PLA2R-IgG4 is more important than PLA2R- accuracy of class 3. IgG for classification accuracy. In addition, the performance of the model (model 4) was the worst (0.6290), when PLA2R-IgG4 and PLA2R-IgG were not contained at the same time. From the above test, it could be found that Table 4: Analysis of statistical significance for different methods via PLA2R-IgG and PLA2R-IgG4 were very helpful for classi- 5-fold cross validation (10 times). fication. However, PLA2R-IgG4 could achieve better results than PLA2R-IgG. P value In this study, we employed t-test to evaluate the sig- Between model 1 and model 2 6.355e − 4 nificant differences of average ACC between different Between model 1 and model 3 0.0085 Between model 1 and model 4 1.8e − 5 models. +e results were list in Table 4, which show that the Between model 2 and model 3 0.1142 differences between model 1 and other three models were Between model 2 and model 4 0.1510 significant. Furthermore, the P value of model 3 and model 4 Between model 3 and model 4 0.0022 was 0.0022. +is means that the PLA2R-IgG4 feature had a more significant performance improvement compared to the ordinary model (without PLA2R-IgG and PLA2R-IgG4). major target antigen, progress was made in understanding the pathogenesis of IMN [6]. Circulating anti-PLA2R an- tibodies not only contribute to distinguish primary mem- 4. Discussion branous nephropathy from secondary membranous It is widely recognized that IMN is an autoimmune glo- nephropathy in diagnosis but also conduce to monitor the merular disease in which autoantibodies combine with immunological activity degree during the treatment period antigens on glomerular podocytes and deposit in glomerular [19]. Accordingly, the quantitative detection of circulating capillary walls [18]. With the discovery of M-type phos- anti-PLA2R antibodies is particularly important in diagnosis pholipase A2 receptor (PLA2R) which was identified as the and treatment of IMN. Out-of-Bag Feature Importance Journal of Healthcare Engineering 5 Among all the PLA2R-IgG antibodies, PLA2R-IgG4 Ethical Approval antibodies are predominant [6, 18]. Lacking of mature +e experimental protocol was established, according to the commercial testing means TRFIA was employed in this ethical guidelines of the Helsinki Declaration and was ap- study for the quantitative detection of anti-PLA2R-IgG4 proved by the Human Ethics Committee (Wuxi People’s antibodies. As a novel nonisotopic labeling technology, −18 Hospital Ethics Committee). +is study had been approved TRFIA has the advantages of high sensitivity (10 mol/L), by the ethics committee of the hospital (ethical approval no. wide monitoring range, and less susceptibility to matrix kyl2016001). interference [14]. Our previous work discovered that using the cutoff value of 161.2 ng/mL, anti-PLA2R-IgG4 had higher sensitivity in diagnosis than anti-PLA2R-IgG by the Consent cutoff value of 1990 ng/mL (90% versus 74%) [14]. +erefore, Written informed consent for publication was obtained we speculate that, in addition to anti-PLA2R-IgG, anti- from all participants. PLA2R-IgG4 may be an efficient biomarker in the assess- ment of the severity and prognosis of IMN too. In recent years, machine learning methods have been Conflicts of Interest widely used in medicine [20–22] and biology [23–25] to solve +e authors declare that they have no conflicts of interest. difficult data analysis problems for researchers. In our study, RF was employed to evaluate the importance of all the features (input variables). And, we found that the 21th (albumin), 37th Authors’ Contributions (PLA2R-IgG), and 38th (PLA2R-IgG4) variables had high +ey are joint first authors: Xiaobin Liu, Jing Xue, Xiaoyi values (>0.3) of 0.3156, 0.3981, and 0.7682, respectively, which Guo. meant these three features were more important to the risk stratification of progression in IMN. At the same time, as shown in Figure 1, PLA2R-IgG4 manifested a better predictive Acknowledgments value compared with PLA2R-IgG (0.7682 versus 0.3981). In +anks to the Department of Nephrology of Wuxi People’s addition, it could be seen from Table 3 that the prediction Hospital for collecting data in this study. +is study was effect of the PLA2R-IgG4 feature (0.6639) was better than that funded by the Top Talent Support Program for Young and of the PLA2R-IgG feature (0.6422). When both PLA2R-IgG4 Middle-Aged People of Wuxi Health Committee and PLA2R-IgG features were input into the model, its pre- (HB2020008), Medical and Public Health Project of Wuxi diction performance was the best (0.6743). +e above test Sci-Tech Development Fund (WX18 II AN047), the Sci- results further validated the importance of PLA2R-IgG4 and entific research project of Wuxi Health Committee PLA2R-IgG in assessing the risk level model. In Table 4, we (MS201927), the Scientific research project of Wuxi Health evaluated the significant differences between different models. Committee (Z201914), Major projects of precision medicine Obviously, the difference between model 1 and other models of Wuxi health Committee (J202001), Maternal and Child was significant (P value< 0.05). Compared with model 4, Health Research Project of Jiangsu Province (F202033), the model 3 also had significant difference (P value � 0.0022). It Scientific Research Projects of Jiangsu Provincial Health was obvious that PLA2R-IgG4 played an important role in Commission (LGY201801), and Jiangsu Province “333” improving the predictive ability of the model. Project (BRA2020142). 5. Conclusions References In this study, we evaluated the value of serum PLA2R-IgG4 [1] J. M. Hofstra and J. F. M. Wetzels, “Management of patients antibodies in predicting risk stratification of IMN by with membranous nephropathy,” Nephrology Dialysis establishing a random forest model. Compared with PLA2R- Transplantation, vol. 27, no. 1, pp. 6–9, 2012. IgG, PLA2R-IgG4 is a more efficient biomarker in predicting [2] F. C. Fervenza, S. Sethi, and U. Specks, “Idiopathic mem- the risk of progression in IMN. +e study results are sat- branous nephropathy: diagnosis and treatment,” Clinical isfactory, while disadvantages remain (1) there is lack of Journal of the American Society of Nephrology, vol. 3, analysis about the value of serum PLA2R-IgG4 in the pp. 905–919, 2008. [3] X. Pan, J. Xu, H. Ren et al., “Changing spectrum of biopsy- prediction of treatment effect; (2) the sample size needs to be proven primary glomerular diseases over the past 15 Years: a further expanded to minimize the prediction bias. In the single-center study in China,” Contributions to Nephrology, next work, we will expand the sample size and survey the vol. 181, pp. 22–30, 2013. predictive effect of serum PLA2R-IgG4 in the therapeutic [4] K. Zuo, Y. Wu, S.-J. Li, F. Xu, C.-H. Zeng, and Z.-H. Liu, regimen and prognosis of IMN. “Long-term outcome and prognostic factors of idiopathic membranous nephropathy in the Chinese population,” Data Availability Clinical Nephrology, vol. 79, pp. 445–453, 2013. [5] A. Schieppati, L. Mosconi, A. Perna et al., “Prognosis of +e data used to support the research can be obtained from untreated patients with idiopathic membranous nephropa- the corresponding authors according to the requirements of thy,” New England Journal of Medicine, vol. 329, no. 2, the institution. pp. 85–89, 1993. 6 Journal of Healthcare Engineering [6] L. H. Beck, R. G. Bonegio, G. Lambeau et al., “M-type transfer TSK fuzzy system,” IEEE Transactions on Intelligent phospholipase A2 receptor as target antigen in idiopathic Transportation Systems, vol. 22, no. 3, pp. 1752–1764, 2021. membranous nephropathy,” New England Journal of Medi- [23] H. Wang, J. Tang, Y. Ding, and F. Guo, “Exploring associa- tions of non-coding RNAs in human diseases via three-matrix cine, vol. 361, pp. 11–21, 2009. [7] J. N. Chi, T. S. Lai, C. F. Wu et al., “+e relationship of anti- factorization with hypergraph-regular terms on center kernel phospholipase A2 receptor antibody and C5a complement alignment,” Briefings in Bioinformatics, 2021. with disease activity and short-term outcome in idiopathic [24] Y. Jiang, X. Gu, D. Wu et al., “A novel negative-transfer- membranous nephropathy,” Journal of the Formosan Medical resistant fuzzy clustering model with a shared cross-domain Association, vol. 118, no. 5, pp. 898–906, 2019. transfer latent space and its application to brain CT image [8] S. Provatopoulou, D. Kalavrizioti, M. Stangou et al., “Cir- segmentation,” IEEE/ACM Transactions on Computational culating anti-phospholipase A2 receptor antibodies as a di- Biology and Bioinformatics, vol. 18, no. 1, pp. 40–52, 2021. agnostic and prognostic marker in Greek patients with [25] Y. Jiang, K. Zhao, K. Xia et al., “A novel distributed multitask idiopathic membranous nephropathy-a retrospective cohort fuzzy clustering algorithm for automatic MR brain image study,” Romanian Journal of Internal Medicine, vol. 57, no. 2, segmentation,” Journal of Medical Systems, vol. 43, no. 5, pp. 141–150, 2019. 118 pages, 2019. [9] A. S. Bomback, “Management of membranous nephropathy in the PLA2R era,” Clinical Journal of the American Society of Nephrology, vol. 13, no. 5, pp. 784–786, 2018. [10] J. Floege, S. J. Barbour, D. C. Cattran et al., “Management and Treatment of Glomerular diseases (part1):Conclusions from a Kidney Disease: Improving Global Outcomes (KDIGO),” Nephrology (Saint-Petersburg), vol. 24, no. 2, pp. 22–41, 2020. [11] W. Schlumberger, N. Hornig, S. Lange et al., “Differential diagnosis of membranous nephropathy with autoantibodies to phospholipase A2 receptor 1,” Autoimmunity Reviews, vol. 13, no. 2, pp. 108–113, 2014. [12] G. Cheng, J. Liu, A. Gilbert et al., “Serum phospholipase A2 receptor antibodies and immunoglobulin G subtypes in adult idiopathic membranous nephropathy: clinical value assess- ment,” Clinica Chimica Acta, vol. 490, pp. 135–141, 2019. [13] B. Huang, L. Wang, Y. N. Cao et al., “Improvement of idi- opathic membranous nephropathy diagnosis with ultrasen- sitive quantitative detection of anti-phospholipase A2 receptor,” e Journal of Allergy and Clinical Immunology, vol. 139, no. 6, pp. 1988–1990 e2, 2017. [14] B. Huang, Y. Zhang, L. Wang et al., “Phospholipase a2 re- ceptor antibody IgG4 subclass improves sensitivity and specificity in the diagnosis of idiopathic membranous ne- phropathy,” Kidney Blood Press Res.vol. 44, no. 4, pp. 848–857, [15] C. Daniel, “Cattran.Idiopathic membranous glomerulone- phritis,” Kidney International, vol. 59, pp. 1983–1994, 2001. [16] P. H. Nachman, J. C. Jennette, and R. J. Falk, “Membranous glomerulopathy,” Brenner BM. Brenner and Rector’s the Kidneypp. 1121–1131, Saunders, Boston, MA, USA, 9th edi- tion, 2012. [17] B. Leo, “Random forests,” Machine Learning, vol. 45, no. 1, pp. 5–32, 2001. [18] L. H. Beck and D. J. Salant, “Membranous nephropathy: from models to man,” Journal of Clinical Investigation, vol. 124, no. 6, pp. 2307–2314, 2014. [19] R. Ayalon and L. H. Beck, “Membranous nephropathy: not just a disease for adults,” Pediatric Nephrology, vol. 30, no. 1, pp. 31–39, 2015. [20] X. Y. Guo, W. Zhou, B. Shi et al., “An efficient multiple kernel support vector regression model for assessing dry weight of hemodialysis patients,” Current Bioinformatics, vol. 16, pp. 284–293, 2021. [21] J. Gao, L. Zhang, G. Yu, G. Qu, Y. Li, and X. Yang, “Model with the GBDT for colorectal adenoma risk diagnosis,” Current Bioinformatics, vol. 15, pp. 971–979, 2020. [22] Y. Jiang, Y. Zhang, C. Lin, D. Wu, and C. T. Lin, “EEG-based driver drowsiness estimation using an online multi-view and

Journal

Journal of Healthcare Engineering – Hindawi Publishing Corporation

Published: Aug 31, 2021

There are no references for this article.