Abstract: FR-PO719
A Novel Multimethylation Site Assay to Predict Steroid-Resistance in Initial Treatment of Nephrotic Syndrome
Session Information
- Top Trainee Posters - 1
October 24, 2024 | Location: Exhibit Hall, Convention Center
Abstract Time: 01:00 PM - 02:00 PM
Category: Pediatric Nephrology
- 1900 Pediatric Nephrology
Authors
- Yinfeng, Wang, Jiangsu University, Zhenjiang, Jiangsu, China
- Zhu, Junzhen, AlloDx Biotech, Shanghai, China
- Jiang, Tingya, AlloDx Biotech, Shanghai, China
Background
Approximately 20% of nephrotic syndrome (NS) patients exhibit steroids-resistant nephrotic syndrome (SRNS). If SRNS can be identified prior to the initial treatment, potential adverse effects associated with subsequent steroid use can be reduced by employing alternative immunosuppressive agents. Current research shows that about 30% of SRNS are caused by genetic mutations associated with podocytes. However, the causes of the other of 70% SRNS cases are unknown. In this study, we confirmed the correlation of gene methylation changes with SRNS and established an REscore to predict SRNS by detecting the multi-CpG sites of specific gene.
Methods
Frozen peripheral blood mononuclear cell (PBMC) samples from 25 non-genetic NS patients (12 SRNS and 13 non-SRNS) were retrospectively enrolled in this study. Systematic investigation into the DNA methylation in the promoter regions of three genes: NLRP3，NR3C1 and CASP1. Multiplexed target bisulfite sequencing (Target-BS) and a risk prediction model (Random Forest, RF) were established to distinguish SRNS and non-SRNS with 15 patients for training and 10 patients for validating.
Results
We investigated 37 CpG sites covered in total，and finally found out that 7 sites showed significant differences between the SRNS and non-SRNS group (p < 0.05). The 7 sites were analyzsed using RF model to obtain a resistance score (REscore) for predicting the response to steroid in NS. The results of the ROC analysis showed an AUC of 0.98，sensitivity of 90%，specificity of 100%，with a cutoff value REscore is 0.74，positive predictive value (PPV) was 100%，and the negative predictive value (NPV) was 93.3% in the training set. In the validation set，the RF model achieved an AUC of 0.92，sensitivity of 83.3%，specificity of 100%，PPV of 100%, and NPV of 80%.
Conclusion
In conclusion, this study identified significant changes methylation markers in the NLRP3，NR3C1 and CASP1 genes in patients with SRNS. Furthermore，the RF model was used to create the REscore, which can effectively predict the risk of SRNS. It is need more samples and more pathological type of nephropathy to verify the conclusion of this study in future.