Combination of the gut microbiota and clinical indicators as a potential index for differentiating idiopathic membranous nephropathy and minimal change disease

Abstract

Objectives

Membranous nephropathy (MN) and minimal change disease (MCD) are two common types of nephrotic syndrome that have similar clinical presentations but require different treatment strategies. Currently, the definitive diagnosis for these conditions relies on invasive renal biopsy, which can be limited in clinical practice.

Methods

In this study, we aimed to differentiate idiopathic MN (IMN) from MCD using clinical data and gut microbiota. We collected clinical data and stool samples from 115 healthy individuals, 115 IMN, and 45 MCD at the onset of disease and performed 16S rRNA sequencing. Through machine learning methods including random forest, logistic regression, and support vector machine, a classifier to differentiate IMN from MCD was constructed.

Results

Baseline clinical data comparing the IMN and MCD groups showed that the MCD had higher levels of hemoglobin, uric acid, cystatin C, β2-microglobulin, α1-microglobulin, total cholesterol, and low-density lipoprotein and lower levels of albumin and CD4⁺ T-cell counts. The gut microbiota of the two groups differed at all levels of the phylum and genus. Differential gut microbiota may disturb the integrity of the intestinal wall and lead to the passage of inflammatory mediators through the intestinal barrier, causing kidney injury. We constructed a noninvasive classifier with a discrimination efficacy of 0.939 that combined the clinical data and gut microbiota information to identify IMN and MCD.

Conclusions

The classifier of the gut microbiota combined with clinical indicators has achieved good performance in identifying IMN and MCD, which provides a new approach for the noninvasive discrimination of different pathological types of kidney disease.

Keywords:

• Idiopathic membranous nephropathy
• minimal change disease
• gut microbiota
• machine learning
• 16S rRNA

Acknowledgments

All of the human samples were obtained from the Biobank of the First Affiliated Hospital of Zhengzhou University and National Human Genetic Resources Sharing Service Platform (Grant No. 2005DKA21300).

Ethics statement

This study was approved by the Ethics Committee of the First Affiliated Hospital of Zhengzhou University (2021-KY-818 and 2021-KY-838).

Author contributions

ZZ and YJ designed the study. YJ and TW drafted the paper. TW, HL and GY collected the data. LW, RG and YZ collected the stool samples. TW, WY, FW analyzed the data and made the figures. ZZ, TW and YJ revised the paper. All authors approved the final version of the manuscript.

Disclosure statement

No potential conflict of interest was reported by the author(s).

Data availability statement

The datasets presented in this study can be found in online repositories. The names of the repositories and accession numbers are as follows: https://www.ncbi.nlm.nih.gov/, PRJNA752445 and PRJNA832071.

Additional information

Funding

This work was supported by the National Natural Science Foundation of China (Grant Nos. 81873611 and 8217033050), and the Science and Technology Innovation Team of Henan (Grant No. 17IRTSTHN020).