Xu Yue, Wang Yuxuan, Zhang Di, Zhang Hao, Wang Yicun, Wang Wei, Hu Xiaopeng
Department of Urology, Beijing Chao-Yang Hospital, Capital Medical University, Beijing 100020, China; Institute of Urology, Capital Medical University, Beijing 100020, China.
Department of Urology, Beijing Chao-Yang Hospital, Capital Medical University, Beijing 100020, China; Institute of Urology, Capital Medical University, Beijing 100020, China.
Transpl Immunol. 2024 Jun;84:102021. doi: 10.1016/j.trim.2024.102021. Epub 2024 Mar 6.
Antibody-mediated rejection (ABMR) emerged as a major cause of graft loss in renal transplantation. Needle biopsy is the gold standard for diagnosis of ABMR in renal allografts. Thus, noninvasive diagnosis methods of ABMR with high accuracy are urgently needed to prevent unnecessary biopsies.
We collected peripheral blood transcriptome data from two independent renal transplantation cohorts with patients with ABMR, stable well-functioning transplants (STA), and T-cell mediated rejection (TCMR). Differentially expressed genes (DEGs) were identified by comparing the ABMR group with the STA group. In addition, functional enrichment analysis and gene set enrichment analysis were performed to seek new key underlying mechanisms in ABMR. Subsequently, we utilized a Boruta algorithm and least absolute shrinkage and selection operator logistic algorithm to establish a diagnostic model which was then evaluated and validated in an independent cohort.
According to functional enrichment analysis, autophagy was found to be the primary upregulated biological process in ABMR. Based on algorithms, three autophagy-associated genes, ubiquitin specific peptidase 33 (USP33), Ras homolog mTORC1 binding (RHEB), and ABL proto-oncogene 2 (ABL2), were selected to establish the diagnostic model in the training cohort. This autophagy-related gene model possessed good diagnostic value in distinguishing ABMR from STA blood samples in the training cohort (AUC = 0.907) and in the validation cohort (AUC = 0.972). In addition, this model also showed good discernibility in distinguishing ABMR from TCMR in the training and validation cohorts (AUCs = 0.908 and 0.833).
We identified and validated an autophagy-associated diagnostic model with high accuracy for renal transplant patients with ABMR. Our study provided a new potential test for the non-invasive diagnosis of ABMR in clinical practice and highlighted the importance of autophagy in ABMR.
抗体介导的排斥反应（ABMR）已成为肾移植中移植物丢失的主要原因。肾活检是诊断肾移植中ABMR的金标准。因此，迫切需要高精度的ABMR无创诊断方法以避免不必要的活检。
我们收集了两个独立肾移植队列的外周血转录组数据，这些队列中的患者分别患有ABMR、移植功能稳定良好（STA）以及T细胞介导的排斥反应（TCMR）。通过比较ABMR组和STA组来鉴定差异表达基因（DEG）。此外，进行功能富集分析和基因集富集分析以探寻ABMR中新的关键潜在机制。随后，我们利用Boruta算法和最小绝对收缩和选择算子逻辑回归算法建立诊断模型，然后在一个独立队列中对其进行评估和验证。
根据功能富集分析，发现自噬是ABMR中主要上调的生物学过程。基于算法，选择了三个自噬相关基因，泛素特异性肽酶33（USP33）、Ras同源mTORC1结合蛋白（RHEB）和ABL原癌基因2（ABL2），在训练队列中建立诊断模型。这个自噬相关基因模型在训练队列中区分ABMR和STA血样时具有良好的诊断价值（AUC = 0.907），在验证队列中也是如此（AUC = 0.972）。此外，该模型在训练和验证队列中区分ABMR和TCMR时也显示出良好的辨别能力（AUC分别为0.908和0.833）。
我们鉴定并验证了一个针对ABMR肾移植患者的高精度自噬相关诊断模型。我们的研究为临床实践中ABMR的无创诊断提供了一种新的潜在检测方法，并突出了自噬在ABMR中的重要性。
