Novel Insights From an Old Technique

Antibody-mediated rejection (AMR) is a significant challenge in kidney transplantation, often leading to reduced graft survival. AMR is characterized by a combination of immunological markers (eg, donor-specific antibodies [DSAs]) and histopathological features detected by light microscopy (LM) (capillaritis, glomerulitis, positive C4d) and electron microscopy (EM), regularly revised by the Banff committee.1 Histological changes, however, often manifest late, limiting early intervention opportunities. EM is widely used in pathology laboratories mostly to refine transplant rejection diagnosis. For instance, glomerular basement membrane duplication and peritubular capillary multilaminations (PTCML) detected by EM often support the diagnosis of AMR. Here, Haririan et al2 investigated whether EM could also identify early ultrastructural changes that would predict AMR. EM findings were analyzed alongside other established markers of AMR as well as graft survival. A notable strength of the study is that it was conducted in a relatively large cohort of >600 transplant recipients. The authors identified several key EM features that correlated with adverse graft outcomes. Notably, glomerular endothelial cell enlargement (GECE) exceeding 50% was present in nearly 30% of biopsies, whereas subendothelial expansion and basement membrane duplication (SEE/BMD) occurred in about 25% and PTCML with >4 extra layers in approximately 18% of cases. These ultrastructural alterations were found to be strongly associated with the presence and number of DSAs: the severity of GECE increased incrementally from patients with no DSA to those with class I, class II, and both courses of DSAs, with the odds of worsening GECE rising significantly as the number of DSAs increased. Similar trends were observed for PTCML. Importantly, multivariate analysis revealed that GECE of >50%, SEE/BMD, and, to a lesser extent, focal podocyte effacement of >50% were independent predictors of graft failure during a mean follow-up of 63 mo. Furthermore, in patients without evidence of graft injury following the main Banff classification histopathological features (score = 0 for glomerulitis, peritubular capillaritis, transplant glomerulopathy, and C4d deposition), severe GECE remained independently predictive of graft failure. This latter finding is crucial as it identifies a pattern of endothelial injury that is not obviously related to known definitions of rejection. The cause of such endothelial injury could be related to hypertension/vascular disease or thrombotic microangiopathy. Overall, these data incorporating systematic EM scoring, especially for parameters such as GECE and SEE/BMD, could facilitate earlier detection of AMR and allow a more accurate risk stratification of transplant recipients even in the absence of conventional LM features. Although the study provides valuable insights, some limitations warrant consideration. First, this is a single-center study, which may affect the generalizability of the findings. Results now need to be confirmed in other centers and across diverse populations. Additionally, the study’s observational nature prevents establishing a causal relationship between the identified ultrastructural changes and the development of AMR. Serial biopsies were only examined for a minority of patients. Although challenging, longitudinal studies tracking patients from transplantation through the development of complications such as AMR are necessary to confirm the predictive value of these early EM markers. Another limitation is the potential variability in EM interpretation between pathology laboratories. Detecting subtle ultrastructural changes requires significant expertise, and interobserver variability could influence diagnostic accuracy. Standardizing EM protocols and training pathologists would mitigate this issue, but this could also be challenging. Furthermore, while the study highlights a correlation between DSA and ultrastructural changes, it does not extensively explore the role of other antibodies, such as non-HLA antibodies.3 Remarkably, GECE and SEE/BMD were observed in a subset of transplant recipients without detectable DSA. Limitations aside, this study underscores the potential of EM to identify early ultrastructural changes associated with graft deterioration. By demonstrating that ultrastructural changes can predict adverse outcomes independently of standard histopathological criteria and DSA status, the study challenges the current diagnostic paradigms and advocates for a broader application of EM in kidney transplant biopsy evaluation. The novel association between the degree of endothelial injury on EM and graft survival, regardless of other LM indicators, represents a significant advance in the field, highlighting the importance of EM in refining diagnostic criteria and guiding therapeutic interventions in kidney transplantation. Further research is necessary to validate these findings and effectively integrate EM into routine clinical practice.