Refinement of humoral rejection effector mechanisms to identify specific pathogenic histological lesions with different graft outcomes

In this issue of the American Journal of Transplantation, Senev et al reported a large retrospective study of kidney transplant patients (n = 935), in whom the impact on graft outcome of the histological phenotype of antibody-mediated rejection following the 2015 Banff classification (ABMRh) was investigated according to the presence or absence of circulating HLA-donor-specific antibody (DSA) (DSAposABMRh and DSAnegABMRh, respectively).1 Additionally, authors assessed whether C4d-positivity in peritubular capillaries (PTCs) could be used as a surrogate for circulating DSA, as proposed by the 2017 Banff classification.2 Interestingly, while the authors found that 208 (22.2%) patients were diagnosed as ABMRh, only 85/208 (40.8%) fulfilled the complete ABMR criteria because they also showed HLA-DSA (DSAposABMRh). Conversely, up to 123/208 (58.2%) were DSAnegABMRh. Notably, DSAposABMRh patients displayed significantly worse graft outcome than DSAnegABMRh, regardless of concomitant T cell–mediated lesions or C4d deposition in PTCs, thus suggesting a more transient and benign phenotype as compared with those with detectable DSA. While the characterization of histopathological lesions of kidney allografts has significantly improved in the past decades, this major progress has been achieved in part by the better identification of the underlying immune mechanism causing such lesions. In fact, with the development of more sensitive tools identifying circulating DSA and the identification of C4d positivity, we have better integrated and grouped other microvascular inflammation and remodeling changes (including PTC and basement membrane reduplication) within the same histological phenotype of ABMR.3 Likewise, certain specific graft lesions such as endarteritis, classically thought to be T cell–mediated, have now been better understood and may be antibody mediated in some patients.3 Moreover, ABMR-related gene expression assessed in tissue graft biopsies has also helped to add further insight on the effector mechanism of allograft damage despite the absence of detectable circulating DSA.4 In the same issue of the journal,5 our group reported in a large number of kidney transplant patients a similar proportion and graft outcome of patients with DSAposABMRh (54/92; 58.7%) and DSAnegABMRh (34/92; 37%) than that found by Senev et al1 (Figure 1A) (Table 1). Using new technology enabling the assessment of circulating donor (HLA)-reactive memory B cells (mBC), we identified that while most of DSAposABMRh showed donor-reactive mBC, an important proportion of DSAnegABMRh (72%) did also show circulating donor-reactive mBC. Most importantly, when graft outcome was stratified according to the histological phenotype and the presence or not of both DSA and donor-reactive mBC (Figure 1B), DSAnegABMRh graft outcome was clearly split into 2 groups; while DSAnegABMRh and no donor-reactive mBC displayed a significantly superior graft survival compared with DSAposABMRh patients, and similar to that of transplant recipients with normal graft parenchyma, DSAnegABMRh with donor-reactive mBC showed worse graft survival, comparable to that of DSAposABMRh, thus strongly suggesting the same immune pathogenic mechanism of allograft damage in these patients. While larger, prospective validation studies are ongoing, we believe that the refinement of the mechanisms of humoral rejection by tracking circulating donor-reactive mBC together with circulating DSA may allow a better discrimination of transplant patients with ABMRh with a clinically benign course from those that may need further attention and eventually guided therapeutic decision-making. The authors of this manuscript have no conflicts of interest to disclose as described by the American Journal of Transplantation.