Recognition of i-IF/TA as a component of the T cell-mediated rejection spectrum: Unselected population approach vs random case selection

To the Editors: Whereas chronic active T cell-mediated rejection (TCMR) has long been acknowledged by the Banff classification, the spectrum of chronic active TCMR lesions was until now restricted to chronic allograft arteriopathy, without any consideration for chronic tubulo-interstitial injury following the acute infiltration of the interstitium and tubules by T cells and macrophages characterizing acute TCMR. To date, active TCMR has been defined only in the non-fibrotic areas of the allograft cortex and this artificial compartmentalization of the alloreactive response seems to have little biological relevance. The recognition of interstitial inflammation in fibrotic areas (i-IF/TA) and tubulitis in atrophic tubules (t-IF/TA) as components of chronic active TCMR represents a step forward in our understanding of the temporality and natural history of the TCMR process, consistent with the natural history of antibody-mediated rejection, which has been modified in the current era of more potent immunosuppression with the emergence of subclinical and chronic active forms. This temporal relationship between acute TCMR leading to chronic active TCMR reflected by i-IF/TA was highlighted in two recent independent studies. Nankivell et al1Nankivell BJ, Shingde M, Keung KL, et al. The causes, significance and consequences of inflammatory fibrosis in kidney transplantation: the Banff i-IFTA lesion [published online ahead of print 2017]. Am J Transplant. https://doi.org/10.1111/ajt.14609.Google Scholar demonstrated on serial protocol allograft biopsies a temporal sequence comprising initial cellular inflammation, followed by subsequent i-IF/TA, followed by accelerated progression of IF/TA and increased incidence of chronic arteriopathy. In our study,2Lefaucheur C, Gosset C, Rambant M, et al. T cell-mediated rejection is a major determinant of inflammation in scarred areas in kidney allografts [published online ahead of print 2017]. Am J Transplant. https://doi.org/10.1111/ajt.14565.Google Scholar we showed that 61% of patients with acute TCMR diagnosed within the first year after transplantation, receiving standard treatment, had post-treatment i-IF/TA and that i-IF/TA severity was associated with increased risks of allograft loss. While we agree that some of the i-IF/TA spectrum may indeed reflect an acute kidney injury/wound healing process, the arguments provided by Famulski and Halloran,3Famulski KS, Halloran PF. Letter re: Nankivell et al. [published online ahead of print 2017]. Am J Transplant. https://doi.org/10.1111/ajt.14653.Google Scholar refuting a relationship between i-IF/TA and chronic active TCMR, are preliminary at this stage since they are based on limited, unpublished data subject to potential flaws and biases, including the following:1Selected population consisting in an unknown number of patients providing 234 kidney allograft biopsies without inclusion and exclusion criteria for patient and biopsy selection.2No consideration of Banff criteria4Haas M, Loupy A, Lefaucheur C, et al. The Banff kidney meeting report: revised diagnostic criteria for chronic active T cell-mediated rejection, antibody-mediated rejection, and prospects for integrative endpoints for next-generation clinical trials [published online ahead of print 2017]. Am J Transplant. https://doi.org/10.1111/ajt.14625.Google Scholar for defining chronic active TCMR. As for cases with inflammation in non-fibrotic areas (i), i-IF/TA is not a specific lesion for chronic active TCMR diagnosis, which requires: (i) a minimal level of i-IF/TA intensity (i-IF/TA≥2); (ii) the presence of concurrent tubulitis, excluding severely atrophic tubules (t ≥ 2); and (iii) the exclusion of alternative etiologies of tubulo-interstitial inflammation such as BK virus nephropathy or bacterial infections, according to the 2017 Banff criteria.3Lack of kinetics data as part of a prospective systematic monitoring protocol: absence of consideration for temporal relationship between i-IF/TA and previous acute rejection episodes.4TCMR molecular classifier training bias, reflected by the fact that the performance of the Molecular Microscope TCMR classifier has not been evaluated in the setting of chronic active TCMR since it was generated on the basis of allograft biopsies showing acute TCMR with inflammation in non-scarred areas and undetermined i-IF/TA and t-IF/TA status.5Reeve J Sellares J Mengel M et al.Molecular diagnosis of T cell-mediated rejection in human kidney transplant biopsies.Am J Transplant. 2013; 13: 645-655Abstract Full Text Full Text PDF PubMed Scopus (157) Google Scholar As Famulski et al, we do believe that molecular approaches are interesting to further address the landscape of chronic active TCMR and better define disease activity in patients showing i-IF/TA. The Paris Transplant Group offers to make the biopsy material from our study available for molecular analysis to further facilitate study of chronic active TCMR. The authors of this manuscript have no conflicts of interest to disclose as described by the American Journal of Transplantation.