Clinical Antibody Mediated Rejection of Lung is Not One Disease, but Includes Distinct Obstructive, Restrictive and Indolent Phenotypes

Purpose Chronic lung allograft dysfunction (CLAD) is the major determinant of mortality in lung transplant (LTX) recipients after 1 year. CLAD is strongly associated with de-novo HLA donor specific antibodies (DSA), but can present as bronchiolitis obliterans syndrome (BOS) or restrictive allograft syndrome (RAS). The term clinical antibody mediated rejection (AMR) is often used to refer to any patient with possible or probable AMR by ISHLT criteria, but this group may include more than one disease. Methods Hierarchical cluster analysis of LTX recipients treated for Clinical AMR 2008-18 at a single center, grouped by radiological opacities and change of FEV1/FVC. P-values adjusted for multi-comparisons (Bonferroni). Results Of 118 patients, by ISHLT criteria, none had definite AMR, 18% probable, 56% possible and 26% were unclassifiable. AMR grade was not correlated with specific clinical features, outcome or treatment response. Cluster analysis identified 3 clinical phenotypes: acute obstruction, restriction and indolent (O: R: I). Rates of possible and probable AMR were similar in each group. Clusters differed in the patterns of FEV1 (figure 1), FVC and ratio (p −8 each). 6 months after treatment, BOS 3 was 87%: 19%: 16% (p=0.003), RAS, 0%: 42%: 19% (p=0.02). The clusters differed in opacities (ground glass 44%: other subtypes 95%: none) (p −8 ), histopathology (A-grade: diffuse alveolar damage: none) (p=0.02), lavage macrophages (50%: 18%: 28%) (p Conclusion The current criteria for AMR do not describe one disease. We identified 3 main phenotypes with distinct clinical features, outcomes and treatment responses. Clinically validated acute phenotypes are needed.