Molecular T-cell mediated rejection in transbronchial and mucosal lung transplant biopsies

We previously developed molecular assessment systems for lung transplant transbronchial biopsies (TBBs) with high surfactant and bronchial mucosal biopsies (3BMBs). Unsupervised machine learning identified T cell-mediated rejection (TCMR) based on expression of validated rejection-associated transcripts (RATs). The relationship to patient outcomes is not known. We aimed to establish whether molecular TCMR in TBBs and 3BMBs has implications for future graft loss. Molecular TCMR scores assigned in all TBBs or high surfactant TBBs agreed closely, indicating that variation in alveolar content in TBBs does not prevent detection of TCMR. Molecular assessments of 3BMB-3BMB pairs showed less variation than TBB-TBB pairs. TBB TCMR scores correlated with those in 3BMB. In both formats, molecular TCMR was associated with increased risk of graft loss whereas the presence of histologic rejection or donor-specific antibodies was not. Molecular TCMR can be detected in TBBs or 3BMBs and is associated with future risk of graft loss. TCMR emerges as a n important contributor to risk of failure. ClinicalTrials.gov NCT02812290. We used microarray analysis and machine learning to assign molecular TCMR scores to an expanded cohort of 457 TBBs (367 high surfactant and 90 low surfactant) and 314 3BMBs from lung transplant recipients. We tested agreement between biopsy pieces within and between TBB and 3BMB formats in the same patient. We also assessed the association between molecular TCMR and the risk of graft loss (death or retransplantation) and compared it to the prognostic associations for histology and presence of donor-specific antibodies in the same cohort.