Patrick T Gauthier, Martina Mackova, Ramsey Hachem, Jan Havlin, Alim Hirji, Peter Jaksch, Stephen Juvet, Shaf Keshavjee, Zsofia Kovacs, Daniel Kreisel, Bartosz Kubisa, Debbie J Levine, Robert Lischke, Maria Piotrowska, Jan Simonek, Irina Timofte, Laura van den Bosch, Greg Snell, Glen Westall, Andrea Zajacova, Kieran Halloran, Philip F Halloran
In lung transplants, the standard-of-care for diagnosing rejection, histologic assessment of transbronchial biopsies, has poor reproducibility and potential complications that limit its use in patients with compromised function. Potentially safer bronchial mucosal biopsies lack a consensus system for histologic evaluation but can be read molecularly. We redefined molecular rejection diagnoses in mucosal biopsies, assessed their ability to predict rejection in transbronchial biopsies, and searched for an antibody-mediated rejection state using clinical and molecular definitions. We used bulk transcriptomics in 526 mucosal biopsies from 433 transplants to cluster biopsies based on expression of rejection-related molecular scores, resulting in 387 No-rejection (NR), 109 T cell-mediated rejection (TCMR), and 30 NK cell-rich rejection-like (NKRL) biopsies. TCMR was associated with chronic lung allograft dysfunction, increased graft loss, and predicted molecular TCMR in transbronchial biopsies (AUC = 0.99). NKRL had mild molecular rejection features but no association with donor-specific antibody and was not associated with the 2016 International Society of Heart and Lung Transplantation definitions of antibody-mediated rejection. The alignment of the molecular profiles of T cell-mediated rejection in mucosal and transbronchial biopsies supports the potential utility of mucosal biopsies for diagnosing lung transplant rejection. Clinical trial registered with www.clinicaltrials.gov (NCT02812290).