Molecular Assessment of Single-Piece Mucosal Biopsies is Highly Reproducible Compared to Transbronchial Biopsies in Lung Transplantation

M. D. Parkes, P. F. Halloran, I. L. Timofte, G. I. Snell, G. P. Westall, A. Roux, J. Havlin, R. Hachem, D. Kreisel, D. Levine, S. Juvet, S. Keshavjee, P. Jaksch, W. Klepetko, A. Hirji, J. Weinkauf, K. M. Halloran

Research output: Contribution to journalMeeting Abstractpeer-review


PURPOSE: Histologic assessment for rejection on transbronchial biopsies (TBB) has poor reproducibility. Mucosal biopsies are safer than TBBs but their histologic assessment has not been systematically studied or standardized. We developed molecular platforms to assess TBBs and mucosal biopsies from the third airway bifurcation (3BMB). We examined the reproducibility of scores assigned by molecular assessment of both biopsy types in lung transplant patients from 9 international centers. METHODS: We trained machine learning models (archetypal analysis, principal component analysis) on microarray data from 457 TBBs and 243 3BMBs. We processed 14 pairs of TBB pieces (28 total) taken during the same procedure and 13 pairs of 3BMBs (26 total). The models assigned each biopsy sets of archetype scores (denoted by "R") and principal component scores (denoted by "PC") describing injury and rejection phenotypes. We studied within-score agreement between biopsy pairs using intraclass correlations (ICC). RESULTS: Most scores had excellent reproducibility in 3BMB pairs compared to TBBs (Table 1). In 3BMBs the R1 (normal), R2 (rejection), R3 (late injury), PC1 (inflammation/rejection), and PC2 (late injury) scores were highly reproducible, but R4 and PC3 (both recent injury) were not. In TBBs the R2 (rejection) and PC1 (inflammation/rejection) scores were moderately reproducible; other scores had poor reproducibility. CONCLUSION: Molecular assessment of paired, single piece 3BMBs achieves excellent reproducibility in its key scores. TBB scores were less reproducible likely because of tissue heterogeneity and greater anatomic distance between samples. Measurements could be stabilized by combining ≥2 pieces per microarray. Importantly, molecular assessment with 3BMBs or TBBs requires fewer pieces (2-3) than TBB histology, which requires up to ten. These data support continued study of the diagnostic and prognostic utility of 3BMBs in lung transplantation.

Original languageEnglish
Pages (from-to)S191-S192
Number of pages2
JournalThe Journal of Heart and Lung Transplantation
Issue number4
Publication statusPublished - Apr 2020
Externally publishedYes
EventAnnual Meeting and Scientific Sessions of the International-Society-for-Heart-and-Lung-Transplantation 2020 - Montreal, Canada
Duration: 22 Apr 202025 Apr 2020
Conference number: 40th

Cite this