Reproducibility of Complement 4d deposition by immunofluorescence and immunohistochemistry in lung allograft biopsies

• Published in: The Journal of heart and lung transplantation Vol. 33; no. 12; pp. 1223 - 1232
• Main Authors: Roden, Anja C., MD, Maleszewski, Joseph J., MD, Yi, Eunhee S., MD, Jenkins, Sarah M, Gandhi, Manish J., MD, Scott, John P., MD, Christine Aubry, Marie, MD
• Format: Journal Article
• Language: English
• Published: United States Elsevier Inc 01.12.2014
• Subjects: Adult; Aged; Allografts; Biopsy; C4d; Complement C4b - metabolism; Fluorescent Antibody Technique - methods; Graft Rejection - diagnosis; Graft Rejection - immunology; Humans; immunofluorescence; immunohistochemistry; Immunohistochemistry - methods; Lung - metabolism; Lung - pathology; Lung Transplantation; Middle Aged; Observer Variation; Peptide Fragments - metabolism; reproducibility; Reproducibility of Results; Retrospective Studies; Surgery; C4d; lung transplantation; immunohistochemistry; reproducibility; immunofluorescence
• ISSN: 1053-2498; 1557-3117
• DOI: 10.1016/j.healun.2014.06.006

Background The significance of Complement 4d (C4d) deposition in the diagnosis of antibody-mediated rejection (AMR) in lung allografts is controversial. A potential cause may be the problematic reproducibility. We studied the reproducibility of C4d by immunofluorescence (IF) and immunohistochemistry (IHC) in lung allograft transbronchial biopsies (TBBx), correlated C4d by IF with IHC, and compared the results with clinical findings. Methods TBBx obtained between 2009 and 2012 were stained with C4d by IHC and available corresponding frozen tissue was stained with C4d by IF. Capillary C4d staining was scored as 0, 1+ (<10% of capillaries), 2+ (10% to 50%) or 3+ (>50%). Four lung transplant pathologists independently scored all slides. Serum donor-specific antibodies (DSA) from within 2 months of the TBBx were noted. Results Of 228 consecutive TBBx, C4d by IHC ( n = 221) and IF ( n = 60) revealed agreement of 46.6% (95% CI 40.4% to 53.0%; κ = 0.13) and 81.4% (95% CI 68.7% to 89.7%; κ = 0.18), respectively. Correlations between IF and IHC varied among reviewers (0.10 to 0.36) (agreement 59.7% to 86.8%). Three patients were clinically diagnosed as having AMR or “possible” AMR. Conclusions Agreement of IF-based C4d staining in lung allografts appears superior to that of IHC. However, overall agreement for IF and IHC among pathologists is sub-optimal and correlation between IF and IHC is low, although this may be influenced in part by the low variability of the results given the mostly negative biopsies. C4d 3+ may be specific for AMR, although rare. A multidisciplinary approach seems to be the best way to diagnose AMR in lung allograft biopsies.