Validating the Banff Classification of Rejection in Vascularized Composite Allografts and a Look Beneath the Surface

Jinny Ha¹, Cinthia Drachenberg², Gerhard Mundinger³, John Papadimitriou², Borisiav Alexiev², Abdolreza Haririan⁴, Luke Jones¹, Stephen Bartlett¹, Eduardo Rodriguez³, Rolf Barth⁵.

¹Department of Surgery, University of Maryland School of Medicine; ²Department of Pathology, University of Maryland School of Medicine; ³Division of Plastics, Reconstructive, and Maxillofacial Surgery, University of Maryland School of Medicine; ⁴Division of Nephrology, University of Maryland School of Medicine; ⁵Division of Transplant Surgery, University of Maryland School of Medicine, Baltimore, MD, USA.

The successes of the limited number of hand and face transplants have made vascularized composite allografts (VCA) a clinical reality; however, pathologists have limited comparative experience in utilizing the Banff criteria for rejection. Therefore, we have assessed the clinical utility and reproducibility of the Banff schema for the skin component of VCA in our non-human primate models.

VCA biopsies from 28 Cynomolgus macaque recipients of a fibula to radius or heterotopic facial subunit VCA were included for analysis. A total of 198 H&E-stained biopsies were reviewed by 3 pathologists. The Banff CTA 2007 grading schema was used to score the severity of rejection. Deeper structures were evaluated when available. Kappa (k) statistical analysis was used to determine inter-rater agreement.

There was substantial inter-rater agreement with k=0.66 (p<0.0001). Highest concordance was seen when identifying grade IV rejection (k=0.75). There was moderate agreement for grades 0 to III (k=0.53-0.60). Perivascular and soft tissue infiltrates were seen in 70% of the specimens examined. These findings were seen in normal skin and in all grades of rejection. Deep structures were observed in 162 specimens. Notable changes in deep tissue vessels were observed in biopsies with grade III/IV rejection. Intimal arteritis was seen in 50% and 42.1% of the samples with grade III and IV rejection, respectively. Arterial necrosis (31.6%) and thrombosis (36.8%) were seen in biopsies with grade IV rejection.

The Banff schema for acute rejection is reproducible and reliable in monitoring graft rejection in our VCA model. Perivascular and soft tissue infiltrates appear to be a non-specific finding, seen in the absence and presence of rejection. Intimal arteritis, arterial necrosis, and thrombosis are associated with advanced rejection. There may be additional diagnostic and prognostic markers of rejection within the deep tissues, which the current Banff schema for VCA does not address.