Cluster Analysis Shows that Chronic Damage in Early Indication Biopsies Predicts Long-Term Graft Survival, While Early Active Inflammation Does Not

Abstract

Background: Recently, it has become clear that specific disease like antibody-mediated rejection and glomerular pathologies are very often the final reason for graft loss. It remains to be determined how early chronic histological damage, without specific pathophysiologic explanation, affects long-term graft survival. Methods: All (N=1197) recipients of a renal allograft or a combined kidney-pancreas graft, transplanted at a single center between January 1, 1991 and December 31, 2001 were included in this study. The data were collected prospectively. All indication biopsies (N=917 in 563 patients) performed within the first year after transplantation were rescored according to the current Banff classification. Principal component analysis, k means clustering and unsupervised hierarchical clustering was applied to reveal the relationships between the different histological lesions. Survival analysis was performed using Cox proportional hazards analysis and log-rank testing. Mean follow-up time is 15.3±2.97 years. Results: Adjusted for baseline demographic covariates, arteriolar hyalinosis (p=0.003) and transplant glomerulopathy (p< 0.0001) independently predict long-term graft survival. Inflammatory lesions, including tubulitis, interstitial inflammation, microcirculation inflammation and C4d deposition in the first year after transplantation associate with early graft loss, but do not affect long-term graft survival. Unsupervised hierarchical clustering analysis of the histological lesions confirms this clear distinction between the chronic histological lesions and lesions associated with inflammatory activity (Figure A). K-means clustering analysis identifies 2 clearly distinct patient groups, which differ from each other with respect to these same chronic histological lesions (interstitial fibrosis, tubular atrophy, arteriolar hyalinosis, mesangial matrix increase, transplant glomerulopathy, vascular intimal thickening and glomerulosclerosis). This clustering of the patients according to the chronic lesions offers a better predictor of long-term graft survival than each of the chronic lesions separately (HR 2.1 [CI 1.5-3.1]; p=0.0001) (Figure B).[Figure 1]Conclusion: Individual chronic lesions like arteriolar hyalinosis, tubular atrophy, interstitial fibrosis, glomerulosclerosis, mesangial matrix increase and vascular intimal thickening closely associate with each other. These lesions cannot be seen as individual entities and likely share common pathophysiology. It is not possible to tell apart the individual contribution of these different lesions to the clinical outcome of these patients. The global burden of early chronic histological damage within the first year after transplantation importantly affects the fate of the allografts, while active inflammation in early indication biopsies is not associated with long-term graft survival.