Histopathologic Clusters Differentiate Subgroups Within the Nonspecific Diagnoses of CAN or CR: Preliminary Data from the DeKAF Study

Abstract

The nonspecific diagnoses ‘chronic rejection’‘CAN’, or ‘IF/TA’ suggest neither identifiable pathophysiologic mechanisms nor possible treatments. As a first step to developing a more useful taxonomy for causes of new-onset late kidney allograft dysfunction, we used cluster analysis of individual Banff score components to define subgroups. In this multicenter study, eligibility included being transplanted prior to October 1, 2005, having a ‘baseline’ serum creatinine ≤2.0 mg/dL before January 1, 2006, and subsequently developing deterioration of graft function leading to a biopsy. Mean time from transplant to biopsy was 7.5 ± 6.1 years. Of the 265 biopsies (all with blinded central pathology interpretation), 240 grouped into six large (n > 13) clusters. There were no major differences between clusters in recipient demographics. The actuarial postbiopsy graft survival varied by cluster (p = 0.002). CAN and CNI toxicity were common diagnoses in each cluster (and did not differentiate clusters). Similarly, C4d and presence of donor specific antibody were frequently observed across clusters. We conclude that for recipients with new-onset late graft dysfunction, cluster analysis of Banff scores distinguishes meaningful subgroups with differing outcomes. The nonspecific diagnoses ‘chronic rejection’‘CAN’, or ‘IF/TA’ suggest neither identifiable pathophysiologic mechanisms nor possible treatments. As a first step to developing a more useful taxonomy for causes of new-onset late kidney allograft dysfunction, we used cluster analysis of individual Banff score components to define subgroups. In this multicenter study, eligibility included being transplanted prior to October 1, 2005, having a ‘baseline’ serum creatinine ≤2.0 mg/dL before January 1, 2006, and subsequently developing deterioration of graft function leading to a biopsy. Mean time from transplant to biopsy was 7.5 ± 6.1 years. Of the 265 biopsies (all with blinded central pathology interpretation), 240 grouped into six large (n > 13) clusters. There were no major differences between clusters in recipient demographics. The actuarial postbiopsy graft survival varied by cluster (p = 0.002). CAN and CNI toxicity were common diagnoses in each cluster (and did not differentiate clusters). Similarly, C4d and presence of donor specific antibody were frequently observed across clusters. We conclude that for recipients with new-onset late graft dysfunction, cluster analysis of Banff scores distinguishes meaningful subgroups with differing outcomes. Kidney allograft recipients with late deterioration of kidney function have often been labeled as having ‘chronic rejection’ or ‘chronic allograft nephropathy’ (CAN). More recently, the terminology, ‘interstitial fibrosis and tubular atrophy, with no evidence of any specific etiology’ (IF/TA) replaced ‘CAN’ and is used to describe biopsies with fibrosis and atrophy and with no obvious underlying pathogenesis (1Solez K Colvin RB Racusen LC et al.Banff ’05 meeting report: Differential diagnosis of chronic allograft injury and elimination of chronic allograft nephropathy (‘CAN’).Am J Transplant. 2007; 7: 518-526Crossref PubMed Scopus (934) Google Scholar). These diagnoses, which have been applied to nearly 50% of recipients with late graft loss, are not specific, and suggest neither identifiable pathophysiologic mechanisms nor possible treatments. As a consequence, it has been difficult to develop intervention trials. The Long-Term Deterioration of Kidney Allograft Function (DeKAF) study, a multicenter observational study conducted at seven transplant centers in the United States and Canada, was designed with the premise that new-onset late graft dysfunction, if studied early in its course, could be characterized as (or associated with) distinct clinical and histopathologic entities with reproducible phenotypes. DeKAF is focused on current nonspecific diagnoses (e.g. CAN-IF/TA) and not on already well-defined entities (e.g. recurrent disease). The current pathogenetic characterization of most late graft dysfunction (CAN–IF/TA) seems somewhat analogous to differentiating subgroups within the diagnosis of ‘Bright’s disease’ 5 decades ago (2Fishberg AM The subdivisions of Bright’s disease.in: Fishberg AM Hypertension and Nephritis. Lea and Febiger, Philadelphia1939: 324-332Google Scholar). Our goal is to determine if distinct subgroups can be defined on the basis of clinical, laboratory and pathologic information available at the time of new-onset graft dysfunction. Ultimately, these defined phenotypes could provide the basis for specific intervention trials. The DeKAF study consists of two cohorts of kidney transplant recipients: (a) a cross-sectional cohort of patients transplanted prior to October 1, 2005, and developing late graft dysfunction leading to a biopsy and (b) a prospective cohort of consecutive consenting recipients entered into the study at the time of transplant (3Gourishankar S, Leduc R, Connett J et al. Pathological and clinical characterization of the ‘troubled transplant’: Data from the DeKAF study. Am J Transplant, in press.Google Scholar). This report focuses on the cross-sectional cohort. We previously described the demographics of this cohort, and also reported on the local pathologists’ interpretations of biopsies done at the time of development of new-onset late graft dysfunction: CAN and calcineurin inhibitor (CNI) toxicity were the most common diagnoses (3Gourishankar S, Leduc R, Connett J et al. Pathological and clinical characterization of the ‘troubled transplant’: Data from the DeKAF study. Am J Transplant, in press.Google Scholar). Importantly, we noted that the local diagnosis of CAN was of no prognostic significance, as patients with or without this diagnosis had similar rates of postbiopsy graft failure (3Gourishankar S, Leduc R, Connett J et al. Pathological and clinical characterization of the ‘troubled transplant’: Data from the DeKAF study. Am J Transplant, in press.Google Scholar). Herein, we describe the use of cluster analysis, which groups patients with similar characteristics, to differentiate subsets within this cohort of recipients with new-onset, late allograft dysfunction (4Available at: http://en.wikipedia.org/wiki/Data_clustering. Accessed September 28, 2009.Google Scholar, 5Available at: http://www.statsoft.com/textbook/stcluan.html.Google Scholar, 6Available at: http://faculty.chass.ncsu.edu/garson/PA765/cluster.html. Accessed September 28, 2009.Google Scholar). Clustering, based on centrally read Banff scores, separated recipients with new-onset late graft dysfunction into meaningful subgroups with differing histologic characteristics, outcomes and presumably, responses to potential therapeutic interventions. Recipients were eligible for enrollment in the cross-sectional cohort if they had undergone a kidney or kidney-pancreas transplant prior to October 1, 2005, had a ‘baseline’ serum creatinine ≤2.0 mg/dL before January 1, 2006, and subsequently developed deterioration of graft function which led to a kidney allograft biopsy. Deterioration of function was defined as (1) an unexplained and persistent ≥25% increase of creatinine over baseline (in the absence of potential confounding factors), or (2) new-onset proteinuria (defined as albumin/creatinine ratio ≥0.2 or a protein/creatinine ratio > 0.5). Importantly, the increase of ≥25% in creatinine over baseline could occur over any follow-up interval (e.g. days to years). At the time of the biopsy, informed consent was obtained and baseline recipient information was recorded. In addition, blood and urine were collected for determination of donor specific antibodies (DSA) and metabolomics, respectively. Follow-up data were prospectively collected every 6 months and at any ‘event’ (e.g. graft loss). Allograft biopsies were read by the local pathologist, and the pathologic diagnosis was used to guide clinical care and any adjustments to immunosuppressive medications. Clinical care was provided using local protocols. Representative histologic sections were submitted to a central laboratory for analysis. Sections were stained for C4d (immunoperoxidase) and interpreted as positive if >10% peritubular capillaries stained positive (7Grande J Matas A Kasiske B et al.C4d deposition as a marker of adverse long-term renal allograft outcome.Am J Transplant. 2009; 9: 583Google Scholar). All biopsies were interpreted by the same pathologist (JG), masked to local diagnoses, clinical information and pathology reports, using the Banff 97 classification (8Racusen LC Solez K Colvin RB et al.The Banff 97 working classification of renal allograft pathology.Kidney Int. 1999; 55: 713-723Abstract Full Text Full Text PDF PubMed Scopus (2783) Google Scholar). Serum samples (2 mL) collected at the time of each biopsy were processed, frozen at –80 degrees and sent to a central laboratory (UCLA) in batches, along with information regarding the patient and donor HLA types and the patient’s pretransplant sensitization status. Sera were tested for anti-donor HLA antibodies, using microparticles with individual purified HLA antigens (Ag) covalently bound as targets (One Lambda, Inc, Canoga Park, CA. (a)Inflammation in areas of atrophy. In a preliminary analysis we examined whether inflammation in areas of atrophy (iatr), tubulitis in areas of atrophy (tatr) and peritubular capillary infiltrates (ptc) (each scored by central pathology) correlated with outcome. Iatr and tatr were scored in the same manner as i and t, except being scored in areas of fibrosis and/or atrophy. Scoring of ptc infiltrates was done using the Banff 2007 classification (9Solez K Colvin RB Racusen LC et al.Banff 07 classification of renal allograft pathology: Updates and future directions.Am J Transplant. 2008; 8: 753-760Crossref PubMed Scopus (1638) Google Scholar). In univariate analysis, each of these correlated with subsequent graft survival (p < 0.001), and were included in our subsequent analyses.(b)Cluster analysis. We used hierarchical cluster analysis of Banff scores (as determined by central pathology) (SAS Proc Cluster) to identify subgroups. Cluster analysis categorizes a collection of observations into subsets (clusters) based on a preselected similarity metric computed from a fixed list of variables (4Available at: http://en.wikipedia.org/wiki/Data_clustering. Accessed September 28, 2009.Google Scholar, 5Available at: http://www.statsoft.com/textbook/stcluan.html.Google Scholar, 6Available at: http://faculty.chass.ncsu.edu/garson/PA765/cluster.html. Accessed September 28, 2009.Google Scholar,10Everitt B Landau S Leese M Cluster Analysis.4th Ed. Oxford University Press, London, Arnold, New York2001Crossref Google Scholar, 11Aldenderfer M Blashfield R Cluster analysis, series: Quantitative applications in the social sciences. Sage Publications Inc, Beverly Hills, CA1984Crossref Google Scholar, 12Milligan GW An examination of the effect of six types of error perturbation on fifteen clustering algorithms.Psychometrika. 1980; 45: 325-342Crossref Scopus (834) Google Scholar), and hence is a particularly useful technique for defining individuals with similar characteristics that may reflect unknown pathogenic mechanisms. Ultimately these clusters may become defined clinical entities that predict outcomes, and suggest potential treatments to improve those outcomes. Cluster analysis was applied to the 265 biopsies having complete centrally determined histology at the time of analysis, and whose local clinical diagnosis did not show BK virus nephropathy or recurrent disease. These diagnoses were eliminated a priori as our focus is on differentiating entities among those with nonspecific diagnoses for which, ultimately, intervention trials could be developed. In preliminary analyses, we explored which variables to use for clustering. Our criteria for selecting variables included: (1) availability at the time of diagnosis; (2) reasonable sample size; and, (3) minimization of strong correlations between variables used in computing dissimilarity. For example, in our preliminary analyses, the Banff i- and t-scores had an observed Pearson correlation of R = 0.81; therefore, both scores were likely to make similar contributions. Including both of these variables in the clustering algorithm without regard to this correlation would potentially disproportionately emphasize this common characteristic. Therefore we only used 1 of these 2 scores in the clustering analyses. Of note, we excluded ‘v’ as there were only 13 positive cases (v > 0); of these, 11 were v= 1, 2 were v= We did correlations of clustering variables and correlated variables variables in with correlation were for a as a in the of selecting which variables to was to to the of variables and to variables with For this analysis, we did the clustering based on the Banff scores and Of the scores tatr and we tatr it was a of inflammation in areas of atrophy and was not as correlated with the i score as was correlation between individual Banff with correlation are = in a We used the to determine the of (1) of the observations to clusters each or more (2) of the between clusters > and (3) a local of the In our preliminary analyses, this at most two for the of clusters to use in a cluster analysis, a of clusters with a large of recipients in each cluster or a of clusters with recipients in each between these two into of the clustering using of cluster differences and with clinical on the clustering are in of clusters. We and developed a to the of the Banff scores for each cluster is as a or with the individual Banff as on the For each Banff the of a within the the of recipients in the cluster have a score For example, in 1, which the for Cluster 1 biopsies have or have the is the of the the score for each Banff is by the of the and The of the the of biopsies with Banff 1 in Cluster 1 have Banff 1 the of the and the of biopsies with Banff 2 and respectively. the between for The cluster provide a to the Banff scores within each cluster and to both the and differences between outcome. For recipients in each we determined actuarial graft survival and of graft In for each we determined local pathologists’ and diagnoses, the of recipients that had C4d positive biopsies or had and the time from transplant to biopsy. In the of clusters to the time to first of biopsy for we J R The analysis of failure time New Scholar). At the time of analysis, had complete central pathology Of these, biopsy from 265 were used in the cluster analysis and the basis of this A of recipients were excluded they had specific recurrent (n = and BK virus nephropathy (n = For studied serum creatinine prior to January was ± creatinine at biopsy was ± Mean time from transplant to biopsy was 7.5 ± 6.1 years). the criteria described clustering was based on Banff scores of and and the score of tubulitis in areas of atrophy Our analysis in six which included 240 of the 265 biopsies to clusters cluster = 2 the 6 clusters (and the for with the of Banff scores used in the analysis. the same clusters Banff scores and the scores for and Of note, of the biopsies were interpreted as having and the also had we are a with 2 for each Banff score in each the score and the and In the we have provided the of each Banff score by based on Banff (n = scores tatr and are observations in seven clusters not tubulitis in areas of inflammation in areas of peritubular capillary and for each Banff score in each = = = = = = in a We no major differences between clusters with regard donor or recipient kidney donor transplant transplant or immunosuppressive (p = In addition, baseline serum creatinine (defined as creatinine prior to January did not between clusters having an and had creatinine and at biopsy creatinine prior to January 1, and creatinine and the of in renal at the time of biopsy, by creatinine at to < Cluster < Cluster < Cluster < Cluster < Cluster < Cluster < Cluster the of in renal in a Importantly, actuarial graft survival by cluster = 5 = Cluster 1, a with Banff 1 scores and no inflammation had the Cluster with more inflammation and the The postbiopsy follow-up interval and the of graft for each cluster are in There was no between clusters in of follow-up for time to graft failure by in a follow-up and of graft by follow-up follow-up ± ± ± ± ± ± in a In the local pathologists’ diagnoses did not differentiate clusters A similar of the biopsies in each cluster were interpreted by local as having a or diagnosis of In addition, of the biopsies in each cluster were interpreted as having CNI was more for Cluster pathologists’ or diagnoses for biopsies in the 6 biopsies were not of these, had local diagnosis of chronic allograft calcineurin inhibitor 1 = 2 = = = 5 = 6 = toxicity biopsies were not of these, had local diagnosis of chronic allograft calcineurin inhibitor in a Of note, in each recipients were in Cluster 1, the with the postbiopsy had the biopsies C4d and the of recipients with from transplant to biopsy by with Cluster 2 being the and 5 and 6 the for time from transplant to biopsy, postbiopsy graft survival differences between clusters 5 = Serum creatinine at the time of biopsy also by cluster at biopsy for each of six 1 = 2 = = = 5 = 6 = from are ± donor specific urine to creatinine ± ± < Cluster ± ± ± < Cluster ± < Cluster positive < Cluster < Cluster < Cluster positive < Cluster < Cluster < Cluster < Cluster < Cluster are ± donor specific urine to creatinine < Cluster in a most kidney transplant recipients with deterioration of graft function not allograft biopsy, or have a biopsy late in show of with no dysfunction is to ‘chronic ‘CAN’ or These nonspecific provide no into potential pathophysiologic and as a may to therapeutic The Banff on allograft pathology (a) elimination of the ‘CAN’ as a or diagnosis; (b) use of specific and the of ‘interstitial fibrosis and tubular atrophy, no evidence of specific etiology’ (IF/TA) for biopsies in which no specific diagnosis could be (1Solez K Colvin RB Racusen LC et al.Banff ’05 meeting report: Differential diagnosis of chronic allograft injury and elimination of chronic allograft nephropathy (‘CAN’).Am J Transplant. 2007; 7: 518-526Crossref PubMed Scopus (934) Google Scholar). a PubMed of the 2007 of to describe In a analysis of our we in patients with late new-onset graft dysfunction, the local pathologists’ diagnosis of CAN no was of no prognostic (3Gourishankar S, Leduc R, Connett J et al. Pathological and clinical characterization of the ‘troubled transplant’: Data from the DeKAF study. Am J Transplant, in press.Google Scholar). The goal of the DeKAF study is to identify subgroups with individual within the nonspecific noted meaningful subgroups are it be possible to pathogenesis and develop intervention trials for For our analyses, we excluded biopsies with specific diagnoses (e.g. recurrent for which trials could be were Of the cohort with new-onset late graft dysfunction, had tubular atrophy > and the had fibrosis > with the nonspecific diagnoses of or in these same our cluster analysis, based on histologic subgroups associated with outcomes. these define a of late allograft that meaningful and clinical criteria which development of clinical trials of to or deterioration of of diagnoses chronic or CNI toxicity into entities also more of This is the first that recipients with new-onset, late graft dysfunction can be grouped into entities with the local pathologists’ diagnoses (3Gourishankar S, Leduc R, Connett J et al. Pathological and clinical characterization of the ‘troubled transplant’: Data from the DeKAF study. Am J Transplant, in press.Google Scholar), the clusters that from our analysis not only in also in outcomes. that the local are of our study, it is likely that future local be by the of the study, with between local and central pathology biopsy there is a to these preliminary observations at this before local as a of the DeKAF study. is possible to describe the (and on the of the individual clusters For example, Cluster 1 has no evidence of inflammation and has fibrosis and tubular atrophy Banff and Cluster 1 is associated with graft survival and not intervention Cluster in to fibrosis and atrophy, has inflammation by Banff i and and more allograft be most in these 1 and 2 are both large and have been in our analyses, they have the variables that are included in the clustering analysis not The cluster that has been is Cluster 6 which consists of recipients whose biopsies show a of fibrosis and tubular atrophy Banff and and evidence of Banff and Recipients in this cluster have had graft survival differentiate to of late biopsies would et capillary in An study on renal J Transplant. PubMed Scopus Google M et occur early in analysis of J Transplant. 2007; 7: PubMed Scopus Google Scholar). of our biopsies were C4d positive and of recipients had This is reported in most K et and antibodies by as for chronic of renal 2009; PubMed Scopus Google J antibodies and chronic to 2008; PubMed Scopus Google Scholar). our patients have had and our cohort not with allograft et al. have reported that of recipients for dysfunction days and had B et antibody at the time of kidney transplant biopsy with pathology and late graft J Transplant. 2009; 9: Full Text Full Text PDF PubMed Scopus Google Scholar). is also that the presence of and the of positive C4d not to be associated with a histologic in the cluster analysis that the that most cases of late graft dysfunction are to may be the of in these patients it difficult to determine are pathogenic or follow-up and more analysis may the of and of antibody into the clusters may provide our cluster have been based on Banff scores the score of information available at the time of development of graft dysfunction, C4d antibody characteristics of the and clinical as donor and and of Our is that this both the of clusters and the of clusters with of biopsies have that recipients with fibrosis and no inflammation to our cluster have long-term those with both fibrosis and inflammation have et subsequent in kidney allograft function from early J Transplant. PubMed Scopus Google Scholar, M M et associated with chronic allograft nephropathy in biopsies as a for late graft J Transplant. PubMed Scopus Google Scholar, S et al.The of on the of chronic allograft Am PubMed Scopus Google Scholar). Our patients in that the biopsies were done for new-onset, late allograft dysfunction and our clusters in fibrosis 1, 2 or and histologic As noted in clusters with fibrosis inflammation and in in addition, clusters with fibrosis and no inflammation 1 and in outcome. There are to this study. be to these clusters and in The that Banff and the data used in the cluster are may be to degrees of with variables for scores may In addition, our follow-up is and the of graft in each cluster is follow-up may differentiate outcomes among clusters. is the of donor kidney et al. have noted the of donor biopsy in long-term follow-up in and biopsy of renal PubMed Scopus Google Scholar), and we have noted that donor pathology can be as chronic or chronic allograft nephropathy in recipients et renal allograft in the first 6 months PubMed Scopus Google Scholar). clustering analyses, with of variables as donor and donor may phenotypes. biopsy by local treatments may have outcomes. As enrollment we be to study whether or not any specific had a our goal was not to show that these clusters have our goal is to differentiate subgroups. may be that 2 subgroups have the same current would be for 2 intervention trials. Our study is based on observations at a in time and on a biopsy in each be to based on a biopsy, a recipient can be characterized into a specific Recipients may subsequently develop causes of graft dysfunction (e.g. recurrent disease). Similarly, that time from transplant to biopsy by it is possible that we are of a in the time from transplant to biopsy was similar for 1 to and for 5 and is that patients were for graft it is possible that we have not included patients with histologic without long-term More it is also possible that we to early in patients et capillary in An study on renal J Transplant. PubMed Scopus Google Scholar, et subsequent in kidney allograft function from early J Transplant. PubMed Scopus Google Scholar, et in renal and J Transplant. 2008; 8: PubMed Scopus Google Scholar). to at a specific in and our goal in DeKAF is to determine if we can subgroups new-onset, late graft dysfunction is first that intervention trials could be developed. The study is in most centers a late biopsy is not done the creatinine or proteinuria our they suggest that it is possible to individual clusters associated with clinical outcomes. As the of recipients and of follow-up we clusters with distinct and of outcomes. This the of intervention trials with reproducible rates of outcomes and sample to clinical and into the current based we that these to of specific pathophysiologic that late kidney allograft We would to our local are a in this study. we for in of the The information is available for this with are not for the or of any by the be to the for the