L. G. Hidalgo

Understanding the Causes of Kidney Transplant Failure: The Dominant Role of Antibody-Mediated Rejection and Nonadherence

We prospectively studied kidney transplants that progressed to failure after a biopsy for clinical indications, aiming to assign a cause to every failure. We followed 315 allograft recipients who underwent indication biopsies at 6 days to 32 years posttransplant. Sixty kidneys progressed to failure in the follow‐up period (median 31.4 months). Failure was rare after T‐cell–mediated rejection and acute kidney injury and common after antibody‐mediated rejection or glomerulonephritis. We developed rules for using biopsy diagnoses, HLA antibody and clinical data to explain each failure. Excluding four with missing information, 56 failures were attributed to four causes: rejection 36 (64%), glomerulonephritis 10 (18%), polyoma virus nephropathy 4 (7%) and intercurrent events 6 (11%). Every rejection loss had evidence of antibody‐mediated rejection by the time of failure. Among rejection losses, 17 of 36 (47%) had been independently identified as nonadherent by attending clinicians. Nonadherence was more frequent in patients who progressed to failure (32%) versus those who survived (3%). Pure T‐cell–mediated rejection, acute kidney injury, drug toxicity and unexplained progressive fibrosis were not causes of loss. This prospective cohort indicates that many actual failures after indication biopsies manifest phenotypic features of antibody‐mediated or mixed rejection and also underscores the major role of nonadherence.

Antibody‐Mediated Microcirculation Injury Is the Major Cause of Late Kidney Transplant Failure

We studied the phenotype of late kidney graft failure in a prospective study of unselected kidney transplant biopsies taken for clinical indications. We analyzed histopathology, HLA antibodies and death‐censored graft survival in 234 consecutive biopsies from 173 patients, taken 6 days to 31 years posttransplant. Patients with late biopsies (>1 year) frequently displayed donor‐specific HLA antibody (particularly class II) and microcirculation changes, including glomerulitis, glomerulopathy, capillaritis, capillary multilayering and C4d staining. Grafts biopsied early rarely failed (1/68), whereas grafts biopsied late often progressed to failure (27/105) within 3 years. T‐cell‐mediated rejection and its lesions were not associated with an increased risk of failure after biopsy. In multivariable analysis, graft failure correlated with microcirculation inflammation and scarring, but C4d staining was not significant. When microcirculation changes and HLA antibody were used to define antibody‐mediated rejection, 17/27 (63%) of late kidney failures after biopsy were attributable to antibody‐mediated rejection, but many were C4d negative and missed by current diagnostic criteria. Glomerulonephritis accounted for 6/27 late losses, whereas T‐cell‐mediated rejection, drug toxicity and unexplained scarring were uncommon. The major cause of late kidney transplant failure is antibody‐mediated microcirculation injury, but detection of this phenotype requires new diagnostic criteria.

De Novo Donor‐Specific Antibody at the Time of Kidney Transplant Biopsy Associates with Microvascular Pathology and Late Graft Failure

We studied whether de novo donor‐specific antibodies (DSA) in sera from patients undergoing kidney transplant biopsies associate with specific histologic lesions in the biopsy and prognosis. DSA were assessed in 145 patients at the time of biopsy between 7 days to 31 years posttransplant. DSA was detected in 54 patients (37%), of which 32 represented de novo DSA. De novo DSA was more frequent in patients having late biopsies (34%) versus early biopsies (4%), and was usually either against class II alone or class I and II but rarely against class I alone. Microcirculation inflammation (glomerulitis, capillaritis) and damage (glomuerulopathy, capillary basement membrane multilayering), and C4d staining were associated with de novo DSA. However, the degree of scarring, arterial fibrosis and tubulo‐interstitial inflammation did not correlate with the presence of de novo DSA. De novo DSA correlated with reduced graft survival after the biopsy. Thus, de novo DSA at the time of a late biopsy for clinical indication is primarily against class II, and associates with microcirculation changes in the biopsy and subsequent graft failure. We propose careful assessment of de novo DSA, particularly against class II, be performed in all late kidney transplant biopsies.

NK Cell Transcripts and NK Cells in Kidney Biopsies from Patients with Donor-Specific Antibodies: Evidence for NK Cell Involvement in Antibody-Mediated Rejection

To explore the mechanisms of antibody‐mediated rejection (ABMR) in kidney transplants, we studied the transcripts expressed in clinically indicated biopsies from patients with donor‐specific antibody (DSA). Comparison of biopsies from DSA‐positive versus DSA‐negative patients revealed 132 differentially expressed transcripts: all were associated with class II DSA but none with class I DSA. Many transcripts were expressed in DSA‐positive ABMR but were also expressed in T‐cell‐mediated rejection (TCMR), reflecting shared molecular features. Removal of shared transcripts created 23 DSA selective transcripts (DSASTs). Some DSASTs (6/23) showed selective high expression in NK cells, whereas others (8/23) were expressed in endothelium or in endothelium plus other cell types (7/23). Of 145 biopsies ranked by DSAST expression, the 25 with highest DSAST expression primarily consisted of ABMR (22/25, 88%), either C4d‐positive or C4d‐negative. By immunostaining, CD56+ and CD68+ cells in peritubular capillaries, but not CD3+ cells, were increased in ABMR compared to TCMR, compatible with a role for NK cells, as well as macrophages, as effectors in endothelial injury during ABMR. Thus, the strategy of using DSASTs in the biopsy to identify mechanism‐related transcripts in biopsies from patients with clinical phenotypes indicates the selective involvement of NK cells in ABMR.

A New Diagnostic Algorithm for Antibody‐Mediated Microcirculation Inflammation in Kidney Transplants

We studied the significance of microcirculation inflammation in kidney transplants, including 329 indication biopsies from 251 renal allograft recipients, who were mostly nonpresensitized (crossmatch negative). Glomerulitis (g) and peritubular capillaritis (ptc) were often associated with antibody‐mediated rejection (65% and 75%, respectively), but were also found in other diseases in the absence of donor‐specific antibody (DSA): T‐cell‐mediated rejection (ptc, g), glomerulonephritis (g) and acute tubular necrosis (ptc). To develop rules for reducing the nonspecificity of microcirculation inflammation and defining the best grading thresholds associated with DSA, we built and validated a decision tree to predict DSA. The decision tree revealed that g + ptc sum (addition of g‐score plus ptc‐score) was the best predictor of DSA, followed by time posttransplant, then C4d, which had a small role. Late biopsies with g + ptc > 0 showed higher frequency of DSA compared to early biopsies with g + ptc > 0 (79% vs. 27%). Microcirculation inflammation in early biopsies was often false positive (antibody‐independent). The decision tree predicted DSA with higher sensitivity and accuracy than C4d staining. Microcirculation inflammation sum score predicted graft failure independently of time, C4d and transplant glomerulopathy. Thus any degree of microcirculation inflammation in late kidney transplant biopsies strongly indicates presence of DSA and predicts progression to graft failure.

Molecular Diagnosis of Antibody‐Mediated Rejection in Human Kidney Transplants

Antibody‐mediated rejection is the major cause of kidney transplant failure, but the histology‐based diagnostic system misses most cases due to its requirement for C4d positivity. We hypothesized that gene expression data could be used to test biopsies for the presence of antibody‐mediated rejection. To develop a molecular test, we prospectively assigned diagnoses, including C4d‐negative antibody‐mediated rejection, to 403 indication biopsies from 315 patients, based on histology (microcirculation lesions) and donor‐specific HLA antibody. We then used microarray data to develop classifiers that assigned antibody‐mediated rejection scores to each biopsy. The transcripts distinguishing antibody‐mediated rejection from other conditions were mostly expressed in endothelial cells or NK cells, or were IFNG‐inducible. The scores correlated with the presence of microcirculation lesions and donor‐specific antibody. Of 45 biopsies with scores >0.5, 39 had been diagnosed as antibody‐mediated rejection on the basis of histology and donor‐specific antibody. High scores were also associated with unanimity among pathologists that antibody‐mediated rejection was present. The molecular score also strongly predicted future graft loss in Cox regression analysis. We conclude that microarray assessment of gene expression can assign a probability of ABMR to transplant biopsies without knowledge of HLA antibody status, histology, or C4d staining, and predicts future failure.

Molecular Diagnosis of T Cell-Mediated Rejection in Human Kidney Transplant Biopsies

Histologic diagnosis of T cell‐mediated rejection is flawed by subjective assessments, nonspecific lesions and arbitrary rules. This study developed a molecular test for T cell‐mediated rejection. We used microarray results from 403 kidney transplant biopsies to derive a classifier assigning T cell‐mediated rejection scores to all biopsies, and compared these with histologic assessments. The score correlated with histologic lesions of T cell‐mediated rejection (infiltrate, tubulitis). The accuracy of the classifier for the histology diagnoses was 89%. Very high and low molecular scores corresponded with unanimity among three pathologists on the presence or absence of T cell‐mediated rejection, respectively. The molecular score had low sensitivity (50%) and positive predictive value (62%) for the histology diagnoses. However, histology showed similar disagreement between pathologists—only 45–56% sensitivity of one pathologist with diagnoses of T cell‐mediated rejection by another. Discrepancies between molecular scores and histology were mostly when histology was ambiguous (“borderline”) or unreliable, e.g. in cases with scarring or inflammation induced by tissue injury. Vasculitis (isolated v‐lesion TCMR) was particularly discrepant, with most cases exhibiting low TCMR scores. We propose new rules to integrate molecular tests and histology into a precision diagnostic system that can reduce errors, ambiguity and interpathologist disagreement.

Changes in the Transcriptome in Allograft Rejection: IFN-γ-Induced Transcripts in Mouse Kidney Allografts

We used Affymetrix Microarrays to define interferon‐γ (IFN‐γ)‐dependent, rejection‐induced transcripts (GRITs) in mouse kidney allografts. The algorithm included inducibility by recombinant IFN‐γ in kidneys of three normal mouse strains, increase in kidney allografts in three strain combinations and less induction in IFN‐γ‐deficient allografts. We identified 40 transcripts, which were highly IFN‐γ inducible (e.g. Cxcl9, ubiquitin D, MHC), and 168 less sensitive to IFN‐γ in normal kidney. In allografts, expression of GRITs was intense and consistent at all time points (day 3 through 42). These transcripts were partially dependent on donor IFN‐γ receptors (IFN‐γrs): receptor‐deficient allografts manifested up to 76% less expression, but some transcripts were highly dependent (ubiquitin D) and others relatively independent (Cxcl9). Kidneys of hosts rejecting allografts showed expression similar to that observed with IFN‐γ injections. Many GRITs showed transient IFN‐γ‐dependent increase in isografts, peaking at day 4–5. GRITs were increased in heart allografts, indicating them as generalized feature of alloresponse. Thus, expression of rejection‐induced transcripts is robust and consistent in allografts, reflecting the IFN‐γ produced by the alloresponse locally and systemically, acting via host and donor IFN‐γr, as well as local IFN‐γ production induced by post‐operative stress.

An integrated view of molecular changes, histopathology and outcomes in kidney transplants.

Data‐driven approaches to deteriorating kidney transplants, incorporating histologic, molecular and HLA antibody findings, have created a new understanding of transplant pathology and why transplants fail. Transplant dysfunction is best understood in terms of three elements: diseases, the active injury–repair response and the cumulative burden of injury. Progression to failure is mainly attributable to antibody‐mediated rejection, nonadherence and glomerular disease. Antibody‐mediated rejection usually develops late due to de novo HLA antibodies, particularly anti‐class II, and is often C4d negative. Pure treated T cell‐mediated rejection does not predispose to graft loss because it responds well, even with endothelialitis, but it may indicate nonadherence. The cumulative burden of injury results in atrophy‐fibrosis (nephron loss), arterial fibrous intimal thickening and arteriolar hyalinosis, but these are not progressive without ongoing disease/injury, and do not explain progression. Calcineurin inhibitor toxicity has been overestimated because burden‐of‐injury lesions invite this default diagnosis when diseases such as antibody‐mediated rejection are missed. Disease/injury triggers a stereotyped active injury–repair response, including de‐differentiation, cell cycling and apoptosis. The active injury–repair response is the strongest correlate of organ function and future progression to failure, but should always prompt a search for the initiating injury or disease.

The molecular phenotype of kidney transplants.

Microarray studies of kidney transplant biopsies provide an opportunity to define the molecular phenotype. To facilitate this process, we used experimental systems to annotate transcripts as members of pathogenesis‐based transcript sets (PBTs) representing biological processes in injured or diseased tissue. Applying this annotation to microarray results revealed that changes in single molecules and PBTs reflected a large‐scale coordinate disturbance, stereotyped across various diseases and injuries, without absolute specificity of individual molecules or PBTs for rejection. Nevertheless, expression of molecules and PBTs was quantitatively specific: IFNG effects for rejection; T cell and macrophage transcripts for T cell‐mediated rejection; endothelial and NK transcripts for antibody‐mediated rejection. Various diseases and injuries induced the same injury–repair response, undetectable by histopathology, involving epithelium, stroma and endothelium, with increased expression of developmental, cell cycle and apoptosis genes and decreased expression of differentiated epithelial features. Transcripts reflecting this injury–repair response were the best correlates of functional disturbance and risk of future graft loss. Late biopsies with atrophy‐fibrosis, reflecting their cumulative burden of injury, displayed more transcripts for B cells, plasma cells and mast cells. Thus the molecular phenotype is best described in terms of three elements: specific diseases, including rejection; the injury–repair response and the cumulative burden of injury.