We read with great interest the recently published overview of Kwun and Knechtle (1) and the accompanying commentary of Terasaki (2). We are in strong agreement with both their opinions and share their frustration with the limited options available to delay the development of chronic rejection/graft sclerosis (CR/GS). As stated by Terasaki (2), there is an unquestionable evidence for a distinctive form of renal (cardiac and pancreatic) graft pathology associated with circulating donor-specific antibodies (DSA+), which relentlessly leads to CR/GS (3–5). Accordingly, we are observing active chronic antibody-mediated rejection (AMR) in more than 25% of our renal patients biopsied for graft dysfunction after the first year of posttransplantation. The main features are transplant glomerulopathy, CD4 positivity, multilayering of basal lamina in peritubular capillaries, parenchymal scarring, and DSA+ (3). Needles to say, subclinical AMR is also identified in some protocol biopsies performed at earlier times. Although routine evaluation of DSA and CD4 staining have helped to delineate the histologic features of AMR, the morphologic diagnosis continues to be problematic, in part because of the inconsistent staining for CD4 which similar to DSA may fluctuate overtime, spontaneously or in response to treatment interventions. Morphologic characterization of chronic AMR is also complicated by bouts of “acute” injury that occur on the background of “chronic” sclerosing changes. In fact, advanced cases of chronic AMR have extensive scarring and remodeling, including glomerular global and segmental glomerular sclerosis that makes an accurate morphologic diagnosis impossible in nephrectomies from these failed grafts. CD4 capillary staining may remain and gives a clue for the cause of graft failure. While acknowledging the importance of DSA in the development of CR/GS, it is important to emphasize that a significant proportion of stable and failing allografts lack features of AMR. In the past years, we have prospectively correlated morphology including electron microscopy with CD4 staining and DSA studies in single or serial biopsies performed 1 to 17 years posttransplantation (>250 patients). With this approach, we have found that at least 50% of patients lack any feature of chronic AMR but may have CR/GS that can be linked to other factors including plain ischemic injury, de novo or recurrent diabetic glomerular sclerosis, de novo or recurrent glomerulopathy often associated with postadaptive segmental sclerosis. The heterogeneous phenotype of CR/GS is, thus, most consistent with a variety of operating mechanisms, which are only beginning to be deciphered. The lack of “specific” pathologic findings in a proportion of cases may be discouraging to some, but in fact, this evidence strengthens the diagnostic significance of microvascular inflammatory and reparative responses typically associated with chronic DSA/AMR. In this area, diagnostic advances were only possible with the integration of morphologic, immunohistochemical, and serologic data. A similar comprehensive approach will be necessary to understand other processes leading to CR/GS. To achieve an accurate determination of the cause of graft failure, there should be concerted efforts to characterize all morphologic confounders (i.e., “idiopathic” thrombotic microangiopathy, hepatitis C-related glomerulonephritis), and all available data collected along the lifespan of the graft taken into account (i.e., infections such as polyomavirus BK nephritis). Systematic collection and analysis of data also incorporating a stratification of risk factors inherent to each specific case (i.e., retransplantation, diabetes mellitus, extended criteria donor) will assure successful incorporation of more advanced technologies such as gene profiling of biopsy samples (6). Although pathologic analysis continues to be crucial, isolated biopsy findings only represent snapshots of the eminently complex and dynamic long-term interactions between graft and host. Integration of morphology with contemporary and previous, clinical and laboratory data will help in understanding CR/GS. Cinthia B. Drachenberg John C. Papadimitriou Department of Pathology University of Maryland School of Medicine Baltimore, MD