Park, Walter

ABSTRACT Predicting delayed graft function (DGF) relies on donor, recipient, and perioperative factors. Despite the growing recognition that DGF duration strongly influences patient outcomes, no models currently address or predict its impact—highlighting an important gap in current research and clinical practice. This study aimed to develop an ensemble‐based machine learning model using perioperative data to predict DGF occurrence and duration. The gradient‐boosted decision trees model was trained and validated on 2725 patients, with k‐fold cross‐validation in an external cohort of 284 patients. Model performance was evaluated based on accuracy, ROC‐AUC, and other metrics using R and Python libraries. The binary DGF prediction model achieved a ROC‐AUC of 0.77, while the DGF duration classification model had an accuracy of 79.2%. DGF duration AUC values by time interval were 0.76 for 0–1 weeks, 0.81 for >1–2 weeks, 0.80 for >2–3 weeks, and 0.87 for >3 weeks; the overall macro‐averaged AUC was 0.81. The mean Brier score for multi‐class predictions was 0.14. External validation showed a 78% accuracy for DGF duration prediction. Acute kidney injury (AKI) and donor donation after circulatory death (DCD) status were key DGF predictors. The use of gradient‐boosted decision trees (GBDT) improves the prediction of both the likelihood and duration of DGF, addressing a current gap in kidney transplant patient care. By facilitating personalized transplant care, this model supports more effective perioperative planning and timely interventions, which may contribute to better patient outcomes.

ABSTRACTIntroductionThe incidence of mortality late in the pandemic, particularly after widespread vaccine availability, is not well understood. Herein, we elucidate the effect of this impact of the COVID pandemic as well as risk factors for mortality during it.MethodsThe primary end point was death with a functioning graft with secondary endpoints of mortality rates in subgroups and at different time intervals during the pandemic.ResultsDespite widespread vaccination, mortality rates for kidney transplant (KTx) recipients almost doubled during the COVID‐19 era (6.40 deaths per 100 person years vs. 3.54 pre‐COVID). Mortality increased in all racial/ethnic groups but increased more in Native Americans, Hispanics, and African Americans compared to non‐Hispanic Caucasians. The highest mortality rate occurred during the Delta and Omicron time frames. In contrast to the general population, mortality was more evenly spread across age groups in KTx recipients.ConclusionsMortality rates during the COVID‐19 era were extremely high, more than doubling in some racial/ethnic groups. We conclude that the KTx population was a particularly vulnerable group during the COVID‐19 era and suggests the need for further research into the management of variants in the future.

Key Points Morphometry allows for a more prognostic multidimensional quantification of interstitial fibrosis and tubular atrophy (IFTA) in kidneys than does visual inspection. The density of IFTA foci is determined by dividing the number of contiguous IFTA patches in the kidney cortex by the area of cortex. Higher density of IFTA foci significantly predicted renal allograft failure beyond %IFTA and other biopsy and clinical characteristics. Background Contraction of interstitial fibrosis and tubular atrophy (IFTA) may cause %IFTA to under-represent the severity of nephron loss. Higher density of IFTA foci is an important predictor of progressive CKD in native kidneys independent of %IFTA. Methods We studied kidney transplant recipients transplanted between 2000 and 2013 who had a 5-year surveillance kidney biopsy and subsequent follow-up. Banff ci score (interstitial fibrosis) was obtained from the pathology reports. After digitizing the biopsies, we traced cortex area and each distinct IFTA focus on a single trichrome-stained section. Percent IFTA area and IFTA foci density (count of IFTA foci/cortex area) were calculated. Cox models assessed the risk of death-censored graft failure after the 5-year biopsy with Banff ci score, morphometric %IFTA, and IFTA foci density. Results There were 58 death-censored allograft failures among 835 kidney recipients during the 5 years of follow-up. Biopsies from grafts that failed had higher mean Banff ci score (1.5 versus 0.7, P < 0.0001), %IFTA (22.6% versus 7.0%, P < 0.0001), and IFTA foci density (1.3/mm 2 versus 0.4/mm 2 , P < 0.0001). After adjusting for other Banff scores or clinical variables, Banff ci did not correlate with allograft failure, but both higher %IFTA (hazard ratio = 1.56, P < 0.0001) and higher IFTA foci density (hazard ratio = 2.34, P < 0.0001) did. All but four allograft failures by 10 years had biopsies in the top quartile of either %IFTA or IFTA foci density at 5 years. A model using just these two morphometric measures without clinical characteristics resulted in a c-statistic of 0.891 with respect to allograft failure. Conclusions Morphometric characterization of IFTA foci density is a strong predictor of death-censored allograft failure not captured in current Banff classification for grading of kidney fibrosis.

AbstractBackgroundThe aim of this study was to correlate peripheral blood gene expression profile (GEP) results during the first post‐transplant year with outcomes after kidney transplantation.MethodsWe conducted a prospective, multicenter observational study of obtaining peripheral blood at five timepoints during the first post‐transplant year to perform a GEP assay. The cohort was stratified based on the pattern of the peripheral blood GEP results: Tx‐all GEP results normal, 1 Not‐TX had one GEP result abnormal and >1 Not‐TX two or more abnormal GEP results. We correlated the GEP results with outcomes after transplantation.ResultsWe enrolled 240 kidney transplant recipients. The cohort was stratified into the three groups: TX n = 117 (47%), 1 Not‐TX n = 59 (25%) and >1 Not‐TX n = 64 (27%). Compared to the TX group, the >1 Not‐TX group had lower eGFR (p < .001) and more chronic changes on 1‐year surveillance biopsy (p = .007). Death censored graft survival showed inferior graft survival in the >1 Not‐TX group (p < .001) but not in the 1 Not‐TX group. All graft losses in the >1 Not‐TX group occurred after 1‐year post‐transplant.ConclusionsWe conclude that a pattern of persistently Not‐TX GEP assay correlates with inferior graft survival.

AbstractBackgroundComputer‐assisted scoring is gaining prominence in the evaluation of renal histology; however, much of the focus has been on identifying larger objects such as glomeruli. Total inflammation impacts graft outcome, and its quantification requires tools to identify objects at the cellular level or smaller. The goal of the current study was to use CD45 stained slides coupled with image analysis tools to quantify the amount of non‐glomerular inflammation within the cortex.MethodsSixty renal transplant whole slide images were used for digital image analysis. Multiple thresholding methods using pixel intensity and object size were used to identify inflammation in the cortex. Additionally, convolutional neural networks were used to separate glomeruli from other objects in the cortex. This combined measure of inflammation was then correlated with rescored Banff total inflammation classification and outcomes.ResultsIdentification of glomeruli on biopsies had high fidelity (mean pixelwise dice coefficient of .858). Continuous total inflammation scores correlated well with Banff rescoring (maximum Pearson correlation .824). A separate set of thresholds resulted in a significant correlation with alloimmune graft loss.ConclusionsAutomated scoring of inflammation showed a high correlation with Banff scoring. Digital image analysis provides a powerful tool for analysis of renal pathology, not only because it is reproducible and can be automated, but also because it provides much more granular data for studies.

Background Obesity is increasingly common in kidney transplant candidates and may limit access to transplantation. Obesity and diabetes are associated with a high risk for post-transplant complications. The best approach to weight loss to facilitate active transplant listing is unknown, but bariatric surgery is rarely considered due to patient- and physician-related apprehension, among other factors. Methods We aimed to determine the magnitude of weight loss, listing, and transplant rates in 28 candidates with a mean BMI of 44.4±4.6 kg/m 2 and diabetes treated conservatively for 1 year post weight-loss consultations (group 1). Additionally, we evaluated 15 patients (group 2) who met the inclusion criteria but received bariatric intervention within the same time frame. All patients completed a multidisciplinary weight management consultation with at least 1 year of follow-up. Results In the conservatively managed group (group 1), the mean weight at the time of initial consultation was 126.5±18.5 kg, and the mean BMI was 44.4±4.6 kg/m 2 . At 1 year post weight-loss consultation, the mean weight decreased by 4.4±8.2 kg to 122.9±17 kg, and the mean BMI was 43±4.8 kg/m 2 , with a total mean body weight decrease of 3% ( P =0.01). Eighteen patients (64%) did not progress to become candidates for active listing/transplantation during the follow-up time of 4±2.9 years, with 15 (54%) subsequently developing renal failure/diabetes-related comorbidities prohibitive for transplantation. In contrast, mean total body weight decreased by 19% at 6 months post bariatric surgery, and the mean BMI was 34.2±4 and 32.5±3.7 kg/m 2 at 6 and 12 months, respectively. Bariatric surgery was strongly associated with subsequent kidney transplantation (HR=8.39 [95% CI 1.71 to 41.19]; P =0.009). Conclusions A conservative weight-loss approach involving multidisciplinary consultation was ineffective in most kidney transplant candidates with diabetes, suggesting that a more proactive approach is needed.

AbstractHistologic findings on 1‐year biopsies such as inflammation with fibrosis and transplant glomerulopathy predict renal allograft loss by 5 years. However, almost half of the patients with graft loss have a 1‐year biopsy that is either normal or has only interstitial fibrosis. The goal of this study was to determine if there was a gene expression profile in these relatively normal 1‐year biopsies that predicted subsequent decline in renal function. Using transcriptome microarrays we measured intragraft mRNA levels in a retrospective Discovery cohort (170 patients with a normal/minimal fibrosis 1‐year biopsy, 54 with progressive decline in function/graft loss and 116 with stable function) and developed a nested 10‐fold cross‐validated gene classifier that predicted progressive decline in renal function (positive predictive value = 38 ± 34%%; negative predictive value = 73 ± 30%, c‐statistic = .59). In a prospective, multicenter Validation cohort (270 patients with Normal/Interstitial Fibrosis [IF]), the classifier had a 20% positive predictive value, 85% negative predictive value and .58 c‐statistic. Importantly, the majority of patients with graft loss in the prospective study had 1‐year biopsies scored as Normal or IF. We conclude predicting graft loss in many renal allograft recipients (i.e., those with a relatively normal 1‐year biopsy and eGFR > 40) remains difficult.

The liver is an immunologically active organ with a tolerogenic microenvironment at a quiescent state. The immunoregulatory properties of the liver appear to be retained after transplantation because liver allografts can reduce alloresponses against other organs that are simultaneously transplanted. Mechanisms of this phenomenon remain unknown. Given the known immunomodulatory properties of mesenchymal stromal cells (MSCs), we hypothesized that liver mesenchymal stromal cells (L‐MSCs) are superior immunomodulators and contribute to liver‐mediated tolerance. L‐MSCs, generated from human liver allograft biopsies, were compared with adipose mesenchymal stromal cells (A‐MSCs) and bone marrow mesenchymal stromal cells (BM‐MSCs). Trilineage differentiation of L‐MSCs was confirmed by immunohistochemistry. Comparative phenotypic analyses were done by flow cytometry and transcriptome analyses by RNA sequencing in unaltered cell cultures. The in vitro functional analyses were performed using alloreactive T cell proliferation assays. The transcriptome analysis showed that the L‐MSCs are different than the A‐MSCs and BM‐MSCs, with significant enrichment of genes and gene sets associated with immunoregulation. Compared with the others, L‐MSCs were found to express higher cell surface levels of several select immunomodulatory molecules. L‐MSCs (versus A‐MSCs/BM‐MSCs) inhibited alloreactive T cell proliferation (22.7% versus 56.4%/58.7%, respectively; P < 0.05) and reduced the frequency of interferon ɤ–producing T cells better than other MSCs (52.8% versus 94.4%/155.4%; P < 0.05). The antiproliferative impact of L‐MSCs was not dependent on cell‐to‐cell contact, could be reversed incompletely by blocking programmed death ligand 1, and required a higher concentration of the competitive inhibitor of indoleamine 2,3‐dioxygenase for complete reversal. In conclusion, L‐MSCs appear to be uniquely well‐equipped immunomodulatory cells, and they are more potent than A‐MSCs and BM‐MSCs in that capacity, which suggests that they may contribute to liver‐induced systemic tolerance.

Background. Glucagon-like peptide-1 receptor agonists (GLP1RA) have been shown to improve glucose control and diabetes-related comorbidities in patients without solid organ transplants. The effectiveness, safety, and tolerability of GLP1RA after kidney transplantation have not been adequately studied. Methods. We retrospectively reviewed data on kidney transplant recipients performed in our institution, who were initiated on GLP1RA either for the treatment of type 2 diabetes diagnosed before transplantation or posttransplant diabetes. We analyzed efficacy, safety, and the effect on kidney allograft function. Results. Seventeen kidney transplant recipients were initiated on GLP1RA therapy, 14 of which remained on the medication for at least 12 months. The use of GLP1RA had no significant impact on weight loss, but was associated with a significant reduction in the total daily insulin dose, from the median of 63 [interquartile range 43-113] IU to 44 [interquartile range 25-88] and reduction in the risk of hypoglycemia in patients who were on therapy for at least approximately 12 months. Kidney function remained stable and none of the recipients experienced acute rejection. Tacrolimus dose was not significantly changed. Five patients (29%) discontinued GLP1RA therapy—4 due to side effects and 1 due to uncontrolled hyperglycemia. Conclusions. GLP1RA may be a relatively safe and effective treatment for kidney transplant recipients with type 2 diabetes that allows for a reduction in insulin requirements. More studies are needed to determine whether the use of these agents will translate into an improvement in allograft and patient survival.