Prediction system for risk of allograft loss in patients receiving kidney transplants: international derivation and validation study

Clinical Events and Renal Function in the First Year Predict Long-Term Kidney Transplant Survival

BACKGROUND

Estimated glomerular filtration rate (eGFR) at 1 year post transplantation has been shown to be a strong predictor of long-term graft survival. However, intercurrent events (ICEs) may affect the relationship between eGFR and failure risk.

METHODS

The OPTN and USRDS databases on single-organ kidney transplant recipients from 2012 to 2016 were linked. Competing risk regressions estimated adjusted subhazard ratios (SHRs) of 12-month eGFR on long-term graft failure, considering all-cause mortality as the competing risk, for deceased donor (DD) and living donor (LD) recipients. Additional predictors included recipient, donor, and transplant characteristics. ICEs examined were acute rejection, cardiovascular events, and infections.

RESULTS

Cohorts comprised 25,131 DD recipients and 7471 LD recipients. SHRs for graft failure increased rapidly as 12-month eGFR values decreased from the reference 60 ml/min per 1.73 m². At an eGFR of 20 ml/min per 1.73 m², SHRs were 13-15 for DD recipients and 12-13 for LD recipients; at an eGFR of 30 ml/min per 1.73 m², SHRs were 5.0-5.7 and 5.0-5.5, respectively. Among first-year ICEs, acute rejection was a significant predictor of long-term graft failure in both DD (SHR=1.63, P<0.001) and LD (SHR=1.51, P=0.006) recipients; cardiovascular events were significant in DD (SHR=1.24, P<0.001), whereas non-CMV infections were significant in the LD cohort (SHR=1.32, P=0.03). Adjustment for ICEs did not significantly reduce the association of eGFR with graft failure.

CONCLUSIONS

Twelve-month eGFR is a strong predictor of long-term graft failure after accounting for clinical events occurring from discharge to 1 year. These findings may improve patient management and clinical evaluation of novel interventions.

Artificial intelligence-enabled decision support in nephrology

Kidney pathophysiology is often complex, nonlinear and heterogeneous, which limits the utility of hypothetical-deductive reasoning and linear, statistical approaches to diagnosis and treatment. Emerging evidence suggests that artificial intelligence (AI)-enabled decision support systems - which use algorithms based on learned examples - may have an important role in nephrology. Contemporary AI applications can accurately predict the onset of acute kidney injury before notable biochemical changes occur; can identify modifiable risk factors for chronic kidney disease onset and progression; can match or exceed human accuracy in recognizing renal tumours on imaging studies; and may augment prognostication and decision-making following renal transplantation. Future AI applications have the potential to make real-time, continuous recommendations for discrete actions and yield the greatest probability of achieving optimal kidney health outcomes. Realizing the clinical integration of AI applications will require cooperative, multidisciplinary commitment to ensure algorithm fairness, overcome barriers to clinical implementation, and build an AI-competent workforce. AI-enabled decision support should preserve the pre-eminence of wisdom and augment rather than replace human decision-making. By anchoring intuition with objective predictions and classifications, this approach should favour clinician intuition when it is honed by experience.

Early Estimated Glomerular Filtration Rate Trajectories After Kidney Transplant Biopsy as a Surrogate Endpoint for Graft Survival in Late Antibody-Mediated Rejection

Background

Late antibody-mediated rejection (ABMR) after kidney transplantation is a major cause of long-term allograft loss with currently no proven treatment strategy. Design for trials testing treatment for late ABMR poses a major challenge as hard clinical endpoints require large sample sizes. We performed a retrospective cohort study applying commonly used selection criteria to evaluate the slope of the estimated glomerular filtration rate (eGFR) within an early and short timeframe after biopsy as a surrogate of future allograft loss for clinical trials addressing late ABMR.

Methods

Study subjects were identified upon screening of the Vienna transplant biopsy database. Main inclusion criteria were (i) a solitary kidney transplant between 2000 and 2013, (ii) diagnosis of ABMR according to the Banff 2015 scheme at >12 months post-transplantation, (iii) age 15-75 years at ABMR diagnosis, (iv) an eGFR > 25 mL/min/1.73 m2 at ABMR diagnosis, and (v) a follow-up for at least 36 months after ABMR diagnosis. The primary outcome variable was death-censored graft survival. A mixed effects model with linear splines was used for eGFR slope modeling and association of graft failure and eGFR slope was assessed applying a multivariate competing risk analysis with landmarks set at 12 and 24 months after index biopsy.

Results

A total of 70 allografts from 68 patients were included. An eGFR loss of 1 ml/min/1.73 m2 per year significantly increased the risk for allograft failure, when eGFR slopes were modeled over 12 months [HR 1.1 (95% CI: 1.01-1.3), p = 0.020] or over 24 months [HR 1.3 (95% CI: 1.1-1.4), p = 0.001] after diagnosis of ABMR with landmarks set at both time points. Covariables influencing graft loss in all models were histologic evidence of glomerulonephritis concurring with ABMR as well as the administration of anti-thymocyte globulin (ATG) at the time of transplantation.

Conclusion

Our study supports the use of the eGFR slope modeled for at least 12 months after biopsy-proven diagnosis of late ABMR, as a surrogate parameter for future allograft loss. The simultaneous occurrence of glomerulonephritis together with ABMR at index biopsy and the use of ATG at the time of transplantation-likely representing a confounder in pre-sensitized recipients-were strongly associated with worse transplant outcomes.

Can Gene Expression Analysis in Zero-Time Biopsies Predict Kidney Transplant Rejection?

Zero-time biopsies are taken to determine the quality of the donor organ at the time of transplantation. Histological analyses alone have so far not been able to identify parameters that allow the prediction of subsequent rejection episodes or graft survival. This study investigated whether gene expression analyses of zero-time biopsies might support this prediction. Using a well-characterized cohort of 26 zero-time biopsies from renal transplant patients that include 4 living donor (LD) and 22 deceased donor (DD) biopsies that later developed no rejection (Ctrl, n = 7), delayed graft function (DGF, n = 4), cellular (T-cell mediated rejection; TCMR, n = 8), or antibody-mediated rejection (ABMR, n = 7), we analyzed gene expression profiles for different types of subsequent renal transplant complication. To this end, RNA was isolated from formalin-fixed, paraffin-embedded (FFPE) sections and gene expression profiles were quantified. Results were correlated with transplant data and B-cell, and plasma cell infiltration was assessed by immunofluorescence microscopy. Both principal component analysis and clustering analysis of gene expression data revealed marked separation between LDs and DDs. Differential expression analysis identified 185 significant differentially expressed genes (adjusted p < 0.05). The expression of 68% of these genes significantly correlated with cold ischemia time (CIT). Furthermore, immunoglobulins were differentially expressed in zero-time biopsies from transplants later developing rejection (TCMR + ABMR) compared to non-rejected (Ctrl + DGF) transplants. In addition, immunoglobulin expression did not correlate with CIT but was increased in transplants with previous acute renal failure (ARF). In conclusion, gene expression profiles in zero-time biopsies derived from LDs are markedly different from those of DDs. Pre-transplant ARF increased immunoglobulin expression, which might be involved in triggering later rejection events. However, these findings must be confirmed in larger cohorts and the role of early immunoglobulin upregulation in zero-biopsies needs further clarification.

Weak Expression of Terminal Complement in Active Antibody-Mediated Rejection of the Kidney

Background

The role of the complement system in antibody-mediated rejection (ABMR) is insufficiently understood. We aimed to investigate the role of local and systemic complement activation in active (aABMR). We quantified complement activation markers, C3, C3d, and C5b-9 in plasma of aABMR, and acute T-cell mediated rejection (aTCMR), and non-rejection kidney transplant recipients. Intra-renal complement markers were analyzed as C4d, C3d, C5b-9, and CD59 deposition. We examined in vitro complement activation and CD59 expression on renal endothelial cells upon incubation with human leukocyte antigen antibodies.

Methods

We included 50 kidney transplant recipients, who we histopathologically classified as aABMR (n=17), aTCMR (n=18), and non-rejection patients (n=15).

Results

Complement activation in plasma did not differ across groups. C3d and C4d deposition were discriminative for aABMR diagnosis. Particularly, C3d deposition was stronger in glomerular (P&lt;0,01), and peritubular capillaries (P&lt;0,05) comparing aABMR to aTCMR rejection and non-rejection biopsies. In contrast to C3d, C5b-9 was only mildly expressed across all groups. For C5b-9, no significant difference between aABMR and non-rejection biopsies regarding peritubular and glomerular C5b-9 deposition was evident. We replicated these findings in vitro using renal endothelial cells and found complement pathway activation with C4d and C3d, but without terminal C5b-9 deposition. Complement regulator CD59 was variably present in biopsies and constitutively expressed on renal endothelial cells in vitro.

Conclusion

Our results indicate that terminal complement might only play a minor role in late aABMR, possibly indicating the need to re-evaluate the applicability of terminal complement inhibitors as treatment for aABMR.

Safety, tolerability, and efficacy of monoclonal CD38 antibody felzartamab in late antibody-mediated renal allograft rejection: study protocol for a phase 2 trial

BACKGROUND

Antibody-mediated rejection (ABMR) is a cardinal cause of renal allograft loss. This rejection type, which may occur at any time after transplantation, commonly presents as a continuum of microvascular inflammation (MVI) culminating in chronic tissue injury. While the clinical relevance of ABMR is well recognized, its treatment, particularly a long time after transplantation, has remained a big challenge. A promising strategy to counteract ABMR may be the use of CD38-directed treatment to deplete alloantibody-producing plasma cells (PC) and natural killer (NK) cells.

METHODS

This investigator-initiated trial is planned as a randomized, placebo-controlled, double-blind, parallel-group, multi-center phase 2 trial designed to assess the safety and tolerability (primary endpoint), pharmacokinetics, immunogenicity, and efficacy of the fully human CD38 monoclonal antibody felzartamab (MOR202) in late ABMR. The trial will include 20 anti-HLA donor-specific antibody (DSA)-positive renal allograft recipients diagnosed with active or chronic active ABMR ≥ 180 days post-transplantation. Subjects will be randomized 1:1 to receive felzartamab (16 mg/kg per infusion) or placebo for a period of 6 months (intravenous administration on day 0, and after 1, 2, 3, 4, 8, 12, 16, and 20 weeks). Two follow-up allograft biopsies will be performed at weeks 24 and 52. Secondary endpoints (preliminary assessment) will include morphologic and molecular rejection activity in renal biopsies, immunologic biomarkers in the blood and urine, and surrogate parameters predicting the progression to allograft failure (slope of renal function; iBOX prediction score).

DISCUSSION

Based on the hypothesis that felzartamab is able to halt the progression of ABMR via targeting antibody-producing PC and NK cells, we believe that our trial could potentially provide the first proof of concept of a new treatment in ABMR based on a prospective randomized clinical trial.

TRIAL REGISTRATION

EU Clinical Trials Register (EudraCT) 2021-000545-40 . Registered on 23 June 2021.

CLINICALTRIALS

gov NCT05021484 . Registered on 25 August 2021.

Post-transplant outcomes in recipients of living donor kidneys and intended recipients of living donor kidneys

Background

Long-term kidney transplant survival at the population level is consistently favorable, but this survival varies widely at an individual level due to both recipient and donor factors. The distinct contribution of recipient and donor factors to individual post kidney transplant outcome remains unclear. Comparing outcomes in deceased donor (DD) recipients with potential but non-actualized living donors (DD1) to those recipients with actualized living donors (LD), and to DD recipients without potential living donors (DD0) may provide transplant candidates with more information about their own post-transplant prognosis.

Methods

We conducted an observational retrospective cohort study of kidney transplant candidates presenting to our centre for evaluation between 01/01/06 and 31/12/18, and who also received a transplant during that time. Patients were followed to 31/08/2019. Candidates were classified as DD0, DD1, or LD based on whether they had an identified living donor at the time of initial pre-transplant assessment, and if the donor actualized or not. Primary outcome was 5-year death-censored graft survival, adjusted for common pre- and post-transplant donor and recipient risk factors. Secondary outcomes analyzed included patient survival and graft function.

Results

There were 453 kidney transplant recipients (LD = 136, DD1 = 83, DD0 = 234) who received a transplant during the study period. DD0 and DD1 did not differ in key donor organ characteristics. The 5-year death censored graft survival of DD1 was similar to LD (p = 0.19). DD0 graft survival was inferior to LD (p = 0.005), but also trended inferior to DD1 (p = 0.052). By multivariate Cox regression analysis, LD demonstrated similar 5-year graft survival to DD1 (HR for graft loss 0.8 [95% CI 0.25–2.6], p = 0.72) but LD graft survival was superior to DD0 (HR 0.34 [0.16–0.72], p = 0.005). The 5-year patient survival in DD1 was similar to LD (p = 0.26) but was superior to DD0 (p = 0.01).

Conclusions

DD recipients with potential but non-actualized living donors exhibit similar mid-term graft and patient survival compared to LD recipients. Having an identified living donor at the time of pre-transplant assessment portends a favorable prognosis for the recipient.

Supplementary Information

The online version contains supplementary material available at 10.1186/s12882-022-02718-6.

Plasma neutrophil gelatinase-associated lipocalin and kidney graft outcome

Background

Plasma neutrophil gelatinase-associated lipocalin (pNGAL) has been investigated extensively in acute kidney injury. This study investigated its pathophysiological significance and utility as marker for graft failure and mortality in stable kidney transplant recipients (KTR).

Methods

Baseline pNGAL was measured in 698 KTR (58% male, age 53 ± 13 years, estimated glomerular filtration rate 52.4 ± 20.4 mL/min/1.73 m²) at median 5.4 (interquartile range 1.8–12.0) years after transplantation, enrolled in the prospective TransplantLines Food and Nutrition Biobank and Cohort Study.

Results

pNGAL concentrations were higher in males, younger patients, patients with a deceased-donor kidney and higher serum creatinine. Independent of these, pNGAL was positively associated with urinary protein excretion, systemic inflammation parameters and calcineurin inhibitor use. During median follow-up of 5.3 (4.5–6.0) years, death-censored graft failure rates were 3.9%, 7.3% and 25.0% across increasing tertiles of pNGAL (P_log-rank < 0.001). Cox-regression analyses showed no independent associations of pNGAL with mortality, but strong associations with graft failure (hazard ratio, per doubling 4.16; 95% confidence interval 3.03–5.71; P < 0.001), which remained independent of adjustment for confounders. These associations were present only in patients with pre-existent proteinuria and poor kidney function.

Conclusions

pNGAL is associated with parameters of kidney graft damage and with graft failure. The latter association is particularly present in KTR with pre-existent poor kidney function and proteinuria.

Trial Registration:  ClinicalTrials.gov NCT02811835.

Two is better than one: when to consider multiorgan transplant

PURPOSE OF REVIEW

Patients with end-stage heart failure often present with concomitant end-stage renal or end-stage liver disease requiring transplantation. There are limited data regarding the risks, benefits and long-term outcomes of heart-kidney (HKT) and heart-liver transplantation (HLT), and guidelines are mainly limited to expert consensus statements.

RECENT FINDINGS

The incidence of HKT and HLT has steadily increased in recent years with favourable outcomes. Both single-centre and large database studies have shown benefits of HKT/HLT through improved survival, freedom from dialysis and lower rates of rejection and coronary allograft vasculopathy. Current guidelines are institution dependent and controversial due to the ethical considerations surrounding multiorgan transplantation (MOT).

SUMMARY

MOT is an effective and necessary option for patients with end-stage heart and kidney/liver failure. MOT is ethically permissible, and efforts should be made to consider eligible patients as early as possible to limit morbidity and mortality. Further research is needed regarding appropriate listing criteria and long-term outcomes.

Intimal Arteritis and Microvascular Inflammation Are Associated With Inferior Kidney Graft Outcome, Regardless of Donor-Specific Antibodies

Background: The prognostic role of intimal arteritis of kidney allografts in donor-specific antibody negative (DSA–) antibody-mediated rejection (ABMR) remains unclear.

Methods: Seventy-two out of 881 patients who had undergone kidney transplantation from 2014 to 2017 exhibited intimal arteritis in biopsies performed during the first 12 months. In 26 DSA negative cases, the intimal arteritis was accompanied by tubulointerstitial inflammation as part of T cell-mediated vascular rejection (TCMRV, N = 26); intimal arteritis along with microvascular inflammation occurred in 29 DSA negative (ABMRV/DSA–) and 19 DSA positive cases (ABMRV, DSA+, N = 17). In 60 (83%) patients with intimal arteritis, the surveillance biopsies after antirejection therapy were performed. Hundred and two patients with non-vascular ABMR with DSA (ABMR/DSA+, N = 55) and without DSA (ABMR/DSA–, N = 47) served as controls. Time to transplant glomerulopathy (TG) and graft failure were the study endpoints.

Results: Transplant glomerulopathy -free survival at 36 months was 100% in TCMRV, 85% in ABMR/DSA–, 65% in ABMRV/DSA-, 54% in ABMR/DSA+ and 31% in ABMRV/DSA+ (log rank p < 0.001). Death-censored graft survival at 36 months was 98% in ABMR/DSA-, 96% in TCMRV, 86% in ABMRV/DSA–, 79% in ABMR/DSA+, and 64% in ABMRV/DSA+ group (log rank p = 0.001). In surveillance biopsies, the resolution of rejection was found in 19 (90%) TCMRV, 14 (58%) ABMRV/DSA–, and only 4 (27%) ABMRV/DSA+ patients (p = 0.006). In the multivariable model, intimal arteritis as part of ABMR represented a significant risk for TG development (HR 2.1, 95% CI 1.2–3.8; p = 0.012) regardless of DSA status but not for graft failure at 36 months.

Conclusions: Intimal arteritis as part of ABMR represented a risk for early development of TG regardless of the presence or absence of DSA. Intimal arteritis in DSA positive ABMR represented the high-risk phenotype.

CRISPR/Cas9-Engineered HLA-Deleted Glomerular Endothelial Cells as a Tool to Predict Pathogenic Non-HLA Antibodies in Kidney Transplant Recipients

BACKGROUND

After kidney transplantation, donor-specific antibodies against human leukocyte antigen donor-specific antibodies (HLA-DSAs) drive antibody-mediated rejection (ABMR) and are associated with poor transplant outcomes. However, ABMR histology (ABMRh) is increasingly reported in kidney transplant recipients (KTRs) without HLA-DSAs, highlighting the emerging role of non-HLA antibodies (Abs).

METHODS

W e designed a non-HLA Ab detection immunoassay (NHADIA) using HLA class I and II-deficient glomerular endothelial cells (CiGEnCΔHLA) that had been previously generated through CRISPR/Cas9-induced B2M and CIITA gene disruption. Flow cytometry assessed the reactivity to non-HLA antigens of pretransplantation serum samples from 389 consecutive KTRs. The intensity of the signal observed with the NHADIA was associated with post-transplant graft histology assessed in 951 adequate biopsy specimens.

RESULTS

W e sequentially applied CRISPR/Cas9 to delete the B2M and CIITA genes to obtain a CiGEnCΔHLA clone. CiGEnCΔHLA cells remained indistinguishable from the parental cell line, CiGEnC, in terms of morphology and phenotype. Previous transplantation was the main determinant of the pretransplantation NHADIA result (P<0.001). Stratification of 3-month allograft biopsy specimens (n=298) according to pretransplantation NHADIA tertiles demonstrated that higher levels of non-HLA Abs positively correlated with increased glomerulitis (P=0.002), microvascular inflammation (P=0.003), and ABMRh (P=0.03). A pretransplantation NHADIA threshold of 1.87 strongly discriminated the KTRs with the highest risk of ABMRh (P=0.005, log-rank test). A multivariate Cox model confirmed that NHADIA status and HLA-DSAs were independent, yet synergistic, predictors of ABMRh.

CONCLUSION

The NHADIA identifies non-HLA Abs and strongly predicts graft endothelial injury independent of HLA-DSAs.

Chronic Allograft Injury

With the incremental improvements in long-term kidney transplant survival, there is renewed focus on what causes failure of the transplanted allograft. Over the past decade, our understanding of the injuries that lead to loss of graft function over time has evolved. Chronic allograft injury includes both immune-mediated and nonimmune-mediated injuries, which may involve the organ donor, the recipient, or both. The targets of injury include the kidney tubular epithelium, the endothelium, and the glomerulus. As a response to injury, there are the expected tissue remodeling and repair processes. However, if inflammation persists, which is not uncommon in the transplant setting, the resulting maladaptive response is matrix deposition and/or fibrosis. This ultimately leads to declining graft function and, finally, failure. With our advancing knowledge of the multiple etiologies and mechanisms, enhanced by more recent cohort studies in humans, there is an opportunity to identify those at greater risk to initiate new strategies to ameliorate the process. Although the most recent studies focus on immune-mediated injuries, there is a critical need to identify both markers of injury and mechanisms of injury. In this review, we highlight the findings of recent studies, highlight the potential therapeutic targets, and identify the continued unmet need for understanding the mechanisms of late graft failure.

Deep learning segmentation of glomeruli on kidney donor frozen sections

Purpose: Recent advances in computational image analysis offer the opportunity to develop automatic quantification of histologic parameters as aid tools for practicing pathologists. We aim to develop deep learning (DL) models to quantify nonsclerotic and sclerotic glomeruli on frozen sections from donor kidney biopsies. Approach: A total of 258 whole slide images (WSI) from cadaveric donor kidney biopsies performed at our institution ( n = 123 ) and at external institutions ( n = 135 ) were used in this study. WSIs from our institution were divided at the patient level into training and validation datasets (ratio: 0.8:0.2), and external WSIs were used as an independent testing dataset. Nonsclerotic ( n = 22767 ) and sclerotic ( n = 1366 ) glomeruli were manually annotated by study pathologists on all WSIs. A nine-layer convolutional neural network based on the common U-Net architecture was developed and tested for the segmentation of nonsclerotic and sclerotic glomeruli. DL-derived, manual segmentation, and reported glomerular count (standard of care) were compared. Results: The average Dice similarity coefficient testing was 0.90 and 0.83. And the F 1 , recall, and precision scores were 0.93, 0.96, and 0.90, and 0.87, 0.93, and 0.81, for nonsclerotic and sclerotic glomeruli, respectively. DL-derived and manual segmentation-derived glomerular counts were comparable, but statistically different from reported glomerular count. Conclusions: DL segmentation is a feasible and robust approach for automatic quantification of glomeruli. We represent the first step toward new protocols for the evaluation of donor kidney biopsies.

Progressive decline of function in renal allografts with normal 1-year biopsies: Gene expression studies fail to identify a classifier

Histologic 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.

Machine learning support for decision making in kidney transplantation: step-by-step development of a technological solution (Preprint)

Background

Kidney transplantation is the preferred treatment option for patients with end-stage renal disease. To maximize patient and graft survival, the allocation of donor organs to potential recipients requires careful consideration.

Objective

This study aimed to develop an innovative technological solution to enable better prediction of kidney transplant survival for each potential donor-recipient pair.

Methods

We used deidentified data on past organ donors, recipients, and transplant outcomes in the United States from the Scientific Registry of Transplant Recipients. To predict transplant outcomes for potential donor-recipient pairs, we used several survival analysis models, including regression analysis (Cox proportional hazards), random survival forests, and several artificial neural networks (DeepSurv, DeepHit, and recurrent neural network [RNN]). We evaluated the performance of each model in terms of its ability to predict the probability of graft survival after kidney transplantation from deceased donors. Three metrics were used: the C-index, integrated Brier score, and integrated calibration index, along with calibration plots.

Results

On the basis of the C-index metrics, the neural network–based models (DeepSurv, DeepHit, and RNN) had better discriminative ability than the Cox model and random survival forest model (0.650, 0.661, and 0.659 vs 0.646 and 0.644, respectively). The proposed RNN model offered a compromise between the good discriminative ability and calibration and was implemented in a technological solution of technology readiness level 4.

Conclusions

Our technological solution based on the RNN model can effectively predict kidney transplant survival and provide support for medical professionals and candidate recipients in determining the most optimal donor-recipient pair.

Application of the iBox prognostication system as a surrogate endpoint in the TRANSFORM randomised controlled trial: proof-of-concept study

OBJECTIVES

Development of pharmaceutical agents in transplantation is currently limited by long waits for hard endpoints. We applied a validated integrative risk-prognostication system integrative Box (iBox) as a surrogate endpoint to the TRANSFORM Study, a large randomised controlled trial, to project individual patient long-term kidney allograft survival from 1 year to 11 years after randomisation.

DESIGN

Post-hoc analysis of a randomised open-label controlled trial.

SETTING

Multicentre study including 186 centres in 42 countries worldwide.

PARTICIPANTS

2037 de novo kidney transplant recipients.

INTERVENTION

Participants were randomised (1:1) to receive everolimus with reduced-exposure calcineurin inhibitor (EVR+rCNI) or mycophenolic acid with standard-exposure CNI (MPA+sCNI).

PRIMARY OUTCOME MEASURE

The iBox scores were computed for each participant at 1 year after randomisation using functional, immunological and histological parameters. Individual long-term death-censored allograft survival over 4, 6 and 11 years after randomisation was projected with the iBox risk-prognostication system.

RESULTS

Overall, 940 patients receiving EVR+rCNI and 932 receiving MPA+sCNI completed the 1-year visit. iBox scores generated at 1 year yielded graft survival prediction rates of 90.9% vs 92.1%, 87.9% vs 89.5%, and 80.0% vs 82.4% in the EVR+rCNI versus MPA+sCNI arms at 4, 6, and 11 years post-randomisation, respectively (all differences below the 10% non-inferiority margin defined by study protocol). Inclusion of immunological and histological Banff diagnoses parameters in iBox scores resulted in comparable and non-inferior predicted graft survival for both treatments.

CONCLUSIONS

This proof-of-concept study provides the first application of a validated prognostication system as a surrogate endpoint in the field of transplantation. The iBox system, by projecting kidney allograft survival up to 11 years post-randomisation, confirms the non-inferiority of EVR+rCNI versus MPA+sCNI regimen. Given the current process engaged for surrogate endpoints qualification, this study illustrates the potential to fast track development of pharmaceutical agents.

TRIAL REGISTRATION NUMBER

TRANSFORM trial: NCT01950819.iBox prognostication system: NCT03474003.

Big Data in Nephrology

A huge array of data in nephrology is collected through patient registries, large epidemiological studies, electronic health records, administrative claims, clinical trial repositories, mobile health devices and molecular databases. Application of these big data, particularly using machine-learning algorithms, provides a unique opportunity to obtain novel insights into kidney diseases, facilitate personalized medicine and improve patient care. Efforts to make large volumes of data freely accessible to the scientific community, increased awareness of the importance of data sharing and the availability of advanced computing algorithms will facilitate the use of big data in nephrology. However, challenges exist in accessing, harmonizing and integrating datasets in different formats from disparate sources, improving data quality and ensuring that data are secure and the rights and privacy of patients and research participants are protected. In addition, the optimism for data-driven breakthroughs in medicine is tempered by scepticism about the accuracy of calibration and prediction from in silico techniques. Machine-learning algorithms designed to study kidney health and diseases must be able to handle the nuances of this specialty, must adapt as medical practice continually evolves, and must have global and prospective applicability for external and future datasets.

The clinical value of donor-derived cell-free DNA measurements in kidney transplantation

Early diagnosis is critical to minimizing the damage rejection can do to the transplanted kidney. Donor-derived cell-free DNA (dd-cfDNA) represents non-encapsulated fragmented DNA that is continuously shed into the bloodstream from the allograft undergoing injury, with a half-life of about 30 min. This article reviews the available evidence regarding the diagnostic value of dd-cfDNA in kidney transplantation, as a result of which two assays, Allosure and Prospera, have garnered Medicare approval. We provide information on important scenarios and contexts including antibody-mediated rejection, T-cell mediated rejection, pre-test probability of rejection, timing of the test, repeat transplants, and background cell-free DNA levels to help our understanding of the test characteristics and utility of these assays in clinical practice. Data on multimodality assays including gene expression profiles and serial monitoring of dd-cfDNA in high risk situations are emerging.

Predicting Kidney Graft Survival Using Machine Learning Methods: Prediction Model Development and Feature Significance Analysis Study

BACKGROUND

Kidney transplantation is the optimal treatment for patients with end-stage renal disease. Short- and long-term kidney graft survival is influenced by a number of donor and recipient factors. Predicting the success of kidney transplantation is important for optimizing kidney allocation.

OBJECTIVE

The aim of this study was to predict the risk of kidney graft failure across three temporal cohorts (within 1 year, within 5 years, and after 5 years following a transplant) based on donor and recipient characteristics. We analyzed a large data set comprising over 50,000 kidney transplants covering an approximate 20-year period.

METHODS

We applied machine learning-based classification algorithms to develop prediction models for the risk of graft failure for three different temporal cohorts. Deep learning-based autoencoders were applied for data dimensionality reduction, which improved the prediction performance. The influence of features on graft survival for each cohort was studied by investigating a new nonoverlapping patient stratification approach.

RESULTS

Our models predicted graft survival with area under the curve scores of 82% within 1 year, 69% within 5 years, and 81% within 17 years. The feature importance analysis elucidated the varying influence of clinical features on graft survival across the three different temporal cohorts.

CONCLUSIONS

In this study, we applied machine learning to develop risk prediction models for graft failure that demonstrated a high level of prediction performance. Acknowledging that these models performed better than those reported in the literature for existing risk prediction tools, future studies will focus on how best to incorporate these prediction models into clinical care algorithms to optimize the long-term health of kidney recipients.

Long-Term Immunosuppression Management: Opportunities and Uncertainties

The long-term management of maintenance immunosuppression in kidney transplant recipients remains complex. The vast majority of patients are treated with the calcineurin inhibitor tacrolimus as the primary agent in combination with mycophenolate, with or without corticosteroids. A tacrolimus trough target 5-8 ng/ml seems to be optimal for rejection prophylaxis, but long-term tacrolimus-related side effects and nephrotoxicity support the ongoing evaluation of noncalcineurin inhibitor-based regimens. Current alternatives include belatacept or mammalian target of rapamycin inhibitors. For the former, superior kidney function at 7 years post-transplant compared with cyclosporin generated initial enthusiasm, but utilization has been hampered by high initial rejection rates. Mammalian target of rapamycin inhibitors have yielded mixed results as well, with improved kidney function tempered by higher risk of rejection, proteinuria, and adverse effects leading to higher discontinuation rates. Mammalian target of rapamycin inhibitors may play a role in the secondary prevention of squamous cell skin cancer as conversion from a calcineurin inhibitor to an mammalian target of rapamycin inhibitor resulted in a reduction of new lesion development. Early withdrawal of corticosteroids remains an attractive strategy but also is associated with a higher risk of rejection despite no difference in 5-year patient or graft survival. A major barrier to long-term graft survival is chronic alloimmunity, and regardless of agent used, managing the toxicities of immunosuppression against the risk of chronic antibody-mediated rejection remains a fragile balance.

The Future Role of Machine Learning in Clinical Transplantation

The use of artificial intelligence and machine learning (ML) has revolutionized our daily lives and will soon be instrumental in healthcare delivery. The rise of ML is due to multiple factors: increasing access to massive datasets, exponential increases in processing power, and key algorithmic developments that allow ML models to tackle increasingly challenging questions. Progressively more transplantation research is exploring the potential utility of ML models throughout the patient journey, although this has not yet widely transitioned into the clinical domain. In this review, we explore common approaches used in ML in solid organ clinical transplantation and consider opportunities for ML to help clinicians and patients. We discuss ways in which ML can aid leverage of large complex datasets, generate cutting-edge prediction models, perform clinical image analysis, discover novel markers in molecular data, and fuse datasets to generate novel insights in modern transplantation practice. We focus on key areas in transplantation in which ML is driving progress, explore the future potential roles of ML, and discuss the challenges and limitations of these powerful tools.

Systems biology approaches in solid organ transplantation

PURPOSE OF REVIEW

Organ transplantation research has led to the discovery of several interesting individual mechanistic pathways, molecules and potential drug targets but there are still no comprehensive studies that have addressed how these varied mechanisms work in unison to regulate the posttransplant immune response that drives kidney rejection and dysfunction.

RECENT FINDINGS

Systems biology is a rapidly expanding field that aims to integrate existing knowledge of molecular concepts and large-scale genomic and clinical datasets into networks that can be used in cutting edge computational models to define disease mechanisms in a holistic manner. Systems biology approaches have brought a paradigm shift from a reductionist view of biology to a wider agnostic assessment of disease from several lines of evidence. Although the complex nature of the posttransplant immune response makes it difficult to pinpoint mechanisms, systems biology is enabling discovery of unknown biological interactions using the cumulative power of genomic data sets, clinical data and endpoints, and improved computational methods for the systematic deconvolution of this response.

SUMMARY

An integrative systems biology approach that leverages genomic data from varied technologies, such as DNA sequencing, copy number variation, RNA sequencing, and methylation profiles along with long-term clinical follow-up data has the potential to define a framework that can be mined to provide novel insights for developing therapeutic interventions in organ transplantation.

Impact of the Mayo Adhesive Probability Score on Donor and Recipient Outcomes After Living-donor Kidney Transplantation: A Retrospective, Single-center Study of 782 Transplants

Supplemental Digital Content is available in the text.

Background.

This study was performed to assess the impact of the Mayo Adhesive Probability (MAP) score on donor and recipient outcomes after living-donor kidney transplantation (LDKT).

Methods.

We retrospectively analyzed 782 transplants involving LDKT between February 2008 and October 2019 to assess the correlation between the MAP score and outcome after LDKT. We divided the transplants into 2 groups according to the donor MAP score: 0 (MAP₀) and 1–5 (MAP_1–5).

Results.

Compared with the MAP₀ group, donors in the MAP_1–5 group were significantly older, had higher body mass index, and were more likely to be men. The prevalences of hypertension, hyperlipidemia, and diabetes were also higher among donors in the MAP_1–5 group than among donors in the MAP₀ group. Operative time, estimated blood loss during donor nephrectomy, and percentage of glomerular sclerosis were significantly greater in the MAP_1–5 group than in the MAP₀ group. Donor and recipient perioperative complications were comparable between the 2 groups; death-censored graft survival rates also did not significantly differ between groups. Although the recipient mean estimated glomerular filtration rate (eGFR) from postoperative d 1 to 7 was significantly higher in the MAP₀ group than in the MAP_1–5 group (P = 0.007), eGFR reductions within 5 y after transplantation were similar between groups. There were no significant differences between groups in recipient mortality and biopsy-proven acute rejection episodes within 1 y after transplantation. Additionally, multivariate analysis showed that the only factors affecting recipient eGFR at postoperative d 7 were donor age, recipient age, and female sex (P < 0.001, <0.001, and =0.004, respectively).

Conclusions.

The MAP score did not influence surgical complications or graft survival; therefore, it should not affect donor selection.

Clinical Rationale for a Routine Testing Schedule Using Donor-Derived Cell-Free DNA After Kidney Transplantation

Kidney transplant recipients require meticulous clinical and laboratory surveillance to monitor allograft health. Conventional biomarkers, including serum creatinine and proteinuria, are lagging indicators of allograft injury, often rising only after significant and potentially irreversible damage has occurred. Immunosuppressive medication levels can be followed, but their utility is largely limited to guiding dosing changes or assessing adherence. Kidney biopsy, the criterion standard for the diagnosis and characterization of injury, is invasive and thus poorly suited for frequent surveillance.

Donor-derived cell-free DNA (dd-cfDNA) is a sensitive, noninvasive, leading indicator of allograft injury, which offers the opportunity for expedited intervention and can improve long-term allograft outcomes. This article describes the clinical rationale for a routine testing schedule utilizing dd-cfDNA surveillance at months 1, 2, 3, 4, 6, 9, and 12 during the first year following kidney transplantation and quarterly thereafter. These time points coincide with major immunologic transition points after transplantation and provide clinicians with molecular information to help inform decision making.

The evolution of histological changes suggestive of antibody-mediated injury, in the presence and absence of donor-specific anti-HLA antibodies

The interplay between donor-specific anti-HLA antibodies (HLA-DSA), histology of active antibody-mediated rejection (aABMR_h ), transplant glomerulopathy (cg) and graft failure in kidney transplantation remains insufficiently understood. We performed a single-center cohort study (n=1000) including 2761 protocol and 833 indication biopsies. Patients with pre-transplant HLA-DSA were more prone to develop aABMRh (OR 22.7, 95% CI, 11.8 - 43.7, p<0.001), cg (OR 5.76, 95% CI, 1.67 - 19.8, p=0.006) and aABMRh/cg (OR 19.5, 95% CI, 10.6 - 35.9, p<0.001). The negative impact of pre-transplant HLA-DSA on graft survival (HR 2.12, 95% CI, 1.41 - 3.20, p<0.001) was partially mediated through aABMRh and cg occurrence. When adjusted for time-dependent HLA-DSA (HR 4.03, 95% CI, 2.21 - 7.15, p=0.002), graft failure was only affected by aABMR_h when cg was evident. In HLA-DSA negative patients, aABMR_h was associated with impaired graft outcome only when evolving to cg (HR 1.32, 95% CI, 1.07 - 1.61, p=0.008). We conclude that the kinetics of HLA-DSA are important to estimate the rate of graft failure, and that histological follow-up is necessary to discover, often subclinical, ABMR and cg. In the absence of HLA-DSA, patients experience similar histological lesions and the evolution to transplant glomerulopathy associates with impaired graft outcome.

US deceased kidney transplantation: Estimated GFR, donor age and KDPI association with graft survival

BACKGROUND

Despite a significant shortage of kidneys for transplantation in the US, kidneys from older deceased donors are infrequently transplanted. This is primarily over concern of graft quality and transplant durability.

METHODS

The US national transplant database (2000-2018) was assessed for deceased donor kidney transplant patient and graft survival, graft durability and stratified by donor age (<65 years>), Kidney Donor Profile Index (KDPI) and estimated glomerual filtration rate (GFR) one year post-transplantation (eGFR-1) were calculated.

FINDINGS

Recipients of kidneys transplanted from deceased donors >65 years had a lower eGFR-1, (median 39 ml/min) than recipients of younger donor kidneys (median 54 ml/min). However, death-censored graft survival, stratified by eGFR-1, demonstrated similar survival, irrespective of donor age or KDPI. The durability of kidney survival decreases as the achieved eGFR-1 declines. KDPI has a poor association with eGFR-1 and lesser for graft durability. While recipients of kidneys > 65 years had a higher one year mortality than younger kidney recipients, recipients of kidneys > 65 years and an eGFR-1 <30 ml/min, had a lower survival than an untransplanted waitlist cohort (p<0.001).

INTERPRETATION

The durability of kidney graft survival after transplantation was associated with the amount of kidney function gained through the transplant (eGFR-1) and the rate of graft loss (return to dialysis) was not significantly associated with donor age. 24.9% of recipients of older donor kidneys failed to achieve sufficient eGFR-1 providing a transplant survival benefit. While there is significant benefit from transplanting older kidneys, better decision-making tools are required to avoid transplanting kidneys that provide insufficient renal function.

FUNDING

None.

Defining a minimal clinically meaningful difference in 12-month estimated glomerular filtration rate for clinical trials in deceased donor kidney transplantation

BACKGROUND

A Minimal Clinically Meaningful Difference (MCMD) has not been defined for Estimated glomerular filtration rate (eGFR). Our goal was to define the MCMD for eGFR anchored to kidney graft failure.

METHODS

A systematic review of studies with 12-month eGFR and subsequent renal graft failure was conducted. For observational studies, we calculated hazard ratio (HR) differences between adjacent eGFR intervals weighted by population distribution. Interventional trials yielded therapeutically induced changes in eGFR and failure risk. OPTN data analysis divided 12-month eGFR into bands for Cox regressions comparing adjacent eGFR bands with a death-censored graft survival outcome.

RESULTS

Observational studies indicated that lower eGFR was associated with increased death-censored graft failure risk; each 5 ml/min/1.73 m2 12-month eGFR band associated with a weighted incremental HR = 1.12 to 1.23. Clinical trial data found a 5 ml/min/1.73 m2 difference was associated with incremental HR = 1.16 to 1.35. OPTN analyses showed weighted mean HRs across 10, 7, and 5 ml/min/1.73 m2 bands of 1.47, 1.30, and 1.19.

CONCLUSIONS

A 5 ml/min/1.73 m2 difference in 12-month eGFR was consistently associated with ~20% increase in death-censored graft failure risk. The magnitude of effect has been interpreted as clinically meaningful in other disease states and should be considered the MCMD in renal transplantation clinical trials.

Image Analysis Pipeline for Renal Allograft Evaluation and Fibrosis Quantification

INTRODUCTION

Digital pathology improves the standardization and reproducibility of kidney biopsy specimen assessment. We developed a pipeline allowing the analysis of many images without requiring human preprocessing and illustrate its use with a simple algorithm for quantification of interstitial fibrosis on a large dataset of kidney allograft biopsy specimens.

METHODS

Masson trichrome-stained images from kidney allograft biopsy specimens were used to train and validate a glomeruli detection algorithm using a VGG19 convolutional neural network and an automatic cortical region of interest (ROI) selection algorithm including cortical regions containing all predicted glomeruli. A positive-pixel count algorithm was used to quantify interstitial fibrosis on the ROIs and the association between automatic fibrosis and pathologist evaluation, estimated glomerular filtration rate (GFR) and allograft survival was assessed.

RESULTS

The glomeruli detection (F1 score of 0.87) and ROIs selection (F1 score 0.83 [SD 0.13]) algorithms displayed high accuracy. The correlation between the automatic fibrosis quantification on manually and automatically selected ROIs was high (r = 1.00 [0.99-1.00]). Automatic fibrosis quantification was only moderately correlated with pathologists' assessment and was not significantly associated with eGFR or allograft survival.

CONCLUSION

This pipeline can automatically and accurately detect glomeruli and select cortical ROIs that can easily be used to develop, validate, and apply image analysis algorithms.

Eudaimonia: An Aristotelian approach to transplantation

Despite extraordinary achievements in over the past 20 years, the field of transplantation remains hindered by relatively narrow metrics for success. Eudaimonia is an Aristotelian concept that refers to flourishing, or achieving the best conditions possible, in every sense. The vast amounts of patient data that are collected throughout the transplant care continuum, ranging from social determinants of health to genomic profiles and patient-reported outcomes, afford us unprecedented opportunity to enhance our definition of success for our transplant patients. We must engage the technologies available for data integration and analysis and apply them in an insightful way, such that our clinical practice evolves beyond patient and graft survival and toward a more comprehensive state of wellness.

A machine learning prediction model for waiting time to kidney transplant

Background

Predicting waiting time for a deceased donor kidney transplant can help patients and clinicians to discuss management and contribute to a more efficient use of resources. This study aimed at developing a predictor model to estimate time on a kidney transplant waiting list using a machine learning approach.

Methods

A retrospective cohort study including data of patients registered, between January 1, 2000 and December 31, 2017, in the waiting list of São Paulo State Organ Allocation System (SP-OAS) /Brazil. Data were randomly divided into two groups: 75% for training and 25% for testing. A Cox regression model was fitted with deceased donor transplant as the outcome. Sensitivity analyses were performed using different Cox models. Cox hazard ratios were used to develop the risk-prediction equations.

Results

Of 54,055 records retrieved, 48,153 registries were included in the final analysis. During the study period, approximately 1/3 of the patients were transplanted with a deceased donor. The major characteristics associated with changes in the likelihood of transplantation were age, subregion, cPRA, and frequency of HLA-DR, -B and -A. The model developed was able to predict waiting time with good agreement in internal validation (c-index = 0.70).

Conclusion

The kidney transplant waiting time calculator developed shows good predictive performance and provides information that may be valuable in assisting candidates and their providers. Moreover, it can significantly improve the use of economic resources and the management of patient care before transplant.

Incidence, risk factors, treatment, and consequences of antibody-mediated kidney transplant rejection: A systematic review

BACKGROUND

Antibody-mediated rejection (AMR) is a leading cause of kidney allograft failure, but its incidence, risk factors, and outcomes are not well understood.

METHODS

We searched Ovid MEDLINE, Cochrane, EMBASE, and Scopus from January 2000 to January 2020 to identify published cohorts of ≥500 incident adult or 75 pediatric kidney transplant recipients followed for ≥1 year post-transplant.

RESULTS

At least two reviewers screened 5061 articles and abstracts; 28 met inclusion criteria. Incidence of acute AMR was 1.1%-21.5%; most studies reported 3%-12% incidence, usually within the first year post-transplant. Few studies reported chronic AMR incidence, from 7.5%-20.1% up to 10 years. Almost all patients with acute or chronic AMR received corticosteroids and intravenous immunoglobulin; most received plasmapheresis, and approximately half with rituximab. Most studies examining death-censored graft failure identified AMR as an independent risk factor. Few reported refractory AMR rates or outcomes, and none examined costs. Most studies were single-center and varied greatly in design.

CONCLUSIONS

Cohort studies of kidney transplant recipients demonstrate that AMR is common and associated with increased risk of death-censored graft failure, but studies vary widely regarding populations, definitions, and reported incidence. Gaps remain in our understanding of refractory AMR, its costs, and resulting quality of life.

Data-driven Derivation and Validation of Novel Phenotypes for Acute Kidney Transplant Rejection using Semi-supervised Clustering

BACKGROUND

Over the past decades, an international group of experts iteratively developed a consensus classification of kidney transplant rejection phenotypes, known as the Banff classification. Data-driven clustering of kidney transplant histologic data could simplify the complex and discretionary rules of the Banff classification, while improving the association with graft failure.

METHODS

The data consisted of a training set of 3510 kidney-transplant biopsies from an observational cohort of 936 recipients. Independent validation of the results was performed on an external set of 3835 biopsies from 1989 patients. On the basis of acute histologic lesion scores and the presence of donor-specific HLA antibodies, stable clustering was achieved on the basis of a consensus of 400 different clustering partitions. Additional information on kidney-transplant failure was introduced with a weighted Euclidean distance.

RESULTS

Based on the proportion of ambiguous clustering, six clinically meaningful cluster phenotypes were identified. There was significant overlap with the existing Banff classification (adjusted rand index, 0.48). However, the data-driven approach eliminated intermediate and mixed phenotypes and created acute rejection clusters that are each significantly associated with graft failure. Finally, a novel visualization tool presents disease phenotypes and severity in a continuous manner, as a complement to the discrete clusters.

CONCLUSIONS

A semisupervised clustering approach for the identification of clinically meaningful novel phenotypes of kidney transplant rejection has been developed and validated. The approach has the potential to offer a more quantitative evaluation of rejection subtypes and severity, especially in situations in which the current histologic categorization is ambiguous.

Artificial Intelligence and Its Impact on Urological Diseases and Management: A Comprehensive Review of the Literature

Recent advances in artificial intelligence (AI) have certainly had a significant impact on the healthcare industry. In urology, AI has been widely adopted to deal with numerous disorders, irrespective of their severity, extending from conditions such as benign prostate hyperplasia to critical illnesses such as urothelial and prostate cancer. In this article, we aim to discuss how algorithms and techniques of artificial intelligence are equipped in the field of urology to detect, treat, and estimate the outcomes of urological diseases. Furthermore, we explain the advantages that come from using AI over any existing traditional methods.

A hybrid data envelopment analysis-artificial neural network prediction model for COVID-19 severity in transplant recipients

In an overwhelming demand scenario, such as the SARS-CoV-2 pandemic, pressure over health systems may outburst their predicted capacity to deal with such extreme situations. Therefore, in order to successfully face a health emergency, scientific evidence and validated models are needed to provide real-time information that could be applied by any health center, especially for high-risk populations, such as transplant recipients. We have developed a hybrid prediction model whose accuracy relative to several alternative configurations has been validated through a battery of clustering techniques. Using hospital admission data from a cohort of hospitalized transplant patients, our hybrid Data Envelopment Analysis (DEA)—Artificial Neural Network (ANN) model extrapolates the progression towards severe COVID-19 disease with an accuracy of 96.3%, outperforming any competing model, such as logistic regression (65.5%) and random forest (44.8%). In this regard, DEA-ANN allows us to categorize the evolution of patients through the values of the analyses performed at hospital admission. Our prediction model may help guiding COVID-19 management through the identification of key predictors that permit a sustainable management of resources in a patient-centered model.

Therapies for Chronic Allograft Rejection

Remarkable advances have been made in the pathophysiology, diagnosis, and treatment of antibody-mediated rejection (ABMR) over the past decades, leading to improved graft outcomes. However, long-term failure is still high and effective treatment for chronic ABMR, an important cause of graft failure, has not yet been identified. Chronic ABMR has a relatively different phenotype from active ABMR and is a slowly progressive disease in which graft injury is mainly caused by de novo donor specific antibodies (DSA). Since most trials of current immunosuppressive therapies for rejection have focused on active ABMR, treatment strategies based on those data might be less effective in chronic ABMR. A better understanding of chronic ABMR may serve as a bridge in establishing treatment strategies to improve graft outcomes. In this in-depth review, we focus on the pathophysiology and characteristics of chronic ABMR along with the newly revised Banff criteria in 2017. In addition, in terms of chronic ABMR, we identify the reasons for the resistance of current immunosuppressive therapies and look at ongoing research that could play a role in setting better treatment strategies in the future. Finally, we review non-invasive biomarkers as tools to monitor for rejection.

Toward Advancing Long-Term Outcomes of Kidney Transplantation with Artificial Intelligence

After decades of pioneering advances and improvements, kidney transplantation is now the renal replacement therapy of choice for most patients with end-stage kidney disease (ESKD). Despite this success, the high risk of premature death and frequent occurrence of graft failure remain important clinical and research challenges. The current burst of studies and other innovative initiatives using artificial intelligence (AI) for a wide range of analytical and practical applications in biomedical areas seems to correlate with the same trend observed in publications in the kidney transplantation field, and points toward the potential of such novel approaches to address the aforementioned aim of improving long-term outcomes of kidney transplant recipients (KTR). However, at the same time, this trend underscores now more than ever the old methodological challenges and potential threats that the research and clinical community needs to be aware of and actively look after with regard to AI-driven evidence. The purpose of this narrative mini-review is to explore challenges for obtaining applicable and adequate kidney transplant data for analyses using AI techniques to develop prediction models, and to propose next steps in the field. We make a call to act toward establishing the strong collaborations needed to bring innovative synergies further augmented by AI, which have the potential to impact the long-term care of KTR. We encourage researchers and clinicians to submit their invaluable research, including original clinical and imaging studies, database studies from registries, meta-analyses, and AI research in the kidney transplantation field.

[New technologies for renal pathology: Transcriptomics on paraffin-embedded fixed tissue]

The development of new high-throughput technologies in genomics and then in transcriptomics has modified clinical approach in nephrology. At the interface between high-throughput technologies (microarray, new generation sequencing «NGS») and few mRNA analysis (reverse transcriptase quantitative PCR [RT-qPCR]), the nCounter® of NanoString® offers a new and complementary approach. Capable of analyzing formalin-fixed paraffin-embedded samples, this technology is a credible candidate for implanting transcriptomics in clinical routine.

A noninferiority design for a delayed calcineurin inhibitor substitution trial in kidney transplantation

Improving long-term kidney transplant outcomes requires novel treatment strategies, including delayed calcineurin inhibitor (CNI) substitution, tested using informative trial designs. An alternative approach to the usual superiority-based trial is a noninferiority trial design that tests whether an investigational agent is not unacceptably worse than standard of care. An informative noninferiority design, with biopsy-proven acute rejection (BPAR) as the endpoint, requires determination of a prespecified, evidence-based noninferiority margin for BPAR. No such information is available for delayed CNI substitution in kidney transplantation. Herein we analyzed data from recent kidney transplant trials of CNI withdrawal and "real world" CNI- based standard of care, containing subjects with well-documented evidence of immune quiescence at 6 months posttransplant-ideal candidates for delayed CNI substitution. Our analysis indicates an evidence-based noninferiority margin of 13.8% for the United States Food and Drug Administration's composite definition of BPAR between 6 and 24 months posttransplant. Sample size estimation determined that ~225 randomized subjects would be required to evaluate noninferiority for this primary clinical efficacy endpoint, and superiority for a renal function safety endpoint. Our findings provide the basis for future delayed CNI substitution noninferiority trials, thereby increasing the likelihood they will provide clinically implementable results and achieve regulatory approval.

The evolution of patient-specific precision biomarkers to guide personalized heart-transplant care

Introduction

In parallel to the clinical maturation of heart transplantation over the last 50 years, rejection testing has been revolutionized within the systems biology paradigm triggered by the Human Genome Project.

Areas Covered

We have co-developed the first FDA-cleared diagnostic and prognostic leukocyte gene expression profiling biomarker test in transplantation medicine that gained international evidence-based medicine guideline acceptance to rule out moderate/severe acute cellular cardiac allograft rejection without invasive endomyocardial biopsies. This work prompted molecular re-classification of intragraft biology, culminating in the identification of a pattern of intragraft myocyte injury, in addition to acute cellular rejection and antibody-mediated rejection. This insight stimulated research into non-invasive detection of myocardial allograft injury. The addition of a donor-organ specific myocardial injury marker based on donor-derived cell-free DNA further strengthens the non-invasive monitoring concept, combining the clinical use of two complementary non-invasive blood-based measures, host immune activity-related risk of acute rejection as well as cardiac allograft injury.

Expert Opinion

This novel complementary non-invasive heart transplant monitoring strategy based on leukocyte gene expression profiling and donor-derived cell-free DNA that incorporates longitudinal variability measures provides an exciting novel algorithm of heart transplant allograft monitoring. This algorithm's clinical utility will need to be tested in an appropriately designed randomized clinical trial which is in preparation.

Development and validation of routine clinical laboratory data derived marker-based nomograms for the prediction of 5-year graft survival in kidney transplant recipients

Background: To develop and validate predictive nomograms for 5-year graft survival in kidney transplant recipients (KTRs) with easily-available laboratory data derived markers and clinical variables within the first year post-transplant.

Methods: The clinical and routine laboratory data from within the first year post-transplant of 1289 KTRs was collected to generate candidate predictors. Univariate and multivariate Cox analyses and LASSO were conducted to select final predictors. X-tile analysis was applied to identify optimal cutoff values to transform potential continuous factors into category variables and stratify patients. C-index, calibration curve, dynamic time-dependent AUC, decision curve analysis, and Kaplan-Meier curves were used to evaluate models’ predictive accuracy and clinical utility.

Results: Two predictive nomograms were constructed by using 0–6- and 0–12- month laboratory data, and showed good predictive performance with C-indexes of 0.78 and 0.85, respectively, in the training cohort. Calibration curves showed that the prediction probabilities of 5-year graft survival were in concordance with actual observations. Additionally, KTRs could be successfully stratified into three risk groups by nomograms.

Conclusions: These predictive nomograms combining demographic and 0–6- or 0–12- month markers derived from post-transplant laboratory data could serve as useful tools for early identification of 5-year graft survival probability in individual KTRs.

Clinical impact of serum bilirubin levels on kidney transplant outcomes

Serum bilirubin, a potent endogenous antioxidant, has been associated with decreased risks of cardiovascular disease, diabetes, and kidney disease. However, the effects of serum bilirubin on kidney transplant outcomes remain undetermined. We analyzed 1628 patients who underwent kidney transplantations between 2003 and 2017. Patients were grouped into sex-specific quartiles according to mean serum bilirubin levels, 3–12 months post-transplantation. Median bilirubin levels were 0.66 mg/dL in males and 0.60 mg/dL in females. The intra-individual variability of serum bilirubin levels was low (9%). Serum bilirubin levels were inversely associated with graft loss, death-censored graft failure, and all-cause mortality, independent of renal function, donor status, and transplant characteristics. Multivariable analysis revealed that the lowest serum bilirubin quartile was associated with increased risk of graft loss (HR 2.64, 95% CI 1.67–4.18, P < 0.001), death-censored graft failure (HR 2.97, 95% CI 1.63–5.42, P < 0.001), and all-cause mortality (HR 2.07, 95% CI 1.01–4.22, P = 0.046). Patients with lower serum bilirubin were also at greater risk of rejection and exhibited consistently lower glomerular filtration rates than those with higher serum bilirubin. Serum bilirubin levels were significantly associated with transplantation outcomes, suggesting that bilirubin could represent a therapeutic target for improving long-term transplant outcomes.

In-silico performance, validation, and modeling of the Nanostring Banff Human Organ transplant gene panel using archival data from human kidney transplants

BACKGROUND

RNA gene expression of renal transplantation biopsies is commonly used to identify the immunological patterns of graft rejection. Mostly done with microarrays, seminal findings defined the patterns of gene sets associated with rejection and non-rejection kidney allograft diagnoses. To make gene expression more accessible, the Molecular Diagnostics Working Group of the Banff Foundation for Allograft Pathology and NanoString Technologies partnered to create the Banff Human Organ Transplant Panel (BHOT), a gene panel set of 770 genes as a surrogate for microarrays (~ 50,000 genes). The advantage of this platform is that gene expressions are quantifiable on formalin fixed and paraffin embedded archival tissue samples, making gene expression analyses more accessible. The purpose of this report is to test in silico the utility of the BHOT panel as a surrogate for microarrays on archival microarray data and test the performance of the modelled BHOT data.

METHODS

BHOT genes as a subset of genes from downloaded archival public microarray data on human renal allograft gene expression were analyzed and modelled by a variety of statistical methods.

RESULTS

Three methods of parsing genes verify that the BHOT panel readily identifies renal rejection and non-rejection diagnoses using in silico statistical analyses of seminal archival databases. Multiple modelling algorithms show a highly variable pattern of misclassifications per sample, either between differently constructed principal components or between modelling algorithms. The misclassifications are related to the gene expression heterogeneity within a given diagnosis because clustering the data into 9 groups modelled with fewer misclassifications.

CONCLUSION

This report supports using the Banff Human Organ Transplant Panel for gene expression of human renal allografts as a surrogate for microarrays on archival tissue. The data modelled satisfactorily with aggregate diagnoses although with limited per sample accuracy and, thereby, reflects and confirms the modelling complexity and the challenges of modelling gene expression as previously reported.