An automated histological classification system for precision diagnostics of kidney allografts

Dynamic monitoring soft tissue healing via visualized Gd-crosslinked double network MRI microspheres

By integrating magnetic resonance-visible components with scaffold materials, hydrogel microspheres (HMs) become visible under magnetic resonance imaging(MRI), allowing for non-invasive, continuous, and dynamic monitoring of the distribution, degradation, and relationship of the HMs with local tissues. However, when these visualization components are physically blended into the HMs, it reduces their relaxation rate and specificity under MRI, weakening the efficacy of real-time dynamic monitoring. To achieve MRI-guided in vivo monitoring of HMs with tissue repair functionality, we utilized airflow control and photo-crosslinking methods to prepare alginate-gelatin-based dual-network hydrogel microspheres (G-AlgMA HMs) using gadolinium ions (Gd (III)), a paramagnetic MRI contrast agent, as the crosslinker. When the network of G-AlgMA HMs degrades, the cleavage of covalent bonds causes the release of Gd (III), continuously altering the arrangement and movement characteristics of surrounding water molecules. This change in local transverse and longitudinal relaxation times results in variations in MRI signal values, thus enabling MRI-guided in vivo monitoring of the HMs. Additionally, in vivo data show that the degradation and release of polypeptide (K2 (SL)6 K2 (KK)) from G-AlgMA HMs promote local vascular regeneration and soft tissue repair. Overall, G-AlgMA HMs enable non-invasive, dynamic in vivo monitoring of biomaterial degradation and tissue regeneration through MRI, which is significant for understanding material degradation mechanisms, evaluating biocompatibility, and optimizing material design.

Cherry on Top or Real Need? A Review of Explainable Machine Learning in Kidney Transplantation

Research on solid organ transplantation has taken advantage of the substantial acquisition of medical data and the use of artificial intelligence (AI) and machine learning (ML) to answer diagnostic, prognostic, and therapeutic questions for many years. Nevertheless, despite the question of whether AI models add value to traditional modeling approaches, such as regression models, their "black box" nature is one of the factors that have hindered the translation from research to clinical practice. Several techniques that make such models understandable to humans were developed with the promise of increasing transparency in the support of medical decision-making. These techniques should help AI to close the gap between theory and practice by yielding trust in the model by doctors and patients, allowing model auditing, and facilitating compliance with emergent AI regulations. But is this also happening in the field of kidney transplantation? This review reports the use and explanation of "black box" models to diagnose and predict kidney allograft rejection, delayed graft function, graft failure, and other related outcomes after kidney transplantation. In particular, we emphasize the discussion on the need (or not) to explain ML models for biological discovery and clinical implementation in kidney transplantation. We also discuss promising future research paths for these computational tools.

Utility of Routine Post Kidney Transplant Anti-HLA Antibody Screening

Introduction

De novo donor-specific antibody (dnDSA) is a strong biomarker associated with the development of antibody-mediated rejection (AMR) and graft loss after kidney transplantation. This procedure is expensive; however, systematic annual screening was recommended by some national organ transplant agencies or societies even though its clinical utility was not clearly established.

Methods

To address this question, we retrospectively assessed the incidence of dnDSA according to the test justification (clinically indicated or systematic) in a cohort of low-immunological risk patients, defined by being nonhuman leukocyte antigen (non-HLA)-sensitized and having no previous kidney transplants.

Results

A total of 1072 patients, for whom 4611 anti-HLA tests were performed, were included in the study. During the follow-up period of 8 (interquartile range, IQR: 5-11) years, 77 recipients developed dnDSA (prevalence of 7.2%). Thirty-five of these dnDSAs (45.5%) were detected during the first year posttransplantation. In 95% of patients with dnDSA, an immunizing event was identified in their medical records. dnDSA was detected in 46 of 4267 systematic screening tests (1.08%) performed. Active and chronic AMR were frequently observed in biopsies performed after systematic DSA testing (17.9% and 15.4%, respectively).

Conclusion

Our results suggest that the detection by systematic screening of dnDSA in low-immunological risk kidney transplant patients without sensitizing events is a rare event, especially after 1 year. Moreover, in real life, systematic annual screening for dnDSA, seems having a limited impact to detect AMR at an earlier stage compared to patients in whom dnDSA was detected after a clinically indicated test.

Biopsy-based transcriptomics in the diagnosis of kidney transplant rejection

PURPOSE OF REVIEW

The last year has seen considerable progress in translational research exploring the clinical utility of biopsy-based transcriptomics of kidney transplant biopsies to enhance the diagnosis of rejection. This review will summarize recent findings with a focus on different platforms, potential clinical applications, and barriers to clinical adoption.

RECENT FINDINGS

Recent literature has focussed on using biopsy-based transcriptomics to improve diagnosis of rejection, in particular antibody-mediated rejection. Different techniques of gene expression analysis (reverse transcriptase quantitative PCR, microarrays, probe-based techniques) have been used either on separate samples with ideally preserved RNA, or on left over tissue from routine biopsy processing. Despite remarkable consistency in overall patterns of gene expression, there is no consensus on acceptable indications, or whether biopsy-based transcriptomics adds significant value at reasonable cost to current diagnostic practice.

SUMMARY

Access to biopsy-based transcriptomics will widen as regulatory approvals for platforms and gene expression models develop. Clinicians need more evidence and guidance to inform decisions on how to use precious biopsy samples for biopsy-based transcriptomics, and how to integrate results with standard histology-based diagnosis.

Deep learning applications for kidney histology analysis

PURPOSE OF REVIEW

Nephropathology is increasingly incorporating computational methods to enhance research and diagnostic accuracy. The widespread adoption of digital pathology, coupled with advancements in deep learning, will likely transform our pathology practices. Here, we discuss basic concepts of deep learning, recent applications in nephropathology, current challenges in implementation and future perspectives.

RECENT FINDINGS

Deep learning models have been developed and tested in various areas of nephropathology, for example, predicting kidney disease progression or diagnosing diseases based on imaging and clinical data. Despite their promising potential, challenges remain that hinder a wider adoption, for example, the lack of prospective evidence and testing in real-world scenarios.

SUMMARY

Deep learning offers great opportunities to improve quantitative and qualitative kidney histology analysis for research and clinical nephropathology diagnostics. Although exciting approaches already exist, the potential of deep learning in nephropathology is only at its beginning and we can expect much more to come.

Design, cohort profile and comparison of the KTD-Innov study: a prospective multidimensional biomarker validation study in kidney allograft rejection

There is an unmet need for robust and clinically validated biomarkers of kidney allograft rejection. Here we present the KTD-Innov study (ClinicalTrials.gov, NCT03582436), an unselected deeply phenotyped cohort of kidney transplant recipients with a holistic approach to validate the clinical utility of precision diagnostic biomarkers. In 2018-2019, we prospectively enrolled consecutive adult patients who received a kidney allograft at seven French centers and followed them for a year. We performed multimodal phenotyping at follow-up visits, by collecting clinical, biological, immunological, and histological parameters, and analyzing a panel of 147 blood, urinary and kidney tissue biomarkers. The primary outcome was allograft rejection, assessed at each visit according to the international Banff 2019 classification. We evaluated the representativeness of participants by comparing them with patients from French, European, and American transplant programs transplanted during the same period. A total of 733 kidney transplant recipients (64.1% male and 35.9% female) were included during the study. The median follow-up after transplantation was 12.3 months (interquartile range, 11.9-13.1 months). The cumulative incidence of rejection was 9.7% at one year post-transplant. We developed a distributed and secured data repository in compliance with the general data protection regulation. We established a multimodal biomarker biobank of 16,736 samples, including 9331 blood, 4425 urinary and 2980 kidney tissue samples, managed and secured in a collaborative network involving 7 clinical centers, 4 analytical platforms and 2 industrial partners. Patients' characteristics, immune profiles and treatments closely resembled those of 41,238 French, European and American kidney transplant recipients. The KTD-Innov study is a unique holistic and multidimensional biomarker validation cohort of kidney transplant recipients representative of the real-world transplant population. Future findings from this cohort are likely to be robust and generalizable.

[Artificial intelligence in kidney transplant pathology]

BACKGROUND

Artificial intelligence (AI) systems have showed promising results in digital pathology, including digital nephropathology and specifically also kidney transplant pathology.

AIM

Summarize the current state of research and limitations in the field of AI in kidney transplant pathology diagnostics and provide a future outlook.

MATERIALS AND METHODS

Literature search in PubMed and Web of Science using the search terms "deep learning", "transplant", and "kidney". Based on these results and studies cited in the identified literature, a selection was made of studies that have a histopathological focus and use AI to improve kidney transplant diagnostics.

RESULTS AND CONCLUSION

Many studies have already made important contributions, particularly to the automation of the quantification of some histopathological lesions in nephropathology. This likely can be extended to automatically quantify all relevant lesions for a kidney transplant, such as Banff lesions. Important limitations and challenges exist in the collection of representative data sets and the updates of Banff classification, making large-scale studies challenging. The already positive study results make future AI support in kidney transplant pathology appear likely.

Electro-Optical Multiclassification Platform for Minimizing Occasional Inaccuracy in Point-of-Care Biomarker Detection

On-site diagnostic tests that accurately identify disease biomarkers lay the foundation for self-healthcare applications. However, these tests routinely rely on single-mode signals and suffer from insufficient accuracy, especially for multiplexed point-of-care tests (POCTs) within a few minutes. Here, this work develops a dual-mode multiclassification diagnostic platform that integrates an electrochemiluminescence sensor and a field-effect transistor sensor in a microfluidic chip. The microfluidic channel guides the testing samples to flow across electro-optical sensor units, which produce dual-mode readouts by detecting infectious biomarkers of tuberculosis (TB), human rhinovirus (HRV), and group B streptococcus (GBS). Then, machine-learning classifiers generate three-dimensional (3D) hyperplanes to diagnose different diseases. Dual-mode readouts derived from distinct mechanisms enhance the anti-interference ability physically, and machine-learning-aided diagnosis in high-dimensional space reduces the occasional inaccuracy mathematically. Clinical validation studies with 501 unprocessed samples indicate that the platform has an accuracy approaching 99%, higher than the 77%-93% accuracy of rapid point-of-care testing technologies at 100% statistical power (>150 clinical tests). Moreover, the diagnosis time is 5 min without a trade-off of accuracy. This work solves the occasional inaccuracy issue of rapid on-site diagnosis, endowing POCT systems with the same accuracy as laboratory tests and holding unique prospects for complicated scenes of personalized healthcare.

Emerging phenotypes in kidney transplant rejection

PURPOSE OF REVIEW

This review focuses on more recently emerging rejection phenotypes in the context of time post transplantation and the resulting differential diagnostic challenges. It also discusses how novel ancillary diagnostic tools can potentially increase the accuracy of biopsy-based rejection diagnosis.

RECENT FINDINGS

With advances in reducing immunological risk at transplantation and improved immunosuppression treatment renal allograft survival improved. However, allograft rejection remains a major challenge and represent a frequent course for allograft failure. With prolonged allograft survival, novel phenotypes of rejection are emerging, which can show complex overlap and transition between cellular and antibody-mediated rejection mechanisms as well as mixtures of acute/active and chronic diseases. With the emerging complexity in rejection phenotypes, it is crucial to achieve diagnostic accuracy in the individual patient.

SUMMARY

The prospective validation and adoption of novel molecular and computational diagnostic tools into well defined and appropriate clinical context of uses will improve our ability to accurately diagnose, stage, and grade allograft rejection.

Exploring the single-cell immune landscape of kidney allograft inflammation using imaging mass cytometry

Kidney allograft inflammation, mostly attributed to rejection and infection, is an important cause of graft injury and loss. Standard histopathological assessment of allograft inflammation provides limited insights into biological processes and the immune landscape. Here, using imaging mass cytometry with a panel of 28 validated biomarkers, we explored the single-cell landscape of kidney allograft inflammation in 32 kidney transplant biopsies and 247 high-dimensional histopathology images of various phenotypes of allograft inflammation (antibody-mediated rejection, T cell-mediated rejection, BK nephropathy, and chronic pyelonephritis). Using novel analytical tools, for cell segmentation, we segmented over 900 000 cells and developed a tissue-based classifier using over 3000 manually annotated kidney microstructures (glomeruli, tubules, interstitium, and arteries). Using PhenoGraph, we identified 11 immune and 9 nonimmune clusters and found a high prevalence of memory T cell and macrophage-enriched immune populations across phenotypes. Additionally, we trained a machine learning classifier to identify spatial biomarkers that could discriminate between the different allograft inflammatory phenotypes. Further validation of imaging mass cytometry in larger cohorts and with more biomarkers will likely help interrogate kidney allograft inflammation in more depth than has been possible to date.

The Banff 2022 Kidney Meeting Report: Reappraisal of microvascular inflammation and the role of biopsy-based transcript diagnostics

The XVI-th Banff Meeting for Allograft Pathology was held at Banff, Alberta, Canada, from 19th to 23rd September 2022, as a joint meeting with the Canadian Society of Transplantation. To mark the 30th anniversary of the first Banff Classification, premeeting discussions were held on the past, present, and future of the Banff Classification. This report is a summary of the meeting highlights that were most important in terms of their effect on the Classification, including discussions around microvascular inflammation and biopsy-based transcript analysis for diagnosis. In a postmeeting survey, agreement was reached on the delineation of the following phenotypes: (1) "Probable antibody-mediated rejection (AMR)," which represents donor-specific antibodies (DSA)-positive cases with some histologic features of AMR but below current thresholds for a definitive AMR diagnosis; and (2) "Microvascular inflammation, DSA-negative and C4d-negative," a phenotype of unclear cause requiring further study, which represents cases with microvascular inflammation not explained by DSA. Although biopsy-based transcript diagnostics are considered promising and remain an integral part of the Banff Classification (limited to diagnosis of AMR), further work needs to be done to agree on the exact classifiers, thresholds, and clinical context of use.

Fluorescence spectroscopic profiling of urine samples for predicting kidney transplant rejection

Rejection is the primary factor affecting the functionality of a kidney post-transplant, where its prompt prediction of risk significantly influences therapeutic strategies and clinical outcomes. Current graft health assessment methods, including serum creatinine measurements and transplant kidney puncture biopsies, possess considerable limitations. In contrast, urine serves as a direct indicator of the graft's degenerative stage and provides a more accurate measure than peripheral blood analysis, given its non-invasive collection of kidney-specific metabolite. This research entailed collecting fluorescent fingerprint data from 120 urine samples of post-renal transplant patients using hyperspectral imaging, followed by the development of a learning model to detect various forms of immunological rejection. The model successfully identified multiple rejection types with an average diagnostic accuracy of 95.56 %.Beyond proposing an innovative approach for predicting the risk of complications post-kidney transplantation, this study heralds the potential introduction of a non-invasive, rapid, and accurate supplementary method for risk assessment in clinical practice.

Banff Digital Pathology Working Group: Image Bank, Artificial Intelligence Algorithm, and Challenge Trial Developments

The Banff Digital Pathology Working Group (DPWG) was established with the goal to establish a digital pathology repository; develop, validate, and share models for image analysis; and foster collaborations using regular videoconferencing. During the calls, a variety of artificial intelligence (AI)-based support systems for transplantation pathology were presented. Potential collaborations in a competition/trial on AI applied to kidney transplant specimens, including the DIAGGRAFT challenge (staining of biopsies at multiple institutions, pathologists' visual assessment, and development and validation of new and pre-existing Banff scoring algorithms), were also discussed. To determine the next steps, a survey was conducted, primarily focusing on the feasibility of establishing a digital pathology repository and identifying potential hosts. Sixteen of the 35 respondents (46%) had access to a server hosting a digital pathology repository, with 2 respondents that could serve as a potential host at no cost to the DPWG. The 16 digital pathology repositories collected specimens from various organs, with the largest constituent being kidney (n = 12,870 specimens). A DPWG pilot digital pathology repository was established, and there are plans for a competition/trial with the DIAGGRAFT project. Utilizing existing resources and previously established models, the Banff DPWG is establishing new resources for the Banff community.

Pathophysiology of Rejection in Kidney Transplantation

Kidney transplantation has been the optimal treatment for end-stage kidney disease for almost 70 years, with increasing frequency over this period. Despite the prevalence of the procedure, allograft rejection continues to impact transplant recipients, with consequences ranging from hospitalization to allograft failure. Rates of rejection have declined over time, which has been largely attributed to developments in immunosuppressive therapy, understanding of the immune system, and monitoring. Developments in these therapies, as well as an improved understanding of rejection risk and the epidemiology of rejection, are dependent on a foundational understanding of the pathophysiology of rejection. This review explains the interconnected mechanisms behind antibody-mediated and T-cell-mediated rejection and highlights how these processes contribute to outcomes and can inform future progress.

Perspective of a Pathologist on Benchmark Strategies for Artificial Intelligence Development in Organ Transplantation

Transplant pathology of donors is a highly specialized field comprising both the evaluation of organ donor biopsy for the oncological risk transmission and to guide the organ allocation. Timing is critical in transplant procurement since organs must be recovered as soon as possible to ensure the best possible outcome for the recipient. To all this is added the fact that the evaluation of a donor causes difficulties in many cases and the impact of these assessments is paramount, considering the possible recovery of organs that would have been erroneously discarded or, conversely, the possibly correct discarding of donors with unacceptable risk profiles. In transplant pathology histology is still the gold standard for diagnosis dictating the subsequent decisions and course of clinical care. Digital pathology has played an important role in accelerating healthcare progression and nowadays artificial intelligence powered computational pathology can effectively improve diagnostic needs, supporting the quality and safety of the process. Mapping the shape of the journey would suggest a progressive approach from supervised to semi/unsupervised models, which would involve training these models directly for clinical endpoints. In machine learning, this generally delivers better performance, compensating for a potential lack in interpretability. With planning and enough confidence in the performance of learning-based methods from digital pathology and artificial intelligence, there is great potential to augment the diagnostic quality and correlation with clinical endpoints. This may improve the donor pool and vastly reduce diagnostic and prognostic errors that are known but currently are unavoidable in transplant donor pathology.