Dynamic prediction of renal survival among deeply phenotyped kidney transplant recipients using artificial intelligence: an observational, international, multicohort study

Raynaud, Marc; Aubert, Olivier; Divard, Gillian; Reese, Peter P.; Kamar, Nassim; Yoo, Daniel; Chin, Chen-Shan; Bailly, Élodie; Büchler, Matthias; Ladrière, Marc; Quintrec, Moglie Le; Jurić, Ivana; Bašić‐Jukić, Nikolina; Crespo, Marta; Silva, Helio Tedesco; Linhares, Kamilla; Castro, Maria Cristina Ribeiro de; Pujol, Gervasio Soler; Empana, Jean‐Philippe; Ulloa, Camilo; Akalin, Enver; Böhmig, Georg A.; Huang, Edmund; Stegall, Mark D.; Bentall, Andrew; Montgomery, Robert A.; Jordan, Stanley C.; Oberbauer, Rainer; Segev, Dorry L.; Friedewald, John; Jouven, Xavier; Legendre, Christophe; Lefaucheur, Carmen; Loupy, Alexandre

doi:10.1016/s2589-7500(21)00209-0

Cited by 36 publications

(35 citation statements)

References 38 publications

Supporting

Mentioning

Contrasting

Order By: Relevance

“…iBox defines a score based on immunologic, histologic, and functional (eGFR and proteinuria) recipient’s criteria [ 32 ]. Other works have also been described with common molecules but different classifier approaches [ 33 , 34 ]. Although such allograft survival prediction models hold promise, these tools work well at the population level but lose accuracy for a specific individual [ 35 ].…”

Section: Molecular Omics After Transplantationmentioning

confidence: 99%

“…This may include for instance-specific buffers and/or adjuvants, which would maximize molecular recovery. Successful implementation into clinical practice hinges on: (i) cost-effectiveness; (ii) availability of clinically realistic sample collection and procedures; (iii) development and validation of viable bioanalytical strategies; and (iv) software tools that allow for data analysis and fast translation into clinically meaningful information [ 33 , 34 , 57 ].…”

Section: Toward An Optimized Use Of Omics In Clinical Application: Wo...mentioning

confidence: 99%

See 1 more Smart Citation

Molecular Markers of Kidney Transplantation Outcome: Current Omics Tools and Future Developments

Lepoittevin

Kerforne

Pellerin

et al. 2022

IJMS

View full text Add to dashboard Cite

Purpose of review: The emerging field of molecular predictive medicine is aiming to change the traditional medical approach in renal transplantation. Many studies have explored potential biomarker molecules with predictive properties in renal transplantation, issued from omics research. Herein, we review the biomarker molecules of four technologies (i.e., Genomics, Transcriptomics, Proteomics, and Metabolomics) associated with favorable kidney transplant outcomes. Recent findings: Several panels of molecules have been associated with the outcome that the majority of markers are related to inflammation and immune response; although. other molecular ontologies are also represented, such as proteasome, growth, regeneration, and drug metabolism. Throughout this review, we highlight the lack of properly validated statistical demonstration. Indeed, the most preeminent molecular panels either remain at the limited size study stage or are not confirmed during large-scale studies. At the core of this problem, we identify the methodological shortcomings and propose a comprehensive workflow for discovery and validation of molecular biomarkers that aims to improve the relevance of these tools in the future. Summary: Overall, adopting a patient management through omics approach could bring remarkable improvement to transplantation success. An increased effort and investment between scientists, medical biologists, and clinicians seem to be the path toward a proper solution.

show abstract

Section: Molecular Omics After Transplantationmentioning

confidence: 99%

Section: Toward An Optimized Use Of Omics In Clinical Application: Wo...mentioning

confidence: 99%

Molecular Markers of Kidney Transplantation Outcome: Current Omics Tools and Future Developments

Lepoittevin

Kerforne

Pellerin

et al. 2022

IJMS

View full text Add to dashboard Cite

show abstract

“…In an impressive multicenter study including data from 13,608 patients (89,328 patient years) a model for prediction of renal survival seven years after initial assessment in transplant recipients was developed based on histological, clinical, and immunological data and repeated measurements of eGFR and proteinuria [26 ▪ ]. However, this study did not use histology directly, but pathologist derived scores and is thus not a direct tool for pathologists.…”

Section: Applicationsmentioning

confidence: 99%

The potential of artificial intelligence-based applications in kidney pathology

Büllow

Marsh

Swamidass

et al. 2022

Current Opinion in Nephrology &Amp; Hypertension

View full text Add to dashboard Cite

Purpose of reviewThe field of pathology is currently undergoing a significant transformation from traditional glass slides to a digital format dependent on whole slide imaging. Transitioning from glass to digital has opened the field to development and application of image analysis technology, commonly deep learning methods (artificial intelligence [AI]) to assist pathologists with tissue examination. Nephropathology is poised to leverage this technology to improve precision, accuracy, and efficiency in clinical practice. Recent findingsThrough a multidisciplinary approach, nephropathologists, and computer scientists have made significant recent advances in developing AI technology to identify histological structures within whole slide images (segmentation), quantification of histologic structures, prediction of clinical outcomes, and classifying disease. Virtual staining of tissue and automation of electron microscopy imaging are emerging applications with particular significance for nephropathology. SummaryAI applied to image analysis in nephropathology has potential to transform the field by improving diagnostic accuracy and reproducibility, efficiency, and prognostic power. Reimbursement, demonstration of clinical utility, and seamless workflow integration are essential to widespread adoption.

show abstract

“…The advent of machine learning and Bayesian concepts are allowing dynamic prediction models based both on pre- and posttransplant data and hold the promise of offering constantly refined individual predictions. 17…”

Section: Historical Perspectivementioning

confidence: 99%

“…72 A recent study developed on the iBox platform advances our understanding of this concept. 17 The underlying concept is that the traditional Cox model uses data from one point in time to examine time to event data posttransplantation. However, trajectories of posttransplant dynamically evolving domains such as eGFR and proteinuria are not captured in the traditional approach.…”

Section: Immune Endpointsmentioning

confidence: 99%

Shifting Clinical Trial Endpoints in Kidney Transplantation: The Rise of Composite Endpoints and Machine Learning to Refine Prognostication

et al. 2022

View full text Add to dashboard Cite

The measurement of outcomes in kidney transplantation has been more accurately documented than almost any other surgical procedure result in recent decades. With significant improvements in short-and long-term outcomes related to optimized immunosuppression, outcomes have gradually shifted away from conventional clinical endpoints (ie, patient and graft survival) to surrogate and composite endpoints. This article reviews how outcomes measurements have evolved in the past 2 decades in the setting of increased data collection and summarizes recent advances in outcomes measurements pertaining to clinical, histopathological, and immune outcomes. Finally, we discuss the use of composite endpoints and Bayesian concepts, specifically focusing on the integrative box risk prediction score, in conjunction with machine learning to refine prognostication.

show abstract

Dynamic prediction of renal survival among deeply phenotyped kidney transplant recipients using artificial intelligence: an observational, international, multicohort study

Cited by 36 publications

References 38 publications

Molecular Markers of Kidney Transplantation Outcome: Current Omics Tools and Future Developments

Molecular Markers of Kidney Transplantation Outcome: Current Omics Tools and Future Developments

The potential of artificial intelligence-based applications in kidney pathology

Shifting Clinical Trial Endpoints in Kidney Transplantation: The Rise of Composite Endpoints and Machine Learning to Refine Prognostication

Contact Info

Product

Resources

About