Novel Biomarker Can Accurately Detect Antibody-mediated Kidney Rejection

An 8-gene assay which can non-invasively identify patients suffering from antibody-mediated rejection of kidney transplant has recently been discovered in a research study led by scientists at KU Leuven. The findings are published in EBioMedicine.

Developed by analyzing blood samples collected at the time of renal allograft biopsies, the 8-gene assay detected the presence of antibody-mediated kidney rejection with high diagnostic accuracy at times of both stable graft function and graft dysfunction.

Currently the diagnosis of antibody-mediated kidney rejection relies on the use of invasive biopsies due to a lack of accurate non-invasive biomarkers. We interviewed Dr Maarten Naesens, Associate Professor at KU Leuven and Nephrologist at University Hospitals Leuven, to learn more about the study and the wider implications the 8-gene assay could have on the field of renal transplantation.

Anna MacDonald (AM): Can you give us a little background to the research and why identifying a biomarker for antibody-mediated rejection of kidney transplant is so important?

Maarten Naesens (MN): Antibody-mediated rejection is a leading cause of kidney allograft failure. The diagnosis of antibody-mediated rejection is made on kidney allograft biopsies performed at the time of decline in glomerular filtration rate or appearance of proteinuria. Because antibody-mediated rejection can occur in the absence of immediate clinical signs or changes in these graft functional characteristics, some centers also perform kidney allograft biopsies at fixed time points (surveillance biopsies). Glomerular filtration rate and proteinuria are non-specific markers for antibody-mediated rejection, as many other immunological and non-immunological injuries can disturb graft function. Based on the association between antibody-mediated rejection and kidney graft failure, and the impossibility to repeatedly perform invasive protocol-specified biopsies, more accurate non-invasive diagnostic markers are thus needed with better sensitivity and specificity.

AM: What were some of the biggest challenges you faced in identifying the biomarker?

MN: We came across some important challenges. First, our reference standard is the pathological phenotype of antibody-mediated rejection in the biopsy. This reference standard is inherently flawed by sampling error and interobserver variability. To minimize the interobserver variability the pathological assessment was centralized. Second, in our first phase, we used untargeted whole genome screening for detection of relevant genes. However, gene signals from the peripheral blood were rather weak, which is why we had to enrich with gene signals from the biopsy samples. For scoring the relevance of each gene transcript for antibody-mediated rejection five different machine learning algorithms were used and integrated. Next, we had to define a more selected list of gene transcripts, which was done in the second derivation phase. This selection was based on the relevance of the genes in pathophysiological pathways and on the scores calculated from the integrated machine learning tools. For further modelling into a ‘signature’, multiple combinations of gene transcripts were evaluated using internal cross validation methods. Finally, we aimed for a robust validation, which was realized in an independent population with real-life disease prevalence and demonstrated the added value of our test on top of already available clinical information.

AM: The study is referred to as “a landmark in the field of biomarker discovery and development in renal transplantation in several aspects”. Can you give us an overview of the reasons why?

MN: This study is a landmark in the field of biomarker discovery most importantly by its robust internal and external validation. This validation in a real-life population is often lacking in other biomarker studies but is indispensable for evaluation of the biomarker for clinical practice. Other aspects are the untargeted whole genome screening used for discovery of genes, whereas other studies often start from a selected set of targets, introducing some selection bias. Furthermore, we demonstrated the performance of our biomarker for detection of subclinical rejection, we demonstrated the additional benefit for clinical decision-making and the added value on top of readily available clinical information. All these aspects are very often lacking in other biomarker studies.

AM: What further clinical validation of the biomarker do you have planned?

MN: The current study was not designed to assess for the kinetics of the biomarker. In a next prospective follow-up study (‘BECS’, BIOMARGIN European, Ambispective Cohort of Adult and Pediatric Renal Transplant Patients) we will evaluate the kinetic evolution of our biomarker, e.g. whether its rise anticipates a rejection episode and whether it normalizes again after therapy. Also, we will look further into the association with outcome (graft failure).

AM: The study was part of the BIOMARGIN consortium. Can you tell us more about this project?

MN: The BIOMARGIN consortium is a European Commission - supported program aiming at discovering and validating robust non-invasive biomarkers for the follow up of renal grafts. BIOMARGIN is a 4-year European project (Collaborative Research Project) started in March 2013 coordinated by INSERM (Institut National de la Santé et de la Recherche Médicale, Professor Pierre Marquet), in Limoges, France. The consortium brings together 13 complementary partners, including three small and medium enterprises, one technology transfer/management company, five academic laboratories, and four University Hospitals from four European Member States (France, Belgium, Germany, and Sweden). More information can be found at www.biomargin.eu. The BIOMARGIN consortium has the ambitious but realistic goal to provide transplant clinicians with validated and easily accessible tools for early prediction of graft lesions and renal function deterioration by integrating multiple ‘omics’ strategies, offering them the possibility to personalize renal transplant patients’ treatment.

Maarten Naesens was speaking to Anna MacDonald, Science Writer for Technology Networks.