Defining Relationships Among Tests for Kidney Transplant Antibody-Mediated Rejection

Abstract

Introduction: Effective therapies for kidney transplant antibody-mediated rejection (ABMR) will require accurate diagnoses plus assessment of ABMR activity, and the new tests that were used to show treatment effects in the clinical trial such as donor-derived cell-free DNA (dd-cfDNA) and molecular biopsy analysis (the Molecular Microscope Diagnostic System) could be useful.

Methods: Trifecta-Kidney (ClinicalTrials.gov #NCT04239703) studied 717 indication biopsies to define the relationships among the following 4 tests used for ABMR: (i) standard-of-care (SOC) local histologic biopsy ABMR diagnosis, (ii) MMDx ABMR diagnosis, (iii) dd-cfDNA, and (iv) donor-specific antibody (DSA).

Results: All 4 tests were correlated in a partial correlation network, with a hierarchy of intertest correlations: MMDx ABMR > dd-cfDNA > histology ABMR > DSA. Surprisingly, DSA correlated at least as strongly with MMDx ABMR as with histologic ABMR, even though DSA is not used in MMDx. When expressed in the same 6 rejection classes, MMDx diagnosed ABMR more frequently than histology. When histology disagreed with MMDx ABMR, dd-cfDNA and DSA correlated more strongly with MMDx assessment. However, histology also detected ABMR lesions in some cases that MMDx called No Rejection, correlating with subthreshold molecular ABMR activity, dd-cfDNA and DSA (AJT 25:72-87, 2024). Molecular rejection predicted graft outcomes better than histologic rejection in Trifecta-Kidney, and this finding was confirmed in the earlier INTERCOMEX study cohort.

Discussion: The 4-way intertest correlations extend below current thresholds for diagnosing ABMR. These results map a network of 4 ABMR-related tests that can add precision to ABMR assessment in trials and clinical management, and highlight the need to establish the clinical significance of subthreshold ABMR activity.

Keywords: DSA; donor-derived cell-free DNA; kidney transplant; molecular diagnostics; rejection.

Graphical abstract

Figure 1

Sample inclusion diagram and study workflow for this Trifecta-Kidney N = 717 study. (a) CONSORT showing sample inclusion and exclusion, as well as sample processing methodology for this study. (b) Study design showing analytical workflow for the biopsy and blood-based assessments. ABMRM, antibody-mediated rejection or mixed rejection; CONSORT, consolidated standards of reporting trials.

Figure 2

Partial correlation network plots and correlograms, illustrating the relationships between the positive assessments for molecular ABMR plus mixed (ABMRM), histologic ABMRM, DSA, and dd-cfDNA. (a) Partial correlation network incorporating dd-cfDNA as a binary variable (%dd-cfDNA ≥ 1.0). (b) Correlogram showing all partial correlation coefficients between each of the 4 independent tests, incorporating dd-cfDNA as a binary variable (%dd-cfDNA ≥ 1.0). (c) Partial correlation network and (d) correlogram expressing dd-cfDNA as a continuous variable. ABMR, antibody-mediated rejection; ABMRM, ABMR or Mixed Rejection; dd-cfDNA, donor-derived cell-free DNA; DSA, donor-specific antibody.

Figure 3

Relationships between dd-cfDNA and the ABMR_Prob classifier score in (a) all biopsies, (b) MMDx No ABMRM biopsies, and (c) MMDx No Rejection biopsies. Scatterplots show %dd-cfDNA (log₁₀ scale) on the y-axis and continuous ABMR activity (the ABMR_Prob classifier score, arcsine transformed) on the x-axis. Biopsies called ABMRM by histology are colored blue, the regression line for ABMR_Prob versus histology ABMRM diagnosis is colored blue, the regression line for ABMR_Prob versus dd-cfDNA is colored red. Correlation coefficients for dd-cfDNA and ABMR_Prob (Spearman), or histology ABMRM and ABMR_Prob (biserial), are shown on each plot for the designated biopsy population. ABMR, antibody-mediated rejection; ABMR_Prob, molecular ABMR activity classifier; ABMRM, ABMR or Mixed Rejection; dd-cfDNA, donor-derived cell-free DNA; DSA, donor-specific antibody; MMDx, Molecular Microscope Diagnostic System.