ABSTRACT
BACKGROUND AND HYPOTHESIS: Isolated Tubulitis, Borderline Changes, and Isolated Arteritis suspicious for histologic T cell-mediated rejection (hTCMR) remain findings of uncertain significance. Although the Molecular Microscope Diagnostics System (MMDx) has not been trained on those lesions, it was suggested that MMDx might reclassify a subgroup to molecular TCMR (mTCMR). METHODS: In this single-center cohort of 326 consecutive, unselected kidney allograft biopsies assessed by histology and MMDx, we analyzed 249 cases with Isolated Tubulitis (i0, t1-3, v0; n=101), Borderline Changes (according to Banff 2022, v0; n=9), Isolated Arteritis (no borderline, v1; n=37), No Inflammation (i0, t0, v0; n=67) and a Positive Control Cohort (hTCMR, n=27; Mixed histologic Rejection, n=8; both according to Banff 2022; total n=35). The first three groups were summarized as TCMR-Suspicion (n=147). Subcategorization included the presence and absence of microvascular inflammation (MVI; g+ptc≥2). Molecular rejection rates and differentiation were investigated. RESULTS: Molecular rejection rates were 37/147 cases (25.2%; 32 with MVI) in TCMR-Suspicion, 6/67 (9%; 4 with MVI) in No Inflammation and 30/35 (85.7%; 19 with MVI) in the Positive Control Cohort. Molecular antibody-mediated rejection (mAMR) was present in 39/73 (53.4%) of cases. The presence of donor-specific antibodies (DSA) at the time of the biopsy was high (127/249, 51%). Only 3 mAMR/TCMR and no pure mTCMR cases were detected in TCMR-Suspicion and No Inflammation, compared to 12 mAMR/TCMR and 10 mTCMR cases in the Positive Control Cohort (p<0.001). Even though the TCMR-specific molecular (Classifier) score differentiated between TCMR-Suspicion and No Inflammation (p=0.005), rejection phenotype scores (R2 and R3) did not (p=0.157 and 0.121). CONCLUSIONS: MMDx did not identify pure mTCMR among Isolated Tubulitis, Borderline Changes, or Isolated Arteritis, likely due to low sensitivity for TCMR-lesions. However, it identified mAMR or mAMR/TCMR, especially in cases with MVI. Subthreshold findings remain to be further studied.
ABSTRACT
Background: The routine use of donor-derived cell-free DNA (dd-cfDNA) assays to monitor graft damage in patients after kidney transplantation is being implemented in many transplant centers worldwide. The interpretation of the results can be complicated in the setting of multiple sequential kidney transplantations where accurate donor assignment of the detected dd-cfDNA can be methodologically challenging. Methods: We investigated the ability of a new next-generation sequencing (NGS)-based dd-cfDNA assay to accurately identify the source of the detected dd-cfDNA in artificially generated samples as well as clinical samples from 31 patients who had undergone two sequential kidney transplantations. Results: The assay showed a high accuracy in quantifying and correctly assigning dd-cfDNA in our artificially generated chimeric sample experiments over a clinically meaningful quantitative range. In our clinical samples, we were able to detect dd-cfDNA from the first transplanted (nonfunctioning) graft in 20% of the analyzed patients. The amount of dd-cfDNA detected from the first graft was consistently in the range of 0.1%-0.6% and showed a fluctuation over time in patients where we analyzed sequential samples. Conclusion: This is the first report on the use of a dd-cfDNA assay to detect dd-cfDNA from multiple kidney transplants. Our data show that a clinically relevant fraction of the transplanted patients have detectable dd-cfDNA from the first donor graft and that the amount of detected dd-cfDNA is in a range where it could influence clinical decision-making.
