Figure 1.
Rationale, validation, and clinical utility of HLA-KMR. (A) Schematic representation of the rationale underlying HLA-KMR. Shown is the simultaneous presence of leukemic blasts (in red) and normal donor hematopoietic cells (in blue) in a typical sample of a patient with relapse after HSCT. In each cell, 4 representative chromosomes are depicted, with a patient-specific HLA marker (in red) and 3 hypothetical non-HLA markers (in black). Leukemic blasts in the classical relapse will be positive for both the HLA and the non-HLA markers (left). Leukemic blasts in the HLA loss relapse will be positive for the non-HLA markers but negative for the HLA markers because of a selective genomic loss of mismatched HLA as an immune escape mechanism of leukemia relapse (right). (B) Validation of the newly developed qPCR assays targeting HLA markers, representatively shown for the KMR505-A assay. Shown are the expected (x-axis) and experimentally determined (y-axis) chimerism measured by serially diluting a target-positive genomic DNA into a target-negative sample. Results from 3 independent experiments are displayed as mean (red dashes) ± standard error of the mean (black whiskers). Data for all other 10 HLA marker-specific qPCR assays developed in this study are reported in supplemental Figure 3. (C-E) Clinical utility of HLA-KMR. (C) Longitudinal posttransplant monitoring by HLA-KMR of a patient who experienced a classical relapse (time of relapse boxed in pink), alongside a patient-specific HLA marker (KMR521-DPB1 reaction, in red) and a patient-specific non-HLA marker (KMRtrack assay KMR033, in black). Note the concordance between the 2 assays at time of relapse, which identifies a classical relapse. (D) Longitudinal posttransplant monitoring by HLA-KMR of a patient who experienced an HLA loss relapse (time of relapse boxed in pink), alongside a patient-specific HLA marker (KMR522-DPB1 reaction, in red) and a patient-specific non-HLA marker (KMRtrack assay KMR017, in black). Note the discordance between the 2 assays at time of relapse, which identifies a HLA loss relapse. (E) Summary of results obtained by chimerism quantification with HLA-KMR on 20 post-HSCT relapses (18 haploidentical, 2 unrelated). As expected, HLA markers (full panel of the newly developed HLA-specific reactions, red dots) and non-HLA markers (KMRtrack assays, black dots) yield concordant results in all classical relapses (left), and discordant results in all HLA loss relapses (right). Because of the relatively low precision of qPCR in measuring high chimerism percentages, quantification of HLA and non-HLA markers in a given sample can have a result slightly different in classical relapses. HLA loss relapses can be unequivocally diagnosed only when HLA markers are negative (<3%, dashed line) and non-HLA markers are positive (>3%).