Impaired B Cell Clonotype Diversification After Allogeneic Hematopoietic Cell Transplantation Predicts Graft-Versus-Host Disease

Graft-versus-host disease (GVHD) is a detrimental complication of allogeneic hematopoietic cell transplantation (HCT) associated with impaired quality-of-life and decreased overall survival. Methods for predicting GVHD may enable use of pre-emptive therapeutic maneuvers to prevent or diminish clinical GVHD symptoms. Studies of GVHD in murine allograft models have shown that GVH reactions are associated with structural and functional derangements of lymphoid tissues, including lymph nodes, spleen, thymus, and bone marrow. Because these organs provide the niche space for B and T cell diversification, these findings suggest that GVHD, even in sub-clinical stages, may impair diversification of the immune repertoire. We evaluated post-HCT B cell immunoglobulin heavy chain (IGH) gene diversity in 42 patients with chronic lymphocytic leukemia who received unmanipulated mobilized peripheral blood grafts following total lymphoid irradiation and anti-thymocyte globulin (TLI/ATG). Twenty-one patients also received 4 weekly doses of Rituximab beginning 2 months post-HCT. Post-HCT peripheral blood samples were prospectively archived at pre-scheduled intervals. The IGH repertoire was quantified in over 400 samples using the LymphoSIGHT(TM) method, which employs consensus IGH V and J segment primers to universally amplify rearranged IGH genes, followed by massively parallel high-throughput sequencing (IGH-HTS) and bioinformatic analysis. Donor graft material was analyzed in the same manner. A median 1.8x10e6 (2.6x10e4 – 2.7x10e6) leukocyte genomes per sample with a median 6.4x10e5 (12 – 7.6x10e5) IGH molecules were analyzed. Compared to healthy donors, patients receiving only TLI/ATG exhibited significantly decreased diversity at days +30 and +90 (p<0.001) (Fig 1A), whereas IGH diversity remained significantly decreased at days +30, +90, +180, +270 (each p<0.001) and +365 (p<0.05) in those who received post-HCT Rituximab (Fig 1B). Clonotype analysis confirmed significant (p=0.009) but incomplete depletion of adoptively transferred IGH clones in those receiving post-HCT Rituximab (Fig 1C). The incidences of relapse (38 vs. 52%) and new-onset GVHD (28 vs. 33%) were similar in the two groups. Amongst patients at risk for new onset GVHD (after censoring for relapse and Rituximab treatment), IGH diversity was significantly decreased at day +30 (54 +/- 13 vs. 18818 +/- 11576 unique IGH clones; p=0.03) and day +90 (2693 +/- 2657 vs. 40316 +/- 16480 unique IGH clones; p=0.02) in those who subsequently developed GVHD in comparison to those who did not (Fig 1D). IGH-HTS analysis of the adoptively transferred and reconstituting B cell repertoires after allo-HCT may provide a method for predicting new onset GVHD and for identifying pathogenically involved adoptively transferred B cell clones.