Quantitative characterization of T-cell repertoire in allogeneic hematopoietic stem cell transplant recipients

Abstract

Allogeneic hematopoietic stem cell transplantation (HSCT) is one of curative treatment options for patients with hematologic malignancies. Although GVHD mediated by the donor's T lymphocytes remains the most challenging toxicity of allo-HSCT, graft-versus-leukemia (GVL) effect targeting leukemic cells, has an important role in affecting the overall outcome of patients with AML. Here we comprehensively characterized the TCR repertoire in patients who underwent matched donor or haplo-cord HSCT using next-generation sequencing approach. Our study defines the functional kinetics of each TCRA and TCRB clone, and changes in T-cell diversity (with identification of CDR3 sequences) and the extent of clonal expansion of certain T-cells. Using this approach, our study demonstrates that higher percentage of cord-blood cells at 30 days after transplant was correlated with higher diversity of TCR repertoire, implicating the role of cord-chimerism in enhancing immune recovery. Importantly, we found that GVHD and relapse, exclusive of each other, were correlated with lower TCR repertoire diversity and expansion of certain T-cell clones. Our results highlight novel insights into the balance between GVHD and GVL effect, suggesting that higher diversity early after transplant possibly implies lower risks of both GVHD and relapse following the HSCT transplantation.

Figure 1

Comparison of diversity index of samples. (a) In all patients, the diversity for TCRA (a) and TCRB (b) before transplantation were compared with that after transplantation.

Figure 2

Reconstruction of TCR repertoires observed in the haplo-cord transplant patients at day 100. (a) Distribution of patients according to the median of percentage of cord-derived cells at day 30 or day 100 after haplo-cord transplant (n=9). The diversity of TCRA (b) and TCRB (c) of patients with >6% (n=4) cord-derived cells at day 30 were compared with that of patients with ⩽6% cord-derived cells (n=5). The diversity of TCRA (d) and TCRB (e) of MD patients were compared with that of haplo-cord patients.

Figure 3

Comparison of the proportion of the top 10 CDR3 sequences and diversity between non-GVHD and GVHD patients. In all patients, the proportion of top 10 CDR3 of TCRA (a) and TCRB (b) of non-GVHD (n=10) were compared with that of GVHD patients (n=11). In the non-relapsed group, the proportion of top 10 CDR3 of TCRA (c) and TCRB (d) of GVHD patients (n=6) were compared with that of non-GVHD patients (n=7). In the relapsed group, the proportion of top 10 CDR3 of TCRA (e) and TCRB (f) of GVHD (n=5) were compared with that of non-GVHD (n=3). In the non-relapsed group, the diversity of TCRA (g) and TCRB (h) of non-GVHD patients (n=7) were compared with that of GVHD patients (n=6). In the relapsed patients, the diversity of TCRA (i) and TCRB (j) of GVHD (n=5) were compared with that of non-GVHD (n=3) patients.

Figure 4

Comparison of the TCR diversity between relapse and non-relapse patients. In all patients, the diversity of TCRA (a) and TCRB (b) of relapsed patients (n=7) were compared with that of non-relapsed patients (n=13). In non-GVHD patients, the diversity of TCRA (c) and TCRB (d) of non-relapsed patients (n=7) were compared with that of relapsed patients (n=3). In GVHD patients, the diversity of TCRA (e) and TCRB (f) of relapsed patients were compared with that of non-relapse patients.