Allogeneic hematopoietic stem cell transplantation has been proven to be one of few curative treatment options for patients with hematological malignancies. One drawback to this procedure is the development of graft-versus-host disease (GVHD), which can lead to high rates of morbidity and mortality. It is well established that donor T cells are primarily responsible for anti-host activity. Removal or delayed-administration of donor T cells reduces the incidence or severity of GVHD, but is often associated with an increased risk of tumor relapse due to lack of donor T cells that mediate graft-versus-leukemia (GVL) effects. Utilizing CDR3-size spectratype analysis we have focused on identifying host- and tumor-reactive T cells. In this study, a MHC-matched, miHA-mismatched murine transplantation model (B10.BR→CBA) was used to characterize CD8+ T cell repertoire anti-host responses as well as anti-tumor responses to a host-derived myeloid leukemia cell line (MMC6). This model is particularly relevant in that it mimics matched-sibling donor transplants for AML. Spectratype analysis has shown that 14 of 20 Vβ families exhibit biased CDR3-size usage in the B10.BR anti-CBA CD8+ T cell repertoire. In response to MMC6, 7 of 20 Vβ families exhibit biased CDR3-size usage, with Vβ4 and Vβ13 uniquely skewed in the anti-tumor response. We sought to test the ability of spectratype-identified Vβ families, uniquely skewed in the B10.BR CD8+ T cell anti-MMC6 response, to mediate GVL effects in the absence of GVHD in tumor challenged transplant recipients. To this end lethally irradiated (11Gy;split dose) CBA recipients were transplanted with B10.BR ATBM alone, ATBM + B10.BR CD8+ T cells, or ATBM + either B10.BR CD8+ Vβ4+ or Vβ13+ T cells. Duplicate groups of mice received the same transplant conditions and were challenged with MMC6 on day +1 post-transplant. All mice were monitored for symptoms of GVHD and tumor growth. Mice receiving MMC6 in the absence of T cells succumbed to tumor within 30 days post-transplant. Tumor challenged mice receiving CD8+ T cells displayed a statistically significant GVL effect (p<0.0001), however, all eventually succumbed to GVHD. Mice receiving either Vβ4+ or Vβ13+ T cells along with MMC6 displayed minimal GVL effects, with all mice succumbing to tumor challenge. As expected based on the spectratype analysis, mice receiving either Vβ4+ or Vβ13+ T cells in the absence of tumor challenge did not develop any signs of GVHD and survived until termination of the experiment. In order to test the possibility of enhancing the GVL potential, duplicate groups of mice received the same transplant conditions and MMC6 tumor challenge, as above with an additional group that was co-transplanted with CD8+Vβ4+ and Vβ13+ T cells. This combined transplantation of CD8+Vβ4+ and Vβ13+ T cells did not mediate severe lethal GVHD (80% survival). However, together, CD8+Vβ4+ and Vβ13+ T cells significantly prolonged survival (p=0.001) of tumor challenged recipients. Taken together, the results of this study provide proof of principle to support the hypothesis that Vβ spectratype analysis can be used to identify the donor GVH- and GVL-reactive T cells. Furthermore, manipulation of the donor inoculum to enrich for the GVL-reactive and/or deplete the GVH-reactive Vβ families can provide an effective way to separate GVHD from GVL effects.

Disclosures: No relevant conflicts of interest to declare.

Author notes

Corresponding author