Use of High-Throughput Sequencing (HTS) of TCRß to Determine the Kinetics of Graft-Versus-Lymphoma (GVL) Effect and T-Cell Repertoire Profiles after Allogeneic Transplant

Introduction:While allogeneic transplant using non-myeloablative preparative regimen provides a viable alternative with low TRM, the kinetics of GVL effect and the T-cell reconstitution may differ from the myeloablative transplant. In general, the donor cell engraftment is slower using a non-myeloablative regimen and a delayed GVL effect is expected. In this study, we applied the TCRß HTS to determine (1) the kinetics of the GVL effect by quantifying the tumor cell burden prior to and after transplant and (2) the pace of the T-cell reconstitution after transplant.

Method: This report included a cohort of 24 patients with advanced stage mycosis fungoides or Sézary Syndrome who underwent allogeneic transplant using a non-myeloablative regimen with total skin electron beam therapy (TSEBT, 24-36 Gy), total lymphoid irradiation (TLI, 8 Gy) and anti-thymocyte globulin (ATG). All patients received G-CSF-mobilized peripheral blood hematopoietic cells with a median donor CD34+ cell dose of 6.9 x 10⁶/kg (range 2.0-12.4) and a median donor CD3+ cell dose of 278.1 x 10⁶/kg (range 134.4-631.0). The unique malignant T-cell clonotype of each individual patient was identified from diagnostic blood/skin samples as the single dominant sequence by TCRß HTS. Blood samples were collected prior to and at different time points after transplant. DNA extracted from PBMC corresponding to approximately 200,000 genomes was used for HTS (Sci. Transl. Med.5:214ra171, 2013) (ImmunoSEQ, Adaptive Biotech). The tumor cell burden (including minimal residual disease, MRD) was expressed as percentage of the malignant clonotype found in the entire T-cell repertoire.

Results: Prior to transplant, 21 patients (88%) had detectable disease by TCRß HTS in the blood (<1%: 10 patients, 1-5%: 3 patients, >5%: 8 patients), and 3 patients had no detectable disease in the blood. The percentage of malignant clone decreased in 19 of these 21 patients at day+30 post-transplant, and 2 patients showed stable minimal disease (0.03 and 0.30%, respectively). The reduction of tumor burden was most pronounced in patients with >5% involvement prior to transplant. In these 8 patients, the pre- and day+30 post-transplant disease burden decreased from 77.9 to 0.9%, 23.6 to 9.0%, 21.1 to 8.8%, 11.8 to 1.9%, 11.3 to 0.9%, 10.0 to 0.1%, 6.0 to 1.5% and 5.5 to 0.2%, respectively. This immediate post-transplant GVL effect was not associated with full donor T-cell engraftment (donor CD3+ >95%). The donor T-cell chimerism at day+30 was 94%, 79%, 1%, 90%, 93%, 93%, 23% and 91% for these 8 patients, respectively. Subsequently, 11 of the 24 patients achieved sustained molecular remission in the blood with a median time to achieve molecular remission of 60 days (range 30-540). Patients with full donor T-cell chimerism (n=16) had a higher chance of achieving molecular remission in the blood than those with mixed donor T-cell chimerism (69% vs 0%, p = 0.002). Of these 11 patients who achieved molecular remission in the blood, 8 also achieved molecular remission in the skin at the same time. Of the remaining 3 patients, 2 achieved molecular remission in the skin more than 4 months after achieving remission in the blood, while one patient has yet to achieve molecular remission in the skin.

We then analyzed the “T-cell repertoire” at different time points post-transplant by assessing the number of unique T-cell clonotypes in each blood sample from 14 patients who had at least 1-year follow-up. While the size of the T-cell repertoire varied significantly between patients, we observed an overall upward trend within individual patients after transplant (Table). The size of T-cell repertoire did not correlate with the donor CD34+ or CD3+ cell dose in the allograft.

Conclusion:By using an extremely sensitive and specific TCRß HTS, we have shown an immediate post-transplant GVL effect in which a full donor engraftment was not required, followed by a subsequent sustained GVL effect that may depend on full donor T-cell engraftment. Our results suggest a different kinetics of GVL effect in different compartments (blood vs skin). We also demonstrate continued expansion of T-cell repertoire profile after allogeneic transplant.