Selective T-Cell Depletion Improves Engraftment, Reduces GVHD in Pediatric/AYA Leukemia

Sarah M. Lo, MD

Eliminating T cells carrying alpha and beta chains of the T-cell receptor reduced the risk for graft-versus-host-disease (GVHD) while producing “excellent” engraftment kinetics for young patients with acute leukemia who underwent hematopoietic stem cell transplant (HSCT).

Sarah M. Lo, MD, pediatric hematologist/oncologist, Children’s Hospital of Philadelphia (CHOP), presented results at the 2018 American Society of Pediatric Hematology/Oncology Conference from an ongoing, prospective, phase II study of 36 pediatric and young adult patients. She said that, so far, investigators have been able to limit GVHD without the use of prophylactic immunosuppression.

“Our strategy of T cell depletion in unrelated donor and partial-matched unrelated donor thus far appears promising and offers opportunities for augmentation of antileukemic benefits, as well as enhancement of immune reconstitution, as well as a decreased risk of viral reactivation” she said.

“In future studies, we can imagine that this might pose a potential platform for additional immunotherapies to prevent relapse, as well as to potentially accelerate immune reconstitution,” Lo added.

HSCT has the potential to cure patients with high-risk leukemias, but only 25% to 30% of patients have human leukocyte antigen (HLA)-identical siblings, who are ideal donors. Patients who receive stem cells from partially-matched or unrelated donors (URD) are at increased risk for acute or extensive chronic GVHD.

Investigators have tried various methods of T-lymphocyte depletion to combat GVHD, but T-lymphocyte depletion can lead to increased risk for nonengraftment and relapse. Lo et al hypothesized that TCRα/β/CD19 depletion could reduce the risk for GVHD while boosting the antileukemic effect by maintaining mature donor-derived alloreactive NK cells and γ/δ(+) T cells.

All patients in the study had HSCT from HLA-identical siblings (n = 8) or URDs (n = 28) from October 2014 to May 2017. Fifteen patients underwent a myeloablative preparative regimen with targeted busulfan, while 21 patients received total body irradiation with 1200 cGy in 6 fractions.

“We did not give any immune suppression to any patient posttransplant as prophylaxis,” Lo said.

Patients were transplanted with mobilized peripheral blood stem cells that underwent TCRα/β/CD19 depletion at the CHOP Cellular Immunotherapy Lab. The median dose of CD34(+) cells was 10.1 x 10⁶, median dose of α/β(+) CD3(+) cells was 5.63 x 10⁶, and median dose of B cells was 8.95 x 10⁴.

The median patient age was 11 years (range, 1.3-21.7). Twelve patients were diagnosed with acute lymphoblastic leukemia (9 B-cell and 3 T-cell) and 24 were diagnosed with acute myeloid leukemia (AML; 21 primary and 3 secondary).

All patients had neutrophil engraftment at a median of 13 days (IQR, 8-30) posttransplant. At a median of 16 days (IQR, 12-40) posttransplant, 94% of patients demonstrated platelet engraftment. Platelet engraftment was 100% among those who received stem cells from HLA-identical donors and 93% for those who received stem cells from URDs.

Lo said that the 2 patients who did not have platelet engraftment had complications early in their course of transplantation. One experienced graft failure and the other developed sepsis.

Fewer than half (47%) of patients experienced viral reactivation. Twenty-two percent of patients experienced grade 1/2 GVHD, while 1 patient (3%) had grade 3. Lo said all cases of acute GVHD were of the skin.

For chronic GVHD, 14% of cases involved the skin while 1 patient (3%) had gut GVHD. GVHD was limited in 11% of patients and extensive in 3%. “Overall, there didn’t seem to be any significant differences based on ATG [antihuman T-lymphocyte immune globulin] status in terms of chronic GVHD,” said Lo.

Eight percent of patients remained on immunosuppressive therapy after 1 year.

Treatment-related mortality was 22% after 3.5 years. Seventeen percent of patients experienced sepsis and 6% experienced acute respiratory distress syndrome. The survival probability was 8.3% (95% CI, 2.8-23.7) at 100 days, 11.7% (95% CI, 4.5-28.5) at 12 months, and 16.9% (95% CI,7.1-37.3) at 24 months.

Overall, 10 (28%) patients relapsed, 7 (19%) in the AML group and 3 (25%) in the ALL group. Median overall time to relapse was 147 days (range, 57-625) post-HSCT.

Across the study population, 31% (n = 11) of patients died. Deaths were twice as common among AML patients (38% vs 17%), but Lo said the difference was not significant considering the small patient populations. The median overall time to death was 298 days (range, 54-803) post-HSCT.

GVHD-free survival was 86.1% (95% CI, 69.8-94.0) at 100 days, 62.2% (95% CI, 43.7-76.1) at 12 months, and 41.3% (95% CI, 24.0-57.9) at 24 months. Relapse-free survival was 86.1% (95% CI, 69.8-94.0) at 100 days, 65.4% (95% CI, 47.1-78.8) at 12 months, and 51.1% (95% CI, 31.0-68.1) at 24 months.

Leukemia- and GVHD-free survival was 66.7% (95% CI, 48.8-79.5) at 100 days, 45.8% (95% CI, 28.7-61.3) at 12 months, and 35.8% (19.0-51.2) at 24 months.

Overall survival at 100 days was 88.9% (95% CI, 73.1-95.7), 80.3% (95% CI, 62.9-90.1) at 12 months, and 67.9% (95% CI, 47.5-81.7) at 24 months.

Lo S, Wang Y, Monos D, et al. Unrelated or partially matched related donor hematopoietic stem cell transplant (HSCT) with TCRα/β/CD19 depletion for children and young adults (CAYA) with acute leukemias. Presented at: the 2018 ASPHO Conference; May 2-5, 2018; Pittsburgh, PA. Poster 1012.

<<< 2018 ASPHO Conference