Memory T-Cell Add Back and Prophylactic blinatumomab Post Tcrαβ Depleted Haploidentical Transplantation Results in Promising Outcomes in Pediatric Patients with High-Risk Acute Leukemia

Abstract

TCRαβ depletion of haploidentical donor (haplo) grafts allows for low rates of GVHD but delays immune reconstitution (IR) in hematopoietic cell transplant (HCT) recipients. CD45RA-negative (memory) T cells have low alloreactivity, retain specificity for viral antigens, and can enhance IR without risk of GVHD. In this prospective clinical trial (NCT03849651), we addback memory T cells following TCRαβ/CD19 depleted haploHCT for children with hematological malignancies and administer prophylactic blinatumomab (blina) for patients with B-cell ALL to mitigate risk of relapse and avoid total body irradiation (TBI).

From 2019 to 2023, 69 patients received haplo HCT (ALL: 34, AML:33, MDS:1, NHL: 1) with a median age of 8.8 years (range: 0.5-21.6). Disease status at HCT was complete remission (CR)1 (n=28), CR2 (n=29), CR≥3 (n=11), and MDS (n=1). Donors included a parent (n=64) or sibling/other (n=5). All patients received a preparative regimen consisting of fludarabine, melphalan, cyclophosphamide, thiotepa, and anti-thymocyte globulin. No post-HCT GVHD prophylaxis was used. Patients received memory T cells in 2 cohorts: a dose escalation cohort (n=29 patients with T cell dose/kg: 1 × 10⁵: n=9; 1 × 10⁶: n=10; 1 × 10⁷: n=10) and a dose expansion cohort (n=40 additional patients based on maximal effective dose of 1 × 10⁷ T cells/kg).

All patients but one engrafted with a median time to neutrophil engraftment of 10 days (range: 9-12). At a median of day +29 post-HCT (range: 22-57), 67 patients received memory T cell addback. At 4 weeks post-memory T cell addback, an increase in the mean CD3, CD8, and CD45RO+ T-cell counts (p<0.0001, p=0.001, p=0.0001) was observed. Emerging memory T cells were functional as judged by Elispot assays. TCR repertoire was comparable to donor by month 6. The incidence of acute GVHD within 28 days of memory T cells addback at dose level 1, 2 and 3 was 0% (0/9), 30% (3/10), and 2% (1/48) respectively. For the entire cohort, cumulative incidence of any grade acute GVHD, grade III-IV acute GVHD and chronic GVHD at 1 year was 26.1%, 10.1% and 4.3% respectively.

At a median of 32 days after memory T cell addback, 26 patients with ALL received blina. Of 21 evaluable patients, 61.9% (13/21) responded with B-cell aplasia (BCA). Median CD3 count prior to blina was significantly higher in patients who developed BCA (440/μL) compared to those did not (135/μL) (p<0.05). Relapses after blina occurred in 3/13 patients with BCA (CD19-pos: 2, unknown: 1), and 5/8 without BCA (CD19-pos: 3, unknown: 2).

For the entire cohort, OS and EFS at 1-year were 87.7% and 73.9%. The cumulative incidence of relapse and non-relapse mortality at 1-year were 23.2% and 2.9%.

In conclusion, our strategy led to overall promising outcomes. Addback of up to 1 × 10⁷ memory T cells/kg led to robust IR with low rates of severe GVHD and use of prophylactic blina was associated with comparable rates of relapse to TBI-containing regimens.