Memory T-cell enriched haploidentical transplantation with NK cell addback results in promising long-term outcomes: a phase II trial

Abstract

Background: Relapse remains a challenge after transplantation in pediatric patients with hematological malignancies. Myeloablative regimens used for disease control are associated with acute and long-term adverse effects. We used a CD45RA-depleted haploidentical graft for adoptive transfer of memory T cells combined with NK-cell addback and hypothesized that maximizing the graft-versus-leukemia (GVL) effect might allow for reduction in intensity of conditioning regimen.

Methods: In this phase II clinical trial (NCT01807611), 72 patients with hematological malignancies (complete remission (CR)1: 25, ≥ CR2: 28, refractory disease: 19) received haploidentical CD34 + enriched and CD45RA-depleted hematopoietic progenitor cell grafts followed by NK-cell infusion. Conditioning included fludarabine, thiotepa, melphalan, cyclophosphamide, total lymphoid irradiation, and graft-versus-host disease (GVHD) prophylaxis consisted of a short-course sirolimus or mycophenolate mofetil without serotherapy.

Results: The 3-year overall survival (OS) and event-free-survival (EFS) for patients in CR1 were 92% (95% CI:72-98) and 88% (95% CI: 67-96); ≥ CR2 were 81% (95% CI: 61-92) and 68% (95% CI: 47-82) and refractory disease were 32% (95% CI: 11-54) and 20% (95% CI: 6-40). The 3-year EFS for all patients in morphological CR was 77% (95% CI: 64-87) with no difference amongst recipients with or without minimal residual disease (P = 0.2992). Immune reconstitution was rapid, with mean CD3 and CD4 T-cell counts of 410/μL and 140/μL at day + 30. Cumulative incidence of acute GVHD and chronic GVHD was 36% and 26% but most patients with acute GVHD recovered rapidly with therapy. Lower rates of grade III-IV acute GVHD were observed with NK-cell alloreactive donors (P = 0.004), and higher rates of moderate/severe chronic GVHD occurred with maternal donors (P = 0.035).

Conclusion: The combination of a CD45RA-depleted graft and NK-cell addback led to robust immune reconstitution maximizing the GVL effect and allowed for use of a submyeloablative, TBI-free conditioning regimen that was associated with excellent EFS resulting in promising long-term outcomes in this high-risk population. The trial is registered at ClinicalTrials.gov (NCT01807611).

Keywords: CD45RA depletion; Graft-versus-leukemia; Haploidentical; Immune reconstitution; Memory T cell; NK cell; Non-TBI regimen; Pediatric hematological malignancies; T-cell depletion; Transplantation.

Fig. 1

Transplant procedure, engraftment, and immune reconstitution. A Outline of transplant procedure. B Median cell doses of infused grafts. Values for CD3 + /CD45RA + cells that were ‘0’ were set to ‘1’ (n = 72). C Cumulative incidence of neutrophil and platelet engraftment (n = 72). D Median CRP levels during the first 21 days post-transplant, shaded area: CI 95% (n = 72). E–K Immune reconstitution in the first 12-months post-transplant. E Median T- (CD3, CD4, CD8), NK- (CD56), and B- (CD19) cell counts, shaded area: 95% CI (n = 69). F–G Percentage of (F) memory (CD3 + /CD45RO +) and naïve and TEMRA (CD3 + /CD45RO-) T-cell subsets (n = 64) and (G) naïve (CD45RO-/CCR7 +), CM (CD45RO + /CCR7 +), EM (CD45RO + /CCR7-), and TEMRA CD45RO-/CCR7-) T cells in CD4 and CD8 T-cell subsets in recipients pre- (n = 54–56) and post (n = 52–53) transplant and donors (n = 54–55). H TCR Vβ spectratyping (n = 38). I Recent thymic emigrants measured as T-cell receptor excision circles/mL (n = 38). Solid horizontal line is median in each violin plot in panels H, I. J–K Elispot assays post HCT, median and individual values are shown, data is plotted as SFC/5 × 10⁵. J Positive (SEB) and negative (DMSO) controls (n = 34). K Frequency of CMV-specific T cells with (n = 10) or without CMV viremia (n = 13) in CMV-seropositive/seropositive or -seropositive/ seronegative donor/recipient pairs, and Adv-specific T cells with (n = 13) or without (n = 13) Adv-positive stool samples. Statistical analysis for panels H–K: Two-way ANOVA, ns = not significant, *P < 0.05, **P < 0.01, ***P < 0.001, ****P < 0.0001. L Elispot assays of donors (n = 20), paired t-test (versus negative Co), ***P < 0.001, ****P < 0.0001. ANC, Absolute neutrophil count; PLT, Platelet; CRP, C-reactive protein; NK, natural killer; CM, central memory; EM, effector memory; TEMRA, Terminally differentiated effector memory; Pre, pre transplant recipients, D, donors, TREC, T-cell receptor excision circles; TCR, T-cell repertoire; CMV, cytomegalovirus; Co, control, pos, positive; neg, negative; SFC, spot-forming cells; SEB, Staphylococcal enterotoxin B; DSMO, dimethyl sulfoxide; D + , Donor positive; D −, Donor negative; Adv, Adenovirus

Fig. 2

Relapse and survival. A Probability of OS and EFS for the full cohort. B Probability of OS and EFS for CR cohort. C Probability of OS and EFS for patients with active disease. D Probability of OS for patients in CR1, ≥ CR2. E Probability of EFS for patients in CR1, ≥ CR2. F Probability of OS for patients with non-detectible and detectible disease at time of transplant. G Probability of EFS for patients with non-detectible and detectible disease at time of transplant. (Fig F,G,- Six patients excluded due to unknown minimal residual disease (MRD) status) H Cumulative incidence of relapse and NRM for full cohort. I Cumulative incidence of relapse for patients in CR1, ≥ CR2, active disease (1 patient with active disease excluded due to death prior to engraftment). OS, overall survival; EFS, event-free survival; NRM, non-relapse mortality; CR, complete remission, CIR, cumulative incidence of relapse. 95% CI is listed in parenthesis

Fig. 3

GVHD and survival. A Cumulative incidence of all aGVHD and grade III-IV aGVHD. B Cumulative incidence of all cGVHD and moderate-severe cGVHD. C Probability of OS from onset of grade III-IV aGVHD. D Probability of OS from onset of cGVHD. aGVHD, acute graft-versus-host disease; cGVHD, chronic graft-versus-host disease, OS, overall survival. 95% CI is listed in parenthesis