Background: Haploidentical hematopoietic cell transplantation (HCT) provides an alternative option for patients without HLA-matched donor. Graft-versus-host disease (GVHD), engraftment failure, and infectious complications continue to be the main causes of non-relapse mortality (NRM). We hypothesized that selective depletion of TCRαβ+ and CD45RA+ naïve T-cells subset will permit hematopoietic engraftment, while effectively reducing GVHD, and provide improved donor immune reconstitution through adoptive transfer of donor's mature NK-cells, γδ T-cells, and CD45RO+ memory T-cells.

Methods: Mobilized PBSC products were divided into two fractions in 9:1 ratio and depleted using CliniMACS device after labeling with TCRαβ and CD45RA reagents (Miltenyi Biotec, Bergish-Gladbach, Germany), respectively. The conditioning regimen consisted of fludarabine 160mg/m2 divided daily over 4 days, thiotepa 10mg/kg divided twice daily for 1 day, and melphalan 70-140mg/m2 for 1 day, in combination with total lymphoid irradiation 6Gy (n=35) or 7.5Gy (n=12) over 3 equal fractions, total body irradiation of 2Gy (n=13) or antithymocyte globuline (n=2). Short term GVHD prophylaxis for 30 days was given to 1 patient using MMF, 50 patients using tacrolimus, and 2 patients using sirolimus.

Results: We transplanted 62 patients, including 55 adults (median age, 47 years; range 20-69) and 7 children (median age, 13 years, range 7-17) with high risk AML (n=34), ALL (n=15), MDS (n=7), plasma cell neoplasm (n=2), mast cell leukemia (n=1), acute undifferentiated leukemia (n=1), and NK/T-cell lymphoma (n=1). The patients were infused with TCRαβ and CD45RA depleted graft containing a median of 6.79 x 106 (range 3.54-20.78) CD34+ cells/kg, 0.00 x 104 (range 0-0.97) CD45RA+CD3+ cells/kg, and 1.09 x 106 (range 0.15-11.67) CD45RO+CD3+ cells/kg. The TCRαβ depleted graft fraction contained a median of 0.20 x 104 (range 0-11.30) TCRαβ+ cells/kg, and 8.52 x 106 (range 0.62-30.00) TCRγδ+ cells/kg. Only 1 patient experienced primary graft failure. All others had engraftment of ANC > 500 cells/µL at a median of 10 days (range 8-22) and platelet > 20,000 cells/µL at a median of 12 days (range 8-22). There was no secondary graft failure. Six patients with high titers of donor-specific HLA antibodies (DSA) engrafted successfully after effective desensitisation with plasma exchange, rituximab, and immunoglobulin.

Twenty-two patients (35%) developed acute GVHD of grade II - IV (Gd II, n=15; Gd III, n=5; Gd IV, n=2). Three patients experienced chronic GVHD, giving 2-year cumulative incidence of 7 %.

Day 180 cumulative incidence of NRM and relapse were 24.8% (95% CI 14.3-36.8%) and 14.9% (95% CI 6.3-27.0%), respectively. Four of the 16 NRM were attributed to aGVHD. Viral reactivation included CMV (n=25), HHV-6 (n=14), EBV (n=11), and adenovirus (n=7). Fourteen patients died of infection, including 5 patients who had fatal blood stream infection (bacteria, n=3; candidemia, n=2) and 1 patient from disseminated adenovirus infection within 180 days. With a median follow-up of 298 days (range 21-1298) in surviving patients, the 2-year overall (OS), event-free (EFS), and GVHD-free/relapse-free (GRFS) survival were 58% (95% CI 37.6-73.4%), 43% (95% CI 27-57%), and 39% (95% CI 24-54%), respectively (Figure 1).

In multivariable analysis, only HCT-comorbidity index (HCT-CI) showed significant impact on OS (HR 6.24; 95% CI 2.05-19.05; p=0.0013), EFS (HR 4.80; 95% CI 1.73-13.35; p=0.0027), and RRM (HR 6.49; 95% CI 1.44-29.25; p=0.015), whereas disease risk index (DRI) impacts risk of relapse (HR 4.50; 95% CI 1.39-14.52; p=0.012). The 2-year OS, EFS, and GRFS for the subset of 30 patients (adults, n=25; children, n=5) with HCT-CI score of 0 and low/intermediate risk DRI were 68%, 68%, and 60%, respectively (Figure 2).

Conclusions: Our preliminary results suggest that RIC haplo-HCT with TCRαβ and CD45RA+ depleted grafts allowed successful allograft in high-risk patients lacking a suitable matched donor, including patients with high level of DSA. Acute GVHD was generally abortive, leading to low incidence of chronic GVHD. The best outcome is seen in patients with favorable HCT-CI and DRI. Further efforts are needed to lower the risk of death due to blood stream infection, and relapse in patients with high risk DRI, such as optimisation of anti-microbial prophylaxis or repeated doses of memory-cell donor lymphocyte infusion (DLI).

Disclosures

No relevant conflicts of interest to declare.

Author notes

*

Asterisk with author names denotes non-ASH members.

Sign in via your Institution