Abstract 1203

Poster Board I-225

HHCT using reduced intensity conditioning (RIC) and immunomagnetic CD3/CD19 graft depletion is of low toxicity and enables fast engraftment even with grafts of low CD34 content. Immune reconstitution may be improved compared to graft CD34 selection due to the cotransplantation of facilitating cells, CD34- progenitors, dendritic cells and natural killer (NK)-cells. A prospective multicenter phase II study of HHCT using CD3/CD19-depleted grafts after RIC with fludarabine (150-200 mg/m2), thiotepa (10 mg/kg), melphalan (120 mg/m2), OKT-3 (5 mg/day, day -5 to +14) was initiated. No post grafting immunosuppression was applied if the graft contained <5×104 CD3+ cells/kg. 60 adults with median age of 45 years (range, 19-61) have been enrolled from 2002 to 2009. Diagnoses were AML (n=37), ALL (n=8), NHL (n=6), myeloma (n=4), CML (n=3), CLL (n=1) and MDS (n=1). Patients were “high risk” because of refractory disease (n=6), cytogenetics (n=5), chemosensitive relapse (n=26) or relapse after prior HCT (allo=16, auto=6, both=1). At HHCT, 30 patients were in CR and 30 in PR. Grafts contained median of 6.8×106 (range, 3.5-22) CD34+ cells/kg, 4.0×104 (range, 0.9-44) CD3+ T cells/kg and 2.8×107 (range, 0.00-37.3) CD56+ cells/kg. Engraftment was rapid with median of 12 days to >500 granulocytes/μL (range, 9-50) and 11 days to >20.000 platelets/μL (range, 7-38). Incidence of grade II-IV acute GVHD and chronic GvHD (cGvHD) was 47% and 15%, respectively. Non-relapse-mortality on day 100 was 25% and 44% for the whole study period. Current overall survival (OS) is 18 of 60 patients (30%) with median follow-up of 525 days (range, 21-1542) of patients alive resulting in a Kaplan-Meier estimate 1-year OS of 41% and 2-year OS of 24%. Kaplan-Meier estimate 1-year OS was 40 % in AML, 38 % in ALL and 67% in NHL patients. No positive impact of KIR-mismatch on survival even in the subgroup of patients with AML was observed. Patients with cGVHD had a trend toward better survival with 56% vs. 39% (p=0.09). 14 patients died because of infections. Detailed immune reconstitution was analyzed in 24 patients. NK-cell engraftment was fast with median of 248 CD16+56+CD3- cells/μl (range, 1-886) on day 20. Increased natural cytotoxicity receptors (NKP30, NKP44, NKP46) and NKG2A, but decreased NKG2D and KIR-expression was observed on NK-cells in the first 3 months. T-cells regenerated delayed with median of 191 CD3+ cells/μl (range, 38-799) on day 100. A slow reconstitution of CD8+ and CD4+ T-cells with median of 66 CD8+ (range, 8-170) vs. 70 CD4+ cells/μl (range, 12-301) on day 100 and 157 (range, 32-379) vs. 181 cells/μl (range, 19-980) on day 400 was observed. The subset of memory T-cells regenerated faster compared to naïve T-cells with median of 25 CD4+45RA+ (range, 4-109) vs. 80 CD4+45RO+ cells/μl (range, 0 to 255) and 46 (range 7-191) vs. 164 cells/μl (range, 66-323) on days 80 and 250, respectively. T-cell repertoire was skewed with oligoclonal T-cell expansion to day 100 and normalization after day 200. B-cell reconstitution was slow with median of 8 (range, 0-407) CD19+20+ cells/μl on day 100. 6 of these 24 patients received donor-lymphocyte-infusions for relapse or mixed chimerism resulting in acceleration of immune recovery in T- and NK-cells. In conclusion, HHCT using RIC and CD3/CD19 depleted grafts is of low toxicity and allows fast engraftment even with low doses of CD34 cells. Overall survival seems promising in a high risk patient cohort. Recovery of NK cells occurs early and fast. KIR receptor expression remains low in the first 3 months. T- and B-cell reconstitution is delayed but seems faster compared to published data after CD34 selection. To evaluate the treatment protocol earlier during the course of disease a new study has just been initiated.


Off Label Use: The use of Fludarabine, Thiotepa, Melphalan and OKT-3 in the conditioning is off-label-use.

Author notes


Asterisk with author names denotes non-ASH members.