Allogeneic natural killer cells may exert cytotoxic activity against HLA-nonidentical tumor cells. This effect may be exploited after T-cell depleted haploidentical hematopoetic stem cell transplantation (HSCT) by using donor NK-cell infusions (NK-DLI) as adoptive immunotherapy.

In a prospective phase II study in 2 centers we treated 15 patients with NK-DLI preemptively with the goals to demonstrate feasibility and to explore whether such infusions would contribute to stabilize engraftment and exert graft versus leukemia activity without causing GvHD.

Patients were adults (6) or children (8) receiving transplants from haploidentical donors to treat AML (7) ALL (5) Hodgkin lymphoma (2) or sarcoma (1). Donors were haplotype mismatched siblings in 3 and parents in 12.

Donor NK cells were selected from unstimulated leukapheresis using immunomagnetic T cell depletion with anti-CD3 and NK cell enrichment with anti-CD56 coated microbeads on the CliniMACS ® device. Per protocol, patients in center A received NK-DLI on days +40 and +100 and in center B on days +3, +40 and +100. NK-DLI were either freshly infused or cryopreserved and thawed.

Center ACenter B
Adults / Children 6/2 0/7 
Disease: AML/ALL/HD/oth 6/1/1/0 1/4/1/1 
Number of NK-DLI per patient 1/2/3 2/5/1 2/3/2 
Median NK-cell recovery (%) 62 37 
NK-cell dose (10e7/kg) (median, range) 1.2 (0.5–3.4) 1.4 (0.66–3.23) 
T-cell dose (10e4/kg) (median, range) 0.2 (0–7.2) 0.23 (0–5.3) 
Events: graft failure / progression / GvHD (III–IV) 2/2/1 1/2/1 
Alive / Dead 3/5 5/2 
Cause of death: graft failure/progression/GvHD 2/2/1 0/2/0 
Center ACenter B
Adults / Children 6/2 0/7 
Disease: AML/ALL/HD/oth 6/1/1/0 1/4/1/1 
Number of NK-DLI per patient 1/2/3 2/5/1 2/3/2 
Median NK-cell recovery (%) 62 37 
NK-cell dose (10e7/kg) (median, range) 1.2 (0.5–3.4) 1.4 (0.66–3.23) 
T-cell dose (10e4/kg) (median, range) 0.2 (0–7.2) 0.23 (0–5.3) 
Events: graft failure / progression / GvHD (III–IV) 2/2/1 1/2/1 
Alive / Dead 3/5 5/2 
Cause of death: graft failure/progression/GvHD 2/2/1 0/2/0 

NK-cell collection and ex vivo purification was successful in all donors. Outcome is shown in the Table. Of note: One patient in each center developed severe (grade III / grade IV) GvHD respectively. These 2 patients had received the highest T-cell dose. The remaining patients tolerated NK-DLI well and did not develop GvHD. Two patients in Center A experienced graft failure after NK-DLI in spite of NK-alloreactivity in GvHD direction.

In conclusion: highly purified NK cells may be infused in recipients of haploidentical HSCT recipients. None of the patients receiving < 10e4/kg CD3+ cells developed GvHD. Further studies are needed to determine the optimal dose and timing of NK-DLI for anti-tumor activity in this setting.

Disclosure: No relevant conflicts of interest to declare.

Author notes

*

Corresponding author

Sign in via your Institution