Safety and Feasibility of Administration of High Doses of Ex Vivo Expanded NK Cells for Prevention of Disease Relapse after Transplantation for Patients with Myeloid Malignancies - Final Results of a Phase I Clinical Trial

Background: Disease relapse remains the most common cause of treatment failure after transplantation. To enhance anti-tumor effect of the graft we explored the use of high doses of donor-derived ex vivo expanded NK cells administered after haploidentical stem cell transplantation (HaploSCT) with the ultimate goal to decrease relapsed rate post-transplant.

Methods: We aimed to study safety and determine the maximum tolerated dose (MTD) of high doses of mbIL-21 ex vivo expanded NK cells in a phase I clinical trial (clinicaltrials.gov NCT01904136) administered post-transplant in patients with myeloid malignancies (AML, CML, MDS). We hypothesized that infusion of mature NK cells would compensate for the lower NK-cell numbers and poor function previously observed by our group in the first month post-transplant. Patients received conditioning with melphalan 140mg/m², fludarabine 160mg/m² and 2GyTBI, and GVHD prophylaxis with post-transplant cyclophosphamide, tacrolimus and mycophenolate (Gaballa S, et al. Cancer. 2016). All patients had a bone marrow graft. NK cells were generated from peripheral blood mononuclear cells of the same donor with infusions on Days -2, +7 and on/after +28. The first infusion was with fresh and the other two were with cryopreserved NK cells. Dose escalation was planned in cohorts of 2 patients starting at 1x10⁵ to 1x10⁹ NK cells/Kg. Having an NK-cell alloreactive donor or certain KIR genotype was not a requirement to participate in this study, although these characteristics were evaluated in all patients.

Results: Thirteen patients were enrolled. Eight patients had AML (7 in CR1, 4 with high-risk cytogenetics, 3 had primary induction failure AML and 3 had +minimal residual disease (MRD) at transplant, and one FLT3+ AML in CR2 with persistent MRD by flow cytometry), and 5 had CML (4 in second chronic phase, 2 with prior CNS disease, one had associated MDS with monosomy 7, and 2 who failed multiple TKIs). The median age was 41.5 years (range 18-60). Five patients were males and 8 females. The NK-cell dose escalation was as follows: 1x10⁵/kg (N=2), 1x10⁶/kg (N=3), 1x10⁷/kg (N=3), 3x10⁷/kg (N=2) and 1x10⁸/kg (N=2). One patient was treated with 1x10⁴/kg (Dose -1). All patients received the 3 planned infusions except one. Median purity of the NK-cell product was 98.98%. All patients achieved primary engraftment (100%) with 100% donor chimerism except one patient (who received 1x10⁴/kg) who had secondary graft failure with concurrent parainfluenza pneumonia and died of treatment-related mortality (TRM). The median time to neutrophil and platelet engraftment was 18 and 26 days, respectively. Of 12 patients evaluable for aGVHD, the maximum grade was II in 7 patients. No grade III-IV aGVHD or cGVHD was observed. Only 5/12 patients had CMV reactivation (41.6% compared with 71% in retrospective data with the same treatment without NK cells), while none developed BK virus hemorrhagic cystitis. All patients (N=12) achieved remission with negative MRD after transplant. One patient treated NK cells at 1x10⁵/kg/dose relapsed, received salvage treatment and was alive at last follow-up. All other patients are alive and in complete remission (N=11) after a median follow-up of 12.8 months (range 6-25.1). Compared with patients treated on a previous clinical trial with the same conditioning without NK cells (AML in CR1/2 and CML in CP), the patients who received NK cells had marked improvement in NK-cell function and cytotoxicity, as well as a significantly increase in INF gamma and TNF alpha production. A lower relapse rate and improved survival was observed compared with the retrospective cohort of patients treated without NK cells, although not statistically significant (p=0.21) (Figure 1).

Conclusions: Doses up to 3x10⁸/kg total dose of mbIL-21 ex vivo expanded NK cells obtained from the same donor can be safely administered in the early post-transplant period after HaploSCT. This was the maximum feasible dose to be manufactured in the current system, while no toxicity or increase in GVHD was observed. Three infusions of 1x10⁸ NK cells per kg will be administered in the phase II study

Figure 1

View largeDownload slide

Relapse and survival for patients treated with the same conditioning regimen with and without NK cells.

Figure 1

View largeDownload slide

Relapse and survival for patients treated with the same conditioning regimen with and without NK cells.

Close modal