With the introduction of the concept of alloreactive Natural Killer (NK) cells, haploidentical transplantations are gaining increasing interest in the treatment of malignant diseases. While it has been shown in patients that alloreactive NK cells can exert an effective Graft-versus-Leukemia (GvL) effect, data in murine haploidentical transplantation models suggest that alloreactive NK cells also facilitate engraftment and reduce the risk of Graft-versus-Host Disease (GvHD). From these findings, we hypothesized that haploidentcial transplantation using Reduced Intensity Conditioning (RIC) regimens should be feasible in a clinical setting if the graft contains not only stem cells, but also large numbers of NK cells. Therefore, we have established a clinical method which allows the large scale depletion of CD3+ T-lymphocytes from mobilized peripheral haploidentical stem cells (PBSC’s). We have initiated a clinical protocol using this approach in a patient population known to be at high risk for transplant-related toxicity and mortality (TRM). All patients suffered from chemorefractory leukemia/lymphoma or had a relapse after a previous myeloablative allogeneic transplant.

As of June 2004, 18 pediatric patients have been transplanted. Seven patients had a chemorefractory disease and 11 patients had a relapse after a previous myeloablative allogeneic transplant. For all patients, haploidentical donors were mobilized using a combination of GM-CSF/G-CSF. Depletion of CD3+ T-cells was performed using the Clinimacs device. The RIC regimen consisted of Fludarabine (200 mg/m2), Thiotepa (10 mg/kg) and Melphalan (60 mg/m2 or 120 mg/m2, depending on the patient’s clinical pretransplant condition). As an additional rejection prophylaxis, anti-CD3 antibody MuronomabR was given. A short course of posttransplant CellceptR was given for GvHD prophylaxis. For in vivo B-cell depletion, patients received one dose of RituxanR at day 0. The number of transplanted CD34+ stem cells was 19.3 x 106/kg (range 2.2 – 37.5) and the number of coinfused CD56+ NK cells was 103 x 106/kg (8.4 – 288). The number of residual CD3+ T-cells in the graft was 1.38 x 105/kg (0.07 – 4.5). Engraftment was rapid in 17 patients with 10.2 days (range 7–12) to reach an ANC>500. One patient did not engraft and one patient rejected his graft at day +12. Seventeen patients were available for chimerism analysis and in all patient, a 100% donor chimerism was seen after 14.5 days (range 6 – 21). Nine out of 15 evaluable patients had GvHD grade I–II and GvHD grade III or IV was not observed. One out of 8 evaluable patients had a limited chronic GvHD. One heavily pretreated patient who already had engrafted died due to a TRM (heart failure at day +10). No lethal infection was seen. One patient developed a slowly improving encephalopathy and one patient has an improving neuropathy. Of the 18 patients, 5 patients relapsed at a median of 73 days (range 35–126) after transplant. Twelve patients are alive without evidence of disease with a short median follow-up of 125 days (range 35 – 370). In 10 of these 12 patients, a 100% donor chimerism is maintained, whereas a decrease in donor chimerism was seen in 2 patients. The feasibility, low toxicity and the fast and sustained engraftment in this study is promising and argues for a wider use of RIC in haploidentical transplantation. In order to augment the antileukemic effect, this approach could be a platform for further posttransplant immune and cell therapy.

Author notes

Corresponding author