Second donation of hematopoietic stem cells from unrelated donors for patients with relapse or graft failure after allogeneic transplantation

Allogeneic hematopoietic cell transplantation (HCT) is a potentially curative approach for a variety of hematologic malignant and non-malignant disorders. With the rising number of stem cell transplantations from unrelated donors over the last years,1 the occurrence of graft failure or disease relapse has increased. Therapeutic options for the majority of these patients include a second HCT using stem cells from the same unrelated donor.

Given the lower stem cell yield obtained by a bone marrow harvest, the easy and convenient access to G-CSF mobilized peripheral blood stem cells (PBSC) has resulted in their preferred usage as the source of hematopoietic progenitor cells.2 Nevertheless, severe complications have been described.3 In fact, we described4 for the first time a significant increase in spleen size after G-CSF stimulation. Although there have been a few single case reports of the occurrence of secondary hematologic malignancies in voluntary stem cell donors after exposure to G-CSF,5 at the moment long-term safety data for repeated G-CSF exposure for stem cell donors are not available. Additionally, a bone marrow harvest as an alternative donation procedure still harbors a minimal risk of morbidity.

Against this background, it is of importance to address the question whether repeated donations of HSC in case of graft failure or relapse are associated with a clinical benefit for patients thus justifying a possibly increased risk for voluntary donors. Earlier studies on this subject were mainly performed with bone marrow (BM) as stem cell source (SCS) and showed a worse outcome compared to results after first transplantation. Nevertheless, even the use of PBSC does not seem to have a positive impact on survival.6–10

Therefore, we retrospectively investigated within the German bone marrow center (DKMS) database the outcome of seventy (n=70) patients with hematologic malignancies (Table 1) receiving a second HCT using peripheral blood stem cells (n=67) or bone marrow (n=3) from the same unrelated donor. To our knowledge this is one of the largest analyses so far for a second HCT using PBSC from unrelated donors as SCS. All patients and donors had given informed consent prior to HCT or donation respectively. The indications for a second donation of stem cells were either relapse (n=36, median time to relapse 6.2 months [range 1.1–69]) or primary graft failure (n=34). Whereas in 21 patients no conditioning was used, the remaining 49 patients, including all but 4 patients with relapsing disease, received a second cytoreductive preparative regimen. Cumulative incidences of graft versus host disease (GvHD), treatment related mortality (TRM) and relapse were determined. Survival rates were calculated by means of Kaplan-Meier analysis. Univariate log rank comparisons were performed to identify parameters predicting the outcome after second transplantation.

After a median follow-up of 33 months (range 7.9–76.2) after second HCT, a total of 55 patients died due to TRM (63%) or relapse (16%) resulting in a median overall survival for all patients of three months. The OS was significantly better in patients receiving a second donation for graft failure (37 vs. 7%, p=0.02) and in patients < 50 years (38 vs. 8%, p=0.01) (Figure 1). Time interval between HCT, HLA-disparity, disease type (AML/MDS vs. CML/MPS vs. ALL), number of infused CD34 cells/kg, CMV-positivity or graft source at first transplantation had no significant impact (data not shown). Additionally, in graft failure patients there was no impact on outcome whether a conditioning prior to second transplantation was used (data not shown). This is an interesting observation suggesting that in a certain subgroup of patients with graft failure, a second stem cell boost without repeated conditioning seems to be equally effective. The infusion of CD34 selected donor cells in patients with poor graft function has been described as a successful strategy by Bagicalupo et al.11 However, due to the limited number of patients results have to be interpreted with caution.

Previous studies using mostly bone marrow for second transplantation showed poor long term survival rates9,12,13 comparable to our results. It is of note that TRM seemed to be slightly higher (63%) in our cohort compared to previous studies. This is surprising, as in about one third of our patients no additional toxic conditioning was used before second transplantation. One explanation for this observation could be a higher median age of patients in our study. In fact, patients' age can predict the outcome after second transplantation, which supports our findings, although the cut-off value separating the favorable from the unfavorable group was significantly higher in this study. Eapen et al. found a better prognosis for patients ≤ 20 years9 whereas in the study of Michallet et al.10 the cut-off was even lower (<16 years). Recent predictors such as time to relapse after first transplantation, remission status at second transplantation, absence of acute and occurrence of chronic GvHD after second transplantation, or conditioning regimen had no impact in this analysis.

Table 1.Patients’ characteristics.

Figure 1.(A , B, C) Treatment-related mortality (A) and overall survival (B, C) according to patient age at and reason for 2^nd HCT.

These data show the limited success of a second infusion of HSC from the same donor, especially in older patients and in cases in which relapse of malignancy is the reason for second transplantation. Given the currently unknown long-term effects of a repeated exposure to G-CSF, the factors mentioned above should be kept in mind whenever a request for a second PBSC donation is considered.

Acknowledgments

we thank Catrin Theuser for data and statistical management of the project.