Background: Graft versus host disease (GvHD) remains the major complication of allogeneic HSCT, as standard prophylaxis regimens are not completely effective in a significant proportion of patients. Recent studies have shown that intravenous administration of high-dose cyclophosphamide (Cy) in the early post-transplant period is an effective GvHD prophylaxis regimen for patients who have undergone allogeneic transplantation with alternative donor marrow grafts. In this report, we extend these findings to patients who have received allogeneic HPC-Apheresis products from alternative donors to determine the impact of Cy-based GvHD prophylaxis on neutrophil engraftment, hematopoietic chimerism, and immune reconstitution.

Methods: In this single center retrospective study, we analyzed data of 103 patients who received matched related (MRD n=30), matched unrelated (MUD n=53), mismatched unrelated (mMUD n=7) and haploidentical (HAPLO n=12) apheresis grafts of which all but MRD received 50 mg/kg/day Cy (HAPLO on post-HSCT days +3 and +4, MUD & mMUD on day +3). All patients were conditioned with either a combination of cyclophosphamide/Fludarabine/TBI or Fludarabine/Busulfan followed by an unmanipulated apheresis graft. Additional GVHD prophylaxis for MUDs, mMUDs, and haplos consisted of tacrolimus (day +5 to +100, followed by taper) and mycophenolate mofetil (day +5 to day +35). MRDs received tacrolimus (day +5 to +100, followed by taper) and low dose, short course methotrexate days (+1, +3, +6, +11). Peripheral blood immune reconstitution was assessed by flow cytometry using antibodies to CD3, CD4, CD8, CD19, CD56, CD25, CD122, CD137, and γδ-T cells. Quantitative T-cell and myeloid specific STR donor chimerism status was assessed with microsatellite variable tandem repeats of peripheral blood leukocytes using standard techniques. Chimerism was assessed on post-HSCT days 30, 60, 100 and 180. Chimerism data are expressed in median and range of donor DNA percentages. If greater than or equal to 95% of donor DNA, they are considered to have achieved full chimerism.

Results: Median time to ANC engraftment was 12d for MRD’s and HAPLO recipients took slightly longer at 15.5 days and is significantly delayed when compared to MRD’s. One dose of Cy does not delay engraftment, but two doses does. Most patients achieved donor T cell (CD3) and myeloid (CD15) chimerism status by day 60. Day +30 median CD3/CD15 chimerism was 77.3+20.7/99.4+0.97 for MRD; 81.3+21.2/98.3+6.7 for MUD; 86.4+15.8/99.9+0.4 for mMUD; and 99.1+1.6/99.5+1.0 for HAP. Day +30 HAPLO CD3 chimerism was significantly improved over MRD (p<0.001), MUD (p = 0.004) but not mMUD (p = 0.10). Day +30 CD15 chimerism did not differ between groups. A greater proportion of patients treated with Cy showed improved 30 day CD3/CD15 chimerism (60%/90%) than MRD patients (7%/86%) who did not receive Cy. T cell count recovery normalized by day +90 (MRD); +180 (MUD); and +270 (HAPLO) with day +30 recovery significantly faster for MRD compared with MUD (p = 0.0001) and HAPLO (p = 0.006) due principally to robust CD8+ T cell recovery in the MRD group. Day +30 T cell recovery for MUD was slightly but not significantly improved over HAPLO (p = 0.09). CD4 count recovery required up to a year in all groups. Recovery of γδ T cells did not differ between groups. Tregcounts were lower in MRD and did not differ significantly per time period in MUD and HAPLO but each showed peak counts at day +60. NK counts recovered by +30 for all groups and B cell recovery was earliest in HAPLO followed by MUD and MRD.

Conclusion: The use of post-HSCT Cy in patients who receive HPC-Apheresis grafts from alternative donors does not significantly impact chimerism or engraftment. In general, T cell recovery is delayed in patients who received Cy when compared MRD HSCT although these differences disappear by day +270. Taken together, HPC-Apheresis is a reasonable alternative donor graft source when accompanied by post-HSCT Cy-based GVHD prophylaxis.


Off Label Use: cyclophosphamide for GVHD prophylaxis.

Author notes


Asterisk with author names denotes non-ASH members.

This icon denotes a clinically relevant abstract