T-cell replete HLA-haploidentical transplantation using post-transplant cyclophosphamide for prevention of GVHD and graft rejection (Haplo-PTCy) has recently emerged as a valid form of alternative-donor transplantation for patients who lack traditional matched-siblings (MSD) or matched-unrelated donors (MUD). We have demonstrated that patients undergoing Haplo-PTCy can have equivalent rates overall and disease-free survival and equivalent or lower cumulative incidences of GVHD and non-relapse mortality to patients transplanted contemporaneously from MSD and MUD at the same center (Bashey et al ASH 2011 abstract #833). In this study we assessed lineage-specific chimerism, together with incidence and outcome of graft-failure in 89 consecutive first Haplo-PTCy performed for hematologic malignancy in our center between Oct 2005 and Jun 2012. Patient characteristics: M 48, F 41; median age 48 (20–74); Diagnosis AML 28, ALL 16, CLL 11, NHL 9, HL 8, CML 8, MDS 7, MPS 2; median number of matched HLA loci were 5/10 (range 5/10 to 8/10); Median CD34+ and CD3+ cell dose infused were 4.01 × 106/kg (0.84–6.27) and 5.35 × 107/kg (1.4–53.82) respectively. Fifty-eight patients received a marrow graft and 31 received G-CSF mobilized PBSC. The preparative regimen was RIC/NST in 59 (fludarabine 30 mg/m2/d d -6 to -2, TBI 200cGy d-1, cyclophosphamide 14.5 mg/kg/d d-6 & -5, and 50 mg/kg/d d+3 & +4) and myeloablative in 30 (regimen A- fludarabine 30 mg/m2/d d -6 to -2, busulfan 110–130 mg/m2/d d-7 to -4, cyclophosphamide 14.5 mg/kg/d d-3 & -2, and 50 mg/kg/g d+3 & +4 [20 patients] and regimen B- fludarabine 30 mg/m2/d d -7 to -5, TBI 1200 cGy given in 8 fractions between days-4 to -1 and cyclophosphamide 50 mg/kg/d d +3 & +4 [10 patients]). The presence of pre-transplant anti-donor HLA antibodies were assessed using a solid phase assay (Panel Reactive Antibody, PRA, Clinimmune, CO) and by anti-donor cross-matching by flow cytometry (Clinimmune). All donors were selected to provide a negative cross-match using recipient serum and donor T-cells prior to transplant. Engraftment was determined using standard CIBMTR criteria. Lineage-specific chimerism was determined using PCR for short tandem repeats on peripheral blood mononuclear cells separated by CD3 and CD33 expression using immunomagnetic beads on d 30,60,90 and 180 following transplant. Median time to ANC > 500/mm3 was 16d (12–27d) and platelets > 20,000/mm3 was 26d (0–26d). Median T-cell (CD3) and myeloid (CD33) donor chimerisms were 100%, at all time-points assessed from d30–180 (Fig 1). All 30 patients who received a myeloablative Haplo-PTCy had full engraftment of T-cells and myeloid cells starting d +30. However six of 59 patients undergoing RIC/NST Haplo-PTCy had primary failure of T-cell engraftment -median CD3 chimerism (range) for these patients on d 30 and 60 were 0% (0–6%) and 0% (0–14%). Median CD33+ cell chimerism for the same patients on d 30 and 60 respectively were 86% (0–100%) and 45% (0–100%). Four of these patients underwent a second Haplo-PTCy, a median of 105d (range 8–123d) following the first transplant using a different haploidentical donor and the same preparative regimen. In each case the second Haplo-PTCy was successful (CD3+ donor chimerism 100% by d 30–60 in all cases). One patient who was too unwell for second Haplo-PTCy had spontaneous improvement in CD3 chimerism (6–14% d 30–90 improving to 100% d 180) and one patient died of progressive malignancy before a second Haplo-PTCy could be performed.
These data demonstrate that full donor chimerism of T-cells and myeloid cells is usual following Haplo-PTCy from the earliest time-points assessed. Engraftment failure was not seen in any patient using the myeloablative regimens described above. Approximately 10% of patients conditioned with the RIC/NST regimen failed to undergo initial T-cell engraftment. However, re-transplantation was successful in all cases when attempted. Late spontaneous improvement of CD3 chimerism is also possible in patients with low level mixed chimerism early post-transplant.
No relevant conflicts of interest to declare.
Asterisk with author names denotes non-ASH members.