Family Mismatched Donor Transplantation for Myelofibrosis: A Retrospective Analysis of the EBMT Chronic Leukaemia Working Party

Background: Allogeneic stem cell transplantation is a treatment option for patients with advanced myelofibrosis. Problems encountered include an increased risk of delayed or poor engraftment and in the mismatched unrelated donor setting a higher rate of GVHD and particularly poor outcomes. As for other indications for allogeneic stem cell transplants, patients for whom either a matched sibling or matched unrelated donor is not available are considered for either a double umbilical cord blood, a mismatched unrelated donor or haploidentical stem cell transplant. Data on the latter option are limited and we reviewed registry data on all family mismatched donor transplants performed between 2001 and 2015 and reported to the EBMT registry.

Results: Records retrieved 69 patients with myelofibrosis transplanted between November 2001 and November 2015. 44 (64%) were male. 50 (74%) had primary myelofibrosis,18 (27%) had secondary myelofibrosis (6 from ET, 5 from PRV and 7 others) and unknown 1(2%). Of 25 patients for whom the JAK2 V617F mutation status was known, 15 (22%) harboured the mutation. Patient Karnofsky performance status was > 70% in 98%.

Of the mismatched family donors, 47 (68%) were male and 22 (31%) female. Donors were HLA mismatched at 1 locus in 12 (17%) and 2 or more loci in 48 (69%). Donor-recipient serology combinations were CMV -/- in 8 (12%), +/- in 4 (6%), -/+ in 15 (22%) and +/+ in 34 (49%) missing 8 (12%).

Bone marrow was the stem cell source in 34 (49%) and peripheral blood in 35 (51%). The median total nucleated cell count (TNC) infused was 7.5x10⁸/kg (range 2.3-21x10⁸/kg) from data available in 17 patients. The median CD34+ cell dose was 6.9x10⁶/kg (range 1.9-18.18x10⁶/kg) from data available in 19 patients.

Conditioning was myeloablative in 48 (70%) and RIC in 21 (30%) The conditioning regimes were varied but the predominant ones were Fludarabine, Busulphan, ATG (FBATG) and Thiotepa, Busulphan, Fludarabine (TBF). TBI was administered in 8 (12%) and in vivo T cell depletion in 22 (32%), ex-vivo T cell depletion in 5 (7%) patients. GVHD prophylaxis varied with post transplant Cyclophosphamide administered in 34/67 (49%) and ATG in 19/67 patients (28%).

Neutrophil engraftment was established in 53 (82%) at a median of 20 days (range 11-83). Primary graft failure occurred in 8 (12%) and secondary graft failure in 4 (6%). This occurred at a median of 12 months (range 4.5-35 months). Eleven of these patients had a second allograft at a mean interval of 6.4 months. Responses to the first allograft (censoring for patients who had a second allograft) with data available in 45 patients, showed that complete remission was achieved in 35 patients (78%), 6 (13%) were never in CR and 4 (9%) were not evaluable.

The cumulative incidence of grade II-IV acute GvHD at 100 days was 12% (95% CI 4-21%) and for grade III-IV acute GvHD at 100 days it was 5% (95% CI 3-11%). Data for chronic GVHD was valid in 49 patients. The cumulative incidence of chronic GvHD at 2 years was 62% (95% CI 47-76%). The cumulative incidence of limited cGvHD was 45% (95% CI 31-59%) whereas the cumulative incidence of extensive cGvHD was 10% (95% CI 2-19%).

The median follow up was 24 months (95% CI 13-35 months). The 2-year OS was 51% (95% CI 37-76%) and the 5-year OS was 38% (95% CI 21-55%). The 2-year RFS was 44% (95% CI 30-59%) and the 5-year RFS was 31% (95% CI 15-48%). The 2 year cumulative incidence of relapse was 14% (95% CI 5-24%). The 2 year NRM was 41% (95% CI 28-55%), which increased to 54% (95% CI 37-72%) at 5 years.

The main causes of death were infection (16, 24%), GVHD (7, 10%) organ damage or failure (3, 5%), relapse/disease progression (1, 2%) and secondary malignancy or PTLD (1, 2%). On univariate analysis there was no significant effect of recipient gender, donor gender, degree of HLA mismatch 1 vs >1 Ag MM, CMV matching between donor and recipient, primary or secondary MF, disease stage at transplant (chronic versus advanced phase), conditioning intensity, conditioning regimen, GVHD prophylaxis with ATG or post transplant cyclophosphamide or stem cell source on overall survival.

Conclusion: Concerns regarding engraftment and secondary graft failure have excluded patients with myelofibrosis from clinical trials of mismatched related donor transplant. The data suggest that engraftment is feasible, and PFS and OS can be attained with limited severe chronic GVHD with family mismatched donors in this setting.