Allogeneic Hematopoietic Stem Cell Transplantation for Patients with Polycythemia Vera or Essential Thrombocythemia Transformed to Myelofibrosis or Acute Myeloid Leukemia: a Single Center Experience

In myelofibrosis (MF) which median overall survival (OS) is about six years allogeneic hematopoietic stem cell transplantation (HSCT) is the sole therapeutic procedure with a curative potential. Recent retrospective and prospective studies have reported that MF patients who receive a related donor graft have a clinical outcome better than that of patients who receive an unrelated donor graft. Thus, in order to verify this issue and identify patient- and HSCT-specific variables that influence this different clinical outcome we conducted this retrospective analysis in 31 pts with chronic and advanced-phase MF who underwent HSCT at our Institution. This series included 6 (19.3%) females and 25 (80.7%) males with a median age of 54.9 years (range 30.2-63.8) whose median follow-up was 1.7 years (range 0-11.8). Fifteen pts were classified as primary MF, 9 as secondary MF and 7 as AML post-MF. From a molecular point of view, 17 pts presented the V617F JAK2 mutation, 7 a CALR mutation, 2 a MPL mutation and 2 none of these mutations ("triple negative"). Eleven pts (35.5%) received the transplant from an identical sibling, 17 (54.8%) from a matched unrelated donor, 3 (9.7%) from a mismatched relative. The graft source was the bone marrow in three pts (9.7%) and the peripheral blood (PBSC) in 28 (90.3%). The median number of CD34+ cell infused was 7.3x106/kg. Neutrophil engraftment occurred in 27 pts (87.0%) at a median time of 16.5 days (range 11-34) and platelet engraftment occurred in 21 pts (67.8%) at a median time of 17 days (14-98). In these pts splenomegaly was associated with a significant delay in neutrophil and platelet recovery. Considering the remaining 4 pts, all with a relevant splenomegaly, 2 did not engraft and 2 experienced an autologous reconstitution. Considering all the cohort, death occurred in 13 pts (41.9%) and the 1-year cumulative incidence of nonrelapse mortality (NRM) was 32.9% (95& CI: 16.3%-50.6%). The 3-year overall survival (OS) was 59.4% (95% CI: 38.1%-75.5%) and the 5-year OS was 47.2% (95% CI: 25.3%-66.4%). The disappearance of the molecular marker present at the onset of the disease on subsequent molecular tests occurred in 10 pts (32.2%) who are still in clinical and molecular complete remission. Relapse occurred in 5 pts (16.1%) and 6-month and 3-year cumulative incidence of relapse (CIR) were 3.5% (95% CI: 0.3%-15.1%) and 19.3% (95% CI: 7.0%-36.1%). Eight pts (27.6%) experienced acute GVHD within 100-day (95% CI: 13.1%-44.3%) and 12 (38.7%) developed chronic GVHD which 6-month, 1-year, 3-years cumulative incidences were 25.2% (95% CI 11.1%-42.2%), 33.1% (95% CI 16.6%-50.6%) and 48.2% (95% CI: 27.3%-66.3%). When the clinical outcome of related donor graft group was compared to that of unrelated donor graft group, the 1-year NRM was 18.2% (95% CI: 2.9%-44.2%) for the first group of pts versus 45.5% (95% CI: 20.5%-67.6%) for the second group (p=0.07); median OS and median progression-free survival (PFS) were not reached for the first group of pts versus 2.2 and 0.7 years respectively for the second group of pts (p=0.05 and p=0.05). In addition, a borderline significance in 2-year OS was noted: 81.8% (95% CI 44.7%-95.1%) versus 51.0% (95% CI: 25.4%-71.8%) (p=0.071). The 100-days cumulative incidence of grade II-IV acute GVHD was 10% (95%CI: 0.6%-35.8%) for the first group of pts versus 43.8% (95% CI: 19.8%-65.6%) for the second group with a similar incidence of chronic GVHD. Re-transplantation of the two pts with graft failure did not result in hematopoietic reconstitution as was followed by rejection of the PBSC infused. One pt with an autologous reconstitution achieved a full donor chimerism after the third transplant from a haplo-identical donor. In conclusion, this retrospective study confirms that HSCT is a potentially curative option for MF patients despite the high relapse incidence and non-relapse mortality. Disclosures No relevant conflicts of interest to declare.