Value and pitfalls of assessing bone marrow morphologic findings to predict response in patients with myelofibrosis who undergo hematopoietic stem cell transplantation

Background: Allogeneic hematopoietic stem cell transplantation (allo-HSCT) is a curative option for patients with myelofibrosis (MF). Bone marrow (BM) morphologic evaluation of myelofibrosis following allo-HSCT is known to be challenging in this context because resolution of morphologic changes is a gradual process. Patients and methods: We compared BM samples of patients with myelofibrosis who underwent first allo-HSCT and achieved molecular remission by day 100 with BM samples of patients who continued to have persistent molecular evidence of disease following allo-HSCT. Results: The study group included 29 patients: 17 primary MF, 7 post-polycythemia vera (PV) MF, and 5 post-essential thrombocythemia (ET) MF. In this cohort there were 18 JAK2 p.V617F, 8 CALR; 1 MPL, and 2 patients had concurrent JAK2 p.V617F and MPL mutations. The control group included 5 patients with primary MF, one with post-PV MF, one with post-ET MF (5 JAK2 p.V617F; 2 CALR). Following allo-HSCT, both groups showed reduction in BM cellularity and number of megakaryocytes. The study cohort also less commonly had dense megakaryocyte clusters and endosteal located megakaryocytes and showed less fibrosis. There was no statistical difference in BM cellularity, presence of erythroid islands, degree of osteosclerosis, or megakaryocyte number, size, nuclear lobation, presence of clusters or intrasinusoidal location. Conclusions: Following allo-HSCT at 100 days, morphologic evaluation of BM in patients with MF cannot reliably predict persistence versus clearance of molecular evidence of MF. Disappearance of BM MF, dense megakaryocyte clusters, and endosteal localization of megakaryocytes are suggestive of disease response.