Effective Cytoreductive Treatment before Allogeneic Hematopoietic Stem Cell Transplantation May Improve the Outcome of Patients with Blast Phase Myelofibrosis

Introduction

The clinical outcome of patients with blast phase myeloproliferative neoplasms (BP-MPN) is dismal. Allogeneic hematopoietic stem cell transplantation (allo-HSCT) may represent a potentially curative treatment option in selected patients. The role of pre-transplant induction therapy is debated. We retrospectively analyzed the outcome of a molecularly annotated cohort of patients with BP myelofibrosis (MF) who underwent allo-HSCT.

Methods

Twenty-four consecutive patients who underwent allo-HSCT for BP-MF between 2008 and 2023 were included. Cytogenetic information was available for 22 patients at the time of BP diagnosis. Cytogenetic risk and response to induction therapy were defined according to the European Leukemia Net 2022 acute myeloid leukemia guidelines. Pre-transplant mutational analysis, using a myeloid gene panel by next-generation sequencing, was performed in 22 patients. Spleen length, measured by ultrasound, was available at blastic evolution in 15 patients and before conditioning in 19 (5 out of 24 had undergone splenectomy). Before conditioning, 20 patients received intensive (“7+3”-like) induction chemotherapy and 1 patient decitabine plus venetoclax while 3 patients moved directly to transplant.

Results

The median age at transplant was 59 years (range, 50-66 years). Patients had progressed to BP from a previous primary MF (n=5), post-essential thrombocythemia MF (n=10), and post-polycythemia vera MF (n=9). The median time to progression from myelofibrosis to BP was 1.3 years. Five patients (23%) had an adverse cytogenetic risk profile. The driver mutation was JAK2V617F in 50% (n = 12), CALR in 33% (n = 8; type 1, n = 5), and MPL in 8% (n = 2), whereas 2 patients (8%) were triple negative. High molecular risk mutations, as defined in the MIPSS70+ version 2.0 score, were present in 45.5% (n = 10). The most frequent non-driver mutations were TET2 (32%), SETBP1 (32%), EZH2 (23%), and RUNX1 (23%). TP53 mutation was detected in 4 patients (18%) (2 multi-hit and 2 single-hit). The donor was matched unrelated, HLA-identical sibling, haploidentical, and mismatched unrelated in 37%, 33%, 16%, and 12%, respectively. Conditioning was reduced intensity in 63% of patients.

Status at transplant was complete remission (CR)/CR with incomplete count recovery (CRi) in 13 patients (54%) and active disease in 11 (46%). Median spleen size was 21 cm at blastic evolution and 19.5 cm before conditioning (17 cm in patients in CR/CRi and 22 cm with active disease). A spleen size reduction was documented in 6 (50%) out of 12 evaluable patients treated with intensive chemotherapy, with a median decrease of 2 cm (range, 1-8 cm). Nine patients (38%) relapsed after allo-HSCT while 3 patients (12%) experienced graft failure. Four patients (17%) underwent a second transplant, 2 for hematological relapse, 1 for molecular relapse, and 1 for graft failure. Deaths were 11 (46%), 8 due to relapse and 3 for transplant-related mortality.

With a median follow-up after transplant of 2.9 years (0.02-15.5 years), the estimated 3-year progression-free survival (PFS) and overall survival (OS) were 49% and 62%, respectively. By univariate analysis, TP53 mutation, EZH2 mutation, and myeloid donor chimerism <95% at day +90 were significantly associated with reduced PFS and OS. A trend towards improved OS and PFS was seen in patients who underwent allo-HSCT in CR/CRi versus those with active disease (HR for OS 0.42, P = 0.16; HR for PFS 0.49, P = 0.21). At 3 years, the cumulative incidence of relapse (CIR) was 38% while the non-relapse mortality was 12%. In patients with either TP53 or EZH2 mutation, the 3-year CIR was 100%.

When comparing these 24 patients with a historical cohort of 50 consecutive patients who underwent HSCT for chronic phase MF in the same timeframe, we observed that SETBP1 (32% vs 0%, P = 0.0001), RUNX1 (23% vs 2%, P = 0.009), and TP53 (18% vs 6%, P = 0.19) mutations were more recurrent in BP-MF. The outcome of patients transplanted in blast versus chronic phase MF was similar (3-year OS 62% versus 58%, respectively).

Conclusions

Our results suggest that intensive chemotherapy or hypomethylating agents and venetoclax may help in bridging patients with BP-MPN to allo-HSCT. Even for these high risk patients, allo-HSCT may potentially represent a curative treatment option. The adverse outcome observed with either TP53 or EZH2 mutation advocates for an early post-transplant intervention to prevent relapse.