Work

Allogeneic stem cell transplant for patients with myeloproliferative neoplasms in blast phase: improving outcomes in the recent era

Year: 2021

Type: article

Abstract: BCR-ABL1 (Philadelphia chromosome)-negative myeloproliferative neoplasms (MPN) are a heterogeneous group of haematological malignancies. The current World Health Organization classification recognizes four variants: primary myelofibrosis (PMF), prefibrotic PMF, polycythaemia vera (PV), and essential thrombocythaemia (ET).1, 2 The propensity to evolve into the myeloid blast phase (MPN-BP), defined as ≥20% blasts in the peripheral blood or bone marrow, depends on the phenotype and is the highest in PMF (14% at 20 years), lower in PV (7% at 20 years), and the lowest (4% at 20 years) in ET.1, 3, 4 Other factors predicting evolution to MPN-BP include age, haematological parameters, cytogenetic features and mutational profile.1, 2, 4-6 MPN-BP is a distinct entity compared to chronic-phase MPN, de novo acute myeloid leukaemia (AML), and even compared to AML evolving from other chronic myeloid malignancies. The management of MPN-BP remains challenging with limited therapeutic options and poor outcomes.5, 7 Median survival without transplant is 1·5–10 months and remains unchanged over the last decade.2, 4, 8 Stem cell transplantation (SCT) is the 'gold standard' due the possibility of cure and improved survival over non-SCT approaches (47% vs 15% at two years). Wider applicability of SCT is limited due to age and comorbidities with as many as 89% of MPN-BP patients not undergoing SCT.8 In addition, being in complete remission (CR) can be a requisite for SCT at many centres — further reducing the fraction of the intended cohort eventually undergoing SCT.9 Limited experience with allogeneic SCT in MPN-BP has been reported.8-12 The European Society for Blood and Marrow Transplantation (EBMT) cohort consisted of 46 MPN-BP patients that underwent allogeneic SCT.10 At three years, progression-free and overall survival (PFS, OS) were 26% and 33% respectively. Achieving CR pre-SCT, consisting of 17% of the cohort, was the only factor predicting improved survival. Kennedy et al. reported a smaller single-institution cohort of patients who underwent high-intensity induction therapy followed by allogeneic SCT with a two-year post-SCT OS of 47%.9 Notably, a prerequisite to proceed for SCT was achieving BM blast ≤ 5%, limiting the number of patients who underwent allogeneic SCT. With the advent of targeted inhibitors for MPN, the availability of alternative donors, improved supportive care, and increased awareness, more patients are being considered for SCT. However, data describing the outcomes of MPN-BP in the current era are lacking.9-11 The aim of our study was to describe patient, disease, and transplant characteristics as well as the outcomes of MPN-BP patients undergoing allogeneic SCT and identify factors predicting improved post-SCT survival. Methods are described in detail in the Supplementary Methods. Briefly, we identified 43 patients that underwent matched sibling (MSD) or matched unrelated donor (MUD) allogeneic SCT for MPN-BP. The clinical and transplant-related characteristics are described in Table SI. When analysed by year of SCT (2001-2009 vs 2010-2016), a higher proportion of patients was treated with JAK2i (36% vs 5%, P = 0·01) and HMA (36% vs 9%, P = 0·04) in 2010–2016. Seven (16%) patients received the combination of JAK2 inhibitor and hypomethylating agents (JAK2i + HMA). Their clinical, molecular, and transplant-related characteristics are described in Table SII. Outcomes of interest were relapse incidence (RI), PFS and OS at four years after SCT. Thirty-six (84%) patients received therapy (median 1, range 1–5), while seven did not receive chemotherapy for MPN-BP. Among 36 patients that received pre-SCT therapy, pre-SCT disease status was: CR1 8%; marrow CR 35%; and no response 50%. Using a combination of morphology, cytogenetics, and JAK2 V617F status, only one patient had no evidence of disease at the time of SCT (Fig 1 and Table SIII). Post SCT, four (9·3%) patients had early death, one (2·3%) had engraftment failure, and 38 (88·4%) patients engrafted. The median time to neutrophil and platelet engraftment was 13 (8–24) and 22 (8–215) days. One patient each died of acute and chronic graft-versus-host disease (GVHD). Acute GVHD was seen in 19 (46%) of 41 evaluable patients — two (5%) grade 1, 14 (34%) grade 2, and three (7%) grade 3. Chronic GVHD was seen in seven (33%) of 21 evaluable patients — three (43%) limited and four (57%) extensive. Twenty-eight deaths were noted with recurrent MPN-BP being the most common cause (n = 17, 61%). Other causes were infection (n = 3, 11%, one bacterial, two fungal); pneumonia (n = 2, 7%), GVHD (n = 2, 7%, one acute, one chronic); organ failure (n = 1, 4%); and secondary malignancy (n = 1, 4%). With a median follow up of 41 months (5–170) among survivors, the median PFS and OS were 6 and 15 months respectively. At four years post SCT, PFS was 31% [95% confidence interval (CI) 17–46], OS 38% (95% CI 22–53), NRM 24% (95% CI 14–41), and RI was 43% (95% CI 30–62, Fig 2A, B; Table SIV). Patients that underwent SCT in 2010–2016 had a significantly improved OS compared to those who underwent SCT in 2001–2009 (55% vs 24%, P = 0·05) and decreased NRM(43% vs 4%, P = 0·02; Fig 2C, D). Other factors predicting the risk of progression at four years on multivariate analysis were MUD [versus MSD, hazard ratio (HR) 2·9, P = 0·02] and RIC (versus MAC, HR 0·2, Table SV). Finally, pre-SCT blast — ≤5% vs >5% (Table SV) or ≤ 10 vs 10% (data not shown) did not predict post-SCT survival. Given the duration of the study, molecular data besides JAK2 V617F were not available for most patients. Duration of pre-SCT therapies, and outcomes thereof, were not consistently available. Finally, as the cohort only included patients undergoing SCT, comparison with non-transplant approaches is not possible. Regardless of pre-SCT therapies employed, all but one patient had evidence of minimal or measurable disease at the time of SCT, potentially explaining the poor outcome with non-transplant modalities. Moreover, despite differences in disease status at the time of SCT, post-SCT outcomes are comparable among various studies,9, 10, 12 suggesting that achieving CR — empirically considered to be a pre-requisite for proceeding with SCT — may not be as feasible in MPN-BP as with de novo or secondary AML.7 Finally, post-SCT outcomes have improved over the last decade. Possible explanations for the improved survival are targeted pre-SCT therapies, improved conditioning regimen, better donor selection, and supportive care. These observations should be tested in larger prospective studies. We would like to thank Julianne Chen (Stem Cell Transplant and Cellular Therapy, the University of Texas MD Anderson Cancer Center, Houston, TX) for meticulous data collection. MVS and UP designed the study, collected and interpreted data, and wrote the manuscript. RMS analysed the data and edited the manuscript; collected data; SOC, BO, AO, PB, QB, SA, AMA, RSM, NP, and SV treated patients, interpreted data, and edited the manuscript. The authors have no relevant conflict of interests to disclose. Table SI. Clinical and transplant characteristics of patients with BCR-ABL1− myeloproliferative neoplasm in the blast phase (MPN-BP) undergoing an allogeneic transplant. Table SII. Characteristics of patients with BCR-ABL1− myeloproliferative neoplasm in the blast phase (MPN-BP) treated with a combination of JAK2 inhibitor and hypomethylating agents (HMA). Table SIII. Disease status in patients with BCR-ABL1− myeloproliferative neoplasm in the blast phase (MPN-BP) with no morphological evidence of MPN at the time of transplant. Table SIV. Summary of outcomes by the year of transplant performed. Table SV. Univariate analysis of factors impacting progression, overall survival non-relapse mortality, progression-free survival, and overall survival for patients with BCR-ABL1− myeloproliferative neoplasms in the blast phase (MPN-BP) undergoing an allogeneic stem cell transplant. Please note: The publisher is not responsible for the content or functionality of any supporting information supplied by the authors. Any queries (other than missing content) should be directed to the corresponding author for the article.

Source: British Journal of Haematology

DOI: 10.1111/bjh.17423