Outcome of Allogeneic Haematopoietic Stem Cell Transplantation in Myeloproliferative Neoplasms-Unclassifiable: A Retrospective Study By the Chronic Malignancies Working Party of EBMT

Introduction

Myeloproliferative Neoplasm (MPN) unclassifiable (MPN-U), is a heterogeneous disease that presents with an MPN-type clinical/ histological phenotype yet fails to meet diagnostic criteria for other MPN entities. Incidence is <5% of all MPN when stringent World Health Organisation (WHO) 2016 classification is followed. No well-defined treatment algorithm exists and therapeutic approaches vary, ranging from observation alone, cytoreductive or experimental agents, high dose chemotherapy and in some instances allogeneic Haematopoietic Cell Transplantation (allo-HCT). Outcome analysis for this allo-HCT cohort is lacking.We hereby report on a retrospective, multicentre, EBMT-registry based study of adult patients with a confirmed diagnosis of MPN-U according to updated WHO 2016 criteria that received an allo-HCT.

Methods

This registry-based analysis was approved by the Chronic Malignancies Working Party of the EBMT. Patient selection was performed by identifying adult patients who underwent first allo-HCT for MPN-U between 2000-2015, using either Reduced Intensity Conditioning (RIC) or Myeloablative conditioning (MAC) as defined by standard EBMT criteria. Further data collection requests (MED-C) forms were sent to centres to improve data completeness. Statistical analyses were performed with SPSS 22 (SPSS Inc./IBM, Armonk, NY) and R. Overall Survival (OS) was calculated from the date of transplant until death or last observation alive. Cumulative incidence functions were used to estimate Non-Relapse Mortality (NRM) and Relapse Incidence (RI) within a competing risk setting.

Results

A total of 70 patients, 48 (69%) male and 22 (31%) female, with a confirmed diagnosis of MPN-U were analysed. Median age was 57 years (range (r), 22-70 years). Of these patients, 37 (53%) underwent allo-HCT in the period 2001-2010 and 33 (47%) between 2011-2015. MAC regimens were utilised in 31 (44%) patients while 39 (56%) received RIC. Patients were most frequently transplanted within the first two years from diagnosis, with a median time to allo-HCT of 13 months (r, 3-244 months). Diagnostic karyotype was normal in 36 (51%) and abnormal in 23 (33%) patients, with data missing for 11 (16%) patients. A total of 45 (64%) patients had received prior treatment, 23 (33%) patients were untreated and data was incomplete in 2 (3%) patients. Regarding donor type, 27 (39%) patients had a Matched Sibling Donor (MSD) and 43 (61%) an Unrelated Donor (URD). Most frequent conditioning regimens were TBI-based in the MAC cohort and Fludarabine-Busulphan in the RIC cohort. A trend towards higher rates of delayed/failed engraftment was noted in the RIC compared to MAC cohort (p=0.09). Where successful, median time to neutrophil engraftment in both cohorts was similar (18 days for MAC and 17 for RIC) and both platforms demonstrated similar platelet engraftment times (17.5 days). Incidence of grade II-IV aGVHD at 3 months was higher in the MAC (37%) compared to RIC cohort (16%; p=0.05) and the 12-month cumulative incidence of cGVHD for MAC was 52% (95%CI: 32.4, 71.6) and for RIC 32.1% (95%CI: 14.8, 49.4; p=0.117)). Median follow-up was 87 months (minimum and maximum of censored cases: 31 and 196 months). The median OS estimates at 1, 3 and 5-year were 77%, 55% and 42% (MAC) and 59%, 44% and 41% (RIC), respectively (p=0.33). No significant difference existed in OS rates between those who had pre-transplant therapy versus not. Relapse remained significant: cumulative incidences of relapse at 1,3 and 5-years were 10%, 23% and 27% (MAC) and 28%, 36% and 36% (RIC), respectively (p=0.28). NRM probabilities at 1, 3 and 5-years post allo-HCT were also considerable: 19%, 29%, and 34% (MAC) and 28%, 28% and 28% (RIC), respectively (p=0.84). Main causes of NRM were infection and GVHD. Univariate analysis associated use of an URD with a significantly worse OS and NRM compared with MSD. Moreover, the presence of abnormal karyotype at time of allo-HCT was associated with a trend towards a higher risk of relapse (p=0.06).

Conclusions

This study highlights the potentially curative role of allo-HCT in MPN-U and provides clinicians with robust engraftment, GVHD and outcome data. Both engraftment and OS rates appear acceptable yet NRM and CIR rates in both settings remain high and need to be addressed. The impact of abnormal karyotype at the time of allo-HCT and a trend towards higher risks of relapse requires further elucidation.