Curative role of allogeneic hematopoietic stem cell transplantation in patients with myeloproliferative neoplasm, unclassifiable

Myeloproliferative neoplasm, unclassifiable (MPN-U) is a heterogeneous disease, which despite clinical presentation and histological phenotype consistent with MPN does not meet diagnostic criteria for any of the defined MPN variants such as polycythemia vera or essential thrombocythemia. Genetically, the most common driver is a JAK2 V617F mutation, but mutations in CALR, MPL, and ‘triple-negative’ cases have also been implicated.¹ Patients with this rare subset of MPN (< 5% of all MPN cases) have worse outcomes compared to patients with other forms of MPN. There is currently no accepted standard therapeutic regimen and allogeneic hematopoietic stem cell transplantation (allo-HSCT) is considered the only potentially curative options, although outcomes have not been fully investigated.

In order to better understand the outcome of allo-HSCT in patients with MPN-U Donal P. McLornan, and colleagues, on behalf of Chronic Malignancies Working Party of the EBMT, performed a retrospective, multicenter study, which was recently published in the British Journal of Hematology.²

Methods

Analysis of overall survival (OS), non-relapse mortality (NRM), and relapse incidence was performed on the data from the EBMT transplant database. Adult patients who underwent first allo-HSCT for MPN-U, between the years 2000 and 2015, using reduced-intensity conditioning (RIC) or myeloablative conditioning (MAC) were eligible.

Patient Characteristics

• In total, 70 patients were included in the analysis. Patient and transplant characteristics are provided in Table 1
• Median follow-up was 87 months (31–196)
• Patient cohorts for MAC and RIC were well balanced for gender (approx. 31% female) and stem cell source (approx. 91% peripheral blood [PB] vs 9% bone marrow [BM])

Table 1. Patient characteristics

MAC, myeloablative conditioning; RIC, reduced-intensity conditioning; TBI, total body irradiation
Characteristics	MAC (n = 31)	RIC (n = 39)	p value
Median age, years (range)	52.2 (21.9–69.7)	58.8 (31.3–69.7)	0.002
Median time to transplant, months (range)	10.58 (2.86–148.5)	16.03 (3.32–244.21)	0.306
Conditioning regimens, % TBI-based Fludarabine and busulphan Fludarabine and melphalan Busulphan and cyclophosphamide Other	42 22 – 10 26	– 56 8 – 36	–
Donor type Matched related Unrelated	45 55	33 67	0.313
In vivo T cell depletion, % Yes No Unknown	52 35 13	72 20 8	0.113

Results

• There was a significantly increased incidence of Grade ≥ 2 acute graft-versus-host disease (aGvHD) with MAC vs RIC (37% vs 16%, respectively; p = 0.05) as presented in Table 2)
• The cumulative incidence of chronic GvHD (cGVHD) was numerically higher in the MAC cohort, however, values did not reach statistical significance (52%; 95% CI, 32.4–71.6 for MAC and 32.1%; 95% CI, 14.8–49.4 for RIC; p= 0.17)
• There was a trend towards lower rates of delayed/ failed neutrophil engraftment in the MAC cohort when compared to the RIC cohort
• There were no significant differences in OS and cumulative incidence of relapse between the MAC and RIC cohorts (p = 0.33 and p = 0.28, respectively)

Table 2. Outcomes

CI, confidence interval; GvHD, graft-versus-host disease; MAC, myeloablative conditioning; RIC, reduced-intensity conditioning;
Characteristics	MAC (n = 31)	RIC (n = 39)	p value
Neutrophil engraftment < 30 days, % Yes No Unknown	97 3 0	82 15 3	0.086
GvHD, % (95% CI) Acute at 3 months, Grade ≥2 Chronic at 12 months	37 (19–54) 52 (32.4–71.6)	16 (4–28) 32.1 (14.8–49.4)	0.05 0.17
Cumulative incidence of relapse, % (95% CI) 1-year 3-year 5-year	10 (0–20) 23 (8–37) 27 (11–43)	28 (14–42) 36 (21–51) 36 (21–51)	0.28
Overall survival rates, % (95% CI) 1-year 3-year 5-year	77 (62–92) 55 (37–72) 42 (23–60)	59 (44–74) 44 (28–59) 41 (25–56)	0.33

• The NRM was similar across both cohorts, with GvHD identified as the main cause of NRM
• There was an association between use of unrelated donor and a decreased OS (p = 0.02) and a higher NRM (p = 0.02) when compared to matched sibling donor
• In both cohorts, T cell replete allo-HSCT was associated with an improved OS compared to T cell deplete (p= 0.03)
• Abnormal karyotype was associated with a trend towards a higher risk of relapse (p = 0.06)

Conclusion

Results of this largest to date analysis of allo-HSCT outcomes in patients with MPN-U demonstrate the potentially curative role of allo-HSCT in MPN-U and provides for the first time more robust data about engraftment, GvHD, and outcome. The risk of relapse in these patients remains high, and occurs early after transplantation, independent of conditioning regimen.

Although the RIC regimen seems to show a trend for lower aGvHD and cGvHD it is offset by a lower neutrophil engraftment rate. Therefore, the authors recommend using MAC in younger and fit patients and reserve RIC for older patients who can still benefit from allo-HSCT.

However, the authors acknowledge the limitation of the study including, relatively low number of patients, changes in transplantation practice, a lack of comprehensive cytogenetic and mutational status, as well as differences in age, the length of disease, and percentage of in vivo T cell depletion between the cohorts.