Outcomes of HSCT in blast phase of BCR-ABL1-negative MPN compared with de novo and post-MDS AML

BCR-ABL1^– myeloproliferative neoplasms (MPN) include essential thrombocythemia, polycythemia vera, and primary myelofibrosis as the three main classical subtypes.¹ Blastic transformation in patients with MPN describes the presence of ≥ 20% blasts in either the peripheral blood or bone marrow — referred to as the blast phase of MPN (MPN-BP). MPN-BP is associated with significant poorer outcomes, with a median survival < 6 months.²

Allogeneic hematopoietic stem cell transplant (HSCT) is established as the only treatment to confer long-term survival and reduction, but only for selected patients with MPN-BP. Little is understood about HSCT outcomes in MPN-BP in comparison with acute myeloid leukemia (AML) patients, specifically those with either post-myelodysplastic syndromes (pMDS-)AML or de novo AML (dnAML) without prior or underlying MDS.

In this large, retrospective case series, Vikas Gupta and colleagues present the outcome data of 117 patients who received HSCT for MPN-BP between 2001 and 2015, compared to control cohorts of both pMDS-AML and dnAML. Their results were recently published in Blood Advances¹ and have been summarized here.

Study design

This was a retrospective, registry-based, case-controlled cohort study. All patients were identified through the International Center for International Blood and Bone Marrow Transplant (CIBTMR) database of > 500 centers.

Case population:

• Patients with MPN-BP (blasts ≥ 20% in peripheral blood or bone marrow) treated with HSCT

Control population:

• Control cohort 1 – patients with pMDS-AML treated with HSCT
• Control cohort 2 – patients with dnAML treated with HSCT

Inclusion criteria:

• ≥ 40 years of age
• Undergoing first HSCT
• Treated between 2001 and 2015

Exclusion criteria:

• Philadelphia-chromosome positive or BCR-ABL⁺ MPN
• Autologous transplants
• Syngeneic donor HSCT
• Cord blood HSCT
• Ex vivo T-cell depletion

Data Collection

A core dataset for both cases and control cohorts was collected (Table 1). For MPN-BP and pMDS-AML patients, remission at HSCT was classed as blast cells < 5% in the peripheral blood or bone marrow.  For dnAML, remission at HSCT was classed as complete remission (first or second) (by accepted international staging criteria).

Primary outcome:

• Overall survival (OS) defined as time from HSCT to death of any cause

Secondary outcomes included:

• Risk of acute (a) or chronic (c) graft-versus-host disease (GvHD)
• Non-relapse mortality (NRM)
• Relapse
• Progression-free survival (PFS; time to progression, relapse, or death)

Table 1. Core data collected¹

AML, acute myeloid leukemia; BM, bone marrow; CMV, cytomegalovirus; D-R, donor-recipient; ET, essential thrombocythemia; G-CSF, granulocyte-colony stimulating factor; GvHD, graft-versus-host disease; HLA, human leukocyte antigen; HSCT, hematopoietic stem cell transplantation; MDS, myelodysplastic syndromes; MPN, myeloproliferative neoplasms; PB, peripheral blood; PMF, primary myelofibrosis; PV, polycythemia vera.
Demographic data	Age (at HSCT), sex
Clinical characteristics	Karnofsky Score, comorbidities (0 to ≥ 3)
Disease characteristics	MPN etiology (PV, ET, PMF, unclassified), time between MDS and AML (where relevant), white blood cell count and diagnosis (% and n×109/L), cytogenetics (poor, intermediate, favourable, missing), disease status at HSCT (remission/active), time from diagnosis to HSCT
HSCT characteristics	Induction therapy, conditioning regimen, graft source (PB vs BM), donor type (HLA-identical sibling, other related, well matched unrelated, partially/mismatched unrelated), D-R sex match, donor, D-R CMV status, GvHD prophylaxis, in vivo T-cell depletion, G-CSF planned within 7 days of HSCT, year of transplant (2001–2005, 2006–2010, 2011–2015), length of follow-up

Results

A total of 6,030 patients were identified, with no statistical difference in baseline characteristics between cases or either of the control cohorts.

• 177 cases (MPN-BP)
• 4,749 dnAML controls
• 1,104 pMDS-AML controls

MPN-BP vs dnAML

Primary outcomes:

• OS reduced in cases vs controls for patients in remission at HSCT (HR, 1.4; 95% CI, 1.12–1.76; p = 0.003)
  • No difference in OS in cases vs controls in patients with active leukemia at HSCT (HR, 0.93; 95% CI, 0.72-1.20; p = 0.59)

Secondary outcomes:

• Increased risk of relapse in cases vs controls in remission at HSCT (HR, 2.18; 95% CI, 1.69–2.80; p < 0.0001)
  • No increased risk of relapse in cases vs controls with active leukemia at HSCT (HR, 1.16; 95% CI, 0.88–1.54; p = 0.30)

• PFS was reduced in cases vs controls (HR, 1.35; 95% CI, 1.15–1.59; p = 0.0003)
• No difference in risk of aGvHD, cGvHD, or NRM between cases and controls

MPN-BP vs pMDS-AML

Primary outcomes:

• OS reduced in cases vs controls (HR, 1.19; 95% CI, 1.00–1.43; p = 0.05)

Secondary outcomes:

• Increased risk of relapse in cases vs controls (HR, 1.6; 95% CI, 1.31–1.96; p < 0.0001)
• PFS reduced in cases vs controls (HR, 1.38; 95% CI, 1.16–1.63; p = 0.0003)

• No difference in risk of aGvHD, GvHD, or NRM between cases and controls

Key findings

For patients in remission, this study showed decreased OS and PFS, and increased risk of relapse for patients with pMDS-BP compared with dnAML. No difference existed between the two groups in patients with active disease at HSCT. OS and PFS were also decreased for patients with MPN-BP compared to pMDS-AML, along with increased risk of relapse.

Conclusion

To summarize, the relapse rate is higher in MPN-BP patients relative to patients with dnAML or pMDS-AML, reducing both OS and PFS. There is no associated change in NRM. Outcomes for MPN-BP patients with active disease at HSCT are similar to patients with active pMDS-AML.  

The study has some limitations, including the lack of central pathological review to confirm relapse, and that remission is confirmed solely on basic morphological criteria. Despite the large size of the study, comorbidity data was missing in one third of patients, precluding adjustment for these variables in the multivariate analysis.

Patients with MPN-BP continue to have poorer survival outcomes relative to pMDS-AML and dnAML in remission at HSCT. There is a subsequent need for further research and improvement in relapse-prevention strategies for MPN-BP patients undergoing HSCT.