Prognostic impact of a cytopenic phenotype in myelofibrosis

Patients diagnosed with primary myelofibrosis (PMF) commonly experience cytopenias, such as anemia, thrombocytopenia, and leukopenia.¹

These features are associated with two distinct clinical phenotypes, myeloproliferative (characterized by increased leukocyte and platelet levels) and myelodysplastic (characterized by cytopenias involving one or more hematopoietic lineages). A myelodysplastic cytopenic phenotype (CP) is often associated with poor prognosis.¹

Coltro et al.¹ recently published a study in Blood Cancer Journal investigating the phenotypic and prognostic correlates associated with a CP in a large cohort of patients diagnosed with PMF.¹ The authors placed an emphasis on distinguishing between prefibrotic and overt PMF. We summarize the results below.

Results

Characteristics

The study included 431 patients, of which 216 were diagnosed with prefibrotic PMF and 215 with overt PMF. Patient characteristics are shown in Table 1.

Table 1. Patient characteristics*

Cytopenia types and frequency

The presence of at least one of the following cytopenias was used to define a cytopenic phenotype:

• Leukopenia (leukocytes <4 × 10⁹/L)
• Sex-adjusted anemia (hemoglobin <11 g/dL for males and <10 g/dL for females)
• Thrombocytopenia (platelets <100 × 10⁹/L)

The cytopenia types recorded in both patient groups are shown in Figure 1 and the total number of cytopenias in both patient groups is shown in Figure 2.

 

Associations between a cytopenic phenotype and clinical and genetic factors

Prefibrotic PMF

Patients diagnosed with prefibrotic PMF with a CP were more likely to exhibit the following factors compared with those with a myeloproliferative phenotype:

• Male
• Older age
• Higher peripheral blood blasts and CD34⁺ cells
• Higher serum lactate dehydrogenase levels
• Higher prevalence of splenomegaly
• Higher prevalence of hepatomegaly
• Constitutional symptoms
• Bone marrow fibrosis (Grade 1)

Patients with a CP were also more likely to have cytogenetic abnormalities and a very high-risk karyotype compared with those with the myeloproliferative phenotype. There was also a higher likelihood of JAK2-unmutated or triple-negative genetic profiles among patients with a CP.

Significantly enriched non-driver mutations in patients with prefibrotic PMF included the following:

• ASXL1
• N/KRAS
• U2AF1
• RUNX1
• SETBP1
• CUX1
• ≥2 high molecular risk mutations

Overt PMF

Patients diagnosed with overt PMF with a CP were associated with the following factors:

• Older age
• Higher CD34⁺ cell count
• Higher prevalence of bone marrow fibrosis (Grade 3)
• Lower JAK2 mutant burden
• Triple negative
• U2AF1 mutation

Patients with two or more cytopenias had a higher frequency of karyotype abnormalities and mutations in CBL and U2AF1.

Associations between phenotype and survival

Among patients with prefibrotic PMF, a total of 76 (35%) deaths were reported after a median follow-up time of 76 months, with a median overall survival (OS) of 149 months. Patients with prefibrotic PMF and a CP had a significantly worse OS compared with patients with prefibrotic PMF and a proliferative phenotype (p < 0.0001; Figure 3). Patients with a CP were also associated with significantly shorter progression-free survival, at a median of 33 months, compared with 193 months in patients with a myeloproliferative phenotype (p < 0.0001).

For overt PMF, the OS of patients with a CP was significantly shorter compared with those with a myeloproliferative phenotype (p = 0.0026; Figure 3).

Moreover, patients with overt PMF and two or more cytopenias experienced a significantly worse OS compared with patients with just one cytopenia (p = 0.0146) and patients with overt PMF and a severe cytopenia were associated with a worse OS compared with those with a moderate cytopenia (p < 0.0001; Figure 4). 

Multivariate analysis

Among patients with profibrotic PMF, multivariate analysis identified the following factors as independent predictors of a worse OS:

• Moderate/severe anemia
• Leukocytosis
• Constitutional symptoms
• High molecular risk

Among patients with overt PMF, multivariate analysis identified the following independent predictors for inferior OS:

• Severe thrombocytopenia

• Severe anemia
• Peripheral blast count ≥2%
• High molecular risk
• ≥2 high molecular risk associated genes

Leukemic transformation

In the study’s most recent follow-up, 20 patients with profibrotic PMF experienced a transformation to acute leukemia. The 5‑year cumulative incidence rate of leukemic transformation in patients with prefibrotic PMF was significantly higher in those with CP compared with those with a proliferative phenotype (p < 0.0001; Figure 5).

Among the patients with over PMF, 28 experienced transformation to acute leukemia. However, the difference in the 5-year cumulative incidence rate of transformation was not statistically different between patients with a CP or a myeloproliferative phenotype (p = 0.4; Figure 5).

Progression of prefibrotic PMF to overt PMF

Patients with prefibrotic PMF with a CP recorded a significantly higher 5-year cumulative incidence rate of progression to overt PMF (67%) compared with patients with a proliferative phenotype (15%) (p < 0.0001). Anemia and thrombocytopenia were also significantly more prevalent among patients with prefibrotic PMF who progressed to overt PMF within 5 years of diagnosis (p < 0.0001 for both anemia and thrombocytopenia; Figure 6).

Conclusion

Results from this study highlight that cytopenic features are more common in patients with overt PMF compared with prefibrotic PMF. Interestingly, mutations in U2AF1 were found to be a prominent abnormality of a CP in both prefibrotic and overt PMF. Importantly, the results also demonstrated that a CP is associated with a significantly reduced OS in both PMF subtypes, a higher risk of leukemic transformation, and fibrotic progression for patients with prefibrotic PMF. The authors highlight that a CP analysis is minimally invasive for patients, simple to achieve, and can be performed over an extended period.