Elsevier

Leukemia Research

Volume 31, Issue 11, November 2007, Pages 1503-1509
Leukemia Research

Monocytosis is an adverse prognostic factor for survival in younger patients with primary myelofibrosis

https://doi.org/10.1016/j.leukres.2006.12.025Get rights and content

Abstract

We recently developed a modified Dupriez prognostic scoring system (PSS) that effectively discriminated between high-, intermediate-, and low-risk young patients (age ≤60 years) with primary myelofibrosis (PMF) based on the respective presence of none, one, or two or more of the following parameters: hemoglobin <10 g/dL, leukocyte count <4 or >30 × 109 L−1, and platelet count <100 × 109 L−1. The current study (n = 129; median age, 52 years; 69 males) reveals, on multivariable analysis, that an absolute monocyte count of ≥1 × 109 L−1 carries an independent predictive value (p = 0.02), for an inferior survival, in addition to that provided by hemoglobin level (p = 0.002), platelet count (0.02), and leukocyte count (p = 0.16). The inclusion of the monocyte count as a fourth risk factor enabled the construction of a new and improved Mayo PSS; median survival was 173, 61, and 26 months in the absence of all four (low-risk), three (intermediate-risk), or two or less (high-risk) adverse features, respectively (p < 0.0001). The independent prognostic value of monocytosis was validated in a separate database of 97 patients with PMF from another institution.

Introduction

Primary myelofibrosis (PMF) is a clonal disorder of the multipotent hematopoietic stem cell, categorized as a chronic myeloproliferative disorder (CMPD) within WHO classification of myeloid neoplasms [1]. PMF is characterized by the clonal proliferation of megakaryocytic and granulocytic elements in the bone marrow, reactive (non-clonal) intramedullary reticulin and collagen fibrosis with extramedullary hematopoiesis, the latter being responsible for the characteristic leukoerythroblastosis and splenomegaly [1], [2]. The disorder is uncommon with a reported annual incidence of 0.5–1.46/100,000 population [3], [4]. Although considered a disease of the elderly (median age at diagnosis of 65 years), approximately 25–30% of patients are younger than 60 years of age at presentation [5], [6], [7], [8], [9].

Clinically, PMF evolves with worsening anemia, increasing splenomegaly and hepatomegaly and ultimately, mortality as a result of complications of marrow failure, portal hypertension or transformation to acute leukemia. The rate of disease progression, with median survivals of between 3.5 and 5 years, are consistently reported in large cohort studies. However, the clinical course of the individual patient is heterogeneous with reported survivals of less than 2 years to more than a decade [7], [10]. This heterogeneity in disease progression has provided the impetus to develop prognostic scoring systems (PSS) that aim to predict of the natural history of PMF for the individual patient. Several such PSS have indeed been shown to be of predictive value in outlining the anticipated clinical course, based on markers of poor prognosis at diagnosis, such as hemoglobin less then 10 g/dL, leukocyte count less than 4 or more than 30 × 109 L−1, more than 10% percentage of circulating immature white cell precursors, platelet count less than 100 × 109 L−1, presence of constitutional symptoms and one percent or more circulating blasts [6], [7], [9], [10]. Such prognostic information is critical in therapeutic decision making, as for patients with PMF options range from observation with best supportive care, to potentially morbid interventions such as investigational agents of unproven safety or efficacy, and allogeneic hematopoietic stem cell transplantation (SCT).

At present, allogeneic SCT represents the only treatment modality with curative potential. However, this procedure continues to be associated with substantial upfront morbidity and mortality that restricts its use in selected patients who could tolerate such an intervention, largely determined by age (generally ≤60 years) and lack of comorbidities [11], [12]. Therefore, age-specific PSS should be utilized when the goal is to identify such patients for whom the potential benefit of transplantation justifies the immediate risks upfront procedure-related mortality or long term morbidity. Two such PSS have been proposed, both based on readily available clinical parameters identified at diagnosis [8], [9]. That of Cervantes identifies two groups at high and low risk of progression based on hemoglobin (<10 g/dL), the presence of constitutional symptoms and circulating blasts (≥1%). That of Dingli modifies the original Dupriez PSS through the addition of a platelet count of less than 100 × 109 L−1 to hemoglobin (<10 g/dL) and leukocyte count (<4 or >30 × 109 L−1) to provide a more refined three-tear (high, intermediate and low risk) stratification based on objective data from the routine complete blood count (CBC) [9].

Recently several other parameters have been reported to carry unexpected prognostic significance in myeloproliferative disorders including the absolute lymphocyte count in chronic myelomonocytic leukemia (CMML) and the absolute monocyte count in bcr/abl negative chronic myeloid leukemia (CML) [13], [14]. In order to clarify the prognostic role of monocytosis and lymphocytosis we investigated these parameters in a consecutive cohort of strictly WHO-defined PMF patients for whom all necessary data at diagnosis was available. Our evaluation identifies an absolute monocyte count of more than 1 × 109 L−1 to be an independent predictor of inferior survival in PMF. Incorporation of monocyte count as one of the parameters in our original CBC-based PSS creates a new Mayo PSS that provides superior stratification of PMF patients into three distinct groups with significantly different prognostic outlook. Furthermore, the independent prognostic value of monocytosis was validated in a separate database of patients with PMF from the MD Anderson Cancer Center (MDACC).

Section snippets

Methods

Following the receipt of approval from the Mayo Foundation Institutional Review Board (IRB) in accordance with federal regulations, medical records of patients diagnosed with PMF before the age of 60 years seen at Mayo Clinic Rochester from 1 January 1976 through 31 December 2003 were reviewed. The diagnosis of PMF was defined according to WHO criteria with special precaution as to exclude cases of CMML [2]. Each chart was evaluated to confirm the diagnosis of PMF and patients with

Results

A total of 129 patients meeting defined inclusion criteria were diagnosed with PMF at or before the age of 60 years. The median age at diagnosis was 52 years (range: 18–60) with 19% and 3% at or below age 45 and 30 years, respectively and 69 (53%) patients were male. Clinical and laboratory parameters at diagnosis are outlined in Table 1. The median and range of absolute monocyte count was 0.39 × 109 L−1 (0.01–13.5) with a median monocyte percentage of the total leukocyte count of 4.5%

Discussion

The wide variability in the rate of disease progression of PMF, as measured in months to several years, mandates an individualized approach to therapy. This is of particular importance when the therapeutic intervention considered has the potential diminish the current quality and/or duration of life. Alternatively, not actively pursuing potentially life-saving therapy in the face of aggressive disease is unacceptable. At this time, allogeneic SCT is the only treatment modality with proven

Acknowledgments

There are no potential author conflicts of interest that relate to the current manuscript.

Contributions. Ayalew Tefferi wrote the paper, performed the statistical analysis, collected patient data, and contributed patients to the study. Michelle Elliott wrote the paper, collected patient data, and contributed patients to the study. Srdan Verstovsek provided validation data from M. D. Anderson Cancer Center database of patients with primary myelofibrosis. David Dingli collected patient data and

References (24)

  • G. Barosi et al.

    A prognostic classification of myelofibrosis with myeloid metaplasia [see comments]

    Br J Haematol

    (1988)
  • G. Visani et al.

    Myelofibrosis with myeloid metaplasia: clinical and haematological parameters predicting survival in a series of 133 patients

    Br J Haematol

    (1990)
  • Cited by (0)

    View full text