Chronic neutrophilic leukemia with plasma cell dyscrasia: friends or relatives?

Chronic neutrophilic leukemia (CNL) is a rare clonal myeloproliferative neoplasm that was first described in 1920 by Tuohy [1]. It typically occurs in elderly patients, equally in males and females. The diagnostic criteria, as defined by the World Health Organization (WHO), are:

1. A sustained peripheral blood leukocytosis (≥ 25 × 10⁹/L) with neutrophils accounting for > 80%. Typically there is toxic granulation and Döhle bodies, minimal left shift (< 10%) and < 1% myeloblasts;

2. Granulocytic hyperplasia in the bone marrow with < 5% myeloblasts;

3. Hepatosplenomegaly;

4. The exclusion of reactive causes of neutrophilia;

5. No Philadelphia (Ph) chromosome or BCR/ABL1 fusion gene;

6. No rearrangement of PDGFRA, PDGFRB or FGFR1;

7. No evidence of another myeloproliferative, myelodysplastic or myelodysplastic/myeloproliferative neoplasm.

Proof of the clonal nature of CNL has been confirmed in some patients on the basis of karyotypic abnormalities [e.g. del(20q), + 8 or + 21], X-inactivation patterns and progenitor cell assays. The JAK2 V617F mutation (homozygous or heterozygous) has been reported in some patients and its detection simplifies the diagnosis by excluding physiological neutrophilias [2–5]. Recently CSF3R mutations (predominantly missense exon 14 mutation CSF3R T618I) have been reported in more than 80% of patients with CNL [6, 7]. Not only will this assist in diagnosis, but it may provide the opportunity for targeted inhibitor therapy.

Although CNL is rare, there are a number of reports associating it with plasma cell dyscrasias, either monoclonal gammopathy of undetermined significance (MGUS) or plasma cell myeloma [8–10]. The paraproteinemias, when occurring with CNL, tend to show a disproportionate excess of lambda light chain restriction [8]. The significance of this association is unclear, and has raised the question of whether the CNL is, in fact, a reactive leukemoid reaction in patients with a primary plasma cell disorder. In this issue, Nedeljkovic and co-workers provide evidence that the neutrophilia in a patient with a paraproteinemia, in a presumed case of CNL, is clonal and not reactive [11]. The patient, an 81-year-old man with a sustained unexplained neutrophilia, had homozygous JAK2 V617F in circulating leukocytes (by high-resolution melt analysis and Sanger sequencing) and a plasma cell neoplasm (16 g/L immunoglobulin G [IgG] kappa paraprotein). The case supports the results from a similar case of JAK2 V617F-positive CNL with MGUS and confirms the clonal nature of the myeloid proliferation [7]. This is an important finding, since it had previously been postulated that the neutrophilic proliferation may be secondary to plasma cell derived cytokines [8]. Interestingly, CSF3R mutations have not been detected in five patients with CNL with MGUS [7].

What is the natural history of this association of a concurrent clonal lymphoid neoplasm in CNL? Is this a “chance” occurrence of two unrelated hematological malignancies in one patient, or are the neoplasms clonally related? Several published cases have described concomitant JAK2 V617F-positive myeloid neoplasms together with chronic myeloid leukemia (CML), BCR/ABL1-positive, with most authors favoring that the two myeloid neoplasms originated from the same aberrant stem cell [12–18]. This was proven in a case of concurrent JAK2 V617F-positive idiopathic myelofibrosis and KIT D816V mutated systemic mastocytosis. Using single cell laser-microdissection, the two mutations, JAK2 and KIT, were shown to be present in the neoplastic cells of both disease components [19].

However, the contrary view, that the two myeloid neoplasms are unrelated, has also been described. One such example is a case of concurrent polcythemia vera JAK2 V617F-positve and CML BCR/ABL1-positive. In this patient the JAK2 V617F allele burden fell when BCR/ABL1 mRNA increased and then reappeared on commencement of imatinib [14]. The implication of this observation was that there were two independent clones; the JAK2 V617F clone only achieved clonal dominance when the BCR/ABL1 clone was suppressed by imatinib. In another example of JAK2 V617F-positive refractory anemia with ring sideroblasts and thrombocytosis and concurrent CML BCR/ABL1-positive, in situ polymerase chain reaction (PCR) demonstrated increasing JAK2 V617F-positive bone marrow cells during imatinib therapy [13]. Both of these reports therefore supported the alternative view that there were two independent abnormal hemopoietic stem cell clones.

Concurrent JAK2 V617F-positive myeloid neoplasms with concurrent lymphoid malignancies also occur. This is not surprising, since chronic lymphocytic leukemia and MGUS occur in 10% of people over 80 years of age, and hence a chance association is plausible. In a recent review of 185 Ph-negative myeloproliferative neoplasms, the most common concurrent lymphoid neoplasm was chronic lymphocytic leukemia. In this study, genetic analysis showed that the JAK2 V617F-positive myeloid and lymphoid neoplasms were unrelated [20]. In another report, Stijnis et al. also showed no evidence of the JAK2 V617F mutation in purified monoclonal B-cells in two patients with polycythemia vera and a concurrent monoclonal B-cell disorder [21]. They concluded that the myeloid and lymphoid neoplasms merely coexisted, and there was no evidence that the proliferative behavior of the neoplastic B-cells was mediated through the JAK2 V617F mutation. Both of these reports therefore favor clonal independence of concurrent lymphoid neoplasms in JAK2 V617F-positive myeloproliferative neoplasms.

The case reported in this journal by Nedeljkovic and colleagues emphasizes the fact that, at least in some cases, the neutrophilia associated with plasma cell myeloma may be clonal in nature, based on their finding of JAK2 V617F [11]. However, the question that still remains is why there is a particular association of CNL with concurrent plasma cell neoplasms. Are the CNL and plasma cell disorder clonally related, or is this an incidental occurrence? Although the literature suggests it unlikely that the plasma cell proliferation would be JAK2 V617F positive, this could be tested [22,23]. Available methods to assess the JAK2 mutation status of the plasma cells include:

1. In situ PCR on bone marrow sections and assessment of morphologically identified plasma cells [13];

2. PCR of laser microdissected plasma cells extracted from sections of bone marrow;

3. PCR of pure populations of plasma cells separated from aspirated bone marrow by flow sorting or magnetic bead separation based on cell phenotype (i.e. CD56/CD138 positive).

The application of one or more of these techniques in this case would have enhanced our understanding of the clonal relationship between the plasma cell neoplasm and CNL. It would have resolved whether this particular association is due to the pluripotency of a neoplastic stem cell with the capacity to differentiate into both neutrophils and plasma cells, or is due to two independent neoplastic clones.

The report by Nedeljkovic and colleagues therefore provides insight into one issue: the JAK2 V617F mutation can be present in neutrophilia associated with a plasma cell dyscrasia [7, 11]. This finding of the clonal nature of the myeloid proliferation is still not a defining diagnostic criterion of CNL, according to the current WHO classification. Since JAK2 V617F-positive neutrophilia has been described in other reports, it is now timely for this, as well as CSF3R mutations, to be included as diagnostic criteria for CNL [2–6]. The outstanding issue that has not been addressed or resolved by this case is the mechanism by which CNL occurs in association with a plasma cell neoplasm. Is this merely the incidental concurrence of two unrelated neoplasms (“friends”), or are the two neoplasms derived from the same mutant progenitor (“relatives”)?

Potential conflict of interest:

Disclosure forms provided by the authors are available with the full text of this article at www.informahealthcare.com/lal.