Ruxolitinib Treatment Induces a Transition of Classical to Non-Classical Monocytes in Patients with Myelofibrosis

Introduction: Myelofibrosis (MF) is a clonal myeloproliferative neoplasm that develops de novo (primary myelofibrosis) or transforms from polycythemia vera or essential thrombocytosis. MF is characterized by stem cell-derived clonal myeloproliferation, abnormal cytokine expression, bone marrow fibrosis, anemia, splenomegaly, extramedullary hematopoiesis, constitutional symptoms, cachexia, leukemic progression, and shortened survival. Several studies suggested that clonal monocytes play a role in the pathobiology of MF. However whether a specific monocyte subpopulation is predominantly present in MF is largely unknown. Traditionally, three subpopulations of CD14⁺ monocytes have been identified: classical (CD14⁺⁺/CD16^-), intermediate (CD14⁺⁺/CD16⁺), and non-classical (CD14^dim+/CD16⁺⁺). Whether MF patients' monocyte subpopulations are different from those of normal individuals and how ruxolitinib treatment affects them has not been elucidated.

Methods: Using flow cytometry we first assessed the distribution of the three monocyte subpopulations in the bone marrow (BM) of healthy individuals and MF and then assessed their distribution in BM samples from phase I/II clinical trial of ruxolitinib in patients with primary or secondary MF (Verstovsek S. etal. N Engl J Med 12:1117, 2010).

Results: Using BM samples from 7 healthy individuals and 12 untreated MF patients we found a significant decrease in the percentage of monocytes (p =0.0061) in the mononuclear gate of untreated MF patients compared to normal individuals. However, the distribution of classical vs. non-classical monocyte subpopulations in MF was similar to that of normal BM (p =0.3, p =0.3, respectively). Remarkably, ruxolitinib treatment significantly altered the distribution of classical vs. non-classical monocyte subpopulations. During treatment (years 0-3, 3-6, 6-8) we identified a progressive increase in the percentage of monocytes in the mononuclear gate (p =0.1; p =0.04, and p =0.03, respectively) with a substantial increase in the non-classical monocyte subpopulation in years 0-3 and 3-6 of treatment (p= 0.04, p= 0.005, respectively) and a decrease in the classical monocytes (p =0.07, p =0.008, respectively). This trend reversed after 6-8 years of therapy (p =0.3, p =0.2, respectively). Importantly, in ruxolitinib-treated patients with a ≥50% spleen size reduction highest percentage of non-classical monocytes was observed during the first 3 years and years 3-6 of treatment (p =0.01, p =0.01, respectively). However during years 6-8 this difference was no longer detected and the percentage of non-classical monocytes was similar to the percentage detected in the pre-treatment BM samples (p =0.4). These changes correlated with response to ruxolitinib treatment. In patients with a <50% spleen size reduction the percentage of non-classical monocytes in years 0-3 of ruxolitinib treatment was significantly lower compared to patients with ≥50% spleen reduction (p =0.005), and patients with ≥50% spleen reduction show correlation between post-treatment spleen size and percentage of non-classical monocytes (p <0.0001, r=−0.4).

Conclusions: Taken together, our results suggest that ruxolitinib induces a transition of classical to non-classical monocyte subpopulation during the first years of ruxolitinib treatment and that this effect correlates with the patients' clinical response. Further studies aimed at exploring the role of monocytes and their subpopulations in the pathobiology of MF are warranted.