BONE Marrow Stromal CELL Paracrine Factors from PMF Patients, but NOT Healthy Donors, Protect JAK2^V617F CELLS from Ruxotinib-Induced Apoptosis

Janus kinases (JAK) comprise a small family of cytoplasmic protein tyrosine kinases, which play an important role in the initiation of cytokine-triggered signaling events via signal transducer and activator of transcription (STAT) proteins. The activating somatic mutation JAK2^V617F is highly prevalent among patients with myeloproliferative neoplasias (MPNs), which has prompted the development of JAK2 inhibitors for patients with these malignancies.

Ruxolitinib (INCB018424), an oral inhibitor of JAK1 and JAK2, induces marked and durable clinical benefits in patients with primary myelofibrosis (Verstovsek S et al. N Engl J Med 366:799-807, 2012) regardless of the presence of the JAK2V617F mutation.

Ruxolitinib induces a dose-dependent suppression of phosphorylated signal transducer and activator of transcription 3 (STAT3), a marker of the JAK signaling pathway; however, after 12 cycles of therapy only a 13% mean suppression of JAK2^V617F allele burden from baseline was observed in COMFORT-1 study. Therefore, we hypothesized that extracellular signals negate the effects of JAK2 inhibitors on malignant cells. To test our hypothesis, we devised a co-culture platform to test the interaction between the JAK2^V617F cells and BM stromal cells. Ruxolitinib inhibited proliferation and induced marked apoptosis of the human JAK2^V617F mutated SET2 (IC50 56nM) and UKE-1 (IC50 329nM) cell lines, but had no effect on wild-type JAK2 human mast MHC1.1 and MHC1.2 cells. However, the anti-proliferative effect of ruxolitinib was markedly attenuated when JAK2^V617F mutated cells were co-cultured with either primary patient samples or immortalized PMF patient-derived human BM mesenchymal stromal cells (TM-R1 MSCs). Specifically, co-culturing JAK2^V617F–positive cells with TM-R1 MSCs hampered the anti-proliferative and pro-apoptotic effect of ruxolitinib and inhibited its ability to dephosphorylate STAT3 and STAT5. However, when JAK2^V617F–positive cells were co-cultured with MSCs from healthy donors the protective effect was not observed. Subsequent experiments demonstrated that the protective effect of the TM-R1 MSCs was due to a paracrine secretion of cytokines. The cytokine expression profile of supernatants from co-cultures of TM-R1 MSCs and JAK2^V617F-positive cells significantly differed from those of supernatants from co-cultures of normal MSCs and JAK2^V617F-positive cells (with or without ruxolitinib). Interleukin (IL)-6 was elevated in both co-cultures whereas basic fibroblast growth factor (b-FGF-basic p= 1.05E-17) and CXCL10/IP10 (p= 2.19E-4) were significantly lower in normal MSC/JAK2^V617F supernatants. Furthermore, immunofluorescence staining demonstrated that nuclear factor-light chain-enhancer p-105 (p-NFK-β-105) was localized to the nucleus in TM-R1 MSCs, whereas in normal MSCs p-NFK-β-105 was found in the cytoplasm. Our results suggest that cytokines secreted by the bone marrow stromal play a significant role in protecting JAK2^V617F–positive cells from JAK2 inhibitor therapy, highlighting the role of the bone marrow microenvironment not only in the pathogenesis of MPNs but also in resistance to JAK2–directed therapies.