Targeting Insulin-like Growth Factor/Insulin Receptor Signaling to Overcome JAK2 Inhibitor Persistence in Myeloproliferative Neoplasms

Classic Philadelphia chromosome-negative myeloproliferative neoplasms (MPNs) are driven by aberrant activity of the JAK2 tyrosine kinase. While JAK2 inhibitors rarely induce molecular remission, they provide quality of life improvements for patients and thus will be a mainstay for MPN treatment for the foreseeable future. This near ubiquitous upfront ineffectiveness of JAK2 inhibitors must be overcome if they are going to contribute to a significant disease-altering therapy for MPNs. The insulin-like growth factor-1 receptor (IGF1R) can contribute to resistance to kinase inhibitor cancer therapeutics and a recent study demonstrated a requirement of IGF1R in JAK2-V617F-induced MPN in mice. Furthermore, IGF is known to play a role in neoplastic erythroid colony formation of hematopoietic progenitors from MPN patients. Thus, there is evidence for a potential role for IGF1R signaling in the context of JAK2 inhibition, particularly in the setting of JAK2 inhibitor resistance/persistence, a concept that remains underexplored.

We observed an increase in activated IGF1R and insulin receptor (IR), which is structurally similar to IGF1R, in JAK2-V617F-driven MPN model cells (UKE1, SET2, and HEL) that are persistently growing in high concentration of the JAK inhibitor ruxolitinib. This suggests cell signaling is re-wired to alter the activity of IGF1R/IR in this state. Interestingly, in ruxolitinib sensitive/drug naïve cells, we observed a rapid increase in IGF1R-beta and IR-beta protein levels in response to ruxolitinib treatment. These observations further suggest IGF1R/IR may contribute to the response of cells to ruxolitinib.

To assess the role of IGF1R/IR activation in both ruxolitinib-persistent cells and drug naïve cells, we utilized the IGF1R/IR selective kinase inhibitor OSI-906. OSI-906 weakly inhibited growth of JAK2-V617F-driven cells (GI50s between 1 and 8 uM), and cells continued to proliferate at a reduced rate in the presence of a high concentration of drug. The IGF1R/IR inhibitor had little effect on cell viability, but in combination with ruxolitinib, it synergistically suppressed cell growth in part due to augmented induction of apoptosis. This combination also synergistically suppressed the neoplastic (erythropoietin-independent) growth of erythroid colonies of hematopoietic progenitors from MPN patients, which supports previous work indicating a role for serum IGF1 in this erythroid colony formation.

Low concentration of IGF1R/IR inhibitor (0.1 uM) suppressed cell growth and viability of cells persistently growing in ruxolitinib yet had no effect on drug naïve cells. This along with our observation of enhanced activation of IGF1R/IR in these cells suggests that in the presence of ruxolitinib, cells may become more dependent on IGF1R/IR activity. We have always observed that high concentration of ruxolitinib is unable to reduce cell viability to zero in certain cell lines (UKE1 & HEL). Ruxolitinib (10 uM) induces a G1 cell cycle arrest in these cells and a sub-population of persister cells remains viable for several weeks. These persister cells remain competent for cell growth once ruxolitinib is removed. However, OSI-906 completely prevented the development of ruxolitinib persister cells. Culturing UKE1 cells in low serum sensitized cells to ruxolitinib and prevented the survival of persister cells. Supplementation of exogenous IGF-1 partially rescued this sensitivity and allowed for persister cells to survive, suggesting IGF-1 may contribute to the development of ruxolitinib persistence.

Finally, in addition to an increase in IGF1R/IR protein, ruxolitinib (10 uM) treatment also led to long-term depletion of IRS1/2 levels, the main signaling docking protein that binds to IGF1R/IR. This correlated with re-activation of the ERK pathway, which has been reported to be due to SHC binding to IGF1R in the absence of IRS1/2. Because these persister cells remain arrested in G1, ERK activation is not sufficient to support cell growth, but perhaps supports cell survival. This may highlight potential therapeutic targets to prevent ruxolitinib persistence. Identifying mechanistic aspects of cell survival in the presence of ruxolitinib could play a significant role in the development of combination therapies for MPNs. Our data support the notion that ruxolitinib treatment may alter the dynamics of IGF1R/IR signaling leading to tolerance of cells to ruxolitinib.