Purpose:

The JAK1/2 inhibitor ruxolitinib has demonstrated significant benefits for patients with myeloproliferative neoplasms (MPN). However, patients often lose response to ruxolitinib or suffer disease progression despite therapy with ruxolitinib. These observations have prompted efforts to devise treatment strategies to improve therapeutic efficacy in combination with ruxolitinib therapy. Activation of JAK–STAT signaling results in dysregulation of key downstream pathways, notably increased expression of cell-cycle mediators including CDC25A and the PIM kinases.

Experimental Design:

Given the involvement of cell-cycle mediators in MPNs, we sought to examine the efficacy of therapy combining ruxolitinib with a CDK4/6 inhibitor (LEE011) and a PIM kinase inhibitor (PIM447). We utilized JAK2-mutant cell lines, murine models, and primary MPN patient samples for these studies.

Results:

Exposure of JAK2-mutant cell lines to the triple combination of ruxolitinib, LEE011, and PIM447 resulted in expected on-target pharmacodynamic effects, as well as increased apoptosis and a decrease in the proportion of cells in S-phase, compared with ruxolitinib. As compared with ruxolitinib monotherapy, combination therapy led to reductions in spleen and liver size, reduction of bone marrow reticulin fibrosis, improved overall survival, and elimination of disease-initiating capacity of treated bone marrow, in murine models of MPN. Finally, the triple combination reduced colony formation capacity of primary MPN patient samples to a greater extent than ruxolitinib.

Conclusions:

The triple combination of ruxolitinib, LEE011, and PIM447 represents a promising therapeutic strategy with the potential to increase therapeutic responses in patients with MPN.

Translational Relevance

The JAK1/2 inhibitor ruxolitinib has demonstrated significant benefits for patients with myeloproliferative neoplasms (MPN). However, patients often lose response to ruxolitinib or suffer disease progression despite therapy with ruxolitinib. Here, we demonstrate that the combination of ruxolitinib, the PIM kinase inhibitor PIM447, and the CDK4/6 inhibitor LEE011 demonstrates improved efficacy as compared with ruxolitinib alone. Taken together, these results offer a novel mechanism-based combination regimen that may provide clinical benefit to patients with MPN beyond the therapeutic benefits of JAK kinase inhibition monotherapy, and which warrant clinical investigation.

The Philadelphia chromosome–negative myeloproliferative neoplasms (MPN) are hematopoietic stem cell (HSC)–initiated malignancies characterized by aberrant JAK–STAT pathway activation, which can result from somatic mutations in the JAK2, MPL, or CALR genes. The MPNs polycythemia vera (PV), essential thrombocythemia (ET), and myelofibrosis (MF) share a propensity toward thrombohemorrhagic events, progressive bone marrow failure, organomegaly, and increased risk of transformation to acute myeloid leukemia (1).

The JAK1/2 inhibitor ruxolitinib is FDA-approved for the treatment of PV and MF. Ruxolitinib has demonstrated significant clinical benefits in these patients, including reduction of splenomegaly and improvement in symptom burden. However, ruxolitinib monotherapy does not markedly alter the natural history and biology of disease in the majority of patients with MPN (2).

JAK2V617F has been demonstrated to promote G1–S cell-cycle progression in leukemia cell lines and Ba/f3 cells (3). This is mediated through several mechanisms including increased expression of CDC25A (a regulator of CDK2) via STAT5-mediated effects on translation (4). Indeed, CDC25A inhibition results in reduced proliferation of JAK2V617F cell lines and erythroid progenitors (4). Further, CDC25A expression is upregulated in primary MPN patient samples (4). JAK2V617F-mediated phosphorylation of STAT5 is also associated with increased cyclin D2 expression and decreased expression of the cell-cycle inhibitor p27Kip1 (3). The Pim family of serine/threonine kinases regulates CDC25A and p27Kip1. PIM-1 phosphorylates CDC25A, leading to an increase in its phosphatase activity. Pim kinases can suppress P27Kip1 transcription, and Pim-mediated phosphorylation of p27Kip1 leads to its nuclear export and proteasomal degradation (5). The expression of PIM kinases is regulated by JAK–STAT signaling (6, 7), and expression of PIM-1 and PIM-2 is upregulated in JAK2-mutant MPN patient samples versus normal controls (8).

These data collectively indicate that cell-cycle mediators regulated by activated JAK–STAT contribute to MPN pathogenesis, thus identifying the cell cycle as a potential therapeutic target in MPN. Importantly, CDK6 and Cyclin D expressions are upregulated in primary MPN patient samples (8), and preclinical studies have demonstrated a role for CDK6 in MPN pathogenesis (9). As well, several preclinical studies have demonstrated efficacy of combination ruxolitinib and PIM kinase inhibition in MPN models and primary patient samples, including reducing colony formation of primary MPN cells and inducing apoptosis (10–12). Collectively, these observations have led us to investigate the hypothesis that combination therapy with ruxolitinib, the pan-PIM kinase inhibitor PIM447 (13), and the CDK4/6 inhibitor LEE011 (14) may result in synergistic therapeutic efficacy. We identify that the combination of ruxolitinib and PIM447 or the combination of ruxolitinib and LEE011, as well as the combination of all three drugs, demonstrates improved efficacy as compared with ruxolitinib alone. Taken together, these results offer a novel potential therapeutic approach for the treatment of patients with MPN.

Murine models

For the UKE-1 model, 10 million viable cells were subcutaneously implanted with Matrigel in the upper right flank of SCID-beige mice. Tumor volume was determined by measurement with calipers and calculated using a modified ellipsoid formula, where tumor volume (mm3) = [((l × w2) × 3.14159)) /6], where l is the longest axis of the tumor and w is perpendicular to l.

Retroviral transplant studies were performed as previously described (15, 16). Briefly, whole bone marrow cells were isolated from Balb/C donor mice, positively selected for cKit+ expression, and transduced with hMPLW515L-IRES-GFP retrovirus and injected into lethally irradiated Balb/C (BALB/cJ stock# 000651 purchased from Jackson Laboratories) recipients. 1.5 × 106 cells were injected into recipients.

Jak2V617F knock-in mice have been described previously (17). Floxed mice were crossed to the IFN-responsive Vav-Cre deleter line (purchased from Jackson Laboratories; cat. #008610). Eight-week-old female CD45.1 (B6.SJL-Ptprc<a>/Boy; Stock #: 4007) mice were purchased from Taconic and were used for secondary Jak2V617F transplantation experiments. All purchased recipient mice were used at 8 to 10 weeks, and only females were used. For secondary transplant experiments, 2.5 × 106 whole bone marrow cells were transplanted via tail vein injection with 1 × 106 CD45.1 support wild-type bone marrow cells into congenic wild-type recipients.

Nonlethal bleeds were performed 14 days after transplantation and every 14 to 30 days thereafter to assess disease severity. Fourteen days following injection, mice were randomized to receive vehicle, 60 mg/kg ruxolitinib twice daily by oral gavage, or combined 60 mg/kg ruxolitinib twice daily by oral gavage, 12.5 mg/kg PIM447 daily, and 37.5 mg/kg LEE011 daily. All mice were bled at day 14 following start of treatment.

Animal care was in strict compliance with institutional guidelines established by the Novartis Institute for Biomedical Research, Memorial Sloan Kettering Cancer Center, and the Association for Assessment and Accreditation of Laboratory Animal Care International. Use of mice for these studies was approved by the Institutional Animal Care and Use Committees of both institutions.

Cells and cell culture conditions

BaF3-JAK2V617F and UKE-1 cell lines were grown in DMEM with 20% FBS. Ba/F3 murine pro B cells stably expressing human erythropoietin receptor (EpoR), JAK2V617F, and luciferase (Ba/F3-EpoRJAK2V617F-luc) cells were generated by Thomas Radimerski's lab at Novartis Institute for Biomedical Research. The cells were expanded in RPMI 1640 (ATCC, Cat. #30-2001) containing 10% inactivated FBS, 1% l-glutamine/penicillin/streptomycin (Corning, Cat. #30-009-C1), 1 μg/mL Puromycin (Sigma, Cat. #P9620), and 100 μg/mL Hygromycin B (Invitrogen, Cat. #10687-010), and stocks were generated for all the studies included in this report. BaF3Luc-EpoRJAK2V617F clone 8 and UKE-1 cells were culture in vitro for less than 2 weeks before use. They were tested free of mycoplasma and viral contamination in the IMPACT VIII PCR assay panel (RADIL, MU Research Animal Diagnostic Laboratory).

In vitro characterization

UKE-1 cells were plated at 1 × 106 cells/mL in 75 cc flasks. Twenty-four hours later, cells were treated with 300 nmol/L ruxolitinib, 300 nmol/L PIM447, and 1000 nmol/L LEE011, cells were removed after 72 hours of culture, and phospho and total proteins were measured using electrochemiluminescence detection with Meso Scale Discovery (MSD) or evaluated for apoptosis and cell-cycle assays according to the manufacturer's protocol. For the protein analysis, cells were washed twice and pellets were lysed, and then 10 to 20 μg of lysate samples were incubated on MSD multiplex microplates for phospho-SER780/total Rb, phospho/Total STAT5a,b, and phospho and Total BAD (MSD; Cat. #N45166B-1, N450IFA-1, and N45103B-1, respectively). Plates were washed and read using SECTOR Imager 2400 software (MSD).

Western blot

UKE-1 cells were cultured in RPMI+20% FBS+PS + 1 uM cortisol and treated with ruxolitinib, LEE011, or PIM447 either individually or in combination (doublet and triplet) for 24 hours. Cells were washed after treatment and lysed with lysis buffer containing protease and phosphatase inhibitors, followed by sonication at 40% amplitude (20-second ON, 2-second OFF) for 1 minute (Branson digital sonifier). Protein concentrations were estimated using the Bradford assay. Forty micrograms of protein was loaded per lane per sample, and proteins were transferred onto a nitrocellulose membrane. The membrane was blocked with 5% nonfat dry milk prepared in 0.1% Tween in TBS buffer. Primary and secondary antibody dilutions were made as recommended by manufacturer. Protein bands were visualized using the GE Amersham Imager 600.

The following antibodies were used from Cell Signaling Technology: Jak2 (Cat. #3230), pJak2(Tyr221; Cat. #3774), STAT5 (Cat. #94205), pSTAT5(Tyr694; Cat. #9359), Rb (Cat. #9313), pRb (Ser807/811; Cat. #8516), Bad (Cat. #9292), pBad (Cat. #5284). GAPDH (Cat. #8884) was used as a loading control. Secondary antibodies anti-Rabbit (Cat. #7074) and anti-mouse (Cat. #7076) also from Cell Signaling Technology were used accordingly.

Patient samples

Collection and utilization of samples from human subjects were conducted in strict accordance with the principles of the Declaration of Helsinki and Good Clinical Practice guidelines. The Institutional Review Board of Memorial Sloan Kettering Cancer Center approved sample collection and all experiments. Written informed consent was obtained from all human subjects.

Methylcellulose assays and liquid culture assays

Methylcellulose assays were performed as previously described (18). Peripheral blood mononuclear cells (PBMC) were isolated by Ficoll-Paque gradient separation. CD34+ cells were isolated from PBMCs using the Miltenyi autoMACS Pro Separator. PBMCs were seeded at a density of 20,000 cells/replicate, and CD34+ cells were seeded at a density of 2,000 cells/replicate into methylcellulose medium supplemented with cytokines (H4435; STEMCELL Technologies) with increasing concentrations of inhibitors. Cord blood units were utilized for normal controls, with CD34+ cells isolation as above. DMSO was added to control wells. Colonies propagated in culture were scored at day 10. All experiments were performed in duplicate. For liquid culture assays, PBMCs were isolated as above and cultured in RPMI 1640 without cytokines containing 10% inactivated FBS with vehicle, ruxolitinib, or triple therapy for 48 hours. Cells were then harvested and used for Western blot analysis.

PIM447 and LEE011 enhance the efficacy of ruxolitinib

In order to assess the therapeutic efficacy of PIM447 (PIM Kinase inhibitor) and LEE011 (CDK4/6 inhibitor) as single agents, each agent in combination with ruxolitinib, and the efficacy of all three agents together, we tested different therapeutic regimens in a JAK2V617F-mutant UKE-1 leukemia xenograft model (Fig. 1A). Tumor growth was observed in mice treated with vehicle and in mice treated with single-agent PIM447, LEE011, or ruxolitinib. Doublet combinations of ruxolitinib and PIM447, ruxolitinib and LEE011, and PIM447 + LEE011 induced tumor regression. However, none of the doublet combinations achieved complete tumor regression. Tumor growth resumed 14 days after discontinuation of treatment. By contrast, treatment with the triple combination of ruxolitinib, PIM447, and LEE011 achieved complete tumor regression during the treatment period. This response was maintained for more than 20 days after the discontinuation of treatment.

We next investigated combination therapeutic regimens in an allograft model in which recipient mice are injected with Ba/F3 cells expressing the Erythropoietin receptor and JAK2V617F, along with firefly luciferase (Ba/F3-EpoR-JAK2V617F-luc). Tumor burden was measured by IVIS imaging system as previously described (19). PIM447 monotherapy and ruxolitinib monotherapy reduced tumor burden by 1.5-fold and 8-fold, respectively, and the combination of these two agents reduced tumor burden by 24-fold (P < 0.0001; Supplementary Fig. S1A). Mice treated with the combination of ruxolitinib and PIM447 had a significantly reduced spleen size compared with ruxolitinib alone (P < 0.0001). In addition, a significantly greater reduction in mutant JAK2 allele burden was observed with dual therapy than with ruxolitinib alone (P < 0.05). The combination of ruxolitinib and LEE011 reduced total tumor burden by 13-fold (P < 0.0001) compared with 4-fold for each drug as a single agent (Supplementary Fig. S1B). Ruxolitinib plus LEE011 also significantly reduced spleen size compared with ruxolitinib monotherapy (P < 0.0001). Treatment with the triple combination of LEE011, PIM447, and ruxolitinib resulted in reduced tumor burden by 238-fold fold (P < 0.001) compared with 3.4-fold for ruxolitinib monotherapy (Fig. 1B), and significantly reduced spleen size compared with ruxolitinib (P < 0.0001). Importantly, the triple combination significantly reduced mutant JAK2 allele burden (85% reduction; P < 0.0001) compared with ruxolitinib monotherapy. Pooled analysis of doublet and triplet experiments demonstrated that triple therapy resulted in a significantly greater decrease in spleen size versus each of the doublet combinations (P < 0.01 for each doublet vs. triplet; Supplementary Fig. S1C). Notably, when the dose of either LEE011 or PIM447 was reduced by 50%, an increase in disease burden (3- and 1.5-fold over the efficacy observed with the triple combination, respectively) was observed. Reduction of the doses of all three agents led to an increase of 20-fold in disease burden (over efficacy with the full dose; Supplementary Fig. S1D). These data indicate that the combination of LEE011, PIM447, and ruxolitinib has increased therapeutic efficacy beyond that observed with ruxolitinib-based doublets or ruxolitinib monotherapy.

PIM447 and LEE011 demonstrate on-target pharmacodynamic effects alone and in combinations

We investigated the pharmacodynamic effects of triple therapy in the JAK2V617F-mutant UKE-1 cells. The triplet combination achieved similar reductions of pSTAT5 levels to those achieved by ruxolitinib alone (Fig. 1C; Supplementary Fig. S2A). The proapoptotic protein Bad is phosphorylated by PIM kinases (20, 21), and PIM447 exposure reduced levels of pBAD compared with ruxolitinib exposure as a single agent (P < 0.05) and in triple combination, consistent with an on-target effect (Fig. 1D; Supplementary Fig. S2B). CDK4/6 phosphorylates Rb (22), and exposure of cells to LEE011 decreased levels of phospho-Rb (pRb) to a greater extent than ruxolitinib (Fig. 1E; P < 0.05). However, the triple combination resulted in a greater decrease in pRb than LEE011 exposure as a single agent (Supplementary Fig. S2C; P < 0.01) or versus ruxolitinib (Fig. 1E; P < 0.01). These observations were confirmed on Western blot analysis of pSTAT5, pBAD, and pRB (Supplementary Fig. S2D). Ruxolitinib exposure resulted in a 50% reduction of cells in S phase. The triple combination significantly reduced the proportion of cells in S-phase versus ruxolitinib-exposed cells (Fig. 1F; P < 0.0001), and versus those treated with a doublet combination of ruxolitinib and PIM447 (Supplementary Fig. S2E; P < 0.001) or a doublet combination of ruxolitinib and LEE011 (Supplementary Fig. S2E; P < 0.01). The proportion of cells in G2–M phase was reduced by triple therapy versus ruxolitinib (Fig. 1F; P < 0.001) and versus ruxolitinib and LEE011 (P < 0.01) or ruxolitinib and PIM447 (P < 0.001; Supplementary Fig. S2F), thus indicating G1–S phase blockade. As well, a significant increase in the proportion of UKE-1 cells undergoing apoptosis (early and late) was observed in triple combination versus ruxolitinib (Fig. 1G; P < 0.0001), and versus the ruxolitinib and PIM447 (P < 0.001) or ruxolitinib and LEE011 (P < 0.01) doublets (Supplementary Fig. S2G). These observations suggest a combinatorial therapeutic effect of triple therapy beyond that observed with single-agent ruxolitinib or ruxolitinib-based doublet therapy.

Combined JAK/PIM/CDK inhibition attenuates MF disease phenotype in a MPLW515L-driven model

We next examined the effects of the triple combination regimen in an adoptive transfer model of MPLW515L-mutant MF. Bone marrow cells were isolated from Balb/C donor mice, selected for ckit, transduced with hMPLW515L-IRES-GFP retrovirus, and injected into lethally irradiated Balb/C recipients. Fourteen days following injection, mice were randomized to receive vehicle, ruxolitinib, or triple therapy. A cohort of mice were subject to terminal sacrifice (n = 3 per arm per experiment) on day 43, whereas the remainder of the cohort was observed for survival. Pooled analysis from two separate experiments demonstrated that overall survival of mice treated with triple therapy was significantly prolonged compared with either ruxolitinib treatment or vehicle (Fig. 2A; P < 0.0001). We observed significant reductions in white blood cell (WBC; Fig. 2B) in triple therapy mice when compared with either vehicle (P < 0.01) or ruxolitinib (P < 0.05), and in platelet (PLT) count (Fig. 2D) when comparing ruxolitinib with triple therapy (P < 0.05), prior to terminal sacrifice of a cohort of mice, on day 43 (data from one representative experiment are shown). No difference in Hgb was observed between arms (Fig. 2C). Of the remaining mice in the cohort after day 43, ruxolitinib-treated mice eventually developed progressive leukocytosis and thrombocytosis, indicative of disease progression, which was not observed in triple therapy–treated mice (Fig. 2B–D). Pooled analysis of three separate experiments demonstrated a significant decrease in spleen weights of triple therapy–treated mice (n = 10) versus vehicle (n = 10; P < 0.01), and significant decreases in liver weights in ruxolitinib (n = 10) and triple therapy–treated mice versus vehicle (P < 0.05 and P < 0.01, respectively), as well as a significant difference between liver weights in ruxolitinib and triple therapy–treated mice (P < 0.05; Fig. 2E and F). An increase in GFP in all treatment arms from days 16 to 37 was observed in a separate cohort of mice, but was significantly lower in triple therapy–treated mice versus ruxolitinib-treated mice at day 37 (P < 0.01; Fig. 2G). The bone marrow of triple therapy–treated mice had reduced cellularity and normal myeloid differentiation, which was not observed in ruxolitinib-treated mice (Fig. 2H). Bone marrow reticulin fibrosis was 2+ in the vehicle group, 1+ in ruxolitinib-treated mice, and was eliminated in the JAK/PIM/CDK inhibition group (Fig. 2I). Splenic extramedullary hematopoiesis was reduced in the ruxolitinib group versus vehicle but was markedly reduced in the triple therapy group, and splenic architecture was preserved in the triple therapy group (Fig. 2J; Supplementary Fig. S3A).

In order to test whether attenuation of disease manifestations differed between triple therapy–treated mice and doublet-treated mice, adoptive transfer was carried out as above, and mice were randomized to treatment with vehicle, ruxolitinib, ruxolitinib+LEE011, ruxolitinib+PIM447, or triple therapy (n = 4 mice per arm). Mice were followed for 42 days before terminal sacrifice. Significant increases in WBC were noted in vehicle- and ruxolitinib-treated mice by day 36, but not in doublet- or triple therapy–treated mice. Notably, a significant increase in thrombocytosis, indicative of progressive MPN, was observed in all treatment arms except the triple therapy arm (Supplementary Fig. S3B–S3D). At day 36 of treatment, only triple therapy–treated mice demonstrated a significantly lower peripheral blood GFP level versus ruxolitinib-treated mice (P < 0.01; Supplementary Fig. S3E). Finally, at the time of euthanasia, lower spleen weights were observed in doublet- and triple therapy–treated mice versus ruxolitinib (with no differences observed between doublet- and triple therapy–treated mice). However, only triple therapy–treated mice demonstrated a significant decrease in liver size (P < 0.05) versus ruxolitinib (Supplementary Fig. S3F and S3G). Analysis of GFP levels at the time of euthanasia demonstrated that triple therapy–treated mice, but not doublet-treated mice, had significantly lower GFP levels in the spleen (P < 0.05) and bone marrow (P < 0.01) compared with ruxolitinib (Supplementary Fig. S3H and S3I). Analysis of bone marrow reticulin fibrosis demonstrated a significant reduction in fibrosis in triple therapy–treated mice (P < 0.01) but not doublet-treated mice, in comparison with ruxolitinib (Supplementary Fig. S4A). A reduction in megakaryocytes was noted in the bone marrow (P = 0.05) and spleen (P < 0.05) of triple therapy–treated mice in comparison with ruxolitinib-treated mice (Supplementary Fig. S4B and S4C). These data support an enhanced efficacy of triple therapy over doublet combinations.

Triple therapy attenuates disease phenotype in Jak2V617F-mutant MPN

We next assessed the efficacy of combined JAK/PIM/CDK inhibition in a model of Jak2V617F-mutant MPN. Bone marrow from vav-Jak2V617F mice, which results in a PV phenotype, was transplanted into lethally irradiated recipients. Mice were allowed to engraft for 2 weeks, and then were randomized to treatment with vehicle, ruxolitinib, or triple therapy. Mice underwent serial complete blood counts every 2 weeks. A cohort of mice from each arm was euthanized approximately 7 weeks after treatment was initiated. When compared with vehicle, ruxolitinib or triple therapy significantly reduced WBC count (Fig. 3A; P < 0.01 and P < 0.001, respectively), with no differences in hemoglobin (Hgb) and hematocrit (HCT) noted (Fig. 3B and C). Significant reductions in PLT count were observed in ruxolitinib- and triple therapy–treated mice compared with vehicle (P < 0.01; Fig. 3D). Long-term follow-up of a separate cohort of mice was followed to day 112 (of which 3 mice per arm underwent planned sacrifice). No significant differences in survival were noted between the treatment arms at last follow-up (Supplementary Fig. S5A). An upward trend was noted in WBC, Hgb, and HCT in both the ruxolitinib- and triple therapy–treated arms through the time of last observation on day 112, but appeared to occur later in the triple therapy arm (Supplementary Fig. S5B–S5E).

Pooled analysis from two separate experiments revealed a significant reduction in spleen size in triple therapy–treated mice (n = 7) when compared with ruxolitinib-treated mice (n = 7; P < 0.05; Fig. 3E), and a significant decrease in liver size in triple therapy– and ruxolitinib-treated mice versus vehicle (n = 9; P < 0.01 for each comparison; Fig. 3F). Bone marrow of ruxolitinib-treated mice was hypercellular with increased myeloid maturation (compared with vehicle), whereas bone marrow from triple therapy–treated mice demonstrated normocellular marrow with appropriate maturation (Fig. 3G). A marked decrease in extramedullary hematopoiesis in the spleen (Fig. 3H) and liver (Fig. 3I) of the ruxolitinib-treated group (vs. vehicle) was observed, whereas the triple therapy group demonstrated little involvement by aberrant extramedullary hematopoiesis in the spleen and no extramedullary hematopoiesis in the liver. Finally, analysis of cell cycle of the LSK population from the bone marrow of treated mice demonstrated a significant increase in the proportion of LSK cells in G0–G1 phase versus LSK cells from ruxolitinib-treated mice (P < 0.05), with a trend toward decreased proportion of LSK cells in S-phase and G2–M phase in triple therapy–treated mice (Supplementary Fig. S5F).These data indicate increased efficacy, including decreased pathologic improvement, by combined JAK/PIM/CDK inhibition in Jak2V617F-mutant MPN.

Triple therapy results in modest effects on normal hematopoiesis

In order to test effect of triple therapy on normal hematopoiesis, wild-type C57BL/6 mice were randomized to treatment with vehicle, ruxolitinib, or triple therapy for 4 weeks. A significant decrease in WBC was noted in triple therapy–treated mice versus ruxolitinib at the end of 4 weeks of treatment (P < 0.01), although the WBC of triple therapy–treated mice was still within normal limits. No significant difference was noted in Hgb, HCT, or PLT over 4 weeks between ruxolitinib- and triple therapy–treated mice (Supplementary Fig. S6A–S6D). No difference in the body weights of dosed mice was noted between ruxolitinib- and triple therapy–treated mice over the 4 weeks of chronic therapy (Supplementary Fig. S6E).

Triple therapy eliminates disease-initiating potential in Jak2V617F mice

Prior studies demonstrate that JAK inhibitors are unable to eliminate the MPN stem cell (17). In order to test whether triple therapy is able to attenuate MPN disease-initiating capacity, whole bone marrow from vavCre-Jak2V617F mice (CD45.2) was transplanted into irradiated congenic mice. Mice were randomized to treatment with vehicle or triple therapy for 7 weeks. Mice were then euthanized, and whole bone marrow was transplanted into lethally irradiated CD45.1 recipients (along with support of wild-type CD45.1 bone marrow; Fig. 4A). Compared with recipients of triple therapy–treated bone marrow, recipients of vehicle-treated bone marrow developed progressive leukocytosis, erythrocytosis, and thrombocytosis, consistent with a PV phenotype (Fig. 4B–E). Median HCT on day 106 was 87.3% in the vehicle-treated group versus 46.1% in the triple therapy–treated group (P < 0.01). In addition, the percentage of CD45.2 cells detectable in the peripheral blood of recipient mice treated was greater in vehicle-treated mice (median 10.53%) versus triple therapy–treated mice (median 0.6%), and declined over time in triple therapy–treated mice, which was not observed in vehicle-treated mice (Fig. 4F). Finally, the LSK and CD71/Ter119 double-positive populations were expanded in recipients of vehicle-treated bone marrow but not in recipients of triple therapy–treated bone marrow (Fig. 4G and H; P < 0.05 for both populations). Analysis of the bone marrow demonstrated a left-shifted myeloid lineage in recipients of vehicle-treated donor bone marrow with morphology consistent with an MPN phenotype, whereas recipients of triple therapy–treated bone marrow or wild-type bone marrow demonstrated orderly myeloid maturation (Supplementary Fig. S7A). 2+ reticulin fibrosis was observed in vehicle-treated recipients, whereas no reticulin fibrosis was observed in triple therapy–treated recipients or wild-type mice (Supplementary Fig. S7B). Review of spleen sections demonstrated a distortion of the splenic architecture with infiltration of red pulp by excessive extramedullary hematopoiesis in vehicle-treated recipients, whereas triple therapy–treated recipients and wild-type mice demonstrated preserved splenic architecture with distinct red and white pulp (Supplementary Fig. S7C and S7D). Collectively, these data indicate that exposure to triple therapy is able to attenuate the disease-initiating potential of MPN stem cells. In order to exclude the possibility that the observed effect was due to nonspecific toxicity on the stem/progenitor populations, whole bone marrow from wild-type CD45.2 mice was transplanted into irradiated congenic mice. Mice were randomized to treatment with vehicle or triple therapy for 7 weeks. Mice were then euthanized and whole bone marrow was transplanted into lethally irradiated CD45.1 recipients (with support of wild-type CD45.1 bone marrow). No statistical difference was observed in the WBC, Hgb, HCT, or PLTs of mice that had been transplanted with vehicle- or triple therapy–treated bone marrow (Supplementary Fig. S7E–S7I). The percentage of peripheral blood CD45.2 cells was greater in the triple therapy–treated mice than in the vehicle-treated mice (Supplementary Fig. S7J; P < 0.05).

Combination therapy results in decreases in MYC and E2F target gene expression

In order to determine the impact of triple therapy on transcriptional output and key downstream pathways, RNA-sequencing from whole bone marrow of ruxolitinib-, doublet-, or triple therapy–treated MPLW515L retroviral transplant mice was carried out. Unsupervised hierarchical clustering demonstrated significantly differentially expressed genes (FDR < 0.05) in triple therapy– and ruxolitinib monotherapy–treated mice (Fig. 5A). Expression of Mpl and Ccnd2 was significantly upregulated in hematopoietic cells from mice treated with triple combination therapy, possible indicating on-target compensatory changes. Gene set enrichment analysis revealed reduced expression of MYC targets (a known target of PIM kinase inhibition; refs. 23, 24) and E2F targets (a known target of CDK4/6 inhibition; refs. 25, 26) with combination therapy (Fig. 5B). Negative enrichment for gene sets involved in DNA repair and the oxidative phosphorylation pathway were also noted, whereas gene sets involved in apoptosis and TNFα signaling via NF-κB were positively enriched. The later observation is consistent with a prior report in a JAK2V617F-Cdk6−/− murine model (9). Similar to this report, Btg2 and Klf4 were among the core-enriched upregulated genes identified, both of which have been reported to act as tumor suppressors (27, 28). As well, a negative regulator of NF-κB–dependent gene expression, Tnfaip3 (29), was also among the core-enriched genes. In order to investigate downstream pathways and provide validation of affected gene groups, we scanned promoter sequences of downregulated genes for the presence of transcription factor motifs implicated in driving these classes. Enrichment was observed for E2f1 and Klf (P = 1e-8 for each) as well as Stat (P = 1e-2) motifs in triple therapy–treated samples, indicating that there is a cis regulatory network consistent with the altered gene expression profiles observed above (Fig. 5C).

Although combining ruxolitinib with either PIM447 or LEE011 resulted in negative enrichment of MYC and E2F targets, only the triple combination demonstrated a significant negative enrichment for both target gene sets. As well, opposing effects on enrichment for angiogenesis and the P53 pathway gene sets were observed in triple therapy compared with ruxolitinib monotherapy (Fig. 5D). In the case of the angiogenesis gene set, Jag2 was noted to be among the core-enriched genes. This gene has been previously reported to play a role in promoting hematopoiesis after myelosuppression (30), and thus may represent a compensatory upregulation. In regard to the TP53 gene set, among the top enriched genes identified was Txnip. TXNIP is known to directly interfere in the MDM2–TP53 interactions and increase TP53 transcriptional activity (31). MDM2, which functions as an E3 ubiquitin ligase and leads to degradation of TP53, is known to be increased in MPN murine models and primary MPN patient samples (32, 33). Thus, disruption of the MDM2–Tp53 interaction may offer one potential explanation for the enrichment of the Tp53 pathway as well as the increased apoptosis, observed with the triple therapy regimen.

Triple therapy reduces colony formation of primary MPN patient samples

Given in vitro and murine data supporting the superior efficacy of combined JAK/PIM/CDK inhibition versus single-agent ruxolitinib, we next sought to determine if triple therapy had similar effects in primary MPN patient samples. We isolated mononuclear cells from the peripheral blood of six patients with MPN, as well as CD34+ cells in select cases (n = 2 PV, n = 4 MF). Cells were plated in methylcellulose with vehicle, ruxolitinib, doublets, or triple therapy. As monotherapy, ruxolitinib exposure induced dose-dependent reduction in colony formation (Fig. 6A–F; Supplementary Fig. S8; Supplementary Table S1). In all cases studied, triple therapy significantly decreased colony formation in comparison with ruxolitinib (P < 0.05 for A–D, F and P < 0.01 for E). This includes samples with high molecular risk mutations (e.g., ASXL1; Fig. 6B, C, and E). By contrast, doublets were only able to significantly reduce colony formation in comparison with ruxolitinib in two samples, (Fig. 6E and F), and in one of these cases, triple therapy significantly reduced colony formation in comparison with a doublet (P < 0.05; Fig. 6E). We next sought to determine the impact of exposure to ruxolitinib, doublets, or triple therapy on colony formation from normal human CD34+ cord blood–derived cells. Compared with vehicle, all treatments resulted in a decrease in colony formation. Compared with ruxolitinib, triple therapy did significantly reduce, but not eliminate, colony formation (P < 0.01, P < 0.05; Fig. 6G and H, respectively). The doublet of ruxolitinib and PIM447 also significantly reduced colony formation compared with ruxolitinib in both samples studied (P < 0.05 in both cases), whereas the doublet of ruxolitinib and LEE011 reduced colony formation in one samples (P < 0.05; Fig. 6G). Western blot analysis of an MF sample incubated in liquid culture for 48 hours with vehicle, ruxolitinib, or triple therapy demonstrated decrease in phosphorylated BAD and Rb with triple therapy treatment, consistent with on-target effects (Fig. 6I). Thus, triple therapy demonstrates activity greater than that observed with ruxolitinib in primary human MPN samples.

Although the majority of patients with MPN derive initial symptom improvement and spleen reduction with JAK inhibitor therapy, 51% of patients discontinue ruxolitinib after 3 years (2) often owing to disease progression, suboptimal response, or anemia (2, 34). Thus, there is a pressing clinical need to extend and enhance the effects of JAK1/2 inhibition.

Combined JAK/PIM/CDK inhibition in Jak2V617F mice or in mice engrafted with MPLW515L-GFP cells significantly attenuated disease phenotype, although treatment did not fully eradicate the disease during the relatively short duration of treatment. Importantly, treatment of Jak2V617F mice was able to attenuate the ability of whole bone marrow from these mice to perpetuate disease in a serial transplant, in contrast to JAK inhibition which cannot abrogate MPN stem cell activity (17). Exposure to CDK4/6 inhibitors has been associated with exit from the cell cycle into states of both cellular quiescence and senescence (35, 36). Transition to senescence of MPN stem cells could render cells unable to re-enter the cell cycle in response to mitogen exposure (as reviewed in ref. 37), thus effectively depleting the MPN stem cell population. Further mechanistic studies in MPN stem cells and in other myeloid malignancies which initiate and propagate in the HSC compartment will elucidate how these agents abrogate aberrant stem cell activity.

Although the rationale for utilizing this triple combination therapy was based on the predicted impact on cell-cycle regulation, other mechanistic possibilities likely contribute to the biological effects observed as well. Further, there appear to be biological effects observed in combination therapy beyond those observed with any of the single agents studied or doublet combinations (doublets were not tested in the Jak2V617F model, and therefore the relative efficacy of doublet therapy in this model remains undetermined). Thus, in aggregate, there appears to be at least an additive effect of the triple combination.

A recent report characterized the loss of Cdk6 in a JAK2V617F-mutant murine model, and demonstrated enhanced survival of JAK2V617F mice and reduced disease features such as splenomegaly, thus credentialing Cdk6 as a contributor to MPN pathogenesis (9). However, the changes in gene expression in LSK cells derived from this model were not fully recapitulated by use of the CDK4/6 inhibitor palbociclib, and thus suggesting that nonenzymatic functions of Cdk6 may play a role in MPN pathogenesis. However, this study did not fully assess in vivo biological effects of palbociclib or the combination of palbociclib and ruxolitinib. It is clear from our data that LEE011 contributes to the activity of the triple therapy regimen. Thus, it is likely that there is both enzymatic and nonenzymatic roles for Cdk6 in MPN pathogenesis, although off-target effects of LEE011 cannot be completely excluded.

In considering translation of these observations into potential clinical studies, additive toxicity of these drugs will require careful attention. Our in vivo studies of wild-type mice treated with the triple combination, as well as primary human cord blood CD34+ cells demonstrate increased myelosuppression relative to that observed with ruxolitinib monotherapy. Thus, development of combinatorial clinical studies of these drugs will likely require conservative dose-escalation schema and perhaps assessment of doublet combinations as a preliminary safety assessment. Notably, a clinical study of the triplet combination has been initiated (NCT02370706), although no data have been reported to date.

In sum, our results demonstrate that combining LEE011 or PIM447 with ruxolitinib enhances the therapeutic efficacy of ruxolitinib, which is more substantive with combined JAK/PIM/CDK inhibition. Clinical trials of these mechanism-based combination regimens are warranted and may provide clinical benefit to patients with MPN beyond the therapeutic benefits of JAK kinase inhibition monotherapy.

R.K. Rampal reports grants and personal fees from Constellation, Incyte, Celgene/BMS, and Stemline, and personal fees from Promedior, CTI, Blueprint, Jazz Pharmaceuticals, Galecto, Pharmaessentia, AbbVie, and Novartis outside the submitted work. M. Pinzon-Ortiz reports personal fees and other from Novartis Institutes for Biomedical Research, Inc. outside the submitted work; in addition, M. Pinzon-Ortiz has a patent for WO2015081083A1 pending and licensed to Novartis Institutes for Biomedical Research, Inc. B. Durham reports grants from NCI and American Society of Hematology during the conduct of the study. X. Rong reports personal fees from Novartis Institutes for Biomedical Research, Inc. outside the submitted work; in addition, X. Rong has a patent for WO2015081083A1 pending and licensed to Novartis Institutes for Biomedical Research. G. Vanasse reports other from Novartis Institutes for Biomedical Research outside the submitted work. Z.A. Cao reports being an employee at Novartis Institute of Biomedical Research during the conduct of the study. There is patent application on the combination of JAK inhibitor, CDK inhibitor, and PIM kinase inhibitor (publication number 20180071296). Z.A. Cao reports being a shareholder of Novartis and BMS during the 36 months prior to publication. He is a shareholder of Merck. R.L. Levine has consulted for Novartis during the conduct of the study (not related to this study). He is on the supervisory board of QIAGEN and is a scientific advisor to Imago, Mission Bio, Zentalis, Ajax, Auron, Prelude, C4 Therapeutics, and Isoplexis. He receives research support from and consulted for Celgene and Roche and has consulted for Incyte, Janssen, Astellas, and Morphosys. He has received honoraria from Roche, Lilly, and Amgen for invited lectures and from Gilead for grant reviews. No disclosures were reported by the other authors.

R.K. Rampal: Conceptualization, data curation, formal analysis, supervision, validation, investigation, visualization, methodology, writing–original draft, writing–review and editing. M. Pinzon-Ortiz: Investigation, writing–original draft. A.V. Hanasoge Somasundara: Formal analysis, investigation. B. Durham: Formal analysis. R. Koche: Formal analysis, methodology, writing–review and editing. B. Spitzer: Data curation, methodology, writing–original draft. S. Mowla: Investigation. A. Krishnan: Formal analysis, investigation, methodology. B. Li: Formal analysis, investigation. W. An: Formal analysis, investigation. A. Derkach: Data curation, formal analysis. S. Devlin: Formal analysis, investigation, writing–review and editing. X. Rong: Investigation. T. Longmire: Investigation. S.E. Eisman: Investigation. K. Cordner: Investigation. J.T. Whitfield: Investigation. G. Vanasse: Conceptualization, resources, investigation. Z.A. Cao: Conceptualization, resources, investigation. R.L. Levine: Conceptualization, resources, funding acquisition, writing–review and editing.

This study was supported by Cancer Center Support Grant/Core Grant to Memorial Sloan Kettering Cancer Center (P30 CA008748). R.K. Rampal is supported by NCI 1K08CA188529-01. This work was also supported by NCI R35 CA197594-01A1 (R.L. Levine) and NCI P01 CA108671 11 (R.L. Levine and R.K. Rampal).

The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked advertisement in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.

1.
Spivak
JL
. 
Myeloproliferative Neoplasms
.
N Engl J Med
2017
;
376
:
2168
81
.
2.
Verstovsek
S
,
Mesa
RA
,
Gotlib
J
,
Levy
RS
,
Gupta
V
,
DiPersio
JF
, et al
Efficacy, safety, and survival with ruxolitinib in patients with myelofibrosis: results of a median 3-year follow-up of COMFORT-I
.
Haematologica
2015
;
100
:
479
88
.
3.
Walz
C
,
Crowley
BJ
,
Hudon
HE
,
Gramlich
JL
,
Neuberg
DS
,
Podar
K
, et al
Activated Jak2 with the V617F point mutation promotes G1/S phase transition
.
J Biol Chem
2006
;
281
:
18177
83
.
4.
Gautier
E-F
,
Picard
M
,
Laurent
C
,
Marty
C
,
Villeval
J-L
,
Demur
C
, et al
The cell cycle regulator CDC25A is a target for JAK2V617F oncogene
.
Blood
2012
;
119
:
1190
9
.
5.
Morishita
D
,
Katayama
R
,
Sekimizu
K
,
Tsuruo
T
,
Fujita
N
. 
Pim kinases promote cell cycle progression by phosphorylating and down-regulating p27Kip1 at the transcriptional and posttranscriptional levels
.
Cancer Res
2008
;
68
:
5076
85
.
6.
Miura
O
,
Miura
Y
,
Nakamura
N
,
Quelle
Fw
,
Witthuhn
Ba
,
Ihle
Jn
, et al
Induction of tyrosine phosphorylation of Vav and expression of Pim-1 correlates with Jak2-mediated growth signaling from the erythropoietin receptor
.
Blood
1994
;
84
:
4135
41
.
7.
Shirogane
T
,
Fukada
T
,
Muller
JM
,
Shima
DT
,
Hibi
M
,
Hirano
T
. 
Synergistic roles for Pim-1 and c-Myc in STAT3-mediated cell cycle progression and antiapoptosis
.
Immunity
1999
;
11
:
709
19
.
8.
Rampal
R
,
Al-Shahrour
F
,
Abdel-Wahab
O
,
Patel
JP
,
Brunel
JP
,
Mermel
CH
, et al
Integrated genomic analysis illustrates the central role of JAK-STAT pathway activation in myeloproliferative neoplasm pathogenesis
.
Blood
2014
;
123
:
e123
33
.
9.
Uras
IZ
,
Maurer
B
,
Nivarthi
H
,
Jodl
P
,
Kollmann
K
,
Prchal-Murphy
M
, et al
CDK6 coordinates JAK2 (V617F) mutant MPN via NF-kappaB and apoptotic networks
.
Blood
2019
;
133
:
1677
90
.
10.
Huang
S-MA
,
Wang
A
,
Greco
R
,
Li
Z
,
Sun
F
,
Barberis
C
, et al
Combination of PIM and JAK2 inhibitors synergistically suppresses MPN cell proliferation and overcomes drug resistance
.
Oncotarget
2014
;
5
:
3362
74
.
11.
Mazzacurati
L
,
Lambert
QT
,
Pradhan
A
,
Griner
LN
,
Huszar
D
,
Reuther
GW
. 
The PIM inhibitor AZD1208 synergizes with ruxolitinib to induce apoptosis of ruxolitinib sensitive and resistant JAK2-V617F-driven cells and inhibit colony formation of primary MPN cells
.
Oncotarget
2015
;
6
:
40141
57
.
12.
Mazzacurati
L
,
Collins
RJ
,
Pandey
G
,
Lambert-Showers
QT
,
Amin
NE
,
Zhang
L
, et al
The pan-PIM inhibitor INCB053914 displays potent synergy in combination with ruxolitinib in models of MPN
.
Blood Adv
2019
;
3
:
3503
14
.
13.
Burger
MT
,
Nishiguchi
G
,
Han
W
,
Lan
J
,
Simmons
R
,
Atallah
G
, et al
Identification of N-(4-((1R,3S,5S)-3-amino-5-methylcyclohexyl)pyridin-3-yl)-6-(2,6-difluorophenyl)- 5-fluoropicolinamide (PIM447), a potent and selective proviral insertion site of moloney murine leukemia (PIM) 1, 2, and 3 kinase inhibitor in clinical trials for hematological malignancies
.
J Med Chem
2015
;
58
:
8373
86
.
14.
Kim
S
,
Loo
A
,
Chopra
R
,
Caponigro
G
,
Huang
A
,
Vora
S
, et al
LEE011: an orally bioavailable, selective samll molecuale inhibitor of CDK4/6-reactinvating Rb in cancer
.
Mol Cancer Therap
2013
;
12
.
15.
Bhagwat
N
,
Koppikar
P
,
Keller
M
,
Marubayashi
S
,
Shank
K
,
Rampal
R
, et al
Improved targeting of JAK2 leads to increased therapeutic efficacy in myeloproliferative neoplasms
.
Blood
2014
;
123
:
2075
83
.
16.
Rampal
R
,
Ahn
J
,
Abdel-Wahab
O
,
Nahas
M
,
Wang
K
,
Lipson
D
, et al
Genomic and functional analysis of leukemic transformation of myeloproliferative neoplasms
.
Proc Natl Acad Sci U S A
2014
;
111
:
E5401
10
.
17.
Mullally
A
,
Lane
SW
,
Ball
B
,
Megerdichian
C
,
Okabe
R
,
Al-Shahrour
F
, et al
Physiological Jak2V617F expression causes a lethal myeloproliferative neoplasm with differential effects on hematopoietic stem and progenitor cells
.
Cancer Cell
2010
;
17
:
584
96
.
18.
Kleppe
M
,
Koche
R
,
Zou
L
,
van Galen
P
,
Hill
CE
,
Dong
L
, et al
Dual targeting of oncogenic activation and inflammatory signaling increases therapeutic efficacy in myeloproliferative neoplasms
.
Cancer Cell
2018
;
33
:
785
7
.
19.
Abd-Elrahman
I
,
Kosuge
H
,
Wises Sadan
T
,
Ben-Nun
Y
,
Meir
K
,
Rubinstein
C
, et al
Cathepsin activity-based probes and inhibitor for preclinical atherosclerosis imaging and macrophage depletion
.
PLoS One
2016
;
11
:
e0160522
.
20.
Macdonald
A
,
Campbell
DG
,
Toth
R
,
McLauchlan
H
,
Hastie
CJ
,
Arthur
JS
. 
Pim kinases phosphorylate multiple sites on Bad and promote 14–3-3 binding and dissociation from Bcl-XL
.
BMC Cell Biol
2006
;
7
:
1
.
21.
Li
YY
,
Popivanova
BK
,
Nagai
Y
,
Ishikura
H
,
Fujii
C
,
Mukaida
N
. 
Pim-3, a proto-oncogene with serine/threonine kinase activity, is aberrantly expressed in human pancreatic cancer and phosphorylates bad to block bad-mediated apoptosis in human pancreatic cancer cell lines
.
Cancer Res
2006
;
66
:
6741
7
.
22.
Kato
J
,
Matsushime
H
,
Hiebert
SW
,
Ewen
ME
,
Sherr
CJ
. 
Direct binding of cyclin D to the retinoblastoma gene product (pRb) and pRb phosphorylation by the cyclin D-dependent kinase CDK4
.
Genes Dev
1993
;
7
:
331
42
.
23.
Zhang
Y
,
Wang
Z
,
Li
X
,
Magnuson
NS
. 
Pim kinase-dependent inhibition of c-Myc degradation
.
Oncogene
2008
;
27
:
4809
19
.
24.
Horiuchi
D
,
Camarda
R
,
Zhou
AY
,
Yau
C
,
Momcilovic
O
,
Balakrishnan
S
, et al
PIM1 kinase inhibition as a targeted therapy against triple-negative breast tumors with elevated MYC expression
.
Nat Med
2016
;
22
:
1321
9
.
25.
Markey
MP
,
Angus
SP
,
Strobeck
MW
,
Williams
SL
,
Gunawardena
RW
,
Aronow
BJ
, et al
Unbiased analysis of RB-mediated transcriptional repression identifies novel targets and distinctions from E2F action
.
Cancer Res
2002
;
62
:
6587
97
.
26.
Knudsen
ES
,
Witkiewicz
AK
. 
Defining the transcriptional and biological response to CDK4/6 inhibition in relation to ER+/HER2- breast cancer
.
Oncotarget
2016
;
7
:
69111
23
.
27.
Yuniati
L
,
Scheijen
B
,
van der Meer
LT
,
van Leeuwen
FN
. 
Tumor suppressors BTG1 and BTG2: beyond growth control
.
J Cell Physiol
2019
;
234
:
5379
89
.
28.
Evans
PM
,
Liu
C
. 
Roles of Krupel-like factor 4 in normal homeostasis, cancer and stem cells
.
Acta Biochim Biophys Sin
2008
;
40
:
554
64
.
29.
Coornaert
B
,
Carpentier
I
,
Beyaert
R
. 
A20: central gatekeeper in inflammation and immunity
.
J Biol Chem
2009
;
284
:
8217
21
.
30.
Guo
P
,
Poulos
MG
,
Palikuqi
B
,
Badwe
CR
,
Lis
R
,
Kunar
B
, et al
Endothelial jagged-2 sustains hematopoietic stem and progenitor reconstitution after myelosuppression
.
J Clin Invest
2017
;
127
:
4242
56
.
31.
Jung
H
,
Kim
MiJ
,
Kim
DOh
,
Kim
WS
,
Yoon
S-J
,
Park
Y-J
, et al
TXNIP maintains the hematopoietic cell pool by switching the function of p53 under oxidative stress
.
Cell Metab
2013
;
18
:
75
85
.
32.
Nakatake
M
,
Monte-Mor
B
,
Debili
N
,
Casadevall
N
,
Ribrag
V
,
Solary
E
, et al
JAK2(V617F) negatively regulates p53 stabilization by enhancing MDM2 via La expression in myeloproliferative neoplasms
.
Oncogene
2012
;
31
:
1323
33
.
33.
Lu
M
,
Wang
X
,
Li
Y
,
Tripodi
J
,
Mosoyan
G
,
Mascarenhas
J
, et al
Combination treatment in vitro with Nutlin, a small-molecule antagonist of MDM2, and pegylated interferon-alpha 2a specifically targets JAK2V617F-positive polycythemia vera cells
.
Blood
2012
;
120
:
3098
105
.
34.
Kuykendall
AT
,
Shah
S
,
Talati
C
,
Al Ali
N
,
Sweet
K
,
Padron
E
, et al
Between a rux and a hard place: evaluating salvage treatment and outcomes in myelofibrosis after ruxolitinib discontinuation
.
Ann Hematol
2018
;
97
:
435
41
.
35.
Choi
YJ
,
Li
X
,
Hydbring
P
,
Sanda
T
,
Stefano
J
,
Christie
AL
, et al
The requirement for cyclin D function in tumor maintenance
.
Cancer Cell
2012
;
22
:
438
51
.
36.
Kovatcheva
M
,
Liu
DD
,
Dickson
MA
,
Klein
ME
,
O'Connor
R
,
Wilder
FO
, et al
MDM2 turnover and expression of ATRX determine the choice between quiescence and senescence in response to CDK4 inhibition
.
Oncotarget
2015
;
6
:
8226
43
.
37.
Rodier
F
,
Campisi
J
. 
Four faces of cellular senescence
.
J Cell Biol
2011
;
192
:
547
56
.

Supplementary data