Improved targeting of JAK2 leads to increased therapeutic efficacy in myeloproliferative neoplasms

Abstract

The discovery of JAK2/MPL mutations in patients with myeloproliferative neoplasms (MPN) led to clinical development of Janus kinase (JAK) inhibitors for treatment of MPN. These inhibitors improve constitutional symptoms and splenomegaly but do not significantly reduce mutant allele burden in patients. We recently showed that chronic exposure to JAK inhibitors results in inhibitor persistence via JAK2 transactivation and persistent JAK-signal transducer and activator of transcription signaling. We performed genetic and pharmacologic studies to determine whether improved JAK2 inhibition would show increased efficacy in MPN models and primary samples. Jak2 deletion in vivo led to profound reduction in disease burden not seen with JAK inhibitors, and deletion of Jak2 following chronic ruxolitinib therapy markedly reduced mutant allele burden. This demonstrates that JAK2 remains an essential target in MPN cells that survive in the setting of chronic JAK inhibition. Combination therapy with the heat shock protein 90 (HSP90) inhibitor PU-H71 and ruxolitinib reduced total and phospho-JAK2 and achieved more potent inhibition of downstream signaling than ruxolitinib monotherapy. Combination treatment improved blood counts, spleen weights, and reduced bone marrow fibrosis compared with ruxolitinib alone. These data suggest alternate approaches that increase JAK2 targeting, including combination JAK/HSP90 inhibitor therapy, are warranted in the clinical setting.

Figure 1

JAK2 is required for initiation of MPLW515L-mediated disease. Deletion of Jak2 following engraftment of MPLW515L-transduced BM leads to significant reduction in (A) WBC counts, (B) platelets, (C) mutant allele burden in terms of GFP-positive cells in PB (P < .005), and (D) spleen (P < .05) and (E) liver size (P < .05) as compared with controls. (F) BM fibrosis assessed by reticulin staining is absent in Jak2-deleted BM. (Data are represented at mean ± standard error of the mean [SEM]; Jak2^+/+, n = 4; Jak2^−/−, n = 5).

Figure 2

JAK2 plays a critical role in the survival of MPN clone. Excision of Jak2 following disease establishment led to reduction in (A) WBC, (B) platelets, (C) spleen sizes, and (D) GFP-positive cells in BM (P < .005). Loss of JAK2 results in significant improvement in disease features including (E) restoration of splenic architecture, (F) decrease in myeloid infiltration in liver, and (G) reduction in BM fibrosis. (H) There was also a reduction in CD11b⁺Gr1⁺ cells and MEP population in MPLW515L-transduced mice lacking Jak2. (Data are represented at mean ± SEM; Jak2^+/+, n = 5; Jak2^−/−, n = 6).

Figure 3

Deletion of Jak2 is more effective than JAK inhibitor therapy at reducing disease burden. MPLW515L-transduced mice were treated with vehicle, 60 mg/kg ruxolitinib or pI:pC to excise Jak2. (A) Deletion of Jak2 led to significant reduction in mutant allele burden in BM as compared with inhibitor treatment (P < .005). It also resulted in further decrease in (B) blood counts and (C) spleen size in Jak2-deleted mice compared with mice that received ruxolitinib alone (P < .005). (D) Loss of JAK2 also leads to reduction in MEP and CD11b⁺Gr⁺ proportions with a dramatic decrease in the contribution of mutant (GFP⁺) cells. (Data are represented at mean ± SEM; vehicle, n = 5; Rux, n = 9; Jak2 deleted, n = 6).

Figure 4

Combination of JAK2 and HSP90 inhibition is more efficacious than JAK inhibitor monotherapy. (A) Combination of ruxolitinib and PU-H71 resulted in JAK2 degradation and inhibition of downstream signaling in splenocytes of MPLW515L-transduced mice (all lanes were run on the same gel). After 2 weeks of drug treatment, the ruxolitinib and PU-H71 combination arm had significantly lower (B) WBC and (C) Plt compared with ruxolitinib alone (P < .0001, n = 10). (D) Spleen size in the combination arm was also lower. (E) Improved histopathology and reduction in BM fibrosis in mice treated with combination of JAK and HSP90 inhibitors. (F) After 4 weeks of drug treatment, the combination group that had ruxolitinib dose increased to 90 mg/kg for 2 weeks (Rux90*+PU-H71) had lowest blood counts and spleen sizes compared with ruxolitinib alone. (Data are represented at mean ± SEM; n = 4).

Figure 5

Genetic loss or pharmacologic degradation of JAK2 subsequent to JAK inhibitor monotherapy can overcome persistence. Deletion of Jak2 following 2 weeks of ruxolitinib treatment resulted in reduction of (A) mutant allele burden (P < .005), (B) bloods counts (P < .001), and (C) spleen weights (P < .005) compared with mice that continued to receive ruxolitinib. (Rux, n = 9; Jak2 deleted, n = 6). Deletion of JAK2 after long-term (5 weeks) of ruxolitinib treatment prevents disease relapse following cessation of treatment by reducing (D) WBC and spleen weights as well as (E) mutant allele burden in Jak2-deleted mice (n = 3). (F) Addition of PU-H71 to ruxolitinib monotherapy resulted in significant reduction in blood counts. (G) Spleen sizes were further reduced by PU-H71 therapy (P < .05, n = 4). (Data are represented at mean ± SEM). (H) Mononuclear cells from ruxolitinib-treated MPN patients were isolated and treated with 0.5 μM PU-H71 for 16 hours, which led to inhibition of JAK2 and downstream STAT-MAPK signaling.