Single agent BMS-911543 Jak2 inhibitor has distinct inhibitory effects on STAT5 signaling in genetically engineered mice with pancreatic cancer

Abstract

The Jak/STAT pathway is activated in human pancreatic ductal adenocarcinoma (PDAC) and cooperates with mutant Kras to drive initiation and progression of PDAC in murine models. We hypothesized that the small-molecule Jak2 inhibitor (BMS-911543) would elicit anti-tumor activity against PDAC and decrease immune suppressive features of the disease. We used an aggressive genetically engineered PDAC model with mutant KrasG12D, tp53R270H, and Brca1 alleles (KPC-Brca1 mice). Mice with confirmed tumor burden were treated orally with vehicle or 30 mg/kg BMS-911543 daily for 14 days. Histologic analysis of pancreata from treated mice revealed fewer foci of adenocarcinoma and significantly decreased Ki67+ cells versus controls. In vivo administration of BMS-911543 significantly reduced pSTAT5 and FoxP3 positive cells within the pancreas, but did not alter STAT3 phosphorylation. Continuous dosing of KPC-Brca1 mice with BMS-911543 resulted in a median survival of 108 days, as compared to a median survival of 87 days in vehicle treated animals, a 23% increase (p = 0.055). In vitro experiments demonstrated that PDAC cell lines were poorly sensitive to BMS-911543, requiring high micromolar concentrations to achieve targeted inhibition of Jak/STAT signaling. Similarly, BMS-911543 had little in vitro effect on the viability of both murine and human PDAC-derived stellate cell lines. However, BMS-911543 potently inhibited phosphorylation of pSTAT3 and pSTAT5 at low micromolar doses in human PBMC and reduced in vitro differentiation of Foxp3+ T regulatory cells. These results indicate that single agent Jak2i deserves further study in preclinical models of PDAC and has distinct inhibitory effects on STAT5 mediated signaling.

Keywords: Jak2; STAT3; STAT5; pancreatic cancer.

Figure 1. The Jak/STAT signaling pathway is active in genetically engineered mouse models (GEMM) of PDAC

Representative 20x H&E, pSTAT3 (Tyr705), and STAT5 (Tyr694) immunohistochemistry images from pancreatic tissue from human PDAC, wild-type murine pancreas, KPC, and KPC-Brca1 mice.

Figure 2. Effect of in vivo administration of BMS-911543 on biomarkers of tumor progression in the KPC-Brca1 GEMM

Tumor burden of KPC-Brca1 mice was confirmed by A. bioluminescent (BLI) imaging at 5–6 weeks of age, at which point mice were treated with 30 mg/kg of BMS-911543 (n = 5) or vehicle (n = 5) by daily gavage for 14 days. B. BLI data were quantified in animals at this time point. C. H&E from pancreatic tissue obtained from two representative animals following euthanasia at day 14 post-treatment were analyzed and D. differentiation status was assessed in all specimens (n = 5). Arrows represent tumor burden assessed by histology. E. Representative IHC analysis of Ki67⁺ cells from day 14 tumors. F. Ki67⁺ cells were quantified in tumors from day 14 mice. G. At 8 weeks of age, KPC-Brca1 mice were treated with 30 mg/kg of BMS-911543 (n = 6) or vehicle (n = 6) by daily gavage until mice met pre-specified IACUC-approved early removal criteria.

Figure 3. Administration of BMS-911543 decreases pSTAT5 expression PDAC tumors

KPC-Brca1 mice (5–6 weeks of age confirmed by BLI were treated for 2 weeks with daily oral gavage of BMS-911543 at 30 mg/kg. A and B. pSTAT3 and C and D. pSTAT5 was assessed in the tissue by IHC (brown staining) at 20x magnification. (B) pSTAT3 and (D) pSTAT5 staining was quantified in the nucleus by spectral measuring of DAB intensity, which is represented as percent positive cells and H-score (staining intensity) from 5 random 20x fields per tissue/mouse (n = 5 per group).

Figure 4. Reduced percentage of intratumoral FoxP3+ T regulatory cells in mice administered BMS-911543

KPC-Brca1 mice were treated with 30 mg/kg BMS-911543 for 2 weeks and pancreatic tissue was A. stained by IHC for CD3⁺ T cells (brown) in well-differentiated and poorly differentiated tumor areas and B. quantified by the average number of positive cells per field (40x magnification; n = 5 per group). C. Representative staining for FoxP3⁺ cells within the tumor of representative mice treated with either vehicle or BMS-911543. D. Quantification of FoxP3⁺ cells per field (40x magnification; n = 5 per group).

Figure 5. In vitro effects of BMS-911543 on murine and human PDAC and stellate cell lines

A and B. Murine KPC-Brca1 PDAC, C. stellate cell lines, and human PDAC D. BxPC3, E. Panc-1, and F. MiaPaca-2 cell lines were treated in vitro for 48 hours with BMS-911543. Cell viability was measured by MTT assay. Error bars represent the standard deviation from 3 biological replicate experiments.

Figure 6. In vitro effects of BMS-911543 on cytokine-stimulated cells

Cells were treated with BMS-911543 for 24 hours and lysates were made for western blotting. A. Murine KPC-Brca1 (CL10–233 and CL-243), B-PSC (KPC-Brca1 derived PSC line), and human PDAC (BxPC3, Panc-1, and MiaPaca-2) cell lines were analyzed for pSTAT3 and pSTAT5 by immunoblot. Healthy human donor PBMC were cultured for 2 hours with BMS-911543 and then stimulated for 20 minutes with B. recombinant IL-6 (10 ng/ml) to test for pSTAT3 or C. IL-2 (4 nM) to test for pSTAT5 expression. For all immunoblots β-actin was used as a loading control.

Figure 7. BMS-911543 inhibits T regulatory cell differentiation in vitro

CD4⁺ T cells were negatively selected from healthy donor PBMC, treated with vehicle control or BMS-911543 (5 and 10 μM) and stimulated with 1 μg/ml CD3, 5 ng/ml IL-2, and 2 ng/ml TGF-β1 for 6 days. A. Cells were harvested and stained by flow cytometry by gating on CD4⁺CD25^hiFoxp3⁺ cells. B. Data shown represent the mean percentage of CD25^hiFoxp3⁺ cells as the percentage of CD4⁺ cells. Error bars represent the standard deviation of independent experiments using cells from n = 6 healthy donors.