Intrinsic and Extrinsic Survival Signals Converge on STAT3 As a Critical Mediator of BCR-ABL-Independent Tyrosine Kinase Inhibitor Resistance,

Abstract 3742

In Philadelphia chromosome positive leukemia, mutations in the BCR-ABL kinase domain are a well established mechanism of resistance to tyrosine kinase inhibitors (TKIs), but fail to explain many cases of primary and acquired resistance. Extrinsic survival signals from the bone marrow microenvironment can protect CML cells from the effects of TKIs in a STAT3-dependent manner. To define the role of pSTAT3 in extrinsic versus intrinsic TKI resistance, we used BCR-ABL⁺ cell lines exhibiting ‘extrinsic resistance' through culture in conditioned media (CM) from human HS-5 bone marrow stromal cells (R^e), or derivative lines expressing native BCR-ABL with ‘intrinsic resistance' adapted for growth in the presence of 1.0–2.5 μM imatinib (Rⁱ).

Immunoblot analysis showed near-complete suppression of BCR-ABL tyrosine kinase activity upon exposure to imatinib in all R^e and Rⁱ cells tested, suggesting BCR-ABL-independent resistance in both cases. K562-Rⁱ and AR230-Rⁱ cells proliferating in 1.0 μM imatinib showed upregulation of pAKT^S473, pJAK2^Y1007/1008, pSTAT3, and pSTAT5 compared to sensitive cells under the same conditions. When imatinib was increased (2.5 μM), pJAK2^Y1007/1008 and pSTAT5 were markedly reduced, suggesting regulation by BCR-ABL kinase activity. In contrast, levels of pSTAT3 and pAKT^S473 were further increased. Similar to Rⁱ cells, extrinsically resistant cells (K562-R^e, LAMA84-R^e, KBM5-R^e) showed increased pSTAT3 when cultured in HS-5 CM; however, upregulation was not observed for pAKT^S473. Of key importance, pSTAT3 levels were also increased by HS-5 CM in CD34⁺ cells from newly diagnosed CML patients.

To further investigate the role of pSTAT3 in TKI resistance, we used lentiviral shRNA to knockdown STAT3 (shSTAT3) in sensitive and resistant BCR-ABL⁺ cells. Cells were kept in culture for 24 hours with and without 1.0 μM imatinib in regular medium or CM, followed by trypan blue exclusion and/or plating in semisolid medium. Compared to cells expressing a scrambled control (shSCR), shSTAT3 reduced the in vitro growth of intrinsically resistant K562-Rⁱ and AR230-Rⁱ cells by 54.6% and 33.3% in the presence of imatinib (1.0 μM), respectively. Consistent with these observations, shSTAT3 impaired the clonogenic potential of K562-Rⁱ cells by 65.0% following culture in 1.0 μM imatinib, and similar results were obtained for AR230-Rⁱ cells, indicating that STAT3 plays a functional role in mediating intrinsic TKI resistance in multiple BCR-ABL-expressing cell lines. In extrinsic resistance, 3–12 hours of culture with CM from HS-5 stromal cells enhanced the clonogenic potential of K562-R^e and LAMA84-R^e cells by 25–40% in the presence of imatinib compared to controls cultured in regular medium. However, the protective effects of CM were abrogated by introduction of shSTAT3. In contrast, shSTAT3 had no effect on colony formation following growth in regular medium, consistent with the low level of pSTAT3 in parental K562 and LAMA-84 cells. These data suggest that STAT3 activation is a prominent feature of both extrinsic and intrinsic BCR-ABL-independent imatinib resistance. Additionally, preliminary experiments implicate pSTAT3 in intrinsic BCR-ABL-independent resistance to the third-generation TKI, ponatinib (AP24534), suggesting that activated STAT3 may have a role in resistance to other TKIs.

Altogether, our data suggest that intrinsic and extrinsic pathways converge on STAT3 as a critical mediator of BCR-ABL-independent TKI resistance, and implicate this pathway as a potential therapeutic target for the treatment of patients with TKI resistance despite BCR-ABL inactivation. We posit that microenvironmental cues activate pathways that initially support survival of leukemia cells despite BCR-ABL inhibition, and that overt resistance develops when these cells establish an intrinsic mechanism to maintain activation of the same pathways.