Combination of Ponatinib with Hedgehog Antagonist Vismodegib for Therapy-Resistant BCR-ABL1–Positive Leukemia

Published OnlineFirst January 14, 2013; DOI: 10.1158/1078-0432.CCR-12-1777

Clinical Cancer Cancer Therapy: Preclinical Research See related commentary by Dao and Tyner, p. 1309

Seiichiro Katagiri1, Tetsuzo Tauchi1, Seiichi Okabe1, Yosuke Minami2, Shinya Kimura3, Taira Maekawa4, Tomoki Naoe2, and Kazuma Ohyashiki1

Abstract Purpose: The Hedgehog signaling pathway is a key regulator of cell growth and differentiation during development. Whereas the Hedgehog pathway is inactive in most normal adult tissues, Hedgehog pathway reactivation has been implicated in the pathogenesis of several neoplasms including BCR-ABL1–positive leukemia. The clear link between the Hedgehog pathway and BCR-ABL1–positive leukemia led to an effort to identify small molecules to block the pathway. Experimental Design: We investigated the combined effects of vismodegib and ponatinib, a pan-ABL1 kinase inhibitor, in nonobese diabetic/severe-combined immunodeficiency (NOD/SCID) repopulating T315I BCR-ABL1–positive cells in vitro and in vivo. Results: We observed that combination with vismodegib and ponatinib helps to eliminate therapy- resistant NOD/SCID repopulating T315I BCR-ABL1–positive cells. The percentage of CD19-positive leukemia cells in peripheral blood was significantly lower in vismodegib þ ponatinib–treated mice than that of the vehicle or ponatinib alone (P < 0.001). Spleen weights were also lower in vismodegib þ ponatinib–treated mice than in ponatinib alone (P < 0.05). Overall tumor burden, as assessed by BCR-ABL mRNA from bone marrow cells, was significantly lower in vismodegib þ ponatinib–treated mice than in ponatinib alone (P < 0.005). We also found that vismodegib significantly reduced BCR-ABL1–positive leukemia cell self-renewal in vitro as well as during serial transplantation in vivo. Conclusions: The combination with a Smo inhibitor and ABL1 tyrosine kinase inhibitors may help eliminate therapy-resistant T315I BCR-ABL1–positive leukemia cells. Our preclinical results indicate that vismodegib has potential as an important option for controlling minimal residual cells in BCR-ABL1– positive leukemia. Clin Cancer Res; 19(6); 1422–32. 2012 AACR.

Introduction common set of critical stemness genes that regulate self- An emerging concept in cancer biology is that a rare renewal (3). Hematopoietic stem cells and leukemic stem population of cancer stem cells exists in the heterogeneous cells share common features, including self-renewal, the cell mass that constitutes a tumor (1). This concept also capacity to differentiate, resistance to apoptosis, and limit- applies to BCR-ABL1–positive leukemia (2). Normal and less proliferative potential (3). Despite these similarities, leukemic hematopoietic stem cell functions are deﬁned by a several stemness factors, such as Hedgehog, Wnt, Notch, and BMI-1, show differential activation in normal versus leukemia stem cells (4). Authors' Afﬁliations: 1First Department of Internal Medicine, Tokyo Hedgehog signaling is increased in BCR-ABL1–positive 2 Medical University, Shinjuku-ku, Tokyo; Department of Hematology and stem and progenitor cells becoming more active with Oncology, Nagoya University Graduate School of Medicine, Nagoya; 3Division of Hematology, Respiratory Medicine and Oncology, Department disease progression (5–7). The Hedgehog signaling path- of Internal Medicine, Faculty of Medicine, Saga University, Saga; and way consists of 3 closely related ligands, Sonic Hedgehog 4Department of Transfusion Medicine and Cell Therapy, Kyoto University Hospital, Faculty of Medicine, Kyoto University, Kyoto, Japan (Shh), Indian Hedgehog (Ihh), and Desert Hedgehog (Dhh), that can each bind to the transmembrane protein Note: Supplementary data for this article are available at Clinical Cancer Research Online (http://clincancerres.aacrjournals.org/). Patched (PTCH; refs. 8–10). Upon ligand binding, PTCH inhibition of the positive effector Smoothened (Smo) is Corresponding Author: Tetsuzo Tauchi, Division of Hematology, Depart- ment of Medicine, Tokyo Medical University, 6-7-1 Nishishinjuku, Shinju- released and leads to production of activating forms of kuku, Tokyo 160-0023, Japan. Phone: 81-3-3342-6111; Fax: 81-3-5381- glioma-associated oncoproteins 1–3 (Gli1–3; refs. 9–12). 6651; E-mail: [email protected] Gli1 is a positive effector of signaling, Gli3 is predomi- doi: 10.1158/1078-0432.CCR-12-1777 nantly a transcriptional inhibitor, and Gli2 can function 2012 American Association for Cancer Research. in both roles (9–12). Smo signals to a cytoplasmic

1422 Clin Cancer Res; 19(6) March 15, 2013

Targeting the Hedgehog Pathway in BCR-ABL1–Positive Leukemia

Materials and Methods Translational Relevance Antibodies and reagents Despite the great success with combination of high- Anti-Gli1 Ab, anti-ABL Ab (11-23), anti-CD19 Ab, anti- dose ABL tyrosine kinases and intensive chemotherapy Bcl2 Ab, anti-Cyclin D2 Ab, and anti-c-Myc Ab were pur- in BCR-ABL1–positive acute lymphoblastic leukemia, chased from Santa Cruz Biotechnology, Inc. Human Gli1 there are still drawbacks that need to be addressed. siRNA and Smo siRNA were also purchased from Santa Cruz Above all, 40% of patients, even with hematopoietic Biotechnology, Inc. Vismodegib, ponatinib, sorafenib, and stem cell transplantation, have relapse of the disease. BEZ235 were obtained from Selleck. Therefore, it is necessary to define targets in BCR- ABL1–positive leukemia stem cells that may be candi- Cells and cell culture dates for new treatment options. In the present study, BaF3 cells expressing wild-type BCR-ABL and mutant we investigated the combined effects of vismodegib forms of BCR-ABL (M244V, G250E, Q252H, Y253F, and ponatinib in mutant forms of BCR-ABL1–expres- T315A, T315I, F317L, F317V, M351T, H396P) were sing leukemia cells. We observed that combination described previously (25, 26). SK-9 was described previ- with a Smo inhibitor and pan-ABL1 tyrosine kinase ously (27). Briefly, SK-9 cell line was established from a inhibitors (TKI) helps to eliminate therapy-resistant BCR-ABL1–positive acute lymphoblastic leukemia (ALL) T315I BCR-ABL1–positive leukemia cells. Thus, patient with T315I mutation (27). K562 cells were obtained exploring a vast array of possible therapeutic combi- from the American Type Culture Collection. OM;922 cells nations will be useful to simultaneously target these were described previously (28). TF-1 BCCR-ABL was pathways. Our preclinical results indicate that vismo- described previously (29) degib has potential as an important option for controlling resistance in BCR-ABL1–positive leukemia. NOD/SCID repopulating BCR-ABL1–positive cells The combined results of cell-based and in vivo studies We serially transplanted human leukemia cells from suggest that vismodegib exhibits sufficient activity patients with chronic myeloid leukemia blast crisis against mutant forms of BCR-ABL1 to warrant consid- (T315I BCR-ABL1: UPN1) or Ph-positive ALL (WT- eration for combined use with ABL TKIs. BCR-ABL1: UPN5) into NOD/SCID/IL-2gc / mice. The þ þ þ cell fractions with CD34 CD38 CD19 and CD34 þ þ CD38 CD19 could self-renew and transfer the leukemia in nonobese diabetic/severe-combined immunodeficien- complex that releases Gli1, which translocates to the cy (NOD/SCID) mice (30). UPN1 cells and UPN5 cells nucleus where it activates the Hedgehog target genes were used as the NOD/SCID repopulating cells. (9, 10, 13). Leukemia cells are believed to rely on autocrine and paracrine Hedgehog signaling. The clear link Apoptosis assay between the Hedgehog pathway and BCR-ABL1–positive The incidence of apoptosis was determined by flow leukemias led to an effort to identify small molecules to cytometric analysis with the fluorescein isothiocyanate block the pathway. (FITC)-conjugated APO2.7 monoclonal antibody (clone Vismodegib is a selective hedgehog pathway inhibitor 2.7), which was raised against the 38-kDa mitochondrial with greater potency and more favorable pharmaceutical membrane protein (7A6 antigen) and is expressed by cells properties than cyclopamine (14–19). Vismodegib has undergoing apoptosis (28). antitumor activity in a mouse model of medulloblastoma and in xenograft models of primary human tumor cells, siRNA experiments including colorectal cancer and pancreatic carcinoma, in siRNA experiments were carried out as described previ- which its effects correlate with blockade of the Hedgehog ously (26). pathway (20–23). In the present study, we investigated the combined effects Transfection of vismodegib and ponatinib, a pan-ABL1 kinase inhibitor Activated Akt1 cDNA in pUSEamp was purchased from (24), in mutant forms of BCR-ABL1–expressing BaF3 cells Millipore. Transfection experiments were carried out as and the T315I-expressing human leukemia cells. We described previously (31). observed that combination with a Smo inhibitor and ABL1 tyrosine kinase inhibitors (TKI) helps to eliminate therapy- Immunoblotting resistant T315I BCR-ABL1–positive leukemia cells. Our Immunoblotting was conducted as described previously preclinical results indicate that vismodegib has potential (31). Nuclear extracts were prepared as described previously as an important option for controlling minimal residual (32) cells in BCR-ABL1–positive leukemia. The combined results of cell-based and in vivo studies suggest that vismodegib Colony-forming assay exhibits sufficient activity against mutant forms of BCR- T315I BCR-ABL BaF3, SK-9, UPN1, and UPN5 cells were ABL1 to warrant consideration for combined use with pan- treated with 1 or 10 mmol/L of vismodegib for 72 hours, ABL1 TKIs. washing free of drug and seeded in triplicate in condition

www.aacrjournals.org Clin Cancer Res; 19(6) March 15, 2013 1423

Katagiri et al.

medium MethoCult GF H4434 (Stem Cell Technologies). of Gli1 in the cytoplasmic extracts and the nuclear extracts At 14 days, the leukemic colonies (>50 cells) were counted (Fig. 1A). Next, we treated SK-9 cells with several kinds of as initial plating. The representative plate was then washed inhibitors, including ponatinib, vismodegib, sorafenib and cells were resuspended and replated. After an additional (Raf-1 kinase inhibitor), and BEZ225 (mTOR/PI3K inhib- 14 days, colonies were counted as secondary replating. itor). Activated Akt1-transfected SK9 cells were also trea- Clonogenic recovery of untreated cells was normalized to ted with ponatinib (Fig. 1B). Treatments with ponatinib, 100%, and plating results from all treatment groups were vismodegib, and BEZ235 reduced the expression of Gli1 expressed as percentage of control. in the nuclear extracts in SK-9 cells; however, Gli1 expression was partially reduced in the activated Akt1-trans- Nude mice xenografts model fected SK-9 cells treated with ponatinib and SK-9 cells Twelve-week-old nude mice were injected with 5 105 treated with sorafenib (Fig. 1B). These results suggest that cells of mixture of BaF3 cells expressing wild-type BCR-ABL Gli1 is regulated by BCR-ABL1 signaling especially, in and mutant forms of BCR-ABL (M244V, G250E, Q252H, part, mTOR/PI3K signaling pathways. To investigate the Y253F, T315A, T315I, F317L, F317V, M351T, H396P). At entire Hedgehog signaling pathway, cell lysates from SK-9 24-hour injection of the leukemia cells, these mice were cells treated with ponatinib, vismodegib, and ponatinib treated with either vehicle or vismodegib (20 mg/kg; every þ vismodegib were immunoblotted with anti-Bcl2 Abs, day) or ponatinib (30 mg/kg; every day), or vismodegib (20 anti-cyclin D2 antibodies, anti-c-Myc antibodies, or anti- mg/kg; every day) þ ponatinib (30 mg/kg; every day). Mice ABL antibody. Treatments with ponatinib and vismode- were observed daily, and body weight as well as signs of gib reduced the expression of Bcl2, cyclin D2, and c-Myc stress (for example, lethargy, ruffled coat, or ataxia) was (Fig. 1C). Next, WT-p210 BCR-ABL BaF3 cells and T315I used to detect possible toxicities. BCR-ABL BaF3 cells were treated with indicated concentrations of vismodegib for 48 hours, and the cytoplasmic NOD/SCID mouse models extracts and the nuclear extracts were immunoblotted On day 1, 5-week-old female NOD/SCID mice were with anti-Gli1 antibody (Fig. 1D). The nuclear fraction injected intravenously with 1 106 cells of BCR-ABL1– of Gli1 was suppressed by 10 mmol/L of vismodegib (Fig. positive leukemia cell line, SK-9 with T315I mutation, or 1D). SK-9 cells were also incubated with nilotinib or NOD/SCID re-populating T315I BCR-ABL1–positive dasatinib or ponatinib for 72 hours, the cytoplasmic cells(UPN1).Onday2,thesemiceweretreatedwith extracts and the nuclear extracts were immunoblotted vehicle (n ¼ 6) or vismodegib (20 mg/kg per os; every day; with anti-Gli1 antibody (Supplementary Fig). Ponatinib n ¼ 6) or ponatinib (30 mg/kg; every day; n ¼ 6) or reducedtheexpressionofGli1inbothcytoplasmic vismodegib (20 mg/kg per os; every day) þ ponatinib (30 extracts and nuclear extracts (Supplementary Fig). mg/kg; every day; n ¼ 6). On day 28, mice were sacrificed for evaluation. Inhibition of the Hedgehog pathway linked the induction of apoptosis and reduced proliferation in Second transplantation BCR-ABL1–positive leukemia NOD/SCID mice were injected with UPN1 cells then T315I BCR-ABL1–expressing SK-9 cells were cultured treated with vismodegib for 28 days. All mice showed with ponatinib (1 nmol/L) or ponatinib (1 nmol/L) and engraftment of leukemia by flow cytometry. We isolated sonic hedgehog (Shh; 30 ng/mL) for 72 hours. The inci- þ human CD45 cells from the spleen of mice from each dence of apoptosis was determined by APO2.7 monoclonal treatment group and injected equivalent numbers of leu- antibody (Fig. 2A). Treatment with ponatinib and Shh kemia cells into secondary recipients. On 28 days, all mice prevented the induction of apoptosis in SK-9 cells (Fig. were sacrificed for evaluated. 2A). However, co-treatment with 10 mmol/L of vismodegib overcame Shh-mediated anti-apoptotic effects (Fig. 2A). Results These results indicate that vismodegib actually inhibits Interactions between Hedgehog activation and BCR- Hedgehog anti-apoptosis pathways in SK-9 cells. To assess ABL1 signaling pathways the functional importance of Smo and Gli1, RNA inter- Major advances have been made in understanding the ference was used to determine whether reductions in Smo interactions between Hedgehog signaling and other path- and Gli1 affect proliferation after ponatinib treatment ways during carcinogenesis (9). Several pathways are (Fig. 2B–D). K562 cells were transfected with control implicated in expression of Gli1 by Ras, TGFb,JUN, siRNA, Smo siRNA, and Gli1 siRNA (Fig. 2B). At 48 hours SCL/TAL1, and EWS-FLI1 oncoproteins (9). Therefore, after transfection, K562 cells were treated with indicated we examined the expression of Gli1 and BCR-ABL1 sig- concentrations of ponatinib for 48 hours. The number of naling pathways (Fig. 1A and B). K562, T315I BCR-ABL, cells in each well was counted by flow cytometry, and cell and TF-1 BCR-ABL cells were treated with ponatinib or of numbers were normalized by dividing the number of cells vismodegib or ponatinib þ vismodegib for 72 hours and (Fig. 2C). In the presence of Smo siRNA or Gli1 siRNA, then cytoplasmic cell lysates and nuclear extracts were K562 cells increased antiproliferative activity with pona- immunoblotted with anti-Gli1 antibodies. Treatments tinib (P < 0.01; Fig. 2C). Co-treatment with Smo siRNA or with ponatinib and vismodegib reduced the expression Gli1 siRNA and ponatinib also enhanced the induction of

1424 Clin Cancer Res; 19(6) March 15, 2013 Clinical Cancer Research

Targeting the Hedgehog Pathway in BCR-ABL1–Positive Leukemia

A B SK-9 K562 T315I BCR-ABL BaF3 TF-1 BCR-ABL Sorafenib Ponatinib Activated Akt1 + Ponatinib Control Ponatinib + Vismodegib BEZ235 Vismodegib Vismodegib Control Ponatinib + Vismodegib Vismodegib Ponatinib Control Ponatinib + Vismodegib Vismodegib Ponatinib Ponatinib + Vismodegib Ponatinib Control

100 8 20 8 105 10 33 (%) α Gli1 α α Gli1 Gli1 BLOT: α Gli1 extracts extracts

α α Cytoplasmic Cytoplasmic α ABL ABL ABL 100 76 112 78 114 98 105 (%) BLOT: α ABL

α Gli1 α Gli1 α Gli1 100 0.7 0.8 0.7 94 0.5 24 (%) Nuclear extracts BLOT: α Gli1

α ABL α ABL α ABL Image quantification Nuclear extracts 100 94 118 93 120 104 121 (%) BLOT: α ABL C D WT BCR-ABL BaF3 T315I BCR-ABL BaF3 Vismodegib: Vismodegib: 0 0.5 1.0 5.0 10.0 μmol/L 0 0.5 1.0 5.0 10.0 μmol/L Control Ponatinib Ponatinib + Vismodegib Vismodegib

100 100 100 82 24 (%) 100 100 94 91 28 (%) α α 100 8 7 5 (%) BLOT: Gli1 Gli1 α Bcl2

100 100 106 101 100 (%) 100 88 82 91 92 (%) 100 12 6 6 (%) BLOT: α ABL α ABL α Cyclin D1 Cytoplasmic extracts

100 100 100 80 2 (%) 100 15 18 7 (%) 100 98 62 58 0.1 (%) BLOT: α Gli1 α

Image quantification: Gli1 α c-myc Image quantification:

100 100 98 96 98 (%) 100 102 78 104 (%) 100 79 82 82 78 (%) Nuclear extracts BLOT: α ABL α ABL α ABL

Figure 1. Interactions between Hedgehog activation and BCR-ABL signaling pathways. A, K562 cells, T315I BCR-ABL BaF3 cells, and TF-1 BCR-ABL cells were treated with 40 nmol/L of ponatinib or 10 mmol/L of vismodegib or 40 nmol/L of ponatinib þ 10 mmol/L of vismodegib for 72 hours and then cytoplasmic cell lysates and nuclear extracts were immunoblotted with anti-Gli1 Abs or anti-ABL Ab. B, SK-9 cells were treated with several kinds of inhibitors, including 40 nmol/L of ponatinib, 10 mmol/L of vismodegib, 40 nmol/L of ponatinib þ 10 mmol/L of vismodegib, 2 mmol/L of sorafenib, and 500 nmol/L of BEZ225 for 48 hours. Activated Akt1-transfected SK9 cells were also treated with 40 nmol/L of ponatinib for 48 hours. The cytoplasmic cell lysates and nuclear extracts were immunoblotted with anti-Gli1 Abs or anti-ABL Ab. C, SK-9 cells were treated with 40 nmol/L of ponatinib, 10 mmol/L of vismodegib, 40 nmol/L of ponatinib þ 10 mmol/L of vismodegib for 48 hours. The cell lysates were immunoblotted with anti-Gli1 Abs, anti-Bcl2 Abs, ant-cyclin D2 Abs, anti-c-Myc Abs, or anti-ABL Ab. D, WT-p210 BCR-ABL BaF3 cells and T315I BCR-ABL BaF3 cells were treated with indicated concentrations of vismodegib for 48 hours. The cytoplasmic cell lysates and nuclear extracts were immunoblotted with anti-Gli1 Abs or anti-ABL Ab. apoptosis in K562 cells (Fig. 2D). We also carried out the week-old nude mice were injected with 5 105 cells of a above experiments using OM9;22 cells and TF1 BCR-ABL mixture of BaF3 cells expressing wild-type BCR-ABL1 and cells (Table 1). Similar results were obtained in these mutant forms of BCR-ABL1 (M244V, G250E, Q252H, cell lines (Table 1). These results showed that inhibition Y253F, T315A, T315I, F317L, F317V, M351T, H396P). At of Smo and Gli1 can play an important role in the 24 hours after injection of leukemia cells, these mice were induction of apoptosis and the antiproliferative effects treated with either vehicle, vismodegib (20 mg/kg; every with ponatinib. day), ponatinib (30 mg/kg; every day), vismodegib (20 mg/ kg; every day) þ ponatinib (30 mg/kg; every day). Vehicle Cotreatment of vismodegib and ponatinib tends to and vismodegib-treated mice died of a condition resem- prolong survival in mouse models of BCR-ABL– bling acute leukemia by 16 days. The combination of mutant—induced leukemia vismodegib þ ponatinib–treated mice survived for more We investigate the in vivo efﬁcacy of vismodegib and than 40 days; tending to improve survival over ponatinib- ponatinib in a BCR-ABL1 BaF3 xenograft (Fig. 3). Twelve- treated mice (P ¼ 0.1005; Fig. 3).

www.aacrjournals.org Clin Cancer Res; 19(6) March 15, 2013 1425

Katagiri et al.

6.7% 25.9% 15.2% Control Gli1 siRNA Smo siRNA

Gli1 Image quantification: 100 16 78 BLOT : α Gli1 Control Ponatinib 1 nmol/L Ponatinib 1 nmol/L + Shh

7.1% 31.8% 30.4% 100 95 8 BLOT : α Smo

Vismodegib 10 μmol/L Ponatinib 1 nmol/L Ponatinib 1 nmol/L + Shh + Vismodegib 10 μmol/L + Vismodegib 10 μmol/L 100 98 92 BLOT : α-Actin

C D Control 8.4% 13.3% 25.3% (%) 100 Control Smo siRNA Gli1 siRNA 80 * Gli1 siRNA 12.2% 22.2% 32.7% 60 * * 40 * 29.2% 29.7% Smo siRNA 14.0%

Ponatinib (−) 0.5 nmol/L 5.0 nmol/L 0 0 2.0 5.0 20.0 Ponatinib concentration (nmol/L)

Figure 2. Inhibition of the Hedgehog pathway enhanced the induction of apoptosis with ABL TKI and reduced proliferation in BCR-ABL1–positive leukemia. A, SK-9 cells were cultured with ponatinib (1 nmol/L) or ponatinib (1 nmol/L) and sonic hedgehog (Shh; 30 ng/mL) for 72 hours. The incidence of apoptosis was determined by APO2.7 monoclonal antibody. B, K562 cells were transfected with control siRNA, Smo siRNA, and Gli1 siRNA. Cell lysates were immunoblotted with anti-Gli1 Ab, anti-Smo Ab, and anti-actin Ab. C, at 48 hours after transfection, K562 cells were treated with indicated concentration of ponatinib for 48 hours. The number of cells in each well was counted by ﬂow cytometry, and cell numbers were normalized by dividing the number of cells. , P < 0.01 compared with control. D, the incidence of apoptosis was determined by APO2.7 monoclonal antibody. Similar results were obtained in 2 independent experiments.

Combination with a Smo inhibitor and ABL tyrosine in peripheral blood was significantly lower in vismodegib þ kinase inhibitors eliminates minimal residual BCR- ponatinib–treated mice than that of the vehicle or ponati- ABL1–positive leukemia cells nib alone (P < 0.001; Fig. 4A). Spleen weights were also We examined the NOD/SCID mouse to elucidate in vivo lower in vismodegib þ ponatinib–treated mice than in efficacy (Fig. 4). On day 1, 5-week-old female NOD/SCID ponatinib alone (P < 0.05; Fig. 4B). Overall tumor burden, mice were injected intravenously with 1 106 cells of the as assessed by BCR-ABL mRNA from bone marrow cells, was BCR-ABL1–positive leukemia cell line, SK-9 with a T315I significantly lower in vismodegib þ ponatinib–treated mice mutation (Fig. 4). We also used the NOD/SCID repopulat- than in ponatinib alone (P < 0.005; Fig. 4C). Histopatho- ing T315I BCR-ABL1–positive leukemia UPN1 cells for this logic analysis of vehicle-treated mice revealed infiltration of study (Fig. 4) The next day, these mice were treated with bone marrow with leukemia cells; however, cotreatment either vehicle (n ¼ 6), vismodegib (20 mg/kg per os; every with ponatinib and vismodegib showed normal hemato- day; n ¼ 6), ponatinib (30 mg/kg; every day; n ¼ 6), poiesis in the bone marrow cavity (Fig. 4D and E). These vismodegib (20 mg/kg per os; every day) þ ponatinib (30 results suggest that combination with a Smo inhibitor and mg/kg; every day; n ¼ 6). On day 28, mice were sacrificed for ABL tyrosine kinase inhibitors may help eliminate BCR- evaluation. The percentage of CD19-positive leukemia cells ABL1–positive leukemia cells.

1426 Clin Cancer Res; 19(6) March 15, 2013 Clinical Cancer Research

Targeting the Hedgehog Pathway in BCR-ABL1–Positive Leukemia

Table 1. Inhibition of the Hedgehog pathway enhanced the induction of apoptosis with ABL TKI and reduced proliferation in BCR-ABL1–positive leukemia

Cell lines

OM922 TF-1 BCR-ABL K562 Apoptosis,a % Control 12.3 5.2 7.5 Ponatinib (1 nmol/L) 29.2 30.2 25.6 Ponatinib (1 nmol/L) þ Shh 18.6 15.7 17.8 Vismodegib (10 mmol/L) 12.5 6.4 8.2 Ponatinib (1 nmol/L) þ vismodegib 32.5 36.6 29.6 (10 mmol/L) Ponatinib (1 nmol/) þ vismodegib 31.5 34.8 28.4 (10 mmol/L) þ Shh Growth inhibitionb,% OM922 Ponatinib 0 nmol/L 2 nmol/L 5 nmol/L 20 nmol/L Control 100 81.2 2.2 75.5 1.8 8.8 0.5 Smo siRNA 100 62.5 1.5 48.6 1.9 7.2 0.2 Glil siRNA 100 60.2 1.6 46.3 1.5 6.8 0.3 TF-1 BCR-ABL Ponatinib 0 nmol/L 2 nmol/L 5 nmol/L 20 nmol/L Control 100 95.5 4.2 83.2 2.2 12.8 0.4 Smo siRNA 100 70.1 1.8 52.4 1.6 9.5 0.2 Glil siRNA 100 72.6 1.6 54.3 1.4 8.8 0.4 SK-9 Ponatinib 0 nmol/L 2 nmol/L 5 nmol/L 20 nmol/L Control 100 70.6 2.2 54.6 1.8 5.5 0.2 Smo siRNA 100 50.6 1.2 38.6 1.3 4.4 0.2 Glil siRNA 100 48.8 1.3 34.2 1.7 3.8 0.1 Apoptosis,c % OM922 Ponatinib 0 nmol/L 0.5 nmol/L 5 nmol/L Control 12.5 16.6 24.8 Smo siRNA 18.6 23.8 35.6 Glil siRNA 17.8 24.6 38.8 TF-1 BCR-ABL Ponatinib 0 nmol/L 0.5 nmol/L 5 nmol/L Control 7.5 12.8 20.6 Smo siRNA 9.6 21.9 36.6 Glil siRNA 8.1 25.9 40.8 SK-9 Ponatinib 0 nmol/L 0.5 nmol/L 5 nmol/L Control 6.6 15.7 26.7 Smo siRNA 8.8 27.2 38.8 Glil siRNA 9.1 24.1 39.9

aThe number of cells in each well was counted by ﬂow cytometry, and cell numbers were normalized by dividing the number of cells. The each cell line was cultured with ponatinib (1 nmol/L) or ponatinib (1 nmol/L) and Shh (30 ng/mL) for 72 hours. The incidence of apoptosis was determined by APO2.7 monoclonal antibody. bEach cell line was transfected with control siRNA, Smo siRNA, or Gli1 siRNA. At 48 hours after transfection, each cell line was treated with indicated concentration of ponatinib for 48 hours. cEach cell line was transfected with control siRNA, Smo siRNA, or Gli1 siRNA. At 48 hours after transfection, each cell line was treated with indicated concentration of ponatinib for 48 hours. The incidence of apoptosis was determined by APO2.7 monoclonal antibody.

www.aacrjournals.org Clin Cancer Res; 19(6) March 15, 2013 1427

Katagiri et al.

from each treatment group and injected equivalent num- Control bers of leukemia cells into secondary recipients. Following Vismodegib 100 Ponatinib 28 days, all mice received UPN1 cells from vehicle-treated Ponatinib+Vismodegib mice engrafted with leukemia (Fig. 5D). In contrast, leuke- 80 mia engraftment was not detected in recipient mice receiving UPN1 cells from initial vismodegib-treated donors (Fig. 60 5D). These results show the persistent effects of Hedgehog inhibition on long-term self-renewing BCR-ABL1–positive 40 leukemia cells.

20 Discussion Survival rate (%) The BCR-ABL1 oncoprotein is found in a subset of 0 patients with ALL carrying the Philadelphia chromosome. 0 10 20 30 40 50 This translocation is the most common cytogenetic abnor- Days mality in adults, with ALL occurring in 25% of patients (33). BCR-ABL1 defines a high-risk group and, as such, patients receive intensive chemotherapy in combination with ABL Figure 3. Cotreatment of vismodegib and ponatinib tends to prolong TKIs and are considered for hematopoietic stem cell trans- survival in mouse models of BCR-ABL–mutant–induced leukemia. Nude mice were injected with 5 105 cells of a mixture of BaF3 cells expressing plantation (HSCT). Despite the great success with combi- the wild-type BCR-ABL and mutant forms of BCR-ABL (M244V, G250E, nation of high-dose ABL TKIs and intensive chemotherapy, Q252H, Y253F, T315A, T315I, F317L, F317V, M351T, H396P). At 24 there are still drawbacks that need to be addressed. Above hours after injection of leukemia cells, these mice were treated with all, 40% of patients, even with HSCT, have relapse of the vehicle, vismodegib (20 mg/kg; q.d.), ponatinib (30 mg/kg; q.d.), vismodegib (20 mg/kg; q.d.) þ ponatinib (30 mg/kg; q.d.). The disease. Furthermore, it is not clear whether responsive combination of vismodegib þ ponatinib–treated mice survived for more patients without HSCT cannot have relapse of the disease, than 40 days, tending to improve survival over ponatinib-treated mice but as there is evidence that BCR-ABL1–positive leukemia stem not significant (P ¼ 0.1005). cells remain present in the patient’s bone marrow even after years of therapy. Therefore, it is necessary to define targets in Hedgehog inhibition with vismodegib limits self- BCR-ABL1–positive leukemia stem cells that may be candi- renewal in vitro and in vivo dates for new treatment options. We further examined the effects of Hedgehog pathway Our study reveals that one candidate could be the Hedge- modulation on in vitro clonogenic growth (Fig. 5A). T315I hog pathway inhibitor, vismodegib. Vismodegib binds to BCR-ABL1 BaF3, SK-9, activated Akt1-transfected SK9, the and inhibits Smo, the 7-transmembrane Hedgehog path- NOD/SCID repopulating BCR-ABL1–positive leukemia way signaling protein (23). The activity of vismodegib was UPN1, and UPN5 cells were treated with 1 or 10 mmol/L first shown in vivo in preclinical models of medulloblasto- of vismodegib for 72 hours, washed free of drugs, and plated ma, colon, and pancreatic tumors (20–22). In a phase I/II in quadruplicate in methylcellulose. At 14 days, colonies study for patients with advanced malignancies, vismodegib were counted as initial plating (Fig. 5A and B). The repre- was well-tolerated, with pharmacodynamic evidence of sentative plate was then washed and cells were resuspended Hedgehog pathway inhibition and tumor regression in and replated. After an additional 14 days, colonies were patients with basal cell carcinoma and medulloblastoma counted as secondary replating (Fig. 5A and B). Upon serial (14–18). In the present study, we observed that cotreatment replating, secondary colony formations were significantly with vismodegib and Shh overcame Shh-mediated anti- inhibited by vismodegib (P < 0.001; Fig. 5A and B). The apoptotic effects (Fig. 2A and B and Table 1). Furthermore, NOD/SCID repopulating UPN1 and UPN5 cells were more the reduction of Gli1 or Smo by siRNA also enhanced the sensitive to vismodegib than T315I BCR-ABL BaF3 cells (Fig. induction of apoptosis with ponatinib (Fig. 2C and D 5A and B). Also, no difference of clonogenic recovery was and Table 1). These results indicate the specificity of this observed between SK-9 and activated Akt1-transfected SK9 compound confirmed by siRNA approaches (Fig. 2 cells (Fig. 5A). To further investigate the effects of Hedgehog and Table 1). We also observed that combination with inhibition on self-renewal and the relevance of the Hedge- vismodegib and an ABL1 TKIs helps to eliminate therapy- hog pathway as a therapeutic target in BCR-ABL1–positive resistant T315I BCR-ABL1–positive leukemia cells (Fig. 4A– leukemia, we examined the activity of vismodegib against E). Overall tumor burden, as assessed by CD19 and BCR- UNP1 cells in vivo (Fig. 5C and D). NOD/SCID mice were ABL mRNA from bone marrow cells, was significantly lower injected with UPN1 cells and then treated with vismodegib in vismodegib þ ponatinib–treated mice than in ponatinib on day 21 for 7 days (Fig. 5C). The treatments with vismo- alone (Fig. 4B and C). Also, cotreatment of vismodegib and þ degib reduced the population of CD34 CD38 cells (Fig. ponatinib seemed to prolong survival in the mice model of 5C). Next, NOD/SCID mice were injected with UPN1 cells BCR-ABL—mutant–induced leukemia (Fig. 3). In a survival then treated with vismodegib on day 2 for 28 days (Fig. 5D). mouse model using BaF3 cells expressing WT BCR-ABL1 All mice showed engraftment of leukemia by flow cytome- and mutant forms of BCR-ABL1 (M244V, G250E, Q252H, þ try. We isolated human CD45 cells from the spleen of mice Y253F, T315A, T315I, F317L, F317V, M351T, and H396P),

1428 Clin Cancer Res; 19(6) March 15, 2013 Clinical Cancer Research

Targeting the Hedgehog Pathway in BCR-ABL1–Positive Leukemia

A B 140 SK-9 UPN1 60 120 Spleen weight

50 100

40 P < 0.0001 Peripheral blood P < 0.0001

CD19-positive cells P < 0.0001 30 60 P < 0.0001 P < 0.0099 P < 0.0001 P < 0.0001 P < 0.0001 P < 0.05 20 P < 0.0001 40

10 20 SK-9 UPN1 P < 0.001 P = 0.0005 Vismodegib Vismodegib Control Control 0 Vismodegib + Ponatinib Vismodegib Control Vismodegib + Ponatinib Vismodegib Control Ponatinib Ponatinib + Ponatinib Vismodegib + Ponatinib Vismodegib Ponatinib 0 Ponatinib (%) (mg)

C D SK-9 SK-9 UPN1 Vismodegib 106 Vehicle Vismodegib Ponatinib + Ponatinib

105 BCR-ABL mRNA

104 H&E staining 103 P < 0.0001 P < 0.0001 102 P < 0.0001 P < 0.0001

101 CD19 staining P < 0.005 P < 0.005 Control Vismodegib Ponatinib + Ponatinib Vismodegib Control Vismodegib Ponatinib + Ponatinib Vismodegib (Copies) Gli1 staining

E UPN1 Vismodegib Vehicle Vismodegib Ponatinib + Ponatinib H&E staining CD19 staining Gli1 staining

Figure 4. Cotreatments with vismodegib and ponatinib eliminate minimal residual SK-9 cells. A, on day 1, 5-week-old female NOD/SCID mice were injected intravenously with 1 106 cells of the BCR-ABL1–positive SK-9 leukemia cell line or with 1 108 cells of the NOD/SCID repopulating T315I BCR-ABL1– positive leukemia UPN1 cells. On day 2, these mice were treated with either vehicle (n ¼ 6), vismodegib (20 mg/kg per os; q.d.; n ¼ 6), ponatinib (30 mg/kg; q.d.; n ¼ 6), vismodegib (20 mg/kg per os; q.d.) þ ponatinib (30 mg/kg; q.d.; n ¼ 6) for each group. On day 28, mice were sacriﬁced for evaluation. CD19-positive cells from peripheral blood from each mouse have been shown. B, spleen weights from each treated mouse. C, BCR-ABL mRNA from bone marrow cells in each treated mouse was signiﬁcantly lower in vismodegib þ ponatinib–treated mice than in ponatinib alone (P < 0.005). D, histopathologic analysis of the bone marrow cavity from each treated mouse with SK-9 transplantation. Similar results were obtained in 2 independent experiments. E, histopathologic analysis of the bone marrow cavity from each treated mouse with UPN1 transplantation. Similar results were obtainedin 2 independent experiments. H&E, hematoxylin and eosin. www.aacrjournals.org Clin Cancer Res; 19(6) March 15, 2013 1429

Katagiri et al.

Control AB% Clonogenic recovery Vismodegib 1 μmol/L Vismodegib 10 μmol/L % Clonogenic recovery Control Vismodegib 1 μmol/L 100 100 Vismodegib 10 μmol/L

80 80

60 60 * 40 * 40 * * * 20 * * 20 * * * * * 0 * * * 0 * * * * * 1st Round 2nd Round 1st Round 2nd Round 1st Round 2nd Round 1st Round 2nd Round 1st Round 2nd Round T315I BCR-ABL SK-9 SK-9 UPN1: Patient 1 UPN5: Patient 5 BaF3 Activated Akt T315I BCR-ABL WT-BCR-ABL

C CD34+ CD38– cells: 63.6% CD34+ CD38– cells: 41.2% D Secondary engraft Vismodegib UNP1: Patient 1 cells 1,000 1,000 Vismodegib Second Transplant CD34 CD34 Day 0 NOD/SCID Spleen cells CD19 staining .1 .1 .1 CD38 1,000 .1 1,000 CD38 Day 0 Dissection Vehicle Control Vismodegib

UNP1: Patient 1 cells Rates of secondary leukemia engraftment following in vivo treatment with vismodegib Vismodegib Engraft Secondary engraft Day 21 Day 28 Vismodegib 6/6 0/6 Vehicle 6/6 6/6 CD19 staining

Figure 5. Hedgehog inhibition with vismodegib limits self-renewal in vitro and in vivo. Clonogenic recovery of T315I BCR-ABL BaF3, SK-9, activated Akt1- transfected SK9 (A); the NOD/SCID repopulating T315I BCR-ABL1–positive leukemia UPN1 cells and UPN5 cells (B). T315I BCR-ABL BaF3, SK-9, activated Akt1-transfected SK9, UPN1, and UPN5 cells were treated with 1 or 10 mmol/L of vismodegib for 72 hours, washed free of drugs, and plated in quadruplicate in methylcellulose. At 14 days, colonies were counted as initial plating. The representative plate was then washed and cells were resuspended and replated. After an additional 14 days, colonies were counted as secondary replating. Clonogenic recovery of untreated cells was normalized to 100% and plating results from all treatment groups were expressed as percentage of control. , P < 0.001 compared with control. Similar results were obtained in 3 independent experiments. C, NOD/SCID mice were injected with UPN1 cells and then treated with vismodegib on day 21 þ for 7 days. The percentage of CD34 CD38 cells was analyzed by ﬂow cytometry. D, an initial cohort of mice was injected with UNP1 cells and treated with either vismodegib or vehicle. All mice were engrafted with leukemia. Following transplantation of harvested spleen cells, leukemia engraftment was not detected in recipient mice receiving UPN1 cells from initial vismodegib-treated donors. Similar results were obtained in 2 independent experiments.

cotreatment with vismodegib and ponatinib tended to serial in vivo transplantation to assess the effects of Hedge- improve survival (Fig. 3). As the impact of Smo on short- hog inhibition on long-term self-renewal (Fig. 5A–D). We term repopulating hematopoietic stem cells indicates that a found that vismodegib signiﬁcantly reduced BCR-ABL1– combination of vismodegib with generally cytotoxic and positive leukemia cell self-renewal in vitro as well as during hematotoxic chemotherapeutic reagents may induce pro- serial transplantation in vivo (Fig. 5A–D). This loss of serial longed neutropenias, anemias, and thrombocytopenias transplantation ability is most consistent with self-renewal, due to delayed bone marrow regeneration, our results from as similarly seen in serial transplantation experiments with these studies suggest that combined use of vismodegib and normal hematopoietic cells. It is notable that such long- ponatinib would be a viable strategy for preventing emer- lasting effects of Hedgehog inhibition were seen following gence of resistant clones in clinical setting. only short exposure to vismodegib either 72 hours in colony Serial transplantation is widely accepted as an assay formation assays (Fig. 5A and B) or after 28 days of treat- measuring long-term self-renewal in normal hematopoietic ment in primary recipient mice (Fig. 5D). On the basis of stem cells and neoplasms. We used a complementary in vitro our in vitro clonogenic data, we believe that this loss of serial assay with serial replating and colony formation, as well as colony formation and transplantation ability is due to the

1430 Clin Cancer Res; 19(6) March 15, 2013 Clinical Cancer Research

Targeting the Hedgehog Pathway in BCR-ABL1–Positive Leukemia

effects of Hedgehog inhibition on the self-renewal proper- The combined results of cell-based and in vivo studies ties of BCR-ABL1 leukemia-initiating cells. However, it is suggest that vismodegib exhibits sufficient activity against possible that the inhibition of engraftment during second- mutant forms of BCR-ABL1 to warrant consideration for ary transplantation is mediated by the effects of Hedgehog combined use with ABL TKIs. Although several Hedgehog inhibition on quiescence of leukemia-initiating cells, their inhibitors have now entered clinical evaluation, it is ability to interact with potential stem cell niches, and proper expected that Hedgehog inhibitors may become extreme- homing during transplantation. ly useful therapeutic interventions in a number of hema- Evidence suggests that BCR-ABL1–positive leukemia- tologic neoplasms where the persistence of cancer stem initiating cell population persists despite BCR-ABL1 inhi- cells and the protective effect of the tumor microenvi- bition, and various mechanisms have been proposed to ronment may be addressed. explain this observation. BCR-ABL1–positive leukemia- initiating cell survival is independent of BCR-ABL1 activ- Disclosure of Potential Conflicts of Interest ity and perhaps depend on other signaling pathways in No potential conflicts of interest were disclosed. vivo (34). Autocrine activation, in which the tumor cells produce and response to own Hedgehog ligands, and Authors' Contributions Conception and design: T. Tauchi paracrine Hedgehog activation have been reported in Development of methodology: S. Okabe, Y. Minami, S. Kimura, T. Mae- BCR-ABL1–positive leukemia (34). We have previously kawa, T. Naoe shown that Hedgehog ligands are produced by stroma Acquisition of data (provided animals, acquired and managed patients, provided facilities, etc.): S. Katagiri, T. Tauchi, S. Okabe, K. Ohyashiki cells in bone marrow and these allow survival and expan- Analysis and interpretation of data (e.g., statistical analysis, biosta- sion of BCR-ABL1–positive cells (35). The activation of tistics, computational analysis): T. Tauchi, Y. Minami, T. Maekawa, T. Naoe the Hedgehog pathway could be due to amplification of Writing, review, and/or revision of the manuscript: T. Tauchi, K. the Hedgehog ligands in BCR-ABL1–positive cells in Smo- Ohyashiki dependent. In the same cells, BCR-ABL1 activates AKT/ Administrative, technical, or material support (i.e., reporting or orga- nizing data, constructing databases): S. Kimura mTOR/S6K1 pathway, which phosphorylates and stabi- Study supervision: T. Tauchi lizes the Gli1 protein in Smo-independent manner. Therefore, vismodegib alone had small effects in BCR- Grant Support ABL1–positive leukemia-transplanted mice (Fig. 4A–E). This work was supported by a "High-Tech Research Center" Project for Recent report showed that a Smo-independent activation private universities: matching fund subsidy from the MEXT (Ministry of Education, Culture, Sports, Science and Technology), and by the "University- of Gli1 by the mTOR/S6K1 pathway, which cannot be Industry Joint Research Project" for private universities: matching fund inhibited by Smo inhibitors but is sensitive to inhibitors subsidy from the MEXT for K. Ohyashiki and by Grants-in-Aid from the National Institute of Biomedical Innovation and the Ministry of Education, of the mTOR pathways (36). Therefore, cotreatment with Culture, Sports, Science, and Technology of Japan for T. Naoe. pan-ABL kinase inhibitor and Smo inhibitor, panatinib The costs of publication of this article were defrayed in part by the payment of page charges. This article must therefore be hereby marked and vismodegib, indeed showed better inhibitory effects advertisement in vivo in accordance with 18 U.S.C. Section 1734 solely to indicate on T315I BCR-ABL leukemia than did single drug this fact. treatment (Fig. 4A–E). Our preclinical results indicate that vismodegib has potential as an important option for Received June 5, 2012; revised November 17, 2012; accepted December controlling resistance in BCR-ABL1–positive leukemia. 17, 2012; published OnlineFirst January 14, 2013.

References 1. Greaves M, Maley GC. Clonal evolution in cancer. Nature 2012;481: 10. Irvine DA, Copland M. Targeting hedgehog in hematologic malignancy. 306–13. Blood 2012;119:2196–204. 2. Cilloni D, Saglio G. Molecular pathways: BCR-ABL. Clin Cancer Res 11. Regl G, Neill GW, Eichberger T, Kasper M, Ikram MS, Koller J, et al. 2012;18:930–7. Human GLI2 and GLI1 are part of a positive feedback mechanism in 3. Bonnet D. Normal and leukemic stem cells. Br J Haematol 2005; basal cell carcinoma. Oncogene 2002;21:5529–39. 130:469–79. 12. Humke EW, Dorn KV, Milenkovic L, Scott MP, Rohatgi R. The output of 4. Huntly BJ, Gilliland DG. Cancer biology: summing up cancer stem Hedgehog signaling is controlled by the dynamic association between cells. Nature 2005;435:1169–70. suppressor of fused and the Gli1 protein. Genes Dev 2010;24:670–82. 5. Dierks C, Beigi R, Guo GR, Zirlik K, Stegert MR, Manley P, et al. 13. Duman-Scheel M, Weng L, Xin S, Du W. Hedgehog regulates cell Expansion of BCR-ABL-positive leukemia stem cells is dependent on growth and proliferation by inducing Cyclin D and Cyclin E. Nature Hedgehog pathway activation. Cancer Cell 2008;14:238–49. 2002;417:299–304. 6. Zhao C, Chen A, Jamieson CH, Fereshteh M, Abrahamsson A, Blum J, 14. Rudin CM, Hann CL, Laterra J, Yauch RL, Callahan CA, Fu L, et al. et al. Hedgehog signaling is essential for maintenance of cancer stem Treatment of medulloblastoma with hedgehog pathway inhibitor GDC- cells in myeloid leukemia. Nature 2009;485:776–9. 0449. N Engl J Med 2009;361:1173–8. 7. Jagani Z, Dorsch M, Warmuth M. Hedgehog pathway activation in 15. Von Hoff DD, LoRusso PM, Rudin CM, Reddy JC, Yauch RL, Tibes R, chronic myeloid leukemia. Cell Cycle 2010;9:3449–56. et al. Inhibition of hedgehog pathway in advanced basal cell carcino- 8. Ingham PW, McMahon AP. Hedgehog signaling in animal develop- ma. N Engl J Med 2009;361:1162–72. ment: paradigms and principles. Genes Dev 2001;15:3059–87. 16. Graham RA, Lum BL, Cheeti S, Jin JY, Von Hoff DD, Rudin CM, et al. 9. Yang L, Xie G, Fan Q, Xie J. Activation of the hedgehog-signaling Pharmacokinetics of hedgehog inhibitor vismodegib (GDC-0449) in pathway in human cancer and the clinical implications. Oncogene patients with locally advanced or metastatic solid tumors: the role of 2010;29:469–81. alpha-1-acid glycoprotein binding. Clin Cancer Res 2011;17:2512–20.

www.aacrjournals.org Clin Cancer Res; 19(6) March 15, 2013 1431

Katagiri et al.

17. LoRusso PM, Rudin CM, Reddy JC, Tibes R, Weiss GJ, Borad MJ, et al. 27. Okabe S, Tauchi T, Ohyashiki K. Establishment of a new Philadelphia Phase I trial of hedgehog pathway inhibitor vismodegib (GDC-0449) in chromosome-positive acute lymphoblastic leukemia cell line (SK-9) patients with refractory, locally advanced or metastatic solid tumors. with T315I mutation. Exp Hematol 2010;38:765–72. Clin Cancer Res 2011;17:2502–11. 28. Komatsu N, Watanabe T, Uchida M, Mori M, Kirito K, Kikuchi S, et al. A 18. Wong H, Alicke B, West KA, Pacheco P, La H, Januario T, et al. member of forkhead transcription factor FKHRL1 is a downstream Pharmacokinetic- pharmacodynamics analysis of vismodegib in pre- effector of STI571-induced cell cycle arrest in BCR-ABL-expressing clinical models of mutational and ligand-dependent Hedgehog path- cells. J Biol Chem 2003;278:6411–9. way activation. Clin Cancer Res 2011;17:4682–92. 29. Tauchi T, Shin-ya K, Sashida G, Sumi M, Nakajima A, Shimamoto T, 19. LoRusso PM, Jimeno A, Dy G, Adjei A, Berlin J, Leichman L, et al. et al. Activity of a novel G-quadruplex-interactive telomerase inhibitor, Pharmacokinetic dose-scheduling study of hedgehog pathway telomestatin (SOT-095), against human leukemia cells: involvement of inhibitor vismodegib (GDC-0449) in patients with locally ATM-dependent DNA damage response pathways. Oncogene advanced or metastatic solid tumors. Clin Cancer Res 2011;17: 2003;22:5338–47. 5774–82. 30. Tanizaki R, Nomura Y, Miyata Y, Minami Y, Abe A, Hanamura A, et al. 20. Yauch RL, Dijkgraaf GJ, Alicke B, Januario T, Ahn CP, Holcomb T, et al. Irrespective of CD34 expression, lineage-committed cell fraction Smoothened mutation confers resistance to a Hedgehog pathway reconstitutes and re-establishes transformed Philadelphia chromo- inhibitor in medulloblastoma. Science 2009;326:572–4. some-positive leukemia in NOD/SCID/IL-2Rgc-/- mice. Cancer Sci 21. Low JA, de Sauvage FJ. Clinical experience with Hedgehog pathway 2010;101:631–8. inhibitors. J Clin Oncol 2010;28:5321–6. 31. Tauchi T, Boswell HS, Leibowitz D, Broxmeyer HE. Coupling between 22. Ailles L, Siu LL. Targeting the Hedgehog pathway in cancer: can the p210bcr-abl and Shc and Grb2 adaptor proteins in hematopoietic cells spines be smoothened? Clin Cancer Res 2011;17:2071–3. permits growth factor receptor-independent link to ras activation 23. Dijkgraaf GJ, Alicke B, Weinmann L, Januario T, West K, Modrusan Z, pathway. J Exp Med 1994;179:167–75. et al. Small molecule inhibition of GDC-0449 refractory smoothened 32. Stwwart MJ, Litz-Jackson S, Burgess GS, Williamson EA, Leibowitz mutants and downstream mechanisms of drug resistance. Cancer Res DS, Boswell HS. Role foe E2F1 in p210 BCR-ABL downstream reg- 2011;71:435–44. ulation of c-myc transcription initiation. Studies in murine myeloid 24. O'Hare T, Shakespeare WC, Zhu X, Eide CA, Rivera VM, Wang F, et al. cells. Leukemia 1995;9:1499–507. AP24534, a pan-BCR-ABL inhibitor for chronic myeloid leukemia, 33. Tauchi T, Broxmeyer HE. BCR-ABL signal transduction. Int J Hematol potently inhibits the T315I mutant and overcomes mutation-based 1995;61:105–12. resistance. Cancer Cell 2009;16:401–12. 34. Mar BG, Amakye D, Aifantis I, Buonamici S. The controversial role of 25. Deguchi Y, Kimura S, Ashihara E, Niwa T, Hodohara K, Fujiyama Y, the Hedgehog pathway in normal and malignant hematopoiesis. Leu- et al. Comparison of imatinib, dasatinib, nilotinib and INNO-406 in kemia 2011;25:1665–73. imatinib-resistant cell lines. Leukemia Res 2008;32:980–3. 35. Okabe S, Tauchi T, Tanaka Y, Ohyashiki K. Effects of the Hedgehog 26. Tauchi T, Okabe S, Ashihara E, Kimura S, Maekawa T, Ohyashiki K. inhibitor GDC-0449, alone or in combination with dasatinib, on BCR- Combined effects of novel heat shock protein 90 inhibitor NVP- ABL-positive leukemia cells. Stem Cells Dev 2012;21:2939–48. AUY922 and nilotinib in a random mutagenesis screen. Oncogene 36. Wang Y, Ding Q, Yen CJ, Xia W, Izzo JG, Lang JY, et al. The crosstalk of 2011;30:2789–97. mTOR/S6K1 and Hedgehog pathways. Cancer Cell 2012;21:374–87.

1432 Clin Cancer Res; 19(6) March 15, 2013 Clinical Cancer Research

Combination of Ponatinib with Hedgehog Antagonist Vismodegib for Therapy-Resistant BCR-ABL1−Positive Leukemia

Seiichiro Katagiri, Tetsuzo Tauchi, Seiichi Okabe, et al.

Clin Cancer Res 2013;19:1422-1432. Published OnlineFirst January 14, 2013.

Updated version Access the most recent version of this article at: doi:10.1158/1078-0432.CCR-12-1777

Supplementary Access the most recent supplemental material at: Material http://clincancerres.aacrjournals.org/content/suppl/2013/01/15/1078-0432.CCR-12-1777.DC1 http://clincancerres.aacrjournals.org/content/suppl/2013/03/12/1078-0432.CCR-12-1777.DC2

Cited articles This article cites 36 articles, 14 of which you can access for free at: http://clincancerres.aacrjournals.org/content/19/6/1422.full#ref-list-1

Citing articles This article has been cited by 3 HighWire-hosted articles. Access the articles at: http://clincancerres.aacrjournals.org/content/19/6/1422.full#related-urls

E-mail alerts Sign up to receive free email-alerts related to this article or journal.

Reprints and To order reprints of this article or to subscribe to the journal, contact the AACR Publications Department at Subscriptions [email protected].

Permissions To request permission to re-use all or part of this article, use this link http://clincancerres.aacrjournals.org/content/19/6/1422. Click on "Request Permissions" which will take you to the Copyright Clearance Center's (CCC) Rightslink site.