WO2016073296A1

WO2016073296A1 - Methods for targeting cancer stem cells

Info

Publication number: WO2016073296A1
Application number: PCT/US2015/058251
Authority: WO
Inventors: Gavin Yong WANG; Bradley SCHULTE; Stephen ETHIER; Aimin Yang
Original assignee: MUSC Foundation for Research and Development
Current assignee: MUSC Foundation for Research and Development
Priority date: 2014-11-04
Filing date: 2015-10-30
Publication date: 2016-05-12
Anticipated expiration: 2017-05-04

Abstract

The present application discloses a method for killing cancer stem cells and methods and kits for predicting the responsiveness of a cancer patient to treatment with a Twist 1 inhibitor.

Description

METHODS FOR TARGETING CANCER STEM CELLS

Field of the Invention

[0001] The instant invention relates to a method for killing cancer stem cells with a Twist 1 inhibitor and methods and kits for predicting the responsiveness of a cancer patient to treatment with a Twist 1 inhibitor. All documents cited to or relied upon below are expressly incorporated herein by reference.

Background of the Invention

[0002] Despite recent advances in the treatment of cancer, many patients ultimately experience therapy-resistance and tumor relapse after initial response to therapies. This disappointing clinical outcome highlights an urgent need for the development of innovative therapeutic strategies to overcome the therapy-resistance and recurrence properties of breast cancer, particularly in triple-negative breast cancer.

[0003] Cancer stem cells (CSCs) have been identified as a subset of cells in tumors that can self-renew and differentiate to regenerate tumors, and are thought to contribute significantly to tumor recurrence and metastasis. Notably, CSCs are resistant to current anticancer therapies, including chemotherapy and radiation. These findings suggest that many current anticancer therapies, although killing the bulk of tumor cells and leading to tumor shrinkage, may ultimately fail to cure cancers because they are unable to eradicate CSCs, leading to tumor recurrence and/or metastatic progression. Therefore, targeting of CSCs could be a novel therapeutic strategy to improve the efficacy of cancer therapies.

Summary of the Invention

[0004] The present invention is directed to an in vitro method for predicting responsiveness of a cancer patient to a therapy, comprising the steps of:

- detecting the relative amount of one or several Twist 1 protein expressing cell types in a sample obtained from said cancer patient prior to treatment and using said relative amount, or percentage (%), of cells expressing Twistl protein as a biomarker for predicting said patient's response to treatment with a pharmaceutical composition comprising a Twistl inhibitor; and

- administering a therapeutically effective amount of said pharmaceutical composition comprising said Twistl inhibitor to said patient in need thereof.

[0005] The present invention is also directed to a method of treating cancer in a patient in need thereof, comprising the steps of detecting the relative amount of one or several Twistl protein expressing cell types in a sample obtained from said cancer patient prior to treatment, and administering a therapeutically effective amount of a Twistl inhibitor to said patient in need thereof.

[0006] The present invention is further directed to a method of killing cancer stem cells in a patient in need thereof, comprising the steps of administering a therapeutically effective amount of camptothecin or an analog or derivative thereof to said patient.

[0007] The present invention is additionally directed to a method of killing cancer stem cells in a patient in need thereof, comprising the steps of administering a therapeutically effective amount of camptothecin, or an analog or derivative thereof, to said patient in combination with radiation therapy or at least one other chemotherapeutic agent.

[0008] The present invention is still further directed to a kit for predicting the response to treatment with a pharmaceutical composition comprising a compound which acts as inhibitor to Twistl, comprising:

a) means for obtaining a sample from a patient suffering from cancer;

b) instructions how to detect the relative amount of one or several Twistl protein expressing cell types from said sample;

c) a comparator module, comprising standard values for the parameters detected in b) and instructions on how to use them; and

d) a pharmaceutical composition comprising a therapeutically effective amount of a Twistl inhibitor. Brief Description of the Figures

[0009] Fig. 1 illustrates that taxol-induced expansion of CSCs is associated with an upregulation of Twist 1 expression. (A) SUM 159 cells were treated with 5 nM paclitaxol (Taxol) for 4 days and the percentage of CD44+/CD24- CSCs in therapy-surviving tumor cells was determined by flow cytometry. (B) The percentage of CD44+/CD24-stem cells in Taxol-treated SUM159 cells vs. DMSO-treated control cells was calculated and graphed. (C-D) The expression levels of Twist 1 in Taxol-treated SUM 149 (C) and SUM 159 cells (D) were determined by Western blotting. Lamin A was probed as a loading control. *** p < 0.001 vs. DMSO control.

[0010] Fig. 2 shows that knockdown of Twist 1 inhibits the mammosphere-forming ability of breast cancer cells. (A) SUM 159 cells were transfected with Twist 1 siRNA (20nM) or equal amount of non-targeting negative control (NC) siRNA using Lipofectamine RNAi MAX

(Invitrogen) according to the manufacturer's instructions. Twistl protein levels were determined by Western blot analyses. (B)Tumor sphere assays were performed to determine the self-renewal capacity of CSCs at 48 hrs after siRNA transfection. (C)The number of tumor spheres was counted and normalized to the percentage of the control. (D) MTS assays were performed to determine the impact of Twistl knockdown on the sensitivity of SUM 159 cells to Taxol treatment. *** p < 0.001 vs. NC-siRNA control.

[0011] Fig. 3 illustrates that silencing of Twistl inhibits the tumorigenicity of breast cancer cells. (A) SUM 149 and SUM 159 human breast cancer cells were transfected with Twistl siRNA or NC-siRNA as described in Figure 2A. The expression levels of Twistl were determined by Western blot analyses at 48 h after transfection. (B) Colony formation assays were performed to determine the effects of Twistl silencing on the colony- forming ability of SUM 149 and

SUM 159 cells. (C) Shown are representative photographs of xenografted tumors derived from SUM 149 cells transfected with Twistl or NC siRNA. (D) Tumor volume was measured twice a week using calipers and is plotted versus days after tumor cell implantation. *, p < 0.05 vs. NC- siRNA.

[0012] Fig. 4 illustrates that Twistl regulates TIMP1 expression in breast cancer cells and TIMP1 is involved in CSC self-renewal and drug resistance. (A) Western blot analyses were performed to determine Twistl and TIMP1 expression levels in SUM 159 cells at 48 h after Twistl siRNA transfection. Lamin A was probed as a loading control. (B) Tumor sphere assays were performed to determine the self-renewal capacity of SUM 159 cells transfected with TIMP1 siRNA or NC control siRNA at 48 h post transfection. (C) MTS assays were carried out to determine the impact of TIMP1 silencing (48 h post siRNA transfection) on the sensitivity of SUM159 cells to Taxol at 48 h after Taxol treatment. (D) Western blot analyses show that knockdown of TIMP1 inhibits the phosphorylation of STAT3.

[0013] Fig. 5 illustrates that ATS11 inhibits Twistl expression in a time and dose-dependent fashion. (A) Immunob lotting analyses were performed to determine the protein levels of Twistl in SUM 159 cells. Actin was probed as a loading control. (B) The expression levels of Twistl at different time after ATS11 (0.5 μΜ) treatment was determined by immunob lotting. (C)

Treatment with a proteosome inhibitor MG-132 (10 μΜ) attenuates ATS11 (0.5 μΜ) induced Twistl degradation. Twistl expression levels were determined by Western blots at 4h after ATS11 and/or MG-132 treatment. (D) The expression levels of Twistl in MCF7 and SUM159 cells were determined by Western blot analyses. (E) MTS assays were carried out to determine cell viability of MCF7 and SUM159 cells at 48 h after ATS11 treatment.

[0014] Fig. 6. shows that pharmacologic inhibition of Twistl by ATS11 selectively depletes CSCs. (A) Tumor sphere assays were performed to determine the self-renewal capacity of CSCs in the presence of different concentrations of ATS11 or DMSO as vehicle control. (B) The number of tumor spheres was counted and normalized to the percentage of the control. (C) Representative flow cytometric graphs showing that ATS 11 treatment selectively depletes the CD44+/CD24- stem cell subpopulations in TNBC cells. (D) MTS assays were performed to determine if low dose of ATS11 (100 nM) treatment enhances the therapeutic effects of CDDP in SUM 159 cells.

[0015] Fig. 7 illustrates the chemical structure of ATS-125.

[0016] Fig. 8 illustrates the efficacy and specificity of ATS-125 for targeting Twistl signaling. [0017] Fig. 9 illustrates the efficacy of ATS-125 to deplete CD44+/CD24-/ESA+ breast cancer stem cell subpopulation in culture dishes in vitro.

[0018] Fig. 10 demonstrates the effectiveness of ATS-125 in suppressing the self-renewal capacity of breast cancer stem cells.

[0019] Fig. 11A demonstrates a dramatic tumor suppression effect of ATS-125 in human breast cancer xenografts in vivo. Fig. 11B shows the reduced tumor weight in mice treated with ATS-125 versus vehicle control. Fig. 11C indicates that mice are well tolerated with ATS-125 treatment. No sign of toxicity was observed in mice treated with ATS-125.

[0020] Fig. 12A demonstrates the efficacy of ATS-125 against Twistl in tumor tissues of human breast cancer xenografts in vivo. Fig. 12B confirms the effectiveness of ATS-125 at eliminating breast CSCs in xenograft human breast cancers in vivo. Fig. 12C depicts the quantitative date of Figure 6B's analyses.

[0021] Fig. 13 demonstrates that Taxol treatment is unable to clear CSCs, but a combinatorial therapy with ATS-125 and Taxol can effectively eradicate CSCs.

Detailed Description of the Invention

[0022] It is to be understood that the figures and descriptions of the present invention have been simplified to illustrate elements that are relevant for a clear understanding of the present invention, while eliminating, for the purpose of clarity, many other elements found in typical pharmaceutical compositions. Those of ordinary skill in the art will recognize that other elements and/or steps are desirable and/or required in implementing the present invention. However, because such elements and steps are well known in the art, and because they do not facilitate a better understanding of the present invention, a discussion of such elements and steps is not provided herein. The disclosure herein is directed to all such variations and modifications to such elements and methods known to those skilled in the art. Furthermore, the embodiments identified and illustrated herein are for exemplary purposes only, and are not meant to be exclusive or limited in their description of the present invention.

[0023] Previous studies have shown that residual tumors after chemotherapy and/or radiation are enriched for CSCs, suggesting that CSCs are resistant to anticancer therapies. However, the mechanisms by which CSCs evade therapies are incompletely understood. It is shown herein that Taxol-induced CSC expansion is associated with a significant upregulation of Twistl expression, suggesting that Twistl may play an important role in CSC drug resistance and the self-renewal expansion of CSCs in response to therapies (Fig. 1). In agreement with this suggestion, the inventors' have shown that Twistl knockdown sensitizes SUM 159 cells to Taxol and inhibits the self-renewal capacity of CSCs (Fig. 2). Consistent with the inventors' findings, it was found that overexpression of Twistl promotes CSC proliferation and may confer drug resistance to tumor cells. More strikingly, the inventors' studies reveal that TIMPl is a novel downstream target of Twistl, and that similar to the effects of Twistl knockdown, silencing of TIMPl inhibits CSC self-renewal and sensitizes tumor cells to Taxol (Figs. 2 & 4). These novel findings suggest that silencing of TIMPl may phenocopy the effects of Twistl knockdown. Therefore, targeting of Twistl and/or TIMPl could be exploited as novel therapeutic strategies to eradicate CSCs for cancer treatment. In support of this hypothesis, the inventors show that treatment with ATS11, a small molecule inhibitor of Twistl, depletes CSCs and selectively kills tumor cells expressing high levels of Twistl (Figs. 5 & 6).

[0024] Twist 1 is a transcriptional factor that is overexpressed in many types of human cancers, including subsets of lung and breast cancers. Moreover, clinical studies showed that high levels of Twistl correlate with poor patient survival. These results suggest that Twistl is a potential target for cancer treatment. In agreement with this suggestion, the inventors show that knockdown of Twistl by siRNA markedly suppresses the growth of breast tumors in xenografted mice (Fig. 3). In addition, the inventors' preliminary studies have demonstrated a novel link between Taxol-induced CSC expansion and the upregulation of Twistl expression in therapy- surviving tumor cells, suggesting a critical role for Twistl in CSC self-renewal proliferation and drug resistance (Fig. 1). In support of this hypothesis, the inventors' preliminary studies have demonstrated that Twistl is required for CSC self-renewal function, and that silencing of Twistl sensitizes breast cancer cells to Taxol treatment (Fig. 2). These novel findings strongly support the hypothesis that targeting of Twistl could be exploited as an innovative therapeutic strategy to eradicate CSCs and thus to improve the efficacy of cancer treatment.

[0025] Previous studies indicated that Twistl is a master modulator of epithelial- mesenchymal transition (EMT) and is involved in tumor metastasis and drug resistance. However, it remains to be determined if and how Twistl contributes to CSC drug resistance and its self-renewal expansion in response to anticancer therapies. To this end, the inventors' preliminary studies have demonstrated that TIMPl is a novel downstream target of Twistl (Fig. 4A). More importantly, the inventors found that knockdown of TIMPl inhibits the self-renewal potential of CSCs and sensitizes breast tumor cells to Taxol at least in part via interacting with STAT3 signaling (Fig. 4). These novel findings suggest that Twistl may modulate CSC self- renewal function and CSC drug resistance via regulating TIMPl expression and its interaction with STAT3 signaling.

[0026] The inventors have, thus,:

• identified a novel link between chemotherapy-induced CSC expansion and the upregulation of Twistl expression (Fig. 1), suggesting that Twistl upregulation may contribute to CSC drug resistance via promoting CSC survival and its self-renewal expansion in response to anticancer therapies. • demonstrated that TIMP1 is a novel downstream target of Twistl involved in modulating CSC self-renewal function and drug resistance, suggesting that TIMP1 may mediate the effects of Twistl in CSCs. This is a significant and novel finding because it may lead to the discovery of novel molecular targets for interventions to target the Twistl signaling pathway.

• identified a small molecule inhibitor of Twistl (ATS11 and an analog thereof, e.g, ATS- 25), which degrades Twistl in cancer cells at nanomolar concentrations through a proteasome- dependent mechanism.

[0027] Taxol-induced CSC expansion is associated with a significant upregulation of Twistl expression. Taxol is widely used for cancer therapy, including breast cancer treatment. However, therapy resistance is a significant problem in clinic. To determine the response of CSCs to chemotherapy, the inventors treated SUM149 and SUM159 human breast cancer cells with Taxol and examined changes in CSCs using flow cytometry based stem cell immunophenotyping analyses. The results show that the proportion of CD44+/CD24- subpopulations is markedly increased in SUM 159 cells that survive Taxol treatment (Figs. 1A & B). Similar results were also observed in SUM149 cells (data not shown). These results demonstrate that Taxol treatment enriches CSCs in therapy-surviving tumor cells. Next, the inventors investigated the mechanisms by which Taxol treatment results in CSC enrichment/expansion. Twistl is a master regulator of EMT, which has been shown to be involved in tumor invasion and metastasis. However, the role of Twistl in CSC drug resistance hasn't been well characterized. Surprisingly, the inventors' data reveal that Taxol-induced CSC expansion is associated with a dramatic upregulation of Twistl expression in both SUM 149 and SUM159 cells (Figs. 1C & D). Given the role of Twistl in tumor cell senescence, the inventors' findings suggest that Twistl upregulation may promote CSC survival and its self-renewal expansion by inhibiting therapy-induced senescence.

[0028] Twistl is required for breast CSC self-renewal function. In light of the newly discovered link between Taxol-induced CSC expansion and the upregulation of Twistl expression (Fig. 1), the inventors decided to investigate the role of Twistl in CSC self-renewal function. To this end, the inventors employed siRNA to knockdown Twistl expression in breast cancer cells and examined if silencing of Twistl affects the self-renewal potential of CSCs. Among a set of 3 Twistl siRNAs, the inventors found that siRNA #3 is the most effective one (Fig. 2A). Therefore, siRNA #3 was employed to knockdown Twistl expression for the inventors' subsequent studies. Tumor sphere assays were performed to determine the self- renewal potential of CSCs. Data show that the number of tumor spheres generated from Twistl - siRNA transfected cells is markedly lower than that of cells transfected with non-targeting control (NC) siRNA, demonstrating that silencing of Twistl inhibits the sphere-forming ability of breast cancer cells (Figs. 2B & C). Given that the sphere-forming capacity is a surrogate of CSC self-renewal potential; these results demonstrate that Twistl is required for CSC self- renewal function. Because previous studies showed that Twistl is involved in therapy resistance, the inventors therefore investigated if silencing Twistl has any impact on the sensitivity of breast tumor cells to Taxol. As shown in Fig. 2D, the results demonstrate that knockdown of Twistl by siRNA significantly enhances the tumor suppressive effect of Taxol, suggesting that silencing of Twistl can sensitize tumor cells to chemotherapy.

[0029] Silencing of Twistl suppresses the tumorigenic potential of breast cancer cells.

Although it has been well established that Twistl plays a significant role in EMT, and that Twistl overexpression correlates with poor outcome in subsets of cancer patients, the role of Twistl in breast tumorigenesis remains elusive. To address this issue, the inventors investigated if silencing of Twistl affects breast tumorigenesis using siRNA-mediated gene silencing and xenograft models. It was found that silencing of Twistl did not cause any significant changes in the number of viable tumor cells at 48 h after transfection, compared with that of cells transfected with NC-siRNA (Fig. 3 A & data not shown). However, these viable Twistl -siRNA transfected SUM 149 and SUM 159 cells showed a significant loss in their ability to regenerate tumor colonies (Fig. 3B). Furthermore, xenograft assays show that silencing of Twistl markedly inhibits the tumorigenic potential of SUM149 cells in vivo (Figs. 3C & D). These results demonstrate a critical role for Twistl in breast tumorigenesis. Together with the inventors' findings that Twistl is required for CSC self-renewal activity (Fig. 2); these findings suggest that silencing of Twistl may suppress breast tumorigenesis via inhibiting the self-renewal function of CSCs. [0030] TIMPl is a downstream target of Twistl involved in modulating CSC function and drug resistance. Preliminary studies have demonstrated a critical role for Twistl in CSC self-renewal function and tumorigenesis (Figs. 2 & 3). However, the mechanisms whereby Twistl regulates CSC function are unknown. Using gene microarrays, the inventors found that silencing of Twistl causes a significant change in the expression of more than 900 genes (> 2- fold) in SUM 159 cells (data not shown). Through bioinformatic analyses, the inventors selected a subset of 5 possible target genes of Twistl, including ALDH3B1 and TIMPl, which are altered more than 5-fold by Twistl knockdown and have been shown to be amplified in invasive breast carcinomas, for subsequent validation and functional studies. Data demonstrate that knockdown of Twistl markedly inhibits TIMPl expression in both SUM149 and SUM159 cells, confirming that TIMPl is a novel downstream target of Twistl (Fig. 4A, and data not shown). To determine the function of TIMPl in CSCs, siR A was used to knockdown TIMPl expression and mammosphere assays were performed. The results reveal that silencing of TIMPl markedly inhibits the mammosphere-forming ability of breast cancer cells, suggesting an important role for TIMPl in CSC self-renewal function (Fig. 4B). Moreover, the inventors show that knockdown of TIMPl by siRNA significantly enhances the antitumor effects of Taxol, suggesting that TIMPl may play a role in drug resistance (Fig. 4C). Consistent with the inventors' findings, it has been shown that a high level of TIMPl expression predicts therapy-resistance and/or poor clinical outcome. Furthermore, the inventors show that TIMPl knockdown diminishes the phosphorylation of STAT3 (Fig. 4D). Given the implications of STAT3 in CSCs and drug resistance, these results suggest that TIMPl may mediate the effects of Twistl in CSCs at least in part via interacting with STAT3 signaling.

[0031] ATS11 degrades Twistl through a proteasome-dependent mechanism.

Preliminary studies and those of others have demonstrated that Twistl is a promising target for cancer treatment (Figs 2 & 3). However, the lack of a drug-like small molecule inhibitor of Twistl has been a major barrier to the development of therapies targeting Twistl . To address this challenge, the inventors performed a screening of small molecule natural compounds to discover potential Twistl inhibitors. Through the screenings, the inventors have identified two natural compounds (ATS 11 and ATS 125) that can significantly inhibit the expression of Twistl in a dose and time-dependent fashion (Figs. 5 A & B). Subsequent studies focused on ATS11 because it can reduce the protein levels of Twistl by 95% in breast cancer cells at nanomolar concentrations, while ATS 125 requires a much higher concentration to achieve the same efficacy (data not shown). Mechanistically, the inventors found that treatment with a proteasome inhibitor, MG-132, inhibits ATSl 1 induced Twistl degradation, suggesting that ATSl 1 degrades Twistl protein through a proteasome-dependent mechanism (Fig. 5C). To determine if AST 11 selectively kills tumor cells expressing high levels of Twistl, the inventors investigated the sensitivity of different breast cancer cells with distinctive Twistl expression statues to ATS 11 treatment. As shown in Fig. 5E, the results demonstrate that SUM 159 cells are dramatically sensitive to ATS 11 treatment than MCF-7 cells (Twistl not detectable, Fig. 5D). These novel findings demonstrate for the first time that ATS 11 selectively kills tumor cells expressing high levels of Twistl . Similar results were also observed in SUM149 and other tumor cells.

[0032] Compounds of the present invention can have one or more asymmetric carbon atoms and can exist in the form of optically pure enantiomers, mixtures of enantiomers such as, for example, racemates, optically pure diastereoisomers, mixtures of diastereoisomers,

diastereoisomeric racemates or mixtures of diastereoisomeric racemates. The optically active forms can be obtained for example by resolution of the racemates, by asymmetric synthesis or asymmetric chromatography (chromatography with a chiral adsorbents or eluant). The invention embraces all of these forms.

[0033] As used herein, the term "pharmaceutically acceptable salt" means any

pharmaceutically acceptable salt of the compound of formula (I). Salts may be prepared from pharmaceutically acceptable non-toxic acids and bases including inorganic and organic acids and bases. Such acids include, for example, acetic, benzenesulfonic, benzoic, camphorsulfonic, citric, ethenesulfonic, dichloroacetic, formic, fumaric, gluconic, glutamic, hippuric, hydrobromic, hydrochloric, isethionic, lactic, maleic, malic, mandelic, methanesulfonic, mucic, nitric, oxalic, pamoic, pantothenic, phosphoric, succinic, sulfuric, tartaric, oxalic, p-toluenesulfonic and the like. Particularly preferred are fumaric, hydrochloric, hydrobromic, phosphoric, succinic, sulfuric and methanesulfonic acids. Acceptable base salts include alkali metal (e.g. sodium, potassium), alkaline earth metal (e.g. calcium, magnesium) and aluminum salts. [0034] In the practice of the method of the present invention, an effective amount of any one of the compounds of this invention or a combination of any of the compounds of this invention or a pharmaceutically acceptable salt thereof, is administered via any of the usual and acceptable methods known in the art, either singly or in combination. The compounds or compositions can thus be administered, for example, ocularly, orally (e.g., buccal cavity), sublingually, parenterally (e.g., intramuscularly, intravenously, or subcutaneously), rectally (e.g., by suppositories or washings), transdermally (e.g., skin electroporation) or by inhalation (e.g., by aerosol), and in the form or solid, liquid or gaseous dosages, including tablets and suspensions. The administration can be conducted in a single unit dosage form with continuous therapy or in a single dose therapy ad libitum. The therapeutic composition can also be in the form of an oil emulsion or dispersion in conjunction with a lipophilic salt such as pamoic acid, or in the form of a biodegradable sustained-release composition for subcutaneous or intramuscular administration.

[0035] Useful pharmaceutical carriers for the preparation of the compositions hereof, can be solids, liquids or gases. Thus, the compositions can take the form of tablets, pills, capsules, suppositories, powders, enterically coated or other protected formulations (e.g. binding on ion- exchange resins or packaging in lipid-protein vesicles), sustained release formulations, solutions, suspensions, elixirs, aerosols, and the like. The carrier can be selected from the various oils including those of petroleum, animal, vegetable or synthetic origin, e.g., peanut oil, soybean oil, mineral oil, sesame oil, and the like. Water, saline, aqueous dextrose, and glycols are preferred liquid carriers, particularly (when isotonic with the blood) for injectable solutions. For example, formulations for intravenous administration comprise sterile aqueous solutions of the active ingredient(s) which are prepared by dissolving solid active ingredient(s) in water to produce an aqueous solution, and rendering the solution sterile. Suitable pharmaceutical excipients include starch, cellulose, talc, glucose, lactose, talc, gelatin, malt, rice, flour, chalk, silica, magnesium stearate, sodium stearate, glycerol monostearate, sodium chloride, dried skim milk, glycerol, propylene glycol, water, ethanol, and the like. The compositions may be subjected to

conventional pharmaceutical additives such as preservatives, stabilizing agents, wetting or emulsifying agents, salts for adjusting osmotic pressure, buffers and the like. Suitable

pharmaceutical carriers and their formulation are described in Remington's Pharmaceutical Sciences by E. W. Martin. Such compositions will, in any event, contain an effective amount of the active compound together with a suitable carrier so as to prepare the proper dosage form for proper administration to the recipient.

[0036] The dose of a compound of the present invention depends on a number of factors, such as, for example, the manner of administration, the age and the body weight of the subject, and the condition of the subject to be treated, and ultimately will be decided by the attending physician or veterinarian. Such an amount of the active compound as determined by the attending physician or veterinarian is referred to herein, and in the claims, as a "therapeutically effective amount". For example, the dose of a compound of the present invention is typically in the range of about 1 to about 1000 mg per day. Preferably, the therapeutically effective amount is in an amount of from about 1 mg to about 500 mg per day.

[0037] Compounds of the present invention can be prepared beginning with commercially available starting materials and utilizing general synthetic techniques and procedures known to those skilled in the art. Chemicals may be purchased from companies such as for example Aldrich, Argonaut Technologies, VWR and Lancaster. Chromatography supplies and equipment may be purchased from such companies as for example AnaLogix, Inc, Burlington, Wis.;

Biotage AB, Charlottesville, Va.; Analytical Sales and Services, Inc., Pompton Plains, N.J.; Teledyne Isco, Lincoln, Nebr.; VWR International, Bridgeport, N.J.; Varian Inc., Palo Alto, Calif, and Multigram II Mettler Toledo Instrument Newark, Del. Biotage, ISCO and Analogix columns are pre-packed silica gel columns used in standard chromatography.

Kit and Device

[0038] In one aspect the invention relates to a kit and in another aspect to a device for predicting the response, preferably of a cancer in a subject (or patient), to a Twistl inhibitor or a pharmaceutically acceptable derivative thereof as defined herein, said kit comprising means, tools or devices for sample collection, especially for collection of blood- or bone marrow samples and instructions how to detect relative amounts of Twistl protein expressing cells as defined herein in said samples. [0039] The kit and device may also preferably comprise a comparator module which comprises a standard value or set of standard values to which the level of said Twistl protein expressing cells as defined herein in the sample is compared.

[0040] In another embodiment, provided is a kit for predicting the response to treatment with a pharmaceutical composition comprising a compound which acts as an inhibitor to Twistl, comprising:

a) means for obtaining a sample from a patient suffering from cancer;

d) a pharmaceutical composition comprising a therapeutically effective amount of a Twistl inhibitor.

Examples

[0041] The disclosure is further illustrated by the following examples, which are not to be construed as limiting this disclosure in scope or spirit to the specific procedures herein described. It is to be understood that the examples are provided to illustrate certain embodiments and that no limitation to the scope of the disclosure is intended thereby. It is to be further understood that resort may be had to various other embodiments, modifications, and equivalents thereof which may suggest themselves to those skilled in the art without departing from the spirit of the present disclosure and/or scope of the appended claims.

Example 1

Pharmacologic inhibition of Twistl by ATSll selectively depletes CSCs

[0042] Preliminary studies have shown that knockdown of Twistl suppresses the self-renewal capacity of CSCs in vitro and inhibits tumorigenesis in vivo (Figs. 2 & 3). However, it remains to be determined if a drug-like small molecule inhibitor of Twistl, ATSl l, can be used to pharmacologically target CSCs. To address this question, mammosphere assays were performed to examine if ATS 11 affects the self-renewal capability of CSCs in culture. The results reveal that ATS11 dramatically inhibits the mammosphere-forming ability of SUM 159 cells, suggesting that pharmacologic inhibition of Twist 1 by ATS 11 may be a novel and effective strategy to target CSCs (Figs. 6A & B). Consistent with this finding, further flow cytometric analyses show that ATS 11 selectively decreases the proportion of CD44+/CD24- stem cell subpopulation in SUM149 cells (Fig. 6C). Together, these novel findings demonstrate for the first time that inhibition of Twist 1 by ATS 11 selectively depletes CSCs, which support the hypothesis that pharmacologic inhibition of Twist 1 can be exploited as a novel therapeutic strategy to eliminate CSCs for cancer treatment.

[0043] In light of the role of CSCs in therapy-resistance, the inventors sought to determine whether targeting of CSCs by ATS 11 can sensitize cancer cells to chemotherapy. It is worth noting that both SUM 149 and SUM 159 cells are derived from triple-negative breast cancer (TNBC), which represents the most virulent subtype of this disease. Chemotherapies such as taxanes are initially effective in most patients with metastatic TNBC, but the majority of these tumors recur at later time. Increasing evidence suggests that therapy-resistant CSCs may be responsible for these tumor relapses after initial response to therapies. Recently, cisplatin (CDDP) has been testing for the treatment of TNBC. Thus, the inventors decided to examine if ATS11 can sensitize TNBC cells to CDDP treatment. The data demonstrate that CDDP (20 μΜ) treatment alone leads to a modest growth suppression in SUM 159 cells. In contrast, a low dose of ATS 11 (100 nM) co-treatment dramatically enhances the tumor cell suppression effects of CDDP (Fig. 6D). These results demonstrate that inhibition of Twist 1 by ATS11 sensitizes TNBC cells to CDDP. Given the effectiveness of ATS 11 at depleting CSCs (Figs. 6A-C), these novel findings suggest that a combined treatment with ATS 11 and CDDP may significantly improve the efficacy of chemotherapy by eradicating drug-resistant CSCs in TNBC patients.

[0044] Experimental design & methods. The inventors will infect tumor cells with lentiviruses expressing Twistl shRNA (Open Biosystems) to silence Twistl expression in SUM149 and SUM159 cells. Cell lines stably expressing Twistl shRNA will be established by puromycin selection. The knockdown of Twistl by shRNA will be confirmed by quantitative real time RT-PCR and Western blot analyses. The inventors will then expose cells stably expressing Twistl shRNA to Taxol treatment (10 nM) or DMSO as vehicle control to see if silencing of Twistl inhibits Taxol-induced CSC expansion. Taxol-induced expansion of CSCs in therapy-surviving tumor cells will be determined by flow cytometric analyses as shown in Figure 1.

[0045] To determine whether silencing of Twistl induces senescence in CSCs, Twistl shRNA transduced SUM149 and SUM159 cells will be stained with FITC-ESA, PE-CD44 and APC-CD24 antibodies (BD Biosciences) to label stem cell surface markers. Then, breast CSCs will be sorted using a Becton Dickson FACS Aria II cell sorter, based on their CD44- /CD24+/ESA+ immunophenotype. Senescence induction in the stem cell subpopulation of SUM149 and SUM159 cells will be determined by senescence associated β-galactosidase (SA-β- gal) staining as the inventors previously reported, while apoptosis in CSCs will be examined by Annexin V staining as well as activated caspase-3 assays.

[0046] Next, the inventors will examine if silencing of Twistl by shRNA inhibits the self- renewal/tumorigenic potential of breast CSCs in vivo using limiting dilution assays as described previously. Briefly, different dilutions (200; 2,000; 20,000; and 100,000 cells) of sorted breast CSCs will be subcutaneously xenotransplanted into the flanks of NOD/SCID mice as shown in Figure 3C. To minimize the variations between animals, non- silencing (NS) shRNA-transduced cells will be injected into the left flank, while Twistl shRNA-transduced cells will be injected into the right flank of the same animal. The formation of tumors will be monitored for 5 weeks and tumor volume will be measured twice a week as shown in Figure 3.

[0047] To confirm the impact of Twistl knockdown on CSC self-renewal activity, secondary xenotransplantation will be performed using cells isolated from the primary xenografts as described previously. Tumor tissues will be histologically examined for senescent and apoptotic cells using SA-P-gal staining and TUNEL assays, respectively. CSCs in tumor tissues will be determined by ALDH1 immunostaining.

[0048] Interpretation of results and statistical considerations. Based on the finding that Twistl is required for CSC self-renewal (Fig. 2), the inventors anticipate that silencing of Twistl by shRNA will inhibit Taxol-induced expansion of CSCs. It is also predicted that silencing of Twistl will inhibit the formation of tumors by CSC transplantation both in 1st and secondary limiting dilution assays. In light of the role of Twistl in tumor cell senescence, the inventors anticipate that silencing of Twistl will cause increased SA-P-gal staining in CSCs. These results will support the hypothesis that silencing of Twistl inhibits therapy-induced CSC expansion and its self-renewal function via increasing senescence induction in CSCs. A total of 160 mice (4 dilutions x 10 mice/dilution x 2 xenotransplantation assays x 2 cell lines = 160) will be needed. With 10 mice per dilution, assuming an average of 7 days difference of tumor formation (see Figure 3D), and a standard deviation of difference in tumor formation time of 6 days (or less), the inventors have at least 89% power with a two-sided alpha of 0.05.

[0049] Alternative approaches. If toxicity is an issue for permanent and stable knockdown of Twistl in CSCs with shRNA, as an alternative approach the inventors will use the tetracycline inducible shRNA system to adjustably knockdown Twistl expression or employ siRNA to transiently silence Twistl in sorted CSCs.

Example 2

Determination of whether overexpression of Twistl promotes the self-renewal potential of

CSCs and enhances therapy-induced CSC expansion

[0050] Rationale. Preliminary studies have demonstrated that Taxol-induced CSC expansion is associated with a dramatic upregulation of Twistl expression, suggesting a role for Twistl in CSC drug resistance and its self-renewal expansion in response to chemotherapy (Fig. 1). In agreement with this suggestion, it has been shown that overexpression of Twistl promotes the generation of CSCs in human breast tumor cells. However, it remains to be determined whether overexpression of Twistl promotes CSC self-renewal activity and enhances Taxol-induced CSC expansion via inhibiting therapy-induced senescence and/or apoptosis.

[0051] Experimental design & methods. To determine if overexpression of Twistl promotes CSC self-renewal and enhances therapy-induced CSC expansion, the inventors will transfect MCF7 and SUM 149 cells with Twistl expressing vector or empty vector as control (OriGene). Stable cell lines will be established by G418 selection. The overexpression of Twistl will be confirmed by quantitative real-time RT-PCR and Western blot analyses. Then, the inventors will expose cells stably over-expressing Twistl to Taxol treatment (10 nM) or DMSO as vehicle control to determine if overexpression of Twistl promotes Taxol-induced CSC expansion in MCF7 and SUM149 cells. The expansion of CSCs by Taxol treatment in therapy- surviving tumor cells will be determined by flow cytometric analyses as shown in Figure 1.

[0052] To determine whether overexpression of Twistl promotes Taxol-induced CSC expansion by inhibiting therapy-induced senescence and/or apoptosis in CSCs, the inventors will stain the cells over-expressing Twistl with FITC-ESA, PE-CD44 and APC-CD24 antibodies (BD Biosciences) to label stem cell surface markers. Breast CSCs will be sorted using a Becton Dickson FACS Aria II cell sorter, based on their CD44-/CD24+/ESA+ immunophenotype. Senescence in the stem cell subpopulations of SUM149 and SUM159 cells will be determined by SA-P-gal staining, while the apoptotic CSCs will be detected by Annexin V staining as well as activated caspase-3 assays.

[0053] Next, the inventors will examine if overexpression of Twistl enhances the self- renewal potential of CSCs in vivo using limiting dilution xenotransplantation assays as described previously. Briefly, different dilutions (200; 2,000; 20,000; and 100,000 cells) of sorted CD44- /CD24+/ESA+ stem cells will be subcutaneously injected into the flanks of NOD/SCID mice as described above. The formation of tumors will be monitored for 5 weeks and tumor volume will be measured twice a week as shown in Figure 3D.

[0054] To confirm the impact of Twistl overexpression on CSC self-renewal potential, secondary xenotransplantation assays will be performed using cells isolated from the primary xenografts as described above. Senescent and apoptotic cells in tumor tissues will be determined by SA-P-gal staining and TUNEL assays, respectively. CSCs in tumor tissues will be detected by ALDH1 immunostaining.

[0055] Interpretation of results and statistical considerations. It is anticipated that overexpression of Twistl will enhance Taxol-induced CSC expansion. It is also predicated that over expression of Twistl will inhibit therapy-induced senescence and/or apoptosis in CSCs. This prediction is based on the observations showing that Twistl knockdown induces senescence and/or apoptosis in tumor cells. Based on the inventors' preliminary studies showing that silencing of Twistl inhibits the self-renewal capacity of CSCs in vitro and suppresses the formation of tumors in vivo (Figs. 2 & 3), the inventors predict that Twistl overexpression will increase the self-renewal potential of CSCs and thus promote the formation of tumors by xenotransplanted CSCs both in primary and secondary limiting dilution assays. A signed-rank test will be used to determine if there is a difference in time to tumor formation on left vs. right flanks. Statistical analysis and sample size considerations are the same as those proposed above. A total of 160 mice (4 dilutions x 10 mice/dilution x 2 transplantation assays x 2 cell lines = 160) will be used for the proposed studies.

Example 3

Silencing of Twistl in established human breast cancer xenografts inhibits tumor growth and progression via inducing CSC senescence and/or apoptosis in tumor tissues

[0056] Rationale. Although the inventors' preliminary studies have demonstrated that silencing of Twistl inhibits the formation of tumors in a xenograft model (Fig. 3), it remains to be determined if genetic silencing of Twistl in established tumors inhibits the growth and metastatic progression of breast cancer.

[0057] Experimental design & methods. To achieve this goal, the inventors will infect SUM 149 and SUM 159 cells with lentiviruses expressing inducible Twistl shRNA (Open Biosystems). Stable cell lines will be established by puromycin selection. Cells expressing inducible Twistl shRNA will be injected into the fat pads of NOD/SCID mice to establish orthotopic xenografts of breast cancer cells as previously described. Approximately two weeks after tumor cell injection, tumor bearing mice will be randomly divided into the following two groups: 1) control and 2) doxycycline (Dox) treatment. Mice will receive Dox (200 μg/ml) in drinking water to induce Twistl shRNA expression in tumor tissues in vivo. Tumors will be measured twice per week using calipers, and tumor volume will be calculated and graphs will be plotted as shown in Figure 3D. [0058] Experiments will end at 4 weeks after Dox treatment. At the end of the experiment, animals will be sacrificed (it is expected all mice will survive to 4 weeks with none sacrificed prior due to morbidity). Tumors will be surgically removed, weighted and subjected to histological tissue section preparations. Lung will be examined histologically to determine the number and size of metastatic tumors as previously described. The expression levels of Twistl in tumor tissues will be determined by immunohistochemistry (IHC). Xenograft derived single cell preparation and flow cytometric analysis of CSCs in tumor tissues will be performed using ALDH1 immunostaining as previously described. Senescent and apoptotic cells in tumor tissues will be determined by SA-P-gal staining and TUNEL assays, respectively.

[0059] Interpretation of results and statistical considerations. Based on the inventors' preliminary studies showing that Twistl knockdown inhibits the formation of breast tumors in a xenograft model (Fig. 3), the inventors predict that Dox treatment will lead to the down- regulation of Twistl expression along with increased SA-P-gal staining in tumor tissues, and that the down-regulation of Twistl will be associated with a significant suppression of tumor growth. Given the important role of Twistl in EMT and tumor metastasis, the inventors also expect that silencing of Twistl by Dox treatment will significantly inhibit the development of metastatic tumors in lungs. Moreover, the inventors predict that the antitumor effect of Dox treatment will be associated with an increased SA-P-gal staining in CSCs. These results will validate the hypothesis that genetic silencing of Twistl inhibits tumor growth and metastatic progression by inducing CSC premature senescence in tumor tissues. Graphical displays will be used to evaluate differences across groups. Random effects linear regression will be used to model (log) tumor volume over time comparing slopes. Number and volume of metastases and gene expression will be compared across groups using two-sample t-tests (with outcomes transformed if needed). A total of 80 mice (10 mice/ x 4 groups x 2 cell lines = 80) will be needed for the studies proposed above. With 10 mice per group, the inventors have 89% power to detect a significant difference of 1.5 standard deviations with a two-sided alpha of 0.05. Preliminary data suggest a difference of this magnitude is reasonable (Fig. 3D). Example 4

Testing the hypothesis that TIMPl mediates the effects of Twistl in CSCs

[0060] Rationale. Although the role of Twistl in CSCs and drug resistance has been documented, the key downstream target(s) of Twistl responsible for mediating Twistl 's effects in CSCs are unknown. To address this issue, the inventors' preliminary data have demonstrated that TIMPl is a novel downstream target of Twistl, and that TIMPl is involved in CSC self- renewal and drug resistance (Fig. 4). Consistent with the inventors' findings, it has been shown that a high level of TIMPl expression correlates with therapy-resistance and/or unfavorable patient survival. More importantly, the inventors' preliminary studies demonstrate that silencing of TIMPl exhibits a similar inhibitory effect on CSC activity as Twistl knockdown does (Figs. 2C & 4B). These novel findings support the hypothesis that TIMPl may mediate the effects of Twistl in CSCs.

Example 5

To validate whether Twistl transcriptionally regulates TIMPl expression

[0061] Rationale. Preliminary studies have shown that silencing of Twistl significantly down-regulates the expression of TIMPl in SUM149 and SUM159 cells, demonstrating that TIMPl is a novel downstream target of Twistl (Fig. 4). However, it remains to be determined whether Twistl transcriptionally regulates TIMPl expression in breast cancer cells.

[0062] Experimental design & methods. First, the inventors will perform ChIP assays to evaluate the interaction between Twistl protein and TIMPl promoter using an EZ-ChIP chromatin immunoprecipitation kit (Millipore) according to the manufacturer's instructions. Briefly, SUM159 cells will be treated with 18.5% formaldehyde to cross-link proteins to DNA. Then the cells will be lysed with a lysis buffer and sonications will be performed to shear the chromatin to a manageable size (200 to 1,000 bp). Immunoselection of cross-linked protein- DNA will be carried out using anti-mouse IgG (negative control) and mouse anti-human Twistl monoclonal antibody (Santa Cruz) along with protein G-conjugated agarose beads. Protein-DNA complexes will be washed and then protein-DNA cross-links will be reversed to free DNA. Finally, the purified DNA will be analyzed for the presence of TIMPl promoter sequence using quantitative real-time PCR analyses.

[0063] In addition, luciferase reporter assays will be conducted to determine the effects of Twistl knockdown on TIMPl promoter activity. Briefly, SUM 159 cells will be co-transfected with firefly luciferase TIMPl reporter plasmid (500 ng) and Renilla luciferase expression plasmid (50 ng) along with Twistl siRNA or NC-siRNA as control for 48 h. Both firefly and Renilla luciferase activities will be measured with the dual luciferase reporter system (Promega). Firefly luciferase activity will be normalized with Renilla luciferase to evaluate the transcriptional activity of TIMPl promoter. Similar approaches will be used to determine if overexpression of Twistl enhances the transcriptional activity of TIMPl promoter in 293T cells.

[0064] Interpretation of results and statistical considerations. It is predicted that ChIP assays will demonstrate a DNA-protein binding association between Twistl protein and TIMPl promoter. It is also predicted that TIMPl luciferase reporter assays will show that knockdown of Twistl attenuates, whereas overexpression of Twistl enhances the transcriptional activity of TIMPl promoter. These results will confirm the hypothesis that Twistl transcriptionally regulates TIMPl expression. Graphical displays will be used to demonstrate differences by conditions. Statistical analyses will be carried out using ANOVA and t-test. Differences will be considered statistically significant if p < 0.05.

Example 6

To examine whether silencing of TIMPl recapitulates the effects of Twistl knockdown and whether overexpression of TIMPl can rescue CSCs from the effects of Twistl knockdown

[0065] Rationale. Although the inventors' preliminary studies have demonstrated that TIMPl is a downstream target of Twistl, and that TIMPl is involved in modulating CSC self-renewal activity and drug resistance (Fig. 4), further studies are needed to confirm that TIMPl mediates the effects of Twistl in CSCs. [0066] Experimental design & methods. To determine if silencing of TIMPl phenocopies the effects of Twistl knockdown, the inventors will transduce SUM 149 and SUM 159 cells with shRNA to knockdown TIMPl expression using lentivirus vectors (Open Biosystems). The knockdown of Twistl expression by shRNA will be confirmed by quantitative real-time PCR and Western blot analyses. Next, the inventors will treat the TIMPl shRNA transduced cells with Taxol to determine if silencing of TIMPl inhibits therapy-induced expansion of CSCs. The expansion of CSCs by Taxol treatment will be determined by flow cytometric analyses as shown in Figure 1.

[0067] To determine whether silencing of TIMPl induces senescence in CSCs, The TIMPl shRNA transduced SUM149 and SUM159 cells will be stained with FITC-ESA, PE-CD44 and APC-CD24 antibodies (BD Biosciences) to label stem cell surface markers. Breast CSCs will be sorted using a Becton Dickson FACS Aria II cell sorter, based on their CD44-/CD24+/ESA+ immunophenotype. Senescent cells will be determined by SA-P-gal staining, while apoptosis will be examined by Annexin V staining and activated caspase-3 assays. In addition, mammosphere assays will be performed to determine the self-renewal potential of sorted CSCs in vitro. Limiting dilution assays will be employed to evaluate the self-renewal/tumorigenic potential of TIMPl shRNA transduced CSCs in vivo using approaches similar to those described above.

[0068] Finally, the inventors will examine if overexpression of TIMPl rescue CSCs from the effects of Twistl knockdown. Briefly, SUM149 cells will transfected with TIMPl expressing vectors (OriGene) or empty vectors as control. Stably cell lines will be established by G418 selection. The overexpression of TIMPl in transfected cells will be confirmed by quantitative real-time PCR and Western blot analyses. Next, the inventors will transduce cells stably over- expressing TIMPl with Twistl shRNA to knockdown Twistl expression to determine whether overexpression of TIMPl attenuates the effects of Twistl knockdown in CSCs. The effects of Twistl knockdown on Taxol-induced CSC expansion, senescence and apoptosis induction in breast CSCs over-expressing TIMPl will be determined.

[0069] Interpretation of results and statistical considerations. It is anticipated that similar to the effects of Twistl knockdown, silencing of TIMPl will increase SA-P-gal staining in sorted CSCs but decrease the ability of CSCs to generate tumors in xenotransplanted mice in both primary and secondary limiting dilution assays. These results will confirm the hypothesis that silencing of TIMPl recapitulates the effects of Twist 1 knockdown. In addition, the inventors also predict that overexpression of TIMPl will block Twist 1 knockdown-induced senescence and attenuate Twistl knockdown-induced inhibitory effects on the self-renewal capacity of CSCs. These results will confirm that overexpression of TIMPl can rescue CSCs from the effects of Twistl knockdown. For the in vitro studies, statistical analyses will be carried out using ANOVA and t-test. Differences will be considered statistically significant if p < 0.05. For the animal studies, statistical analysis and sample size considerations are the same as those proposed above. A total of 80 mice (10 mice/dilutions x 4 dilutions x 2 transplantation assays = 80) will be used.

Example 7

To determine the correlation between Twistl and TIMPl expression in breast tumor tissues and the co-localization of Twistl and TIMPl immunostaining with ALDH1 CSC maker in tumor cells

[0070] Rationale. Preliminary studies have demonstrated that TIMPl is a downstream target of Twistl (Fig. 4), suggesting that a correlation between Twistl and TIMPl expression may be established in tumor tissues.

[0071] Experimental design & methods. To determine the correlation between Twistl and TIMPl expression in tumor tissues, sixty snap-frozen tumor tissues will be obtained from the Tissue Biorepository core of the Hollings Cancer Center (HCC) at the Medical University of South Carolina (MUSC). Tumor samples will be retrospectively selected from breast cancer patients undergoing curative surgery at the MUSC hospital. The identity of the patients will not be known and patient confidentiality and anonymity will be maintained at all times. The expression levels of Twistl and TIMPl in tumor tissues will be determined by immunohistochemical (IHC) analysis. The IHC staining results will be scored 0 (negative), 1 (weak positive), 2 (positive), and 3 (strong positive). The expression levels of Twistl and TIMPl will be dichotomized as expressed (2 or 3) or not expressed (0 or 1). [0072] In addition, the inventors also will examine whether TIMPl and Twistl co-localize with the immunostaining for ALDHl CSC marker in breast tumor cells. This will provide evidence demonstrating that Twistl and/or TIMPl are physically associated with CSCs. To achieve this goal, the inventors will label SUM 149 and SUM 159 cells with a rabbit anti-Twist 1 antibody (Santa Cruz) in conjunction with an Alexoflour-488 (Green) conjugated goat anti-rabbit secondary antibody. Meanwhile, a mouse anti-ALDHl monoclonal antibody (BD Biosciences) along with a rabbit anti-mouse secondary antibody labeled with Alexofluor-555 (Red) will be used to probe the ALDHl CSC marker. Nuclei will be visualized by DAPI (Blue) counter staining. The co-localization of Twistl immunostaining with ALDHl staining will be examined using a confocal fluorescent microscope. Similar approaches will be employed to determine the distribution and co-localization of TIMPl and ALDHl in breast tumor cells.

[0073] Expected results and statistical considerations. Based on the inventors' preliminary studies showing that TIMPl is a downstream target of Twistl (Fig. 4), the inventors predict that a correlation between Twistl and TIMPl expression will be established in breast tumor tissues. Given the role of Twistl in CSCs and that TIMPl is a downstream target of Twistl ; the inventors also predict that Twistl and TIMPl immunostaining will co-localize with the staining for ALDHl in breast tumor cells. 60 tumor samples obtained from the HCC's tissue Biorepository core will be examined. This sample size yields 80% power to detect a difference in TIMPl versus Twistl expression of 0.66 SDs based on a two-sided paired t-test with a = 0.05.

Example 8

Todetermine if pharmacologic inhibition of Twistl sensitizes breast tumors to

chemotherapy by targeting CSCs

[0074] Rationale. Mounting evidence has indicated that therapy-resistant CSCs may play a significant role in tumor recurrence and metastatic progression after initial response to treatment, which emphasizes an urgent need for the development of novel strategies to eradicate CSCs and thus to improve the efficacy of cancer therapies. Recent studies suggest that Twistl is a promising target for cancer treatment. More importantly, the inventors' preliminary studies have demonstrated that silencing of Twistl inhibits CSC self-renewal in vitro and suppresses tumor growth in vivo (Figs. 2 & 3). These novel findings prompted us to hypothesize that targeting of Twistl can be exploited as a novel therapeutic strategy to eliminate CSCs for cancer treatment. In support of this hypothesis, the inventors have identified a small molecule inhibitor of Twistl, ATS11 , that can selectively kill tumor cells expressing high levels of Twistl and deplete CSCs in culture (Figs. 5 & 6).

Example 9

To determine if ATS11 sensitizes CSCs to therapy-induced senescence and/or apoptosis

[0075] Rationale. Preliminary studies have shown that knockdown of Twistl inhibits CSC self-renewal and sensitizes SUM159 cells to Taxol (Fig. 2). It is also shown that treatment with a Twistl small molecule inhibitor ATS 11 inhibits the mammosphere-forming capacity of CSCs and sensitizes TNBC cells to CDDP (Fig. 6). These results suggest that pharmacologic inhibition of Twistl by ATS 11 can enhance the efficacy of anticancer therapies by eliminating CSCs. Given the important role of Twistl in inhibiting tumor cell senescence, the inventors hypothesize that ATS 11 may sensitize tumors to chemotherapy via enhancing therapy-induced senescence and/or apoptosis in CSCs.

[0076] Experimental design & methods. To determine if ATS 11 sensitizes CSCs to therapy-induced senescence and/or apoptosis, the inventors will treat SUM149, SUM159 and MCF7 cells with different doses of CDDP (10, 20 or 50 μΜ) in the presence of 100 nM ATS 11 or DMSO as vehicle control. This ATS 11 dose selection is based on the observations showing that treatment with 100 nM of ATS 11 markedly enhances the antitumor effects of CDDP in tumor cells expressing high levels of Twistl, but exhibits no significant effect on the proliferation of cells expressing undetectable Twistl (Figs. 5E & 6D). At 24 and 72 h after CDDP treatment, cell viability will be determined using MTS assays. Therapy-induced senescent cells in the CSC subpopulation will be determined by SA-P-gal staining. Apoptosis in CSC subpopulations will be examined using Annexin V staining along with flow cytometric analyses. In addition, tumor sphere assays will be performed to examine whether there is any change in the self-renewal potential of CSCs after different dose of CDDP treatments. Similar approaches will be employed to determine whether ATS 11 sensitizes CSCs to the treatment of other chemotherapeutic agents, including Taxol and doxorubicin.

[0077] Expected results and statistical considerations. Because CSCs have been shown to be resistant to chemotherapy and radiation, the inventors predict that CDDP treatment alone will not cause any significant changes in senescence and/or apoptosis induction in CSCs. In contrast, the inventors anticipate that ATS 11 and CDDP combined treatment will lead to a significant increase in senescence and/or apoptosis induction in CSCs. This prediction is based on the observations that ATS 11 inhibits CSC self-renewal activity and markedly enhances the antitumor effects of CDDP (Fig. 6). Statistical analysis will follow.

Example 10

To examine if pharmacologic inhibition of Twistl by ATS11 inhibits tumor cell invasion in vitro and prevents tumor progression in vivo

[0078] Rationale. Given the implications of Twistl in CSCs and tumor metastasis, the inventors hypothesize that pharmacologic inhibition of Twistl by ATS11 may inhibit tumor cell invasion in vitro and prevent metastatic progression in vivo via targeting CSCs and/or EMT signaling.

[0079] Experimental design & methods. To determine if ATS 11 treatment inhibits breast tumor cell migration and invasion, the inventors will incubate SUM 149 and SUM 159 cells with different doses of ATS 11 (50, 100 and 200 nM) or DMSO as vehicle control. At 6 h after ATS11 treatment, wound-healing assays and transwell analyses will be performed to determine the effects of Twistl inhibition by ATS11 on tumor cell migration and invasion using procedures as previous described.

[0080] To examine if ATS 11 treatment prevents tumor metastatic progression, first the inventors will determine the optimal dose of ATS 11 required for the inhibition of Twistl in tumor tissues in vivo. Briefly, SUM 149 cells will be used to establish xenograft tumors in NOD/SCID mice (5 per dose) as shown in Fig. 3C. Then, tumor bearing mice will receive a single i.p. injection of different doses of ATS 11 (2, 5, 10 and 20 mg/kg body weight) or DMSO as a vehicle control. The expression levels of Twistl in tumor tissues will be determined at 24h after ATS11 treatment by IHC analyses. The minimum dose will be selected that can reduce the expression levels of Twistl in tumor tissues by 80% in at least 4 of 5 mice as a starting dose for the follow-up studies.

[0081] Next, the SUM 149 orthotopic xenograft model will be established as described above and will be used to determine if pharmacologic inhibition of Twistl by ATS 11 treatment prevents tumor metastatic progression in vivo in a preclinical setting. Approximately 2 weeks after tumor cell implantation, tumor bearing mice will be randomly divided into 3 groups: 1) vehicle control; 2) ATS 11 low dose (the minimum dose established by a pilot experiment as described above, for example, 5 mg/kg); and 3) ATS 11 high dose (3-fold of the minimum dose, for example 15 mg/kg). ATS 11 will be dissolved in DMSO and diluted with PBS (DMSO final concentration will be less than 5%) before i.p. injection, twice per week for 4 weeks. Changes in tumor volume will be measured twice per week using calipers. Experiments will end at 3 days after the final dose of ATS11. At the end of this experiment, animals will be sacrificed. Tumors will be surgically removed, weighted and subjected to histological tissue section analyses. Lung will be examined histologically to assess the number and size of metastatic tumors as previously described. The expression levels of Twistl, TIMP1 and EMT markers including E-cadherin, ZO- 1, Vimentin and N-cadherin in tumor tissues will be determined by IHC. The number of CSCs in tumor tissues will be determined by ALDH1 immunostaining as previously described. Senescence and apoptosis will be assessed by SA-P-gal staining and TUNEL assays. The same approaches will be used to examine the efficacy of ATS11 in SUM159 cells derived xenograft tumors.

[0082] Expected results and statistical considerations. Based on the observations that Twistl is required for CSC self-renewal (Fig. 2), and that Twistl is involved in EMT and tumor cell invasion, the inventors predict that ATS 11 treatment will significantly inhibit the migration and invasion of SUM 149 and SUM 159 cells in vitro as determined by wound-healing assays and transwell analyses. In addition, the inventors anticipate that ATS 11 treatment will reduce the number and size of metastatic nodules in the lungs of xenotransplanted mice. It is also expected that the tumor progression inhibition effects of ATS 11 will be associated with a significant down-regulation of the expression of Twist 1, TIMP1 and EMT markers along with a decrease in CSC numbers in tumor tissues. A total of 60 NOD/SCID mice will be required (10 mice/group x 3 groups x 2 cell lines = 60). Statistical analysis and sample size considerations are the same as those proposed above.

Example 11

To determine if ATS11 treatment sensitizes breast tumors to chemotherapy by eliminating

CSCs in tumor tissues

[0083] Rationale. Drug resistance is a significant problem in cancer treatment and CSCs are resistant to therapies. Notably, increasing evidence indicates that targeting CSCs could sensitize tumors to chemotherapy. In view of the observations showing that knockdown of Twist 1 inhibits the self-renewal capacity of CSCs in vitro and suppresses tumor growth in vivo (Figs. 2 & 3), it is logical to hypothesize that pharmacologic inhibition of Twist 1 by ATS 11 can sensitize breast tumors to CDDP by eliminating CSCs. In agreement with this hypothesis, recent studies have shown that CSCs are resistant to CDDP treatment, and that targeting CSCs sensitizes tumors to platinum therapy. Moreover, the inventors' preliminary studies have demonstrated that ATS 11 sensitizes TNBC cells to CDDP treatment most likely via depleting CSCs (Fig. 6).

[0084] Experimental design & methods. To determine if ATS11 sensitizes breast tumors to chemotherapy, SUM 149 orthotopic xenografts will be established as described previously. Tumor bearing mice will be randomly divided into 4 groups: 1) vehicle control; 2) ATS 11 (optimal dose will be determined by a pilot experiment described in earlier Examples, for example, 5 mg/kg); 3) CDDP (20 mg/kg); and 4) ATS 11 and CDDP combined treatment (i.p. injection twice per week for 4 weeks). Tumor size will be measured twice per week using calipers. Experiments will end at 3 days after the final dose of ATS 11 and CDDP treatment, at which time it is expected that all mice will still be alive. At the end of this experiment, animals will be sacrificed. Tumors will be surgically removed, weighted and subjected to histological tissue section analyses. The expression levels of Twist 1, TIMP1 and EMT markers in tumor tissues will be determined by IHC. The number of CSCs, apoptotic and senescent cells in tumor tissues will be determined using the same methods as described above. The same approaches will be employed to examine the efficacy of ATS 11 to sensitize SUM 159 human breast cancer cells derived xenograft tumors to chemotherapy.

[0085] Expected results and statistical considerations. Based on the inventors' preliminary studies showing that ATS 11 markedly sensitizes TNBC cells to CDDP treatment (Fig. 6), the inventors predict that ATS 11 treatment will significantly enhances the anticancer efficacy of CDDP in the above xenograft animal models. It is also predicted that tumors from ATS11 treated mice will show decreased expression of Twist 1 and TIMP1 along with reduced number of CSCs but increased number of senescent and/or apoptotic cells in tumor tissues. A total of 80 NOD/SCID mice will be used (10 mice/group x 4 groups x 2 cell lines = 80). Statistical analysis and sample size considerations are the same as those proposed in earlier Examples.

Example 12

[0086] ATS-125 is an analog of camptothecin having the structure shown in Figure 7. It exhibits efficacy and specificity to targeting Twistl signaling (Fig. 8). ATS-125 also shows efficacy in depleting CD44+/CD24-/ESA+ breast cancer stem cell subpopulation in culture dishes in vitro (Fig. 9). Figure 10 demonstrates the effectiveness of ATS-125 in suppressing the self-renewal capacity of breast cancer stem cells (BCSCs) in vitro. The number of tumor spheres (BCSC readout) decrease with ATS-125 doses. As shown in Fig. 11A, ATS-125 has a dramatic tumor suppression effect in human breast cancer xenografts in vivo. Mice treated with ATS-125 resulted in reduced tumor weight and tolerance (Figs. 1 IB and 11C).

[0087] ATS-125 is efficacious against Twistl in tumor tissues of human breast cancer xenografts in vivo and is effective at eliminating breast CSCs in xenograft human breast cancers in vivo. (Figs 12A and 12C) Further, as shown in Fig. 13, Taxol treatment alone is unable to clear CSCs, but a combinatorial therapy with ATS-125 and Taxol can effectively eradicate CSCs. [0088] Thus, it is expected that ATS-125 is a potent Twistl inhibitor and would be useful in killing CSCs either alone or in combination with radiation or other chemotherapeutic agents.

[0089] Aspects of the invention are further described in the following numbered paragraphs:

1. An in vitro method for predicting responsiveness of a cancer patient to a therapy, comprising the steps of:

- detecting the relative amount of one or several Twistl protein expressing cell types in a sample obtained from said cancer patient prior to treatment and using said relative amount, or percentage (%), of cells expressing Twistl protein as a biomarker for predicting said patient's response to treatment with a pharmaceutical composition comprising a Twistl inhibitor; and

2. The method according to paragraph 1, comprising the steps of:

a) taking a sample from the cancer patient;

b) detecting said relative amount of Twistl protein in the sample obtained in a);

c) comparing the relative amount of Twistl protein obtained in b) to standard values from a patient with the same cancer; and

d) administering a therapeutically effective amount of said Twistl inhibitor to said patient in need thereof if the amount detected in b) is above said standard values.

3. The method according to paragraph 1, wherein the cancer is breast cancer, prostate cancer, cervical cancer, ovarian cancer, gastric cancer, colorectal cancer, pancreatic cancer, liver cancer, brain cancer, neuroendocrine cancer, lung cancer, kidney cancer, hematological malignancies, melanoma or sarcomas.

4. The method according to paragraph 3, wherein said cancer is breast cancer.

5. The method according to paragraph 1, wherein said Twistl inhibitor is camptothecin or an analog or derivative thereof. 6. The method according to paragraph 1 , wherein said Twistl inhibitor is ATS-11.

7. The method according to paragraph 1 , wherein said Twistl inhibitor is ATS-125 having the structure:

8. A method of treating cancer in a patient in need thereof, comprising the steps of detecting the relative amount of one or several Twistl protein expressing cell types in a sample obtained from said cancer patient prior to treatment, and administering a therapeutically effective amount of a Twistl inhibitor to said patient in need thereof.

9. The method according to paragraph 8, wherein said Twistl inhibitor is ATS-11.

10. The method according to paragraph 8, wherein said Twistl inhibitor is ATS-125 having the structure:

11. A method of killing cancer stem cells in a patient in need thereof, comprising the steps of administering a therapeutically effective amount of camptothecin or an analog or derivative thereof to said patient.

12. The method according to paragraph 11 , wherein said Twistl inhibitor is ATS-1 1.

13. The method according to paragraph 11, wherein said Twistl inhibitor is ATS-125 having the structure:

14. A method of killing cancer stem cells in a patient in need thereof, comprising the steps of administering a therapeutically effective amount of camptothecin, or an analog or derivative thereof, to said patient in combination with radiation therapy or at least one other chemotherapeutic agent.

15. The method according to paragraph 14, wherein said Twistl inhibitor is ATS-1 1.

16. The method according to paragraph 14, wherein said Twistl inhibitor is ATS-125 having the structure:

* * *

It is to be understood that the invention is not limited to the particular embodiments of the invention described above, as variations of the particular embodiments may be made and still fall within the scope of the appended claims.

Claims

Claims:

2. The method according to claim 1, comprising the steps of:

a) taking a sample from the cancer patient;

3. The method according to claim 1, wherein the cancer is breast cancer, prostate cancer, cervical cancer, ovarian cancer, gastric cancer, colorectal cancer, pancreatic cancer, liver cancer, brain cancer, neuroendocrine cancer, lung cancer, kidney cancer, hematological malignancies, melanoma or sarcomas.

4. The method according to claim 3, wherein said cancer is breast cancer.

5. The method according to claim 1, wherein said Twistl inhibitor is camptothecin or an analog or derivative thereof.

6. The method according to claim 1 , wherein said Twistl inhibitor is ATS-11.

7. The method according to claim 1, wherein said Twistl inhibitor is ATS-125 having the structure:

9. The method according to claim 8, wherein said Twistl inhibitor is ATS-11.

10. The method according to claim 8, wherein said Twistl inhibitor is ATS-125 having the structure:

12. The method according to claim 11 , wherein said Twistl inhibitor is ATS-11.

13. The method according to claim 11, wherein said Twistl inhibitor is ATS-125 having the structure:

15. The method according to claim 14, wherein said Twistl inhibitor is ATS-11.

16. The method according to claim 14, wherein said Twistl inhibitor is ATS-125 having the structure:

17. A kit for predicting the response to treatment with a pharmaceutical composition comprising a compound which acts as inhibitor to Twist 1, comprising:

a) means for obtaining a sample from a patient suffering from cancer;

b) instructions how to detect the relative amount of one or several Twist 1 protein expressing cell types from said sample;

d) a pharmaceutical composition comprising a therapeutically effective amount of a Twist 1 inhibitor.