Breast Cancer Therapeutics Brustkrebstherapeutika Produits Thérapeutiques Contre Le Cancer Du Sein - DocsLib

sign in sign up

<<

Home , Tanaproget

(19) TZZ _¥¥_T

(11) EP 2 618 833 B1

(12) EUROPEAN PATENT SPECIFICATION

(45) Date of publication and mention (51) Int Cl.: of the grant of the patent: A61K 31/565 (2006.01) A61K 39/00 (2006.01) 02.05.2018 Bulletin 2018/18 A61K 39/395 (2006.01) C07K 16/28 (2006.01) A61P 35/00 (2006.01) A61K 31/57 (2006.01) (21) Application number: 11758242.9 (86) International application number: (22) Date of filing: 22.09.2011 PCT/EP2011/066511

(87) International publication number: WO 2012/038504 (29.03.2012 Gazette 2012/13)

(54) BREAST CANCER THERAPEUTICS BRUSTKREBSTHERAPEUTIKA PRODUITS THÉRAPEUTIQUES CONTRE LE CANCER DU SEIN

(84) Designated Contracting States: • ELLIS GEORGIANA K ET AL: "Effect of AL AT BE BG CH CY CZ DE DK EE ES FI FR GB denosumab on bone mineral density in women GR HR HU IE IS IT LI LT LU LV MC MK MT NL NO receiving adjuvant aromatase inhibitors for PL PT RO RS SE SI SK SM TR non-metastatic breast cancer: subgroup analyses of a phase 3 study.", BREAST CANCER (30) Priority: 22.09.2010 EP 10178346 RESEARCH AND TREATMENT NOV 2009 LNKD- PUBMED:19308727, vol. 118, no. 1, November (43) Date of publication of application: 2009 (2009-11), pages 81-87, XP019746615, ISSN: 31.07.2013 Bulletin 2013/31 1573-7217 • TOMETSKO M ET AL: "Efficacy of a RANKL (73) Proprietor: IMBA-Institut für Molekulare inhibitor, OPG-Fc, relative to zoledronic acid to Biotechnologie GmbH inhibit bone metastasis of a RANK-expressing 1030 Wien (AT) human breast cancer cell", BONE, PERGAMON PRESS., OXFORD, GB, vol. 47, no. Suppl.2, 1 July (72) Inventors: 2010 (2010-07-01), pages S325-S326, • PENNINGER, Josef XP009143866, ISSN: 8756-3282 1130 Wien (AT) • CANONJ ET AL: "Prevention of tumor growth and • SCHRAMEK, Daniel tumor-inducedosteolysis by the RANKL inhibitor 1190 Wien (AT) osteoprotegerin (OPG) is associated with significant improvement in survival using a (74) Representative: Sonn & Partner Patentanwälte mouse model of breast cancer bone metastasis", Riemergasse 14 BONE, PERGAMON PRESS., OXFORD, GB, vol. 1010 Wien (AT) 40, no. 6, Suppl. 2, 1 June 2007 (2007-06-01), page S145, XP009143894, ISSN: 8756-3282 (56) References cited: WO-A1-99/53942 WO-A2-02/098362

Note: Within nine months of the publication of the mention of the grant of the European patent in the European Patent Bulletin, any person may give notice to the European Patent Office of opposition to that patent, in accordance with the Implementing Regulations. Notice of opposition shall not be deemed to have been filed until the opposition fee has been paid. (Art. 99(1) European Patent Convention). EP 2 618 833 B1

Printed by Jouve, 75001 PARIS (FR) (Cont. next page) EP 2 618 833 B1

• CANON J ET AL: "549 POSTER The RANKL • LEE RICHARD J ET AL: "Treatment and inhibitor osteoprotegerin (OPG) inhibits tumor prevention of bone complications from prostate growth, prevents tumor-induced osteolysis, and cancer", BONE, PERGAMON PRESS., OXFORD, significantly improves survival in a mouse model GB, vol. 48, no. 1, 1 June 2010 (2010-06-01), pages of breast cancer bone metastasis", EUROPEAN 88-95, XP009144092, ISSN: 8756-3282, DOI: JOURNAL OF CANCER. SUPPLEMENT, DOI:10.1016/J.BONE.2010.05.038 [retrieved on PERGAMON, OXFORD, GB, vol. 4, no. 12, 1 2010-06-01] November 2006 (2006-11-01), pages 166-167, • SCHRAMEK DANIEL ET AL: "Osteoclast XP025026924, ISSN: 1359-6349, DOI: differentiation factor RANKL controls DOI:10.1016/S1359-6349(06)70554-7 [retrieved development of progestin-driven mammary on 2006-11-01] cancer", NATURE, NATURE PUBLISHING • CANON J ET AL: "P13. The RANKL inhibitor GROUP, LONDON, GB, vol. 468, no. 7320, 1 OPG-Fc delays the de novo establishment of November 2010 (2010-11-01), page 98, breast cancer skeletal metastases in an XP009144000, ISSN: 0028-0836, DOI: MDA-MB-231 mouse model", CANCER DOI:10.1038/NATURE09387 TREATMENT REVIEWS, SAUNDERS, US, vol. 34, • LEIBBRANDTA ETAL: "ESCI award lecture:from 1 January 2008 (2008-01-01), page 17, a little mouse to rationale medicine for bone XP022694153, ISSN: 0305-7372, DOI: loss", EUROPEAN JOURNAL OF CLINICAL DOI:10.1016/J.CTRV.2008.03.054 [retrieved on INVESTIGATION, WILEY-BLACKWELL 2008-01-01] PUBLISHING LTD, GB, vol. 39, no. 10, 1 October • TOMETSKO M ET AL: "RANKL inhibition causes 2009 (2009-10-01), pages 842-850, XP009144002, a greater suppression of tumor-induced ISSN: 0014-2972, DOI: osteoclastogenesis than zoledronate treatment DOI:10.1111/J.1365-2362.2009.02215.X[retrieved in vivo and RANKL rescues osteoclasts from on 2009-09-15] zoledronate killing in vitro", JOURNAL OF BONE • WU XIAOJUN ET AL: "RANKL regulates Fas AND MINERAL RESEARCH, AMERICAN expression and Fas-mediated apoptosis in SOCIETYFOR BONEAND MINERAL RESEARCH, osteoclasts",JOURNAL OF BONE AND MINERAL NEW YORK, NY, US, vol. 21, no. Suppl. 1, 1 RESEARCH, AMERICAN SOCIETY FOR BONE September 2006 (2006-09-01), page S346, AND MINERAL RESEARCH, NEW YORK, NY, US, XP009143893, ISSN: 0884-0431 vol. 20, no. 1, 1 January 2005 (2005-01-01), pages • HOLLANDPAMELA M ET AL: "Combined therapy 107-116, XP009143980, ISSN: 0884-0431, DOI: with the RANKL inhibitor RANK-Fc and DOI:10.1359/JBMR.041022 [retrieved on rhApo2L/TRAIL/dulanermin reduces bone 2004-10-25] lesions and skeletal tumor burden in a model of • BHARTI A C ET AL: "Ranking the role of RANK breast cancer skeletal metastasis", CANCER ligand in apoptosis", APOPTOSIS ; AN BIOLOGY AND THERAPY, LANDES INTERNATIONAL JOURNAL ON PROGRAMMED BIOSCIENCE, US, vol. 9, no. 7, 1 April 2010 CELL DEATH, KLUWER ACADEMIC (2010-04-01), pages539-550, XP009143898, ISSN: PUBLISHERS, BO, vol. 9, no. 6, 1 November 2004 1538-4047 (2004-11-01), pages677-690, XP019204852, ISSN: • AKIYAMA T ET AL: "Systemic RANK-Fc protein 1573-675X, DOI: therapy is efficacious against primary DOI:10.1023/B:APPT.0000045780.10463.C6 osteosarcoma growth in a murine model via • MOLYNEUX SAM D ET AL: "Prkar1a is an activity against osteoclasts", JOURNAL OF osteosarcoma tumor suppressor that defines a PHARMACY AND PHARMACOLOGY 2010 molecular subclass in mice", JOURNAL OF PHARMACEUTICAL PRESS GBR LNKD- CLINICAL INVESTIGATION, AMERICAN DOI:10.1211/JPP/62.04.0009, vol. 62, no. 4, April SOCIETY FOR CLINICAL INVESTIGATION, US, 2010 (2010-04), pages 470-476, XP009143985, vol. 120, no. 9, 1 September 2010 (2010-09-01), ISSN: 0022-3573 pages3310-3325, XP009144001, ISSN: 0021-9738, • ZHANG JIAN ET AL: "Osteoprotegerin inhibits DOI: DOI:10.1172/JCI42391 prostatecancer-induced osteoclastogenesis and • PARK HYE-RIM ET AL: "Expression of prevents prostate tumor growth in the bone", osteoprotegerin and RANK ligand in breast JOURNAL OF CLINICAL INVESTIGATION, cancerbone metastasis", JOURNAL OF KOREAN AMERICAN SOCIETY FOR CLINICAL MEDICAL SCIENCE, KOREAN ACADEMY OF INVESTIGATION, US, vol. 107, no. 10, 1 May 2001 MEDICAL SCIENCE, SEOUL, KR, vol. 18, no. 4, 1 (2001-05-01), pages 1235-1244, XP009143975, August 2003 (2003-08-01) , pages 541-546, ISSN: 0021-9738 XP009143976, ISSN: 1011-8934

Remarks: Thefile contains technical information submitted after the application was filed and not included in this specification

2 1 EP 2 618 833 B1 2

Description mourigenesis. PARK et al., J. Korean Med. Sci., 18 (4) (2003): [0001] The present invention relates to the field of can- 541-546, contains a discussion of the role of OPG cer therapeutics. and RANKL in breast cancer bone metastasis. 5 ELLIS et al., Breast Cancer Research and Treat- [0002] It is a goal of the invention to provide alternative ment, 118 (1) (2009): 81-87, describe the adminis- anti-cancer therapies that do not suffer from this draw- tration of Denosumab to increase bone mineral den- back. sity in breast cancer patients during the treatment [0003] Receptor Activator of NF-κB Ligand (RANKL, with aromatase inhibitor as anti-cancer agent. 10 also known as ODF, TRANCE, OPGL, TNFSF11) and TOMETSKO et al., Bone, 47 (2010): 325-326, re- its receptor RANK (TRANCE-R, TNFRSF11A) are es- lates to the role of RANK/RANKL in bone metastas- sential for the development and activation of osteoclasts. es. RANKL is said to be a critical mediator of oste- RANKL inhibition was approved for potentially millions oclast activity, which in turn favours tumour growth. of patients to prevent bone loss. RANK and RANKL have Some breast and prostate cancer cell lines are dis- 15 been cloned and characterized (US 6,017,729, EP 0 873 closed to express RANK, thus having the ability to 998, EP 0 911 342, US 5,843,678, WO 98/46751, WO form bone metastases. 98/54201). Both RANKL and RANK expression have CANON et al., Bone, 40 (6) (2007):145, CANON et been observed in primary breast cancers in humans and al.,Europ. J.of Cancer,4 (12) (2006): 166-167, CAN- breast cancer cells lines and it has been proposed that ON et al., Cancer Treatment Reviews, 34 (2008): 20 the RANKL/RANK system can regulate bone metastases 60,and TOMETSKO et al.,J. Bone and MineralRes., of epithelial tumors 14 without an effect on proliferation or 21 (2006): 346 (Abstract) disclose the use of RANKL death susceptibility. inhibitor OPG to treat tumour growth of cancer bone [0004] US 2008/107597 and WO 2010/022120 A1 de- metastases. scribe anti-RANKL antibodies and their use for inhibiting HOLLAND et al., Cancer Biology and Therapy, 9 (7) 25 bone loss associated with a multitude of diseases or for (2010): 539-550, contains a comparison of adminis- the treatment of "RANKL-associated diseases". tering a RANKL inhibitor (RANK-Fc) alone or togeth- [0005] In WO 02/098362 the treatment of cancer in er combination with an apoptosis-inducing ligand non-hypercalcemic patients with a RANK antagonist is (rhApo2L) in the treatment of breast cancer skeletal proposed. However no therapy was developed and this metastases. 30 patent family was finally abandoned. AKIYAMA et al., J. Pharmacy and Pharmacology, [0006] Established uses of RANKL for treating diseas- 62 (4) (2010): 470-476, relates to the administration es associated with bone and calcium metabolism and of RANK-Fc in the treatment of osteosarcoma inflammatory conditions are described e.g. in WO ZHANG et al., J. Clin. Invest., 107 (10) (2001): 2007/128564. 1235-1244, relates to bone metastatic prostate can- 35 [0007] Breast cancer is one of the most common cancer cells in vitro. cers in humans and will on average affect up to one in LEE et al., Bone, 48 (1) (2010): 88-95 relates to bone eight women in their life time in the US and Europe. The metastasis of prostate cancer. This document sum- Women’s Health Initiative (WHI) and the Million Women marizes different pharmaceutical agents and studies Study have shown that hormone replacement therapy investigating the effects of metastatic prostate can- 40 (HRT) is associated with an increased risk of incident cer. and fatal breast cancer. In particular synthetic progester- LEIBBRANDT et al., Europ. J. of Clin. Invest., 39 one derivatives (progestins) such as medroxyprogester- (10) (2009): 842-850, is a review article on the func- one acetate (MPA), used in millions of women for HRT tions of RANK, RANKL and OPG on osteoblasts and and contraceptives, markedly increase the risk of devel- osteoclasts, and their involvement in bone metasta- 45 oping breast cancer. sis formation. [0008] Estrogene as well as progesterone including its WO 99/53942 discloses the use of OPG in the treat- synthetic derivatives (progestins) are used in combined ment of cardiovascular disease. hormone replacement therapies (HRT) in postmenopau- WU et al., J. Bone and Mineral Res., 20 (1) (2005): sal women to ameliorate menopausal symptoms. Estro- 107-116 discloses RANKL as regulator of Fas-me- 50 gene and progestins may also be used as hormonal con- diated apoptosis of osteoclasts. traceptives. BHARTI et al., Apoptosis, 9 (6) (2004): 677-690, re- [0009] Tamoxifen is an antagonist of the estrogen relates the role of RANKL in apoptosis of osteoclast ceptor in breast tissue. It is a standard endocrine (anti- precursors. estrogen) therapy for hormone-positive early breast can- MOLYNEUX et al., J. Clin. Invest., 120 (9) (2010): 55 cer in post-menopausal women. It would also inhibit hor- 3310-3325, relates to research on Prkar1a as bone mone replacement therapies by blocking the estrogen tumour suppressor gene, which is capable of induc- receptor. ing RANKL overexpression during osteosarcoma tu- [0010] It is a goal of the present application to provide

3 3 EP 2 618 833 B1 4 therapeutic methods and means for preventing or treat- ation. Importantly, RANKL/RANK inhibition results in a ing cancer. markedly reduced incidence and delayed onset of pro- [0011] The present invention relates to the specific role gestin-driven cancer and reduced self-renewal capacity of RANKL in cancer and methods and means to interfere of cancer stem cells. Although RANKL is known to influ- in the cancer associated RANKL mechanism for thera- 5 ence bone metastasis of tumors, the present invention peutic or prophylactic purposes. It was found that RANKL provides for the first time the possibility to treat cancer in is responsible for protecting cells from cancerogenous general and cancer independent of metastasis, which mutations as it prevents cell death after such mutations metastasis is not target of the inventive treatment. Pref- induced by DNA damage. Survival of cells despite trans- erably the cancer is a primary tumor. The cancer may be forming mutations is one key property of cancer cells. 10 non-metastatic. The newly discovered role of RANKL in this activity allows [0015] Inprior publications (WO 02/098362) it was sug- the inhibition of RANKL activity to treat and prevent can- gested to use RANK antagonists in cancer therapy. How- cer development and progression. Although RANKL in- ever no therapy was finally developed based on this sug- volvement was known during bone metastasis formation gestion. According to the present invention, particular due to effects on osteoclast and osteoblasts, it was not 15 preferred targets are cells dependent on hormone sig- known that RANKL can have a direct effect on cancer nalling for growth. A significant factor in RANK/RANKL cell formation. Disclosed is cancer therapy using a RAN- driven cancer development is the hormonal background. KL inhibitor as active agent, especially in the prevention According to the invention RANK/RANKL’s canceroge- or treatment of primary tumours apart from metastatic nous effect has been associated with sexual hormones, cancer cells. A cancer therapy is to be understood as the 20 in particular progesterone and its derivatives (pro- reduction of cancer cells in the body of a patient or at gestins), that are widely administered to females in hor- least the prevention of further progression of the disease. mone replacement therapies or as contraceptives. RANKL inhibitors include antibodies widely known e.g. Therefore a RANK/RANKL antagonizing therapy is for from US 2008/107597, WO 2010/022120 A1, US treating patients with elevated sexual hormones or of 6,740,522, US 7,411,050, EP 2003203 A1, or polypep- 25 their functional derivatives. One cancer treated or pre- tide inhibitors disclosed in US 2004/167072. RANKL in- vented according to the present invention is primary hibitors being anti-RANKL antibodies are used according breast cancer. to the present invention. [0016] In preferred embodiments the cancer is a can- [0012] RANKL is a known ligand of cell surface recep- cer dependent on hormones for growth. Such hormones tor RANK that regulates function of dendritic cells and 30 may be sexual hormones, such as female sexual hor- osteclasts. According to the present invention, a further mones like progesterone or estrogene. The cancer cells mechanism in the development of cancer has been dis- may have hormone receptors, especially progesterone covered. RANKL drives hormone-influenced cancer de- receptors and/or estrogene receptors. Examples of es- velopment. Such hormones may be of the normal hor- trogene receptors are ESR1 (comprising ER alpha monal background in any individual or may have been 35 chains), ESR2 (comprising ER-beta chains) or hetero- artificiallyadministered (such as in hormone replacement meric receptors, such as of mixed ER-alpha and ER-beta therapies, in menopause treatment or as contraceptive). chains. Presence of such receptors may indicate a re- Furthermore, the cellular mechanism and activity of this quirement of hormone signalling in the cell. In particular, ligand in cancer has been investigated and character- this signalling may be RANKL-mediated. Preferably the ized. 40 hormone signalling is mediated by the progesterone re- [0013] The effect of RANKL, "RANKL activity", in- ceptor. In special embodiments the cancer comprises cludes binding of RANKL to RANK and its resulting ac- cancer cells with a progesterone receptor but without tivation. RANK in turn further activates IKKα, IκBα, P- functional estrogene receptor. NFκB and cyclinD1 as well as the Id2-p21, MAPK Erk [0017] Activation of RANKL by hormones protects the and p38 pathway. Modifying activity of any of these fac- 45 cancer or a pre-cancerous cell from DNA damage in- tors can be used for a therapeutic or prophylactic method duced cell death. Thus these hormones may support can- or to reduce cancer cell viability. Most of these proteins cer via increased RANKL activity. Preferred types of can- are intracellular and it is possible to inhibit their function cerdiagnosed according to thepresent invention arecan- by intracellular inhibition of their activity or expression cers with sexual hormone dependency during develop- such as by RNAi. RANKL and RANK are extracellular 50 ment, especially breast cancer or prostate cancer. Fur- targets but may also be targeted intracellular (e.g. by ther types of cancer include Hodgkin’s lymphoma; non- RNAi). Hodgkin’s lymphoma; B-cell acute lymphoblastic leuke- [0014] According to the invention it was found that in mia/lymphoma; T-cell acute lymphoblastic leuke- vivo administration of progestins such as MPA triggers mia/lymphoma; peripheral T-cell leukemia, adult T-cell massive induction of the key osteoclast differentiation 55 leukemia/T-cell lymphoma; NK cell tumor; large granular factor Receptor Activator of NF-B Ligand (RANKL) in lymphocytic leukemia; Langerhans cell histiocytosis; my- cells. Inhibition or inactivation of RANKL or its receptor eloidneoplasia; acute myelogenous leukemia;acute pro- RANK in these cells prevents progestin-induced prolifer- myelocytic leukemia; acute myelomonocytic leukemia;

4 5 EP 2 618 833 B1 6 acute monocytic leukemia; a myelodysplastic syndrome; tions, in particular of sexual hormones such as proges- and a chronic myeloproliferative disorder. In particular terone (or the synthetic functional analogons, proges- preferred the cancer is selected from lung cancer, breast tines) or estrogenes. The administration of female sexual cancer, mammary cancer, melanoma, sarcoma, prostate hormones to a patient, be it for a hormone replacement cancer, head and neck cancer, cancer of unknown pri- 5 therapy or as hormone contraceptive (particular proges- mary origin, lymphoma, leukemia, kidney cancer, and terone and its derivatives) increases the risk of cancer gastrointestinal cancer. Preferably the cancer is a hor- via the RANKL pathway, in particular of hormone driven mone driven-cancer like breast or prostate cancer. Such cancers such as breast cancer or prostate cancer. Fur- hormones may be female sexual hormones like estrogen ther, any deregulation of the endogenous progesterone or progesterone, or artificial functional equivalents there- 10 system such as in pre-menopause may increase the risk of like progestins or other factors known to induce RANKL of cancer. In turn, administrating a RANKL inhibitor can such as PTHrP (Parathyroid hormone-related protein) or decrease the risk of breast cancer or be used to treat a Vitamin D. The cancer may be of epithelial origin. Pref- breast cancer. Mechanistically, RANKL activity protects erably the cancer is treated or prevented in its origin tis- breast cancer cells from apoptosis and increases breast sue, such as in the breast in the case of breast or mam- 15 cancer survival in view of cancerogenic mutations. Re- mary cancer or in the prostate for prostate cancer. ducing RANKL activity therefore prevents this protective [0018] According to the invention the cancer is a pri- effect and results in increased cancer cell death. mary cancer. As shown herein a primary cancerous de- [0023] The patient might have or might have had a ther- velopment can be prevented or delayed by inhibiting apy with sexual hormones. It was found according to the RANKL activity. Therefore the primary cancer can be 20 present invention that RANKL activity in cancer develop- treated or prevented. In addition it is also possible to pre- ment is influenced by sexual hormones, in particular by vent or treat the development of re-occuring estrogenes or progesteron or its dervatives (progestins). [0019] According to one embodiment the invention Therefore, in preferred embodiments the patient is treat- provides a RANKL inhibitor being an anti-RANKL anti- ed by a hormone, preferably receives hormone replace- body for use in the treatment or prevention of primary 25 ment therapy, preferably with estrogene, progesterone breast or mammary cancer in a patient, wherein the pa- or a progestin, or with a hormone contraceptive. Exam- tient has received a hormone treatment, preferably a hor- ples of progestins are medroxyprogesterone (or its ace- mone replacement therapy, preferably with estrogene, tate, e.g. the 17-acetate), norethisterone, ethisterone, progesterone or a progestin, or a hormone contraceptive. norethynodrel, norethindrone acetate, ethynodiol diace- [0020] "Treatment" shall be construed as a beneficial 30 tate, levonorgestrel, norgestrel, desogestrel, gestodene, effect on cancer patients, in particular as a reduction of norgestimate, drospirenone, dienogest, nesterone, no- cancer cells, including preventing further progression of megestrol acetate, trimegestrone, tanaproget, megestrol cancer, but not necessarily in an absolute curative sense, acetate, pranone, etonogestrel. The hormone or deriva- which is of course possible but not necessarily required tive preferably has progestinic effects. In particular the by the invention. 35 hormone may induce RANKL. [0021] Similar, "preventing" shall not be construed as [0024] In hormone replacement therapies or by using an absolute success to always prevent the onset of can- hormone contraceptives, the hormone level, in general cer, but as a relative reduction of the risk of developing of sexual hormones, is upregulated leading to increased cancer in a patient or of delaying onset of cancer, i.e. as progesterone levels that could be tied to development of a prophylactic treatment. The prevention of cancer is a 40 cancer via the RANKL pathway according to the present particular advantage of the present invention. Since invention. Therefore, patients which receive or have been RANKL-induced protection against apoptosis is a funda- treated by a hormone or hormone contraceptive are at mental step in the development of cancer, it is now pos- an increased risk of developing breast cancer. For these sible to inhibit this step in cancerous developments and patients, administering a RANKL inhibitor is particularly prevent for a certain time or delay cancer manifestation. 45 effective for treating or preventing cancer, cancer devel- The treatment or prevention according to the present in- opment or cancer progression. vention can be used to treat benign tumors or nodules [0025] Inparticular high progestinand/or progesterone and thus inhibit further development in cancer formation. levels that may be found in a patient to be treated are [0022] The normal hormonal levels of a patient may e.g. serum concentration or at least 0.2 ng/ml, preferably usually be sufficient to trigger the RANKL pathway that 50 at least 0.3 ng/ml. has been involved with cancer development according [0026] A "RANKL inhibitor" is to be understood as any to the present invention. At certain conditions hormone compound or agent, that reduces RANKL activity. It in- level may be increased in a patient, be it for natural or cludes any RANKL ligand, in particular anti-RANKL-an- artificial causes. Such increased hormone level may in- tibodies, that inactivates free RANKL and prevents com- crease the RANKL associated cancer development and 55 petitively its binding to and activation of RANK. Further progression and therefore in a preferred embodiment the RANKL inhibitors include RANK antagonists that block patient to be treated according to the present invention RANK, RANKLs cellular receptor. Furthermore, any fac- has increased hormone levels, e.g. blood concentra- tor in RANKLs signalling pathway, including RANKL it-

5 7 EP 2 618 833 B1 8 self, RANK, IKKα, IκBα, P-NFκB, cyclinD1 (NFκB path- include altered siRNAthat is a hybridDNA/RNA construct way inhibition), Id2, MAPK Erk and p38 can be antago- or any equivalent thereof, double-stranded RNA, micro- nized or p21 increased or agonized (Id2 pathway inhibi- RNA (miRNA), as well as siRNA forms such as siRNA tion) to prevent RANKL signalling and reduce RANKL duplications, small hairpin RNA (shRNA) in viral and non- activity. The Id2 pathway functions as a transcriptional 5 viral vectors and siRNA or shRNA in carriers. RANKL repressor of p21, on of the major cyclin-dependent kinase RNAi is e.g. described in WO 2005/028633. inhibitors, which inhibits cell cycle progression. There- [0029] There exists several methods in the art for in- fore, the Id2 pathway has to be inhibited for p21 to be hibiting gene expression using RNAi such as described upregulated hence stopping proliferation. for example in WO 02/055692, WO 02/055693, EP 1 144 [0027] A RANKL inhibitor can inhibit or antagonize any 10 623 B1 and WO 03/074654. By using a siRNA therapy one of RANKL itself, RANK, α IKK, κ I Bα, P-NFκB, any cellular factor can be targeted and inhibited for a cyclinD1, Id2, MAPK Erk or p38 or increase or agonize RANKL antagonizing and inhibiting therapy. Therefore, p21 or any combination thereof, preferably of at least 2, any such compound can be used as a RANKL inhibitor. 3, 4, 5, or all of these factors, in particular preferred with [0030] The RANKL inhibitor of the present invention is RANKL. A RANKL inhibitor may be any compound that 15 an anti-RANKL antibody, such as Denosumab, or as dis- lowers RANKL concentration, in particular serum con- closed in US 2008/107597. "Anti-RANKL-antibody" in- centrations, or lower RANKL expression. A RANKL in- cludes derivatives therof, including antibody fragments hibitor may further lower RANK concentration on cell sur- such as Fab, F(ab)2, Fv, or single chain antibodies face, in particular cancer cell surface, and/or lower RANK (scAb). expression and/or lower RANK activation in response to 20 The antibodies may be produced by immunization with RANKL binding. A RANKL inhibitor may further lower in- full-length protein, soluble forms of the protein, or a frag- tracellular concentrations and/or expression and/or acti- ment thereof. The antibodies of the invention may be vation of IKKα. A RANKL inhibitor may further lower in- polyclonal or monoclonal, or may be recombinant anti- tracellular concentrations and/or expression and/or acti- bodies, such as chimeric antibodies wherein the murine vation of IκBα. A RANKL inhibitor may further lower in- 25 constant regions on light and heavy chains are replaced tracellular concentrations and/or expression and/or acti- by human sequences, or CDR-grafted antibodies where- vation of P-NFκB. A RANKL inhibitor may further lower in only the complementary determining regions are of intracellular concentrations and/or expression and/or ac- murine origin. Antibodies of the invention may also be tivation of cyclinD1. A RANKL inhibitor may further lower human antibodies prepared, for example, by immuniza- intracellular concentrations and/or expression and/or ac- 30 tion of transgenic animals capable of producing human tivation of Id2. A RANKL inhibitor may further increase antibodies (WO 93/12227). The antibodies are useful for intracellular concentrations and/or expression and/or ac- detecting RANKL in biological samples, thereby allowing tivation of p21. A RANKL inhibitor may further lower in- the identification of cells or tissues which produce the tracellular concentrations and/or expression and/or acti- protein in addition, antibodies which bind to RANKL (and vation of MAPK Erk. A RANKL inhibitor may further lower 35 block interaction with other binding compounds) have intracellular concentrations and/or expression and/or ac- therapeutic use as RANKL inhibitor. tivation of p38. [0031] RANKL inhibitors may be suitable and used for [0028] For antagonizing cellular factors preferably siR- reducing the free RANKL concentration in the circulation NA molecules are administered to reduce the expression of a patient. RANK is not the only natural receptor for and function of these factors. RNA interference (RNAi) 40 RANKL. Osteoprotegin (OPG) is a secreted decoy re- is a mechanism to suppress gene expression in a se- ceptor that can reduce the RANKL activity (binding of quence specific manner. RNA interference (RNAi) is RANKL to RANK and its signaling pathway via IKK α, Iκ highly effective methodology for suppression of specific Bα, P-NFκB and cyclinD1 or via Id2-p21, MAPK Erk and gene function in eukaryotic cells. When applied to cells p38). and organisms, RNAi entails the degradation of target 45 [0032] The inhibitor is usually administered in a thera- mRNA upontransfection of shortinterfering RNA (siRNA) peutically effective amount, an amount that reduces oligos or short-hairpin RNA (shRNA) encoding vectors. RANKL activity to significantly decrease cancer cell via- Various methods of RNAi have been described and are bility. In particular embodiments the RANKL activity may generally known for the altering gene expression in plant be suppressed to normal levels of an average person, cells, drosophila and human melanoma cells as is de- 50 furthermore, the RANKL activity may be reduced below scribed for example in US 2002/0162126 and US average levels. Preferably the RANKL activity is reduced 2002/0173478. The siRNA for use in methods and com- by at least 25%, by at least 30%, by at least 40%, by at positions least 50%, by at least 60%, by at least 70% by at least are selected to target a desired molecule of the RANKL 80% or by at least 90%. In preferred embodiments this signaling pathway or combinations of such molecules. In 55 reduction equates to RANKL serum levels. this manner they are targeted to various RNAs corre- [0033] The inhibitor may be used in combination with sponding to a target gene. It is understood by one of skill or for priming a further anti-cancer therapy, preferably a in the art that the siRNA as herein described may also radiation- or chemotherapy. According to this aspect of

6 9 EP 2 618 833 B1 10 the present invention a method for increasing the efficacy cially for prophylactic methods) or a cancer cell. Further- of cancer therapy in a subject is provided, comprising more said cell can be an epithelial cell or a cancer stem administering to a subject in need of an effective amount cell. Especially cancer of epithelial origin can be effec- of an RANKL inhibitor, wherein said subject is also being tively treated by the inventive methods. In particular pre- administered a cancer therapy selected from the group 5 ferred embodiments the cancer is of mammary gland ep- consisting of small-molecule drugs, angiogenesis inhib- ithelial cells. Furthermore RANKL inhibition as a signifi- itors, tumor vaccine, chemotherapy, immunotherapy, ra- cant effect on cancer stem cells and can prevent or re- diation therapy, gene therapy and combinations thereof. duce cancer comprising such stem cells. [0034] RANKL acts as a protective agent, immunizing [0037] According to preferred therapeutic or preven- cancer cells against DNA damage associated cell death. 10 tive methods characterized in that the patient has re- By reducing RANKL activity and therefore reducing RAN- ceived a hormone treatment, preferably a hormone re- KLs protective effect on cancer, the cells become more placement therapy, preferably with progesterone or a susceptible to DNA damage. Therefore, in a preferred progestin, or a hormone contraceptive, such as de- embodiment a cancer is treated in addition with a RANKL scribed above. inhibitor and a further anti-cancer therapy that induces 15 [0038] As mentioned above, hormone treatment, in further cell damage such as radiation or chemotherapies. particularwith femalesexual hormones increasesthe risk Preferably such further cancer therapies are specific for of breast cancer and therefore, in a inventive therapeu- cancer in a certain tissue, like breast, e.g. by administer- tical preventive method preferably the patient has been ing localised a chemotherapeutic agent or a homing treated with such a hormone therapy previously or during agent that targets the tissue cancer cells and induces 20 the inventive therapeutic method. Previous administra- further DNA damage either by radiation or a chemother- tions of hormones may have been in the past 1, 2, 3, 4 apeutic agent. or 5 years or more. [0035] According to this embodiment the present in- [0039] In a further aspect the present invention pro- vention relates also to a kit comprising a chemothera- vides a pharmaceutical composition comprising a RAN- peutic agent and a RANKL inhibitor being an anti-RANKL 25 KL inhibitor being an anti-RANKL antibody and a hor- antibody for use in a treatment or prevention a primary mone or derivative thereof, selected from progesterone breast or mammary cancer. According to a further em- or a progestin, or a hormone contraceptive. bodiment, said RANKL inhibitor is provided for use in [0040] As mentioned, RANKL inhibition can reduce the combination with a hormone replacement therapy or for risk of developing cancer by female sexual hormones, in use in combination with a hormone contraceptive, in par- 30 particular of estrogene, progesterone or progestins. ticular to reduce the risk of developing breast cancer. Therefore, it is advised to administer such hormones to- [0036] The patientis preferably a mammal,in particular gether with the RANKL inhibitor provided by the present preferred a primate, even more preferred a human, in invention. Such hormones might be used for hormone particular a female. The patient might have or might have replacement therapies or as hormonal contraceptives. had a therapy with hormones, in particular female sexual 35 [0041] Most common types of hormonal contraceptive hormones. Such a past therapy might have been in the formulations include both an estrogen and a progestin, past 5, 4, 3, 2, years or past 36, 32, 30, 28, 26, 24, 22, or are progestogen-only formulations which contain only 20, 18, 16, 14, 12, 10, 8, 7, 6, 5, 4, 3, 2, or 1, month(s). progesterone or one of its synthetic analogues (pro- Patients, in particular females, receiving a hormone re- gestins). In order to reduce the risk of hormone depend- placement therapy or are taking a hormone contracep- 40 ent cancer development, the present invention provides tive, in particular a progesterone or a progestine are at a combined use of hormone therapeutics, including fe- an increased risk to develop cancer. To reduce this in- male sexual hormone contraceptives, together with a creased risk, the present invention provides the com- RANKL inhibitor. bined use of a RANKL inhibitor as described above. Also [0042] Pharmaceutical compositions or formulations disclosed is a method of reducing RANKL activity in a 45 for therapeutic or prophylactic use may comprise a phar- cancer patient comprising administering a RANKL inhib- maceutically acceptable diluent, carrier, solubilizer, itor to said patient. Also disclosed is a method of reducing emulsifier, preservative and/or adjuvant. The invention the amount of cancer cells in a patient comprising ad- also provides for pharmaceutical compositions compris- ministering a RANKL inhibitor to said patient. Also dis- ing a therapeutically effective amount of an RANKL in- closed is a method of increasing cell death due to DNA 50 hibitor. The term "therapeutically effective amount" damage of a cell comprising inhibiting RANK or RANKL means an amount which provides a therapeutic effect for or IKK-alpha or IkB-alpha, P-NF-kappa-B or CyclinD1 or a specified condition and route of administration. The Id2 or MAPK Erk or p38 or increasing p21 or any com- composition may be in a liquid or lyophilized form and bination thereof, in particular a combination with inhibiting comprises a diluent (Tris, acetate or phosphate buffers) RANKL, in said cell, preferably by administering a RAN- 55 having various pH values and ionic strengths, solubilizer KL inhibitor to said cell. The cell receiving the inventive such as Tween or Polysorbate, carriers such as human treatment may be a tissue cell wherein hormone depend- serum albumin or gelatin, preservatives such as thime- ent cancers may develop, especially the breast (espe- rosalor benzylalcohol, and antioxidantssuch as ascrobic

7 11 EP 2 618 833 B1 12 acid or sodium metabisulfite. Selection of a particular sham treatment and MPA implantation. composition will depend upon a number of factors, in- Proliferation was determined byin situ Ki67 cluding the condition being treated, the route of admin- immunostaining. e,f, Marked increase of the istration and the pharmacokinetic parameters desired. stem cell-enriched CD24+CD49high population siRNA formulations are preferably administered in lipo- 5 (MaSC) in MPA-treated mammary glands in some formulations. control but not in RANK Δmam mammary glands. [0043] The present invention also relates to a kit com- e, Representative FACS profiles showing CD24 prising a hormone therapeutic, as described above, and and CD49 expression of lineage negative a RANKL inhibitor being an anti-RANKL antibody. Such (CD31- (endothelial cells) CD45 - (hematopoietic a kit may allow separate administration of the hormone 10 cells) TER199- (erythroid cells)) of mammary therapeutic and the RANKL inhibitor. MaSCs from MPA- or sham-treated 8-week old [0044] Also encompassed are compositions compris- virgin females, f, Quantification of MaSC- ing RANKL inhibitors modified with water soluble poly- enriched CD24+CD49high population from mers to increase solubility, stability, plasma half-life and mammary glands of MPA- or sham-treated bioavailability. Compositions may also comprise incor- 15 virginRANK Δmamand littermate controlfemales. poration of RANKL inhibitors into liposomes, microemul- For all MaSCs experiments mice were treated sions, micelles or vesicles for controlled delivery over an with MPA for 3 days. n=4 per group +/- sem. * extended period of time. Soluble RANKL inhibitors may P < 0.05; *** P < 0.001 (Student’s t-test). be formulated into microparticles suitable for pulmonary administration. 20 Figure 2. RANK controls the incidence and onset of [0045] Compositions of the invention may be adminis- progestin-driven mammary cancer. tered by injection, either subcutaneous, intravenous or intramuscular, or by oral, nasal, pulmonary or rectal ad- a, MPA/DMBA carcinogenesis scheme. ministration. The route of administration eventually cho- Nulliparous, six-week old female mice were s.c. sen will depend upon a number of factors and may be 25 implanted with MPA pellets and treated orally ascertained by one skilled in the art. with DMBA as indicated for 8 week. b, Onset of [0046] The present invention is further illustrated by palpable mammary tumors in MMTV-Cre the following figures and examples, without being limited rankfloxed/Δ females (RANKΔmam) (n=14) and to such specific examples. age-matched littermate control females (n=19) 30 treated with MPA pellets and DMBA as indicated Figures: in Fig. 2a. Data are shown as percentage of tumor free mice after the last DMBA challenge. [0047] Median tumor onset for controls was 11 days after last DMBA treatment and 30 days for Figure 1. The progesterone-derivative MPA triggers 35 RANKΔmam females. c, Representative in vivo RANKL expression and mammary epithelial histological sections of mammary tumors cell proliferation via RANK. isolated from control littermate and RANKΔmam females 22 days after the last DMBA treatment. a,b, Induction of RANKL expression by the Cytokeratin5 staining is shown. Magnifications progesterone-derivative MPA. Nulliparous wild 40 X 20 d,e, Numbers of carcinomas in situ and type females were s.c. implanted with slow- invasive mammary cancers in control and release MPA pellets or treated with surgery RANKΔmam females on day 7 ( d) and day 22 ( e) sham. a, RANKL mRNA was determined in after the final DMBA treatment. Data are shown purified mammary epithelial cells by qRT-PCR as mean values per mouse +/- sem. n=3 mice three days after implantation. Data are shown 45 per genotype. All 10 mammary glands were as fold change compared to sham treatment (+/- analysed for each mouse. * P < 0.05 (Student’s sem) (n=3). b, In situ Ki67 immunostaining of t-test). Bottom panels show representative progesterone receptor (PR, red) and RANKL histological sections with typical invasive (green) inmammary epithelial cells after 3d MPA adenocarcinomas in the control females. For treatment. c, Induction of soluble RANKL protein 50 RANKΔmam females, normal acinar morphology assayed on isolated mammary gland epithelial (day 7) and a carcinoma in situ (day 22) are cells from control, but not in prolactin receptor shown. H&E stained sections and mutant (PRL-R KO) females on day 3 after s.c. immunostaining for the proliferation marker Ki67 MPA implantation. Mammary glands from sham are shown. Magnifications X20. operated control littermates and PRL-R KO55 females are shown as controls.d , Epithelial Figure 3. RANK induces NFBκ signaling, anchor- proliferation in mammary glands of control age-independent growth, and protects from radia- littermates and RANK Δmam females 3 days after tion-induced epithelial apoptosis.

8 13 EP 2 618 833 B1 14

a, Western blotting for phosphorylated (P) IKK α, mined 7 days after replating. b, Quantification of pri- total IKKα, phosphorylated (P) p65 NFκB, total mary and secondary passage (F1 and F2) mammos- p65 NFκB, phosphorylated (P) IκBα, and total pheres from MPA/DMBA-treated RANKΔmam and IκBα in isolated primary mouse mammary gland control littermate females. Results shown are mean epithelial cells (MECs) in response to RANKL 5 values +/- sem. n = 3 mice per group. * P < 0.05 stimulation (1mg/ml). β-actin is shown as loading (Student’s t-test). control. b, Western blotting for IKK α, IKKβ, IKKγ, Figure 5. The progesterone-derivative MPA triggers phosphorylated (P) p65 NFκB, total p65 NFκB, in vivo RANKL expression. phosphorylated (P) κ IBα, and total κB I α in pooled late stage mammary adenocarconimas 10 a-d, Induction of RANKL expression by the pro- (n=4 for each lane) that developed in control, gesterone-derivative MPA. Nulliparous wild RANKΔmam, and IKKαΔmam females. β-actin is type females were s.c. implanted with slow-re- shown as loading controlc, Expression of lease MPA pellets or treated with surgery sham. RANK, CyclinDl, and p21 mRNA in late stage a, Expression of transmembrane (tm) and solu- mammary adenocacinomas that developed in 15 ble (s) RANKL protein was assayed on isolated control,RANK Δmam,and IKK αΔmamfemales. Ex- mammary gland epithelial cells by Western blot pression was determined by qRT-PCR. Data are 3 and 6 days after MPA pellet implantation or mean values +/- sem. n=4 per group.d , Soft- sham surgery. β-actin is shown as a loading con- Agar Colony Formation Assay. Growth of hu- trol. Recombinant murine sRANKL and recom- man SKBR3 breast cancer cells in soft agar in 20 binant human tmRANKL protein are shown for response to stimulation with RANKL (1 mg/ml) or molecular size comparison in the right panels. EGF (100ng/ml). Anchorage-independent, Data are representative of 8 animals tested, b, macroscopic colonies formed after 18 days in Schematic representation of the three RANKL culture with RANKL, which was prevented by isoforms adapted from Ikeda et al.. The trans- the decoy receptor OPG m (1g/ml). Controls 25 membrane domain of RANKL1 and RANKL2 is were unstimulated SKBR3 cells. e, Onset of pal- indicated (TM, black box, spanning aa 48-71). pable mammary tumors in IKK αΔmam (n=10) and RANKL2 is lacking aa 14-44 of the intracellular age matched littermate control (n=11) females domain while RANKL3 is lacking aa 1-118 span- treated with MPA pellets and DMBA. Data are ning the whole intracellular domain as well as shown as percentage of tumor free mice after 30 the transmembrane domain. Numbers indicate the last DMBA challenge. Median tumor onset amino acid residues. c, Expression of total (t) for controls was 10 days after last DMBA treat- RANKL, RANKL1, RANKL2, and RANKL3 mR- ment and 24 days for IKKαΔmam females. f,g γ- NA in purified mammary epithelial cells after irradiation (5 Gray) induced mammary epithelial three days MPA treatment. β-actin is shown as cell apoptosis in control and RANK Δmam female 35 a control. d, Representative agarose gel of RT- littermates either sham operated or implanted PCR products as in c. e, mRNA expression of with a MPA pellet. Apoptosis was detected by various proteases known to shed RANKL from immunostaining for active Caspase 3. f, Apop- the surface of cells. mRNA was determined in totic nuclei of epithelial cells (arrows) are shown purified mammary epithelial cells by qRT-PCR for representative mammary gland sections.40 three days after MPA implantation. Data are Magnifications X 40. g, Quantification of mam- shown as fold change compared to sham treat- mary epithelial apoptosis. A minimum of 5000 ment (+/- sem) (n=3 per group). * P< 0.05; ** P nuclei was counted for each mouse. Results < 0.005; *** P < 0.001 (Student’s t-test). shown are mean values +/- sem. n = 3 mice per group. * P < 0.05; ** P< 0.02 (Student’s t-test). 45 Figure 6. MPA does not trigger RANKL expression in human breast tumor cell lines. Figure 4. RANK controls cancer stem cell self renewal a, Self-renewal of mammary cancer stem cells, a, Analysis of RANKL mRNA expression in hu- i.e. tumor initiating cells (TICs) requires RANK ex- man breast cancer cells as well as cultured pression. Mammary tumor cells from MPA/DMBA- 50 MECs after MPA stimulation (0.1mM). Expres- treated RANKΔmam and control littermate female sion was determined by qRT-PCR in unstimu- were cultured for 7 days and a small percentage of lated cells or cells stimulated with MPA for the primary cells formed mammospheres (1 st passage). indicated time points. We never fond induction Primary mammospheres were then digested into or RANKL with the exception of SKBR3 cells; single cells and assayed for their ability to form sec- 55 however, even in these cells induction was only ondary mammospheres (2nd passage). Tumor cells slightly increased as compared to the more than from control littermate but not from RANKΔmam fe- 2000 time increase of RANKL we observed after males could form secondary mammospheres deter- in vivo MPA challenge. b-c, Progesterone re-

9 15 EP 2 618 833 B1 16

ceptor (PR) mRNA was determined in a panel hereafter. of human breast cancer cell lines ( b) and short- term cultured (passage < 3) mammary epithelial Figure 9. MPA induces RANKL expression and ep- cells (MECs) and freshly purified primary MECs ithelial proliferation in mammary glands. (c). Data were obtained by qRT-PCR and are 5 shown as fold change compared to MCF10A ( b) a-c, Quantification of epithelial proliferation in and purified primary MECs (c) (+/- sem) (n=3). mammary glands of control littermates and RANKΔmamfemales 3 days after sham treatment Figure 7. RANKf1/Δ females crossed to the K5-Cre and MPA implantation as shown in Fig. Id.a , mice show defective lobulo-alveolar development in 10 Data are shown as relative changes in total pregnancy. Ki67+ epithelial cells compared to sham-operated females of the respective genotype. At least a, Whole-mount analyses of mammary tissue of 1000 mammary gland epithelial cells were nulliparous and lactating (L1) RANK f1/A females counted per mouse. n = 3 mice per genotype. crossed with K5-Cre compared to mammary15 ** P < 0.005; *** P < 0.001 (Student’s t-test). b, glands of control littermates. Alveoli in gestating Quantification of in situ Ki67 immunostained wild-type females (arrow) from lobulo-alveolar cells per acinus from mammary glands of control structures, whereas this development is arrest- littermate and RANK Δmam females on day 3 after ed at a rudimentary alveolar bud (arrow) in K5- MPA s.c. implantation. Whereas in control fe- Cre RANKf1/Δ females. b, Southern blot of puri- 20 males ∼ 80% of acini showed signs of medium fied mammary epithelial cells derived from to high proliferation, more than 60% of acini in RANKflox/Δ and RANK-flox/Δ; K5-Cre+ females. RANKΔmam females exhibited very low prolifer- The wild type or floxed RANK allele (fl/+; 9.6 kb) ation rates. c, To control for estrus-dependent and the deleted RANK allele ( Δ;3.9 kb) are indi- background proliferation levels, superovulated cated after digestion of genomic DNA with PvuII 25 and sham operated control littermate and and SphI. c, Whole-mount analyses of mamma- RANKΔmam females were analysed. At least ry glands from control BALBc nude nu( /nu) fe- 1000 cells were counted per mammary gland. n males showing normal lobulo-alveolar struc- = 3 per genotype. In b, and c, data are shown tures at day 1 of lactation (L1), normal lobulo- as percentage of acini/ducts with low (<20% of alveolar structures at L1 in "cleared" nu/nu mice 30 epithelial cells are Ki67+), medium (20-80% of transplanted with wild type mammary gland tis- epithelial cells are Ki67 +) and high (>80% of ep- sue, and defective lobulo-alveolar development ithelial cells are Ki67+) numbers of proliferating at L1 in "cleared" nu/nu mice transplanted with cells +/- sem. *** P < 0.001; * P < 0.03 (Student’s RANKf1/Δ; K5-Cre mammary gland tissue. Fat t-test). d, Epithelial proliferation in mammary pads of nu/nu mice after surgical clearing are 35 glands of control littermates and RANK Δmam fe- also shown. All magnifications X 5. males 1 day after i.p. injection of PBS or RANKL (1mg). Proliferation was determined in situ Ki67 Figure 8. Normal formation of a lactating mammary immunostaining. Magnifications X 40. e, Quan- gland in pregnant MMTV-Cre, RANKf1/Δ females. tification of epithelial proliferation 1 day after i.p. 40 injection of PBS or RANKL (1 mg).Mean percent- a, H&E analyses of mammary tissue of ages of Ki67 positive cells +/- sem are shown. * nulliparous littermate control and RANKf1/Δ; P < 0.03; n=5 (Student’s t-test). MMTV-Cre females showing normal alveolar/ductal epithelial structures.Figure 10. Survival curves of progestin-driven mam- Magnifications X 10 (top) and X 40 (bottom45 mary cancers in RANKΔmam mice. panels). b, Whole-mount analyses of mammary tissue of nulliparous and lactating (L1) RANK f1/Δ a,b, MPA induces RANKL expression in females crossed with MMTV-Cre compared to mammary epithelial cells. Nulliparous wild type mammary glands of control littermates. MMTV- females were treated with oral gavage of DMBA Cre mediated deletion of RANK did not affect 50 or oil vehicle, s.c. implanted with slow-release formation of a lactating mammary gland.c, MPA pellets, or treated with sham surgery,a , Southern blot of purified mammary epithelial RANKL mRNA was determined in purified cells derived from RANKflox/Δ and RANKflox/Δ; mammary epithelial cells by qRT-PCR 3 days MMTV-Cre females. The wild type or floxed after implantation/oral gavage. Data are shown RANK allele (fl/+; 9.6 kb) and the deleted RANK 55 as fold change compared to sham treatment +/- allele (Δ;3.9 kb) are indicated after digestion of sem. n=3 mice per group.b , Expression of genomic DNA with PvuII and SphI. RANKflox/Δ; soluble (s) RANKL protein (19kDa) was assayed MMTV-Cre animals are denoted RANKΔmam on cell lysates from purified mammary epithelial

10 17 EP 2 618 833 B1 18

cells by Western blot 3 days after treatment with activated by four poly I:C injections i.p. over the oil vehicle, DMBA, MPA, or sham surgery.β - course of 8 days (200mg in 200 ml PBS). Data actin is shown as a loading control. Of note, only are shown as percentage of tumor free mice MPA but not DMBA or the oil vehicle alone after the last DMBA challenge. No significant induced RANKL mRNA and protein expression. 5 differences were found. b, Southern blot of the c, Onset of palpable mammary tumors in non-deleted RANKfloxed allele (fl/+) and after RANKΔmam (n=14) and age-matched littermate induction of deletion (Δ) in Mx-Cre rankfloxed/Δ control females with MMTV-Cre (Cre+ control; mice. c, Southern blot of various organs after n=13) or without MMTV-Cre (Cre- control; n=9) induction of deletion (Δ) in Mx-Cre rankfloxed/Δ treated with MPA pellets and DMBA as indicated 10 mice. While various degrees of deletion in Fig. 2a. Data are shown as percentage of (50-100%) can be seen in thymus, heart, liver, tumor-free mice after the last DMBA challenge. and spleen, deletion of the RANK floxed allelewas No significant difference was found between the not induced in purified mammary epithelial cells Cre+ and the Cre- control groups. Median tumor (MECs). d, Onset of palpable mammary tumors onset for Cre’ controls was 9 days, for Cre- 15 in NeuT RANKΔmam (NeuT, MMTV-Cre+ controls 11 days, and for RANKΔmam females rankfloxed/Δ) females (n=18) and age matched 30 days after the last DMBA treatment.d, NeuT (NeuT MMTV-Cre+ rank+/Δ) control Representative Southern blot of MPA/DMBA- littermate females (n=17). No significant induced mammary tumors derived from difference was found. e, Southern blot of tumors RANKflox/+; MMTV-Cre+ and RANKflox/Δ ; 20 derived from NeuT MMTVCre+ rankfloxed/Δ and MMTV-Cre+ (RANKΔmam) females. The wild mammary cancers from littermateNeuT type or floxed RANK allele (fl/+; 9.6 kb) and the MMTVCre+ RANK+/Δ females to control for deleted RANK allele Δ (;3.9 kb) are indicated RANK deletion in tumors. after digestion of genomic DNA with PvuII and SphI. All tumors derived from RANKΔmam 25 Figure 13. Tumor formation in NeuT RANKΔmam fe- females showed almost complete deletion.e , males. Kaplan Mayer analysis for overall survival of RANKΔmam (n=9) and age-matched littermate a-c, Representative histological sections with control females (n=9) after treatment with MPA typical adenocarcinomas detected in control /DMBA. Median survival was 48 days for control 30 and NeuT RANKΔmam female mice. and 93 days after the last DMBA treatment for Cytokeratin5 (a), Cytokeratin14 (b) and E-cad- RANKΔmam females. herin c ( ) stainings are shown in mammary glands of 6 month-old females. Mammary Figure 11. Development of squamous adenocarci- glands from littermate RANKΔmam females not nomas in RANKΔmam females. 35 expressing NeuT (NeuT-) are shown as controls. d, RANKL mRNA expression in mammary a,b, Representative histological sections of epithelial cells (MECs) isolated from either mammary tumors isolated from control sham- or MPA-treated virgin mice, in MPA-driv- littermate and RANKΔmam females 7 (a) and 21 en mamamry adenocarcinomas, and mammary (b) days after the last DMBA treatment. H&E and 40 tumors derived from NeuT, Pym, Neu, and E-cadherin stainings are shown indicating Neu/TGFβ transgenic mice. RANKL mRNA was typical features of ductal adenocarcinomas in determined by qRT-PCR. Data are shown as tumors from control littermates and RANKΔmam fold change compared to sham treatment (+/- females. Cytokeratin14 (K14) expression sem). n=3 per group. e, RANKL protein expres- demonstrates the basal cell origin in both45 sion in various tumors isolated from the indicat- controls and RANKΔmam females. However, ed genetic mouse models and MPA-driven tu- RANKΔmam females tend to show mors in control RANKΔ/+ and RANKΔmam fe- characteristics of squamous metaplasia. All males. Recombinant murine sRANKL protein is magnifications X 20. shown for molecular size comparison. β-actin is 50 shown as a loading control. Figure 12. Mammary cancer onset in Mx-Cre RANKfloxed/Δ and NeuT RANKΔmam mice. Figure 14. RANKL/RANK downstream signaling in MECs. a, Onset of palpable mammary tumors in Mx- Cre rankfloxed/Δ (n=4) and age matched Mx-Cre 55 a, Schematic outline of genetically confirmed rank+/Δ littermate control females (n=6) treated signalling pathways that control RANKL-RANK with MPA pellets and DMBA as indicated in Fig. mediated formation of a lactating mammary 2a. The Mx-driven Cre recombinase was gland during pregnancy. b, Western blotting for

11 19 EP 2 618 833 B1 20

phosphorylated (P) AKT, total AKT, phosphor- a, Western Blot analysis for γH2AX, phosphorylated (P) ERK1/2, total ERK1/2, phosphorylat- ylated (P) Chk1, total Chk1, p53 and β-actin in ed (P) p38-MAPK, and p38-MAPK in isolated isolated primary mouse mammary gland epithe- primary mouse mammary gland epithelial cells lial cells in response to γ-irradiation (2 Gray) in in response to RANKL stimulation (1 mg/ml). Da- 5 the presence (1ug/ml) or absence of RANKL ta are representative of 4 experiments. stimulation. b,c, FACS analysis of propidium io- dide(PI) stained b, mammaryepithelial cells and Figure 15. MPA activates the NFBκ pathway and c, SKBR3 human breast cancer cells after γ-ir- triggers CyclinDl expression via RANKL/RANK. radiation (2 Gray) in the absence or presence 10 (1mg/ml) of RANKL. Data are representative of a, Activation of the NFκB pathway and CyclinDl at least 3 experiments. Apoptotic cells with a expression by MPA. Nulliparous RANKΔmam DNA content < 2n appear in the sub-Gl region. and littermate control females were s.c. implant- Percent of cells in sub-Gl (M1), G1-phase (M2), ed with slow-release MPA pellets or treated with S/G2/M-phase (M3) as well as polyploid cells sham surgery. a, In situ immunostaining to de- 15 with a DNA content > 4n are given for the indi- tect phosphorylated (P) IκBα in mammary epi- cated time points. thelial cells of RANK Δmam and littermate control females after 3d MPA treatment. b, Western blot Figure 18. RANKL protects primary murine mamma- analysis of CyclinDl and RANKL in isolated ry epithelial cells and human SKBR3 breast cancer mammary epithelial cells from RANKΔmam and 20 cells from apoptosis in response to doxorubicin. littermate control females after 3d MPA treatment. Recombinant, murine sRANKL protein is a,b, FACS analysis ofa , mammary epithelial shown for molecular size comparison. β-actin is cells and b, SKBR3 human breast cancer cells shown as a loading control. incubated with the genotoxic agent doxorubicin 25 (Dox, 1mM) in the presence (1 mg/ml) or absence Figure 16. RANKL/RANK downstream signaling of RANKL. Data are representative of at least 3 pathways. experiments. Percent of cells in sub-Gl (M1), G1-phase (M2), S/G2/M-phase (M3) as well as a, Western blotting for phosphorylated (P) p65 polyploid cells with a DNA content >4n are given NFκB,total p65 NF κB, phosphorylated(P) I κBα, 30 for 24 and 36 hours after doxorubicin treatment. total IκBα, phosphorylated (P) ERK1/2, total c, mRNA expression of pro-apoptotic genes ERK1/2, phosphorylated (P) p38-MAPK, and Bim, Puma, and Noxa 6 hours after γ-irradiation p38-MAPK in human SKBR3 breast cancer cells (2 Gray) in the presence (1 mg/ml) or absence of in response to RANKL stimulation (1 mg/ml). Da- RANKL stimulation. Data are shown as fold ta are representative of 3 experiments.b, 35 change compared to control (+/- sem). n=3). * Growth curve of SKBR3 breast cancer cells cul- P < 0.05; ** P < 0.005 (Student’s t-test). tured in normal growth medium (control, DMEM supplemented with 10% FCS) or in the presence Figure 19. IKKα mediates MPA-induced protection of RANKL (1mg/ml). Cell numbers were deter- from radiation-induced epithelial apoptosis. mined by counting live cells (trypan blue-exclu- 40 sion) over 3 days. c, Onset of palpable mamma- a,b, Reduced induction of mammary epithelial ry tumors in NFATc1Δmam (n=10) and age cell apoptosis in response toγ-irradiation in matched littermate control (n=16) females treat- IKKαΔmam females. Littermates control and ed with MPA pellets and DMBA. Data are shown IKKαΔmam females were either sham operated as percentage of tumor free mice after the last 45 or implanted with an MPA pellet and γ-irradiated DMBA challenge. No significant difference was (5 Gray). MPA pellets were implanted 3 days found. d, Quantification of NFATc1 mRNA ex- before γ-irradiation. 24 hours after irradiation, pression in purified mammary epithelial cells apoptosis was detected by immunostaining for (MECs) and MPA-driven mammary tumors from active Caspase 3. a, Apoptotic nuclei of epithe- NFATc1Δmam and littermate control females.50 lial cells (arrows) are shown for representative mRNA was determined by qRT-PCR. Data are mammary gland sections. Magnifications X 40. shown as fold change compared to control (+/- b, Quantification of mammary epithelial apopto- sem). n=5 Per group. * P < 0.05 (Student’s t- sis. A minimum of 5000 nuclei was counted for test). each mouse. Results shown are mean values 55 +/- sem. n = 3 mice per group. * P < 0.05 (Stu- Figure 17. RANKL protects primary murine mamma- dent’s t-test). ry epithelial cells and human SKBR3 breast cancer cells from apoptosis in response to γ-irradiation. Figure 20. Quantification of Western blots.

12 21 EP 2 618 833 B1 22

[0048] Densitometry was performed on at least three Example 3: Mammary tissue transplants. For independent Western blots per experiment. Data are transplantation studies, mammary epithelial tissue shown for Western blots in Fig. 1b, Fig. 1c, Fig. 3a, and wasisolated from nulliparous 3-week-old donorsand Fig. 3b. Expression values for the indicated proteins were implanted into cleared mammary fat pads (devoid of normalized to β-actin loading controls. For quantification 5 endogenous epithelium) of 3-week-old hostnu /nu of phosphorylation data were normalized to the respec- mice as described7. Three weeks after surgery, tive total protein bands. * P <0.05; ** P < 0.001 (Student’s hosts were mated and mammary tissue was isolated t-test). for analysis. Example 4: Histology, whole-mount, and immu- Examples : 10 nohistochemistry. For histological analysis, 5pm sections were cut and stained with hematoxylin and [0049] eosin (H&E). Whole-mount staining of mammary glands was performed as described8. For immu- Example 1: Mice. Rankfloxed mice have been noperoxidase staining paraffin-embedded sections recently generated1. Briefly, to generate mice15 were dehydrated and antigenic epitopes exposed carrying a null allele of Rank (rankΔ allele), rankfloxed using a 10 mM citrate buffer or microwaving. Sec- mice were crossed to β-actin-Cre ubiquitous deleter tions were incubated with antibodies to cytokeratine mice. Mice carrying the rankfloxed or rankΔ alleles as 5, cytokeratine 14, E-cadherin, anti-Ki67 (Novocas- well as the MMTV-Cre mice were backcrossed tra) and anti-active Caspase 3 (Cell Signaling) and seven times onto a BALBc background before20 visualized using peroxidase-conjugated secondary generating the MMTV-Cre rankΔ/floxed mice. MMTV- antibodies. Histomorphometric indices (proliferation NeuT mice were kindly provided by Guido Forni, and apoptosis) were calculated as the number of Milan. MMTV-Cre (stock # 003553) and Mx-Cre mice positive epithelial cells divided by the total number (stock # 003556) were obtained from the Jackson of epithelial cells, with no fewer than 1000 nuclei for Laboratory. K5-Cre, IKKαfloxed and NFATc1floxed 25 Ki67 stainings and no fewer than 5000 cells for active mice have been previously described2-4. Mouse Caspase 3 staining counted per section. genotypes were determined by PCR and Southern Example 5: Western blotting. The human epithelial blot analysis. In all experiments, only littermate mice breast tumor cell line SKBR3 and primary non-trans- from the same breedings were used. All mice were formed mouse mammary epithelial cells were left un- bred and maintained according to institutional30 treated or stimulated with recombinant murine guidelines. RANKLref. 9. Adenocarcinomas were isolated from RANK deletion in tumors and Cre effects. control and mutant mice and total protein lysates pre- Southern blotting of the tumors that developed in pared. Western blotting was carried out using stand- RANKΔmam femalesshowed deletion of RANK, albeit ard protocols. Briefly, blots were blocked with 5% some residual wild type band was observed (Fig. 35 BSA in 1xTBS 0.1% Tween-20 (TBST) for 1 hour 10c) that may be explained by the presence of other andincubated with primary antibody overnightat 4°C cell types and/or escaper cells. Differences in tumor (diluted in TBST according to the manufactures pro- onset in Cre-negative control females and tocol). Primary antibodies reactive to mouse RANKL littermates expressing the MMTV-Cre transgene (AF462; R&D), Cyclin D1 (Santa Cruz #Sc-8396), β- were not observed indicating that Cre expression per 40 actin (Sigma), phosphorylated (P) NFκB (#3033), se does not alter tumor incidence in the MPA/DMBA NF-κB (#4767), phosphorylated (P) κIBα (#2859), mammary tumor model (Fig. 10d). IκBα (#4814), phosphorylated (P) IKK α (#2681), IK- Example 2: MPA/DMBA-induced mammary car- Kα (#2678), IKKβ (#2678), IKKγ (#2685), phospho- cinogenesis. MPA/DMBA treatment was performed rylated (P) Akt (#3787), Akt (#9272), phosphorylated as described5,6. Briefly, six-week old female mice 45 (P) Erk1/2 (#9101), Erk1/2 (#9102), and p38-MAPK were anesthetized with ketamine-xylazine and sur- (#9212), p53 (#2524), phosphorylated (P) Chk1 gically implanted with slow-release Medroxyproges- (#2348), Chk1 (#2345) (all from Cell Signaling), p38- terone Acetat (MPA) pellets (50mg, 90-day release; MAPK (AF869; R&D), and γH2Ax (Ser139 #07-164 Innovative Research of America) subcutaneously on Millipore) were used. Blots were washed 3 times in the right flank. 200ml DMBA (5mg/ml diluted in cot- 50 TBST for 30 minutes, incubated with HRP-conjugat- tonseed oil) was administered by oral gavage 6 times ed 2nd antibodies (1:2000, Promega) for 1 hour at throughout the following 8 weeks as outlined in Fig. room temperature, washed 3 times in TBST for 30 2a. Onset of mammary tumors was determined by minutes, and visualized using ECL. palpation. Differences intumor onset in Cre-negative Example 6: qRT-PCR. Total RNA of tumors was pre- control females and littermates expressing the55 pared using the RNeasy Mini Kit (Qiagen), according MMTV-Cre transgene were not observed indicating to the manufacturer’s instructions. Total RNA (2 mg) that Cre expression per se does not alter tumor in- was subjected to quantitative (q)RT-PCR analysis. cidence in the MPA/DMBA mammary tumor model. The following primers were used:

13 23 EP 2 618 833 B1 24

β-actin forward primer: 5’- GCTCATAGCTCT- tive of apoptosis. TCTCCAGGG -3’; Example 8: FACS analysis of primary mammary β-actin reverse primer: 5’-CCTGAACCCTAAG- epithelial cells. For FACS analysis of mammary ep- GCCAACCG -3’. ithelial subpopulations the following protocol for tis- RANKL forward primer: 5’ - CTGAG-5 sue dissociation was used to generate single cell GCCCAGCCATTTG -3’ suspensions: lymph nodes were removed from both RANKL reverse primer: 5’ - GTTGCTTAACGT- inguinal mammary fat pads and fat pads were then CATGTTAGAGATCTTG -3’ digested in 2ml complete EpiCult medium (EpiCult- RANK forward primer: 5’ - CTTGGACACCT- B basal medium (StemCell Technologies SCT Cat- GGAATGAAG -3’ 10 alog #05610) supplemented with EpiCult-B prolifer- RANK reverse primer: 5’ - CAGCACTCG- ation supplements, 10ng/ml bFGF (SCT Catalog CAGTCTGAGTT -3’ #02634), 10 ng/ml EGF (SCT Catalog #02633), CyclinDl forward primer: 5’ - CTGT-4mg/ml Heparin (SCT Catalog #07980), 2.5ml FBS GCGCCCTCCGTATCTTA-3’ (5%) and antibiotics) with 2.5x collagenase/hyaluro- CyclinDl reverse primer: 5’ - GGCGGCCAGGT- 15 nidase (e.g. 500ml 10x collagenase + 1.5ml Epicult TCCACTTGAG-3’ per mouse) (SCT Catalog #07912) in 50ml Falcon p21 (Cdkn1a) forward primer: 5’ - GTGGCCTT- tubes at 37°C for 2.5 hours. After vigorous vortexing, GTCGCTGTCTT -3’ pellets were washed with 10mL HF medium (Hanks p21 (Cdkn1a) reverse primer: 5’ - GCGCTT- Balanced Salt Solution Modified SCT Catalog GGAGTGATAGAAATCTG-3’ 20 #07913 + 2% FBS). Pellets were then resuspended tRANKL forward primer: 5’ - GCGCCG- in 2ml of pre-warmed trypsin-EDTA. After another GGCCAGCCGAGACTAC-3’ wash with 10 mL HF, pellets were resuspend in 2mL RANKL1 forward primer: 5’ - GTCCCACAC- pre-warmed (37°C) dispase (SCT Catalog #07913) GAGGGTCCGCTGC-3’ supplemented with 200 ml of 1 mg/mL DNAse I (SCT RANKL2 forward primer: 5’ - TGCGCACTCCG- 25 Catalog #07900). After a final wash in HF, cells were GCGTCCCG-3’ countedand prepared for FACSstaining. One million RANKL3 forward primer: 5’ - CCGAGACTACG- cells were incubated with the following antibodies: GCGGATCCTAACAG-3’ biotin-conjugated anti-CD31 (#553371; BD), which RANKLcom. reverse primer: 5’ - TCAGTCTAT- labels endothelial cells, and biotinylated CD45 + and GTCCTGAACTTTGAAAGCCCC-3’ 30 Ter119+ (StemSep murine chimera cocktail Puma forward primer: 5’ -CCGCCTGAT- #13058C; Stem Cell Technologies; 3.5 ml/100ul vol), GCCCTCCGCTGTAT -3’ which labels hematopoietic cells, for 10min at room Puma reverse primer: 5’ -CGGGCCCACTC- temperature. Hematopoietic and endothelial cells CTCCTCCTCCAC -3’ were excluded by FACS using Strepavidin-conjugat- Noxa forward primer: 5’ -ACTTTGTCTC-35 ed-APC (#554067; BD). Staining with anti-CD49f CAATCCTCCG -3’ (#551129; BD) and CD24 (#553261; BD) was used Noxa reverse primer: 5’ - GTGCACCGGACAT- to identify the mammary stem cell population as pre- AACTGTG-3’ viously described10,11. Bim forward primer: 5’ - GTTGAACTCGTCTC- Example 9: Cancer stem cell assays . Self-renewal CGATCC-3’ 40 of mammary cancer stem cells, i.e. tumor initiating Bim reverse primer 5’ - GCCCCTACCTCCCT- cells (TICs), was assayed using a mammosphere ACAGAC-3’ assay as described previously6,12. Briefly, similar sized tumors (1cm3 volume) were minced and di- Example 7: DNA damage responses. For meas- gested in complete EpiCult medium with 2.5x colla- urement of cell cycle arrest and apoptosis primary 45 genase/hyaluronidase (SCT Catalog #07912). mouse mammary epithelial cells and SKBR3 human 2x105 cells were then cultured in serum-free EpiCult breast cancer cells were seeded at a cell-density of medium supplemented with B27 (Invitrogen), 100000 cells/well in a 6-well plate and allowed to 20ng/ml EGF (Protech), and 20 ng/ml bFGF (Sigma) grow for 24 hours. Cells where then treated with dox- using 6 well ultra-low attachment plates (Corning orubicin (1mM) or γ-irradiation (2 Gray) in the ab- 50 Costar). The primary mammospheres, which formed sence or presence of recombinant RANKL (1 mg/ml). over 7 days, were collected by gentle centrifugation Cell cycle arrest and numbers of dead cells were (800rpm), digested into single cell suspensions with determined using propidium iodine staining. To de- trypsin (0.05%, 10min), and assayed for their ability termine in vivo mammary gland epithelial cell death, to form secondary mammospheres as above. control and RANKΔmam littermate females were γ- 55 Example 10: Anchorage-independent growth. irradiated with a total dose of 5 Gray (Gy). Six hours The ability of cells to grow in soft agar was assayed later, mammary glands were isolated and immunos- as described previously13. Briefly, DNA grade aga- tained for active Caspase 3 (Cell Signaling) indica- rose (1% in DMEM) was used as bottom layer (2ml

14 25 EP 2 618 833 B1 26

in 6 well plates) and 2x10 4 SKBR3 cells were seeded of experimental animals. For the study involving the in 1,5ml agarose (0,3% in DMEM). Cells were over- RANKΔmam animals the null hypothesis can be reject with layed with 1.5ml DMEM supplemented with 10% a probability (power) of 0,933 and for the IKKαΔmam an- FCS and cultured for 24 days. imals with a probability of 0,766. The Type I error prob- 5 ability associated with this test of this null hypothesis is Example 11: Progesterone assay 0,05. Human data were analysed using a paired t-Test, the Mann Whitney U test, or a Spearman rank test as [0050] indicated.

• 1st incubation: 30 mL sample - in the presence of a 10 Example 13: Effects of progestin on RANKL levels: biotinylated monoclonal progesterone-specific antibody and a progesterone derivative labeled with ru- [0055] MPA (medroxyprogesterone acetate, a repre- thenium complex - are incubated with Danazol to sentive progestin) pellets were implanted into female release progesterone. Progesterone from the sam- mice and RANKL expression was assayed. MPA treat- ple competes with the labeled progesterone deriva- 15 ment resulted in a more than 2000 fold induction of RAN- tive for the antibody binding site. KL mRNA in isolated mammary epithelial cells (Fig. 1a). • 2nd incubation: After addition of streptavidin-coated mRNA expression of RANK, of the decoy receptor of microparticles, the complex becomes bound to the RANKL osteoprogesterin (OPG), of PTHrP (known to in- solid phase via interaction of biotin and streptavidin. duce RANKL) and of TRAIL (known to bind to OPG) were The amount of the labeled progesterone derivative 20 not changed in isolated mammary epithelial cells. Immu- bound to the solid phase is inversely proportional to nohistochemistry confirmed induction of RANKL protein the progesterone content of the sample. in the mammary gland (Fig. 1b). Western blotting of iso- • The reaction mixture is aspirated into the measuring lated mammary epithelial cells also showed induction of cell where the microparticles are magnetically cap- the 19kDa soluble form of RANKL (Fig. 5a), most likely tured onto the surface of the electrode. Unbound 25 due to alternative splicing (Fig. 5b-d) as well as induction substances are then removed with ProCell. Applica- of MMP14 and ADAM17/TACE (Fig. 5e), two proteases tion of a voltage to the electrode then induces chemi- known to shed RANKL from the cell surface. RANKL in- luminescent emission which is measured by a pho- duction in response to MPA in in vitro cell culture assays tomultiplier. was not observed (Fig. 6a-c) suggesting that progestins • Results are determined via a calibration curve which 30 per se are not sufficient to induce RANKL expression. is instrument-specifically generated by 2-point cali- Since MPA can induce transcription of prolactin and the bration and a master curve provided via the reagent prolactin receptor in breast cancer cells and prolactin can barcode. directly induce RANKL via Jak2/Stat5 signaling, it was tested whether MPA induces RANKL in the mammary Reagents - working solutions 35 gland of prolactin receptor knock-out mice. Whereas MPA triggers substantial production of RANKL protein in [0051] M Streptavidin-coated microparticles (transpar- female control mice, we failed to detect RANKL induction ent cap), 1 bottle, 6.5mL: Streptavidin-coated micropar- in prolactin receptor mutants (Fig. 1c). This data indicates ticles, 0.72mg/mL; preservative. that MPA triggers massive RANKL expression in the [0052] R1 Anti-progesterone-Ab∼biotin (gray cap), 1 40 mammarygland via theprolactin receptor and/orthat pro- bottle, 10 mL: Biotinylated monoclonal anti-progesterone lactin and MPA cooperate in RANKL induction. antibody (mouse) 0.15mg/L, phosphate buffer 25mmol/L, pH 7.0; preservative. Example 14: Effects of progestins on RANKL levels [0053] R2 Progesterone-peptide∼Ru(bpy)2+ (black in RANKL/RANK impaired mice: cap), 1 bottle, 8 mL: Progesterone (of vegetable origin) 45 coupled to a synthetic peptide labeled with ruthenium [0056] To examine the potential role of RANKL/RANK compley, 10 ng/mL; phosphate buffer 25 mmol/L, pH 7.0; in MPA-mediated tumorigenesis RANK was deleted in preservative. mammary epithelial cells using K5-Cre and MMTV-Cre [0054] Example 12: Statistics. All values herein are mediated excision of an inducible RANK allele (K5-Cre given as means sem. Comparisons between groups50 rankflox/Δ mice and MMTV-Crerank flox/Δ mice). Both were made by Student’s t-test. For the Kaplan-Meier mouse lines appear healthy and exhibit normal bone analysis of tumor onset a log rank test was performed. structures and lymph node formation. As expected K5- P 0.05 was accepted as statistically significant. In addi- Cre rankflox/Δ mice exhibited apparently normal mamma- tion to the Log RANK test a post-hoc power analysis was ry gland development in puberty; however, these mice performed (PS Power and Sample Size Calculations, in 55 did not develop milk-secreting lobuloalveolar structures the web at biostat.mc.vanderbilt.edu/PowerSampleSize) during pregnancy (Fig. 7a,b). These effects were cell au- to calculate the probability of correctly rejecting the null tonomous using transplantation experiments (Fig. 7c). In hypothesis of equal tumor onset times given the number MMTV-Cre rankflox/Δmice, mammary gland development

15 27 EP 2 618 833 B1 28 in nulliparous females and formation of milk-secreting mas of the Cytokeratine (CK) 5 and CK14 positive basal lobuloalveolar structures in pregnancy appeared normal cell subtype (Fig. 2c, Fig. 11a,b). However, ductal ade- (Fig. 8a-c). To exclude any issue of development effects nocarcinomas arising in RANKΔmam females frequently in K5-Cre rankflox/Δ mice that might affect certain cell pop- developed extensive areas of squamous metaplasia ulations in normal physiology, MMTV-Cre rankfloxlΔ mice 5 (Fig. 2c, Fig. 11a,b). In line with these histopathological were therefore used for all further experiments. These changes, gene expression profiling of mammary carci- MMTV-Cre rankflox/A mutant mice are hereafter termed nomas from control and RANK Δmam females showed dis- RANKΔmam. tinctive differences in their molecular signatures as assessed by gene clustering. Example 15: Mechanism of RANKL activation upon 10 progestin administration: Example 17: Cancer onset after progestin challenge and DNA damage: [0057] In a wild type population MPA treatment triggers massive proliferation of mammary epithelial cells. MPA- [0060] Since RANKΔmam showed a delayed onset in induced proliferation of mammary epithelial cells was sig- 15 progestin-induced mammary cancer, next the incidence nificantly reduced in RANKΔmam females (Fig. 1d; Fig. of mammary tumors at early stages after MPA/DMBA 9a-c). Accordingly, RANKL i.p. injections into nulliparous challenge was analysed. One week after the last DMBA females triggered proliferation of mammary gland epithe- treatment, all RANK expressing control females already lial cells via RANK (Fig. 9d,e). Recently it has been re- exhibited multiple in situ carcinomas and even invasive ported that endogenous progesterone affects the num- 20 mammary tumors. By contrast, very few carcinomas bers of Lin-CD24+CD49f hi stem cells during pregnancy 15 were observed in situ and never any invasive mammary and the estrous cycle16. In both studies, the RAN- carcinomas in RANKΔmam animals one week after the KL/RANK system was implicated based on in vivo whole last DMBA challenge (Fig. 2d). Three weeks after the last body Ab blocking studies and RT-PCR expression, how- DMBA challenge the incidence of carcinomas in situ was ever, it was not known whether this is a direct effect of 25 similar among control and RANKΔmam females, but RANKL-RANK in mammary epithelial cells rather than a invasive carcinomas were still very infrequent in the secondary effect. Therefore it was assayed whether pro- RANKΔmam females (Fig. 2e). Moreover, proliferation gestins such as MPA can also expand was Lin- typically reduced in tumors from RANK Δmam females CD24+CD49fhi cells. MPA treatment resulted in a two- (Fig. 2d,e). Deletion of RANK in multiple other tissues fold expansion of Lin-CD24+CD49fhi cells. Such expan- 30 including the liver, heart, muscle and the haematopoiet- sion did not occur in MPA-treated RANKΔmam females ic compartment, but not in mammary epithelial cells, (Fig. 1e,f). These data provide the first genetic proof that using Mx-Cre rankflox/flox mice did not delay the onset of the RANKL/RANK system controls expansion of Lin- MPA/DMBA-induced mammary cancer (Fig. 12a-c), CD24+CD49fhi cells. suggesting that the effects of RANK/RANKL impairment 35 are restricted to mammary epithelial cells. Moreover, Example 16: Influence of progestins on cancer devel- mammary gland specific deletion of RANK did not alter opment via RANK/RANKL in control and RANK/RAN- mammary cancer incidence (Fig. 12d,e) or the KL deficient mice: histopathology of adenocarcinomas (Fig. 13a-c) in MMTV-NeuT transgenic mice, a genetic model of [0058] In The Women’s Health Initiative (WHI) and the 40 mammary cancer. In these NeuT tumors very low levels Million Women Study, the use of progestins has been of RANKL were observed (Fig. 13d,e). Thus, genetic epidemiologically linked to the onset and incidence of inactivation of RANK in mammary epithelial cells results breast cancer. Progestin-driven mammary cancer can in a markedly delayed onset and reduced incidence of be modeled in female mice, where implantation of slow progestin-driven mammary cancer. release MPA pellets in the presence of the DNA damag- 45 ing agent dimethylbenz[a]anthracene (DMBA) triggers Example 18: RANKL signaling: mammary cancer (Fig. 2a; Fig. 10a,b). [0059] In control females, MPA/DMBA treatment in- [0061] RANKL-RANK signaling via α-NF IKKκB- duced a rapid onset of palpable mammary tumors. In- CyclinD1 in mammary epithelial cells is illustrated in Fig. triguingly, in RANKΔmam female mice, a marked delay in 50 14a. RANKL stimulation indeed resulted in p65 NFκB the onset of MPA/DMBA-induced mammary cancer was and IκBα phosphorylation in primary mouse mammary observed (Fig. 2b; Fig. 10c,d). Delayed tumor onset on gland epithelial cells (MECs) (Fig. 3a). In addition, RAN- RANKΔmam females also resulted in markedly enhanced KL stimulationof MECs triggeredphosphorylation of p38- survival (Fig. 10e). Southern blotting of the tumors that MAPKs and ERK (Fig. 14b). To directly show whether developed in RANK Δmam females confirmed efficient de- 55 RANK mediates NFκB-CyclinD1 activation downstream letion of RANK (Fig. 10c). All tumors that developed in of progestins in vivo, mice were challenged with MPA. A control and RANK Δmamfemales exhibitedtypical histopa- three day MPA treatment resulted in nuclear accumula- thology of E-Cadherin expressing ductal adenocarcino- tion of phosphorylated I κBα, indicative of an active NF κB

16 29 EP 2 618 833 B1 30 pathway, and induction of CyclinDl protein expression in massive proliferation of mammary epithelial cells, MPA mammary epithelial cells, both of which were severely alone is not sufficient to trigger mammary cancer which reduced in RANK Δmam female (Fig. 15a,b). Moreover, in requires a carcinogen to induce DNA mutations. To an- MPA/DMBA-induced mammary adenocarcinomas iso- alyze the role of RANKL in the cellular response to DNA lated from control and RANK Δmam females we found im- 5 damage such as cell cycle arrest and apoptosis, mouse paired activation of the NFKB pathway (Fig. 3b) and primary mammary epithelial cells (MECs) and the RAN- downregulated mRNA expression of Cyclin D1 (Fig. 3c). KL-responsive human breast cancer cell line SKBR3 In these primary tumors also upregulation of p21 mRNA were treated with DNA damaging agents doxorubicin or (Fig. 3c) was observed, a gene that is transcriptionally γ-irradiation. RANKL treatment did not alter induction of suppressed by the Id2 pathway17. The Id2 pathway is a 10 γH2AX and p53 or activation of Chk1, prototypic markers second genetically confirmed downstream pathway for of a functional DNA damage response (Fig. 17a). More- RANKL/RANK in mammary epithelial cells17. To extend over, RANKL did not alter the early cell cycle arrest after these findings to human, human SKBR3 breast tumor DNAdamage with γ-irradiation (Fig. 17b,c) or doxorubicin cells were assayed. RANKL stimulation induced NFκB, (Fig. 18a,b). Surprisingly, RANKL treatment resulted in p38-MAPKs and ERK activation and proliferation in15 a marked protection from cell death in response toγ- SKBR3 cells (Fig. 16a,b). To further test the effects of irradiation (Fig. 17b,c) and doxorubicin-induced DNA RANKL stimulation the ability of these cells to grow in an damage (Fig. 18a,b). Mechanistically,γ -irradiation-in- anchorage-independent manner was assessed, which duced upregulation of the pro-apoptotic molecules Bim, correlates well with tumorigenicity in vivo18. Importantly, Puma, and Noxa did not occur in the presence of RANKL similar to EGFR stimulation, it was observed that RANKL 20 (Fig. 18c). In vivo it has been shown that γ-irradiation of induced growth of SKBR3 cells in soft agar (Fig. 3d), i.e. female mice results in a 5-fold induction of apoptosis of RANK signaling not only triggers proliferation but also mammary epithelial cells 19. Therefore this previously es- acts as a transforming agent to induce anchorage-inde- tablished system was used to assess the effects of the pendent growth. MPA-RANKL/RANK axis onγ -irradiation-induced cell [0062] In osteoclasts, besides the NFκB pathway, the 25 death. MPA treatment indeed protected fromγ -irradia- calcineurin-NFATcl signaling pathway has been found to tion-induced apoptosis of mammary epithelial cells in vi- be essential for RANKL-RANK mediated osteoclas- vo. Loss of RANK expression on mammary epithelial togenesis. NFATcl can also be regulated by the Id2 path- cells abrogated the protective effects of MPA on γ-irradi- wayduring RANKL-mediatedosteoclastogenesis. Toas- ation-induced cell death (Fig. 3f,g). Moreover, the IKKα sess whether these key RANKL-RANK activation path- 30 pathway was involved in MPA-induced protection of the ways are also operational in MPA/DMBA-induced mam- mammary epithelium afterγ -irradiation (Fig. 19a,b). mary cancer, MMTV-Cre nfatc1flox/Δ (NFATc1Δmam) and Thus, in addition to promoting cell cycle progression, MMTV-Cre Ikkαflox/flox (IKKaAmam) mice were generated MPA can protect from cell death after DNA damage via to delete NFATcl and IKKα in mammary epithelial cells. RANKL/RANK and IKKα signaling. Both mutant mouse strains appear healthy and exhibit 35 no overt defects in any organs assayed. When chal- Example 20: Tumor inhitiating stem cells: lenged with MPA/DMBA, IKK αΔmam mice exhibited a delayed onset of mammary cancer (Fig. 3e). In tumors from [0064] Recently it has been shown in humans and mice IKKαΔmam mice normal expression of IKK β and IKK γ was that mammary tumors might arise from stem cell found but reduced NF κB activation as determined by in- 40 populations20. Therefore it was tested whether loss of creased IκB protein levels and decreased p65 NFκB RANK affects mammary cancer stem cells, i.e. tumor in- phosphorylation (Fig. 3b) and downregulated mRNA ex- itiating cells (TICs). TICs can be functionally assayed by pression of the NFκB target gene Cyclin D1 (Fig. 3c) their ability to form non-adherent mammospheres20. suggesting that IKKα is indeed required for NFκB sign- Freshly isolated cancer cells from control and RANK Δmam aling in these tumors. As expected, the Id2 pathway gene 45 females were able to form primary mammospheres, how- p21 was not affected in tumors from IKK αΔmam mice (Fig. ever, after dispersion into single cells the ability to form 3c). No significant differences in the tumor onset between secondary mammospheres was significantly impaired control and NFATc1Δmam mice were observed (Fig. using TICs from RANK Δmam mice (Fig. 4a,b). These data 16c,d), suggesting that downstream signaling require- indicate that loss of RANK activity or expression mark- ments are different in osteoclast progenitors and mam- 50 edly impairs the self-renewal capacity of cancer stem mary gland epithelial cells. Thus, deletion of IKKα, but cells. not NFATcl, in mammary gland epithelial cells affects the onset of progestin-driven mammary cancer. Example 21: Discussion:

Example 19: Hormone driven cancer development 55 [0065] Based on these results the following molecular due to DNA damage: mechanism how hormones such as MPA drive cancer via RANKL is apparent: MPA triggers an enormous in- [0063] AlthoughMPA treatmentinduces very rapid and duction of RANKL in the mammary gland. The induction

17 31 EP 2 618 833 B1 32 of RANKL in response to MPA requires expression of the gland development. Cell 103, 41-50 (2000). prolactin receptor and possibly other intermediates. 9. Lacey, D.L. et al. Osteoprotegerin ligand is a cy- RANKL via RANK on mammary epithelial cells drives tokine that regulates osteoclast differentiation and these cells into the cell cycle and, importantly, protects activation. Cell 93, 165-76 (1998). mouse as well as human mammary gland epithelial cells 5 10. Stingl, J. et al. Purification and unique properties from apoptosis in response to DNA damage including γ- of mammary epithelial stem cells. Nature 439, 993-7 irradiation. Moreover, RANKL-/RANK control self renew- (2006). al of mammary cancer stem cells and anchorage-inde- 11. Shackleton, M. et al. Generation of a functional pendent growth. Thus, progestin-induced RANKL/RANK mammary gland from a single stem cell. Nature 439, provide a growth and survival advantage to damaged 10 84-8 (2006). mammary epithelium, a pre-requisite to initiate mammary 12. Dontu, G. et al. In vitro propagation and tran- cancer21. These effects are, at least in part, mediated via scriptional profiling of human mammary stem/pro- the IKKα-NFκB signaling pathway. Millions of women genitor cells. Genes Dev 17, 1253-70 (2003). take progesterone-derivatives in contraceptives and for 13. Freedman, V.H. & Shin, S.I. Cellular tumorigenic- hormonal replacement therapy. Such hormones have 15 ity in nude mice: correlation with cell growth in semi- been epidemiologically linked to an increased risk to de- solid medium. Cell 3, 355-9 (1974). velop breast cancer. The present invention shows that 14. Jones, D.H. et al. Regulation of cancer cell mi- the RANKL-RANK system is an important molecular link grationand-bone metastasis by RANKL. Nature 440, between progestins and epithelial carcinogenesis. RAN- 692-6 (2006). KL inhibition is therefore a novel approach to prevent 20 15. Asselin-Labat, M.L. et al. Control of mammary and/or treat cancer. RANKL inhibition can also be used stem cell function by steroid hormone signalling. Na- to prime cancer cells for anti-cancer therapy such as ture. based on radiation or other DNA damaging agents. 16. Joshi, P.A. et al. Progesterone induces adult mammary stem cell expansion. Nature. References : 25 17. Kim, N.S. et al. Receptor activator of NF-kappaB ligand regulates the proliferation of mammary epi- [0066] thelial cells via Id2. Mol Cell Biol 26, 1002-13 (2006). 18. Freedman, V.H. & Shin, S.I. Cellular tumorigenic- 1. Hanada, R. et al. Central control of fever and fe- ity in nude mice: correlation with cell growth in semi- male body temperature by RANKL/RANK. Nature 30 solid medium. Cell 3, 355-9 (1974). 462, 505-9 (2009). 19. Ewan, K.B. et al. Transforming growth factor- 2. Tarutani, M. et al. Tissue-specific knockout of the betal mediates cellular response to DNA damage in mouse Pig-a gene reveals important roles for GPI- situ. Cancer Res 62, 5627-31 (2002). anchored proteins in skin development. Proc Natl 20. Pece, S. et al. Biological and molecular hetero- Acad Sci USA 94, 7400-5 (1997). 35 geneity of breast cancers correlates with their cancer 3. Gareus, R. et al. Normal epidermal differentiation stem cell content. Cell 140, 62-73. but impaired skin-barrier formation upon keratinoc- 21. Hanahan, D. & Weinberg, R.A. The hallmarks of yte-restricted IKK1 ablation. Nat Cell Biol 9, 461-9 cancer. Cell 100, 57-70 (2000). (2007). 4. Aliprantis, A.O. et al. NFATcl in mice represses 40 SEQUENCE LISTING osteoprotegerin during osteoclastogenesis and dis- sociates systemic osteopenia from inflammation in [0067] cherubism. J Clin Invest 118, 3775-89 (2008). 5. Aldaz, C.M., Liao, Q.Y., LaBate, M. & Johnston, <110> IMBA - Institut für Molekulare Biotechnologie D.A. Medroxyprogesterone acetate accelerates the 45 GmbH development and increases the incidence of mouse mammary tumors induced by dimethylbenzanthra- <120> Breast Cancer Therapeutics cene. Carcinogenesis 17, 2069-72 (1996). 6.Cao, Y., Luo, J.L. & Karin, M.IkappaB kinase alpha <130> r59661 kinase activity is required for self-renewal 50 of ErbB2/Her2-transformed mammary tumor-initiating <150> EP 10178346.2 cells. Proc Natl Acad Sci USA 104, 15852-7 (2007). <151> 2010 09 22 7. Robinson, G.W. & Hennighausen, L. Inhibins and activins regulate mammary epithelial cell differenti- <160> 21 ation through mesenchymal-epithelial interactions. 55 Development 124, 2701-8 (1997). <170> PatentIn version 3.5 8. Fata, J.E. et al. The osteoclast differentiation factor osteoprotegerin-ligand is essential for mammary <210> 1

18 33 EP 2 618 833 B1 34

<211> 21 <220> <212> DNA <223> Primer <213> Artificial Sequence <400> 6 <220> 5 cagcactcgc agtctgagtt 20 <223> Primer <210> 7 <400> 1 <211> 21 gctcatagct cttctccagg g 21 <212> DNA 10 <213> Artificial Sequence <210> 2 <211> 21 <220> <212> DNA <223> Primer <213> Artificial Sequence 15 <400> 7 <220> ctgtgcgccc tccgtatctt a 21 <223> Primer <210> 8 <400> 2 <211> 21 cctgaaccct aaggccaacc g 21 20 <212> DNA <213> Artificial Sequence <210> 3 <211> 18 <220> <212> DNA <223> Primer <213> Artificial Sequence 25 <400> 8 <220> ggcggccagg ttccacttga g 21 <223> Primer <210> 9 <400> 3 30 <211> 19 ctgaggccca gccatttg 18 <212> DNA <213> Artificial Sequence <210> 4 <211> 28 <220> <212> DNA 35 <223> Primer <213> Artificial Sequence <400> 9 <220> gtggccttgt cgctgtctt 19 <223> Primer 40 <210> 10 <400> 4 <211> 23 gttgcttaac gtcatgttag agatcttg 28 <212> DNA <213> Artificial Sequence <210> 5 <211> 20 45 <220> <212> DNA <223> Primer <213> Artificial Sequence <400> 10 <220> gcgcttggag tgatagaaat ctg 23 <223> Primer 50 <210> 11 <400> 5 <211> 22 cttggacacc tggaatgaag 20 <212> DNA <213> Artificial Sequence <210> 6 55 <211> 20 <220> <212> DNA <223> Primer <213> Artificial Sequence

19 35 EP 2 618 833 B1 36

<400> 11 <210> 17 gcgccgggcc agccgagact ac 22 <211> 23 <212> DNA <210> 12 <213> Artificial Sequence <211> 22 5 <212> DNA <220> <213> Artificial Sequence <223> Primer

<220> <400> 17 <223> Primer 10 cgggcccact cctcctcctc cac 23

<400> 12 <210> 18 gtcccacacg agggtccgct gc 22 <211> 20 <212> DNA <210> 13 15 <213> Artificial Sequence <211> 19 <212> DNA <220> <213> Artificial Sequence <223> Primer

<220> 20 <400> 18 <223> Primer actttgtctc caatcctccg 20

<400> 13 <210> 19 tgcgcactcc ggcgtcccg 19 <211> 20 25 <212> DNA <210> 14 <213> Artificial Sequence <211> 25 <212> DNA <220> <213> Artificial Sequence <223> Primer 30 <220> <400> 19 <223> Primer gtgcaccgga cataactgtg 20

<400> 14 <210> 20 ccgagactac ggcggatcct aacag 25 35 <211> 20 <212> DNA <210> 15 <213> Artificial Sequence <211> 30 <212> DNA <220> <213> Artificial Sequence 40 <223> Primer

<220> <400> 20 <223> Primer gttgaactcg tctccgatcc 20

<400> 15 45 <210> 21 tcagtctatg tcctgaactt tgaaagcccc 30 <211> 20 <212> DNA <210> 16 <213> Artificial Sequence <211> 23 <212> DNA 50 <220> <213> Artificial Sequence <223> Primer

<220> <400> 21 <223> Primer gcccctacct ccctacagac 20 55 <400> 16 ccgcctgatg ccctccgctg tat 23

20 37 EP 2 618 833 B1 38

Claims 8, wherein the RANKL inhibitor is Denosumab.

1. A RANKL inhibitor for use in a method of preventing 10. Inhibitor for use according to any one of claims 1 to a primary breast or mammary cancer in a subject, 9, wherein said cancer comprises cancerous cells comprising administering a therapeutically effective 5 selected from a breast tissue cell or breast cancer amount of the RANKL inhibitor to said patient, cell, and/or an epithelial cell or basal cell and/or a wherein the RANKL inhibitor is an anti-RANKL anti- cancer stem cell. body. 11. Pharmaceutical composition comprising a RANKL 2. A RANKL inhibitor for use in a method of treating a 10 inhibitor and a hormone selected from estrogen and primary breast or mammary cancer in a subject, progesterone or a progestin or a hormone contra- comprising administering a therapeutically effective ceptive for use ina treatmentor prevention of primary amount of a RANKL inhibitor to said patient for the breast or mammary cancer, wherein the RANKL in- reduction of cancer cells or the prevention of cancer hibitor is an anti-RANKL antibody. progression in the tissue of origin of said primary 15 breast or mammary cancer, 12. Composition for use according to claim 11 further wherein the RANKL inhibitor is an anti-RANKL anti- comprising a pharmaceutically acceptable diluent, body. carrier, solubilizer, emulsifier, preservative and/or adjuvant. 3. A RANKL inhibitor for use according to claim 1, 20 wherein said 13. Composition for use according to claim 11 further for treatment is for the reduction of cancer cells or the use in a hormone therapy. prevention of cancer progression in the tissue of origin of said primary breast or mammary cancer, 14. Composition for use according to claim 13, wherein wherein the RANKL inhibitor is an anti-RANKL anti- 25 the hormone therapy is contraception or a hormone body. replacement therapy.

4. A RANKL inhibitor for use according to claims 1 or 15. Kit comprising a chemotherapeutic agent and a 2, wherein the RANKL inhibitor for use in a treatment or prevention patient has received a treatment with female sexual 30 of primary breast or mammary cancer, wherein the hormones, preferably a hormone replacement ther- RANKL inhibitor is an anti-RANKL antibody. apy, preferably with progesterone or a progestin, or a hormone contraceptive, wherein the RANKL inhibitor is an anti-RANKL antibody. Patentansprüche 35 5. A RANKL inhibitor for use according to claims 1 or 2, 1. RANKL-Inhibitor zur Verwendung in einem Verfah- in combination ren zur Prävention eines primären Brust- bzw. Mam- with a therapy of female sexual hormones, in partic- makarzinoms in einem Subjekt, umfassend das an ular a hormone replacement therapy or for use in das Subjekt erfolgende Verabreichen einer thera- combination with a hormone contraceptive, in par- 40 peutisch wirksamen Menge des RANKL-Inhibitors, ticular to reduce the risk of developing cancer, wobei es sich bei dem RANKL-Inhibitor um einen wherein the RANKL inhibitor is an anti-RANKL anti- Anti-RANKL-Antikörper handelt. body. 2. RANKL-Inhibitor zur Verwendung in einem Verfah- 6. Inhibitor for use according to any one of claims 1 to 45 ren zur Behandlung eines primären Brust- bzw. 5, characterized in that the cancer is of epithelial Mammakarzinoms in einem Subjekt, umfassend das cell origin. an das Subjekt erfolgende Verabreichen einer the- rapeutisch wirksamen Menge eines RANKL-Inhibi- 7. Inhibitor for use according to any one of claims 1 to tors zur Reduktion der Krebszellen oder zur Präven- 6 in combination with or for priming a further anti- 50 tion der Progression der Krebserkrankung in dem cancer therapy, preferably a radiation- or chemo- Ursprungsgewebe des primären Brust- bzw. Mam- therapy. makarzinoms, wobei es sich bei dem RANKL-Inhi- bitor um einen Anti-RANKL-Antikörper handelt. 8. Inhibitor for use according to any one of claims 1 to 7 for treating or preventing a primary cancer or for 55 3. RANKL-Inhibitor zur Verwendung nach Anspruch 1, treating or preventing reoccurence of a cancer. wobei die Behandlung zur Reduktion der Krebszel- len oderzur Prävention der Progression der Krebser- 9. Inhibitor for use according to any one of claims 1 to krankung in dem Ursprungsgewebe des primären

21 39 EP 2 618 833 B1 40

Brust- bzw. Mammakarzinoms vorgesehen ist, wo- 12. Zusammensetzung zur Verwendung nach Anspruch bei es sich bei dem RANKL-Inhibitor um einen Anti- 11, des Weiteren umfassend einen pharmazeutisch RANKL-Antikörper handelt. verträglichen Diluenten, Träger, Lösungsvermittler, Emulgator, Konservierungsstoff und/oder Hilfsstoff. 4. RANKL-Inhibitor zur Verwendung nach Anspruch 1 5 oder 2, wobei das Subjekt bereits eine Behandlung 13. Zusammensetzung zur Verwendung nach Anspruch mit weiblichen Sexualhormonen, bevorzugt eine 11, die des Weiteren zur Verwendung im Rahmen Hormonersatztherapie, bevorzugt mit Progesteron einer Hormontherapie vorgesehen ist. oder einem Progestin, oder mit einem hormonellen Kontrazeptivum erhalten hat, wobei es sich bei dem 10 14. Zusammensetzung zur Verwendung nach Anspruch RANKL-Inhibitor um einen Anti-RANKL-Antikörper 13, wobei es sich bei der Hormontherapie um Kon- handelt. trazeption oder eine Hormonersatztherapie handelt.

5. RANKL-Inhibitor zur Verwendung nach Anspruch 1 15. Kit, umfassend ein chemotherapeutisches Agens oder 2 in Kombination mit einer Therapie mit weib- 15 und einen RANKL-Inhibitor, zur Verwendung bei der lichen Sexualhormonen, insbesondere einer Hor- Behandlung oder Prävention eines primären Brust- monersatztherapie, oder zur Verwendung in Kombi- bzw. Mammakarzinoms, wobei es sich bei dem nation mit einem hormonellen Kontrazeptivum, ins- RANKL-Inhibitor um einen Anti-RANKL-Antikörper besondere zur Reduktion des Risikos der Entwick- handelt. lung einer Krebserkrankung, wobei es sich bei dem 20 RANKL-Inhibitor um einen Anti-RANKL-Antikörper handelt. Revendications

6. Inhibitor zur Verwendung nach einem der Ansprüche 1. Inhibiteur de RANKL destiné à une utilisation visant 1 bis 5,dadurch gekennzeichnet, dass die25 à empêcher un cancer primaire mammaire ou du Krebserkrankung ihren Ursprung in Epithelzellen sein chez un sujet, comprenant une étape consistant hat. à administrer une quantité efficace de manière thé- rapeutique d’un inhibiteur de RANKL à ladite 7. Inhibitor zur Verwendung nach einem der Ansprüche patiente ; 1 bis 6 in Kombination mit oder zum Zwecke des30 dans lequel l’inhibiteur de RANKL est un anticorps Vorbereitens bzw. Primings einer weiteren Anti- anti-RANKL. krebstherapie, bevorzugt einer Strahlen- oder Che- motherapie. 2. Inhibiteur de RANKL destiné à une utilisation dans un procédé visant à traiter un cancer primaire mam- 8. Inhibitor zur Verwendung nach einem der Ansprüche 35 maire ou du sein chez un sujet, comprenant une éta- 1 bis 7 zur Behandlung oder Prävention einer primä- pe consistant à administrer une quantité efficace de renKrebserkrankung oder zur Behandlungoder Prä- manière thérapeutique d’un inhibiteur de RANKL à vention eines Rezidivs einer Krebserkrankung. ladite patiente, de façon à réduire les cellules can- céreuses ou à empêcher la progression du cancer 9. Inhibitor zur Verwendung nach einem der Ansprüche 40 dans le tissu qui est à l’origine dudit cancer primaire 1 bis 8, wobei es sich bei dem RANKL-Inhibitor um mammaire ou du sein ; Denosumab handelt. dans lequel l’inhibiteur de RANKL est un anticorps anti-RANKL. 10. Inhibitor zur Verwendung nach einem der Ansprüche 1bis 9, wobei dieKrebserkrankung kanzeröse Zellen 45 3. Inhibiteur de RANKL destiné à une utilisation selon umfasst, die ausgewählt sind aus einer Brustgewe- la revendication 1, dans lequel ledit traitement vise be- oder Brustkarzinomzelle und/oder einer Epithel- à réduire les cellules cancéreuses ou à empêcher la oder Basalzelle und/oder einer Krebsstammzelle. progression du cancer dans le tissu qui est à l’origine dudit cancer primaire mammaire ou du sein, dans 11. Pharmazeutische Zusammensetzung, umfassend 50 lequel l’inhibiteur de RANKL est un anticorps anti- einen RANKL-Inhibitor und ein Hormon, das aus Ös- RANKL. trogen und Progesteron oder einem Progestin oder einem hormonellen Kontrazeptivum ausgewählt ist, 4. Inhibiteur de RANKL destiné à une utilisation selon zur Verwendung bei der Behandlung oder Präventi- la revendication 1 ou la revendication 2, dans lequel on eines primären Brust- bzw. Mammakarzinoms, 55 la patiente a reçu un traitement avec des hormones wobei es sich bei dem RANKL-Inhibitor um einen sexuelles femelles, de préférence une thérapie hor- Anti-RANKL-Antikörper handelt. monale de remplacement, de préférence avec une progestérone ou un progestatif, ou un contraceptif

22 41 EP 2 618 833 B1 42

hormonal ; vendication 11, destinée en outre à une utilisation dans lequel l’inhibiteur de RANKL est un anticorps dans une thérapie hormonale. anti-RANKL. 14. Composition destinée à une utilisation selon la re- 5. Inhibiteur de RANKL destiné à une utilisation selon 5 vendication 13, dans laquelle la thérapie hormonale la revendication 1 ou la revendication 2, en associa- est une contraception ou une thérapie hormonale de tion avec une thérapie d’hormones sexuelles femel- remplacement. les, en particulier une thérapie hormonale de remplacement, ou à une utilisation en association avec 15. Kit comprenant un agent chimiothérapeutique et un un contraceptif hormonal, en particulier de façon à 10 inhibiteur de RANKL, destiné à une utilisation dans réduire le risque de développer un cancer ; un traitement ou dans une prévention d’un cancer dans lequel l’inhibiteur de RANKL est un anticorps primaire mammaire ou du sein ; anti-RANKL. dans lequel l’inhibiteur de RANKL est un anticorps anti-RANKL. 6. Inhibiteur destiné à une utilisation selon l’une quel- 15 conque des revendications 1 à 5, caractérisé en ce que le cancer est d’origine cellule épithéliale.

7. Inhibiteur destiné à une utilisation selon l’une quel- conque des revendications 1 à 6, en association20 avec une autre thérapie anticancéreuse, ou pour amorcer celle-ci, de préférence une radiothérapie ou une chimiothérapie.

8. Inhibiteur destiné à une utilisation selon l’une quel- 25 conque des revendications 1 à 7, destiné à traiter ou à empêcher un cancer primaire, ou à traiter ou à empêcher la récidive d’un cancer.

9. Inhibiteur destiné à une utilisation selon l’une quel- 30 conque des revendications 1 à 8 ; danslequel l’inhibiteurde RANKL est leDénosumab.

10. Inhibiteur destiné à une utilisation selon l’une quel- conque des revendications 1 à 9, dans lequel ledit 35 cancer comprend des cellules cancéreuses sélec- tionnées dans le groupe constitué par une cellule du tissu du sein ou une cellule de cancer du sein, et / ou une cellule épithéliale ou une cellule basale et / ou une cellule souche cancéreuse. 40

11. Composition pharmaceutique comprenant un inhibiteur de RANKL et une hormone sélectionnée dans le groupe constitué par un oestrogène et une pro- gestérone ou une progestine ou un contraceptif hor- 45 monal, destinées à une utilisation visant à traiter ou à empêcher un cancer primaire mammaire ou du sein ; dans lequel l’inhibiteur de RANKL est un anticorps anti-RANKL. 50

12. Composition destinée à une utilisation selon la revendication 11, comprenant en outre un diluant, un vecteur, un solubilisant, un émulsifiant, un conser- vateur et / ou un adjuvant, acceptables de manière 55 pharmaceutique.

13. Composition destinée à une utilisation selon la re-

23 EP 2 618 833 B1

24 EP 2 618 833 B1

25 EP 2 618 833 B1

26 EP 2 618 833 B1

27 EP 2 618 833 B1

28 EP 2 618 833 B1

29 EP 2 618 833 B1

30 EP 2 618 833 B1

31 EP 2 618 833 B1

32 EP 2 618 833 B1

33 EP 2 618 833 B1

34 EP 2 618 833 B1

35 EP 2 618 833 B1

36 EP 2 618 833 B1

37 EP 2 618 833 B1

38 EP 2 618 833 B1

39 EP 2 618 833 B1

40 EP 2 618 833 B1

41 EP 2 618 833 B1

42 EP 2 618 833 B1

43 EP 2 618 833 B1

REFERENCES CITED IN THE DESCRIPTION

This list of references cited by the applicant is for the reader’s convenience only. It does not form part of the European patent document. Even though great care has been taken in compiling the references, errors or omissions cannot be excluded and the EPO disclaims all liability in this regard.

Patent documents cited in the description

• WO 9953942 A [0001] • US 7411050 B [0011] • US 6017729 A [0003] • EP 2003203 A1 [0011] • EP 0873998 A [0003] • US 2004167072 A [0011] • EP 0911342 A [0003] • US 20020162126 A [0028] • US 5843678 A [0003] • US 20020173478 A [0028] • WO 9846751 A [0003] • WO 2005028633 A [0028] • WO 9854201 A [0003] • WO 02055692 A [0029] • US 2008107597 A [0004] [0011] [0030] • WO 02055693 A [0029] • WO 2010022120 A1 [0004] [0011] • EP 1144623 B1 [0029] • WO 02098362 A [0005] [0015] • WO 03074654 A [0029] • WO 2007128564 A [0006] • WO 9312227 A [0030] • US 6740522 B [0011] • EP 10178346 A [0067]

Non-patent literature cited in the description

• ELLIS et al. Breast Cancer Research and Treatment, • TARUTANI, M. et al. Tissue-specific knockout of the 2009, vol. 118 (1), 81-87 [0001] mouse Pig-a gene reveals important roles for GPI-an- • TOMETSKO et al. Bone, 2010, vol. 47, 325-326 chored proteins in skin development. Proc Natl Acad [0001] Sci USA, 1997, vol. 94, 7400-5 [0066] • CANON et al. Bone, 2007, vol. 40 (6), 145 [0001] • GAREUS, R. et al. Normal epidermal differentiation • CANON et al. Europ. J. of Cancer, 2006, vol. 4 (12), but impaired skin-barrier formation upon keratinoc- 166-167 [0001] yte-restricted IKK1 ablation. Nat Cell Biol, 2007, vol. • CANON et al. Cancer Treatment Reviews, 2008, vol. 9, 461-9 [0066] 34, 60 [0001] • ALIPRANTIS, A.O. et al. NFATcl in mice represses • TOMETSKO et al. J. Bone and Mineral Res., 2006, osteoprotegerin during osteoclastogenesis and dis- vol. 21, 346 [0001] sociates systemic osteopenia from inflammation in • HOLLAND et al. Cancer Biology and Therapy, 2010, cherubism. J Clin Invest, 2008, vol. 118, 3775-89 vol. 9 (7), 539-550 [0001] [0066] • AKIYAMA et al. J. Pharmacy and Pharmacology, • ALDAZ, C.M. ; LIAO, Q.Y. ; LABATE, M. ; JOHN- 2010, vol. 62 (4), 470-476 [0001] STON, D.A. Medroxyprogesterone acetate acceler- • ZHANG et al. J. Clin. Invest., 2001, vol. 107 (10), ates the development and increases the incidence of 1235-1244 [0001] mouse mammary tumors induced by dimethylben- • LEE et al. Bone, 2010, vol. 48 (1), 88-95 [0001] zanthracene. Carcinogenesis, 1996, vol. 17, 2069-72 • LEIBBRANDT et al. Europ. J. of Clin. Invest., 2009, [0066] vol. 39 (10), 842-850 [0001] • CAO, Y. ; LUO, J.L. ; KARIN, M. IkappaB kinase al- • WU et al. J. Bone and Mineral Res., 2005, vol. 20 pha kinase activity is required for self-renewal of (1), 107-116 [0001] ErbB2/Her2-transformed mammary tumor-initiating • BHARTI et al. Apoptosis, 2004, vol. 9 (6), 677-690 cells. Proc Natl Acad Sci USA, 2007, vol. 104, [0001] 15852-7 [0066] • MOLYNEUX et al. J. Clin. Invest., 2010, vol. 120 (9), • ROBINSON, G.W. ; HENNIGHAUSEN, L. Inhibins 3310-3325 [0001] and activins regulate mammary epithelial cell differ- • PARK et al. J. Korean Med. Sci., 2003, vol. 18 (4), entiation through mesenchymal-epithelial interac- 541-546 [0001] tions. Development, 1997, vol. 124, 2701-8 [0066] • HANADA, R. et al. Central control of fever and female • FATA, J.E. et al. The osteoclast differentiation factor body temperature by RANKL/RANK. Nature, 2009, osteoprotegerin-ligand is essential for mammary vol. 462, 505-9 [0066] gland development. Cell, 2000, vol. 103, 41-50 [0066]

44 EP 2 618 833 B1

• LACEY, D.L. et al. Osteoprotegerin ligand is a cy- • ASSELIN-LABAT, M.L. et al. Control of mammary tokinethat regulates osteoclast differentiation and ac- stem cell function by steroid hormone signalling. Na- tivation. Cell, 1998, vol. 93, 165-76 [0066] ture [0066] • STINGL, J. et al. Purification and unique properties • JOSHI, P.A. et al. Progesterone induces adult mam- of mammary epithelial stem cells. Nature, 2006, vol. mary stem cell expansion. Nature [0066] 439, 993-7 [0066] • KIM, N.S. et al. Receptor activator of NF-kappaB lig- • SHACKLETON, M. et al. Generation of a functional and regulates the proliferation of mammary epithelial mammary gland from a single stem cell.Nature, cells via Id2. Mol Cell Biol, 2006, vol. 26, 1002-13 2006, vol. 439, 84-8 [0066] [0066] • DONTU, G. et al. In vitro propagation and transcrip- • EWAN, K.B. et al. Transforming growth factor-betal tional profiling of human mammary stem/progenitor mediates cellular response to DNA damage in situ. cells. Genes Dev, 2003, vol. 17, 1253-70 [0066] Cancer Res, 2002, vol. 62, 5627-31 [0066] • FREEDMAN, V.H. ; SHIN, S.I. Cellular tumorigenic- • PECE, S. et al. Biological and molecular heteroge- ity in nude mice: correlation with cell growth in neity of breast cancers correlates with their cancer semi-solid medium. Cell, 1974, vol. 3, 355-9 [0066] stem cell content. Cell, vol. 140, 62-73 [0066] • JONES, D.H. et al. Regulation of cancer cell migra- • HANAHAN, D. ; WEINBERG, R.A. The hallmarks of tion and-bone metastasis by RANKL. Nature, 2006, cancer. Cell, 2000, vol. 100, 57-70 [0066] vol. 440, 692-6 [0066]

45

Web Analytics