- Email: [email protected]
BRCA genes: lessons learned from experimental and clinical cancer
Annals of Oncology 22 (Supplement 1): i7–i10, 2011 doi:10.1093/annonc/mdq659
symposium article BRCA genes: lessons learned from experimental and clinical cancer F. Muggia1*, T. Safra1 & L. Dubeau2 1
New York University Cancer Institute, New York; 2Department of Pathology, USC/Norris Comprehensive Cancer Center, Los Angeles, USA
The assignation of mutations within specific genes to a familial predisposition to develop breast and ovarian cancers in the 1990s was followed by a flurry of laboratory and clinical studies. Several groups set out to identify the function of BRCA genes, with work in the laboratory of David Livingston being instrumental in recognizing the roles of BRCA1 and BRCA2 in pathways involved in repair of DNA damage. In the subsequent decade, BRCA1 has emerged as playing a central role in the repair of double-strand breaks by homologous recombination, with BRCA2 and a number of accompanying proteins (including the genetically well studied Fanconi proteins) in a supportive role for the recognition of such DNA damage. The function of these proteins has been the subject of seminal publications, reviewed at this symposium by Daniel Silver [1–4]. Concomitantly, clinical and epidemiologic studies set out to assign cancer risks to families identified to harbor BRCA mutations, and to provide guidance to the afflicted individuals. Most prominent were the risks involved in the development of breast and ovarian cancer, with a number of other inherited cancers arising in the pancreas, prostate and colon also emerging within certain families. Preventive strategies relying on risk-reducing surgery were developed for known carriers of deleterious mutations, while also testing the usefulness of surveillance. Such studies led to detection of early neoplastic and preneoplastic lesions, exemplified by those not previously recognized in the Fallopian tubes [5, 6]. As a result, we now *Correspondence to: Franco Muggia, NYU Cancer Institute, 550 First Avenue, New York, NY 10016, USA. Tel: +1-212-263-6485; Fax: +1-212-263-8210; E-mail: [email protected]
acknowledge Mullerian epithelium at extrauterine sites such as fimbria of the Fallopian tubes and others as a site of origin of high-grade serous carcinomas [7]. The clinical significance of these findings is being pursued, primarily though studies of pathologic specimens obtained at risk-reducing surgery. As a result of these evolving notions of gene function linked to hereditary cancer risks, a number of animal models have been developed to further investigate the role of specific genes in the pathogenesis of these cancers. We describe below selected examples of breast and ovarian cancer models, and discuss how they contribute to our knowledge of their pathogenesis and to their clinical relevance.
conditional BRCA and p53 knockout in the murine breast This model developed at the Netherlands Cancer Institute provided insight into the therapeutic issues surrounding mammary gland adenocarcinomas lacking BRCA function [8]. Most notably, these tumors were very sensitive to DNAdamaging agents, and were utilized to investigate reasons for eventual resistance to drugs such as anthracyclines. The most useful agents were the platinums and recurrences were always sensitive to the reintroduction of cisplatin [9–11]. The model has also been hailed as a useful representation of ‘triplenegative’ breast cancers, providing a contrast to many murine models that are considered to be driven by estrogenic stimulation via estrogen receptors (ERs)—recently identified as ‘luminal’, ER-positive subtypes [12]. In brief, this model was created to develop ‘spontaneous tumors’ arising in mice by introducing conditional deletions in tumor suppressor genes. Because they are non-immunogenic,
ª The Author 2011. Published by Oxford University Press on behalf of the European Society for Medical Oncology. All rights reserved. For permissions, please email: [email protected]
symposium article
introduction
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 23, 2015
Advances in the study of BRCA1 and BRCA2 gene functions have relied on the development of animal models for seeking to explore further what we have learned from the human disease. Specifically, mouse models of a ‘triplenegative’ breast cancer (utilizing conditional knockout of BRCA1 and p53 in the breast), of an endometrioid ovarian cancer (based on oncogenic kras and loss of function of pten), and of anatomic and functional consequences of BRCA1 mutations in granulosa cells, have led to further inquiry into the pathogenesis and therapeutic consequences of genetic alterations. A striking susceptibility of these murine malignancies to platinum drugs has emerged, providing further confidence in their relevance to the human disease. In addition to these models, the pathogenesis of highgrade serous disease derived from risk-reducing surgeries in mutation carriers has pointed to a role of mutations in p53 commonly encountered in tubal intraepithelial carcinomas.
symposium article
endometrioid ovarian cancer model In this particular model, the focus was on altering gene pathways known to be implicated in a subtype of ovarian cancer of ovarian surface epithelial cells [16, 17]. The discovery of frequent somatic PTEN mutations and loss of heterozygosity at the 10q23 PTEN locus in endometrioid ovarian cancer implicates a key role for PTEN in the etiology of this epithelial ovarian cancer subtype. Similarly, K-RAS oncogene is also mutated in endometrioid ovarian cancer, albeit at a lower frequency. This was achieved by introducing a recombinant adenoviral vector expressing Cre recombinase into the bursal cavity that encloses the ovaries in genetically engineered mouse models carrying the conditional oncogenic K-ras and/or loss of function tumor suppressor Pten mutation. Introduction of
i8 | Muggia et al.
adenoviral Cre resulted in activation of the oncogenic KrasG12D mutation within cells of the ovarian surface epithelium, which in turn gave rise to benign ovarian endometriotic-like lesions with simple endometrioid glandular morphology [17]. Furthermore, when adenoviral Cre was introduced into the uterine tubal junction, activation of oncogenic K-ras was achieved within uterine and tubal cells, which acquired a growth advantage and were able to implant and expand at ectopic peritoneal sites [17]. As a result, some mice developed pelvic endometriosis, making this the first model of de novo endometriosis. When oncogenic K-ras mutation was further combined with loss of function Pten expression in LSL–KrasG12D/+; Ptenloxp/loxp mice, frank endometrioid carcinomas arose from the ovaries and metastasized to pelvic and peritoneal locations similar to human tumors [18]. In addition to its histological characteristics, several observations such as high levels of ER immunostaining, activation of Akt and mTOR pathways, peritoneal spread, eventual development of ascites and therapeutic benefit from cisplatin [19] suggested recapitulation of molecular and clinical findings. Adding interest, subsequent work has expanded on features of ‘side populations’ that arise under pressure of treatment, and may be interrogated further for ‘stem cell’ features [20]. While this model does not address specifically the role of BRCA genes in ovarian tumorigenesis, it links the emergence of endometrioid cancers to alterations in K-ras oncogenic signals coupled with Pten loss within this specialized epithelium. Of interest, BRCA1 dysfunction has been recently linked with PTEN mutations in basal-like breast carcinogenesis [21]. The Dinulescu model, therefore, provides further substantiation of how these pathway alterations lead to phenotypic expression (including sensitivity to cisplatin) in tumors arising from the female genital tract.
pathogenesis of serous ovarian, tubal and primary peritoneal cancers A mouse model for the predominant subtype of extrauterine neoplasms of Mullerian epithelial origin has not been yet been developed. Nevertheless, risk-reducing surgery in BRCA1 and BRCA2 mutation carriers has identified a likely precursor to high-grade serous carcinomas: the tubal intraepithelial carcinoma [22]. As noted earlier, TP53 mutations are a recognizable step in this evolution [6]—a situation somewhat analogous to the pathogenesis of BRCA1 (but not BRCA2)related breast cancer (refer to Mangia et al., this supplement [23])
granulosa cell dysfunction due to BRCA loss of function Given that the menstrual cycle is the strongest risk factor for non-familial ovarian cancers, Dubeau and co-workers [24] hypothesized that ovarian cancer predisposition in BRCA1 mutation carriers may be driven, at least in part, by consequences of such mutations on menstrual cycle regulation. They introduced a Brca1 knockout in granulosa cells, which play a central role in menstrual cycle control, using a mouse
Volume 22 | Supplement 1 | January 2011
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 23, 2015
these tumors could prove useful in allowing drug resistance studies. A mammary tumor arising in K14 Cre; Brca1flox/flox; p53 flox/flox mice had knocked out p53 and BRCA function within the breast. Histologically, tumors arising in the breast of these mice are poorly differentiated and lack ER, progesterone receptor and human epidermal growth factor receptor 2 (Her2) expression—hence, the designation ‘triple negative’. The sensitivity of these tumors to cisplatin, coupled with the recognition that BRCA1 mutation carriers often developed the ‘triple-negative’ breast cancer phenotype, spurred clinical trials of platinum compounds directed towards patients with metastatic disease fulfilling the criteria for such designation. Some encouraging therapeutic results have been reported [8], but additional refinements in the characterization of breast cancers within this phenotype will be needed to provide more convincing evidence of the usefulness of platinum-based therapeutic strategies against breast cancers arising in a BRCA mutation background [13, 14]. Platinum-based therapies, of course, have formed the cornerstone of treatment of epithelial ovarian cancer, with the added emerging feature that this treatment is more effective stage-by-stage in BRCA mutation carriers than in those cancers not arising in a hereditary background [15]. Interfering with the poly(ADP-ribose) polymerase (PARP) pathway involved in base excision repair leads to marked differential cytotoxicity of BRCA-deficient tumors versus wild-type tumors, which have an intact homologous recombination (HR) pathway of DNA repair of double-strand breaks [16]. Concepts emanating in part from this model have influenced the design of clinical trials that demonstrated the usefulness of adding a drug with presumed PARP-inhibitor properties to a carboplatin–gemcitabine doublet: patients previously untreated for metastatic breast cancer were simply selected for randomization to PARP-1 inhibitor or no PARP-1 inhibitor added to chemotherapy on the basis of being ‘triple negative’. A remarkable improvement in response rates and progressionfree survival was documented and this has led to a large phase III trial as a pivotal registration trial seeking approval for inhibitors of PARP-1. The trial also encourages the extrapolation of certain clinical subtypes as being not only reflective of lack of BRCA function (linked to inherited deleterious BRCA mutations) but also arising from other causes of genetic instability and epigenetic silencing of these pathways.
Annals of Oncology
Annals of Oncology
model. Noteworthy findings were the emergence of serous cystadenomas. Results presented at the Bari symposium demonstrated a pattern of menstrual/estrus dysfunction when these engineered mice were compared with the wild-type species. Clinical validation of these findings hopefully will be sought. A pattern of menstrual dysfunction occasionally occurring in women with a strong family history of breast and ovarian neoplasia has been reported [25]. Also, it is attractive to be able to pursue further a hypothesis that links BRCA gene dysfunction with a propensity towards breast and ovarian neoplasia.
discussion
(i) Contribution of other genes in the eventual development of neoplasia related to BRCA1 and BRCA2 mutations in hereditary carriers. (ii) What accounts for the susceptibility of the breast and ovary to neoplastic transformation in relation to these inherited mutations. (iii) The relationship of estrogens and estrogen–receptor signaling in the pathogenetic steps leading to breast and extrauterine Mullerian carcinogenesis. (iv) Characterization of premalignant lesions in these tissues. (v) Approaches to interfering with the emergence of an eventual malignancy (e.g. chemo-prevention) in the presence of known hereditary mutations (vi) Characterization of epigenetic changes that lead to phenotypic expression similar to the hereditary type. (vii) Future clues to some of these questions will emerge not only from these models but also from epidemiologic
Volume 22 | Supplement 1 | January 2011
study and clinical interventions. Several articles in this supplement deal with such studies, particularly as they relate to determining and/or altering breast cancer risks [23, 26–29].
disclosures The authors declare no conflict of interest.
references 1. Chen JJ, Silver D, Cantor S et al. BRCA1, BRCA2, and Rad51 operate in a common DNA damage response pathway. Cancer Res 1999; 59(7 Suppl): 1752s–1756s. 2. Scully R, Livingston DM. In search of the tumour-suppressor functions of BRCA1 and BRCA2. Nature 2000; 408: 429–432. 3. Silver DP, Dimitrov SD, Feunteun J et al. Further evidence for BRCA1 communication with the inactive X chromosome. Cell 2007; 128: 991–1002. 4. Livingston DM, Silver DP. Cancer: crossing over to drug resistance Nature 2008; 451: 1111–1115. 5. Crum CP, Drapkin R, Miron A et al. The distal Fallopian tube: a model for pelvic serous carcinogenesis. Curr Opin Obstet Gynec 2007; 19: 3–9. 6. Levanon K, Ng V, Piao HY et al. Primary ex vivo cultures of human Fallopian tube epithelium as a model for serous ovarian carcinogenesis. Oncogene 2010; 29: 1103–1113. 7. Dubeau L. The cell of origin of ovarian epithelial tumours. Lancet 2008; 9: 1191–1197. 8. Fasano J, Muggia F. Breast cancer arising in a BRCA-mutated background: therapeutic implications from an animal model and drug development. Ann Oncol 2008; 4: 609–614. 9. Borst P, Rottenberg S, Jonkers J. How do real tumors become resistant to cisplatin? Cell Cycle 2008; 7: 1353–1359. 10. Muggia F. BRCA-deficient animal models and cisplatin resistance. Ann Oncol 2009; 20: 962. 11. Rottenberg S, Mygren AOH, Pajic M et al. Selective induction of chemotherapy resistance of mammary tumors in a conditional mouse model for hereditary breast cancer. Proc Natl Acad Sci USA 2007; 104: 12117–12122. 12. Perou CM, Sorlie T, Eisen MB et al. Molecular portraits of human breast tumors. Nature 2000; 406: 747–752. 13. Chappuis PO, Goffin J, Wong N et al. A significant response to neoadjuvant chemotherapy in BRCA1/2 related breast cancer. J Med Genet 2002; 39: 608–610. 14. Kennedy RD, Quinn JE, Mullan PB et al. The role of BRCA1 in the cellular response to chemotherapy. J Natl Cancer Inst 2004; 96: 1659–1668. 15. Muggia F. Platinum compounds 30 years after the introduction of cisplatin: implications for the treatment of ovarian cancer. Gynecol Oncol 2009; 112: 275–281. 16. Ashworth AA. A synthetic lethal therapeutic approach: poly(ADP) ribose polymerase inhibitors for the treatment of cancers deficient in DNA double-strand break repair. J Clin Oncol 2008; 26: 3785–3790. 17. Dinulescu DM, Ince TA, Quade BJ et al. Role of k-ras and Pten in the development of mouse models of endometriosis and endometrioid ovarian cancer. Nat Med 2005; 11: 63–70. 18. Martzuk MM. News and views: gynecologic diseases get their genes. Nat Med 2005; 11: 24–26. 19. Paraskar AS, Shivani S, Chin KT et al. Harnessing structure-activity relationship to engineer a cisplatin nanoparticle for enhanced antitumor efficacy. Proc Natl Acad Sci 2010; 107: 12435–12440. 20. Szotek PP, Pieretti-VonMarckeMasiakos R, Masiakos PT et al. Ovarian cancer side population defines cells with stem cell-like characteristics and Mullerian inhibiting substance responsiveness. Proc Natl Acad Sci USA 2006; 103: 11154–11159.
doi:10.1093/annonc/mdq659 | i9
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 23, 2015
We have expanded on preclinical models that have recently explored the molecular changes involved in the pathogenesis of breast and ovarian cancers. A key link between BRCA gene dysfunction and sensitivity to platinum drugs has emerged from these models—emphasizing the importance of defective HR in therapeutic responses to cisplatin. Therapeutic interest has been further stimulated by the introduction of PARP-1 inhibitors: these drugs not only potentiated the effect of DNAdamaging drugs that cause single-strand breaks, but were noted particularly effective in the absence of HR. A corollary to these findings has been the hypothesis that sporadically occurring cancers sharing the phenotype associated with BRCA mutation carriers may result as a consequence of dysfunction in these genes arising from epigenetic changes or gene deletions. As these associations come to light, the emerging questions converge towards pathogenetic pathways leading to such breast and ovarian tumorigenesis. A further dimension has been added by experimental findings indicating that granulosa cell dysfunction may have profound consequences in a mouse model deleting Brca1 in these cells. The animal models described earlier may be viewed as just an initial step in understanding the pathogenesis of human disease. Such models may allow us to look into additional preventive and therapeutic interventions both in models and in the clinic. Specifically, issues that need to be explored include the following.
symposium article
symposium article 21. Saal LH, Gruvberger-Sall SK, Persson C et al. Recurrent gross mutations of the PTEN tumor suppressor gene in breast cancers with deficient DSB repair. Nat Genet 2008; 40: 102–107. 22. Vang R, Shir I-M, Kurman RJ. Ovarian low-grade and high-grade serous carcinoma: pathogenesis, clinicopathologic and molecular biologic features, and diagnostic problems. Adv Anat Pathol 2009; 16: 267–282. 23. Mangia A, Malfettone A, Simoni G, Darvishian F. Old and new concepts in histopathological characterization of familial breast cancer. Ann Oncol 2011; 22(Suppl1): mdq662. 24. Chodankar R, Kwang S, Sangiorgi F et al. Cell-nonautonomous induction of ovarian and uterine serous cystadenomas in mice lacking a functional Brca1 in ovarian granulosa cells. Curr Biol 2005; 15: 561–565.
Annals of Oncology
25. Matalliotakis IM, Cakmak H, Fragouli YG et al. Epidemiological characteristics in women with and without endometriosis in the Yale series. Arch Gynecol Obstet 2008; 277: 389–393. 26. Milne RL, Antoniou AC. Genetic modifiers of cancer risk for BRCA1 and BRCA2 mutation carriers. Ann Oncol 2011; 22 (Suppl 1): mdq660. 27. Radice P, De Summa S, Caleca L, Tommasi S. Unclassified variants in BRCA genes: guidelines for interpretation. Ann Oncol 2011; 22 (Suppl 1): mdq661. 28. Warner E. Impact of MRI surveillance and breast cancer detection in young women with BRCA mutations. Ann Oncol 2011; 22 (Suppl 1): mdq665. 29. Smith J, Axelrod D, Singh B, Kleinberg D. Prevention of breast cancer: the case for studying inhibition of IGF-1 actions. Ann Oncol 2011; 22 (Suppl 1): mdq666.
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 23, 2015
i10 | Muggia et al.
Volume 22 | Supplement 1 | January 2011
symposium article BRCA genes: lessons learned from experimental and clinical cancer F. Muggia1*, T. Safra1 & L. Dubeau2 1
New York University Cancer Institute, New York; 2Department of Pathology, USC/Norris Comprehensive Cancer Center, Los Angeles, USA
The assignation of mutations within specific genes to a familial predisposition to develop breast and ovarian cancers in the 1990s was followed by a flurry of laboratory and clinical studies. Several groups set out to identify the function of BRCA genes, with work in the laboratory of David Livingston being instrumental in recognizing the roles of BRCA1 and BRCA2 in pathways involved in repair of DNA damage. In the subsequent decade, BRCA1 has emerged as playing a central role in the repair of double-strand breaks by homologous recombination, with BRCA2 and a number of accompanying proteins (including the genetically well studied Fanconi proteins) in a supportive role for the recognition of such DNA damage. The function of these proteins has been the subject of seminal publications, reviewed at this symposium by Daniel Silver [1–4]. Concomitantly, clinical and epidemiologic studies set out to assign cancer risks to families identified to harbor BRCA mutations, and to provide guidance to the afflicted individuals. Most prominent were the risks involved in the development of breast and ovarian cancer, with a number of other inherited cancers arising in the pancreas, prostate and colon also emerging within certain families. Preventive strategies relying on risk-reducing surgery were developed for known carriers of deleterious mutations, while also testing the usefulness of surveillance. Such studies led to detection of early neoplastic and preneoplastic lesions, exemplified by those not previously recognized in the Fallopian tubes [5, 6]. As a result, we now *Correspondence to: Franco Muggia, NYU Cancer Institute, 550 First Avenue, New York, NY 10016, USA. Tel: +1-212-263-6485; Fax: +1-212-263-8210; E-mail: [email protected]
acknowledge Mullerian epithelium at extrauterine sites such as fimbria of the Fallopian tubes and others as a site of origin of high-grade serous carcinomas [7]. The clinical significance of these findings is being pursued, primarily though studies of pathologic specimens obtained at risk-reducing surgery. As a result of these evolving notions of gene function linked to hereditary cancer risks, a number of animal models have been developed to further investigate the role of specific genes in the pathogenesis of these cancers. We describe below selected examples of breast and ovarian cancer models, and discuss how they contribute to our knowledge of their pathogenesis and to their clinical relevance.
conditional BRCA and p53 knockout in the murine breast This model developed at the Netherlands Cancer Institute provided insight into the therapeutic issues surrounding mammary gland adenocarcinomas lacking BRCA function [8]. Most notably, these tumors were very sensitive to DNAdamaging agents, and were utilized to investigate reasons for eventual resistance to drugs such as anthracyclines. The most useful agents were the platinums and recurrences were always sensitive to the reintroduction of cisplatin [9–11]. The model has also been hailed as a useful representation of ‘triplenegative’ breast cancers, providing a contrast to many murine models that are considered to be driven by estrogenic stimulation via estrogen receptors (ERs)—recently identified as ‘luminal’, ER-positive subtypes [12]. In brief, this model was created to develop ‘spontaneous tumors’ arising in mice by introducing conditional deletions in tumor suppressor genes. Because they are non-immunogenic,
ª The Author 2011. Published by Oxford University Press on behalf of the European Society for Medical Oncology. All rights reserved. For permissions, please email: [email protected]
symposium article
introduction
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 23, 2015
Advances in the study of BRCA1 and BRCA2 gene functions have relied on the development of animal models for seeking to explore further what we have learned from the human disease. Specifically, mouse models of a ‘triplenegative’ breast cancer (utilizing conditional knockout of BRCA1 and p53 in the breast), of an endometrioid ovarian cancer (based on oncogenic kras and loss of function of pten), and of anatomic and functional consequences of BRCA1 mutations in granulosa cells, have led to further inquiry into the pathogenesis and therapeutic consequences of genetic alterations. A striking susceptibility of these murine malignancies to platinum drugs has emerged, providing further confidence in their relevance to the human disease. In addition to these models, the pathogenesis of highgrade serous disease derived from risk-reducing surgeries in mutation carriers has pointed to a role of mutations in p53 commonly encountered in tubal intraepithelial carcinomas.
symposium article
endometrioid ovarian cancer model In this particular model, the focus was on altering gene pathways known to be implicated in a subtype of ovarian cancer of ovarian surface epithelial cells [16, 17]. The discovery of frequent somatic PTEN mutations and loss of heterozygosity at the 10q23 PTEN locus in endometrioid ovarian cancer implicates a key role for PTEN in the etiology of this epithelial ovarian cancer subtype. Similarly, K-RAS oncogene is also mutated in endometrioid ovarian cancer, albeit at a lower frequency. This was achieved by introducing a recombinant adenoviral vector expressing Cre recombinase into the bursal cavity that encloses the ovaries in genetically engineered mouse models carrying the conditional oncogenic K-ras and/or loss of function tumor suppressor Pten mutation. Introduction of
i8 | Muggia et al.
adenoviral Cre resulted in activation of the oncogenic KrasG12D mutation within cells of the ovarian surface epithelium, which in turn gave rise to benign ovarian endometriotic-like lesions with simple endometrioid glandular morphology [17]. Furthermore, when adenoviral Cre was introduced into the uterine tubal junction, activation of oncogenic K-ras was achieved within uterine and tubal cells, which acquired a growth advantage and were able to implant and expand at ectopic peritoneal sites [17]. As a result, some mice developed pelvic endometriosis, making this the first model of de novo endometriosis. When oncogenic K-ras mutation was further combined with loss of function Pten expression in LSL–KrasG12D/+; Ptenloxp/loxp mice, frank endometrioid carcinomas arose from the ovaries and metastasized to pelvic and peritoneal locations similar to human tumors [18]. In addition to its histological characteristics, several observations such as high levels of ER immunostaining, activation of Akt and mTOR pathways, peritoneal spread, eventual development of ascites and therapeutic benefit from cisplatin [19] suggested recapitulation of molecular and clinical findings. Adding interest, subsequent work has expanded on features of ‘side populations’ that arise under pressure of treatment, and may be interrogated further for ‘stem cell’ features [20]. While this model does not address specifically the role of BRCA genes in ovarian tumorigenesis, it links the emergence of endometrioid cancers to alterations in K-ras oncogenic signals coupled with Pten loss within this specialized epithelium. Of interest, BRCA1 dysfunction has been recently linked with PTEN mutations in basal-like breast carcinogenesis [21]. The Dinulescu model, therefore, provides further substantiation of how these pathway alterations lead to phenotypic expression (including sensitivity to cisplatin) in tumors arising from the female genital tract.
pathogenesis of serous ovarian, tubal and primary peritoneal cancers A mouse model for the predominant subtype of extrauterine neoplasms of Mullerian epithelial origin has not been yet been developed. Nevertheless, risk-reducing surgery in BRCA1 and BRCA2 mutation carriers has identified a likely precursor to high-grade serous carcinomas: the tubal intraepithelial carcinoma [22]. As noted earlier, TP53 mutations are a recognizable step in this evolution [6]—a situation somewhat analogous to the pathogenesis of BRCA1 (but not BRCA2)related breast cancer (refer to Mangia et al., this supplement [23])
granulosa cell dysfunction due to BRCA loss of function Given that the menstrual cycle is the strongest risk factor for non-familial ovarian cancers, Dubeau and co-workers [24] hypothesized that ovarian cancer predisposition in BRCA1 mutation carriers may be driven, at least in part, by consequences of such mutations on menstrual cycle regulation. They introduced a Brca1 knockout in granulosa cells, which play a central role in menstrual cycle control, using a mouse
Volume 22 | Supplement 1 | January 2011
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 23, 2015
these tumors could prove useful in allowing drug resistance studies. A mammary tumor arising in K14 Cre; Brca1flox/flox; p53 flox/flox mice had knocked out p53 and BRCA function within the breast. Histologically, tumors arising in the breast of these mice are poorly differentiated and lack ER, progesterone receptor and human epidermal growth factor receptor 2 (Her2) expression—hence, the designation ‘triple negative’. The sensitivity of these tumors to cisplatin, coupled with the recognition that BRCA1 mutation carriers often developed the ‘triple-negative’ breast cancer phenotype, spurred clinical trials of platinum compounds directed towards patients with metastatic disease fulfilling the criteria for such designation. Some encouraging therapeutic results have been reported [8], but additional refinements in the characterization of breast cancers within this phenotype will be needed to provide more convincing evidence of the usefulness of platinum-based therapeutic strategies against breast cancers arising in a BRCA mutation background [13, 14]. Platinum-based therapies, of course, have formed the cornerstone of treatment of epithelial ovarian cancer, with the added emerging feature that this treatment is more effective stage-by-stage in BRCA mutation carriers than in those cancers not arising in a hereditary background [15]. Interfering with the poly(ADP-ribose) polymerase (PARP) pathway involved in base excision repair leads to marked differential cytotoxicity of BRCA-deficient tumors versus wild-type tumors, which have an intact homologous recombination (HR) pathway of DNA repair of double-strand breaks [16]. Concepts emanating in part from this model have influenced the design of clinical trials that demonstrated the usefulness of adding a drug with presumed PARP-inhibitor properties to a carboplatin–gemcitabine doublet: patients previously untreated for metastatic breast cancer were simply selected for randomization to PARP-1 inhibitor or no PARP-1 inhibitor added to chemotherapy on the basis of being ‘triple negative’. A remarkable improvement in response rates and progressionfree survival was documented and this has led to a large phase III trial as a pivotal registration trial seeking approval for inhibitors of PARP-1. The trial also encourages the extrapolation of certain clinical subtypes as being not only reflective of lack of BRCA function (linked to inherited deleterious BRCA mutations) but also arising from other causes of genetic instability and epigenetic silencing of these pathways.
Annals of Oncology
Annals of Oncology
model. Noteworthy findings were the emergence of serous cystadenomas. Results presented at the Bari symposium demonstrated a pattern of menstrual/estrus dysfunction when these engineered mice were compared with the wild-type species. Clinical validation of these findings hopefully will be sought. A pattern of menstrual dysfunction occasionally occurring in women with a strong family history of breast and ovarian neoplasia has been reported [25]. Also, it is attractive to be able to pursue further a hypothesis that links BRCA gene dysfunction with a propensity towards breast and ovarian neoplasia.
discussion
(i) Contribution of other genes in the eventual development of neoplasia related to BRCA1 and BRCA2 mutations in hereditary carriers. (ii) What accounts for the susceptibility of the breast and ovary to neoplastic transformation in relation to these inherited mutations. (iii) The relationship of estrogens and estrogen–receptor signaling in the pathogenetic steps leading to breast and extrauterine Mullerian carcinogenesis. (iv) Characterization of premalignant lesions in these tissues. (v) Approaches to interfering with the emergence of an eventual malignancy (e.g. chemo-prevention) in the presence of known hereditary mutations (vi) Characterization of epigenetic changes that lead to phenotypic expression similar to the hereditary type. (vii) Future clues to some of these questions will emerge not only from these models but also from epidemiologic
Volume 22 | Supplement 1 | January 2011
study and clinical interventions. Several articles in this supplement deal with such studies, particularly as they relate to determining and/or altering breast cancer risks [23, 26–29].
disclosures The authors declare no conflict of interest.
references 1. Chen JJ, Silver D, Cantor S et al. BRCA1, BRCA2, and Rad51 operate in a common DNA damage response pathway. Cancer Res 1999; 59(7 Suppl): 1752s–1756s. 2. Scully R, Livingston DM. In search of the tumour-suppressor functions of BRCA1 and BRCA2. Nature 2000; 408: 429–432. 3. Silver DP, Dimitrov SD, Feunteun J et al. Further evidence for BRCA1 communication with the inactive X chromosome. Cell 2007; 128: 991–1002. 4. Livingston DM, Silver DP. Cancer: crossing over to drug resistance Nature 2008; 451: 1111–1115. 5. Crum CP, Drapkin R, Miron A et al. The distal Fallopian tube: a model for pelvic serous carcinogenesis. Curr Opin Obstet Gynec 2007; 19: 3–9. 6. Levanon K, Ng V, Piao HY et al. Primary ex vivo cultures of human Fallopian tube epithelium as a model for serous ovarian carcinogenesis. Oncogene 2010; 29: 1103–1113. 7. Dubeau L. The cell of origin of ovarian epithelial tumours. Lancet 2008; 9: 1191–1197. 8. Fasano J, Muggia F. Breast cancer arising in a BRCA-mutated background: therapeutic implications from an animal model and drug development. Ann Oncol 2008; 4: 609–614. 9. Borst P, Rottenberg S, Jonkers J. How do real tumors become resistant to cisplatin? Cell Cycle 2008; 7: 1353–1359. 10. Muggia F. BRCA-deficient animal models and cisplatin resistance. Ann Oncol 2009; 20: 962. 11. Rottenberg S, Mygren AOH, Pajic M et al. Selective induction of chemotherapy resistance of mammary tumors in a conditional mouse model for hereditary breast cancer. Proc Natl Acad Sci USA 2007; 104: 12117–12122. 12. Perou CM, Sorlie T, Eisen MB et al. Molecular portraits of human breast tumors. Nature 2000; 406: 747–752. 13. Chappuis PO, Goffin J, Wong N et al. A significant response to neoadjuvant chemotherapy in BRCA1/2 related breast cancer. J Med Genet 2002; 39: 608–610. 14. Kennedy RD, Quinn JE, Mullan PB et al. The role of BRCA1 in the cellular response to chemotherapy. J Natl Cancer Inst 2004; 96: 1659–1668. 15. Muggia F. Platinum compounds 30 years after the introduction of cisplatin: implications for the treatment of ovarian cancer. Gynecol Oncol 2009; 112: 275–281. 16. Ashworth AA. A synthetic lethal therapeutic approach: poly(ADP) ribose polymerase inhibitors for the treatment of cancers deficient in DNA double-strand break repair. J Clin Oncol 2008; 26: 3785–3790. 17. Dinulescu DM, Ince TA, Quade BJ et al. Role of k-ras and Pten in the development of mouse models of endometriosis and endometrioid ovarian cancer. Nat Med 2005; 11: 63–70. 18. Martzuk MM. News and views: gynecologic diseases get their genes. Nat Med 2005; 11: 24–26. 19. Paraskar AS, Shivani S, Chin KT et al. Harnessing structure-activity relationship to engineer a cisplatin nanoparticle for enhanced antitumor efficacy. Proc Natl Acad Sci 2010; 107: 12435–12440. 20. Szotek PP, Pieretti-VonMarckeMasiakos R, Masiakos PT et al. Ovarian cancer side population defines cells with stem cell-like characteristics and Mullerian inhibiting substance responsiveness. Proc Natl Acad Sci USA 2006; 103: 11154–11159.
doi:10.1093/annonc/mdq659 | i9
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 23, 2015
We have expanded on preclinical models that have recently explored the molecular changes involved in the pathogenesis of breast and ovarian cancers. A key link between BRCA gene dysfunction and sensitivity to platinum drugs has emerged from these models—emphasizing the importance of defective HR in therapeutic responses to cisplatin. Therapeutic interest has been further stimulated by the introduction of PARP-1 inhibitors: these drugs not only potentiated the effect of DNAdamaging drugs that cause single-strand breaks, but were noted particularly effective in the absence of HR. A corollary to these findings has been the hypothesis that sporadically occurring cancers sharing the phenotype associated with BRCA mutation carriers may result as a consequence of dysfunction in these genes arising from epigenetic changes or gene deletions. As these associations come to light, the emerging questions converge towards pathogenetic pathways leading to such breast and ovarian tumorigenesis. A further dimension has been added by experimental findings indicating that granulosa cell dysfunction may have profound consequences in a mouse model deleting Brca1 in these cells. The animal models described earlier may be viewed as just an initial step in understanding the pathogenesis of human disease. Such models may allow us to look into additional preventive and therapeutic interventions both in models and in the clinic. Specifically, issues that need to be explored include the following.
symposium article
symposium article 21. Saal LH, Gruvberger-Sall SK, Persson C et al. Recurrent gross mutations of the PTEN tumor suppressor gene in breast cancers with deficient DSB repair. Nat Genet 2008; 40: 102–107. 22. Vang R, Shir I-M, Kurman RJ. Ovarian low-grade and high-grade serous carcinoma: pathogenesis, clinicopathologic and molecular biologic features, and diagnostic problems. Adv Anat Pathol 2009; 16: 267–282. 23. Mangia A, Malfettone A, Simoni G, Darvishian F. Old and new concepts in histopathological characterization of familial breast cancer. Ann Oncol 2011; 22(Suppl1): mdq662. 24. Chodankar R, Kwang S, Sangiorgi F et al. Cell-nonautonomous induction of ovarian and uterine serous cystadenomas in mice lacking a functional Brca1 in ovarian granulosa cells. Curr Biol 2005; 15: 561–565.
Annals of Oncology
25. Matalliotakis IM, Cakmak H, Fragouli YG et al. Epidemiological characteristics in women with and without endometriosis in the Yale series. Arch Gynecol Obstet 2008; 277: 389–393. 26. Milne RL, Antoniou AC. Genetic modifiers of cancer risk for BRCA1 and BRCA2 mutation carriers. Ann Oncol 2011; 22 (Suppl 1): mdq660. 27. Radice P, De Summa S, Caleca L, Tommasi S. Unclassified variants in BRCA genes: guidelines for interpretation. Ann Oncol 2011; 22 (Suppl 1): mdq661. 28. Warner E. Impact of MRI surveillance and breast cancer detection in young women with BRCA mutations. Ann Oncol 2011; 22 (Suppl 1): mdq665. 29. Smith J, Axelrod D, Singh B, Kleinberg D. Prevention of breast cancer: the case for studying inhibition of IGF-1 actions. Ann Oncol 2011; 22 (Suppl 1): mdq666.
Downloaded from http://annonc.oxfordjournals.org/ by guest on August 23, 2015
i10 | Muggia et al.
Volume 22 | Supplement 1 | January 2011