- Email: [email protected]
The pathology of inherited breast cancer
Pathology (2002 ) 34, pp. 309– 314
T I M E LY
TOPIC
The pathology of inherited breast cancer JANE E. ARMES*
AND
DEON J. VENTER†
Pathology Downloaded from informahealthcare.com by York University Libraries on 07/09/14 For personal use only.
*Melbourne Pathology and Department of Pathology, Royal Women’s Hospital, Melbourne, and Molecular Pathology Laboratory, Victorian Breast Cancer Research Consortium, Department of Pathology, The University of Melbourne; † Cancer Genomics Laboratory, Murdoch Children’s Research Institute, Department of Molecular Diagnostics, Royal Women’s and Royal Children’s Hospitals, and Department of Pathology, University of Melbourne, Melbourne, Australia
Summary Familial breast cancer has been recognised for many years. In the 1990s the genetic mechanism of inheritance of a proportion of these familial cancers was found to be attributable to germline mutation in either of two newly discovered genes, namely BRCA1 and BRCA2. Since the discovery of these genes, studies have been performed in which the pathological characteristics of familial cancers arising in patients with germline BRCA1 and BRCA2 mutation have been examined. A distinct pathological phenotype of high-grade, oestrogen receptor-negative breast cancer, often with medullary features, has been consistently described for BRCA1 cancers. A less distinct phenotype has been described for BRCA2 cancers. The discovery of genotypephenotype correlation has significant implications for patient management and novel treatment strategies, not only for inherited cancers, but for breast cancer in general. Key words: Breast cancer, BRCA1, BRCA2, familial cancer, pathology, immunohistochemistry, gene expression microarrays, genotype-phenotyp e correlation. Received 16 April, accepted 22 April 2002
INTRODUCTION Breast cancer develops due to an interaction between genetic predisposition and exposure to environmental risk factors. More than 7500 breast cancers are diagnosed in Australia annually. Whilst the majority of these cancers are sporadic ( i.e., not associated with a family history of disease), a significant minority of breast cancers occur in multiple-case families. The majority of familial breast cancers are believed to arise on the basis of an inherited predisposition, although it is possible that some ‘familial’ cancers are due to chance clustering of a common disease. In recent years, two genes have been discovered which, when inherited in a mutant form, strongly predispose to the development of breast cancer. These genes, namely BRCA1 and BRCA2, are estimated to account for approximately 6% of all breast cancers and up to 15% of breast cancers occurring in women less than 30 years of age.1 BRCA1 and BRCA2 are large genes, located at chromosome 17q and 13q, respectively. The functions of the proteins encoded by these genes are not fully understood and, due to the large protein size, are likely to be multiple
and complex. However, experimental data suggest that the BRCA1 and BRCA2 proteins participate in cellular DNA repair mechanisms and in the maintenance of genomic integrity.2 Therefore, it is understandable that an inherited mutation in these genes predisposes to cancer. It has recently been suggested that the tissue specificity of the cancer predisposition ( notably breast and ovarian tissue) in the case of BRCA1, is due to wild-type BRCA1 acting as a controller of oestrogen receptor regulated gene transcription. 3 Although BRCA1 and BRCA2 are known breast cancer predisposition genes, it is important to note that only a minority of breast cancers occurring in a familial setting are due to mutations in these genes. Further, not all people who inherit mutations in these genes will develop breast cancer. Both BRCA1 and BRCA2 were identified by linkage studies using approximately 200 extraordinarily large, multigeneration kindreds, with numerous individuals affected by breast and ovarian cancer. It is not surprising, therefore, that almost all families with six or more individuals with breast cancer ( i.e., families similar to those used for linkage analysis) will harbour mutation in BRCA1 or BRCA2.4 However, less than 50% of families with four or five individuals with breast cancer will have a germline BRCA1 or BRCA2 mutation.5 These latter families represent a large proportion of patients with a significant family history and are strong evidence to support the proposition that there are other breast cancer predisposition genes. Penetrance estimates of developing breast cancer also differ depending on the families studied. Thus, following the localisation and identification of BRCA1 and BRCA2, these same multigeneration kindreds were used to predict the penetrance of breast cancer in individuals who carried a mutation. The estimates were a 20-fold increased risk, or an 80% risk of developing breast cancer to the age of 70 years. 6 However, subsequent population-based studies have shown that these penetrance figures are overestimates. The phenotypic penetration of mutation in these very large families used for linkage analyses were not representative of the penetrance of mutations in the population as a whole. One of our groups has shown that the population-based penetrance rate is much less than that predicted from the multigeneration kindreds, and is more likely 40%.1 Whilst not all cases of familial breast cancer are due to inheritance of germline mutation in BRCA1 and BRCA2, it is also true that some patients who develop breast cancer and have a germline mutation in these genes do not have a
ISSN 0031–3025 printed/ISSN 1465– 3931 online/02/040309 – 06 © 2002 Royal College of Pathologists of Australasia DOI:10.1080/00313020220147113
Pathology Downloaded from informahealthcare.com by York University Libraries on 07/09/14 For personal use only.
310
ARMES AND VENTER
family history of breast cancer. When an Australian population-based cohort of early onset breast cancer was studied ( The Australian Breast Cancer Family Study7), fewer than 50% of women with germline BRCA1 or BRCA2 mutations had a family history of breast or ovarian cancer. Further, less than 10% of these women had a family history of two or more affected individuals.8 The lack of family history is probably multifactorial, possibly relating to factors affecting the penetrance of the cancer phenotype, including the site and type of mutation, inheritance of mutations or polymorphisms in other genes which may modify the effect of the BRCA1 or BRCA2 gene mutation, and environmental factors. Additionally, family history may be masked if carriers belong to small kindreds, with few female relatives, particularly if inheritance of the mutation is through the paternal allele. Despite the imperfect association between inherited BRCA1 and BRCA2 mutation and family history of breast cancer, genetic testing for mutations in these genes is available to selected patients with breast cancer and their family members. Typically, a strong family history of breast cancer, especially if associated with early onset or multiple primary tumours, has been proposed to form the basis of genetic testing. Unfortunately, since the genes are large and mutation sites can be spread uniformly across their entire coding regions, genetic testing is time-consuming and labour-intensive. 9 Given these complexities, there is a need to consider additional criteria to better screen for women most likely to carry a germline mutation. It is of great interest to pathologists, therefore, that several publications have indicated that breast cancers which arise in patients with germline mutations in BRCA1 and BRCA2 have a recognisable histological phenotype.
PATHOLOGY OF FAMILIAL BREAST CANCER Historical aspects Familial breast cancer has been recognised for many years, and several early attempts were made to define the pathology of lesions occurring in this context. Not surprisingly, prior to the identification of BRCA1 and BRCA2, workers were unable to clearly categorise the familial cancers by mutation type, and there was little consensus as to the pathological appearance of familial breast cancer. Invasive tubular, lobular and medullary carcinoma and in situ lobular carcinoma have all been reported as common in familial breast cancer.10–12 The studies were further confounded since breast cancer is a common disease and, without confirmation of mutation status, it is possible that some cancers which clustered within a large pedigree were a mixture of inherited cancers and sporadic cancers. The cloning of BRCA1 and BRCA2 allowed pathologists to categorise subsets of familial breast cancer according to the underlying germline mutation, and thus paved the way for identifying the histological phenotype associated with the different subtypes of familial breast cancer. The pathology of BRCA1 and BRCA2 breast cancer Several studies have reported the morphology and immunohistochemical phenotypes of breast cancers arising in patients with germline mutation in BRCA1 and BRCA2.13–18 These include our own analysis of an
Pathology ( 2002 ), 34, August
Fig. 1 Low power view of BRCA1 cancer showing medullary-type features including the circumscribed border ( H&E, original magnification, ´ 20 ).
Australian population-based series of early onset breast cancer ( The Australian Breast Cancer Family Study19,20). In general, these studies indicate that BRCA1 cancers have a characteristic and relatively specific appearance, whilst the appearance of BRCA2 cancers is less distinct. Medullary-type cancers ( as defined by Ridolfi et al. 21) are over-represented in cancers arising in BRCA1 mutation carriers18,19,22 ( Fig. 1). Importantly, our own series demonstrated that this over-representation was not due to the relatively young age of mutation carriers, since the cancers in patients with germline BRCA1 mutation were compared with other early onset cancers ( i.e., diagnosed before 40 years) without mutation and with early onset cancers in BRCA2 mutation carriers. Whilst true medullary cancers have been reported in BRCA1 mutation carriers,18,22 only atypical medullary cancers were seen in our own series, which generally lacked completely circumscribed pushing margins.19 Nonetheless, extensive pushing margins ( usually > 50% of the tumour edge) were a characteristic feature of BRCA1 tumours ( Fig. 2). Two of the other secondary features of medullary carcinomas are absence of tubule formation and presence of a moderate or marked lymphocytic
Fig. 2 BRCA1 cancer showing high power view of circumscribed, pushing margin, trabecular pattern of growth and lack of tubule formation ( H&E, original magnification, ´ 250 ).
Pathology Downloaded from informahealthcare.com by York University Libraries on 07/09/14 For personal use only.
FAMILIAL BREAST CANCER
311
Fig. 3 ( a) High power view of BRCA1 cancer showing marked nuclear pleomorphism, lack of tubule formation and abundant mitotic figures ( H&E, original magnification, ´ 400 ). ( b) Low grade, early onset control cancer without BRCA1 or BRCA2 mutation. This low grade type of cancer is extremely rare in association with germline BRCA1 or BRCA2 mutation ( H&E, original magnification, ´ 400 ).
infiltrate. Absence of tubule or acinar formation appears to be a strong characteristic of BRCA1 cancers, the tumours usually growing as solid trabeculae of cells ( Fig. 2). The specificity of the lymphocytic infiltrate is more controversial, being shown as significant in BRCA1-associated cancers in some reports,22 but not significant compared with BRCA2 cancers and early onset controls in other series.19 Almost all reports of BRCA1-associated breast cancers have described high-grade tumours, due to the pronounced nuclear pleomorphism, characteristic lack of tubule formation and the exceptionally high mitotic rate23 ( Fig. 3a). The mitotic index is commonly greater than 50 per 10 high power fields.16,17,19,22 Low-grade tumours with few mitotic figures and well-formed tubules are almost never seen in BRCA1 cancers ( Fig. 3b). Confluent necrosis is also common in BRCA1-associated cancers. These high-grade features are independent of the relatively early onset of the majority of BRCA1-related cancers. It has been proposed that ductal carcinoma in situ ( DCIS) around the invasive lesion is rare in BRCA1related carcinomas.22 This may reflect the rapid proliferation rate of BRCA1 tumours, since a rapidly growing invasive tumour is likely to obliterate the preceding in situ component. However, the above study could not address the extent of DCIS within the breast tissue away from the invasive lesion, since only one slide ( chosen to represent the invasive tumour) was available for analysis in each case. Further studies have shown that when all the pathology material is available for examination, DCIS is present in BRCA1 cancers, although likely to be in lesser amounts than in BRCA2 cancers or early onset controls.19 The issue of a pre-existing in situ phase requires clarifica-
tion, as identification of a pre-invasive stage of disease is important for cancer screening strategies. A distinct histological phenotype of BRCA1 cancers has emerged from these studies; that of a high-grade tumour, with a trabecular growth pattern, pushing margins, central necrosis and an exceptionally high mitotic rate. Unfortunately, the phenotype for BRCA2 cancers is less clear. Whilst BRCA2-associated lesions may be distinguishable from BRCA1 cancers, this is mainly due to their lack of features characteristic of BRCA1 cancers, rather than specific features arising from the BRCA2 mutation itself. Indeed, there appears to be a dichotomy of phenotype related to BRCA2 cancers. Several studies have shown few distinguishing features between BRCA2 cancers when compared with early onset controls,16,19 other than an over-representation of pleomorphic lobular cancers and lack of grade 1 lesions ( Fig. 4). However, the European-based Breast Cancer Linkage Consortium,22 an Icelandic group24 and our own ( unpublished ) experience from cancers referred to a family cancer clinic, have identified a high-grade tumour arising in BRCA2 cancers, somewhat reminiscent of BRCA1 cancers, but with a lower mitotic rate and a propensity for acinar rather than trabecular-type growth. The molecular pathogenesis of BRCA1 and BRCA2 cancers The morphological phenotypes of BRCA1 and BRCA2 cancers would suggest distinct mechanisms of molecular pathogenesis of these cancers. Immunohistochemical studies identified a specific immunophenotype of BRCA1- but not BRCA2-associated cancers. Thus, BRCA1 cancers and their precursor DCIS component, often show stabilisation of
Pathology Downloaded from informahealthcare.com by York University Libraries on 07/09/14 For personal use only.
312
Pathology ( 2002 ), 34, August
ARMES AND VENTER
Fig. 4 BRCA2 cancer. ( a) Low power view of lobular growth pattern with targetoid appearance and single cell files ( H&E, original magnification,
´ 50). ( b) High power view showing pleomorphic lobular phenotype ( H&E, original magnification, ´ 400 ). p53 protein expression, detected as strong nuclear staining with antibodies directed to p53 ( Fig. 5).20,25,26 Somatic mutations in p53 have been confirmed in these cancers, as would be expected from the immunohistochemical studies.20,27 BRCA1 cancers also lack expression of the steroid hormone receptors for oestrogen (ER) and progesterone ( PR). They are also ERBB2 negative.13,15,20,28,29 There is little difference between expression of p53, ER, PR or ERBB2 in BRCA2 cancers compared with early onset controls.20 The significance of the molecular phenotype In many instances, BRCA1 cancers can be distinguished from cancers arising in patients with germline BRCA2
Fig. 5 BRCA1 cancer showing immunohistochemical staining with antibodies to p53. Strong nuclear staining in almost all invasive carcinoma cells is characteristic of BRCA1 cancers ( p53 immunohistochemical stain, original magnification, ´ 250 ).
mutations or cancers in patients with no germline mutation in these genes, due to their characteristic morphology and immunophenotype. This finding has practical importance in the screening for germline mutations. Given that the majority of breast cancers in families with six or more affected people are due to inheritance of BRCA1 or BRCA2 mutation, pathological assessment of their tumour would indicate which gene to screen first. Also, since the most economically efficient form of standard genetic testing ( the protein truncation test; PTT) is acknowledged not to detect all possible germline mutations in BRCA1 or BRCA2, a characteristic phenotype could prompt a more extensive search for germline mutation, if not at first detected by the PTT, perhaps including analyses for large deletions within the gene.30 Further, there are many individuals in the Australian population with only a moderate ( e.g., one other affected relative) or no family history of breast cancer, who are not usually offered genetic testing for BRCA1 or BRCA2 germline mutation. In such cases, a pathology element should be entered into the risk calculations for germline mutation in such people, as has been suggested by Lalloo and Evans.31 As a consequence of the relatively recent recognition of the existence of a characteristic phenotype occurring in patients with a germline BRCA1 mutation, there are no reliable estimates of the relative risk of having a germline mutation in a patient with the characteristic cancer phenotype. Such studies are relatively complex due to ethical implications of finding germline mutations in unselected archival series; therefore, such studies would have to be performed on patient cohorts who are aware that germline mutation analysis is proceeding. The documented specificity of the phenotype is also reliant on the sensitivity of the type of mutation testing performed. The ability to recognise distinct morphological and molecular phenotypes has important implications for
Pathology Downloaded from informahealthcare.com by York University Libraries on 07/09/14 For personal use only.
FAMILIAL BREAST CANCER
anatomical pathology.32 Despite acknowledgement that pathology analysis is pivotal in patient management, until now treatment strategies have been relatively crude, so extensive subcategorisation of tumours by genotype and phenotype has been unnecessary. Nonetheless, some subcategorisation is standard practice in pathology (exemplified by the important distinction between ER-positive versus ER-negative breast cancer or small cell and nonsmall cell lung cancer). More recently, there has been rapid progress in the move to individualise cancer therapy, based on the molecular drivers of disease. Thus, specific gene-alteration-based treatment is available for patients with metastatic breast cancer believed to be driven by amplification and overexpression of ERBB2, using trastuzumab ( Herceptin; Roche, Switzerland). In another example, carboplatin-based therapy trials are underway for BRCA1 and BRCA2-associated breast cancers, due to the effect of mutation in these genes on double-stranded DNA repair. It is possible that, if such trials prove effective in known mutation carriers, there may be implications for treatment of sporadic cancers with phenotypes similar to BRCA1 and BRCA2. Further, the advent of microarray-based analysis of tumours, including breast cancer, is likely to stratify cancers more accurately, since this technology is able to analyse the expression of thousands of gene sequences simultaneously per tumour and enhance our repertoire of diagnostic probes. The few studies performed on breast cancer using this powerful technology to date, and our own unpublished data, would predict that poor-outcome breast cancers are likely to be driven primarily by aberrations in genes involved in cell cycle control.33 Finally, the molecular phenotype of BRCA1 and BRCA2 cancers can be used in conjunction with experimental approaches to elucidate the mechanisms of carcinogenesis induced by aberrant function of these genes. Both BRCA1 and BRCA2 are involved in the maintenance of genomic integrity.34,35 However, given the experimental data indicating the anti-proliferation properties of wild-type BRCA1 36,37 and the notable high mitotic rate in BRCA1 cancers, it is likely that the BRCA1 mutation itself leads to the key event of oncogenesis, that of uncontrolled cellular proliferation. Once early inactivation of the normal BRCA1 allele and p53 stabilisation occurs, there is little selection pressure for the recruitment of other genes involved in oncogenesis. In contrast, the more heterogeneous phenotype of BRCA2-associated cancers, and the less striking proliferation rate, would support the hypothesis that a defect in DNA repair mechanisms would drive the pathogenesis of BRCA2 breast cancers. The more variable phenotype could then be explained by dependence on the aberrant function of other genes with unrepaired mutations. Clearly, the pathology of inherited breast cancer highlights the current inextricable interaction between the diagnostic discipline of pathology and molecular pathogenesis of disease. ACKNOWLEDGEMENTS The authors would like to thank Dr Gino Somers for his careful reading of the manuscript and Mr John Ciciulla for figure preparation. Sources of support were The Victorian Breast Cancer Research Consortium, National Health and Medical Research Council and the National Institutes of Health.
313
Address for correspondence: Associate Professor J. E. Armes, Department of Pathology, Royal Women’s Hospital, Grattan Street, Carlton, Vic 3053, Australia. E-mail: [email protected] u
References 1. Hopper JL, Southey MC, Dite GS, et al. Population-based estimate of the average age-specific cumulative risk of breast cancer for a defined set of protein-truncating mutations in BRCA1 and BRCA2. Australian Breast Cancer Family Study. Cancer Epidemiol Biomarkers Prev 1999; 8: 741–7. 2. Welcsh PL, Owens KN, King MC. Insights into the functions of BRCA1 and BRCA2. Trends Genet 2000; 16: 69–74. 3. Zheng L, Annab LA, Afshari CA, Lee WH, Boyer TG. BRCA1 mediates ligand-independent transcriptional repression of the estrogen receptor. Proc Natl Acad Sci USA 2001; 98: 9587– 92. 4. Ford D, Easton DF, Peto J. Estimates of the gene frequency of BRCA1 and its contribution to breast and ovarian cancer incidence. Am J Hum Genet 1995; 57: 1457–62. 5. Ford D, Easton DF, Stratton M, et al. Genetic heterogeneity and penetrance analysis of the BRCA1 and BRCA2 genes in breast cancer families. The Breast Cancer Linkage Consortium. Am J Hum Genet 1998; 62: 676–89. 6. Easton DF, Bishop DT, Ford D, Crockford GP. Genetic linkage analysis in familial breast and ovarian cancer: results from 214 families. The Breast Cancer Linkage Consortium. Am J Hum Genet 1993; 52: 678–701. 7. McCredie MR, Dite GS, Giles GG, Hopper JL. Breast cancer in Australian women under the age of 40. Cancer Causes Control 1998; 9: 189– 98. 8. Cui J, Hopper JL. Why are the majority of hereditary cases of earlyonset breast cancer sporadic? A simulation study. Cancer Epidemiol Biomarkers Prev 2000; 9: 805–12. 9. Couch FJ, Weber BL. Mutations and polymorphisms in the familial early-onset breast cancer ( BRCA1) gene. Breast Cancer Information Core. Hum Mutat 1996; 8: 8–18. 10. Rosen PP, Lesser ML, Senie RT, Kinne DW. Epidemiology of breast carcinoma III: relationship of family history to tumor type. Cancer 1982; 50: 171–9. 11. Lagios MD, Rose MR, Margolin FR. Tubular carcinoma of the breast: association with multicentricity, bilaterality, and family history of mammary carcinoma. Am J Clin Pathol 1980; 73: 25– 30. 12. Marcus JN, Watson P, Page DL, Lynch HT. Pathology and heredity of breast cancer in younger women. J Natl Cancer Inst Monogr 1994; 16: 23– 34. 13. Robson M, Gilewski T, Haas B, et al. BRCA-associated breast cancer in young women. J Clin Oncol 1998; 16: 1642– 9. 14. Lakhani SR, Jacquemier J, Sloane JP, et al. Multifactorial analysis of differences between sporadic breast cancers and cancers involving BRCA1 and BRCA2 mutations. J Natl Cancer Inst 1998; 90: 1138– 45. 15. Karp SE, Tonin PN, Begin LR, et al. Influence of BRCA1 mutations on nuclear grade and estrogen receptor status of breast carcinoma in Ashkenazi Jewish women. Cancer 1997; 80: 435–41. 16. Marcus JN, Watson P, Page DL, et al. BRCA2 hereditary breast cancer pathophenotype. Breast Cancer Res Treat 1997; 44: 275–7. 17. Eisinger F, Stoppa-Lyonnet D, Longy M, et al. Germ line mutation at BRCA1 affects the histoprognostic grade in hereditary breast cancer. Cancer Res 1996; 56: 471– 4. 18. Marcus JN, Watson P, Page DL, et al. Hereditary breast cancer: pathobiology, prognosis, and BRCA1 and BRCA2 gene linkage. Cancer 1996; 77: 697–709. 19. Armes JE, Egan AJ, Southey MC, et al. The histologic phenotypes of breast carcinoma occurring before age 40 years in women with and without BRCA1 or BRCA2 germline mutations: a population-base d study. Cancer 1998; 83: 2335– 45. 20. Armes JE, Trute L, White D, et al. Distinct molecular pathogeneses of early-onset breast cancers in BRCA1 and BRCA2 mutation carriers: a population-based study. Cancer Res 1999; 59: 2011–7. 21. Ridolfi RL, Rosen PP, Port A, Kinne D, Mike V. Medullary carcinoma of the breast: a clinicopathologic study with 10 year follow-up. Cancer 1977; 40: 1365– 85. 22. Breast Cancer Linkage Consortium. Pathology of familial breast cancer: differences between breast cancer in carriers of BRCA1 or BRCA2 mutations and sporadic cases. Lancet 1997; 349: 1505–10. 23. Elston CW. Grading of invasive carcinoma of the breast. In: Page DL, Anderson TJ, editors. Diagnostic histopathology of the breast. Edinburgh: Churchill Livingstone, 1987; 300–11.
314
24. Agnarsson BA, Jonasson JG, Bjornsdottir IB, Barkardottir RB, Egilsson V, Sigurdsson H. Inherited BRCA2 mutation associated with high grade breast cancer. Breast Cancer Res Treat 1998; 47: 121–7. 25. Struewing JP, Hartge P, Wacholder S, et al. The risk of cancer associated with specific mutations of BRCA1 and BRCA2 among Ashkenazi Jews. N Engl J Med 1997; 336: 1401– 8. 26. Newman B, Mu H, Butler LM, Millikan RC, Moorman PG, King MC. Frequency of breast cancer attributable to BRCA1 in a populationbased series of American women. JAMA 1998; 279: 915– 21. 27. Phillips KA, Nichol K, Ozcelik H, et al. Frequency of p53 mutations in breast carcinomas from Ashkenazi Jewish carriers of BRCA1 mutations. J Natl Cancer Inst 1999; 91: 469–73. 28. Johannsson OT, Idvall I, Anderson C, et al. Tumour biological features of BRCA1-induced breast and ovarian cancer. Eur J Cancer 1997; 33: 362–71. 29. Robson M, Rajan P, Rosen PP, et al. BRCA-associated breast cancer: absence of a characteristic immunophenotype. Cancer Res 1998; 58: 1839– 42. 30. Puget N, Stoppa-Lyonnet D, Sinilnikova OM, et al. Screening for germ-line rearrangements and regulatory mutations in BRCA1 led to
Pathology Downloaded from informahealthcare.com by York University Libraries on 07/09/14 For personal use only.
Pathology ( 2002 ), 34, August
ARMES AND VENTER
31. 32. 33. 34. 35. 36. 37.
the identification of four new deletions. Cancer Res 1999; 59: 455– 61. Lalloo F, Evans DG. The pathology of familial breast cancer: Clinical and genetic counselling implications of breast cancer pathology. Breast Cancer Res 1999; 1: 48–51. MacGregor DP. Anatomical pathology. Med J Aust 2002; 176: 7. van ‘t Veer LJ, Dai H, van de Vijver MJ, et al. Gene expression profiling predicts clinical outcome of breast cancer. Nature 2002; 415: 530– 6. Sharan SK, Morimatsu M, Albrecht U, et al. Embryonic lethality and radiation hypersensitivity mediated by Rad51 in mice lacking Brca2. Nature 1997; 386: 804–10. Scully R, Anderson SF, Chao DM, et al. BRCA1 is a component of the RNA polymerase II holoenzyme. Proc Natl Acad Sci USA 1997; 94: 5605– 10. Holt JT, Thompson ME, Szabo C, et al. Growth retardation and tumour inhibition by BRCA1. Nat Genet 1996; 12: 298– 302. Thompson ME, Jensen RA, Obermiller PS, Page DL, Holt JT. Decreased expression of BRCA1 accelerates growth and is often present during sporadic breast cancer progression. Nat Genet 1995; 9: 444– 50.
T I M E LY
TOPIC
The pathology of inherited breast cancer JANE E. ARMES*
AND
DEON J. VENTER†
Pathology Downloaded from informahealthcare.com by York University Libraries on 07/09/14 For personal use only.
*Melbourne Pathology and Department of Pathology, Royal Women’s Hospital, Melbourne, and Molecular Pathology Laboratory, Victorian Breast Cancer Research Consortium, Department of Pathology, The University of Melbourne; † Cancer Genomics Laboratory, Murdoch Children’s Research Institute, Department of Molecular Diagnostics, Royal Women’s and Royal Children’s Hospitals, and Department of Pathology, University of Melbourne, Melbourne, Australia
Summary Familial breast cancer has been recognised for many years. In the 1990s the genetic mechanism of inheritance of a proportion of these familial cancers was found to be attributable to germline mutation in either of two newly discovered genes, namely BRCA1 and BRCA2. Since the discovery of these genes, studies have been performed in which the pathological characteristics of familial cancers arising in patients with germline BRCA1 and BRCA2 mutation have been examined. A distinct pathological phenotype of high-grade, oestrogen receptor-negative breast cancer, often with medullary features, has been consistently described for BRCA1 cancers. A less distinct phenotype has been described for BRCA2 cancers. The discovery of genotypephenotype correlation has significant implications for patient management and novel treatment strategies, not only for inherited cancers, but for breast cancer in general. Key words: Breast cancer, BRCA1, BRCA2, familial cancer, pathology, immunohistochemistry, gene expression microarrays, genotype-phenotyp e correlation. Received 16 April, accepted 22 April 2002
INTRODUCTION Breast cancer develops due to an interaction between genetic predisposition and exposure to environmental risk factors. More than 7500 breast cancers are diagnosed in Australia annually. Whilst the majority of these cancers are sporadic ( i.e., not associated with a family history of disease), a significant minority of breast cancers occur in multiple-case families. The majority of familial breast cancers are believed to arise on the basis of an inherited predisposition, although it is possible that some ‘familial’ cancers are due to chance clustering of a common disease. In recent years, two genes have been discovered which, when inherited in a mutant form, strongly predispose to the development of breast cancer. These genes, namely BRCA1 and BRCA2, are estimated to account for approximately 6% of all breast cancers and up to 15% of breast cancers occurring in women less than 30 years of age.1 BRCA1 and BRCA2 are large genes, located at chromosome 17q and 13q, respectively. The functions of the proteins encoded by these genes are not fully understood and, due to the large protein size, are likely to be multiple
and complex. However, experimental data suggest that the BRCA1 and BRCA2 proteins participate in cellular DNA repair mechanisms and in the maintenance of genomic integrity.2 Therefore, it is understandable that an inherited mutation in these genes predisposes to cancer. It has recently been suggested that the tissue specificity of the cancer predisposition ( notably breast and ovarian tissue) in the case of BRCA1, is due to wild-type BRCA1 acting as a controller of oestrogen receptor regulated gene transcription. 3 Although BRCA1 and BRCA2 are known breast cancer predisposition genes, it is important to note that only a minority of breast cancers occurring in a familial setting are due to mutations in these genes. Further, not all people who inherit mutations in these genes will develop breast cancer. Both BRCA1 and BRCA2 were identified by linkage studies using approximately 200 extraordinarily large, multigeneration kindreds, with numerous individuals affected by breast and ovarian cancer. It is not surprising, therefore, that almost all families with six or more individuals with breast cancer ( i.e., families similar to those used for linkage analysis) will harbour mutation in BRCA1 or BRCA2.4 However, less than 50% of families with four or five individuals with breast cancer will have a germline BRCA1 or BRCA2 mutation.5 These latter families represent a large proportion of patients with a significant family history and are strong evidence to support the proposition that there are other breast cancer predisposition genes. Penetrance estimates of developing breast cancer also differ depending on the families studied. Thus, following the localisation and identification of BRCA1 and BRCA2, these same multigeneration kindreds were used to predict the penetrance of breast cancer in individuals who carried a mutation. The estimates were a 20-fold increased risk, or an 80% risk of developing breast cancer to the age of 70 years. 6 However, subsequent population-based studies have shown that these penetrance figures are overestimates. The phenotypic penetration of mutation in these very large families used for linkage analyses were not representative of the penetrance of mutations in the population as a whole. One of our groups has shown that the population-based penetrance rate is much less than that predicted from the multigeneration kindreds, and is more likely 40%.1 Whilst not all cases of familial breast cancer are due to inheritance of germline mutation in BRCA1 and BRCA2, it is also true that some patients who develop breast cancer and have a germline mutation in these genes do not have a
ISSN 0031–3025 printed/ISSN 1465– 3931 online/02/040309 – 06 © 2002 Royal College of Pathologists of Australasia DOI:10.1080/00313020220147113
Pathology Downloaded from informahealthcare.com by York University Libraries on 07/09/14 For personal use only.
310
ARMES AND VENTER
family history of breast cancer. When an Australian population-based cohort of early onset breast cancer was studied ( The Australian Breast Cancer Family Study7), fewer than 50% of women with germline BRCA1 or BRCA2 mutations had a family history of breast or ovarian cancer. Further, less than 10% of these women had a family history of two or more affected individuals.8 The lack of family history is probably multifactorial, possibly relating to factors affecting the penetrance of the cancer phenotype, including the site and type of mutation, inheritance of mutations or polymorphisms in other genes which may modify the effect of the BRCA1 or BRCA2 gene mutation, and environmental factors. Additionally, family history may be masked if carriers belong to small kindreds, with few female relatives, particularly if inheritance of the mutation is through the paternal allele. Despite the imperfect association between inherited BRCA1 and BRCA2 mutation and family history of breast cancer, genetic testing for mutations in these genes is available to selected patients with breast cancer and their family members. Typically, a strong family history of breast cancer, especially if associated with early onset or multiple primary tumours, has been proposed to form the basis of genetic testing. Unfortunately, since the genes are large and mutation sites can be spread uniformly across their entire coding regions, genetic testing is time-consuming and labour-intensive. 9 Given these complexities, there is a need to consider additional criteria to better screen for women most likely to carry a germline mutation. It is of great interest to pathologists, therefore, that several publications have indicated that breast cancers which arise in patients with germline mutations in BRCA1 and BRCA2 have a recognisable histological phenotype.
PATHOLOGY OF FAMILIAL BREAST CANCER Historical aspects Familial breast cancer has been recognised for many years, and several early attempts were made to define the pathology of lesions occurring in this context. Not surprisingly, prior to the identification of BRCA1 and BRCA2, workers were unable to clearly categorise the familial cancers by mutation type, and there was little consensus as to the pathological appearance of familial breast cancer. Invasive tubular, lobular and medullary carcinoma and in situ lobular carcinoma have all been reported as common in familial breast cancer.10–12 The studies were further confounded since breast cancer is a common disease and, without confirmation of mutation status, it is possible that some cancers which clustered within a large pedigree were a mixture of inherited cancers and sporadic cancers. The cloning of BRCA1 and BRCA2 allowed pathologists to categorise subsets of familial breast cancer according to the underlying germline mutation, and thus paved the way for identifying the histological phenotype associated with the different subtypes of familial breast cancer. The pathology of BRCA1 and BRCA2 breast cancer Several studies have reported the morphology and immunohistochemical phenotypes of breast cancers arising in patients with germline mutation in BRCA1 and BRCA2.13–18 These include our own analysis of an
Pathology ( 2002 ), 34, August
Fig. 1 Low power view of BRCA1 cancer showing medullary-type features including the circumscribed border ( H&E, original magnification, ´ 20 ).
Australian population-based series of early onset breast cancer ( The Australian Breast Cancer Family Study19,20). In general, these studies indicate that BRCA1 cancers have a characteristic and relatively specific appearance, whilst the appearance of BRCA2 cancers is less distinct. Medullary-type cancers ( as defined by Ridolfi et al. 21) are over-represented in cancers arising in BRCA1 mutation carriers18,19,22 ( Fig. 1). Importantly, our own series demonstrated that this over-representation was not due to the relatively young age of mutation carriers, since the cancers in patients with germline BRCA1 mutation were compared with other early onset cancers ( i.e., diagnosed before 40 years) without mutation and with early onset cancers in BRCA2 mutation carriers. Whilst true medullary cancers have been reported in BRCA1 mutation carriers,18,22 only atypical medullary cancers were seen in our own series, which generally lacked completely circumscribed pushing margins.19 Nonetheless, extensive pushing margins ( usually > 50% of the tumour edge) were a characteristic feature of BRCA1 tumours ( Fig. 2). Two of the other secondary features of medullary carcinomas are absence of tubule formation and presence of a moderate or marked lymphocytic
Fig. 2 BRCA1 cancer showing high power view of circumscribed, pushing margin, trabecular pattern of growth and lack of tubule formation ( H&E, original magnification, ´ 250 ).
Pathology Downloaded from informahealthcare.com by York University Libraries on 07/09/14 For personal use only.
FAMILIAL BREAST CANCER
311
Fig. 3 ( a) High power view of BRCA1 cancer showing marked nuclear pleomorphism, lack of tubule formation and abundant mitotic figures ( H&E, original magnification, ´ 400 ). ( b) Low grade, early onset control cancer without BRCA1 or BRCA2 mutation. This low grade type of cancer is extremely rare in association with germline BRCA1 or BRCA2 mutation ( H&E, original magnification, ´ 400 ).
infiltrate. Absence of tubule or acinar formation appears to be a strong characteristic of BRCA1 cancers, the tumours usually growing as solid trabeculae of cells ( Fig. 2). The specificity of the lymphocytic infiltrate is more controversial, being shown as significant in BRCA1-associated cancers in some reports,22 but not significant compared with BRCA2 cancers and early onset controls in other series.19 Almost all reports of BRCA1-associated breast cancers have described high-grade tumours, due to the pronounced nuclear pleomorphism, characteristic lack of tubule formation and the exceptionally high mitotic rate23 ( Fig. 3a). The mitotic index is commonly greater than 50 per 10 high power fields.16,17,19,22 Low-grade tumours with few mitotic figures and well-formed tubules are almost never seen in BRCA1 cancers ( Fig. 3b). Confluent necrosis is also common in BRCA1-associated cancers. These high-grade features are independent of the relatively early onset of the majority of BRCA1-related cancers. It has been proposed that ductal carcinoma in situ ( DCIS) around the invasive lesion is rare in BRCA1related carcinomas.22 This may reflect the rapid proliferation rate of BRCA1 tumours, since a rapidly growing invasive tumour is likely to obliterate the preceding in situ component. However, the above study could not address the extent of DCIS within the breast tissue away from the invasive lesion, since only one slide ( chosen to represent the invasive tumour) was available for analysis in each case. Further studies have shown that when all the pathology material is available for examination, DCIS is present in BRCA1 cancers, although likely to be in lesser amounts than in BRCA2 cancers or early onset controls.19 The issue of a pre-existing in situ phase requires clarifica-
tion, as identification of a pre-invasive stage of disease is important for cancer screening strategies. A distinct histological phenotype of BRCA1 cancers has emerged from these studies; that of a high-grade tumour, with a trabecular growth pattern, pushing margins, central necrosis and an exceptionally high mitotic rate. Unfortunately, the phenotype for BRCA2 cancers is less clear. Whilst BRCA2-associated lesions may be distinguishable from BRCA1 cancers, this is mainly due to their lack of features characteristic of BRCA1 cancers, rather than specific features arising from the BRCA2 mutation itself. Indeed, there appears to be a dichotomy of phenotype related to BRCA2 cancers. Several studies have shown few distinguishing features between BRCA2 cancers when compared with early onset controls,16,19 other than an over-representation of pleomorphic lobular cancers and lack of grade 1 lesions ( Fig. 4). However, the European-based Breast Cancer Linkage Consortium,22 an Icelandic group24 and our own ( unpublished ) experience from cancers referred to a family cancer clinic, have identified a high-grade tumour arising in BRCA2 cancers, somewhat reminiscent of BRCA1 cancers, but with a lower mitotic rate and a propensity for acinar rather than trabecular-type growth. The molecular pathogenesis of BRCA1 and BRCA2 cancers The morphological phenotypes of BRCA1 and BRCA2 cancers would suggest distinct mechanisms of molecular pathogenesis of these cancers. Immunohistochemical studies identified a specific immunophenotype of BRCA1- but not BRCA2-associated cancers. Thus, BRCA1 cancers and their precursor DCIS component, often show stabilisation of
Pathology Downloaded from informahealthcare.com by York University Libraries on 07/09/14 For personal use only.
312
Pathology ( 2002 ), 34, August
ARMES AND VENTER
Fig. 4 BRCA2 cancer. ( a) Low power view of lobular growth pattern with targetoid appearance and single cell files ( H&E, original magnification,
´ 50). ( b) High power view showing pleomorphic lobular phenotype ( H&E, original magnification, ´ 400 ). p53 protein expression, detected as strong nuclear staining with antibodies directed to p53 ( Fig. 5).20,25,26 Somatic mutations in p53 have been confirmed in these cancers, as would be expected from the immunohistochemical studies.20,27 BRCA1 cancers also lack expression of the steroid hormone receptors for oestrogen (ER) and progesterone ( PR). They are also ERBB2 negative.13,15,20,28,29 There is little difference between expression of p53, ER, PR or ERBB2 in BRCA2 cancers compared with early onset controls.20 The significance of the molecular phenotype In many instances, BRCA1 cancers can be distinguished from cancers arising in patients with germline BRCA2
Fig. 5 BRCA1 cancer showing immunohistochemical staining with antibodies to p53. Strong nuclear staining in almost all invasive carcinoma cells is characteristic of BRCA1 cancers ( p53 immunohistochemical stain, original magnification, ´ 250 ).
mutations or cancers in patients with no germline mutation in these genes, due to their characteristic morphology and immunophenotype. This finding has practical importance in the screening for germline mutations. Given that the majority of breast cancers in families with six or more affected people are due to inheritance of BRCA1 or BRCA2 mutation, pathological assessment of their tumour would indicate which gene to screen first. Also, since the most economically efficient form of standard genetic testing ( the protein truncation test; PTT) is acknowledged not to detect all possible germline mutations in BRCA1 or BRCA2, a characteristic phenotype could prompt a more extensive search for germline mutation, if not at first detected by the PTT, perhaps including analyses for large deletions within the gene.30 Further, there are many individuals in the Australian population with only a moderate ( e.g., one other affected relative) or no family history of breast cancer, who are not usually offered genetic testing for BRCA1 or BRCA2 germline mutation. In such cases, a pathology element should be entered into the risk calculations for germline mutation in such people, as has been suggested by Lalloo and Evans.31 As a consequence of the relatively recent recognition of the existence of a characteristic phenotype occurring in patients with a germline BRCA1 mutation, there are no reliable estimates of the relative risk of having a germline mutation in a patient with the characteristic cancer phenotype. Such studies are relatively complex due to ethical implications of finding germline mutations in unselected archival series; therefore, such studies would have to be performed on patient cohorts who are aware that germline mutation analysis is proceeding. The documented specificity of the phenotype is also reliant on the sensitivity of the type of mutation testing performed. The ability to recognise distinct morphological and molecular phenotypes has important implications for
Pathology Downloaded from informahealthcare.com by York University Libraries on 07/09/14 For personal use only.
FAMILIAL BREAST CANCER
anatomical pathology.32 Despite acknowledgement that pathology analysis is pivotal in patient management, until now treatment strategies have been relatively crude, so extensive subcategorisation of tumours by genotype and phenotype has been unnecessary. Nonetheless, some subcategorisation is standard practice in pathology (exemplified by the important distinction between ER-positive versus ER-negative breast cancer or small cell and nonsmall cell lung cancer). More recently, there has been rapid progress in the move to individualise cancer therapy, based on the molecular drivers of disease. Thus, specific gene-alteration-based treatment is available for patients with metastatic breast cancer believed to be driven by amplification and overexpression of ERBB2, using trastuzumab ( Herceptin; Roche, Switzerland). In another example, carboplatin-based therapy trials are underway for BRCA1 and BRCA2-associated breast cancers, due to the effect of mutation in these genes on double-stranded DNA repair. It is possible that, if such trials prove effective in known mutation carriers, there may be implications for treatment of sporadic cancers with phenotypes similar to BRCA1 and BRCA2. Further, the advent of microarray-based analysis of tumours, including breast cancer, is likely to stratify cancers more accurately, since this technology is able to analyse the expression of thousands of gene sequences simultaneously per tumour and enhance our repertoire of diagnostic probes. The few studies performed on breast cancer using this powerful technology to date, and our own unpublished data, would predict that poor-outcome breast cancers are likely to be driven primarily by aberrations in genes involved in cell cycle control.33 Finally, the molecular phenotype of BRCA1 and BRCA2 cancers can be used in conjunction with experimental approaches to elucidate the mechanisms of carcinogenesis induced by aberrant function of these genes. Both BRCA1 and BRCA2 are involved in the maintenance of genomic integrity.34,35 However, given the experimental data indicating the anti-proliferation properties of wild-type BRCA1 36,37 and the notable high mitotic rate in BRCA1 cancers, it is likely that the BRCA1 mutation itself leads to the key event of oncogenesis, that of uncontrolled cellular proliferation. Once early inactivation of the normal BRCA1 allele and p53 stabilisation occurs, there is little selection pressure for the recruitment of other genes involved in oncogenesis. In contrast, the more heterogeneous phenotype of BRCA2-associated cancers, and the less striking proliferation rate, would support the hypothesis that a defect in DNA repair mechanisms would drive the pathogenesis of BRCA2 breast cancers. The more variable phenotype could then be explained by dependence on the aberrant function of other genes with unrepaired mutations. Clearly, the pathology of inherited breast cancer highlights the current inextricable interaction between the diagnostic discipline of pathology and molecular pathogenesis of disease. ACKNOWLEDGEMENTS The authors would like to thank Dr Gino Somers for his careful reading of the manuscript and Mr John Ciciulla for figure preparation. Sources of support were The Victorian Breast Cancer Research Consortium, National Health and Medical Research Council and the National Institutes of Health.
313
Address for correspondence: Associate Professor J. E. Armes, Department of Pathology, Royal Women’s Hospital, Grattan Street, Carlton, Vic 3053, Australia. E-mail: [email protected] u
References 1. Hopper JL, Southey MC, Dite GS, et al. Population-based estimate of the average age-specific cumulative risk of breast cancer for a defined set of protein-truncating mutations in BRCA1 and BRCA2. Australian Breast Cancer Family Study. Cancer Epidemiol Biomarkers Prev 1999; 8: 741–7. 2. Welcsh PL, Owens KN, King MC. Insights into the functions of BRCA1 and BRCA2. Trends Genet 2000; 16: 69–74. 3. Zheng L, Annab LA, Afshari CA, Lee WH, Boyer TG. BRCA1 mediates ligand-independent transcriptional repression of the estrogen receptor. Proc Natl Acad Sci USA 2001; 98: 9587– 92. 4. Ford D, Easton DF, Peto J. Estimates of the gene frequency of BRCA1 and its contribution to breast and ovarian cancer incidence. Am J Hum Genet 1995; 57: 1457–62. 5. Ford D, Easton DF, Stratton M, et al. Genetic heterogeneity and penetrance analysis of the BRCA1 and BRCA2 genes in breast cancer families. The Breast Cancer Linkage Consortium. Am J Hum Genet 1998; 62: 676–89. 6. Easton DF, Bishop DT, Ford D, Crockford GP. Genetic linkage analysis in familial breast and ovarian cancer: results from 214 families. The Breast Cancer Linkage Consortium. Am J Hum Genet 1993; 52: 678–701. 7. McCredie MR, Dite GS, Giles GG, Hopper JL. Breast cancer in Australian women under the age of 40. Cancer Causes Control 1998; 9: 189– 98. 8. Cui J, Hopper JL. Why are the majority of hereditary cases of earlyonset breast cancer sporadic? A simulation study. Cancer Epidemiol Biomarkers Prev 2000; 9: 805–12. 9. Couch FJ, Weber BL. Mutations and polymorphisms in the familial early-onset breast cancer ( BRCA1) gene. Breast Cancer Information Core. Hum Mutat 1996; 8: 8–18. 10. Rosen PP, Lesser ML, Senie RT, Kinne DW. Epidemiology of breast carcinoma III: relationship of family history to tumor type. Cancer 1982; 50: 171–9. 11. Lagios MD, Rose MR, Margolin FR. Tubular carcinoma of the breast: association with multicentricity, bilaterality, and family history of mammary carcinoma. Am J Clin Pathol 1980; 73: 25– 30. 12. Marcus JN, Watson P, Page DL, Lynch HT. Pathology and heredity of breast cancer in younger women. J Natl Cancer Inst Monogr 1994; 16: 23– 34. 13. Robson M, Gilewski T, Haas B, et al. BRCA-associated breast cancer in young women. J Clin Oncol 1998; 16: 1642– 9. 14. Lakhani SR, Jacquemier J, Sloane JP, et al. Multifactorial analysis of differences between sporadic breast cancers and cancers involving BRCA1 and BRCA2 mutations. J Natl Cancer Inst 1998; 90: 1138– 45. 15. Karp SE, Tonin PN, Begin LR, et al. Influence of BRCA1 mutations on nuclear grade and estrogen receptor status of breast carcinoma in Ashkenazi Jewish women. Cancer 1997; 80: 435–41. 16. Marcus JN, Watson P, Page DL, et al. BRCA2 hereditary breast cancer pathophenotype. Breast Cancer Res Treat 1997; 44: 275–7. 17. Eisinger F, Stoppa-Lyonnet D, Longy M, et al. Germ line mutation at BRCA1 affects the histoprognostic grade in hereditary breast cancer. Cancer Res 1996; 56: 471– 4. 18. Marcus JN, Watson P, Page DL, et al. Hereditary breast cancer: pathobiology, prognosis, and BRCA1 and BRCA2 gene linkage. Cancer 1996; 77: 697–709. 19. Armes JE, Egan AJ, Southey MC, et al. The histologic phenotypes of breast carcinoma occurring before age 40 years in women with and without BRCA1 or BRCA2 germline mutations: a population-base d study. Cancer 1998; 83: 2335– 45. 20. Armes JE, Trute L, White D, et al. Distinct molecular pathogeneses of early-onset breast cancers in BRCA1 and BRCA2 mutation carriers: a population-based study. Cancer Res 1999; 59: 2011–7. 21. Ridolfi RL, Rosen PP, Port A, Kinne D, Mike V. Medullary carcinoma of the breast: a clinicopathologic study with 10 year follow-up. Cancer 1977; 40: 1365– 85. 22. Breast Cancer Linkage Consortium. Pathology of familial breast cancer: differences between breast cancer in carriers of BRCA1 or BRCA2 mutations and sporadic cases. Lancet 1997; 349: 1505–10. 23. Elston CW. Grading of invasive carcinoma of the breast. In: Page DL, Anderson TJ, editors. Diagnostic histopathology of the breast. Edinburgh: Churchill Livingstone, 1987; 300–11.
314
24. Agnarsson BA, Jonasson JG, Bjornsdottir IB, Barkardottir RB, Egilsson V, Sigurdsson H. Inherited BRCA2 mutation associated with high grade breast cancer. Breast Cancer Res Treat 1998; 47: 121–7. 25. Struewing JP, Hartge P, Wacholder S, et al. The risk of cancer associated with specific mutations of BRCA1 and BRCA2 among Ashkenazi Jews. N Engl J Med 1997; 336: 1401– 8. 26. Newman B, Mu H, Butler LM, Millikan RC, Moorman PG, King MC. Frequency of breast cancer attributable to BRCA1 in a populationbased series of American women. JAMA 1998; 279: 915– 21. 27. Phillips KA, Nichol K, Ozcelik H, et al. Frequency of p53 mutations in breast carcinomas from Ashkenazi Jewish carriers of BRCA1 mutations. J Natl Cancer Inst 1999; 91: 469–73. 28. Johannsson OT, Idvall I, Anderson C, et al. Tumour biological features of BRCA1-induced breast and ovarian cancer. Eur J Cancer 1997; 33: 362–71. 29. Robson M, Rajan P, Rosen PP, et al. BRCA-associated breast cancer: absence of a characteristic immunophenotype. Cancer Res 1998; 58: 1839– 42. 30. Puget N, Stoppa-Lyonnet D, Sinilnikova OM, et al. Screening for germ-line rearrangements and regulatory mutations in BRCA1 led to
Pathology Downloaded from informahealthcare.com by York University Libraries on 07/09/14 For personal use only.
Pathology ( 2002 ), 34, August
ARMES AND VENTER
31. 32. 33. 34. 35. 36. 37.
the identification of four new deletions. Cancer Res 1999; 59: 455– 61. Lalloo F, Evans DG. The pathology of familial breast cancer: Clinical and genetic counselling implications of breast cancer pathology. Breast Cancer Res 1999; 1: 48–51. MacGregor DP. Anatomical pathology. Med J Aust 2002; 176: 7. van ‘t Veer LJ, Dai H, van de Vijver MJ, et al. Gene expression profiling predicts clinical outcome of breast cancer. Nature 2002; 415: 530– 6. Sharan SK, Morimatsu M, Albrecht U, et al. Embryonic lethality and radiation hypersensitivity mediated by Rad51 in mice lacking Brca2. Nature 1997; 386: 804–10. Scully R, Anderson SF, Chao DM, et al. BRCA1 is a component of the RNA polymerase II holoenzyme. Proc Natl Acad Sci USA 1997; 94: 5605– 10. Holt JT, Thompson ME, Szabo C, et al. Growth retardation and tumour inhibition by BRCA1. Nat Genet 1996; 12: 298– 302. Thompson ME, Jensen RA, Obermiller PS, Page DL, Holt JT. Decreased expression of BRCA1 accelerates growth and is often present during sporadic breast cancer progression. Nat Genet 1995; 9: 444– 50.