1. Department of Medical Oncology and Hereditary Cancer Clinic, Prince of Wales Hospital Randwick, NSW
2. Gynaecologic Oncology Centre, Royal Hospital for Women, Randwick, NSW
3. Department of Medical Oncology and Cancer Family Clinic, Peter MacCallum Cancer Institute, Melbourne, VIC
Epidemiological studies have identified family history as one of the major risk factors for epithelial ovarian cancer (subsequently referred to as “ovarian cancer”) and in recent years we have gained a greater understanding of the specific genes associated with inherited susceptibility to the disease1,2,3. This has led to the ability to identify individuals and families who are at significantly increased risk of developing ovarian cancer. These women require counselling about their risks and advice as to what can be done to possibly decrease risk as well as information regarding screening and early diagnosis. This is a specialised area that is rapidly evolving and we strongly recommend that women at increased genetic risk are referred to familial cancer clinics for multidisciplinary management which includes confirmation of the family history and diagnoses, counselling prior to genetic testing, communication and explanation of the results of genetic testing, advice about screening and prophylactic surgery and follow up in high risk management clinics4,5.
The management of these women at increased risk of ovarian cancer is difficult and presents many challenges. It should be recognised that much of our knowledge regarding the clinical applications and advice to at risk individuals is still preliminary and will undoubtedly change over time. The purpose of this paper is to briefly review the current state of knowledge about the genetics of hereditary ovarian cancer and update interested readers on the genetic epidemiology, prevalence and penertrance of the specific germline mutations associated with ovarian cancer, as well as review the clinical features and overall management of women at increased genetic risk of ovarian cancer.
The majority of women who develop epithelial ovarian cancer do not have a family history and have what is currently termed sporadic ovarian cancer, the causes of which are not well understood. The two major hereditary cancer syndromes associated with a substantially increased risk of ovarian cancer are the hereditary breast/ovarian cancer syndrome (HBOC) and hereditary non-polyposis colorectal cancer syndrome (HNPCC)2(table one). It should be noted that the HBOC syndrome also predisposes women to associated gynaecological cancers, specifically cancer of the fallopian tube and primary peritoneal cancer.
Approximately 10% of women who develop ovarian cancer have a germline mutation in either BRCA1 or BRCA2 with a smaller percentage having a mutation in one of the mismatch repair genes associated with HNPCC and are the focus of discussion6,7. There are some other rare genetic syndromes associated with increased risk of ovarian cancer, which are beyond the scope of this paper. Germline mutations in BRCA1 and BRCA2 account for the majority of families with hereditary ovarian cancer with about 60% being attributed to BRCA1 and 30% to BRCA2.
The likelihood of finding a mutation in a women with ovarian cancer increases with the number of blood relatives with ovarian or breast cancer in her family as well as ethnicity and other associated cancers (table two). It is not clear what proportions of ovarian cancer in unselected general populations are due to mutations in these genes but the percentage appears to be higher than previously anticipated. In a recent population-based series of 649 unselected incident cases of ovarian cancer diagnosed in Canada, 11.7% of women were found to have pathogenic mutations in BRCA1 and BRCA2, which is more than previous estimates and may even be higher as the genetic tests used would not be expected to detect all mutations8. Mutations were found in 19% of women with first-degree relatives with breast/ovarian cancer and in 6.5% of women reporting no family history. No mutations were found in women with borderline tumours, which is consistent with other studies, but there was an association with mutations and the histology of invasive ovarian cancer. Specifically, mutations were detected in 16.4% of serous cancers, 4.3% of endometrioid cancers and in none of the other histological subtypes8. This study needs to be confirmed as it raises the possibility that mutations in BRCA1 and BRCA2 may be more common in unselected populations with ovarian cancer than previously anticipated and it also challenges some of the epidemiological data relating family history to risk of ovarian cancer.
First degree relatives with one affected family member with ovarian cancer have generally been advised that they have a 5% lifetime risk which is about three times the 1.4% lifetime risk for women without a family history1,9,10. For women with two affected relatives the lifetime risk has been said to rise to about 7%, and these figures have been used to counsel women about risk1,8,9,10. However, the confidence values of these risk estimates are wide and studies such as the Canadian population study among others challenge these figures, which may be too simplistic8. It is essential that an accurate family history is taken in order to estimate risk of ovarian cancer based on family history and ethnicity should also be taken into account. Women of Ashkenazi Jewish background with a family history of even one case of ovarian cancer and women in the general population with two affected relatives with ovarian cancer should be considered to be at potentially higher risk and be referred to a familial cancer clilnic for further advice and counselling11,12.
The lifetime risk of ovarian cancer in women with mutations in BRCA1 and BRCA2 varies considerably in different studies and is dependent in part on the ascertainment of families in these studies. Recent results of pooled data from 22 studies demonstrated that women with BRCA1 mutations have a cumulative risk of 39% (confidence intervals 18-54%) of developing ovarian cancer by age 70 while women with BRCA2 mutations have a lower risk of 11% (CI 2.4-19%)13. There are age-related differences in the penetration of these genes. The mean age of onset of ovarian cancer in women with BRCA1 mutations is in the mid- to late-forties and for BRCA2-related ovarian cancers about a decade later. BRCA1 and particularly BRCA2-related ovarian cancer are very uncommon under 40 and ovarian cancers under 30 are rarely, if ever, associated with mutations in BRCA1 or BRCA213,14,15. These data have important practical significance as it influences the age at which screening should commence and the age at which prophylactic surgery should be considered. We find it difficult to reconcile these data regarding age of onset of ovarian cancer in mutation carriers with the general advice which has been promulgated by many committees around the world that screening begin at age 25-30 and prophylactic surgery be considered at completion of childbearing or at age 35 years16,17. While some may view our comments as heresy, it makes little sense to advise a screening test which is potentially associated with many false positives to young women with a mutation in say BRCA2 knowing that ovarian cancers in such women are rarely diagnosed under 40 and are usually only observed in the late 50s or early 60s. Even women with mutations in BRCA1 rarely develop ovarian cancer under 40 years old13. This is in distinct contrast to ovarian cancers that occur in association with HNPCC which are usually seen at a younger age18,19,20.
Extracolonic cancers occur in 69% of women with MSH2 mutations and 19% of women with MLH1 mutations. There is a 40% risk of uterine cancer in women with HNPCC and a 10% risk of ovarian cancer. The median age of ovarian cancer is 42 years with 30% of invasive ovarian cancer occurring in women under 40. In a review of 120 families entered on the HNPCC register in Victoria, the mean age of diagnosis was 48.3 years with a range of 29 to 74 years20. In contrast to BRCA1 and BRCA2-related ovarian cancers which are commonly high-grade serous cancers and at advanced stage at diagnosis, HNPCC-associated ovarian cancers are usually well to moderately-well differentiated and are more likely to be endometrioid or mucinous tumours and are often confined to the ovaries14,18,19,20. Furthermore, synchronous endometrial cancers are observed in 20% of women. These differences in biology should influence the advice given to women regarding age to commence screening and the age at which prophylactic surgery should be considered as well as the type of surgery that should be done.
Finally, before discussing management issues it is worthwhile to draw the reader’s attention to the importance of ethnicity and risk of BRCA1 and BRCA2 mutations. The overall prevalence of pathogenic mutations in BRCA1 and BRCA2 has been estimated to be 1 in 500 to 1 in 1000 although can be lower in certain populations1. A number of founder effects have been observed where the same mutation has been found in multiple, unrelated families and can be traced back to a common ancestor. Founder mutations have been identified in many populations, but have been particularly well studied in the Ashkenazi Jewish population (of central/eastern European ancestry) where three specific mutations (185 delAG and 5382 insC in BRCA1 and 6174 delT in BRCA2) are present in approximately 1 in 40 of the population21,22,23. A woman with ovarian cancer and Ashkenazi Jewish ancestry has a 40-60% chance of having a founder mutation identified in BRCA1 or BRCA2 and therefore her relatives are considered to be at potentially high risk and should be offered genetic counselling and testing if a germline mutation is identified in the proband. If the proband is not living it is still worth considering screening first-degree relatives for the three common founder mutations. These mutations may account for up to 25-30% of early onset breast cancers and up to 90% of cancers in families with both breast and ovarian cancers and Ashkenazi ancestry,12,23.
While we are now in the position to better identify women who are at increased genetic risk of ovarian cancer the challenge is how to manage them and their families. Of particular importance is whether it will be possible to modify risk and also be able to diagnose these cancers when they are still confined to the ovaries and would therefore be expected to have a good prognosis. The majority of women with epithelial ovarian cancer have advanced disease at initial presentation and only 25% are likely to survive five years24. In contrast, the majority of women with stage 1 ovarian cancer are cured and it has been assumed that early diagnosis would result in a better outcome. There has understandably been a lot of interest in ovarian cancer screening in both women at population risk and increased genetic risk, but there is no evidence to date that screening is associated with a reduction in mortality. There are currently three large population-based ovarian screening studies in progress which will be determine whether regular ultrasound screening and estimation of serum CA125 will result in early diagnosis and improve survival.
Screening studies to date have largely focused on determining the sensitivity and specificity of the various screening modalities. There are problems with specificity and positive predictive value of the screening tests and many women undergo unnecessary surgical exploration25,26,27. It is beyond the scope of this paper to review this in detail, but it is clear that there are inherent problems associated with screening tools currently available and we also do not know enough about the natural history of ovarian cancer. While it may sound counterintuitive we really don’t know if there is a stepwise progression from stage 1 to 2 to 3 and so on, and if so, what the sojourn time in each stage is. We have not as yet convincingly identified a precursor lesion for serous ovarian cancer, which is the most common histological subtype in the general population as well as being almost exclusively seen in hereditary ovarian cancer. There are very real differences between the various histological subtypes of epithelial ovarian cancer with respect to natural history and biological behavior which seem to be only appreciated by a small number of people with a particular interest in the biology of ovarian cancer24,28,29. For example, endometrioid and mucinous cancers are usually confined to the ovaries at initial clinical presentation and have an excellent prognosis. While these may be detected by screening it is unlikely that this will improve their already very good prognosis. In contrast, serous cancers are usually poorly differentiated and widely disseminated within the peritoneal cavity at diagnosis and only 20% are stage 124. Most of the stage 1 serous cancers are well differentiated and may have progressed from serous borderline tumours and biologically are likely to be a distinct subgroup30. Indeed, Singer et al have proposed a dualistic model for ovarian serous carcinogenesis with one pathway involving stepwise progression from serous borderline tumours to invasive well differentiated micropapillary serous cancer, and the other more common pathway, from ovarian surface epithelium or inclusion cysts to high grade serous cancer, which seems to develop rapidly and is unlikely to be detected at an “early” phase with our current technology30. A good understanding of ovarian cancer biology is essential if we are to be able to design screening studies and interpret the studies that have been published. A major limitation in all the studies that have been published is the difficulty the interested reader has in trying to determine the histological subtype and grade of screen detected cancers or even their stage. It is not uncommon for the tumours to be listed as “adenocarcinomas” or just stage 1 ovarian cancer, which is inadequate and limits interpretation of the studies. There has also been a tendency to lump all tumours together including non-epithelial subtypes such as granulosa cell tumours and germ cell tumours.
It is important that these limitations are appreciated and also conveyed to women at increased genetic risk of ovarian cancer who are undergoing regular screening. All the guidelines recommend regular ovarian cancer screening, with the usual caveats, and have generally recommended annual screening commence at ages 25-3016,17,31. This has been modified in the recently revised Australian guidelines with the appreciation of the ages at which ovarian cancer is likely to develop in women who have germline mutations in either BRCA1 or 2. We would encourage high-risk women to consider participating in the GOG 199 study which will be carried out by the Australian and New Zealand Gynecological Oncology Group (ANZGOG). Women who opt for screening will have a three-monthly CA125 and a yearly ultrasound and in addition blood samples will be stored for newer tests such as proteomics. The mathematical algorithm developed by Skates et al which takes into account the individual’s age and the rate of change in CA125 (ROCA) together with the absolute CA125 value and allows women to be triaged into normal, intermediate or elevated risk groups32. It is thought to improve the sensitivity and specificity of CA125 as a screening test as it appears to reduce false positives and false negatives that would be generated by a single CA125 set at a cut off level of 35 U/ml. The GOG 199 study is not a randomised study and will also include women who opt for prophylactic or risk-reducing bilateral salpingo-oophorectomy (RRBSO) who will have the surgery performed following a strict protocol and careful pathological assessment. All women will be followed, and in addition will have regular quality of life assessments. This study should allow us to determine what the value of screening is in a high-risk population.
There have been a number of reports of screening in high-risk populations. Dorum et al reported on the results of a screening study of 803 women33. Of the 16 ovarian cancers diagnosed five were stage 1 or borderline tumours, one was stage 2 and 10 were stage 3. Liede et al reported on a study that included 290 Jewish women and they diagnosed eight incident cases of ovarian or related cancers (fallopian tube or peritoneal cancer)34. Only three were diagnosed at screening and five of the eight women presented with symptoms between screening visits and had advanced cancers. Taylor and Schwartz reported the results of screening 252 high-risk women and diagnosed two advanced cancers, one at screening and one in a symptomatic woman35. It is clear from reviewing all the published studies of screening in high-risk women that ovarian cancers are diagnosed relatively infrequently and that most of these women have advanced cancers at diagnosis. In contrast, about 4% of women who have a RRBSO are found with evidence of occult cancers after careful pathological assessment and has been reported to be as high as 13%36-39. Microscopic foci of serous cancer have been found in the fallopian tubes, ovaries and at times also in peritoneal washings, and most of these women have had negative screening tests in the 6-12 months preceding surgery. There are recent studies to show that RRBSO in high-risk women significantly reduces the risk of subsequently developing both ovarian cancer and breast cancer (table three). Rebbeck et al reported a 96% reduction in risk of BRCA-related gynecologic cancers in 259 women who had undergone BSO compared with 292 matched mutation carriers who had not had surgery40. Six ovarian cancers were identified at the time of prophylactic surgery and all were stage 1. This should be contrasted and compared with the 20% incidence of ovarian cancer in the control group, 11% of which were stage 1. In this study there was a 53% reduction in the risk of breast cancer as well in mutation carriers who had a BSO compared to those who had not. The mean age at BSO was 42 years. In a prospective study reported by Kauff et al there was a 75% reduction in risk of BRCA associated gynaecologic cancer or breast cancer in 98 mutation carriers who underwent RRBSO compared with 72 mutation carriers who opted for screening41.
It has been appreciated for some years that women who undergo RRBSO may be at risk of subsequently developing peritoneal cancer as the ovarian surface epithelium and peritoneal mesothelium are both derived form coelomic epithelium42. The risk of this appears to be much lower than previously thought and only two women in the Rebbeck study subsequently were diagnosed with a peritoneal cancer at 3.8 years and 8.6 years after prophylactic surgery, and one out of 98 women who had prophylactic surgery in the Kauff study was subsequently diagnosed with peritoneal cancer40,41. The reasons for this difference are unclear but could be explained by a number of factors including careful pathological assessment at the time of initial surgery including peritoneal washings and complete removal of the ovaries and both fallopian tubes. It is likely that just clamping the broad ligament and removing the ovary as has occurred in the past would leave ovarian tissue remnants in situ and increase the risk of subsequent “peritoneal“ cancer.
The available data strongly suggests that women at increased genetic risk should be counseled about prophylactic surgery. As we have mentioned above the risk of ovarian cancer is very uncommon in women under 40 and women with known BRCA1 mutations could have a RRBSO in her late 30s or possibly early 40s depending on her family history and age of onset of ovarian cancer13,43. Women with BRCA2 mutations generally only develop ovarian cancer when they are post-menopausal and RRBSO could arguably be delayed until the late 40s or even early 50s in some women. It is difficult to be too prescriptive and the decision regarding timing of surgery can be only made after detailed discussion with the individual woman. The age of onset of ovarian cancer in her family needs to be considered as well as the possibility of genetic anticipation as cancers in subsequent generations tending to occur at a younger age. A small study of anticipation in hereditary breast cancer was recently published by Dagan et al who found that among women with a BRCA1 mutation there was no real difference in age of onset of breast cancer in mothers and daughters, but mutations in BRCA2 were associated with a significantly younger age of breast cancer in the second generation suggesting specific gene environment interactions44. The importance of genetic anticipation in BRCA 1 and BRCA2-related ovarian cancer is unclear at present.
The use of hormonal replacement in these women following RRBSO is controversial as these women are also at increased genetic risk of breast cancer and should be individualised with short-term use considered only in women with acute troublesome menopausal symptoms45. It is of course important to try to prevent some of the other possible consequences of premature menopause such as osteoporosis46. It has been suggested that in women in whom hormone replacement therapy is likely to be requested or used that a hysterectomy is performed as well as a BSO. Estrogen replacement alone is used in women without a uterus rather than combined estrogen and progestagens, which appear to be associated with higher risk of thromboembolic disease as well as breast cancer45. This is still an area for which we have no clear answers and detailed individual discussion is required.
Chemoprevention is likely to be a far more acceptable approach to management of high risk women and ideally modification of risk through lifestyle/environmental changes would also be important if we knew what modifications made a difference and at what age they should be introduced. Epidemiological studies have strongly suggested that the oral contraceptive pill (OCP) is associated with a significant reduction in risk of ovarian cancer in the general population47,48. Similarly, case control studies have demonstrated that women who have a family history of breast/ovarian cancer also have a similar reduction in ovarian cancer risk of about 50% if they have been on the OCP49. However, this was not confirmed in a population study of Jewish women in Israel, which raises some questions regarding the value of OCP in reducing ovarian cancer incidence in the high-risk population50. Of particular concern as well is the uncertainty regarding the risk of the OCP on breast cancer risk in women who have germ line mutations in BRCA1 or BRCA251. This also needs to be discussed in detail rather than recommending the OCP to all high-risk women to reduce risk of ovarian cancer. Tubal ligation, interestingly, has been reported to be associated with a reduction in risk of ovarian cancer in the general population as well as in a study of women with known BRCA1 mutations but not BRCA2 mutations52,53. The sample size of women with BRCA2 mutations was small and so one cannot be confident in assuming that they would not benefit and this requires more study. While the mechanism is not known it is of interest that Piek et al have recently reported that up to 36% of women with mutations in BRCA1 or 2 have dysphasic changes noted in the tubal epithelium and he has postulated that most serous “ovarian” cancers arise in the tubal epithelium and “spill over” onto the ovarian surface54. Supporting this is the relatively common finding of occult cancers in the fallopian tube of women at the time of RRBSO. While the mechanism for the protective benefit of tubal ligation is not known it is tempting to speculate that tubal ligation could alter the micro-environment in the tube and in addition there is vascular impairment and this could result in reduction in risk. This is being further studied but it would seem reasonable to recommend tubal ligation as a means of contraception to high risk women after completion of childbearing and before RRBSO is performed at a later age.
The aim of this paper was to briefly review and summarise a large and complex literature that is rapidly growing. Our increasing ability to identify women at increased genetic risk of developing ovarian cancer has not been matched by good level 1 evidence regarding how best we manage these women and their families. We have briefly discussed the natural history and biology of ovarian cancer, the appeal of screening and early diagnosis as well as the limitations of screening using currently available screening tests. The most convincing data regarding reducing risk is seen in the studies of RRBSO, but clearly this is not an attractive option to many women, and alternative strategies are needed. We have not dealt with the large literature on the psychological impact of being a mutation carrier and the many issues that these women and their families face. This should not be interpreted as lack of importance but rather a lack of space to cover a large and important literature, and the interested reader is directed to papers on this55-60. The management guidelines and advice about familial aspects of ovarian cancer recently have been updated and will provide health professionals with a detailed overview of familial ovarian cancer.
3. AS Whittemore, R Harris, J Itnyre. “Collaborative Ovarian Cancer Group: Characteristics relating to ovarian cancer risk: Collaborative analysis of 12 U.S. case-control studies. II. Invasive epithelial ovarian cancers in white women.” Am J Epidemiol, 136 (1992): 1184-203.
8. HA Risch, JR McLaughlin, DEC Cole, B Rosen, L Bradley, E Kwan. “Prevalence and penetrance of germline BRCA1 and BRCA2 mutations in a population series of 649 women with ovarian cancer.” Am J Hum Genet, 68 (2001): 700-10.
9. P Hartge, AS Whittemore, J Itnyre, L McGowan, D Cramer. “Rates and risks of ovarian cancer in subgroups of white women in the United States. The Collaborative Ovarian Cancer Group.” Obstet Gynecol, 84 (1994): 760-4.
10. AS Whittemore, R Harris, J Itnyre. “Collaborative Ovarian Cancer Group. Characteristics relating to ovarian cancer risk: collaborative analysis of 12 US case-control studies.” Am J Epidemiol. 136 (1992): 1184-203.
11. JP Struewing, P Hartge, S Wacholder, SM Baker, M Berlin, M McAdams. The risk of cancer associated with specific mutations of BRCA1 and BRCA2 among Ashkenazi Jews.” N Engl J Med, 336 (1997): 1401-8.
12. D Abeliovich, L Kaduri, I Lerer, N Weinberg, G Amir, M Sagi, et al. “The founder mutations 185delAG and 5382insC in BRCA1 and 6174delT in BRCA2 appear in 60% of ovarian cancer and 30% of early-onset breast cancer patients among Ashkenazi women.” Am J Hum Genet, 60, 3 (1997): 505-14.
13. A Antoniou, PDP Pharoah, S Narod, et al. “Average risks of breast and ovarian cancer associated with BRCA1 or BRCA2 mutations detected in case series unselected for family history: a combined analysis of 22 studies.” Am J Hum Genet, 2003.
14. PA Shaw, MT Deavers, GB Mills. “Clinical characteristics of genetically determined ovarian cancer.” In: I VG Vogel (ed) Management of patients at high risk for breast cancer, Blackwell Science, Malden, 2001, pp 94-107.
17. F Eisinger, N Alby, A Bremond, J Dauplat, et al. “Inserm ad hoc committee: Recommendations for the management of women with a genetic risk for developing cancer of the breast and/or the ovary” Cancer, 86, 3 (1999): 307-13.
19. P Watson, R Bützow, HT Lynch, J-P Mecklin, HJ Järvinen, HFA Vasen, et al. “The clinical features of ovarian cancer in hereditary nonpolyposis colorectal cancer.” Gynecol Oncol, 82, 2 (2001): 223-8.
20. GJ Brown, DJ St John, FA Macrae, K Aittomaki. “Cancer risk in young women at risk of hereditary nonpolyposis colorectal cancer: Implications for gynecologic surveillance.” Gynecol Oncol, 80, 3 (2001): 346-9.
21. E Levy-Lahad, R Catane, S Eisenberg, B Kaufman, et al. “Mutations in Ashkenazi Jews in Israel: Frequency and differential penetrance in ovarian cancer and in breast-ovarian cancer families.” Am J Hum Genet, 60 (1997): 1059-67.
22. AY Bahar, PJ Taylor, L Andrews, A Proos, et al. “The frequency of founder mutations in the BRCA1, BRCA2, and APC genes in Australian Ashkenazi Jews: implications for the generality of U.S. population data.” Cancer, 92, 2 (2001): 440-5.
29. M Pieretti, C Hopenhayn-Rich, NH Khattar, et al. “Heterogeneity of ovarian cancer: relationships among histological group, stage of disease, tumor markers, patient characteristics, and survival.” Cancer Invest, 20 (2002): 11-23.
43. CM Julian-Reynier, LJ Bouchard, DG Evans, FA Eisinger, et al. “Women’s attitudes toward preventive strategies for hereditary breast or ovarian carcinoma differ from one country to another: differences among English, French, and Canadian women.” Cancer, 92, 4 (2001): 959-68.
45. I Persson, E Weiderpass, L Bergkvist, R Bergstrom, C Schairer. “Risks of breast and endometrial cancer after estrogen and estrogen-progestin replacement.” Cancer Causes Control, 10, 4 (1999): 253-60.
50. B Modan, P Hartge, G Hirsh-Yechezkel, A Chetrit, et al. “Parity, oral contraceptives, and the risk of ovarian cancer among carriers and noncarriers of a BRCA1 or BRCA2 mutation.” N Engl J Med, 345, 4 (2001): 235-40.
53. KA Rosenblatt, DB Thomas. “Reduced risk of ovarian cancer in women with a tubal ligation or hysterectomy. The World Health Organization Collaborative Study of Neoplasia and Steroid Contraceptives.” Cancer Epidemiol Biomarkers Prev, 5, 11 (1996): 933-5.
55. K Tiller, B Meiser, P Butow, M Clifton, B Thewes, M Friedlander, K Tucker. “Psychological impact of prophylactic oophorectomy in women at increased risk of developing ovarian cancer: a prospective study.” Gynecol Oncol, 86, 2 (2002): 212-9.
57. B Meiser, P Butow, A Barratt, M Friedlander, et al. “Attitudes toward prophylactic oophorectomy and screening utilization in women at increased risk of developing hereditary breast/ovarian cancer.” Gynecol Oncol, 75, 1 (1999): 122-9.
58. C Lerman, S Narod, K Schulman, C Hughes, et al. “BRCA1 testing in families with hereditary breast-ovarian cancer. A prospective study of patient decision making and outcomes.” JAMA, 275 (1996): 1885-92.