- Open Access
High penetrances of BRCA1 and BRCA2 mutations confirmed in a prospective series
Hereditary Cancer in Clinical Practice volume 8, Article number: 2 (2010)
Penetrances of BRCA1 and BRCA2 mutations have been derived from retrospective studies, implying the possibility of ascertainment biases to influence the results.
We have followed women at risk for breast and/or ovarian cancer for two decades, and report the prospectively observed age-related annual incidence rates to contract breast or ovarian cancer for women with deleterious BRCA1 or BRCA2 mutations based on 4830 observation years. Patients were grouped according to mutation, age and having/not having had previous cancer.
In women not having had previous cancer and aged 40-59 years, the annual incidence rate to contract breast or ovarian cancer in those having the most frequent BRCA1 founder mutations was 4.0%, for women in this age group and with less frequent BRCA1 mutations annual incidence rate was 5.9%, and for women with BRCA2 mutations 3.5%.
The observed figures may be used for genetic counseling of healthy mutation carriers in the respective age groups. The results may indicate that less frequent BRCA1 mutations have higher penetrances than BRCA1 founder mutations.
Mutations in the two genes BRCA1 and BRCA2 may cause breast or ovarian cancer. Estimates on penetrances have been based on retrospective studies and have arrived at diverging results.
Retrospective reports include the possibility of presenting the selection criteria as results . Some studies tested one affected proband and calculated on number of affected close relatives [2, 3]. In such studies, prevalences of close relatives to be mutation carriers, as well as age of onset of disease in the relatives, is related to ascertainment of the proband (age, complete or incomplete ascertainment), and the reported results reflect assumptions on these factors. Others reported findings based on mutation testing in extended families to avoid some of these confounders and arrived at higher penetrance estimates [4–7]. Reports including prospective series are limited [8–10].
The suggestion that genetic modifying factors of penetrance should underly the conflicting results has not been confirmed, but the possibility of three modifiers inferring hazard ratio (HR) of 1.1 to 1.3 in BRCA2 mutation carriers, and one locus with HR 1.1 for BRCA1 remain. The three loci are outside linkage distance for both BRCA1 and BRCA2 and will combined with the HRs reported have marginal impact on total risk estimates for relatives . Discussing the theoretical concepts of modifiers of penetrances and expressions in depth, as well as the pitfalls in examining for such factors, are outside the scope of this reports.
We have suggested that all incident cases of breast or ovarian cancer in Norway should be offered testing for the Norwegian deleterious BRCA1 and BRCA2 mutations. While considering our suggestion, the Norwegian Government asked for documentation of the penetrances of the mutations in question. We decided to analyze our prospective series to arrive at prospectively, empirically observed annual incidence rates to answer the question, and report our findings here. The main goal was to describe annual incidence rate in those who had had no cancer before inclusion and who had no cancer demonstrated at first (prevalence) round.
Materials and methods
We have subjected healthy women at risk for breast or ovarian cancer by family history to prospective follow-up for two decades. Details on ascertainment methods have been previously published [12–17]. Genetic testing was facilitated by the demonstration of Norwegian founder mutations and rapid and cheap tests to demonstrate them. BRCA founder mutations is not specific for Norway, but the mutations are different in different ethnic groups [6, 18]. Close to all families assumed by family history to have inherited breast or ovarian cancer have been tested for at least the 10 most frequent Norwegian mutations (for review, test panel and geographical distribution of the Norwegian founder mutations, see Møller et al. ).
The families were ascertained according to preset criteria as previously published . We examined at least one affected per family with sequencing and MLPA, if available. If no affected was available, we have in many families sequenced obligate male carriers or young daughters.
As mentioned above, most BRCA mutations demonstrated in Norway have passed through the bottleneck caused by the Bubonic plagues 25 generations ago, and the population parts have lived separately until a few generations ago . The consequence is that most mutations still are located to one single geographical area, and the families themselves are aware of this and give a precise description if asked. We have over two decades classified the families according to both geographical origin and present address of current members. Our electronic medical files include zip-codes for all women either with breast or ovarian cancer or at risk according to family history, and were used to identify families for geographical locations. Whenever a mutation not included in the general test panel was described, the family was extended as far as possible, the likely geographical origin(s) of the mutation determined, and all families connected to that area were tested for that mutation.
We included mutations causing direct stop, frameshift or large insertions/deletions and splice defects. It is agreed that such mutations are considered deleterious (see BIC: http://research.nhgri.nih.gov/bic/). To avoid discussions on validity of results, we excluded missense mutations in the finger domains considered deleterious by some , all forms of mutations in the last coding exon in BRCA2, and all intron variations outside +/- two bases from the exons considered to be splice donor or receptor sites. We did, however, include the BRCA c.1A>G mutation as it is locally frequent in one area and obviously follows the disease in all families affected (to be reported separately). A list of the deleterious mutations included in the present report, is posted on our web-site http://www.inherited-cancer.com.
Whenever a mutation was found, predictive testing was offered to the family members pending consent from the proband to do so. Compliance to predictive testing has been demonstrated to be high . Most families are expanded far beyond any selection criteria, many of them to distant relatives often more than 5 meioses apart from the index person. In one family, a de-novo BRCA2 c.8090_8115del16 mutation was demonstrated (both parents without mutation, paternity confirmed by DNA testing).
Mutation testing was performed under national legislation including genetic counseling and written informed consent for each single patient. No named information was exported from the medical files, no research registry including patient identifications was erected. The medical database and application was constructed by Oracle 10 g© and Delphi 2007© and the data for this report extracted by TOAD© by PM and NC.
All healthy women with a deleterious BRCA1 or BRCA2 mutation were offered annual follow-up from age 25 years on, aiming at early diagnosis and treatment. The same offer was given women with past or present cancer, if they were assumed to benefit.
The patients have been followed with annual examinations [12–17], and some have opted for risk-reducing surgery [17, 20]. Observation time was calculated as time between first and last recorded results. All cases of breast or ovarian cancers in the observation period were recorded as events irrespective of detection method, including interval cancers. No other cancer or any other disease was scored as event. Each woman was counted once only for having or not having breast or ovarian cancer prior to inclusion. Each woman was scored once only for cancer at first examination irrespective of how many tumours possibly detected. Each woman was scored once only for having or not having demonstrated breast or ovarian cancer at follow-up. In theory, one woman may have been scored for having cancer in all three groups mentioned, but not more than once in any group.
All data from the reporting centres including March 2009 are reported. Founder BRCA1 mutation series from Bergen was incomplete, and all BRCA1 founder mutation carriers from Bergen were excluded.
Observation time was censored at first demonstrated breast or ovarian cancer at follow-up. Observation time was censored at both bilateral risk-reducing mastectomy and oophorectomy. No correction was done for those having had only the one or the other risk-reducing intervention. As previously reported, ovarian cancer may have some time in preclinical detectable stage, and we did diagnose some through risk-reducing oophorectomy . Censoring the study at that time, might have given false high penetrance estimates. The way we did it, implies the risk of arriving at too low penetrance estimates, which we decided to prefer.
Age related annual incidence rates were calculated as number of women who contracted one or more cancer divided by numbers of observation years of women having the ages at first control as specified in Table 1 and Table 2.
Confidence intervals of means were considered by assuming Poisson distributions, groups were compared two-by-two by Fishers' exact p.
All together 1055 women were identified as having a deleterious BRCA1/2 mutation and examined once or more. At first (prevalence) round, 54 among them were diagnosed as having breast and/or ovarian cancer, arriving at a prevalence of 5.1%.
Among these, 870 were examined twice or more. Through 4830 follow-up years 147 among them were diagnosed as having breast and/or ovarian cancer, arriving at an overall annual incidence rate irrespective of age to be 3.0%.
These 870 were fractionated into three groups: a) the four BRCA1 founder mutations for which we have previously reported retrospective cumulative incidence rates, b) other BRCA1 mutations, and c) BRCA2 mutations. These three groups were split into two groups each: Those with no cancer prior to inclusion as well as with no cancer at first control, and those who did have a cancer prior to inclusion or demonstrated cancer at first control. The overall annual incidence rates irrespective of age in those who had no prior or prevalent cancer, were 2.7% for BRCA1 founder mutations, 3.6% for less frequent BRCA1 mutations, and 2.5% for BRCA2 mutations.
Then we fractionated each of the six groups defined above, into three groups each: those aged 25-39 years at inclusion, those aged 40-59 years at inclusion, and those aged 60 years or more at inclusion. The results for women without a previous cancer and no cancer at prevalence round are detailed in Table 1. In the age group 40-59 years at inclusion, annual incidence rates were 4.0% for BRCA1 founder mutations, 5.9% for less frequent BRCA1 mutations, and 3.5% for BRCA2 mutations. For all age groups, the annual incidence rates for women with founder mutations were less than for women with less frequent BRCA1 mutations. None of the differences between the three mutation groups were statistically significant.
The corresponding details for patients having had breast or ovarian cancer before inclusion, or at first control, are given in Table 2.
This prospective report confirms the previously reported penetrance estimates for the four most frequent BRCA1 mutations in Norway. As the healthy mutation carriers reported here are the next generation in the families previously reported, the finding was that disease continued to occur prospectively as previously reported retrospectively in these families.
The point estimates for annual incidence rates for the less frequent BRCA1 mutation carriers were higher than for those having a BRCA1 founder mutation. This may be a methodological artifact, as only one third of the less frequent BRCA1 mutations carried by the prospective cases, were included in the test panel applied to all families (data not given in results). However, as described in methods above we have over years searched for all the rare mutations in the geographical areas they are connected to, and have described a number of delineated areas with local founder mutations not present in the other population parts. The majority of families carrying rare mutations were not detected through sequencing a prospectively demonstrated cancer case: In the files at The Norwegian Radium Hospital which was available for detailed analysis, 547 women in 127 different kindreds were identified to have rare BRCA1 mutations, while the number of prospectively detected cancers in women with no cancer prior to or at first control and with rare BRCA1 mutations not included in the test panel used for all families, was 24. The influence of a selection error remains, but the probability that the selection problem had a major effect on the results may be limited.
An alternative speculation may be that there may have been an element of selection: The rare mutations may be rare because they have reduced fitness. If so, previous reports based on the most frequent mutations may be underestimates if applied to the less frequent mutations. If true, this may explain some of the discrepancies in the literature discussed above. The resources needed to clarify this, would imply full-scale sequencing in numbers which were not available to us and still not is. An answer may await new technology making sequencing of large numbers awailable and affordable.
We here report incidence rate as number of cases affected among the patients exposed during a given period of time , and grouped the patients according to age at first examination. As seen in the tables, the average follow-up period was about 6 years, meaning that the observed annual incidence rates reflected the risk per year of becoming affected the next 6 years for healthy women in the given age groups.
Retrospective reports are most often constructed differently: they report cumulative incidence rate by age, starting at birth. We did so in our previous retrospective reports [4, 7]. For comparison with the current results, we recalculated our previous figures to reflect age related annual incidences among those who had experienced no cancer before reaching the various age groups. Doing so, the annual incidence rates in the retrospective series were 1.0%, 3.0%, 3.5% and 2.5% for the age groups 30-39, 40-49, 50-59 and 60-69 years, respectively (see Heimdal et al , table three, coloumn C for the data used for calculations). This is close to what we prospectively observed in the present series. The younger group <40 years in the present series had an average age at inclusion of 32 years and was followed for an average of 6 years, meaning that the mean age was about 38 years and the annual incidence rate was expected to be higher than derived from the retrospective series for the age group 30-39. The lower annual incidence rate observed in the older group 60+ years may reflect a diminishing annual incidence rate in older ages. All variations discussed here were, however, statistically insignificant and may have been caused by random variation in small numbers.
Numbers included with a previous cancer or a cancer demonstrated at first round, were insufficient to draw a firm conclusion. The impression was, however, that those having had a cancer had the same risk for a new primary as those who had not had a previous cancer. This group was not the goal for the current study, and we do not discuss this further.
Penetrance observed for BRCA2 mutation carriers was similar to BRCA1 founder mutations. From the literature, we had expected lower. Both King et al. 2003 and Antoniou et al 2003 did report increasing penetrance of BRCA mutations today compared to previous generations. Our findings may support their notion. If this were to be true, it may explain some of the controversies in retrospective studies in different countries and ethnic groups. This adds to variation in penetrances and methodological problems, as discussed above, to explain controversies between retrospective reports.
Prospectively observed survival of patients with BRCA1 mutations and with breast or ovarian cancer, is 50% or less in 10 years [13, 15]. It follows that the risk of dying from breast or ovarian cancer is at least half the annual incidence rates of contracting cancer reported here. Prospects for better treatment modalities and prevention strategies are outside the aims of the current report and are not discussed here.
In conclusion, we have confirmed high penetrances of BRCA1 and BRCA2 mutations in a prospective series. The annual incidence rates presented are empirical observations which may be used for genetic counseling of healthy women a with stop, frameshift or large deletion/insertion mutation in the BRCA1 or BRCA2 genes. In the given age groups, our results may be considered annual risk of breast or ovarian cancer in the next 6 years to come. In addition, our findings may give reason to examine further whether or not rare BRCA1 mutations may be rare because they have lower fitness than the most frequent BRCA1 mutations.
Easton DF, Ford D, Bishop DT: Breast and ovarian cancer incidence in BRCA1 -mutation carriers. Breast Cancer Linkage Consortium. Am J Hum Genet 1995,56(1):265–71.
Antoniou A, Pharoah PD, Narod S, Risch HA, Eyfjord JE, Hopper JL, Loman N, Olsson H, Johannsson O, Borg A, Pasini B, Radice P, Manoukian S, Eccles DM, Tang N, Olah E, Anton-Culver H, Warner E, Lubinski J, Gronwald J, Gorski B, Tulinius H, Thorlacius S, Eerola H, Nevanlinna H, Syrjäkoski K, Kallioniemi OP, Thompson D, Evans C, Peto J, Lalloo F, Evans DG, Easton DF: 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 2003,72(5):1117–30. 10.1086/375033
Begg CB, Haile RW, Borg A, Malone KE, Concannon P, Thomas DC, Langholz B, Bernstein L, Olsen JH, Lynch CF, Anton-Culver H, Capanu M, Liang X, Hummer AJ, Sima C, Bernstein JL: Variation of breast cancer risk among BRCA1 /2 carriers. JAMA 2008,299(2):194–201. 10.1001/jama.2007.55-a
Dørum A, Heimdal K, Hovig E, Inganäs M, Møller P: Penetrances of BRCA1 1675delA and 1135insA with respect to breast and ovarian cancer. Am J Hum Genet 1999, 65: 671–679. 10.1086/302530
Einbeigi Z, Bergman A, Kindblom LG, Martinsson T, Meis-Kindblom JM, Nordling M, Suurküla M, Wahlström J, Wallgren A, Karlsson P: A founder mutation of the BRCA1 gene in Western Sweden associated with a high incidence of breast and ovarian cancer. Eur J Cancer 2001,37(15):1904–9. 10.1016/S0959-8049(01)00223-4
King MC, Marks JH, Mandell JB, New York Breast Cancer Study Group: Breast and ovarian cancer risks due to inherited mutations in BRCA1 and BRCA2 . Science 2003,302(5645):643–6. 10.1126/science.1088759
Heimdal K, Mæhle L, Apold J, Pedersen JC, Møller P: The Norwegian founder mutations in BRCA1 : high penetrance confirmed in an incident cancer series and differences observed in the risk of ovarian cancer. Europ J Cancer 2003, 39: 2205–2213. 10.1016/S0959-8049(03)00548-3
Evans DG, Lennard F, Pointon LJ, Ramus SJ, Gayther SA, Sodha N, Kwan-Lim GE, Leach MO, Warren R, Thompson D, Easton DF, Eeles R, UK Study of MRI Screening for Breast Cancer in Women at High Risk (MARIBS): Eligibility for magnetic resonance imaging screening in the United Kingdom: effect of strict selection criteria and anonymous DNA testing on breast cancer incidence in the MARIBS Study. Cancer Epidemiol Biomarkers Prev 2009,18(7):2123–31. 10.1158/1055-9965.EPI-09-0138
Meijers-Heijboer H, van Geel B, van Putten WL, Henzen-Logmans SC, Seynaeve C, Menke-Pluymers MB, Bartels CC, Verhoog LC, van den Ouweland AM, Niermeijer MF, Brekelmans CT, Klijn JG: Breast cancer after prophylactic bilateral mastectomy in women with a BRCA1 or BRCA2 mutation. N Engl J Med 2001,345(3):159–64. 10.1056/NEJM200107193450301
Evans DG, Shenton A, Woodward E, Lalloo F, Howell A, Maher ER: Penetrance estimates for BRCA1 and BRCA2 based on genetic testing in a Clinical Cancer Genetics service setting: risks of breast/ovarian cancer quoted should reflect the cancer burden in the family. BMC Cancer 2008, 8: 155. 10.1186/1471-2407-8-155
Antoniou AC, Spurdle AB, Sinilnikova OM, Healey S, Pooley KA, Schmutzler RK, Versmold B, Engel C, Meindl A, Arnold N, Hofmann W, Sutter C, et al.: Common breast cancer-predisposition alleles are associated with breast cancer risk in BRCA1 and BRCA2 mutation carriers. Am J Hum Genet 2008,82(4):937–48. 10.1016/j.ajhg.2008.02.008
Møller P, Evans G, Haites N, Vasen H, Reis MM, Anderson E, Apold J, Hodgson S, Eccles D, Olsson H, Stoppa-Lyonnet D, Chang-Claude J, et al.: Guidelines for follow-up of women at high risk for inherited breast cancer. Consensus statement from the Biomed2 Demonstration Programme on Inherited Breast Cancer. Dis Markers 1999, 15: 207–2011.
Evans GR, Gaarenstroom KN, Stirling D, Shenton A, Maehle L, Dørum A, Steel M, Lalloo F, Apold J, Porteous M, Vasen HF, van Asperen C, Moller P: Screening for Familial Ovarian Cancer: Poor survival of BRCA1 /2 related cancers. J Med Genet 2008, 46: 593–597. 10.1136/jmg.2008.058248
Hagen Anne I, Kvistad Kjell, Maehle Lovise, Holmen Marit, Aase Hildegunn, Styr Bodil, Vabø Anita, Apold Jaran, Skaane Per, Møller Pål: Sensitivity of MRI versus mammography to diagnose BRCA-associated breast cancer validated in a national prospective series. The Breast 2007,16(4):367–74. 10.1016/j.breast.2007.01.006
Moller P, Evans DG, Reis MM, Gregory H, Anderson E, Maehle L, Lalloo F, Howell A, Apold J, Clark N, Lucassen A, Steel CM: Surveillance for familial breast cancer: Differences in outcome according to BRCA mutation status. Int J Cancer 2007,121(5):1017–20. 10.1002/ijc.22789
Møller P, Hagen AI, Apold J, Maehle L, Clark N, Fiane B, Løvslett K, Hovig E, Vabø A: Genetic epidemiology of BRCA mutation - family history identifies less than 50% of mutation carriers. Eur J Cancer 2007,43(11):1713–7. 10.1016/j.ejca.2007.04.023
Mæhle L, Apold J, Paulsen T, Hagen B, Løvslett K, Fiane B, vGhelue M, Clark N, Møller P: High risk for ovarian cancer in a prospective series is restricted to BRCA mutation carriers. Clin Cancer Res 2008,14(22):7569–73. 10.1158/1078-0432.CCR-08-0112
Górski B, Jakubowska A, Huzarski T, Byrski T, Gronwald J, Grzybowska E, Mackiewicz A, Stawicka M, Bebenek M, Sorokin D, Fiszer-Maliszewska Ł, Haus O, Janiszewska H, Niepsuj S, Góźdź S, Zaremba L, Posmyk M, Płuzañska M, Kilar E, Czudowska D, Waśko B, Miturski R, Kowalczyk JR, Urbañski K, Szwiec M, Koc J, Debniak B, Rozmiarek A, Debniak T, Cybulski C, Kowalska E, Tołoczko-Grabarek A, Zajaczek S, Menkiszak J, Medrek K, Masojæ B, Mierzejewski M, Narod SA, Lubiñski J: A high proportion of founder BRCA1 mutations in Polish breast cancer families. Int J Cancer 2004,110(5):683–6. 10.1002/ijc.20162
Bodd TL, Reichelt J, Heimdal K, Møller P: Uptake of BRCA1 genetic testing in adult sisters and daughters of known mutation carriers in Norway. J Genet Counselling 2003, 5: 405–417. 10.1023/A:1025864703405
Evans G, Baildam A, Brain A, Anderson E, Shenton A, Vasen HF, Eccles D, Lucassen AM, Pichert G, Hamed H, Møller P, Mahle L, Morrison PJ, Stoppat-Lyonnet D, Gregory H, Smyth E, Niederacher D, Nestle-Krämling C, Campbell J, Hopwood P, Lalloo F, Howell A: Risk reducing mastectomy: outcomes in 10 European Centres. J Med Genet 2008, 46: 254–8. 10.1136/jmg.2008.062232
Hill AB: Principles of medical statistics. The Lancet Limited, London; 1971:327.
The authors declare that they have no competing interests.
The study was designed by PM. Data were collected by by PM, LM, LFE, TL, CJ, JA and AV. IT systems were constructed and data for current analysis extracted by PM and NC. The manuscript was written and approved by all authors.
About this article
Cite this article
Møller, P., Mæhle, L., Engebretsen, L.F. et al. High penetrances of BRCA1 and BRCA2 mutations confirmed in a prospective series. Hered Cancer Clin Pract 8, 2 (2010). https://doi.org/10.1186/1897-4287-8-2
- Ovarian Cancer
- Mutation Carrier
- BRCA1 Mutation
- BRCA2 Mutation Carrier
- Founder Mutation