|
Advertisement |
|
|
||
 |
|
|||||
|
|||||
|
 |
||||||
|
||||||
|
||||||
|
© 2001 American Society for Clinical Oncology Germline Mutations in BRCA1 and BRCA2 in Breast-Ovarian Families From a Breast Cancer Risk Evaluation ClinicFrom the Divisions of Hematology and Oncology, Medical Genetics, Department of Medicine, Cancer Center, and Abramson Family Cancer Research Institute, University of Pennsylvania School of Medicine, Philadelphia, PA; and Institute of Cancer Research, Sutton, Surrey, United Kingdom. Address reprint requests to K.L. Nathanson, MD, Rm 513, BRBII/III, University of Pennsylvania School of Medicine, 421 Curie Blvd, Philadelphia, PA 19104; email: knathans{at}mail.med.upenn.edu
PURPOSE: Data from the Breast Cancer Linkage Consortium suggest that the proportion of familial breast and ovarian cancers linked to BRCA1 or BRCA2 may be as high as 98% depending on the characteristics of the families, suggesting that mutations in BRCA1 or BRCA2 may entirely account for hereditary breast and ovarian cancer families. We sought to determine what proportion of families with both breast and ovarian cancers seen in a breast cancer risk evaluation clinic are accounted for by coding region germline mutations in BRCA1 and BRCA2 as compared to a linkage study group. We also evaluated what clinical parameters were predictive of mutation status. PATIENTS AND METHODS: Affected women from 100 families with at least one case of breast cancer and at least one case of ovarian cancer in the same lineage were screened for germline mutations in the entire coding regions of BRCA1 and BRCA2 by conformation-sensitive gel electrophoresis, apolymerase chain reaction–based heteroduplex analysis, or direct sequencing. RESULTS: Unequivocal deleterious mutations were found in 55% (55 of 100) of the families studied. Mutations in BRCA1 and BRCA2 accounted for 80% and 20% of the mutations overall, respectively. Using multivariate analysis, the strongest predictors of detecting a mutation in BRCA1 or BRCA2 in this study group were the presence of a single family member with both breast and ovarian cancer (P < .0009; odds ratio [OR], 5.68; 95% confidence interval [CI], 2.04 to 15.76) and a young average age at breast cancer diagnosis in the family (P < .0016; OR, 1.69; 95% CI, 1.23 to 2.38). CONCLUSION: These results suggest that at least half of breast/ovarian families evaluated in a high-risk cancer evaluation clinic may have germline mutations in BRCA1 or BRCA2. Whether the remaining families have mutations in noncoding regions in BRCA1, mutations in other, as-yet-unidentified, low-penetrance susceptibility genes, or represent chance clustering remains to be determined.
BREAST CANCER IS a complex, multifactorial disease in which there is thought to be a strong interplay between genetic and environmental factors. Two genes that, when mutated, confer a high risk of breast cancer have been isolated, namely BRCA1 and BRCA2. Previous studies using linkage analysis suggest that germline mutations in BRCA1 or BRCA2 account for greater than 90% of rigorously defined families with both breast and ovarian cancer.1-3 However, the frequency of mutations in both genes among families with both breast and ovarian cancer attending a high-risk clinic, a more clinically relevant study group than families ascertained for linkage analysis, has not been determined. It is clear that the in the presence of ovarian cancer, mutation frequencies for BRCA1 and BRCA2 are high. Gayther et al4 found 43% of families ascertained for multiple cases of ovarian cancer had a germline BRCA1 or BRCA2 mutation. In another study in which women were ascertained on the basis of Ashkenazi Jewish ancestry and prevalent cases of ovarian cancer, mutations were identified in 41% of families.5 Even in a cohort of 18 Ashkenazi Jewish women with ovarian cancer and minimal or no family history of breast cancer or ovarian cancer, founder mutations were identified in seven (39%) of 18.6 Thus, women with a personal or family history of breast or ovarian cancer represent a study group with a high likelihood of having BRCA1 or BRCA2 mutations, necessitating accurate estimates of mutation frequencies for counseling purposes as these women seek risk assessment. This analysis of 100 families with at least one breast cancer and one ovarian cancer identified in the high-risk clinic setting provides a means of estimating the frequency and predictors of mutations in this clinically relevant population.
Ascertainment Families were eligible for this study if they had a single breast cancer and a single ovarian cancer in the same lineage; cancer diagnoses could be made at any age. The families were ascertained either through the cancer risk evaluation programs at the University of Pennsylvania (1994 to 1998) or at the University of Michigan (1993 to 1994). Families were either self-referred or physician referred for the evaluation of a perceived risk of breast or ovarian cancer based on family history. All cancer diagnoses were confirmed by pathology reports for probands and in relatives when available. Ethnic origin was self-reported. Samples were consecutively collected between 1993 and 1998 with no recruitment or advertising strategy. Families with male breast cancer were included only if there was also a female breast cancer case in the same lineage. One hundred families met these criteria ( Table 1). DNA samples were collected from at least one affected member of each family, and the sample was used for screening purposes. In cases in which more than one sample was available for testing, the woman with the earliest age at diagnosis of breast cancer was selected. Three probands had breast and ovarian cancers on both sides of the family. In these cases, the lineage counted for analytic purposes was randomly determined by flipping a coin. The average distance between the tested proband and the ovarian cancer case in the family is depicted in Table 2. The study was reviewed and approved by the institutional review boards of both institutions.
Mutation Analysis DNA was extracted from peripheral-blood mononuclear cells and stored in Tris EDTA (TE) at 4°C. The entire 22 exon-coding domain and flanking splice site regions of BRCA1 were amplified by 31 sets of polymerase chain reaction (PCR) primers. One set of primers was used to amplify each exon and the adjacent intronic region except for exon 11, in which 10 sets of overlapping primer sets were used to amplify the entire exon. Similarly, the 26 coding exons of BRCA2 were amplified using 47 sets of intron-based primers, with overlapping primer sets for exons 10 and 11. All primer sequences are available through the Breast Cancer Information Core Web site (http://www.nhgri.nih.gov/Intramural_research/Lab_transfer/Bic/index.html) and are available on request. PCR amplification was performed in a final volume of 20 µL containing 80 ng of DNA, 1.5 mmol/L MgCl2, 10 mmol/L Tris HCl (pH 8.3), 50 mmol/L KCl, 0.2 mmol/L each of dCTP, dATP, dTTP, and dGTP (Amersham Pharmacia Biotech, Piscataway, NJ), each primer at 1.0 µmol/L, and 1.0 unit of Taq polymerase (Roche Diagnostics, Indianapolis, IN). Amplifications were for 20 cycles in a 9700 Thermocycler (PE Applied Biosystems, Foster City, CA). For generation of fluorescent PCR products, each forward primer was labeled at the 5'-end with one of the following phosphoramidite dyes: HEX (green), 6-FAM (blue), and NED (yellow). None of the reverse primers was labeled. All reactions were cold-started with an initial denaturation of 94°C for 1 minute, followed by a touchdown of 1°C per minute beginning 10°C above the annealing temperature. Annealing temperatures were optimized for each primer set and ranged from 55°C to 60°C. Elongation was at 72°C and PCR products were analyzed by conformation-sensitive gel electrophoresis (CSGE) analysis, using either a conventional protocol7 or a recently developed fluorescence-based protocol.8
Conventional CSGE
Fluorescent CSGE
Sequence Analysis
Statistical Analysis
Both univariate and multivariate analyses were used to examine the associations between specific familial characteristics (phenotypes) and the presence of a mutation in BRCA1, BRCA2, or both (genotype). The following potential predictor variables were examined: number of female breast cancers in a family, number of ovarian cancers in a family, number of breast and ovarian cancer multiple primary cases, the average age at diagnosis of breast cancer, the average age at diagnosis of ovarian cancer, Ashkenazi Jewish ancestry, average age at breast cancer diagnosis for the proband, and average age at ovarian cancer diagnosis for the proband. All variables were analyzed as ordinal or continuous variables and as categorical variables. The average age at diagnosis of breast or ovarian cancer in each family was calculated by dividing the variable into 5-year categories (< 35, 35 to 39, 40 to 44, 45 to 49, 50 to 54, 55 to 59, and > 60). All numbers were treated as categorical for the multivariate analysis. For univariate analyses, the Kruskal-Wallis
Unequivocal disease-associated mutations in BRCA1 and BRCA2 were found in 55 (55%) of 100 families. Twenty-five different mutations in BRCA1 were identified in 45 families ( Fig 1A). Of the recurrent mutations found in BRCA1, 185delAG was identified in 16 families (all known Ashkenazi Jewish families), 5382insC in six families, 3875del4 in three families, and Y978X in two families. To our knowledge, all families were unrelated to one another. These four mutations account for 61% of the mutations found in BRCA1 in this study. Twenty-four of 25 BRCA1 mutations were frameshift or nonsense mutations; the remaining mutation was a splice-site mutation that disrupts a conserved splice-site acceptor sequence (IVS20-1 G  A). One family (no. 961) had disease-associated mutations in both BRCA1 and BRCA2 (3888delGA and 6174delT), as previously described.9
Eleven (11%) of 100 families had disease-associated mutations in BRCA2; an additional six families had variants of uncertain significance in BRCA2. Of the 11 families with deleterious mutations, six had 6174delT (two of these families were of known Ashkenazi Jewish ancestry). Ten of the 11 mutations were frameshift mutations; a single family had a mutation in a conserved splice-site donor consensus sequence (IVS13-2 A G) (Fig 1B). Five variants of uncertain significance were found in BRCA2, each only once ( Table 3). None of these variants was included in our statistical analyses.
Predictors of Mutation Status To evaluate predictors of detecting a mutation, a number of parameters were analyzed. Variables were analyzed both as categorical variables ( Table 4) and as continuous or ordinal variables (data not shown); all of the results were consistent in the two types of analyses. BRCA1 mutation status was associated with the number of ovarian cancer cases per family (P = .011). However, the number of ovarian cancer cases did not predict BRCA2 or overall (BRCA1 and BRCA2) mutation status. The presence and number of individuals with both breast and ovarian cancers within a family were strongly predictive of BRCA1 mutation status (P = .006) and overall mutation status (P = .0009). The presence of a family member with both breast and ovarian cancer was the only significant variable predictive of a BRCA2 mutation when BRCA2-positive families were compared with families without detectable coding region mutations (P = .0162; OR, 6.0; 95% CI, 1.39 to 25.86). The numbers of breast cancer cases in the family and ethnicity (Ashkenazi Jewish v non-Ashkenazi Jewish) were not significantly associated with mutation status (P = .293 and P = .142, respectively).
Average age at breast cancer diagnosis within the family was predictive of both BRCA1 mutation (P = .0001) and overall mutation status (P = .006). For each 5-year increase in average age at breast cancer diagnosis, there was a 50% decrease in the probability of finding a BRCA1 mutation and a 41% decrease in the probability of finding a mutation overall. The average age at ovarian cancer diagnosis was significantly associated with a mutation in BRCA1 (P = .036), with a 25% reduction in the probability of finding a mutation for each 5-year increment in average age at diagnosis, but it was only of borderline significance in families with mutations overall. Similar to the average age at breast cancer diagnosis, the proband’s age at breast cancer diagnosis was associated with a 45% decrease in probability of finding a BRCA1 mutation (P = .01) for each 5-year increase in age and with a 38% decrease in probability of finding an overall mutation. The proband’s age at diagnosis of ovarian cancer was only significantly associated with a BRCA1 mutation (P = .036), and for each 5-year increase in proband’s age, there was a 65% decrease in the probability of finding a BRCA1 mutation.
Multivariate analysis of the data was also performed using variables that had generated P values
We identified germline BRCA1 and BRCA2 mutations in more than half (55%) of the breast/ovarian families tested from a high-risk clinic setting. The ratio of BRCA1 mutations to BRCA2 mutations in this study group was 4:1. However, the proportion of BRCA1 to BRCA2 mutations may vary depending on the population. For instance, Gayther et al4 also observed a ratio of 4:1 in families with ovarian cancer. However, in Ashkenazi Jewish women with breast cancer unselected for age of diagnosis, the ratio of BRCA1 to BRCA2 founder mutations was 1:110, compared with a 6:1 ratio among Ashkenazi women with an early age at breast cancer diagnosis.11 These differences between populations likely reflect the higher ovarian cancer penetrance and earlier age at breast cancer diagnosis associated with BRCA1 mutations; as the number of ovarian cancer cases increases in families, the proportion of BRCA1 and BRCA2 mutations also increases.1 In our study group, a mutation could not be identified in almost half of the women tested. Some mutations, such as large genomic rearrangements and deletions, will undoubtedly have been missed by the PCR-based methods. We and others have reported the presence of an exon 13 duplication and genomic rearrangements in BRCA1,12-15 which may account for up to 10% of mutations in families with breast and ovarian cancers where mutations are not detected using PCR-based screening.13,15 However, even when considering the presence of the exon 13 duplication and genomic rearrangements in the remaining individuals, the frequency of BRCA1 and BRCA2 mutations in this cohort still does not reach the Breast Cancer Linkage Consortium estimated frequency of approximately 98%. The disparity is almost certainly explained by the characteristics of the two study populations, where families ascertained for linkage studies are not representative of those attending a breast cancer risk evaluation clinic. Whether additional low-penetrance susceptibility genes can be identified in this study group remains to be determined. As in previous studies,16,17 a young average age at breast cancer diagnosis was a strong predictor of a BRCA1 germline mutation and mutation status overall by the logistic regression model. As age at breast cancer diagnosis increased in the current study group, the probability of finding a mutation in BRCA1 decreased by 50% for each 5-year increment and by 41% for a mutation overall, consistent with previously published data.17-21 When we examined the age of breast or ovarian cancer diagnosis of the proband, as opposed to average age in the family, we found that the proband’s age at diagnosis of ovarian cancer also was significantly associated with a BRCA1 mutation only, with each 5-year increase in proband’s age associated with a 65% decrease in the probability of finding a BRCA1 mutation. While the effect of age at diagnosis of breast cancer on the likelihood of finding BRCA1 mutations has been described in many studies, the association with age at ovarian cancer diagnosis has been less clear. In this study, early age at diagnosis of breast cancer was not a predictor of a BRCA2 mutation. Also consistent with previous studies, the number of breast cancers in a family was not a significant predictor of the presence of any mutation in this study group.16 When families are selected on the basis of breast and ovarian cancers, their underlying mutation risk is quite high, and Ashkenazi Jewish ancestry is not predictive of mutation status in this setting, which also was found by Frank et al.22 The other strong predictor of mutation status in this study group was the presence of breast and ovarian cancers in a single individual, which was predictive of overall mutation status and BRCA1 mutation status when compared with families without detectable mutations. In addition, this was the only variable that was predictive of a BRCA2 mutation. Previous reports16,17 have also shown that the presence of breast and ovarian cancers in one family member significantly predicts the presence of a BRCA1 mutation, and in most cases this is the single strongest predictor. When the non-Ashkenazi Jewish group was analyzed separately for predictors of mutation status overall using the stepwise selection multivariate model, young average age at ovarian cancer diagnosis emerged as a novel predictor of mutation; the presence of an individual with both breast and ovarian cancers and young average age at breast cancer diagnosis remained predictors of mutation status. The finding that young average age at ovarian cancer diagnosis is predictive of mutation status only in the non-Ashkenazi Jewish group is consistent with the finding that young average age at ovarian cancer diagnosis is predictive of BRCA1 mutation as the vast majority of mutations in the non-Ashkenazi group are in BRCA1. While average age at ovarian cancer diagnosis has not been consistently examined in the models predicting the prior probability of mutations in BRCA1 or BRCA2, young age at ovarian cancer diagnosis has been associated with mutations in BRCA1 as compared with sporadic cases.4 Because the sample size used for the analysis in the non-Ashkenazi families was small, these results will need to be verified with a larger sample set. An ovarian cancer cluster region (OCCR) in exon 11 of BRCA2 has been postulated, in which mutations lead to an increased relative risk (OR, 2.5) of ovarian cancer.23 This may occur because mutations in this region of BRCA2 seem to confer a lower risk of breast cancer per se, or a lower breast cancer risk relative to ovarian cancer risk. In this analysis, we compared the proportions of breast cancer and ovarian cancer cases among the families with mutations in the OCCR in the BRCA2 gene with those with mutations elsewhere. The majority of BRCA2 mutations did fall within the OCCR (nine in this region compared with three elsewhere in the gene), and six out of nine OCCR mutations were 6174delT. Although difficult to assess with accuracy due to the small numbers, the proportional size of the OCCR to the remainder of BRCA2 does support previous observations that this region confers either increased risk of ovarian cancer or a relatively decreased risk of breast cancer. Of note, one family in this study had both a BRCA1 (3888delGA) and BRCA2 (6174delT) mutation.9 Although several families with a founder mutation in each gene have been described in the Ashkenazi Jewish population, this family has a distinct BRCA1 mutation. The BRCA1 mutation was reported as being inherited from the proband’s father and the BRCA2 mutation from her mother. Interestingly, another patient carrying identical mutations in BRCA1 and BRCA2 has been reported from Australia.24 In the latter case, the proband also inherited the BRCA2 mutation from her mother and the BRCA1 mutation was a de novo mutation that arose during spermatogenesis. The BRCA1 3888delGA mutation is uncommon, and to our knowledge these are the only two reported cases. Whether there is a functional link between these two mutations is not known. In this study group of breast and ovarian cancer families seen at our high-risk breast evaluation clinic, approximately half had an identifiable germline mutation in BRCA1 and BRCA2. That fraction increased to 65% in families with more than two cases of ovarian cancer. Clearly, some mutations are missed by PCR-based screening methods, such as large genomic rearrangements and deletions. CSGE is thought to be approximately 95% sensitive,16 and therefore some mutations may be missed due to technical issues. Another variable to consider that may decrease mutation detection sensitivity in research-based mutation testing is the possibility of human error. Finally, it is possible that the proband is a phenocopy and does not carry a germline mutation in BRCA1 or BRCA2 while other family members do. The best way to minimize the possibility of missing BRCA1 or BRCA2 mutations because of phenocopies is to sample multiple family members, which adds prohibitively to the cost of such studies, or to test the family members with the youngest age at diagnosis or both breast and ovarian cancers, the approach that was used in our study. In this study, there were two probands with an age at breast cancer diagnosis over 60 years; one of them had both breast and ovarian cancer and thus was very unlikely to be a phenocopy, but it is more difficult to be certain with the other proband. Our analyses provide further support for previous observations that early average age at diagnosis and the presence of breast and ovarian cancers in a single individual are strong predictors for detecting a germline mutation in a family. In families that have an early age at breast cancer diagnosis and that contain an individual with both breast and ovarian cancer, it may be prudent to recommend mutation screening using non-PCR–based methods. Given the study by the Exon 13 Duplication Consortium,14 women of British descent, in whom germline mutations in BRCA1 and BRCA2 are not detected using PCR-based screening methods, exon 13 duplication screening should be considered before using other more complex non-PCR–based methodologies, such as Southern blotting. In terms of the clinical care of women with a compelling history of ovarian cancer but without a detectable mutation in BRCA1 and BRCA2, we recommend that they undergo prophylactic oophorectomy after childbearing age, due to the fact that we are unable to detect a significant proportion of mutations. While prophylactic oophorectomy is particularly important because it substantially decreases the rate of ovarian cancer, for which optimal screening methods have yet to be developed, it also has been demonstrated to decrease the risk of breast cancer in BRCA1 and BRCA2 mutation carriers.25,26 To our knowledge, this is the first comprehensive analysis of coding region mutations in BRCA1 and BRCA2 in families with both breast and ovarian cancer seeking risk evaluation and consideration of genetic testing. It should provide guidance to clinicians estimating the likelihood of finding mutations in women from these families considering genetic testing.
1. Easton DF, Bishop DT, Ford D, et al: Genetic linkage analysis in familial breast and ovarian cancer: Results from 214 families—The Breast Cancer Linkage Consortium. Am J Hum Genet 52: 678-701, 1993[Medline] 2. Narod SA, Ford D, Devilee P, et al: An evaluation of genetic heterogeneity in 145 breast-ovarian cancer families: Breast Cancer Linkage Consortium. Am J Hum Genet 56: 254-264, 1995[Medline] 3. 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 62: 676-689, 1998[Medline] 4. Gayther S, Russell P, Harrington P, et al: The contribution of germline BRCA1 and BRCA2 mutations in familial ovarian cancer: No evidence for other ovarian cancer-susceptibility genes. Am J Hum Genet 65: 1021-29, 1999[Medline] 5. Moslehi R, Chu W, Karlan B, et al: BRCA1 and BRCA2 mutation analysis of 208 Ashkenazi Jewish women with ovarian cancer. Am J Hum Genet 66: 1259-1272, 2000[Medline] 6. Levy-Lahad E, Catane R, Eisenberg S, et al: Founder BRCA1 and BRCA2 mutations in Ashkenazi Jews in Israel: Frequency and differential penetrance in ovarian cancer and in breast-ovarian cancer families. Am J Hum Genet 60: 1059-1067, 1997[Medline]
7.
Ganguly A, Rock MJ, Prockop DJ: Conformation-sensitive gel electrophoresis for rapid detection of single-base differences in double-stranded PCR products and DNA fragments: Evidence for solvent-induced bends in DNA heteroduplexes. Proc Natl Acad Sci U S A 90: 10325-10329, 1993 (published erratum appears in Proc Natl Acad Sci U S A 91:5217, 1994) 8. Ganguly T, Dhulipala R, Godmilow L, et al: High throughput fluorescence-based conformation-sensitive gel electrophoresis (F-CSGE) identifies six unique BRCA2 mutations and an overall low incidence of BRCA2 mutations in high-risk BRCA1-negative breast cancer families. Hum Genet 102: 549-556, 1998[Medline] 9. Randall TC, Bell KA, Rebane BA, et al: Germline mutations of the BRCA1 and BRCA2 genes in a breast and ovarian cancer patient. Gynecol Oncol 70: 432-434, 1998[Medline]
10.
Hodgson SV, Heap E, Cameron J, et al: Risk factors for detecting germline BRCA1 and BRCA2 founder mutations in Ashkenazi Jewish women with breast or ovarian cancer. J Med Genet 36: 369-373, 1999
11.
Krainer M, Silva-Arrieta S, FitzGerald MG, et al: Differential contributions of BRCA1 and BRCA2 to early-onset breast cancer. N Engl J Med 336: 1416-1421, 1997 12. Puget N, Sinilnikova OM, Stoppa-Lyonnet D, et al: An Alu-mediated duplication in the BRCA1 gene: A new founder mutation? Am J Hum Genet 64: 300-302, 1999[Medline]
13.
Puget N, Stoppa-Lyonnet D, Sinilnikova OM, et al: Screening for genomic rearrangements and regulatory mutations in BRCA1 led to the identification of four new deletions. Cancer Res 59: 455-461, 1999 14. Mazoyer S, Group TBEDS: The exon 13 duplication in the BRCA1 gene is a founder mutation present in geographically diverse populations. Am J Hum Genet 67: 207-212, 2000[Medline] 15. Unger M, Nathanson KL, Calzone K, et al: Screening for genomic rearrangements in breast and ovarian cancer families identifies BRCA1 mutations previously missed by CSGE or sequencing. Am J Hum Genet 67: 841-581, 2000[Medline]
16.
Couch FJ, DeShano ML, Blackwood MA, et al: BRCA1 mutations in women attending clinics that evaluate the risk of breast cancer. N Engl J Med 336: 1409-1415, 1997
17.
Shattuck-Eidens D, Oliphant A, McClure M, et al: BRCA1 sequence analysis in women at high risk for susceptibility mutations: Risk factor analysis and implications for genetic testing. JAMA 278: 1242-1250, 1997 18. 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 80: 435-441, 1997[Medline] 19. Fodor FH, Weston A, Bleiweiss IJ, et al: Frequency and carrier risk associated with common BRCA1 and BRCA2 mutations in Ashkenazi Jewish breast cancer patients. Am J Hum Genet 63: 45-51, 1998[Medline] 20. Hartge P, Streuwing JP, Wacholder S, et al: The prevalence of common BRCA1 and BRCA2 mutations among Ashkenazi Jews. Am J Hum Genet 64: 963-970, 1999[Medline] 21. Foulkes WD, Brunet J-S, Warner E, et al: The importance of a family history of breast cancer in predicting the presence of a BRCA mutation. Am J Hum Genet 65: 1776-1779, 1999[Medline] 22. Frank TS, Manley SA, Olopade OI, et al: Sequence analysis of BRCA1 and BRCA2: Correlation of mutations with family history and ovarian cancer risk. J Clin Oncol 16: 2417-2425, 1998[Abstract] 23. Gayther SA, Mangion J, Russell P, et al: Variation of risks of breast and ovarian cancer associated with different germline mutations of the BRCA2 gene. Nat Genet 15: 103-105, 1997[Medline] 24. Tesoriero A, Andersen C, Southey M, et al: De novo BRCA1 mutation in a patient with breast cancer and an inherited BRCA2 mutation. Am J Hum Genet 65: 567-569, 1999[Medline]
25.
Rebbeck T, Levin A, Eisen A, et al: Breast cancer risk after bilateral prophylactic oophorectomy in BRCA1 mutation carriers. J Natl Cancer Inst 91: 1475-1479, 1999
26.
Eisen A, Rebbeck TR, Wood WC, et al: Prophylactic surgery in women with a hereditary predisposition to breast and ovarian cancer. J Clin Oncol 18: 1980-1995, 2000 Submitted August 15, 2000; accepted January 17, 2001.
This article has been cited by other articles:
|
|||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||||
|
|
|||||||||||
|
|||||||||||
|
|
Copyright © 2001 by the American Society of Clinical Oncology, Online ISSN: 1527-7755. Print ISSN: 0732-183X
|
TOP
ABSTRACT
INTRODUCTION
2 approximation was used to evaluate the differences among quantitative traits, such as mean age at diagnosis of breast cancer and/or ovarian cancer; as well as the age at diagnosis of both cancers in the proband. Contingency table and Fisher’s exact test analyses were used to evaluate differences in the frequency of BRCA1/BRCA2 mutations across groups. Relationships were evaluated using the Mantel-Haenszel test for linear trend. Odds ratios (ORs) and 95% confidence intervals (CIs) were also computed to estimate predictors of mutation status. Factors that were significant at the P = .05 level were entered into the multivariate models. Models were examined using both no selection and a stepwise selection technique. The predictor variables (phenotypes) were treated as ordinal/continuous and categorical variables. Because results from the two modeling techniques were in agreement, the data presented are for the best-fitting model using the step-wise selection technique, and all data presented are for the categorical treatment of predictor variables only. One family was excluded from these analyses because there was only a single affected individual in this family, albeit with both breast and ovarian cancer. 
.05 in the univariate analyses. In the best-fitting model, using a stepwise comparison of variables, the ORs of the presence of a family member with both breast and ovarian cancer (P = .0009; OR, 5.68; 95% CI, 2.04 to 15.76) as well as average age at breast cancer diagnosis in the family (P = .0016; OR, 1.69; 95% CI, 1.23 to 2.38) were significant predictors of overall mutation status (
