Breast CancerResearch and Treatment15: 63-71, 1990. © 1990KluwerAcademic Publishers. Printedin the Netherlands. Viewpoints
Clinical/genetic features in hereditary breast cancer Henry T. Lynch, Patrice Watson, Theresa A. Conway and Jane F. Lynch Department of Preventive Medicine~Public Health, Creighton University School of Medicine, and the Hereditary Cancer Consultation Center, Omaha, Nebraska 68178, USA
Key words: genetics, hereditary breast cancer, clinical features, cancer control, family studies Summary Patients from hereditary breast cancer-prone (HBC) families provide one of the most powerful and potentially cost effective models for cancer prevention and control. Paradoxically, this opportunity is often missed in the clinical practice setting due, in part, to inattention to the family history and/or lack of knowledge about breast cancer genetics. We describe the family study process, wherein documentation of genealogy, medical, and cancer history through pathology verification is attained, often on extended families. The findings of such studies are illustrated by description of nine breast cancer-prone families. These families illustrate several important clinical/genetic features such as age of cancer onset, bilaterality and/or multiple primary cancer occurrences, incomplete gene penetrance, and the identification of putative obligate gene carriers. Interpretation of HB C pedigrees is dependent upon the understanding of these issues, which in turn may enable the physician to more readily identify those individuals who might benefit from highly targeted breast cancer control measures.
Introduction All women are not at equal risk for breast cancer. The highest known risk for breast cancer occurs to a woman whose first degree relative has had breast cancer and who is in the direct lineage of a hereditary breast cancer-prone (HBC) family. The significance of this risk factor is frequently missed, since the family history of cancer is often given short shrift in the clinical practice setting [1]. This is unfortunate since HBC's natural history typically differs from ordinary breast cancer's natural history. These differences make it possible to differentiate families where multiple breast cancers were likely to have occurred by chance from those where HBC is more likely. We have here described nine families from more
than 100 well-documented HBC kindreds in our resource so that we could illustrate common clinical/ genetic features which appear to be pervasive in HBC kindreds. Attention will then be focused upon how this knowledge might enable at risk relatives to be more readily identified, thereby enabling their opportunity for participation in highly organ-targeted surveillance and management strategies.
Subjects and methods
Operational definitions Sporadic breast cancer. This is defined as a patient who manifests breast cancer but whose family history shows an absence of this disease through two
Addressfor offprints: H.T. Lynch,Dept. of PreventiveMedicine,CreightonUniversitySchoolof Medicine,California@24th Street, Omaha, NE 68178USA
64
H T Lynch et al.
generations involving sibs, offspring, parents, and both sets of aunts, uncles, and grandparents. An unknown fraction of such patients may be hereditary in that they may be the result of a new germinal mutation or reduced penetrance of the gene in their parent.
Familial breast cancer. This is defined as breast cancer in a patient who has one or more first or second degree relatives with breast cancer in the absence of evidence of hereditary breast cancer. Familial breast cancer is common and accounts for about 25% of the total breast cancer burden. Its etiology is multifactorial and/or due to chance. Given the high frequency of breast cancer in the general population, wherein about 1 in 10 women who live to age 70 or more will develop this disease, it is not surprising that by chance alone, one should encounter many families with two or more relatives affected. Nevertheless, the empirical risk to a first degree relative of a breast cancer affected individual in the familial setting is as much as three-fold increased over that which exists for patients who lack a familial background of breast cancer [1].
Hereditary breast cancer. By hereditary breast cancer (HBC), we mean a pattern within a particular family which shows Mendelian segregation of breast cancer. As the term is used in this report, H B C encompasses all families evidencing Mendelian inheritance of breast cancer susceptibility, and thus does not refer to a homogeneous category. In fact, HBC families are heterogeneous with regard to integral associations with cancer sites other than the breast, and with regard to age of onset. Such variations in cancer natural history between H B C families have led to the description of putative H B C syndromes (see Table 1). The lack of a biomarker of acceptable sensitivity and specificity to HBC's genotype limits the boundaries of this operational definition. Specifically, while vertical transmission of breast cancer in many of our pedigrees and findings from formal segregation analysis of families from our resource are compatible with an autosomal dominant mode of inheritance [2], autosomal recessive inheritance or mere chance may nevertheless be responsible in some of our kin-
dreds. These issues will be further clarified through continued investigations of homogeneous groupings of pedigrees based on breast cancer on a sitespecific basis vs. its integral association with cancer of differing anatomic sites. When coupled with biomolecular determinants, we may then be able to resolve whether susceptibility to H B C in these syndromes is allelic or determined by differing chromosomal loci.
Family study methods Probands in our H B C kindreds were ascertained through self or physician referral, or through their identification in our oncology clinic during the course of their breast cancer evaluation and treatment. Signed permission forms enabled us to contact their relatives. Questionnaires were distributed to selected first and second degree relatives (both sets of grandparents, aunts, and uncles). Permission forms for release of primary medical and pathology documents, when signed by living cancer-affected patients or their legal next of kin, enabled us to secure these materials for review. Personal interview and/or telephone contact was made with selected relatives in order to increase the accuracy of our dataset.
Family case reports
Family #2711 (Fig. 1). The proband (IV-28) is age 53 and was worried about her risk for cancer. Note that three of her sisters (IV-25, IV-26, IV-27) manifested breast cancer, one of whom (IV-26) had bilateral breast cancer. Two of her brothers (IV-23, IV-24) had carcinoma of the prostate. The mother of this sibship (III-5) had breast cancer at age 52 and died of carcinoma of the esophagus. Three of this lady's brothers (III-1, III-3, III-4) had carcinoma, and each in turn had children with breast cancer. The brother with prostate cancer (III-1) had a son (IV-4) with carcinoma of the breast at age 51 and a daughter (IV-5) with carcinoma of the breast at age 39. An unaffected brother (III-2) had two daughters (IV-6, IV-7) who, by family report, had
Clinical/genetic features in hereditary breast cancer breast cancer. He is a putative obligate gene carrier. Another brother who reportedly had cancer (possibly prostate) (III-3) had a daughter (IV-12) with bilateral breast cancer and ovarian cancer. This lady had a daughter (V-5) with breast cancer. A third brother (III-4) with cancer, possibly of the lung, had four daughters with breast cancer, one at age 34 (IV-16), a second at age 72 (IV-17), a third at age 42 (IV-18), and the fourth at age 48 (IV-19). The proband's maternal grandmother (II-1) had breast cancer by family history and died from cancer of the cervix at age 78. Her great-grandmother (I-1) had breast cancer in her 40s.
65
patient I-2, who had breast cancer at age 49, through two separate marriages. From her first marriage, note that she had a daughter (II-1) with breast cancer and ovarian cancer at age 64 and a daughter (II-3) with breast cancer at age 48 and bilateral ovarian cancer at age 56. From her second marriage, her 60 year old daughter (II-5) did not manifest cancer. However, this lady did have a daughter (lII-6) who had breast cancer at age 34. Thus, patient II-5 is a putative obligate gene carrier. These observations support the need to trace cancer through each of the marriages of an affected individual as well as to recognize the status of an obligate gene carrier.
Family #1852 (Fig. 2). The proband (II-1) is age 70 years and is cancer-free. However, she is a putative obligate gene carrier by virtue of the fact that her daughter (III-2) and her mother (I-1) both manifested breast cancer, and her sister (II-2) had bilateral breast cancer. Note also that the proband's maternal aunt (I-2) had breast cancer at age 34 and this lady's daughter (II-5) had breast cancer at age 54. Another maternal aunt 0-4) died at age 66 and was cancer-free. This lady is a putative obligate gene carrier. This is inferred by the fact that she had two daughters with cancer; one (II-6) manifested colon cancer at age 68, while the second (II-7) had breast cancer at age 30 and bilateral ovarian cancer at age 55. Therefore, an important clinical consideration in this family is the recognition of putative obligate gene carriers (II-1, I-4), a phenomenon which to the unwary may suggest that we are dealing with 'skipped generations' when in fact we are most probably dealing with reduced penetrance of the deleterious breast cancer-prone gene(s).
Family #1879 (Fig. 2). This family is of interest in that we observe breast cancer transmitted from Table 1. Heterogeneity in hereditary breast cancer
site-specific breast cancer [13] Cowden's disease [14] breast/ovarian cancer syndrome [15] breast/gastrointestinal cancer syndrome [1] SBLA syndrome [16, 17] extraordinarily early onset breast cancer [4, 5]
Family #2191 (Fig. 2). The proband (III-1) had breast cancer at age 42. Her daughter (IV-l), age 24, had been adopted out at birth. While this daughter was in college, she initiated contact with a physician who was knowledgeable about her biological mother's whereabouts, and her breast cancer family history. This resulted in a referral to us for purposes of studying the family and advising the daughter of her cancer risk and her need for surveillance. Note that the proband's mother (II-1) had breast cancer at age 50 and that the maternal grandmother 0-1) had breast cancer at age 83. A maternal great-uncle (I-3) is a putative obligate gene carrier by virtue of the fact that one of his daughters (II-7) had breast cancer at age 44. Another of his daughters (II-6) had colon cancer and she, in turn, had a daughter (III-6) with breast cancer, by unconfirmed family report.
Family #2775 (Fig. 2). The proband (IV-l) had breast cancer at age 41, while her sister (IV-2) had breast cancer at age 35. The proband's mother (III-1) had breast cancer at age 47 and this lady had three sisters (III-3, III-5, III-7) with breast cancer. The proband's maternal grandmother (II-1) had a liposarcoma at age 54 and the proband's maternal great-grandmother (I-2) had breast cancer and died at age 35. One of the proband's maternal aunts (II1-3) had a daughter with breast cancer at age 31 (IV-8), while a second maternal aunt (III-7) had a daughter (IV-14) with breast cancer at age 33. One of the proband's maternal great-uncles (II-6), al-
H T Lynch et al.
66
FamiLy
2711
}
a,45
Cx75 ~78
I
d.St
e.82 5
9
11
1
4
5
ci,63
0L39
d,68
d,4~
1
75
cL47
3~]v48
/sr~ / d.37
OL35
7~
56
74
-
~'4S
g
1]
O
Sites
UnaFfected by CQncer
~d ~a g
Lu7~ 75
59
40-28 4~-33
Cancer
£ J
64
¢
•
.t
o4,~t~
Center es~Q~tmhed ~,y p~thc~oey C ~ n c e r sr~t ¢ ~ =~ ~ a o s l s Age a t w*e o¢ study CQOCep I S t G ~ l ~ d by ~ ¢ h mtd4c~{ r e c o r d $
Co
C~on
Cx
Ca,r v ~
Lu
LeUkt,~ Lung
kyx
L~ry~x
ceP¢~
gl. . . . . 0 t ~ u s
~ostQte
by I~¢hc~o~y
[] T,~S
Pro~r,d
Fig. 1. Pedigree of a large, extended, hereditary breast cancer family. Note the occurrence of male breast cancer in patient IV-4. Four males in generation III have transmitted breast cancer to their offspring.
though not affected with cancer, had a daughter with breast cancer (III-10) at age 45. Again, we see examples of putative obligate gene carriers with one example of a sarcoma in a patient (II-1) in the direct genetic lineage.
Family #2651 (Fig. 2). This kindred is of interest in that it shows a familial aggregation of carcinoma of the breast and ovary. Note that the proband (III-1) had an endometrioid ovarian carcinoma at age 55. Her three sisters (III-4, III-5, III-6) had breast cancer at ages 48, 42, and 36, respectively, and one of these women (III-5) also had a carcinoma of the ovary at age 49. The daughter (IV-l) of the proband had breast cancer at age 32. The proband's mother (II-1) did not have cancer and is still living
at age 80. However, her sister (II-6), who died at age 87, was also reported to have had breast cancer. By family report, she had a daughter (III-7) with breast cancer. The proband's maternal grandmother 0-2) had breast cancer at age 48. Thus, we see some of the problems involved in pedigree interpretation, including determination of putative obligate gene carriers.
Late age of breast cancer onset pedigrees In certain families, a clustering of late age of onset of breast cancer may occur. Examples are depicted in Families #2637, #2485, and #1882 (Fig. 3). In Family #2637, we note four siblings (III-1 to
Clinical~genetic features in hereditary breast cancer FamiLy
Fo.mlty
i ~
Family
1852
3
2
1879
I[[
~)
38
67
11.159
Br30
42
II
;~8 ~ r 30
6.....
40 47
9 Br64
78
DV64 67
7
44
?
"
/~r48 ~ x 4 9
. . . . . . . . . . d18
22
d.59
IV
24 ~1 2o'~ 18
1
2775
~
:,2"
II[
~lDv56 ~,49
d,35
e'a~_-
5
3
III
45
Family
a.80
a, 5
2 _/
; II
2191
;
4
67
~
~
~
FamiLy
II
8651
~(@Je~0e 0~
~)e 6~a.B~7
d'155 II1
/BrSO
Be55 d
49 5
%m IV Br41 42 25 22
41 38 31 28
6
Be45
58
13
53
fly55
17
1
57
61-43Br48 3
Br42~r36
Br
5
IV 35
25
31
r 33
,
× 43
38
r 43
41 ~ 45 I / I
23
25
24
20 14
r 34
~ 22
9
e7
d, ~o
Fig. 2. Pedigrees of hereditary breast cancer families which illustrate obligate gene carriers and male transmission of breast cancer. 4) with breast cancer. The proband (III-1) had breast cancer at age 65 and his three sisters manifested breast cancer at ages 60, 62, and 66. Note that their father (II-3) had colon cancer at age 71. In Family #2485, the proband (II-1) had breast cancer at age 66. H e r mother 0-1) had colon cancer at age 55 and breast cancer at age 57. Two of her mother's sisters (I-2, 3) had breast cancer in their 70s and 80s, while one sister (I-4) had breast cancer at age 31. Note that the patient with breast cancer at age 31 had a daughter 01-5) with bilateral breast cancer, the first lesion at age 70 and the second at age 73. Thus, we see enormous age variation, ranging from a low age of breast cancer onset of 31 to the late onset in the 80s. In Family #1882, the proband (III-1) is an identical twin who manifested breast cancer at age 51. H e r twin sister (III-2) had breast cancer at age 67. Another sister (III-3) had ovarian cancer at age 73. A fourth sister (IIl-4) was cancer free at age 75. However, this lady had a daughter (IV-4) who had breast cancer at age 44. The mother (II-1) of this informative sibship had breast cancer at age 43, with a second primary of the breast at age 79. This lady had a sister (II-4) with breast cancer at age 55. Note that this 55 year old lady had 8 children, of
whom five had cancer. One of her daughters (III-10) had breast cancer at age 75.
Discussion
Several clinical/genetic features in H B C kindreds have been identified in our pedigrees (Figs 1-3). These include the following: 1) variation in age of onset, some extremely early, some mixed, and some clustering at a late age; 2) integral patterns of tumor combinations; and 3) incomplete gene penetrance. All of these features appear to be pervasive in HBC. In addition, the presence of nonbreast cancer types in putative obligate gene carrier males is of concern, particularly with respect to carcinoma of the prostate. However, their etiologic significance is uncertain and will require further biomolecular and epidemiologic study. Hereditary breast cancer is exceedingly heterogeneous, as evidenced by the variability in age at onset and tumor spectrum seen in these pedigrees. Apparent obligate gene carriers may live through a normal lifespan and fail to manifest cancer, a finding which we have attributed to incomplete penetrance, under the assumption that HB C is autosomal dominantly inher-
68
H T L y n c h et al.
I
n
~
#2637 3
4
~-m
a,49
ill
U',:TS
L e u 11 e9
Fig. 3. Hereditarybreast cancerfamilieswhichshowlater age at canceronset than previouslyreported in this disorder.
ited [2]. However, because of the likely extant heterogeneity in the genetics of HBC, the significance for other modes of inheritance (autosomal recessive, X-linked dominant or recessive) remains elusive. These facts of HBC's natural history are best observed through longitudinal study of large, extended pedigrees wherein a high degree of pathology documentation of cancer (all sites) has been attained (Figs 1-3). It is important to understand that knowledge of the cardinal features of hereditary breast cancer may not necessarily contribute to the elucidation of the etiology and pathogenesis of the overall breast cancer burden. Family #2711 (Fig. 1) clearly shows the importance of extending the family for HBC diagnosis, including the need to recognize obligate gene carrier males. Most investigators consider breast cancer to be autosomal dominantly inherited, with equal transmission of the deleterious gene by both males and females [1, 2]. There is also the possibility that carcinoma of the prostate may, in certain families, affect obligate gene carrier males. This is consistent with Thiessen's [3] finding of higher incidence of prostate cancer in relatives of breast cancer patients compared to relatives of controls. Although breast cancer in males is uncommon, two of the families reported here (#2711 [Fig/1] and #2637 [Fig. 3]) have males with this malady. It is important that adopted children become informed about their genetic heritage. Family
#2191 (Fig. 2) portrays an example of hereditary cancer risk which was identified sensitively through the referring physician, the consultant, the patient's adoptive mother, and ultimately, the patient's biological mother. Confidences were never violated, and the patient may benefit from highly targeted education and surveillance. An interesting clinical/genetic observation was the concordance for breast cancer in identical twins (Family #1882, Fig. 3). It is important to realize that in a hereditary breast cancer setting, concordance for breast cancer in identical twins should theoretically approach 100%, and unaffected twins of affected family members are putative gene carriers. Zygosity of the twins must be assessed and it is essential to determine whether or not we are dealing with a hereditary breast cancer syndrome. Families #2637, #2485, and #1882 (Fig. 3) show consistently late age of onset among cases of breast cancer. In contrast, we have recently described several families showing consistently early ages of onset [4]. Very little is known about the role of genetics in late age of onset breast cancer. Indeed, there is exceedingly limited information available in the cancer genetics literature relevant to late age of onset clustering for virtually all forms of alleged hereditary varieties of cancer. However, it is clear that a subset of older at-risk patients warrants equally intensive surveillance for breast cancer.
Clinical~genetic features in hereditary breast cancer
69
7°T 0
o
80-
~:
60
r=0,48
o
°
o o
e
dCE
u~
0 in
O
o
.~ 70
o
o
o
o e
W
Clinical/genetic features in hereditary breast cancer Henry T. Lynch, Patrice Watson, Theresa A. Conway and Jane F. Lynch Department of Preventive Medicine~Public Health, Creighton University School of Medicine, and the Hereditary Cancer Consultation Center, Omaha, Nebraska 68178, USA
Key words: genetics, hereditary breast cancer, clinical features, cancer control, family studies Summary Patients from hereditary breast cancer-prone (HBC) families provide one of the most powerful and potentially cost effective models for cancer prevention and control. Paradoxically, this opportunity is often missed in the clinical practice setting due, in part, to inattention to the family history and/or lack of knowledge about breast cancer genetics. We describe the family study process, wherein documentation of genealogy, medical, and cancer history through pathology verification is attained, often on extended families. The findings of such studies are illustrated by description of nine breast cancer-prone families. These families illustrate several important clinical/genetic features such as age of cancer onset, bilaterality and/or multiple primary cancer occurrences, incomplete gene penetrance, and the identification of putative obligate gene carriers. Interpretation of HB C pedigrees is dependent upon the understanding of these issues, which in turn may enable the physician to more readily identify those individuals who might benefit from highly targeted breast cancer control measures.
Introduction All women are not at equal risk for breast cancer. The highest known risk for breast cancer occurs to a woman whose first degree relative has had breast cancer and who is in the direct lineage of a hereditary breast cancer-prone (HBC) family. The significance of this risk factor is frequently missed, since the family history of cancer is often given short shrift in the clinical practice setting [1]. This is unfortunate since HBC's natural history typically differs from ordinary breast cancer's natural history. These differences make it possible to differentiate families where multiple breast cancers were likely to have occurred by chance from those where HBC is more likely. We have here described nine families from more
than 100 well-documented HBC kindreds in our resource so that we could illustrate common clinical/ genetic features which appear to be pervasive in HBC kindreds. Attention will then be focused upon how this knowledge might enable at risk relatives to be more readily identified, thereby enabling their opportunity for participation in highly organ-targeted surveillance and management strategies.
Subjects and methods
Operational definitions Sporadic breast cancer. This is defined as a patient who manifests breast cancer but whose family history shows an absence of this disease through two
Addressfor offprints: H.T. Lynch,Dept. of PreventiveMedicine,CreightonUniversitySchoolof Medicine,California@24th Street, Omaha, NE 68178USA
64
H T Lynch et al.
generations involving sibs, offspring, parents, and both sets of aunts, uncles, and grandparents. An unknown fraction of such patients may be hereditary in that they may be the result of a new germinal mutation or reduced penetrance of the gene in their parent.
Familial breast cancer. This is defined as breast cancer in a patient who has one or more first or second degree relatives with breast cancer in the absence of evidence of hereditary breast cancer. Familial breast cancer is common and accounts for about 25% of the total breast cancer burden. Its etiology is multifactorial and/or due to chance. Given the high frequency of breast cancer in the general population, wherein about 1 in 10 women who live to age 70 or more will develop this disease, it is not surprising that by chance alone, one should encounter many families with two or more relatives affected. Nevertheless, the empirical risk to a first degree relative of a breast cancer affected individual in the familial setting is as much as three-fold increased over that which exists for patients who lack a familial background of breast cancer [1].
Hereditary breast cancer. By hereditary breast cancer (HBC), we mean a pattern within a particular family which shows Mendelian segregation of breast cancer. As the term is used in this report, H B C encompasses all families evidencing Mendelian inheritance of breast cancer susceptibility, and thus does not refer to a homogeneous category. In fact, HBC families are heterogeneous with regard to integral associations with cancer sites other than the breast, and with regard to age of onset. Such variations in cancer natural history between H B C families have led to the description of putative H B C syndromes (see Table 1). The lack of a biomarker of acceptable sensitivity and specificity to HBC's genotype limits the boundaries of this operational definition. Specifically, while vertical transmission of breast cancer in many of our pedigrees and findings from formal segregation analysis of families from our resource are compatible with an autosomal dominant mode of inheritance [2], autosomal recessive inheritance or mere chance may nevertheless be responsible in some of our kin-
dreds. These issues will be further clarified through continued investigations of homogeneous groupings of pedigrees based on breast cancer on a sitespecific basis vs. its integral association with cancer of differing anatomic sites. When coupled with biomolecular determinants, we may then be able to resolve whether susceptibility to H B C in these syndromes is allelic or determined by differing chromosomal loci.
Family study methods Probands in our H B C kindreds were ascertained through self or physician referral, or through their identification in our oncology clinic during the course of their breast cancer evaluation and treatment. Signed permission forms enabled us to contact their relatives. Questionnaires were distributed to selected first and second degree relatives (both sets of grandparents, aunts, and uncles). Permission forms for release of primary medical and pathology documents, when signed by living cancer-affected patients or their legal next of kin, enabled us to secure these materials for review. Personal interview and/or telephone contact was made with selected relatives in order to increase the accuracy of our dataset.
Family case reports
Family #2711 (Fig. 1). The proband (IV-28) is age 53 and was worried about her risk for cancer. Note that three of her sisters (IV-25, IV-26, IV-27) manifested breast cancer, one of whom (IV-26) had bilateral breast cancer. Two of her brothers (IV-23, IV-24) had carcinoma of the prostate. The mother of this sibship (III-5) had breast cancer at age 52 and died of carcinoma of the esophagus. Three of this lady's brothers (III-1, III-3, III-4) had carcinoma, and each in turn had children with breast cancer. The brother with prostate cancer (III-1) had a son (IV-4) with carcinoma of the breast at age 51 and a daughter (IV-5) with carcinoma of the breast at age 39. An unaffected brother (III-2) had two daughters (IV-6, IV-7) who, by family report, had
Clinical/genetic features in hereditary breast cancer breast cancer. He is a putative obligate gene carrier. Another brother who reportedly had cancer (possibly prostate) (III-3) had a daughter (IV-12) with bilateral breast cancer and ovarian cancer. This lady had a daughter (V-5) with breast cancer. A third brother (III-4) with cancer, possibly of the lung, had four daughters with breast cancer, one at age 34 (IV-16), a second at age 72 (IV-17), a third at age 42 (IV-18), and the fourth at age 48 (IV-19). The proband's maternal grandmother (II-1) had breast cancer by family history and died from cancer of the cervix at age 78. Her great-grandmother (I-1) had breast cancer in her 40s.
65
patient I-2, who had breast cancer at age 49, through two separate marriages. From her first marriage, note that she had a daughter (II-1) with breast cancer and ovarian cancer at age 64 and a daughter (II-3) with breast cancer at age 48 and bilateral ovarian cancer at age 56. From her second marriage, her 60 year old daughter (II-5) did not manifest cancer. However, this lady did have a daughter (lII-6) who had breast cancer at age 34. Thus, patient II-5 is a putative obligate gene carrier. These observations support the need to trace cancer through each of the marriages of an affected individual as well as to recognize the status of an obligate gene carrier.
Family #1852 (Fig. 2). The proband (II-1) is age 70 years and is cancer-free. However, she is a putative obligate gene carrier by virtue of the fact that her daughter (III-2) and her mother (I-1) both manifested breast cancer, and her sister (II-2) had bilateral breast cancer. Note also that the proband's maternal aunt (I-2) had breast cancer at age 34 and this lady's daughter (II-5) had breast cancer at age 54. Another maternal aunt 0-4) died at age 66 and was cancer-free. This lady is a putative obligate gene carrier. This is inferred by the fact that she had two daughters with cancer; one (II-6) manifested colon cancer at age 68, while the second (II-7) had breast cancer at age 30 and bilateral ovarian cancer at age 55. Therefore, an important clinical consideration in this family is the recognition of putative obligate gene carriers (II-1, I-4), a phenomenon which to the unwary may suggest that we are dealing with 'skipped generations' when in fact we are most probably dealing with reduced penetrance of the deleterious breast cancer-prone gene(s).
Family #1879 (Fig. 2). This family is of interest in that we observe breast cancer transmitted from Table 1. Heterogeneity in hereditary breast cancer
site-specific breast cancer [13] Cowden's disease [14] breast/ovarian cancer syndrome [15] breast/gastrointestinal cancer syndrome [1] SBLA syndrome [16, 17] extraordinarily early onset breast cancer [4, 5]
Family #2191 (Fig. 2). The proband (III-1) had breast cancer at age 42. Her daughter (IV-l), age 24, had been adopted out at birth. While this daughter was in college, she initiated contact with a physician who was knowledgeable about her biological mother's whereabouts, and her breast cancer family history. This resulted in a referral to us for purposes of studying the family and advising the daughter of her cancer risk and her need for surveillance. Note that the proband's mother (II-1) had breast cancer at age 50 and that the maternal grandmother 0-1) had breast cancer at age 83. A maternal great-uncle (I-3) is a putative obligate gene carrier by virtue of the fact that one of his daughters (II-7) had breast cancer at age 44. Another of his daughters (II-6) had colon cancer and she, in turn, had a daughter (III-6) with breast cancer, by unconfirmed family report.
Family #2775 (Fig. 2). The proband (IV-l) had breast cancer at age 41, while her sister (IV-2) had breast cancer at age 35. The proband's mother (III-1) had breast cancer at age 47 and this lady had three sisters (III-3, III-5, III-7) with breast cancer. The proband's maternal grandmother (II-1) had a liposarcoma at age 54 and the proband's maternal great-grandmother (I-2) had breast cancer and died at age 35. One of the proband's maternal aunts (II1-3) had a daughter with breast cancer at age 31 (IV-8), while a second maternal aunt (III-7) had a daughter (IV-14) with breast cancer at age 33. One of the proband's maternal great-uncles (II-6), al-
H T Lynch et al.
66
FamiLy
2711
}
a,45
Cx75 ~78
I
d.St
e.82 5
9
11
1
4
5
ci,63
0L39
d,68
d,4~
1
75
cL47
3~]v48
/sr~ / d.37
OL35
7~
56
74
-
~'4S
g
1]
O
Sites
UnaFfected by CQncer
~d ~a g
Lu7~ 75
59
40-28 4~-33
Cancer
£ J
64
¢
•
.t
o4,~t~
Center es~Q~tmhed ~,y p~thc~oey C ~ n c e r sr~t ¢ ~ =~ ~ a o s l s Age a t w*e o¢ study CQOCep I S t G ~ l ~ d by ~ ¢ h mtd4c~{ r e c o r d $
Co
C~on
Cx
Ca,r v ~
Lu
LeUkt,~ Lung
kyx
L~ry~x
ceP¢~
gl. . . . . 0 t ~ u s
~ostQte
by I~¢hc~o~y
[] T,~S
Pro~r,d
Fig. 1. Pedigree of a large, extended, hereditary breast cancer family. Note the occurrence of male breast cancer in patient IV-4. Four males in generation III have transmitted breast cancer to their offspring.
though not affected with cancer, had a daughter with breast cancer (III-10) at age 45. Again, we see examples of putative obligate gene carriers with one example of a sarcoma in a patient (II-1) in the direct genetic lineage.
Family #2651 (Fig. 2). This kindred is of interest in that it shows a familial aggregation of carcinoma of the breast and ovary. Note that the proband (III-1) had an endometrioid ovarian carcinoma at age 55. Her three sisters (III-4, III-5, III-6) had breast cancer at ages 48, 42, and 36, respectively, and one of these women (III-5) also had a carcinoma of the ovary at age 49. The daughter (IV-l) of the proband had breast cancer at age 32. The proband's mother (II-1) did not have cancer and is still living
at age 80. However, her sister (II-6), who died at age 87, was also reported to have had breast cancer. By family report, she had a daughter (III-7) with breast cancer. The proband's maternal grandmother 0-2) had breast cancer at age 48. Thus, we see some of the problems involved in pedigree interpretation, including determination of putative obligate gene carriers.
Late age of breast cancer onset pedigrees In certain families, a clustering of late age of onset of breast cancer may occur. Examples are depicted in Families #2637, #2485, and #1882 (Fig. 3). In Family #2637, we note four siblings (III-1 to
Clinical~genetic features in hereditary breast cancer FamiLy
Fo.mlty
i ~
Family
1852
3
2
1879
I[[
~)
38
67
11.159
Br30
42
II
;~8 ~ r 30
6.....
40 47
9 Br64
78
DV64 67
7
44
?
"
/~r48 ~ x 4 9
. . . . . . . . . . d18
22
d.59
IV
24 ~1 2o'~ 18
1
2775
~
:,2"
II[
~lDv56 ~,49
d,35
e'a~_-
5
3
III
45
Family
a.80
a, 5
2 _/
; II
2191
;
4
67
~
~
~
FamiLy
II
8651
~(@Je~0e 0~
~)e 6~a.B~7
d'155 II1
/BrSO
Be55 d
49 5
%m IV Br41 42 25 22
41 38 31 28
6
Be45
58
13
53
fly55
17
1
57
61-43Br48 3
Br42~r36
Br
5
IV 35
25
31
r 33
,
× 43
38
r 43
41 ~ 45 I / I
23
25
24
20 14
r 34
~ 22
9
e7
d, ~o
Fig. 2. Pedigrees of hereditary breast cancer families which illustrate obligate gene carriers and male transmission of breast cancer. 4) with breast cancer. The proband (III-1) had breast cancer at age 65 and his three sisters manifested breast cancer at ages 60, 62, and 66. Note that their father (II-3) had colon cancer at age 71. In Family #2485, the proband (II-1) had breast cancer at age 66. H e r mother 0-1) had colon cancer at age 55 and breast cancer at age 57. Two of her mother's sisters (I-2, 3) had breast cancer in their 70s and 80s, while one sister (I-4) had breast cancer at age 31. Note that the patient with breast cancer at age 31 had a daughter 01-5) with bilateral breast cancer, the first lesion at age 70 and the second at age 73. Thus, we see enormous age variation, ranging from a low age of breast cancer onset of 31 to the late onset in the 80s. In Family #1882, the proband (III-1) is an identical twin who manifested breast cancer at age 51. H e r twin sister (III-2) had breast cancer at age 67. Another sister (III-3) had ovarian cancer at age 73. A fourth sister (IIl-4) was cancer free at age 75. However, this lady had a daughter (IV-4) who had breast cancer at age 44. The mother (II-1) of this informative sibship had breast cancer at age 43, with a second primary of the breast at age 79. This lady had a sister (II-4) with breast cancer at age 55. Note that this 55 year old lady had 8 children, of
whom five had cancer. One of her daughters (III-10) had breast cancer at age 75.
Discussion
Several clinical/genetic features in H B C kindreds have been identified in our pedigrees (Figs 1-3). These include the following: 1) variation in age of onset, some extremely early, some mixed, and some clustering at a late age; 2) integral patterns of tumor combinations; and 3) incomplete gene penetrance. All of these features appear to be pervasive in HBC. In addition, the presence of nonbreast cancer types in putative obligate gene carrier males is of concern, particularly with respect to carcinoma of the prostate. However, their etiologic significance is uncertain and will require further biomolecular and epidemiologic study. Hereditary breast cancer is exceedingly heterogeneous, as evidenced by the variability in age at onset and tumor spectrum seen in these pedigrees. Apparent obligate gene carriers may live through a normal lifespan and fail to manifest cancer, a finding which we have attributed to incomplete penetrance, under the assumption that HB C is autosomal dominantly inher-
68
H T L y n c h et al.
I
n
~
#2637 3
4
~-m
a,49
ill
U',:TS
L e u 11 e9
Fig. 3. Hereditarybreast cancerfamilieswhichshowlater age at canceronset than previouslyreported in this disorder.
ited [2]. However, because of the likely extant heterogeneity in the genetics of HBC, the significance for other modes of inheritance (autosomal recessive, X-linked dominant or recessive) remains elusive. These facts of HBC's natural history are best observed through longitudinal study of large, extended pedigrees wherein a high degree of pathology documentation of cancer (all sites) has been attained (Figs 1-3). It is important to understand that knowledge of the cardinal features of hereditary breast cancer may not necessarily contribute to the elucidation of the etiology and pathogenesis of the overall breast cancer burden. Family #2711 (Fig. 1) clearly shows the importance of extending the family for HBC diagnosis, including the need to recognize obligate gene carrier males. Most investigators consider breast cancer to be autosomal dominantly inherited, with equal transmission of the deleterious gene by both males and females [1, 2]. There is also the possibility that carcinoma of the prostate may, in certain families, affect obligate gene carrier males. This is consistent with Thiessen's [3] finding of higher incidence of prostate cancer in relatives of breast cancer patients compared to relatives of controls. Although breast cancer in males is uncommon, two of the families reported here (#2711 [Fig/1] and #2637 [Fig. 3]) have males with this malady. It is important that adopted children become informed about their genetic heritage. Family
#2191 (Fig. 2) portrays an example of hereditary cancer risk which was identified sensitively through the referring physician, the consultant, the patient's adoptive mother, and ultimately, the patient's biological mother. Confidences were never violated, and the patient may benefit from highly targeted education and surveillance. An interesting clinical/genetic observation was the concordance for breast cancer in identical twins (Family #1882, Fig. 3). It is important to realize that in a hereditary breast cancer setting, concordance for breast cancer in identical twins should theoretically approach 100%, and unaffected twins of affected family members are putative gene carriers. Zygosity of the twins must be assessed and it is essential to determine whether or not we are dealing with a hereditary breast cancer syndrome. Families #2637, #2485, and #1882 (Fig. 3) show consistently late age of onset among cases of breast cancer. In contrast, we have recently described several families showing consistently early ages of onset [4]. Very little is known about the role of genetics in late age of onset breast cancer. Indeed, there is exceedingly limited information available in the cancer genetics literature relevant to late age of onset clustering for virtually all forms of alleged hereditary varieties of cancer. However, it is clear that a subset of older at-risk patients warrants equally intensive surveillance for breast cancer.
Clinical~genetic features in hereditary breast cancer
69
7°T 0
o
80-
~:
60
r=0,48
o
°
o o
e
dCE
u~
0 in
O
o
.~ 70
o
o
o
o e
W
