The Sarcoma, Breast Cancer, Lung Cancer, and Adrenocortical Carcinoma Syndrome Revisited
Childhood Cancer
Henry T. Lynch, MD; David A. Katz, MD; Patrick J. Bogard, MD; Jane F. Lynch, BSN
. We studied two children who had rhabdomyosarcoma and glioblastoma and who were from a family with a heredi- tary cancer syndrome that was charac- terized by sarcoma, breast cancer, brain tumors, lung cancer, laryngeal carci- noma, leukemia, and adrenocortical car- cinoma. The deleterious genotype has now been expressed through the fourth generation of this large kindred. The pedigree emphasizes the need for an extended history of several generations to arrive at a hereditary-syndrome diag- nosis. A limited pedigree may result in nonappreciation of the genetic compo- nent. The pedigree Illustrates that, in certain circumstances, the highly spe- cific varieties of cancer may occur in children before it is expressed In the parent who carries the putative gene. Pediatricians, in evaluating the causes of childhood cancer, must be cognizant of cancer among adult relatives, since this recognition may ald in the diagnosis of those hereditary cancer syndromes that are characterized by cancer occur- rence in children as well as adults. (AJDC 1985;139:134-136)
A significant but still unknown frac- tion of childhood cancers are due to primary genetic factors.1 The diag- nosis of those hereditary cancer syn- dromes in which childhood cancers oc- cur as genetic components will be ex- pedited by knowledge of the particular disorder’s natural history and possible pattern of tumor types. This must in- clude those lesions that may appear in adults and are consonant with the vari-
From the Departments of Preventive Medi- cine/Public Health (Dr Lynch and Ms Lynch) and Pathology (Drs Katz and Bogard), Creighton University School of Medicine, Omaha; and the Hereditary Cancer Institute, Omaha (Dr Lynch). Reprint requests to Department of Preventive Medicine, Creighton University School of Medi- cine, California at 24th Street, Omaha, NE 68178 (Dr Lynch).
able expressivity of the deleterious cancer-prone genotype. Unfortu- nately, many clinicians are not cog- nizant of the extant pleiotropy of can- cer-prone genes, some of which may predispose to a bizarre spectrum of cancer types within a given family and which, on occasion, may involve all three germinal layers. The age at on- set of a specific cancer may also be extremely variable in hereditary forms of cancer, although the average age at onset will be significantly ear- lier than the sporadic counterpart.
Most studies of cancer in children, including the Oxford Survey of Child- hood Cancer,2 have focused almost ex- clusively on cancers of infancy and childhood. They have given only lim- ited attention to cancer in family mem- bers. When hereditary factors have been considered, in-depth evaluation of cancer in adults has often been given little attention. Hence, those heredi- tary cancer syndromes that are char- acterized by both childhood- and adult- onset cancers have been frequently missed. However, one noteworthy ex- ception is the recent study by Birch et al3 in which the health status or cause of death in the mothers of 143 children with soft-tissue sarcomas were thoroughly investigated. Inter- estingly, six of these mothers had the premenopausal onset of breast cancer and two of them had bilateral disease. This represented a threefold excess risk of breast cancer.
We previously described an ex- tended kindred that showed a broad spectrum of cancer; namely, sarcoma, breast cancer and brain tumors, lung cancer, laryngeal carcinoma, leuke- mia, and adrenocortical carcinoma, which we referred to as the SBLA syndrome.4 This disorder had been
previously discovered in four nuclear kindreds by Li and Fraumeni,5 who subsequently published a prospective observation of these families that cov- ered a 12-year time frame (1969 through 1981).6 Of interest was the fact that in 31 surviving family members, 16 additional cancers developed (the expected number was 0.5). Five of these were carcinomas of the breast, four were soft-tissue sarcomas, and seven were cancers at other anatomic sites. Eight of the patients had multi- ple primary cancers. Four cancers oc- curred at sites of prior radiotherapy (three soft-tissue sarcomas and one mesothelioma).
Our purpose is to describe two sib- lings who were affected by cancer: namely, a 5-year-old boy with rhab- domyosarcoma, and his sister, who died at the age of 11 years with a glioblastoma. These children repre- sent a fourth generation with syn- drome cancers in the original kindred with the SBLA syndrome.4
REPORT OF CASES
CASE 1 .- The first tumor, a rhabdomyo- sarcoma, was from a soft-tissue lesion above the right eye in a 2/2-year-old boy (Figure, V-5). This consisted grossly of a gray-white fragment of soft tissue. The lesion was composed of small, spindle- shaped cells that contained round to ovoid hyperchromatic nuclei with little cyto- plasm. Numerous mitotic figures were present. The patient received chemother- apy based on a protocol for rhabdomyosar- coma, stage I, which included vincristine, dactinomycin, and oral cyclophosphamide. He had an excellent response, as evidenced by his disease-free survival to the age of 5 years at the time of this writing.
CASE 2 .- The second instance was that of a glioblastoma of the cerebral cortex in the 10-year-old sister of patient 1. Grossly,
AJDC-Vol 139, Feb 1985
1
2
I
B-83
Pr d65
4 82
3
4
5
O
7
II
1
2
-
C M
66
Sa
B
37
30
D
38
Sa
42
B 81
41
BT d
51
Lx
48 57
66
d
d
d
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
III
30
31
.
66
64
d24
Lx
50
BT6
59
Sa d
46 1.8
AC23 d24
47
44
d12
Lu
59 60
64
37
AC3 d4
14
Lu d 14
d14
61
d
Lk
35
58 di
61
LE13 ₫13
43
Sa8
B29 Pa43 d44
46
$2
Lk12 di3
BT11 d11
B
23
Lu
55
de
d 36
d9
d 25
d
55
1
2
3
4
5
6
7
8
9
10
11
12
13
1
15
16
17
18
9
20
21
22
23
24
IV
3
5
3
.
3
T
Ly32 d32
34
31
2
BT18
Sa10
BT3
22
26
18
BT4
14
da
39
39
35
d20
d11
d3
d5
19
31
Site
1
2
3
4
5
V
Code
AC, Adrenocortical
BT10 14 ₫11
13
9
Sa2
5
Male, Female, Unaffected Pedigree Code
B, Breast
1
2
BT, Brain Tumor
Cancer Verified by Pathology
C, Colon
C43
B54 Cancer Site and Age, yr, When Diagnosed
Lk, Leukemia
56
d60 Age, yr, Living or Deceased
Lu, Lung
.
Multiple Primaries Verified by Pathology
Lx, Larynx
☐
:: Cancer Verified by Death Certificate
Ly, Lymphoma Pa, Pancreas
Cancer Reported by Family History
Pr, Prostate
4
No. of Progeny
Sa, Sarcoma
Proband
First-degree Relatives
Extended to Include Second-degree Relatives
Pedigree of a family with sarcoma, breast cancer, lung cancer, and adrenocortical carcinoma syndrome (updated from Lynch et al4).
the specimen consisted of four hemorrhagic fragments of tissue. The sections showed a highly cellular anaplastic glial tumor with an unusual perivascular pattern. The neo- plastic cells showed hyperchromatic nuclei with prominent nucleoli. Numerous mitotic figures as well as areas of necrosis were present. Following a craniotomy, the pa- tient received irradiation therapy, but died of her primary disease one year later.
COMMENT
Patients with hereditary forms of cancer (including the SBLA syn- drome), which lack distinguishing physical signs such as those that are found in more than 50 cancer-associ- ated genodermatoses7 or biomarkers such as the elevation in the calcitonin level in multiple endocrine neoplasia IIa and IIb, must await the expression of the specific varieties of cancer (phe- notype) before the deleterious geno- type can be identified.
Four new patients have manifested cancer in the subjects’ family since our original report in 1978.4 The proband’s son (Figure, IV-1) had a lymphoma at the age of 32 years and the paternal grandfather (Figure, III-12) of our pa- tients died of lung cancer at the age of 60 years. Tumors in the children in- cluded an embryonal rhabdomyosar- coma in soft tissue from above the right eye (Figure, V-5), and a glio- blastoma of the brain (Figure, V-1). There were no distinct histologic fea- tures in these two tumors that differ- entiated them from similar tumors in other patients. It is essential that the clinician obtain a detailed and accurate family history concerning previous and current cancers throughout the pedigree in the search for the pos- sibility that hereditary factors may be operative.
Problems in pedigree analysis in the SBLA syndrome are confounded by the fact that in addition to reduced penetrance of the deleterious gene, one encounters two age-specific modes of cancer expression, one in childhood and the second in adult life. In the subjects’ family, the affected siblings’ 39-year old father (Figure, IV-15) was in the direct genetic lineage, but he had not yet manifested cancer at the time of writing. Nevertheless, two of his five children (Figure, 1, V-1 and V-5) manifested cancers consonant
with the SBLA syndrome. Each of the remaining children (Figure, V-2 through 4) are at a 50% risk for the development of cancer. Their mother (Figure, IV-16) and her relatives have been cancer-free.
We wish to emphasize that syn- drome identification may not be possi- ble when one deals with only a limited portion of the family. Note that the highlighted segment of the pedigree (Figure) involving the subject cancer- affected children and their unaffected but putative cancer-prone, genotype- carrier father does not harbor suffi- cient information to signify hereditary cancer syndrome diagnosis (SBLA). Given this much information, one would not even appreciate the fact that the father of the affected children is himself at inordinately high risk for cancer. However, by extending the pedigree to involve second-degree rel- atives, we note that the paternal grandfather had lung cancer. Still, these observations are insufficient in themselves to delineate a hereditary cancer syndrome. However, when the pedigree is extended to include addi- tional direct lineage relatives, a pat- tern begins to emerge for some of the important tumors that constitute this syndrome; namely, adrenocortical car- cinoma in a paternal great-aunt (Fig- ure, III-15), lung cancer at the age of 44 years in a paternal great-uncle (Fig- ure, III-16), and early-onset breast cancer in the paternal great-grand- mother (Figure, II-3). To date, this large kindred had showed a total of 26 syndrome tumors that have been iden- tified in patients in the direct genetic lineage. Six patients had sarcoma (three were children), four patients had breast cancer (all had an ex- tremely early onset), seven patients had brain tumors (five were children), two patients had both laryngeal and lung carcinoma, two patients had lung cancer, two patients had adrenocor- tical carcinoma (one was a child), and three patients had leukemia (two were children).
These observations are exceedingly important in that they clearly depict the need for surveillance/management programs for the progeny of affected as well as unaffected persons in the direct genetic lineage for syndrome
cancers in the SBLA syndrome. When a parent in the direct genetic lineage is unaffected, the risk to his progeny is 25%, but when any of the children in a sibship becomes affected, we must then reassess a 50% risk to each of his/ her siblings. This is of major impor- tance for genetic counselling.
It is the physician’s responsibility to provide empathetic and accurate infor- mation to members of families with the SBLA syndrome as well as any of the more than 100 mendelian inherited cancer-predisposing diseases (more than 50 fulfill the criteria for cancer- associated genodermatoses).7,8 This can only be accomplished by meticu- lous family study. As a minimal pre- caution, we recommend extending the pedigree to include siblings, parents, paternal and maternal grandparents, and aunts and uncles (modified nuclear pedigree). Because of the generally later age at the onset of cancer, older adults will usually be more informative than the progeny, nephews, and nieces of the proband. However, in syn- dromes where both childhood- and adult-onset cancers are characteristic (as in the SBLA), one must be keenly aware of the natural history and tumor pattern to capitalize on the potential significance of childhood cancer occur- rence for the facilitation of hereditary cancer syndrome identification.
This study was supported in part by the Ne- braska Division-Fraternal Order of Eagles and grant 1297-AR2 from the Council for Tobacco Research, USA, Inc.
References
1. Lynch HT: Hereditary factors in childhood cancer. Paediatrician 1982;11:205-221.
2. Draper GJ, Heaf MM, Kinnier-Wilson LM: Occurrence of childhood cancers among sibs and estimation of familial risk. J Med Genet 1977;14: 81-90.
3. Birch JM, Hartley AL, Marsden HB, et al: Excess risk of breast cancer in the mothers of children with soft tissue sarcomas. Br J Cancer 1984;49:325-331.
4. Lynch HT, Mulcahy GM, Harris RE, et al: Genetic and pathologic findings in a kindred with hereditary sarcoma, breast cancer, brain tumors, leukemia, lung, laryngeal, and adrenal cortical carcinoma. Cancer 1978;14:2055-2064.
5. Li FP, Fraumeni JF: Soft tissue sarcomas, breast cancer, and other neoplasms: a familial syndrome? Ann Intern Med 1969;71:747-752.
6. Li FP, Fraumeni JF: Prospective study of a family cancer syndrome. JAMA 1982;247: 2692-2694.
7. Lynch HT, Fusaro RM: Cancer-Associated Genodermatoses. New York, Van Nostrand Rein- hold Co, 1982.
8. Lynch HT: Cancer Genetics. Charles C Thomas Publisher, Springfield, Ill, 1976.