Download editorials - CiteSeerX

EDITORIALS
Molecular Epidemiology of Basal
Cell Carcinoma
Curtis C. Harris*
Journal of the National Cancer Institute, Vol. 88, No. 6, March 20, 1996
p53 mutants that provide cells with a selective clonal expansion
advantage during the multistep process of carcinogenesis.
BCC is the most common histologic type of skin cancer in
humans and rarely metastasizes or causes death. In this issue of
the Journal, Gailani et al. (16) have found an association between sunlight exposure and molecular changes, i.e., loss of
heterozygosity (LOH) of alleles on chromosome 9, and the
spectrum and frequency of p53 gene mutations in BCC from 58
patients. A striking finding of their study was the high frequency
(68%) of LOH for chromosome 9q alleles irrespective of histologic characteristics or clinical behavior of BCC. This finding
suggests that a "gatekeeper" cancer susceptibility gene(s)
resides at chromosome 9q22, which also is the locus associated
with the nevoid BCC syndrome (12,13,17,18). B. Vogelstein
(personal communication) proposed the gatekeeper gene concept whereby genetic or epigenetic inactivation of the gatekeeper is needed to start the process of carcinogenesis. The
gatekeeper gene may vary among tissue sites, e.g., the Rb (also
known as RBI) gene in retinoblastoma and the APC gene in
colon carcinoma, and may lead to a clonal survival advantage of
the cell containing the inactivated gatekeeper gene. Because
members of the cell cycle G, checkpoint pathway, CDK4 and
p 16INK4, are both candidate gatekeeper genes of melanomas, one
can extend this concept to a gatekeeper "pathway." Within a tissue such as the skin, in which BCC and melanoma arise from
different cell types, more than one gatekeeper pathway may be
found in a single tissue type.
Downloaded from http://jnci.oxfordjournals.org/ at Pennsylvania State University on February 26, 2014
The skin is a large organ directly exposed to physical and
chemical carcinogens, which produce frequent cancers. Studies
of skin cancer have figured prominently in cancer epidemiology
(including that of occupational cancers such as scrotal cancers in
chimney sweeps) and in laboratory investigations of chemical, physical, and viral carcinogenesis [reviewed in (/)].
Ultraviolet (UV) radiation (UVB [wavelengths of 280-320 nm]
and to a lesser extent UVA [wavelengths of 320-400 nm] in sunlight) are considered to be the major etiologic agent of squamous
cell carcinoma (SCC), basal cell carcinoma (BCC), and melanoma (2-4). Autosomal recessive, e.g., xeroderma pigmentosum, and autosomal dominant, e.g., nevoid BCC syndrome
(Gorlin's syndrome), inherited cancer-prone conditions have
been identified [reviewed in (7)]. Germline mutations in specific
cancer susceptibility genes involved in nucleotide excision
repair (XPA, XPB, XPC, XPD, XPF, and XPG) [reviewed in
(5)] or cell cycle control [pl6 INK4 or cyclin-dependent kinase
type 4 (CDK4)] (6-11) have been identified in xeroderma pigmentosum and familial melanoma, respectively. The search for
cancer susceptibility genes is continuing for other inherited cancer-prone conditions, e.g., nevoid BCC, where linkage analysis
has identified a locus at chromosome 9q22 (12,13). Somatic abnormalities in skin carcinomas have been observed in genes encoding proteins controlling cell cycle (e.g., Ha-ras, cyclin D 1;
pl6 INK4 , and p53), apoptosis (e.g., p53), and genomic stability
(e.g., p53 and cyclin D|) [reviewed in (/)].
The analysis of germline and somatic mutation spectra of the
p53 tumor suppressor gene has contributed significantly to the
development of the field of molecular epidemiology of human
cancer. Characteristic p53 mutation spectra have been associated with dietary aflatoxin B) exposure and hepatocellular
carcinoma, sunlight exposure and skin carcinoma, cigarette
smoking and lung cancer, and occupational vinyl chloride exposure and hepatic angiosarcoma [reviewed in (74,75)]. These
studies have provided a molecular link between carcinogen exposure and specific types of human cancer. In contrast to the
somatic mutation spectra of many cancer types, the analysis of
the germline p53 mutation spectrum is consistent with the hypothesis that endogenous mutagenic mechanisms such as
deamination of 5-methylcytosine at CpG dinucleotides in
germline DNA are responsible for the majority of inherited cancer-prone conditions. The mutation spectrum also reveals those
Gailani et al. (76) also used the UVB-related mutations (C to
T and CC to TT transitions) at dipyrimidines in the p53 tumor
suppressor gene as a "molecular dosimeter" of sunlight exposure and concluded that the pathogenesis of mutations in a
gene(s) at chromosome 9q22 may involve factors other than
sunlight in a significant proportion of tumors. As the authors
recognize, this conclusion is based on a small dataset, and the
gene(s) at chromosome 9q22 has not as yet been identified.
* Affiliation of author: Laboratory of Human Carcinogenesis, Division of
Basic Science, National Cancer Institute, Bethesda, MD.
Correspondence to: Curtis C. Harris, M.D., National Institutes of Health,
Bldg. 37, Rm. 2C07, Bethesda, MD 20892-4255.
EDITORIALS
315
Basal Cell Carcinoma (n=64)
CC -> TT
Squamous Cell Carcinoma (n=33)
CC -> TT ( ; : C .> C : G
9%
Del.+Ins. ftVr
Del.+lns.
6%
6%
A :T->C:G
A:T->G:
5%
A:T->T:A
6%
9%
G:C->A:T
60%
-> T:A
21%
/
/
A: T->T:A
^
3%
•
^
\
G:C -> A:T
46%
Mutation Frequency 44%
Mutation Frequency 44%
XP: Basal Cell Carcinoma (n=7>
A:T->G:C
14%
Fig. 1. p53 mutation spectra
in skin carcinoma (27). Del. =
deletion; Ins. = insertion.
XP: Sauamous Cell Carcinoma (n=l(>)
cc •> n
40%
CC -> TT
86%
Mutation Frequency 27 %
316
A:T->C:G
10%
Mutation Frequency 48%
Therefore, one can better consider this conclusion as a hypothesis.
Epidemiologic studies have identified other etiologic agents
of skin cancers, including polycyclic aromatic hydrocarbons,
ionizing radiation, thermal injury, and arsenicals [reviewed in
(7)]. As Brash et al. (19) first pointed out, the majority of the
p53 mutations are transitions at dipyrimidines (including CC to
TT) and are characteristic of UVB exposure (Fig. 1). The p53
mutations caused by bulky chemical carcinogens such as aflatoxin B, or by physical carcinogens occur generally on the nontranscribed strand of DNA and reflect a more rapid repair of the
transcribed DNA strand (20,21). The most striking difference
between the p53 mutation spectra of BCC and SCC is the higher
frequency of G:C to T:A transversions in SCC, which may reflect exposure either to chemical carcinogens such as benzofo]pyrene, which have been found to cause these transversions in
an animal model of skin carcinogenesis (22), or to oxidative
damage (23-25). An inherited deficiency in nucleotide excision
repair of the nontranscribed DNA strand (xeroderma pigmentosum C) increases the proportion of C to T and CC to TT transitions and decreases the frequency of G:C to T:A transversions
but not the absolute frequency of p53 mutations (Fig. 1). The
dose of the mutagen may also affect the p53 mutation spectrum.
Therefore, the p53 mutation spectrum is molded by both
mutagenic agents and DNA repair rates. Because DNA repair
rates can be sequence dependent (26), the p53 mutation spectrum can be influenced by both the type and the location of the
promutagenic lesion. Because about one half of the BCC do not
harbor p53 mutations, one can speculate that inactivating mutations are found in other genes in the p53 pathway(s) or that an
independent pathway(s) is involved.
Considering the rapid pace of cancer research, we can soon
expect that the cancer susceptibility gene(s) on chromosome
9q22 will be identified and that the normal function of the
gene(s) and its product(s) will be elucidated. Determination of
the mutational spectrum of the identified gene(s) will provide
further insights into the etiology and molecular pathogenesis of
BCC and suggest additional strategies for its prevention and
treatment.
EDITORIALS
Journal of the National Cancer Institute, Vol. 88, No. 6, March 20, 1996
References
(/) Yuspa SH, Dlugosz AA. Cutaneous carcinogenesis: natural and experimental. In: Goldsmith L, editor. Physiology, biochemistry, and molecular
biology of the skin. New York: Oxford Univ Press, 1991:1365-402.
(2) Lee JA, Scotto J. Melanoma: linked temporal and latitude changes in the
United States. Cancer Causes Control 1993;4:4l3-8.
(3) Scotto J, Fears TR, Fraumeni JF Jr. Incidence of non-melanoma skin cancers in the United Stales. Washington, DC: National Institutes of Health;
Publication. 83-2433, 1983.
(4) Fitzpatrick TB, Sober AJ. Sunlight and skin cancer [editorial], N Engl J
Med I985;313:818-2O.
(5) Friedberg EC, Walker GC. Siede W, editors. DNA repair and mutagenesis.
Washington (DC): ASM Press, 1995;663-87.
(6) Wolfel T, Hauer M, Schneider J, Serrano M, Wolfel C. Klehmann-Hieb E,
et al. A pl61NK4-insensitive CDK4 mutant targeted by cytolytic T lymphocytes in a human melanoma. Science 1995;269:1281-4.
(7) Hussussian CJ, Struewing JP, Goldstein AM, Higgins PA, Ally DS.
Sheahan MD, et al. Germline pl6 mutations in familial melanoma [see
comment citation in Medline], Nat Genet 1994;8:15-2I.
(8) Ohta M. Nagai H, Shimizu M. Rasio D. Berd D, Mastrangelo M, et al.
Rarity of somatic and germline mutations of the cyclin-dependent kinase 4
inhibitor gene, CDK4I, in melanoma. Cancer Res 1994:54:5269-72.
(9) Kamb A, Gruis NA, Weaver-Feldhaus J. Liu Q. Harshman K. Tavtigian
SV, et al. A cell cycle regulator potentially involved in genesis of many
tumor types [see comment citations in Medline]. Science 1994;264:436-40.
(10) Gruis NA, van der Velden PA, Sandkuijl LA, Prins DE, Weaver-Feldhaus
J, Kamb A, et al. Homozygotes for CDKN2 (pl6) germline mutation in
Dutch familial melanoma kindreds. Nat Genet 1995:10:351 -3.
Downloaded from http://jnci.oxfordjournals.org/ at Pennsylvania State University on February 26, 2014
G:C->A:T
50%
(20) Evans MK. Taffe BG, Harris CC, Bohr VA. DNA strand bias in the repair
of the p53 gene in normal human and xeroderma pigmemosum group C
fibroblasts. Cancer Res 1993:53:5377-81.
(2/) Ford JM, Lommel L, Hanawalt PC. Preferential repair of ultraviolet lightinduced DiNA damage in the transcribed strand of the human p53 gene.
Mol Carcinog 1994;10:105-9.
(22) Ruggeri B, DiRado M, Zhang SY, Bauer B, Goodrow T, Klein-Szanto AJ.
Benzo[a]pyrene-induced murine skin tumors exhibit frequent and characteristic G to T mutations in the p53 gene. Proc Natl Acad Sci U S A
1993:90:1013-7.
(23) Hussain SP, Aguilar F, Amstad P, Cerutti P. Oxy-radical induced
mutagenesis of hotspot codons 248 and 249 of the human p53 gene. Oncogene 1994:9:2277-81.
(24) Cheng KC, Cahill DS, Kasai H, Nishimura S, Loeb LA. 8hydroxyguanine, an abundant product of oxidative DNA damage, causes G
to T and A to C substitutions. J Biol Chem 1992;267:166-72.
(25) Moriya M, Ou C, Bodepudi V, Johnson F, Takeshita M, Grollman AP.
Site-specific mutagenesis using a gapped duplex vector: a study of translesion synthesis past 8-oxodeoxyguanosine in E. coli. Mutat Res
1991:254:281-8.
(26) Tornaletti S, Pfeifer GP. Slow repair of pyrimidine dimers at p53 mutation
hotspots in skin cancer [see comment citations in Medline]. Science
1994:263:1436-8.
(27) Hollstein M, Rice K, Greenblatt MS, Soussi T, Fuchs R, Sorlie T, et al.
Database of p53 gene somatic mutations in human tumors and cell lines.
Nucleic Acids Res 1994;22:3547-51.
Do Our Current Cervical Cancer Control
Strategies Still Make Sense?
Nancy B. Kiviat, Laura A. Koutsky*
on acceptance of the idea [proposed almost 30 years ago by
Richart and Barron (J)] that CIS evolves from CIN 1 (and perhaps from ASCUS) and that CIN 1, CIN 2, and CIN 3/CIS represent a morphologic and biologic continuum of progressive,
consecutive stages in the development of invasive cancer. This
high rate of progression of CIN 1 lesions has recently been
questioned (2). Most recent natural history studies [reviewed in
(3)] suggest that fewer than 30% of the women with CIN 1 (who
had not had a biopsy or who had not been treated) will develop
CIN 3. It has become clear that many CIN 1 lesions are simply
self-limited cervical infections with either high- or low-risk
types of HPV. Data presented by Park et al. (4) support this
idea. The realization that the majority of low-grade CIN appears
to spontaneously resolve, along with the high costs incurred by
follow-up of all women with ASCUS and LGSIL, has prompted
a search for alternative methods for management of women with
these lesions. The possible use of DNA assays in the management of women with LGSIL or ASCUS is now being explored
by a multicenter National Cancer Institute-funded trial in which
women with ASCUS and LGSIL are randomly assigned to
The dramatic decrease in cervical cancer mortality seen over
the last 50 years is the result of cervical cancer control programs
conceived and established prior to understanding the importance
of human papillomavirus (HPV) in the development of both
benign and malignant neoplastic cervical lesions. At present in
the United States, cervical cancer control is achieved by routine
cytologic screening to identify women with Pap smears showing
"low- or high-grade squamous intraepithelial lesions (SIL)/carcinoma in situ (CIS)" or repeated "atypical squamous cells of
undetermined significance (ASCUS)." Women with SIL/CIS
and repeated ASCUS are referred for diagnostic colposcopy (examination of the cervix through a magnifying lens) and biopsy.
Those with biopsy-confirmed cervical intraepithelial neoplasia
grade 2 or 3 (CIN 2-3) and frequently those with CIN grade 1
(CIN 1) undergo ablative treatment of the transformation zone
(area on the cervix originally surfaced by columnar epithelium
that has undergone squamous metaplasia). Close follow-up of
women with ASCUS only is also deemed important by many
clinicians. This relatively aggressive approach to cervical cancer
control is based on the hypothesis that invasive cervical cancer
is preceded by an intraepithelial stage termed carcinoma in situ,
which, unless detected and eradicated in a timely fashion, evolves
into invasive cervical cancer within an average time span of 1020 years. Referral of women with low-grade SIL (LGSIL) for
colposcopy, biopsy, and close follow-up or treatment is based
*Affiliation of authors: Department of Pathology/Cytology, University of
Washington, Seattle.
Correspondence to: Nancy B. Kiviat, M.D., HPV Research Group, University
of Washington, Ann Bldg., Suite 310,6 Nickerson St., Seattle, WA 98109.
Journal of the National Cancer Institute, Vol. 88, No. 6, March 20, 1996
EDITORIALS
Downloaded from http://jnci.oxfordjournals.org/ at Pennsylvania State University on February 26, 2014
(//) Holland EA, Beaton SC, Becker TM, Grulet OM, Peters BA, Rizos H, et
al. Analysis of the pl6 gene, CDKN2, in 17 Australian melanoma
kindreds. Oncogene 1995; 11:2289-94.
(12) Gailani MR. Bale SJ, Leffell DJ, DiGiovanna JJ. Peck GL, Poliak S. et al.
Developmental defects in Gorlin syndrome related to a putative tumor suppressor gene on chromosome 9. Cell 1992;69:111-7.
(J3) Chenevix-Trench G, Wicking C, Berkman J, Sharpe H, Hockey A, Haan E.
et al. Further localization of the gene for nevoid basal cell carcinoma
syndrome (NBCCS) in 15 Australasian families: linkage and loss of
heterozygosity. Am J Hum Genet 1993;53:76O-7.
(14) Hollstein M, Sidransky D, Vogelstein B, Harris CC. p53 mutations in
human cancers. Science 1991;253:49-53.
(15) Greenblatt MS, Bennett WP, Hollstein M, Harris CC. Mutations in the p53
tumor suppressor gene: clues to cancer etiology and molecular
pathogenesis. Cancer Res 1994;54:4855-78.
(16) Gailani MR, Leffell DJ, Ziegler AM, Gross EG, Brash DE, Bale AE.
Relationship between sunlight exposure and a key genetic alteration in
basal cell carcinoma. J Natl Cancer Inst 1996;88:349-54.
(17) Quinn AG, Campbell C, Healy E, Rees JL. Chromosome 9 allele loss occurs in both basal and squamous cell carcinomas of the skin. J Invest Dermatol 1994:102:300-3.
(18) Farndon PA, Del Mastro RG, Evans DG, Kilpatrick MW. Location of gene
for Gorlin syndrome. Lancet 1992:339:581-2.
(19) Brash DE, Rudolph JA, Simon JA, Lin A, McKenna GJ, Baden HP, et al.
A role for sunlight in skin cancer: UV-induced p53 mutations in squamous
cell carcinoma. Proc Natl Acad Sci U S A 1991 ;88:10124-8.
317