BCC is the commonest cancer in European-derived populations and Australia has the
highest recorded incidence in the world, creating enormous individual and societal cost in
management of this disease. The incidence of this cancer has been increasing
internationally, with evidence of a 1 to 2% rise in incidence in Australia per year over the
last two decades.
The main four epidemiological risk factors for the development of BCC are ultraviolet
radiation (UVR) exposure, increasing age, male sex, and inability to tan. The pattern and
timing of UVR exposure is important to BCC risk, with childhood and intermittent UVR
exposure both associated with an increased risk. The complex of inherited characteristics
making up an individual’s ‘sun sensitivity’ is also important in determining BCC risk.
Very little is known about population genetic susceptibility to BCC outside of the rare
genodermatosis Gorlin syndrome. Mutations in the tumour suppressor gene patched
(PTCH) are responsible for this BCC predisposition syndrome and the molecular
pathway and target genes of this highly conserved pathway are well described.
Derangments in this pathway occur in sporadic BCC development, and the PTCH gene is
an obvious candidate to contribute to non-syndromic susceptibility to BCC.
The melanocortin 1 receptor (MC1R) locus is known to be involved in pigmentary traits
and the cutaneous response to UVR, and variants have been associated with skin cancer
risk. Many other genes have been considered with respect to population BCC risk and
include p53, HPV, GSTs, and HLAs. There is preliminary evidence for specific familial
aggregation of BCC, but very little known about the causes.
56 individuals who developed BCC under the age of 40 in the year 2000 were recruited
from the Skin and Cancer Foundation of Australia’s database. This represents the
youngest 7 – 8% of Australians with BCC from a database that captures approximately
10% of Sydney’s BCCs. 212 of their first degree relatives were also recruited, including
89 parents and 123 siblings of these 56 probands.
All subjects were interviewed with respect to their cancer history and all reports of cancer
verified with histopathological reports where possible. The oldest unaffected sibling for
each proband (where available) was designated as an intra-family control. All cases and
control siblings filled out a questionnaire regarding their pigmentary and sun sensitivity
factors and underwent a skin examination by a trained examiner. Peripheral blood was
collected from these cases and controls for genotyping of PTCH. All the exons of PTCH
for which mutations have been documented in Gorlin patients were amplified using PCR.
PCR products were screened for mutations using dHPLC, and all detectable variants
Prevalence of BCC and SCC for the Australian population was estimated from incidence
data using a novel statistical approach. Familial aggregation of BCC, SCC and MM
occurred within the 56 families studied here. The majority of families with aggregation of
skin cancer had a combination of SCC and BCC, however nearly one fifth of families in
this study had aggregation of BCC to the exclusion of SCC or MM, suggesting that BCCspecific
risk factors are also likely to be at work. Skin cancer risks for first-degree
relatives of people with early onset BCC were calculated: sisters and mothers of people
with early-onset BCC had a 2-fold increased risk of BCC; brothers had a 5-fold increased
risk of BCC; and sisters and fathers of people with early-onset BCC had over four times
the prevalence of SCC than that expected. For melanoma, the increased risk was
significant for male relatives only, with a 10-fold increased risk for brothers of people
with early-onset BCC and 3-fold for fathers.
On skin examination of cases and controls, several phenotypic factors were significantly
associated with BCC risk. These included increasing risk of BCC with having fair, easyburning
skin (ie decreasing skin phototype), and with having signs of cumulative sun
damage to the skin in the form of actinic keratoses. Signs reflecting the combination of
pigmentary characteristics and sun exposure - in the form of arm freckling and solar
lentigines - also gave subjects a significantly increased risk BCC. Constitutive red-green
reflectance of the skin was associated with decreased risk of BCC, as measured by
spectrophotometery. Other non-significant trends were seen that may become significant
in larger studies including associations of BCC with propensity to burn, moderate tanning
ability and an inability to tan. No convincing trend for risk of BCC was seen with the
pigmentary variables of hair or eye colour, and a non-significant reduced risk of BCC
was associated with increasing numbers of seborrhoeic keratoses.
Twenty PTCH exons (exons 2, 3, 5 to 18, and 20 to 23) were screened, accounting for
97% of the coding regions with published mutations in PTCH. Nine of these 20 exons
were found to harbour single nucleotide polymorphisms (SNPs), seen on dHPLC as
variant melting curves and confirmed on direct sequencing. SNPs frequencies were not
significantly different to published population frequencies, or to Australian general
population frequencies where SNP database population data was unavailable. Assuming a
Poisson distribution, and having observed no mutations in a sample of 56, we can be
97.5% confident that if there are any PTCH mutations contributing to early-onset BCC in
the Australian population, then their prevalence is less than 5.1%.
Overall, this study provides evidence that familial aggregation of BCC is occurring, that
first-degree relatives are at increased risk of all three types of skin cancer, and that a
combination of environmental and genetic risk factors are likely to be responsible. The
PTCH gene is excluded as a major cause of this increased susceptibility to BCC in
particular and skin cancer in general. The weaknesses of the study design are explored,
the possible clinical relevance of the data is examined, and future directions for research
into the genetics of basal cell carcinoma are discussed.