Breast Cancer

Breast cancer is now known to occur as part of a high penetrance predisposition such as in LFS, and in BRCA1/ 2 families, but may also be caused by mutations in genes such as ATM and PTEN which confer a risk of <50%. Breast cancer has long been known to have a familial tendency, as discussed earlier, and there is a profusion of supporting literature. Evidence from meticulous epide-miological studies shows that 4---5% of breast cancer is due to a dominant cancer gene with high penetrance and a population frequency in the USA of 0.003 (Claus et al., 1990). Studies in the UK have confirmed this population frequency and gave useful data on which risk estimation can be based. Important factors which point towards a possible familial predisposition are the number of relatives, particularly first degree, who have been affected,

Figure 7 A family tree showing a dominant inheritance pattern of a TP53 codon 191 two base pair deletion in a family with LFS.

the age at which they developed the disease (early onset more significant) and whether bilateral or associated with other tumours.

The search for the gene or genes responsible for dominantly inherited breast cancer was dogged by some of the problems found in LFS. Ascribing status is difficult in unaffected cases even late in life and many of the known affected cases have died. Obtaining samples in some cases may depend on the use of stored paraffin block material, which may be unavailable and relies on having polymerase chain reaction technology and suitable probes, which are of course the norm nowadays. Additionally, as breast cancer is so common, affecting one in 11 — 12 women in their lifetime in the UK (HMSO, 1998), chance aggregations are likely to occur and non-gene mutation carriers in dominant families may be affected. The other problem, which could only be found by trial and error, was that of locus heterogeneity. Many chromosomal locations had been implicated by cytogenetic and LOH studies on tumour material. Chromosomal regions known to show involvement in more than 20% of tumours by molecular studies are 1p, 1q, 3p, 11p, 13q, 14q, 15q, 17p, 17q and 18q. Many of these regions were already known to contain tumour suppressor genes, e.g. RB on 13q, TP53 on 17p and DCC on 18q, and these genes are likely to be involved in a multistage process towards malignancy. In a major breakthrough, Hall et al. (1990) were able to show linkage in some breast cancer families to 17q12— 21. They looked at over 20 families from whom they had collected samples over many years, including many cases that had subsequently died. Nevertheless, they still had to use a PCR probe to work with paraffin block tumour samples in some cases. When all families were included in the linkage analysis the region on 17q was excluded. However, when the families were stratified in terms of their average age at onset, the first seven families showed a significant linkage to 17q12 (lod score approaching 6). They argued that a large proportion of early-onset breast cancer families (<46 years) were caused by a mutated gene on 17q. Without the work of Hall et al., and subsequently by the Breast Cancer Linkage Consortium, it could have been many years before research was focused on this region. The problem of genetic locus heterogeneity was only overcome by a combination of meticulous collection of samples, innovative ideas and luck. Another possible gene that was implicated at about this time was the oestrogen receptor gene on chromosome 6. However, this has not since been confirmed. Following the discovery of linkage of breast cancer to 17q (Hall et al., 1990), Narod et al. (1991) undertook linkage on five families with breast/ovarian aggregation. They found that three of the families were linked to a locus at 17q12— q23 and their additive lod scores reached statistical significance. Subsequent work by the Breast Cancer Linkage Consortium showed that 80% of breast/ovarian families with four or more affected patients were linked the 17q locus (Easton et al., 1993). The following year heralded the identification of the first major breast cancer predisposing gene BRCA1 (Miki et al., 1994). Surprisingly, BRCA1 does not appear to be involved as a significant acquired mutation (somatic) in non-hereditary breast cancer. In the same month that the BRCA1 gene was identified, the location of a second gene dubbed BRCA2 was announced on chromosome 13. A year later BRCA2 was cloned and again there was little evidence of involvement in sporadic disease (Wooster et al., 1995). It is now clear that although mutations in BRCA1 and BRCA2 account for the majority of high-risk breast cancer families (85%) and nearly all breast/ovarian families, in smaller aggregations they account for <50% of the hereditary element (Ford et al., 1998). While there is no doubt that BRCA1/2 are highly penetrant genes, initial estimates of the lifetime risk of 85% (Easton et al., 1993) appear slightly high. Population studies do detect BRCA1 and BRCA2 mutations in blood samples from apparently sporadic breast cancer patients (Peto et al., 1999). Furthermore, founder mutations in the Jewish and Icelandic populations where BRCA1/2 mutation frequencies can be as high as 2—2.5%, are associated with lifetime risks of breast cancer of 40—60% (Struewing et al., 1997). Outside populations with significant founder effects the frequencies of BRCA1/2 mutations combined is probably no higher than 0.2%.

Having identified the most important high penetrance genes, the search is on for lower penetrance genes. Aggregation of breast cancer has been shown to occur in ataxia telangiectasia heterozygotes (Swift et al., 1987, 1991), who are the carriers of the recessive gene which causes a disease which predisposes especially to haema-tological malignancy in childhood. A mother of an affected child is at 3-5-fold risk of breast cancer, which would fulfil a dominant gene model with 25---40% penetrance and a population frequency of about 0.01. Since the isolation of the ATM gene (Savitsky et al., 1995) there have been conflicting studies as to whether this gene is a significant cause of breast cancer. Breast cancer is also thought to occur in 30% of women with Cowden's disease (a condition predisposing to multiple hamartomas), but since the discovery of the underlying gene defects in the PTEN gene, no studies have found the gene to be involved in familial aggregations of breast cancer.

Quit Smoking For Good

Quit Smoking For Good

Quit smoking for good! Stop your bad habits for good, learn to cope with the addiction of cigarettes and how to curb cravings and begin a new life. You will never again have to leave a meeting and find a place outside to smoke, losing valuable time. This is the key to your freedom from addiction, take the first step!

Get My Free Ebook

Post a comment