We have presented data from a case/control study which suggest that the low-activity SULT1A1*His allele is associated with reduced colorectal cancer risk in Macedonia, primarily in females >60 years of age. This is an observation that has not been reported before. Published studies on the influence of the SULT1A1 variant on colo­rectal cancer risk either do not show any effect of this variant [15-19] or suggest that the SULT1A1*Arg allele is protective in patients <80 years of age [1]. It is difficult to explain these results outside the context of the overall influence of low penetrance genes as susceptibility factors of a disease that has proven dietary and lifestyle components. Although our current understanding is rather limited, it appears that gene-gene and gene-environment interactions of low penetrance alleles have a significant influence on susceptibility to colorectal cancer, and that there might be a different set of predisposition alleles peculiar to each population [4]. The findings we record here support our previous results on the different influences of TbR-I(6A) and Cyclin D1 G480A alleles between Macedonian and US/Northwestern European populations [21,22], and strengthen the notion that colorectal cancer in various geographic areas has a dissimilar molecular pathogenesis, probably reflecting different genetic backgrounds and environmental exposures [23]. Since these molecular differences could affect prevention strategies, therapeutic efficacy and transferability of clinical trial results in this part of the world, a large prospective multicenter study is warranted to evaluate this issue.
The finding that the low-activity SULT1A1*His allele is under represented primarily in women >60 years of age is very intriguing, and tempts speculation that the primary mechanism for this is the low potential of this variant for estrogen inactivation. Numerous studies have demon­strated the influence of female hormones on suppression of colorectal carcinogenesis, as manifested by a lower incidence of colorectal cancer in women during their reproductive period [2,24]. In postmenopausal women, estrogen production is rather low and “below the protective threshold” so the sex-difference in incidence of the disease disappears in women who are not using hormone replacement therapy [25,26]. We speculate that our female patients, who are homozygotes for the low-activity SULT1A1*His allele, have a slightly higher level of estrogen that is sufficient to suppress the formation of colorectal cancer. This is a testable hypothesis, that could contribute to the elucidation of the mechanisms of the action of estrogen on the colorectal carcinogenesis and to the definition of a low-dose hormone therapy for the prevention of colorectal cancer.