Aetiology of sex differences in colorectal cancer.

Abstract

The aetiology of sex differences in colorectal cancer Ž is controversial. Fernandez et al. European Journal ́ . of Cancer Pre ention 9: 99 104, 2000 explained favourable trends for women analysing colorectal mortality sex ratio trends in major European countries over the last four decades. Their conclusions were based on the protective effect of female hormones, including replacement therapy and oral conŽ . traceptives. McMichael and Potter 1980 raised this hypothesis for the first time after narrative review Ž . study. Meanwhile, MacLennan et al. 1995 reported no association after their meta-analysis study. We think that this protective effect could be excluded on seeing the colorectal cancer mortality rates among women during the last 40 years. It is obvious that in Ž nearly all Northern countries Sweden, Switzerland, UK, Scotland, Austria, Belgium, The Netherlands . and Norway , in spite of lowering use of oral contraŽ . ceptives IARC, 1999 , these rates are decreasing. In general, the colorectal cancer mortality rate among females in nearly all Northern countries is falling compared with the increase in all Southern countries. Perhaps these contradictions come from other unconsidered risk factors, which must be taken into account. Sex-specific mortalities could be an explanation for the favourable trends for women in colorectal cancer. According to WHO, among males in the age group 35 39 years suicide is the principal cause of mortality, and between 40 and 64 years acute myocardial infarction, leaving the field open for different types of cancer, such as lung cancer and colorectal cancer. For women between 35 and 64 years breast cancer is the principal cause of mortality, leaving a small percentage of women vulnerable for colorectal Ž cancer Anuario de la Sanidad y del Medicamento . en Espana, 2000 . Moreover, breast cancer, the most ̃ frequent malignancy among females with rising incidence in all European countries over the last 50 years, represents the majority of female proportional Ž mortality attributed to cancer Boring et al., 1993; . Morillo et al., 2000 . These point at selective mortality among females by breast cancer, a cancer known to be more frequent among females using exogenous Ž . hormones Kelsey and Bernstein, 1996 . The possible carcinogenic role of smoking in colorectal cancer is a cause for concern. Many previous Ž studies Potter, 1996; Tulinius et al., 1997; World Cancer Research Fund and American Institute for . Cancer Research, 1997; Slattery et al., 1998 have considered smoking as a possible risk factor for colorectal cancer, especially rectal cancer. Smoking is well known to be more prevalent among males than females. In Spain, the male to female ratio of smoking in the population older than 15 years is 1.7, a ratio that coincides with that recorded for colorecŽ tal cancer mortality rate Castillejo and Mercade,́ . 2000 . We agree with Fernandez et al. in singling out ́ alcohol drinking as a possible risk factor for colorectal cancer. For example, Hungary in 1994, where the Ž alcohol consumption was quite high 10.2 litres of . pure alcohol per person yearly , had the highest age-standardized mortality rate both for men and women. In general, countries with high colorectal cancer mortality rates, especially from 1990 to 1994, are the countries with the highest consumption of alcohol, such as, for example; France, Portugal, Hungary, Denmark, Germany, Austria and SwitzerŽ . land Castillejo and Mercade, 2000 . Simanowski et ́ Ž . al. 1995 described and explained the effect of alcohol on colorectal cell regeneration as a possible mechanism of alcohol-associated carcinogenesis. Alcohol was found to be both an independent and a tobacco-interactive risk factor, influencing the canŽ cer incidence for the colorectum Chhabra et al., 1996; World Cancer Research Fund and American Institute for Cancer Research, 1997; World Cancer . Research Fund, 1998 . Based on these findings, we would like to raise concerns about the rising incidence of colorectal cancer in Europe among both men and women, and the evidence regarding the carcinogenic effects of alcohol drinking and smoking. Finally, we acknowledge such well-designed eco-