The epidemiology of large-bowel cancer

Familial colon cancer in the Tel-Aviv area and the influence of ethnic origin

The family history of colon cancer was investigated in 38,823 individuals (2,129 families) who comprised a control and an oncology patient series from Tel-Aviv and nearby areas. A significant increased risk for colon cancer was observed among first-degree relatives of colon cancer patients when compared to controls. When the patient sample was divided into two groups based on country and continent of birth--European (Ashkenazim) and other (nonAshkenazim)--the relatives of the nonAshkenazi subjects showed a greater relative risk for colon cancer (P less than 0.05). Colon cancer was found to be less frequent in nonAshkenazim than in Ashkenazim controls. These findings suggest that although the colon cancer frequency in the nonAshkenazi group is lower, the genetic component may be more important than for the Ashkenazi sample. The nonAshkenazi Jews may represent distinct subgroups that differ with respect to either primary genetic susceptibility to colorectal cancer and/or they may have been subjected to peculiar, environmental carcinogenic exposures when compared to their Ashkenazim brethren.

Dietary fiber and colon cancer: animal model studies

Support for a protective role for certain dietary fibers in the etiology of colon cancer has come from nutritional epidemiologic studies. Recommendations to increase consumption of fiber-containing food and decrease the intake of dietary fat should form the basis of a diet that is unlikely to do harm, and may have the potential for reducing the development of colon cancer, in humans. Studies examining the role of dietary fiber as an inhibitor of colon cancer in animal models appear to have provided some conflicting results, due mostly to differences in the nature and amount of carcinogen used to induce colon tumors, variation in the composition of the experimental diets, and relative difference in food intake by animals, to cite a few of the methodological problems. However, overall, the feeding of wheat bran appears to inhibit color tumor development to a greater degree than do other dietary sources of fiber.

Effect of long term lactulose ingestion on secondary bile salt metabolism in man: potential protective effect of lactulose in colonic carcinogenesis

This study investigated whether colonic absorption of secondary bile acids, especially deoxycholate in patients with adenomas could be decreased by oral lactulose. Bile acid metabolism was studied using bile sampling and 14C-deoxycholate kinetics in patients with colonic adenomatous polyps before and after four and 12 weeks of lactulose, 60 g/day. The results indicate that lactulose decreased the deoxycholate pool size from a mean of 22.0 (SD: +/- 13.8) to 14.3 (+/- 7.6) mumol/kg (p less than 0.025). Deoxycholate absorption fell from 3.8 (+/- 2.3) to 2.9 (+/- 1.4 mumol/kg/d (ns). The biliary bile acid composition decreased significantly in deoxycholate after four and 12 weeks with a rise in primary bile acids. There was a highly significant correlation between the %-change in DCA input and the %-change in DCA pool size (r = 0.89). Intestinal transit measured by the pellet method (4.1 +/- 1.9 to 2.4 +/- 0.6 day; p less than 0.01) and faecal pH decreased, while stool frequency and weight rose significantly. Significant correlations between the %-change in gut transit time and the %-change in DCA pool size or %-change in DCA input were absent. The results show that it is possible to lower colonic secondary bile acid absorption by long term lactulose feeding. This effect can be mediated by accelerated transit and the acidification of the colonic contents.

Morphological and cell kinetic effects of dietary manipulation during colorectal carcinogenesis

The effect of dietary manipulation of fat and fibre on the structural and cell kinetic characteristics of colonic mucosa was studied before and during experimental carcinogenesis in 232 male Albino Swiss rats. Carcinogen treated animals were given 12 weekly injections of azoxymethane (10 mg/kg/week). The animals were divided between four dietary groups (1) high fat, high fibre, (2) low fat, high fibre, (3) high fat, low fibre and (4) low fat, low fibre. Pathological and cell kinetic information together with details of certain faecal characteristics was collected when the animals were killed 4, 20, and 28 weeks after starting their experimental diet. Tumour induction was significantly influenced by diet. The highest risk of colorectal tumour development was found in groups fed diet 3: high fat, low fibre (p less than 0.03). In contrast, diet 2: low fat, high fibre was associated with the lowest risk. The proportion of histologically proven colonic tumours occurring in each dietary group was: diet 1-10.9%, diet 2-3.6%, diet 3-63.7%, diet 4-21.8%. Scanning electron microscopic (SEM) studies done on selected samples indicated both dietary and azoxymethane related alterations in crypt unit integrity. The most marked surface architectural changes were seen in carcinogen treated animals maintained on diet 3 (high fat, low fibre). Stathmokinetic analysis revealed considerable intergroup variability. Both fat and fibre produced significant effects, principally during the preneoplastic phase of carcinogenesis. Faster proliferative activity tended to be found in animals at low risk of tumour induction (diet 2), slower proliferation being more characteristic of animals at high risk (p less than 0.05). The findings suggest that both topographical and cell kinetic parameters have an important relationship with promoting and protecting dietary factors during the development of colorectal cancer.

Is the 7,12-dimethylbenz[a]anthracene-induced rat mammary tumor model suitable as a preclinical model to study mammary tumor malignancy?

To study the biological characteristics of 7,12-dimethylbenz[a]anthracene (DMBA)-induced mammary tumors in rats, 20 Sprague Dawley female rats received a single oral dose of 5 mg of this carcinogen. During the 35-week observation time 78 primary tumors were removed. While in most cases the primary tumor could be removed completely, 7 out of 20 animals eventually had to be sacrificed for inoperable local recurrence of the primary tumor. Notwithstanding, the long period of time given for tumor metastases to develop (mean time between tumor removal and termination was 18.5 weeks), tumor spread either to lungs or regional lymph nodes could not be established. This relatively benign behavior of the tumor was in contrast with the morphological characteristics of the tumor, which uniformly showed the features of adenocarcinomas. The difference in biological behavior between DMBA-induced mammary tumors in rats and malignant mammary tumors in humans suggests that as a model this system is of limited value for investigations of mechanisms of malignant behavior of human tumors.

Ulcerative colitis: the result of an altered bacterial metabolism of bile acids or cholesterol

It is proposed that in genetically susceptible individuals ulcerative colitis is caused by a bacterial metabolite of bile acids or cholesterol and that this substance is similar or identical to the bacterial metabolites implicated in the development of colon cancer. Since the responsible bacterial reactions may be dependent on vitamin K as an electron acceptor it is suggested that poorly absorbed vitamin K antagonists, specifically alpha-tocopherylquinone, may be capable of inhibiting these reactions and may therefore prove effective in treating ulcerative colitis and in preventing the development of colon cancer.

Familial heterogeneity of colon cancer risk

The authors have assembled detailed family histories of cancer on 857 cancer probands, of whom 180 manifested colorectal carcinoma. This study determines if some families had a greater risk for colorectal cancer than others, and if so, what factors were associated with an increase in risk. To test for the possibility of heterogeneity of risk, a parameter called the Z-score, was calculated for each family. The Z-score is a measure of the number of cancer cases in the family adjusted for the number of expected cases. A permutation test was employed to test whether or not the variance of Z-scores from the sample was greater then expected by random chance. The variance for families ascertained through colon cancer probands, but not in any of the other groups, was significantly increased. Of the colon group, 10.6% fell into a high-risk category, as did 5.56% of the rectal cancer families, but only 3.95% of the other groups combined were at high risk. Anatomic sites (in the proband) with the highest Z-score variances were sigmoid and transverse colon, whereas lower variances were seen for cecum and descending colon. Risk status therefore may be partially dependent upon exact anatomic sites within the colon. The effect of proband's age of diagnosis was not significant, but did show the possibility of an effect on heterogeneity of risk for both the younger and older groups.

Colorectal polyps and the risk of subsequent carcinoma

Of 323 patients with colorectal polyps that either were larger than 1 cm in diameter or were 1 cm or smaller but had been subjected to biopsy or excision, 20 (6%) had subsequent development of a colorectal carcinoma, an incidence rate of 4.6 per 1,000 person-years of follow-up. These 20 cases of cancer compared with 7.38 cases expected; thus, the risk of colorectal carcinoma in these patients was 2.7 times that in the general population (P less than 0.001). Although risks were greater for certain subgroups of patients, the differences detected were usually modest. We conclude that patients with large colorectal polyps (more than 1 cm in diameter) should be closely followed up after treatment of the initial polyp, regardless of the size, site, or histologic type of the polyp, the age or sex of the patient, or the type of initial treatment (excision or fulguration). Those patients with more than one polyp, especially if more than one segment of the colon is involved, need special attention. Examination of the entire colon is important because the site of development of carcinoma may be distant from the site of the initial polyp.

Bowel symptoms in an apparently well population

The aim of this study was to determine the prevalence of various kinds of bowel behavior and symptoms thought to be indicative of colorectal cancer in people randomly selected from the community. A probability sample of 330 dwellings in the inner western suburbs of Sydney yielded 202 completed interviews with occupants aged 30 years and older. Eight percent reported annoying abdominal pain that had lasted for two weeks or more in the preceding six months, while 19 percent reported a feeling of incomplete evacuation at least once every two weeks. Blood on the toilet paper was reported by 14 percent and blood in the toilet bowl by 2 percent. Twenty-one percent said they always looked at their stool in the toilet bowl and 34 percent always looked at the toilet paper after using it, but 43 percent seldom or never looked at either their stool or the paper. Of the 75 who said they looked at their stool about half the time or more, two (3.1 percent) reported seeing blood during the preceding six months. Symptoms that may be associated with colorectal cancer are common in apparently well adults. Whilst this includes bleeding from the rectum in toto, it may not be true for blood seen specifically in the toilet bowl. Because this latter symptom has potential discriminating value, it may be worthwhile to promote public education encouraging people to inspect their stools regularly, and to visit their doctor if blood is seen.

[Epidemiology of colorectal cancer--internationally and in Switzerland]

The descriptive epidemiology of colorectal cancer in Switzerland is presented and compared to international geographic patterns and time trends. Switzerland has a typical western pattern of relatively high incidence and mortality rates. Mortality from colon cancer is stable with a slight male preponderance; rectal cancer mortality is decreasing, especially in females. The proportion of rectal cancers in the incidence data of colorectal cancer is not as prominent as shown by French or German cancer registries. There is an urban-rural-gradient, more prominent in colon than in rectal cancer. As in other countries, white collar occupations show elevated PMRs and manual occupations low ones, a finding which is confounded with the geographical distribution of these occupations.

Metabolic epidemiology of colon cancer: fecal mutagens in healthy subjects from rural Kuopio and urban Helsinki, Finland

Fecal mutagenic activity and dietary pattern of rural and urban Finnish population groups with distinct risk for the development of colon cancer were studied in a low-risk population in rural Kuopio and an intermediate-risk population in urban Helsinki. The average daily intake of protein and fat was the same in the two groups but the frequency of consumption of whole-grain cereals and whole-grain bread, as well as the amount of fiber from the bread were higher in Kuopio as compared to Helsinki. Fecal samples collected for 2 days were incubated under anaerobic conditions at 37 degrees C for 96 h, extracted with hexane: peroxide-free diethyl ether, partially purified on a silica Sep-Pak cartridge, and assayed for mutagenic activity using the Salmonella typhimurium/mammalian microsome system. The fecal mutagenic activity was observed with the tester strains TA98 and TA100 with and without microsomal activation in both the population groups. The percentage of samples showing fecal mutagenic ratio greater than 3 with TA98 and TA100 with microsomal activation, was higher in volunteers from Helsinki than in Kuopio.

Hereditary nonpolyposis colorectal cancer (Lynch syndromes I and II): a common genotype linked to oncogenes?

We hypothesize that: hereditary nonpolyposis colorectal cancer (HNPCC) accounts for a significant proportion of the colon cancer burden; the premalignant colonic mucosa in HNPCC patients may be comparable in many respects to the premalignant polyps in its hereditary multiple adenomatous polyposis counterpart; and finally, study of HNPCC colonic mucosa may harbor important clues for elucidation of the multistep process of carcinogenesis, including relationships between activated oncogenes, genetics, and environmental perturbations.

Descriptive epidemiology of colorectal cancer

The purpose of this report is to present the descriptive epidemiology of colorectal cancer using the most recent data available to highlight two characteristics of the disease. First is the great variation which takes place in the frequency of this disease over geographic areas of all sizes. Colorectal cancer is common in most countries of North America and Europe, is rare in Asia and is particularly uncommon in Africa. Internationally, the variation in colon cancer is 60-fold, and within Europe there is a 4-fold difference in the incidence of colon cancer between areas with the highest and lowest rates. For cancer of the rectum, variation internationally is 18-fold and within Europe it is 3-fold. Within the United Kingdom, colon cancer is uniformly higher in the 5 Scottish Cancer Registry Regions than in their counterparts in England and Wales. Even within Scotland there is a 4-fold range of colon cancer incidence in rates, with the North and South clearly demarcated by a striking difference in colon cancer incidence in both sexes. Secondly, examination of international mortality rates for colorectal cancer demonstrates remarkable differences in trends over time between countries. In countries where colorectal cancer mortality rates were initially low, rates have increased substantially. In many countries where rates circa 1950 were moderately high, they have increased slightly or become stabilized. However, in countries such as Scotland, Canada, England and Wales and the United States, where rates were initially high, there have been gradual falls in mortality over time.

Colorectal cancer in a nuclear family. Familial or hereditary?

Because of the high incidence of colorectal cancer, familial aggregations of this disease are common. Differentiation between etiologies contributing to familial clustering (which may have resulted either from common environmental exposure or from mere chance) and primary genetic factors may prove vexing to the physician. This report deals with the myriad problems encountered when attempting to make such etiologic distinctions in order to provide appropriate surveillance and management, based upon tumor spectrum and natural history, for patients at increased cancer risk.

Treatment of small colorectal polyps: a population-based study of the risk of subsequent carcinoma

We conducted a retrospective cohort study of 751 residents of Rochester, Minnesota, whose small (1 cm or less) colorectal polyps had been treated without biopsy. Eighteen colorectal carcinomas were found in this group in the course of more than 10,000 person-years of follow-up, whereas 15.27 cases were expected (relative risk, 1.2). The overall rate for the development of cancer and the relative risks for various subgroups of the cohort were not significantly elevated, and survival in this cohort was not impaired. We conclude that fulguration of selected small colorectal polyps without prior biopsy does not subject patients to an unusual risk for development of colorectal carcinoma. Customary periodic medical examination should suffice for patients with these lesions.

Geographic distribution and demographic correlates of colorectal cancer mortality in Sydney, New South Wales

Colorectal cancer (CC) has the highest incidence of any internal malignancy in Australia and is a major cause of death. A knowledge of the geographic distribution of CC will aid in the identification of high risk groups for whom diagnostic and treatment facilities should be provided and may give clues to aetiological factors. Total CC deaths and age-standardized mortality ratios for Local Government Areas in Sydney, Australia were mapped and the relationships between mortality and selected demographic measures were examined. Several statistically significant zero-order correlation coefficients were found but multiple regression analysis showed a dominant independent negative effect of distance from the centre of the city on CC mortality. No substantial independent effects of demographic variables were observed. The existence of presently unmeasured environmental aetiological factors is postulated.

Cancer risks in the optical manufacturing industry

A mortality odds ratio (MOR) study has been conducted to explore the cancer risks of exposures experienced in the production of optical lenses and metal spectacle frames. Male death certificates were obtained from a Massachusetts town where a large optical industry is located. Craftsmen, foremen, and operatives of non-optical industries, such as woollen textile workers and workers in the optical company with short-term or no exposure, were chosen as reference workers their incomes were similar to those of the exposed workers. Cardiovascular disease (total 714) is chosen as the reference disease to explore cancers (total 232). An excess risk of total cancers observed = 70, expected = 48) has formed among lens workers. The excess may be accounted for mainly by the excess risk of gastrointestinal cancers; the standardised MORs (sMOR) for medium and long-term exposure were 2.2 and 2.5. The excess was especially evident for colorectal cancers; the sMORs for medium and long-term exposures were 3.2 and 2.6. Excess risks of gastrointestinal cancers (sMOR = 2.9) and colorectal cancers (sMOR = 3.4) were found among metal frame workers with long-term (employed for more than 29 years) exposure, but the number of exposed cases was small (9 and 6 respectively). These results suggest that exposure to abrasives or cutting oil mists or both, possibly by ingestion, might increase the risk of gastrointestinal (especially colorectal) cancers among lens and metal spectacle frame manufacturers.

Metabolic epidemiology of colon cancer: dietary pattern and fecal sterol concentrations of three populations

The nutrient intake, fecal neutral sterol concentration, and bile acid concentration of populations with a varied risk for colon cancer development were investigated. High-risk populations in the metropolitan New York area and Malmo, Sweden, were compared with an intermediate-risk population in Umea, Sweden. The mean daily intake of protein and fat was comparable in all groups, but the total daily fiber intake was higher in Umea, as was the total daily stool output. There was no difference in the total fiber intake and stool output between Malmo and metropolitan New York. The fecal secondary bile acid concentration was lower in Umea than in the other two areas; no difference was observed between Malmo and metropolitan New York. These results suggest that high fiber intake may be considered protective against colon cancer even in a population with a high risk intake. A high dietary fiber intake may limit colon cancer risk by increasing stool bulk, and thus diluting and/or binding tumor promoters.

Relative carriage rates of nuclear dehydrogenating clostridia in two populations of different colorectal cancer risk

Carriage of nuclear dehydrogenating clostridia has been associated with colon cancer and implicated in its aetiology. This study has compared the carriage of these organisms in a British population at high risk for the development of colon cancer with a low risk Nigerian population. Clostridia were found in all of the stools from both populations. Nuclear dehydrogenating clostridia were only found in the stools of the British subjects (32%). These results support the suggestion that the carriage rate of nuclear dehydrogenating clostridia in a population is related to the risk of colon cancer.

Socio-economic class differences in cancer incidence in Cali, Colombia

Based on census tract information, cancer incidence rates for three socio-economic strata of the city of Cali, Colombia, were calculated. Strong negative associations with socio-economic status were found for cancers of the cervix and stomach. Colon cancer and endocrine-related cancers were positively associated with socio-economic status, while no such association was found for rectal cancer. Contrary to data from developed countries, all smoking-related cancers were positively associated with socio-economic class. The role of socio-economic gradients in developing countries is stressed as a basis for etiological research.

Mutagens in human faeces. Are they relevant to cancer of the large bowel?

1. Mutagenic activity has been detected in faecal extracts, prepared by a number of methods, from donors living under widely differing geographical, cultural and dietary circumstances. Faecal extracts cause point mutations in bacteria and chromosomal damage in cultured mammalian cells. 2. The claims that nitroso compounds are present in human faeces have been retracted, and the chemical nature of faecal mutagens is still unknown. Indirect evidence suggests the presence of several classes of mutagen. 3. The use of different methods of mutation assay gives conflicting estimates of the proportion of people who excrete mutagenic faeces. There is wide variation in mutagenic activity between different stool samples from one person, and between different stool samples from different people. There is conflicting evidence for inhibition or enhancement of the mutagenicity of reference mutagens by faecal extracts. The effects of air oxidation on the mutagenicity of faecal extracts have not been investigated in detail. 4. It has been claimed that the proportion of people excreting mutagenic faeces is higher in groups representing populations at high risk of large-bowel cancer than in groups at low risk of large-bowel cancer. For the reasons given in paragraph 3, these claims must be regarded as premature. 5. The part played by faecal mutagens in the aetiology of large-bowel cancer has yet to be determined.

Fecal sterols and bacterial beta-glucuronidase activity: a preliminary metabolic epidemiology study of healthy volunteers from Umea, Sweden, and metropolitan New York

The dietary pattern, fecal bile acid and neutral sterol concentrations, and the bacterial beta-glucuronidase activity of 2 population groups with a varied risk for colon cancer development (i.e., a high-risk population in the metropolitan New York area and an intermediate-risk population in Umea, Sweden) were investigated. The average daily intake of dietary protein was the same in the 2 groups, but the fat intake was higher in Umea than in New York. The daily total fiber intake was also higher in Umea, as was the daily total stool output. The concentration of fecal secondary bile acids and beta-glucuronidase activity was lower in Umea than in New York, but the total daily excretion of these constituents was the same in both groups. The data suggest that one of the factors contributing to the lower risk of colon cancer in Umea, despite the high dietary fat intake, is the high intake of dietary whole grain and cereal fiber, which leads to an increase in stool bulk, thus diluting and/or binding promoters.

Problems in the assessment of relative risk of chronic disease among biological relatives of affected individuals

A question often asked in regard to a chronic disease is whether the risk to a biological relative of a case is evaluated, and if so by how much the risk is altered. To answer this question, data may be collected directly with genetic objectives in mind by ascertaining population of pedigrees. More often, the initial assessment of the question comes from family history data collected in an incidental manner in the course of a case-control or similar type of study. This paper discusses some limitations to the inferences which can be derived from such casual family history data. These include (i) poor statistical properties of standard relative risk measures, (ii) interpretational problems of observed relative risks when affected cases arise from genetic as well as nongenetic causes and when genes may not always be expressed in individuals in whom they are present, and (iii) confounding effects which may occur when a high risk allele alters the age of onset pattern of the disease. These problems result largely from a loss of design control over the degree of exposure of individuals ascertained and can lead to a small observed relative risk value even when genetic factors are important. Suggestions for handling such casual family history data are offered.

Atherosclerosis and polyps of the colon. Quantification of precursors of coronary heart disease and colon cancer

The prevalence of adenomatous and hyperplastic polyps of the colon and the degree of involvement of the aorta by atherosclerotic lesions were evaluated in 842 autopsies from New Orleans. A positive statistical association between the two variables was found in white males under 75 years of age. In black males a positive correlation between atherosclerotic lesions and adenomatous (but not hyperplastic) polyps was found. The most extensive involvement of the aorta by atherosclerosis was found in patients in whom adenomatous and hyperplastic polyps coincided. The possibilities that dietary factors may account for these findings and its relevance in relation to colon cancer and coronary heart disease are discussed.

Age dependent differences in human bile acid metabolism and 7 alpha-dehydroxylation

It has been suggested that transformation of secondary bile acids into (co)carcinogenic compounds may have a role in the development of cancer of the large bowel. Because of age dependent differences of this disease we undertook a study of cholic and deoxycholic acid metabolism of eleven young adults (group A, 20-30 years old) and eleven elderly persons (group B, 55-75 years old) with a double isotope dilution method. Daily food intake was standardized individually and gut transit time measured with radioopaque pellets and labelled chromium chloride. The 7 alpha-dehydroxylation fractions (the ratio of deoxycholic acid input rate from the large bowel to cholic acid synthesis rate) were higher in group B (P less than 0.01) due to higher deoxycholic acid input rates (P less than 0.005), especially when individuals from both groups with rapid gut transit were compared. As contributory factor was recognized the higher fractional turnover rate of cholic acid in group B. Pool sizes and synthesis rates of cholic acid and gut transit times were similar. In group A, but not in B, gut transit times correlated with deoxycholic acid input rates (P less than 0.01). The differences in bile acid metabolism may be related to a more effective colonic absorption of deoxycholic acid in the elderly persons with a concomitant decrease of active ileal absorption of cholic acid in the elderly persons. Differences in diet or gut transit time between both groups do not seem to be the underlying mechanism.

Knowledge of colorectal cancer and attitude to occult blood testing among recent medical graduates in New South Wales

The extent to which new information on colorectal cancer (C.C.) is being transmitted to medical undergraduates was assessed in a repeated survey of new interns' knowledge of and attitudes on C.C. A probability sample of 214 graduates of the two Sydney medical schools was drawn as they commenced their internships in 1978. A comparable sample of 186 was drawn in 1980. Interns in each sub-sample completed a questionnaire of 54 pre-coded items of knowledge of C.C. epidemiology, early diagnosis, and followup, and on their attitudes towards C.C. On most topics knowledge was not well developed, and 1980 graduates were not significantly better informed than those of 1978. Almost three-quarters of 1980 interns intend using faecal occult blood testing.

Diet and cancer of the large bowel

The incidence of large bowel cancer correlates internationally with the consumption of fat, protein or meat. Etiological hypotheses have been proposed, suggesting that a high-fat diet is related to an increased risk of large bowel cancer, while dietary fiber is claimed to have a modifying effect. However, while these hypotheses were derived from the analysis of the geographical distribution of colon cancer in humans, they are not unequivocally supported by the results of further epidemiological studies. This is particularly true for case-control studies which generally have not confirmed an association between diet and the risk of colorectal cancer at the individual level. Lack of uniformity among the findings of epidemiological studies concerned with the role of diet in the etiology of bowel cancer could be explained by 1. lack of sensitive and reproducible methods for assessment of diet in epidemiological studies, 2. lack of knowledge of the mode of action of dietary factors responsible either for the development of the tumor or for protection against it. Future epidemiological research on diet and large bowel cancer should deal with more precisely defined hypotheses in which the possible multifactorial nature of this tumor would be taken into account.