Trends in cancer incidence in Cali, Colombia.

Trends in cancer incidence from 1962 to 177 provided by the cancer registry of Cali, Colombia, are presented. A decline in rates of cancer of the oral cavity, esophagus, and larynx, has coincided with an increase in the rates of lung cancer in women of all ages and in men over 65 years of age. No such rise in lung cancer has been detected in young males. Colon cancer incidence has increased, and a larger proportion of tumors in later years is concentrated around the sigmoid colon. Cervical cancer incidence has declined, but carcinoma in situ of the cervix is being more frequently diagnosed probably as a consequence of vaginal cytology screening programs. Similarities of these findings with those of Puerto Rico and contrasts with trends in the United States and Europe are discussed.

Social class differences among patients with large-bowel cancer in Cali, Colombia.

The incidence of large-bowel cancer in Cali, Colombia, for 1962-71 shows the upper socioeconomic classes to be a higher risk. This is the first report of a socioeconomic gradient in risk for this site. The gradients were most marked for cancer of the ascending through rectosigmoid colon and were minimal for cancer of the cecum and rectum. The Cali experience presents several parallels with information derived from comparisons of developed and developing countries and also appears consistent with recent information on the possible role of dietary factors in bowel cancer.

Epidemiologic pathology of gastric ulcer and gastric carcinoma among Japanese in Hawaii.

This case-control study of Hawaiian Japanese indicated that gastric ulcer in the proximal portion of the pyloric antrum has features similar to those of gastric cancer. Such ulcers occurred at sites most frequently and most severely affected by intestinal metaplasia, although metaplasia tended to be more extensive with cancer than with ulcer. Metaplastic mucosa was more vulnerable to the action of pepsin and acid than was normal mucosa. The risk of ulceration would rise when a sufficiently lagrge area of the antrum was intestinalized and when the corpus continued to produce significant quantities of these substances. This study showed a strong association between salt intake, ulcer, and metaplasia. Significant but less dramatic associations were demonstrated between metaplasia and the use of traditional Japanese foods and smoking. The question was raised as to whether salt promotes ulceration or whether it potentiates the action of a mutagen that causes intestinal metaplasia.

A case-control study of large bowel cancer in Japan.

This investigation of diet and other environmental factors reports on 588 patients with colorectal cancer and 1,176 hospitalized controls in three prefectures of Japan. Weak (not statistically significant) positive effects were found for social class and urbanization. The significant association of colorectal cancer with consumption of beef, string beans, or starches previously described for Hawaiian Japanese were not reproduced here. An association with hakusal (cabbage) agreed with other reports on a negative association with cruciferous vegetables. An analysis of the subset of cases in the low rectum yielded results similar to those for the total series. The failure to uncover important food effects in Japan is attributed to the difficulty of detecting case-control differences in areas with homogeneous diet practices. Further epidemiologic research aided by leads from ongoing work with animals may provide ideas for more sharply defined questions, should stress new approaches for more accurate diet histories, and should continue to emphasize tumor localization.

Premalignant lesions of the cervix in women of Cali, Colombia.

A detailed histologic study was made of uterine cervices from 441 autopsy specimens from the population of Cali, Colombia. Women in this city have one of the highest registered incidence rates of cancer of the uterine cervix. The following prevalence (in percent) of lesions in adult women was found: cervicitis, 79; reserve cell hyperplasia, 14; squamous metaplasia, 41; dysplasia, 9. Contrary to expectations, no increase in prevalence of these lesions was found with age, lower socioeconomic status, or number of pregnancies. A similar study of a sample of hysterectomies performed for uterine prolapse showed approximately the same results. The findings suggest that promotional factors are more important than initiator factors in uterine cervical carcinogenesis.

Stomach cancer in Japan.

A study of 783 patients with stomach cancer and 1,566 hospital controls in Hiroshima and Miyagi prefectures of Japan showed that farmers, representing mostly the lowest socioeconomic class, had higher risk of developing stomach cancer. The usual inverse gradient in risk by social class was in the urban population of Miyagi, but not Hiroshima, prefecture. The study in Japan did not reproduce the association of stomach cancer with consumption of salted/dried fish and salt-pickled vegetables described for the Hawaiian Japanese. Salted/dried fish and pickled vegetables were more widely used by farmers than by nonfarmers in Japan or by Japanese migrants to Hawaii. The ability to detect associations for these typical Japanese foods in Hawaii stemmed from the fact that these reduced levels of use were more completely expressed by the Hawaiian-Japanese controls than by patients. The lower risk of developing stomach cancer for lettuce and celery users agreed with the Hawaiian-Japanese findings, and the combined results supported conjectures on possible protective food effects. Lettuce, in particular, warranted attention from this viewpoint, since similar findings have been consistently reported.

A cohort study of stomach cancer risk in men after gastric surgery for benign disease.

BACKGROUND: For the past 40 years, investigators have suggested that there exists an increased risk of stomach cancer following gastric surgery for benign disease. Recent cohort studies have consistently identified an increased risk of stomach cancer beginning 20 years or more following gastric surgery. Validation of this association and elucidation of risk factors related to gastric cancer have been complicated by variability in study designs. PURPOSE: This cohort study was designed to investigate the risk of stomach cancer following gastric surgery and to identify patient and treatment characteristics that may alter this risk. METHODS: Medical admission records of 17077 male military veterans hospitalized during 1970-1971 in U.S. Department of Veterans Affairs (VA) hospitals were examined. From this initial cohort, 1094 patients who died within the 1st year following gastric surgery were excluded. Data analysis was performed on the final cohort consisting of 15,983 patients divided into the following two groups: 1) an exposed group (gastric surgery group) that included 7609 patients receiving gastric surgery for a documented benign disorder and 2) an unexposed group (comparison group) that included 8374 male patients randomly selected from all other hospitalized male patients in the patient database. The comparison group was matched to the gastric surgery group by age (within 10 years), race, hospital, and year of admission. Mortality follow-up utilized the following three sources to identify vital status: 1) the VA Patient Treatment File (1970-1988), 2) the VA Beneficiary Identification Record Linkage System (1970-1989), and 3) the National Death Index (1979-1988). Death certificates were obtained for 99% of the deceased patients. Analyses included estimations of risk using standardized rate ratios (SRRs) and proportional hazards techniques. RESULTS: A statistically significant increase in risk of stomach cancer was demonstrated among males during the 20 years following gastric surgery (SRR = 1.9; 95% confidence interval [CI] = 1.3-2.4; P = .0001). The risk of developing gastric cancer was greatest during the 2nd to 5th postoperative years (SRR = 2.8; 95% CI = 1.6-4.5; P < .01) and during years 11-15 (SRR = 2.5; 95% CI = 1.2-4.8; P < .01). Also, the risk of developing gastric cancer was greatest among those treated by gastrectomy for any type of ulcer (SRR = 2.6; 95% CI = 1.2-4.9; P < .01) and those having any type of gastric surgery when the primary diagnosis was gastric ulcer (SRR = 2.9, 95% CI = 1.4-5.3; P < .01). CONCLUSIONS: This study confirms that men undergoing gastrectomy for benign disease and men receiving any gastric surgery for gastric ulcer are at increased risk for developing gastric cancer. Unlike earlier studies, we find that the increased risk is not delayed for 20 years.

Diet and gastric cancer: nutrition survey in a high-risk area.

A nutrition survey was done in four Colombian villages situated in the rural area of Nariño, previously identified as exhibiting different levels of risk for gastric cancer and its precursor lesions. The survey was based on personal visits by trained interviewers to a random sample of families in each village. The major findings of the survey reflected a higher consumption of fava beans in villages with high indices of gastric cancer risk and a higher consumption of fresh fruits and vegetables in villages where such indices were lower. Excessive salt intake and lower potassium intake in villages with a high risk of gastric cancer were suggested, but further work is required to test this preliminary observation. A generally low intake of animal proteins and a high intake of cereals in Nariño did not discriminate between villages at different levels of risk to gastric cancer and its precursors.

Gastric cancer in Colombia. I. Cancer risk and suspect environmental agents.

A case control study of patients discharged from hospitals revealed fourfold differences in geographic variation in stomach cancer risk within the Department of Narino (Colombia). Data from gastroscopic surveys of population groups, samples of water supplies, and urine and saliva in Narino also indicated a generally positive correlation among the following parameters: 1) gastric cancer risk, 2) prevalence of chronic atrophic gastritis and intestinal metaplasia, 3) nitrate content of well waters, and 4) nitrate excretion by the population. Urinary excretion reflected the ingestion of nitrates, and this implied a higher average intake of nitrates in the populations at high risk for stomach cancer. The Narino data could be construed as presumptive epidemiologic evidence for the role of nitrate availability in the etiology of stomach cancer.

Gastric cancer in Colombia. II. Case-control epidemiologic study of precursor lesions.

Endoscopic surveys in the Department of Narino (Colombia) showed that natives of areas of very high risks for stomach cancer have a higher prevalence of atrophic gastritis and intestinal metaplasia than do persons born in other parts of Narino. In these surveys, diets were compared between individuals with and without evidence of these suspect precursor lesions of stomach cancer and between individuals born in the areas designated as high and low risk for stomach cancer. The composite evidence from the comparisons indicated lettuce to be negatively associated with stomach cancer and raised the possibility to a positive association for corn. The relationship of these findings to other epidemiologic evidence and the implications for applying the epidemiology of stomach cancer to the epidemiology of precursor lesions were examined.

Dietary determinants of gastric cancer in south Louisiana inhabitants.

In south Louisiana, 391 recently diagnosed gastric cancer patients and an equal number of controls were interviewed. Questions asked covered residential and occupational histories, environmental exposures, tobacco use, diet, alcohol consumption, and pertinent demographic characteristics. Elevated relative risks were found for use of tobacco and alcohol products. Diet was found to be the main determinant of gastric cancer risk in south Louisiana. Both dietary patterns and dietary risk factors differed for blacks and whites, although fruits as a group and dietary vitamin C were found to exert strong protective effects for both blacks and whites. Consumption of smoked foods and homemade or home-cured meats increased risk of gastric cancer for blacks but not for whites. The findings are discussed in the light of the prevailing etiologic hypotheses.

Diet and chronic atrophic gastritis: a case-control study.

A hospital-based case-control study of gastric cancer precursor lesions was conducted in a high-risk black population in southern Louisiana. Ninety-three subjects with biopsy-proved advanced chronic atrophic gastritis were compared to two control series: a gastroscopy clinic series and a general hospital-admission series. Dietary case-control differences indicated a protective effect associated with fruit and vegetable intake and with dietary vitamin C and a risk elevation associated with milk consumption. The protective effect associated with consumption of fruits, vegetables, and vitamin C is consistent with findings for gastric cancer and with the etiologic hypothesis of intragastric nitrosation. A twofold increased risk was associated with cigarette smoking. Gastric juice pH, NO3-, and NO2- were determined for subjects undergoing gastroscopy, and comparisons were made between this high-risk U.S. group and a Colombian population with a much greater magnitude of risk; the latter had higher NO3- and NO2- levels. An increase in pH was associated with increasing severity of gastric lesions. Levels of pH and NO2- concentration were significantly correlated (P less than .0005); however, in Louisiana the large difference in NO2- concentration associated with pH elevation is not associated with histopathologic severity. Divergent trends with severity of lesions for NO3- concentration were seen in the two populations.

Gastric cancer in Colombia. III. Natural history of precursor lesions.

The premalignant process in the gastric mucosa was studied by gastroscopic surveys of Colombian populations, and the prevalence of superficial gastritis, chronic atrophic gastritis, and intestinal metaplasia was calculated for population samples having a very high gastric cancer risk (Nariño), very low risk (Cartagena), and intermediate risk (Cali). The prevalence of individuals with normal mucosa in successive age groups was used to estimate "depletion" curves, which were taken as indicators of the dynamics of the premalignant process in each community. Differences corresponding to the geographic variation in stomach cancer risk were found: In the high-risk areas of Nariño, around 75% of the population developed some type of gastritis by 45 years of age, whereas in the low- and intermediate-risk population of Cartagena and Cali, the proportion of such lesions did not exceed 50% at age 45 or thereafter. The effect of environmental factors in early life seemed to be important in determining the prevalence of lesions in each population.

Developments in the epidemiology of stomach cancer over the past decade.

The history of stomach cancer epidemiology is reviewed. The introduction of migrant population studies in the 1960 decade that described the critical role of exposures to this disease in early life was a key event. Companion pathology studies have indicated different epidemiological patterns for 2 histological entities, intestinal and diffuse type carcinomas, and confirmed an excess of intestinal metaplasia in populations at high risk to stomach cancer. Recent results suggest that epidemiology of stomach cancer can be transformed into the epidemiology of precursor lesions, and introduction of the fiberoptic gastroscope makes technically feasible detailed studies of the relationship of precursor lesions to suspect factors, including diet, in selected geographic areas. Nitroso compounds have been identified as candidate carcinogens and the epidemiological, pathological, and chemical data display signs of internal consistency. Feeding experiments with N-methyl-N'-nitro-N-nitrosoguanidine have led to animal models that permit a coordinated epidemiological-experimental approach to stomach cancer.

A case-control study of large bowel cancer and hormone exposure in women.

Several lines of evidence indicate a potential role for hormonal or reproductive factors in the subsequent development of large bowel cancer in women. To evaluate the relationship between hormone exposure and large bowel cancer a case-control study was carried out in 18 Illinois hospitals. Female cases, ages 45-74 (n = 90), and controls (n = 208) were identified from an ongoing large bowel cancer study. Data were obtained from medical records, personal interviews, and a subsequent mail survey with a questionnaire specific to hormone usage. Menopausal estrogen use was found to be protective with respect to the subsequent development of large bowel cancer with an odds ratio of 0.6 (95% CI, 0.33-0.99). This effect remained after controlling individually for age at diagnosis, ever pregnant (yes/no), parity, age at first birth, hysterectomy with documented oophorectomy, cholecystectomy, and appendectomy. Simultaneous adjustment, using logistic regression, for age at diagnosis, parity, hysterectomy, and cholecystectomy resulted in an adjusted odds ratio for menopausal estrogen use and large bowel cancer of 0.5 (95% CI, 0.27-0.90). Subsite analysis revealed the protective effect to be strongest for the rectal cancer cases. These data support the hypothesis that exogenous hormones may alter the risk of large bowel cancer in women.

Gastric precancerous process in a high risk population: cross-sectional studies.

The gastric precancerous process is evaluated in 1788 participants in a gastroscopy survey in the population of Nariño, Colombia, which has one of the highest gastric cancer incidence rates on record. A detailed histological classification is used, and a hierarchical distribution of lesions is described with the main stages being gland neck hyperplasia, atrophy (gland loss), intestinal metaplasia and dysplasia. Acute inflammation was not found to be a specific stage in the sequence but rather a common finding in all stages of the precancerous spectrum. Indices of disease progression for the different steps are calculated and found to increase with gastric pH and nitrate and nitrite content of the gastric juice. The effects of high pH and nitrite content are intimately correlated. Relative risks of specific lesions, namely, hyperplasia, atrophy, metaplasia, and dysplasia, increase linearly with higher pH, nitrate, and nitrite values in the gastric juice. The severity of atrophy correlates with the prevalence of metaplasia, suggesting a sequential relationship between the described stages, a finding supported by all parameters examined. The model of progression described may serve as a basis for comparisons with populations at different levels of gastric cancer risk but it fails to provide information concerning the time required for each change, which should be provided by follow-up (cohort) studies.

Gastric precancerous process in a high risk population: cohort follow-up.

In an attempt to characterize the natural history of the gastric precancerous process, 1422 residents of a high risk area of Nariño, Columbia, have been followed from 3-16 years (average 5.1) with repeated gastric biopsies, for a total of 7290 person-years. The original cohort consisted of 1788 individuals yielding a successful completion rate of 79.5%. Comparison of initial and subsequent biopsies revealed a very complex dynamic flow of both progressive and regressive events, suggesting sporadic environmental forces of modulation. One-time measurement of gastric juice, pH, and nitrite failed to predict future events in the gastric mucosa. The net loss of individuals whose gastric mucosa initially showed normal histology or superficial gastritis was 3.3%/year, representing a net gain of 1.7% for chronic atrophic gastritis, 0.9% for intestinal metaplasia, and 0.7% for dysplasia. The incidence rate of gastric cancer in this population was 0.16/100 person-years. The net rates of progression were higher and those of regression lower in older compared to younger individuals. The general pattern detected is that of a slow forward movement in the previously described hierarchical organization of precursor lesions. The presence of progressive as well as regressive changes and the slow pace of change offer special opportunities to inhibit progression through intervention strategies targeting previously identified etiological factors. The difficulties and opportunities offered by the long term follow-up studies as well as the congruency of the findings with current etiological hypotheses are discussed.

Sodium intake and gastric cancer.

To test the hypothesis that excessive intake of sodium chloride is a factor in gastric carcinogenesis, urinary excretion of sodium and creatinine was measured in Colombian subjects. Age, sex, weight, and height regression slopes for creatinine excretion were more similar in Colombia than in other countries. Sodium/creatinine (S/C) ratios correlated with 24-h urinary excretion of sodium revealed higher sodium excretion in populations with higher gastric cancer rates. The S/C ratios were not affected by circadial rhythms, making it possible to use single urine samples to investigate interpopulation differences in sodium excretion.

[Risk factors for cancer of the esophagus: a case control study in a metropolitan area of south-eastern Brazil]. / Factores de risco para o câncer de esôfago: estudo caso-controle em área metropolitana da região Sudeste do Brasil.

The association between esophageal cancer and smoking and drinking habits, instruction and nutritional factors was examined by means of the utilization of data of a case-control study which was undertaken in the city of S. Paulo (Brazil). Eighty five cases were compared with two hundred and ninety two hospital controls, with different diseases, including other kinds of cancer. The crude estimations of the odds ratios were calculated for all the variables. Logistic regression was used in the next steps of the analysis. The drinking habit [odds ratio = 3.68, 95% confidence interval (1.74 - 7.78)], the smoking habit [odds ratio 4.86; 95% confidence interval (1.95 - 12.13)] and the frequent eating of hot pepper [odds ratio = 2.48; 95% confidence interval (1.46 - 4.23)] are important risk factors for the disease. The estimate of odds ratio for smoking ordinary cigarette was 3.43 (1.31-8.97) and for smoking of corn straw hand-rolled cigarrette was of 4.18 (1.38-12.66).