Search

It is well-known that second primary tumours occur at elevated levels in patients undergoing cancer treatment with cytotoxic drugs. While this is clearly an undesirable outcome, it represents one of the few situations where the relationship between dose and effect of a carcinogen can be studied in humans. Current epidemiological methods for studying this relationship are described, as well as some of the issues involved in the quantification of risk from human data. Finally, the possibility of measuring biochemical markers and other parameters of DNA damage in patients undergoing treatment with antineoplastic agents is discussed.

A randomised double-blind intervention trial was carried out in Huixian, Henan Province, People's Republic of China, to determine whether combined treatment with retinol, riboflavine, and zinc could lower the prevalence of precancerous lesions of the oesophagus. 610 subjects in the age group 35-64 were randomised to receive once a week the active treatment (15 mg [50 000 IU] retinol, 200 mg riboflavine, and 50 mg zinc) or placebo. Both at entry to the study and at the end of the treatment, 13.5 months later, the subjects were examined, with an emphasis on signs of vitamin A and riboflavine deficiences, and riboflavine, retinol, beta-carotene, and zinc levels were measured. Compliance was excellent. The final examination, on 567 (93%) subjects, included oesophagoscopy and at least two biopsies. The intervention did not affect the prevalence of oesophageal lesions: after one year, the prevalence of oesophagitis with or without atrophy or dysplasia was 45.3% in the placebo group and 48.9% in the vitamin/zinc treated group.

The purpose of this review has been to demonstrate that a wide range of epidemiological phenomena can be described in terms of simple multistage models of carcinogenesis. The relationship of cancer risk with the different time variables considered corresponds closely with the behaviour predicted by theories of multistage process. Furthermore, the different behaviour associated with different agents enables one to attempt some classification as to how an agent is acting. There is considerable conformity between the tentative classification one might propose for the few agents for which epidemiological data are adequate, and the activity seen experimentally for these agents. The purpose of using multistage models in epidemiology is not simply descriptive. A distant goal would be the use of experimental information on mechanisms of action for the prediction of human risk. Information currently available is clearly insufficient, but the consistency seen between the epidemiological and experimental results suggests that some progress may be possible.

Since the inception of cancer registration in Singapore in 1968, the incidence rates of malignancies have been monitored among the 3 major ethnic groups--Chinese (77% of population), Malays (15%) and Indians (6%). Substantial changes in incidence at various sites were observed in the period 1968-1982, which seem to represent a shift from the pattern of cancer incidence in China towards that prevailing in the West. We now report on the Singapore Chinese experience, in which the following changes were notable: decreases in rates of cancers of the stomach and oesophagus and increases in rates of cancers of the lung, colon, rectum, skin (excluding melanoma), breast and ovary. For several sites, the secular changes differed among age groups; among women under 50 years of age breast cancer increased particularly sharply and lung cancer decreased. Environmental and personal factors, potentially responsible for the change, are discussed.

In 1974 a non-randomised study of the effect of mass screening by physical examination and xeromammography on mortality from breast cancer was started. Of the 20 555 eligible women in the city of Utrecht born between 1911 and 1925 (aged 50-64 at the start of the study), 14 796 attended for screening. Four rounds of screening were carried out. The relative risk of dying from breast cancer among women ever screened compared with women never screened was 0.30 (95% confidence interval 0.13-0.70).

In recent years, epidemiological methods have been used increasingly to study the carcinogenicity of drugs used in the chemotherapy of cancer. Such studies are useful to clinicians in identifying therapeutic agents with particular long-term risk to patients. They can also provide information on the dose- and time-related risks of cancer in one of the few human populations intentionally exposed to known levels of carcinogens. Aspects of epidemiological studies of second cancer risk are described, including sources of cases, study design, statistical methods, and possible biases. Results from a cohort study of second cancers following ovarian cancer, testicular cancer and Hodgkin's disease and from a case-control study of leukaemia following Hodgkin's disease are also given.

The high incidence of lung cancer in Chinese females in Singapore, especially among those belonging to the Cantonese dialect group, and the relatively high rates in Chinese males have been studied by means of interviews of cases and controls. A significant dose-response effect of cigarette smoking was found for all male and female groups, but neither smoking nor any other exposure explains the high incidence of lung cancer observed in Cantonese females who exhibit high rates of adenocarcinoma appraently unrelated to smoking. In general, persons with a low consumption of green vegetables were at higher risk for lung cancer. This finding might be due to an increased susceptibility in the presence of a relative deficiency of vitamin A.

Back projection methods are used to predict the yearly number of new AIDS diagnoses and the number of new HIV infections, to the end of 1992. The AIDS, but not the HIV, predictions are insensitive to the choice of incubation period distribution. A wide range of predictions is consistent with the AIDS diagnoses in years up to 1987, but limited ancillary information on the relative rates of new HIV infection in 1984 to 1987 can be used to narrow this range. The range of prediction based on AIDS reports to the end of 1988 is lower and narrower than the range based on reports to the end of 1987. The number of new AIDS cases in 1992 appears likely to fall in the range 1000-3000.

It has been well established that many of the drugs used in cancer therapy are themselves potentially carcinogenic. It is therefore important to quantify the carcinogenic risk associated with specific agents, and to investigate ways of predicting their risk from animal and in vitro studies. In this paper, an index of carcinogenic potency is defined, and applied to published data on acute non-lymphocytic leukemia following therapy with cytotoxic drugs used as single agents. Carcinogenic potency estimates for rats and mice are also obtained for 15 antineoplastic drugs, and the potency correlation between humans and rodents is examined for the five agents for which there are data in common. The broader implications for quantitative cancer risk prediction are discussed.

The evidence that the latent period of Burkitt's lymphoma may be short, that is, months rather than years, stems from the observation of seasonal variation in disease onset and time-space clustering. The evidence for both of these effects is equivocal, but both are quite strong in the West Nile district of Uganda, where the disease has been most intensively studied epidemiologically. We have assumed that the interval between some final triggering event and onset of disease follows a log-normal distribution, and we have examined the range of values of the mean and variance of that distribution which are consistent with the observed seasonal variation in incidence. These analyses would suggest that to explain the seasonal variation in the West Nile the latent period must be less than one year on average, and rarely exceed two years.

The results of a cervical cancer screening program in a population of 22,106 women aged over 25 years in a rural county of China are evaluated in a case-control analysis. The relative risk of cervix cancer was 0.33 in women with three or more negative tests (compared with one or less) and 11.4 for women whose last negative smear was 8 or more years previously (compared with 2 years or less). Attendance at the screening program was positively associated with some risk factors for cervix cancer (sexual behavior of the woman and her partner) and negatively associated with others (genital hygiene). Controlling for the confounding effect of these risk factors resulted in only small changes in the estimate of the protective effect of screening.

Eleven population-based cancer registries tabulated second cancers among 133,411 patients diagnosed with testicular cancer, ovarian cancer or Hodgkin's disease between 1945 and 1984. Overall, 3,157 second cancers were observed, as compared with 2,420 expected at least one year after the first cancer. Survivors of testicular and ovarian cancer experienced 30% and 20% more cancers respectively than the general population comparison group, and patients previously diagnosed with Hodgkin's disease had an 80% excess of cancer. No information was available either on treatment for the first cancer, or other risk factors. However, temporal patterns in the risk of specific second cancers were analysed, with particular reference to the possible role of therapy for the first cancer. Leukaemia of the acute or non-lymphatic type, which has been previously linked to alkylating agent therapy, occurred in excess following all 3 first cancers, as did non-Hodgkin's lymphoma (overall relative risks of 6.1 and 1.8 respectively, with considerably higher relative risks following Hodgkin's disease). Other cancers for which important and plausibly therapy-induced excesses occurred were lung cancer following Hodgkin's disease (relative risk 1.9), breast cancer following Hodgkin's disease (relative risk 1.4) and bladder cancer following ovarian cancer and Hodgkin's disease (relative risks 1.7 and 2.2 in women, respectively). Rarer sites at which striking excesses occurred were the salivary gland, thyroid, bone and connective tissue. There were smaller, but clear excesses for cancers of the rectum and colon following ovarian cancer and testicular cancer, skin cancer following Hodgkin's disease, and kidney cancer following ovarian cancer. Overdiagnosis, misclassification of metastases and confounding by other risk factors were all considered as explanations of observed excesses. Nonetheless, it appeared that there are clear excess risks for cancers other than acute leukaemia which must be ascribed to therapy for the first cancer, especially in view of the possible under-reporting in registry material. Case-control studies are under way to provide information on the role of specific aspects of therapy.

Tobacco smoking and alcohol drinking histories were obtained from 194 patients with hepatocellular carcinoma (HCC) and 456 hospital controls, and the results were analysed in conjunction with the results of serological determinations of hepatitis B surface antigen (HBsAg), antibody to HBsAg (anti-HBs) and antibody to hepatitis B core antigen (anti-HBc) in all subjects, as well as the presence or absence of cirrhosis in HCC patients. The relative risk (RR) of HCC (and 95% confidence interval) among HBsAg-positive subjects was 13.7 (8.0-23.5), whereas the excess risk among antibody-positive subjects was small and statistically non-significant. In the presence of cirrhosis the RR for HBsAg-positive subjects was considerably higher (30.7 vs. 7.1 among HBsAg-positive subjects without cirrhosis) indicating that HBV may affect the development of HCC through at least two different and potentially multiplicative mechanisms (DNA integration and liver regeneration). Moderate ethanol consumption does not affect the risk of HCC, but there is a statistically significant and dose-dependent association between tobacco smoking and HBsAg-negative HCC. In most of the developed countries of Europe and North America, where the prevalence of HBsAg carrier state is very low and tobacco smoking very common, more cases of HCC may be due to tobacco smoking than to HBV, even though the RR for HCC is much higher among HBsAg carriers than among tobacco smokers.

The results of the first 3 years of cancer registration on the Caspian Littoral are described. The main finding, confirming previous reports, is a very large variation within the region of the incidence of oesophageal cancer. Possible sources of bias are considered and shown to contribute little to the pattern of incidence. Among women there is a thirty-fold variation in the incidence across the regions; among men a ten-fold variation. In the north-east of the region the tumour is at least as common in women as in men, and is more common than almost any tumour anywhere in the world. Among other tumours, stomach cancer has a strikingly uniform incidence by comparison; breast cancer shows an incidence gradient of opposite slope.

Temperatures at the lower end of the oesophagus were taken during the ingestion of hot coffee at varying temperatures and varying volume, by using a modified duodenal tube. TEMPERATURES HIGHER THAN EXPECTED WERE FOUND: of 209 measurements on 17 patients and one volunteer, standardized for size of sip and temperature of coffee, 43% were 44 degrees C or higher (up to 53 degrees C). Statistical analysis showed a linear relationship between the intraoesophageal and the initial coffee temperatures.Volume has a larger effect than the temperature at which the coffee is taken.The average increase in intraoesophageal temperatures at the point of measurement was 0.381 degrees C per ml coffee ingested. Some implications of the findings are discussed from the epidemiological point of view.