No abstract

SUMMARY Results are given of a case-control study on bladder cancer in West Yorkshire. The aim of the study was to assess what risks exist for cigarette smoking in the United Kingdom and also to investigate if a dose response effect was present. The study shows for the first time in the United Kingdom a positive but weak relationship between cigarette smoking and bladder cancer. Close examination of the data with regard to this effect shows that a complex set of relationships result when the quantity smoked and the period of smoking are taken into account. A dose response effect is weakly demonstrated when a medium quantity of cigarettes are smoked but it is not present at all in those who smoked most cigarettes for the longest periods. No risk appears to exist for those who have smoked only filter cigarettes. The results are contrasted with similar studies and the significance of the observations are discussed.An association between cigarette consumption and bladder cancer is known as a result of various types of epidemiological inquiry. Correlation studies by Lea,' Fraumeni,2 and Stevens and Moolgavkar,3 several follow up and birth cohort studies," and some case control studies7-13 all indicate that risks exist. There are, however, some deficiencies in our knowledge, in particular no other English case-control study has yet shown a risk.14 Part of the work presented here has already been published in a preliminary form15 but it is the purpose of this study to Tables 1 and 2 show the number of cases and controls in the analyses and their place of interview. Some of the later present fewer results than shown here due to lack of certain information. Table 3 gives the overall result of "ever" versus "never" cigarette smokers from the entire series. These results are similar to those reported in a preliminary study15 for men Table 4 indicates that for men the risks from cigarette smoking decline after abstinence of at least five years; the resulting risk for current smokers Table 5 shows that the risk of the current smokers varies slightly with age at diagnosis but not to any significant degree.

Data are presented on a group of cases of primary carcinoma of the bladder detailing their ABO, MNSs and rhesus blood antigenic phenotypes. Differences, either between all cases and controls or between particular stages of presentation and controls, have been found in the ABO red cell system as well as the MNSs red cell system. In the case of the ABO system an excess of the A gene is present but detailed study shows this to be only in patients with T1 and T2 tumours. No excess was seen in patients with carcinoma in situ or T3 or T4 stage disease. The S gene of the MNSs system is also in excess in cases of bladder cancer.

A case-control study in West Yorkshire has been used to investigate possible risks of bladder cancer amongst those fishermen who used azo-based dyestuffs to stain maggots or who purchased ready coloured maggots. No risks have been found, although the confidence limits of the estimates are wide. These data refer to exposures over ten years ago and it is possible recent changes in fishing practice, if they have occurred, are not yet assessable due to the long mean latency seen in bladder cancer when exposure to carcinogens is not great. This paper does not suggest the general use of these substances is without hazard, merely that no risk is associated with the study group and that the chemicals under study are not a major cause of bladder cancer.

SUMMARY Data are presented on a group of cases of primary carcinoma of the bladder, detailing red cell surface blood group antigenic phenotypes, serum haptoglobin phenotypes, and some red cell isoenzyme phenotypes. Account is taken of the stage of the disease at presentation. The results are compared with corresponding phenotype frequencies in groups of presumed healthy persons originating either in Yorkshire or County Durham. Differences in relative incidences were found in the haptoglobin, phosphoglucomutase (PGM), and some other systems. These are both differences between all cases and controls and between particular stages at presentation and controls.