Endogenous and exogenous factors in carcinogenesis: limits to cancer prevention.

Organ-specific cancer incidence rates can vary dramatically between low- and high-incidence areas. Such differences are due to (1) heritable susceptibility determinants, (2) risk factors associated with the environmental and local living conditions (e.g., viruses, pollution), and (3) personal life-style factors. For organs showing large differences between cancer registries, exogenous factors might be most important, while for organs showing only small differences, endogenous and unavoidable factors are expected to be more important. In this paper, a working hypothesis based on descriptive cancer epidemiology is presented to estimate, in a quantitative manner, the unavoidable contribution to the process of carcinogenesis and to discuss limitations to individual cancer prevention. Cumulative cancer incidence rates for a 75-year period of life (CR74, in percent) were taken from IARC Scientific Publication No. 120 (1992). For each organ, values were ranked in ascending order, and the ratio between high-rate and low-rate registries (90th percentile/10th percentile) was determined. This measure of variability among registries differed strongly between organs. Largest ratios were seen for organs with well-known exogenous risk factors, such as pharynx, lip, tongue, mouth, liver, esophagus, and melanoma in males, and lung, esophagus, gallbladder, liver, and bladder in females. Small ratios were seen for rectum, brain, colon, and Hodgkin's disease in males, and breast, rectum, ovary, brain, and colon in females. It is concluded that the process of carcinogenesis in the latter organs has a stronger endogenous/unavoidable component, for some tissues possibly of hormonal type. A fictitious population was composed where, for each organ, the minimum reported cancer rate was taken. When based on all cancer registries world-wide, CR74 sums over all sites of 2.0% and 2.3% resulted in males and females, respectively. When only Central/Western European countries were included in the analysis in order to reduce differences in risk factors nos. 1 and 2, the sum of the minimum values was 10.4% and 8.7%. After correction of the data for smoking, 'minimum' cancer incidence rates in males and females were estimated to be 7.6% and 6.8%. Based on a median cancer incidence rate for nonsmoking males in Europe of about 21%, therefore, individual preventive measures taken by a nonsmoker can reduce the cancer risk, on average, 'only' by a factor of about 3. A considerable fraction of cases thus appears to be hardly avoidable.