The modifying influence of diet and the physical environment on spontaneous tumour frequency in rats

The results are reported of an investigation into the influence of diet and geography on the kinds of neoplasms and their frequency in three strains of albino rats. A group of 586 GG strain rats reared in Johannesburg showed a high frequency of spontaneous neoplasms, 74 per cent of all males and 50 per cent of all females being tumour-bearers. Eighty per cent of all neoplasms were found in the endocrine glands, phaeochromocytoma being the commonest tumour in both sexes and interstitial cell tumours of the testis and fibroadenoma of the breast ranking second in order of frequency in males and females respectively. By radical modification in the diet, without interference with life expectation, it was possible to achieve profound changes in tumour frequency rates including, amongst others, a reduction in phaeochromocytoma and an increase in the carcinoma of the uterus in GG strain rats. A high protein, carbohydrate-free diet reduced the overall frequency of tumours as well as of tumour-bearing rats. The modifications in the frequency of endocrine tumours suggested that the diets had created at least a favourable endocrine context for retarding or accelerating an underlying cancerization process which may have been determined in the first instance by factors other than diet. A comparison of the frequency of spontaneous neoplasms in two migrant strains of albino rats, namely, Utrecht and Copenhagen, with that of the GG strain, when maintained under similar conditions of diet and environment, disclosed significant differences in respect of some neoplasms. Some of these differences were preserved in the first and subsequent generations of migrant strains whereas others were submerged with the result that the tumour frequency converged on a pattern similar to that of the GG strain. It was suggested that until the significant environmental factors, depending on diet and possibly also on the physical attributes of the environment, such as altitude, climate etc. were identified, it would not be possible to assess the extent to which so-called strain differences in tumour frequency could be attributed to one or more specific factors in the environment or to genetic factors. © 1958, The British Empire Cancer Campaign for Research. All rights reserved.