Cancer And Hormones
( Originally Published Early 1960's )
Internal regulators of growth.
One of the most interesting and best studied areas of biological research is endocrinology. It concerns the glands of internal secretion, which produce chemicals that are important in the growth and regulation of many functions of the body. The anterior pituitary gland, from its position at the base of the brain, is the conductor of this orchestra; perhaps a better or an alternative metaphor is one of a thermostat. At any rate, an intimate relationship exists between the pituitary and the thyroid, the adrenals, and the ovary and testis: when the action of the latter glands is inadequate, the pituitary puts out increased amounts of stimulating protein substances known as the trophic hormones.
Abnormally deficient or excessive hormone production leads to some of the more bizarre clinical manifestations, and these have attracted medical interest since antiquity. During the 1930's, the first pure hormones were isolated and chemically characterized, and there became available an extensive number of chemicals with remarkable physiological activity. The female sex hormones, estrogens and progesterones; the male sex hormones, androgens; the adrenal corticoids are the chief examples of steroid chemicals. Thyroxin and parathormone, pitressin and the anterior pituitary hormones are examples of nonsteroid chemicals.
The relationship of hormones to cancer was one of the earliest areas of interest in cancer research. The late Leo Loeb of St. Louis, Mo., demonstrated by 1918 that breast cancers could be prevented in female mice if the ovaries were removed. Furthermore, breast cancers could be induced in male mice, which do not develop breast cancer, if they were castrated and had ovaries implanted under the skin. With the advent of pure estrogenic compounds, A. Lacassagne of France in 1936 reported that these also would evoke breast cancers in males of susceptible strains. It was shown subsequently that breast cancer in mice required the presence of three complexes: genetic susceptibility, the hormonal status, and the milk transmitted Bittner virus.
Large doses of estrogen, given over protracted periods, will also lead to the development of testicular tumors of the interstitial cell type, cancers of the uterus, pituitary tumors and leukemia in specific strains of mice. William Gardner of Yale attributed these effects to the production of a hormonal imbalance, which could be achieved in several procedures other than exogenously administered hormones. The appearance of cancers and other tumors of the adrenal cortex was reported following castration, an effect that could be prevented by the administration of estrogens. Ovarian tumors were produced in rats following the transplantation of the ovaries to the spleen. In this location, the ovarian hormones were secreted into the blood supply to the liver, where they are inactivated. The pituitary was thus stimulated to produce increased amounts of gonadotrophic hormones that forced the ovaries in the spleen into increased activity. The tumor reaction also could be prevented by injections of estrogen, or by leaving one of the ovaries in its normal position.
Mere increased activity, however, is not a carcinogenic precursor, and other factors are undoubtedly operative. The lactating breast, for example, represents the most active hyperplastic phase of the tissue, yet is not especially susceptible to becoming cancerous. Cancers appear to arise more readily during regressive phases when small islands of active tissue remain in an otherwise somewhat atrophic organ. On the other hand, mice that develop testicular tumors manifest an increased resistance to estrogens that allows them to protect the sperm producing elements of the testis much better than mice of strains that respond to estrogens by a severe early atrophy of the testis. It is probable that in many of the cancers evoked by estrogens, the reaction is mediated through the pituitary. These inter-relationships may involve not only other endrocrines, but detoxification of steroids by the liver, folic acid metabolism and nutritional factors. The complex situation is well illustrated by Howard Andervont's findings at the National Cancer Institute, that the incidence and age of occurrence of breast cancer in mice are affected by the number of animals that are kept in one cage. The interpretation here is that a single mouse in a cage has an unlimited access to the food supply, but is deprived of the warmth achieved by communal nesting. These factors affect the estrous cycles of the animals, which are related to hormone levels and, thus, to the development of breast cancer.
We have already noted that in some strains of mice, the occurrence of leukemia can be increased by injections of estrogens. This reaction probably involves the adrenal cortical hormones, which affect the lymphoid tissues. The hormonal status of the animal also affects other neoplastic reactions. The production of liver cancer with azo dyes in the rat is markedly inhibited by the removal of the pituitary gland. It has been shown recently that hypophysectomy also prevents the development of "spontaneous" liver tumors in mice. Hypophysectomy also leads to the inhibition of the formation of subcutaneous sarcomas in mice receiving polycyclic hydrocarbons.
The endrocrine system must be particularly reactive in laboratory rodents as compared with man, because many of the striking effects just noted are hard to demonstrate in man. The great use of estrogens and other hormones since their introduction into clinical practice almost 30 years ago has not led to an increase in the incidence of breast cancer in women or in men. A careful study of the rare breast cancer in males failed to show a greater antecedent history of the use of estrogens and other hormones than matched control patients. Earlier reports on the occurrence of breast cancer in men treated with estrogens for prostatic cancer need reevaluation, because the cancers may have been metastases from the prostate rather than being primary in the breast.
The role of the pituitary gland in the origin of human cancer is also not clear. The risk of developing cancer among patients with acromegaly, one type of hyperpituitarism, and among pituitary dwarfs, one type of hypopituitarism, was the same as that to be expected in the general population of a comparable age. An interesting recent report on the effect of thyroid activity and cancer occurrence suggests that cancer is less frequent among patients with hyperthyroidism and more frequent among patients with hypothyroidism than among people with normal thyroid function.
We have touched upon some of the interrelationships between the development of cancer and hormones. There are also some interesting effects of hormones on established tumors and their growth. It has been known since 1896 that some premenopausal women with advanced breast cancer had temporary regressions of the cancer following the removal of the ovaries. In prostatic cancer castration also often leads to the regression of the tumor. Here we have two cancers in man that even at the stage when the cancer manifests its ability to metastasize are still in part "hormone dependent."
Hormone dependency is also observed in some experimental tumors in animals, and studies on these tumors have established that there are gradients of malignancy among tumors. Some tumors, such as the breast adenomas of rats, or the interstitial cell tumors of the testes in mice, can grow only if the animal continues to receive estrogen. Some thyroid tumors in rats receiving thiouracil also continue to grow only if the normal thyroid function is kept suppressed.
When mice of a susceptible strain are implanted with estrogen pellets, microscopic tumors of the testis are seen 8 to 12 weeks later; if the pellets are removed at this time, practically no mice develop gross tumors. But if the pellets are reimplanted several months later, the tumors promptly start growing. Many of the large tumors no longer need estrogen for further growth, showing a loss of hormone dependence.
The hormone dependence of some human tumors can be exploited clinically in several ways. Castration for prostatic cancer and breast cancer can be substituted or complemented by the use of large doses of appropriate sex hormones. Alexander Haddow of London showed that diethylstilbestrol and other estrogens also have regressive effects in advanced breast cancer in post-menopausal women. The paradox of presumably opposing hormones having similar end effects is probably resolved by the fact that both hormones in large doses suppress the anterior pituitary gland, and this is the possible mode of action. Support for this is derived from the direct attack on the pituitary, and its removal also produces beneficial effects in many patients with advanced breast cancer.