This section is from the book "Symposium Phenomena Of The Tumor Viruses", by U.S. Dept. of Health. Also available from Amazon: Tumor Suppressing Viruses, Genes, and Drugs: Innovative Cancer Therapy Approaches.
Investigations with the Bittner factor (the mammary-tumor agent or the milk factor) have aroused considerable interest ever since its discovery by Bittner (1). This agent, which behaves in many respects like a virus, is related to the development of mammary cancer, an epithelial tumor, in mice. The characteristic host response to this agent is therefore the development of mammary cancer at a late age in these animals.
This review considers the natural factors influencing host response to the mammary-tumor agent. The various aspects involved will be illustrated only with the investigations performed in the Netherlands Cancer Institute, Amsterdam.
Table 1 records the results of an experiment to demonstrate the presence of this agent. Susceptible female mice given the agent by foster-nursing have a high incidence (96%) of mammary tumors at an average age of 438 days. The controls have a low incidence of 3 percent at a considerable later age (700 days). The demonstration of the effect of the agent is as yet only possible by statistical means and after a long observation period- severe handicaps for the work with this agent.
Mammary tumor (percent)
Average tumor age (days)
Average age at death without tumor (days)
With agent, foster-nursed by 9 DBA
This agent is instrumental only in the development of mammary cancer and does not influence the development of other tumors, as the results in table 2 show. Male mice which do not develop mammary cancer were given a potent extract containing the agent at weanling age. The incidences of lung tumors and of hepatomas were recorded. Another group of the same mice were given an extract of mammary tumors without the agent. No significant differences in the tumor percentages were found, and the lifespan of these animals was not influenced by the presence of t.hp. furont,.
Lung tumor (percent)
Average age at death (days)
With agent, given as extract 91 from C3H tumors
Without agent + extract 90 from C3Hf tumors
There are no obvious signs of the agent in the early life of the animals. Since the mammary-tumor agent is related to the development of mammary cancer, the first visible effects of the agent can be expected in the mammary gland. Hyperplastic nodules are generally considered to represent a precancerous condition. Hyperplastic nodules and carcinomas are also observed in agent-free mice, but the frequency of nodules and carcinomas is markedly increased in the presence of the agent. The influence of the mammary-tumor agent may be twofold: (1) The agent causes the development of more nodules, and the risk of development of carcinomas is proportionally increased. (2) The agent is associated with the development of more nodules, and these nodules show a more marked tendency to develop into carcinomas than they would in the absence of the agent.
From the experiments (2-4), in which the number of nodules was compared with the frequency of carcinomas in different strains of mice, with and without the agent, the following conclusion can be drawn (table 3): In the C3H and DBA strain, the nodules are numerous and the number of nodules is proportional to the mammary-cancer incidence. But in the O20 strain, the development of malignancy from these nodules is relatively more frequent. The reverse is true in the C57BL strain. Thus the odds that precancerous hyperplastic nodules in the mammary gland become malignant are different in various strains of mice, and depend primarily on genetically determined factors, which are independent of the presence of the mammary-tumor agent.
If this assumption is accepted, then the primary host response to the mammary-tumor agent would be the development of more hyperplastic nodules in the mammary gland at an early age.
There is no doubt that this response to the mammary-tumor agent is governed by genetic factors too. The number of the hyperplastic nodules and the age at which they can be first observed are quite different in various inbred strains of mice. This brings up the question whether the mammary-tumor agent is the same in the various strains of mice; this is probably not the case (6, 6). The agent of 5 different strains (C3H, DBA, ROT, WLL, and A) was administered to the same susceptible hybrids (9 O20 X & DBAf)Fi, either by the injection of extracts of mammary tumors or by foster-nursing (table 4). The agents of the first 4 strains gave a high incidence of mammary tumors, whereas the effect of the agent of strain A was quite different. Only a few mammary tumors developed in the test animals, though there can be no doubt that the A strain contains a mammary-tumor agent.
+ + + +
+ + + +
+ + +
+ + +
If the agent of the A strain is administered to different strains free of their own agent (C3Hf, DBAf, WLLf), the surprising fact is that it is effective only in the C3Hf strain (6).
Sources of agent
Number of animals
Number with tumors
Average tumor age (days)
It may be concluded that the mammary-tumor agent from various sources differs in effect, either by a difference in quantity or, more probably, in quality.
Another experiment (6) showed that the agent of the A strain can enhance the effect of that of the C3H strain. Hybrids F1 were treated in the following manner: (1) One group was treated with the agent of the C3H strain at the age of 4 weeks; 50 percent developed mammary carcinoma. (2) Another group was treated with the C3H extract at 3 months of age; only 35 percent had mammary carcinoma. This confirms the long-known phenomenon that administration of the agent at a late age is less effective. (3) This group received the agent of the A strain at the age of 4 weeks; the mammary-tumor incidence was the same as in the control group. (4) Another group received the agent of the A strain when 4 weeks old, and the agent of the C3H strain 3 months later. This group showed the highest incidence of mammary cancer (88%). Thus there is an additive effect of the 2 agents.
Another factor that is genetically determined and that can influence the host response is the susceptibility to the mammary-tumor agent. Four different strains, which do not harbor the agent, were foster-nursed by the same strain (DBA). The virgin females of the IF strain had a mammary-tumor incidence of 100 percent; of the O20 strain, 30 percent; the CBA strain and the C57BL strain showed no mammary tumors.
As Gardner (7) showed many years ago, there is no doubt that the development of hyperplastic nodules is dependent on the presence of hormones. These are the same hormones that govern the growth of the mammary gland: hormones of the ovary (estrogen and progesterone) and the mammotropic hormone, prolactin, of the anterior lobe of the pituitary gland.
The experimental evidence indicates that the amount of hormones acting is important. Generally it may be stated that a higher dose enhances the development of the nodules more effectively. It has long been known that pregnancy, too, shows the same effect. Pregnancy is a complex phenomenon, but there are reasons to assume that the pregnancy effect is due to a stronger hormonal influence on the mammary gland.