The role of psychological factors in the pathogenesis of cancer, although still obscure, has been of increasing interest in recent years. Various theoretical considerations (i.e. the relationship between the known effect of emotions upon hormonal and biochemical balances and the possible effect, in turn, of the latter upon neoplasms), as well as a number of clinical reports and experimental and statistical studies, point toward psychological influence in cancer pathogenesis, but little is actually known in this area.

In order to explore this matter further, a research program has been carried on in our Institute since 1952 by Dr. L. LeShan. This study has included the evaluation of projective personality tests given to over 300 cancer patients; interviews of 2 to 8 hours each with over 150 patients; and extensive exploratory psychotherapy (of from 60 to 400 hours) for 25 patients. Control groups were also included in each category. For patients undergoing psychotherapy, regular comparisons were made between the personality picture and various biochemical activities reflected in blood and urine analyses.

A "back and forth" method between the three techniques of personality evaluation has been employed. Hypotheses formulated from data obtained with one technique have been evaluated, refined and clarified by data from the others. When a hypothesis was consistently supported by all three approaches, an attempt was made to formulate it in terms permitting it to be subjected to critical test by experimental or statistical technique.

As an example, an hypothesis was developed that the cancer patient, more often than chance would allow, had lost a major emotional relationship, and had been unable to find a satisfactory substitute, some time before the first apparent symptom of cancer. This hypothesis appeared to be validated by data from all three techniques. It was then formulated in terms by which it could be tested. If the hypothesis were true, then certain social groups which, a priori, had known higher rates of such losses should also have a higher cancer mortality rate. Thus, for example, if marital status were taken as the only variable, then, after age was cancelled out, we should expect the highest cancer mortality rate in the "widowed," the next in the "divorced," the next in the "married" and the next in the "single." Published data, such as census material, could be used to explore the accuracy of this prediction. Various predictions of this type—all based on the hypothesis—were made. When tested against published statistical data, all were demonstrated to be valid. (116, 117)

At this point of the research, one general, emotional pattern has been found in over 50% of the 300 studied cancer patients and in approximately 10% of the equated controls: An early life history with much self doubting and some anxiety over relating to others; the establishment of one personal relationship that afforded a high degree of satisfaction, meaning and validity to the individual and provided him with a "raison d'etre"; and the loss of this relationship, followed by inability to find a substitute, and a period of intense (if often concealed) depression. This has been elaborated upon, and case histories presented in various publications. (118, 119, 120, 121, 122)

In summing up this research in a recent paper (123), the following conclusions could be reached by LeShan:

1. There seems to be a correlation between the existence of neoplastic disease and the persistence of certain types of psychological situations.

2. The most consistently reported, relevant psychological state has been the loss of a major emotional relationship. Often the psychic state resulting from this loss could be traced to a period shortly before the first noted symptoms of cancer.

3. There appears to be some relationship between personality organization and the evolution of the cancerous condition.

4. There may be some relationship between personality organization and the type or location of a cancer.

It would seem as if future research in this area, to be as useful as possible, must focus upon the chemicophysiological changes which result from variations in psychic states in general. It is highly probable that these changes are mediated through the endocrine system. Through the linking of psychic states with hormonal changes, we may be able not only to integrate psychological factors with the many other factors influencing development of cancer, but also to relate them to certain levels of organization. It may be possible to establish the relationship of psychological factors to other influences which favor or even induce passage of cancer from one phase to another. This may permit a more complete understanding of cancer and help in finding new points at which some therapeutic value might be expected from psychological intervention.

The relationship between the adrenals and psychic states on the one hand, and between the adrenals and the lymphatic system on the other hand, could explain why the influence of a psychologically unresolved problem is most evident, among all cancerous conditions, in lymphomas. LeShan has been able, by analyzing a sizable number of these lymphoma cases, to recognize more clearly and more consistently than in other conditions the existence of a pattern of psychological changes occurring prior to clinical illness. Recurrences of symptoms, or periods of exacerbation, also could be connected to events which had deep repercussions upon the psychological state.

Research along these lines, seeking information on psychic factors in the pathogenic mechanism of cancer, is being pursued actively by LeShan and his co workers at our Institute.