Discussion of the Papers

WILLIAM A. GREENE (University of Rochester Medical Center, Rochester, N . Y . ) : About a year ago I received a telephone call late one night from a man with whom I grew up, who has become an eminent pathologist and who continues as a long-time friend. After we had briefly reconnoitered, he let me know the reason for his call. Approximately six weeks before, his elderly uncle, who had been disabled with recurring cerebral thromboses for the previous year, had died. My friend’s elderly aunt, who had been married to this man for some 55 years, had been stunned and grief-stricken following the death of her husband, and three days later developed fulminating symptoms and was diagnosed as acute leukemia. She announced, at the time, that she was going to die on a certain date, two days hence. It evolved that she was noit quite accurate and that she died two days before she predicted, the morning of the day that my friend called. After my appropriate condolences and references to the long life together of the elderly couple, my friend inquired, “Didn’t you write something about this stuff?” I indicated that I had written something about this stuff and that this was not an uncommon sequence of experiences for the patient who developed manifest leukemia. He then said, “What I want to know is, how does it work?’ I indicated that I didn’t know how it worked. There were a number of possibilities. I returned the challenge indicating that he was a good pathologist and perhaps he could find out. This is one of the ways of dealing with the “How does it work” question which arises when one entertains or is struck by an association between psychosocial factors and the manifest development of leukemia or lymphoma. This is a question, an answer to which appears to be necessary for us to consider seriously any relevant relationship between psychological parameters and neoplasia. Complementing this reluotance to consider seriously any relationship between psychological parameters and neoplasia without an explanation is the fact that psychological factors as they may pertain to cancer have come under systematic study only in the recent past. This is in part because prevailing psychosomatic hypotheses have not fit as readily as in a number of other disorders with any conceivable relation of psychologic process and a pathophysiologic manifestation such as cancer. This was significantly emphasized by George Engel’ in his presidential address to the American Psychosomatic Society back in 1954. His address on the “Seleotion of Clinical Material in Psychosomatic Medicine” was the prelude to a panel discussion on neoplastic disease and psychological processes which, I believe, was the first formal public discussion of these matters. Most of the members of that panel are participating in this conference. I would like to make three points in my discussion today. I do so with the working assumption that there is a significant relationship between the manifest development of neoplasia, as represented by the leukemias and lymphomas, at least, and the psychological reaction of the individual to various life events. First, I think there is no evidence for any specific mechanism for understanding the relationship between psychological parameters and neoplasia. Second, I think that the approach to the question of possible mechanisms should be done with more versatility and with the assumption that there are likely a variety of possible mechanisms. Just as the assumption that there is a single cause of cancer is attractive, so is the assumption that there is a single mechanism to explain any relationship between psychological reactions and neoplasia. It seems to me highly unlikely that there is a common psychophysiologic mechanism to account for the variety of pathological manifestations subsumed under the very broad term cancer, for instance leukemia, carcinoma of the breast or brain tumor, particularly in the age