This view puts the problem of diagnosis of cancer in a new light. The recognition of a cancerous condition by itself, although important, has little clinical meaning. The presence of "cells with cancerous characteristics" in the prostate of almost all men 40 years of age and older, and in the thyroid, lung and stomach in a high proportion of the population, has failed to produce a general feeling of despair only because such findings are commonplace. While they still mean cancer, they do not implicitly indicate malignant disease. It appears very clear that a diagnosis of cancer is incomplete without immediate qualification as to its phase. We can no longer speak of cancer with any degree of practical meaning unless we add a descriptive adjective—noninvasive, invasive, tissular, organic or systemic.

And the search for a test to detect cancer will have no meaning as long as we have not defined in advance the information we want. A test, biochemical or immunological, which indicates the existence even of a specific anomaly in the noninvasive phase or before, while interesting, will have little significance since this anomaly exists in so many subjects and for most does not go beyond the noninvasive phase. The test will not indicate malignant cancer in the clinically frightening sense. On the other hand, the processes which are added to a noninvasive phase form of cancer and turn it into the invasive, tissue, organic or systemic phase, have no character of specificity. Similar growth changes, or the appearance of lipidic predominance, which represent added factors are seen in many other conditions. By using them for diagnostic purposes we will not recognize the cancerous condition but only nonspecific intervening factors. While these factors are responsible for the changes, malignancy develops only when, and because, these factors operate on already abnormal entities, i.e., cancerous entities. This explains the nonspecificity of many proposed tests and the misleading positive results obtained in conditions such as pregnancy where one of these added factors, (active growth processes) is always present.

A test for cancer, to have clinical value, would have to indicate two things: one, the specific early change which is widely distributed but represents the essential condition for the potential development of malignancy; and, two, the concomitant presence and concomitant operation of the nonspecific factors which can cause the actual development of malignancy. This kind of diagnostic test undoubtedly will come from further systematized study of biochemical changes induced by the simultaneous action of the two groups of factors.

Immunological studies represent an approach of value for diagnosis. The different phases of cancer can be interpreted, in the final analysis, to correspond largely to the intervention of the defense mechanism at different stages at the different levels. As mentioned above, a change in a phase results also from a change in the defense stage at the respective level. We have seen that the immunological aspect of cancer cannot be understood without accepting a relative independence of the levels in their different stages of defense. This view explains some seemingly paradoxical occurrences.

Cancerous cells are frequently found circulating in the blood yet this does not indicate generalized cancer. While the organism defends itself successfully at the systemic level against cancerous cells, the cancer can still progress at the lower level of the tissues. The loss at this low level of an effective defense, principally primary or allergic, which is still persistent at the systemic level, explains why the cancerous cells invade the tissues. A test indicating the presence or absence of any immunological reaction would consequently have value only when related to hierarchic levels. It must furnish indications only of what is happening at a specific level. The nature of the immunological reaction in cancer is also different from the reaction in other conditions. Defense capacity—natural defense capacity— at different levels is lost as the respective level participates in the disease. This is in distinction to the immunological processes in other conditions in which the normal individual lacks specific immune bodies. An immunological test for cancer would have to reveal the loss of a previously existing defense mechanism rather than the appearance of an immunological response. This loss can be revealed in different ways. In one, the response to a cancerous antigen is investigated, and its lack would indicate the existence of a cancerous condition at this level.

In a study now in progress, we utilize pooled human tumoral tissues as antigen, and try to see if an allergic reaction can be induced with it, in two administrations, sensitizing and trigger. Two intradermic injections at the same site are made 12 days apart. They induce an allergic reaction in normal individuals, but are without effect in patients with active malignancy. If the effect is negative, a third injection is given 15 days later. A similar test is made for the conjunctiva, with sensitizing and trigger instillations of a similar antigen. No allergic reaction indicates a positive result, while a reaction is considered to be normal.

Another test which we are studying is based on the same lack of efficient defense mechanism at the tissue level. Such a lack of defense would permit a cancerous antigen to be present without the body offering a sufficiently effective defense against it. The presence of such an antigen in the tissues is revealed by inducing an allergic reaction, through the administration of specific coagulant antibodies. Sera of guinea pigs injected with pooled human tumors and having a high precipitin content are used in intradermic injections or in eye instillation. An immediate reaction indicates a positive result. As control, we use normal guinea pig sera. The studies are now in progress and the diagnostic value of these tests will be reported in a later publication.

Circulating Cancer Cells And Surgery

These immunological considerations have appeared important in considering a problem related to the use of surgery in cancer. The existence of a veritable flow of cancerous cells in the general circulation, largely induced by the manipulations inherent in operative procedures, has produced grave doubts as to the value of the measures taken by surgeons to prevent local spread of cancerous cells through the surgical act itself. Paradoxically, however, these precautions have been followed by good clinical results. Analysis from the point of view of the defense mechanism involved can clearly explain this situation. In the invasive phase, the systemic defense mechanism is still adequate to insure destruction of cancerous cells which get into the blood. This is not true at the level of the interstitial formations, that is, at the tissular level, where such defense means are failing. The real danger during surgery consequently is not so much the presence of cancer cells in the blood, since the blood can still take care of them, but the spread of these cells at the tissular level where the defense capacity has been lost, and where a cancerous cell consequenUy has every chance not only to remain alive but also to grow.

The independence of the defense mechanism at different levels also must be taken into account in explaining the differences in the events which follow the appearance of a spontaneous tumor and those which occur after experimental tumor transplantation.