Cancer, A Diversion Into The Meaning Of The Cause
( Originally Published Early 1960's )
In the last three chapters we have concerned ourselves with three major classes of external stimuli that have been defined as carcinogenic, or being able to evoke a reaction that eventuates in cancer. Are such relationships causal? This question requires the examination of what we mean by cause.
Reference to an unabridged dictionary reveals how difficult it is to define "cause" and how many diverse meanings have been given to the word since the days of Aristotle.
The human mind is first confronted with the effect and not with the cause. It is only after the effect is defined that the reasons for (or, more correctly, events preceding) the effect are sought. The relationship between the two is then clarified through an analysis of the intermediate steps, usually referred to as the "mechanism of action" or, for disease, "pathogenesis."
There are few, if any, simple causes in biology. There are, instead, complex situations and environments in which the probability of certain events is increased. The appearance of cancer following exposure to a chemical agent or a surgical procedure merely means that the incidence of this type of cancer has been increased over the incidence that would have occurred without the interference. The difference between an "induced" or a "spontaneous" tumor is that with the latter no conscious procedure is undertaken, so that it is of unknown etiology. In neither situation is knowledge implied of the "direct" or "indirect" mechanisms that are involved.
Nevertheless, in specific medical situations we have evolved sets of facts, or evidence, which have to be fulfilled before by common acceptance we conclude that a cause effect relationship has been demonstrated. An excellent example is available in bacterial diseases. Robert Koch, the great German bacteriologist of the last century, described these steps so well that they became known as Koch's postulates. The postulates state that in order to show that a specific microorganism causes a specific disease, we have to find and to recover the microorganism on an artificial medium as a pure culture; we must reproduce the disease regularly in healthy individuals who are given the pure culture of the microorganism; and we must again isolate the same microorganism from the individuals who develop the disease under experimental conditions.
These rules of evidence have been most useful in bacterial diseases, but they are not entirely applicable to diseases evoked by nonreproducing agents such as chemicals. With chemicals, the practical acceptance of a causative role is usually based on the regular observation of a specific reaction following exposure of individuals to the chemical, the study of the chemical and the reaction it produces under controlled laboratory conditions, and the ability to prevent the occurrence of the reaction by withdrawing the chemical or protecting the individual from exposure to it.
If this limited, practical definition of cause is made, we know the causes of many specific cancers. For example, the exposure of dye workers to beta-naphthylamine will lead to the appearance of cancer of the urinary bladder many years later. We can prevent the occurrence of these cancers by eliminating betanaphthylamine and related chemicals from the environment of dye workers. We can also reproduce the sequence of events in dogs and in rabbits. We can thus accept that beta-naphthylamine causes bladder cancer. This statement does not imply, however, that we know the "mechanism," or the steps between the initial stimulus of the chemical and the eventual reaction of bladder cancer. This statement also does not imply that all cancers of the urinary bladder are due to exposure to betanaphthylamine or related chemicals.
These exceptions or limitations apply just as well to bacterial diseases. In pneumococcus pneumonia, for example, the reaction of the lung to the microorganism that eventuates in a disease called pneumonia is almost as incompletely understood as the transformation of normal cells to cancer cells. And, as in the case of our bladder cancer example, not all pneumonias are caused by pneumococci. Very similar disease processes can be evoked by a wide variety of bacterial and fungal agents, and by chemicals as well.
The reaction of an organism to a stimulus is the property of the organism and not of the stimulus. Obviously, a chemical composed of carbon, hydrogen and oxygen in a polycyclic structure becomes a "carcinogenic hydrocarbon" because it evokes a cancer response in some organism.
In order to produce a cancer response, the chemical, or the virus, or the quantum of physical energy, must gain access into the reactive site of the organism. This site of action in cancer must be within body cells. The agent thus must gain access to the cell, and produce a permanent change in the cell that is transmitted to the progeny of the cell. Furthermore, in order that the cancer be recognized, the cellular change must progress through division and growth until a large population of cells is formed.
No tissue is composed of one type of cell, and cells of the same type vary in age and in their relations to each other and to the environment, as well as in such presumably basic characteristics as the number of chromosomes. A completely uniform response of cells or tissues is, therefore, a biological impossibility. Despite the wide variations in individual responses, however, specific neoplastic reactions of organisms of specific strains and species are surprisingly reproducible. This indicates that the reactions are mediated through rather specific biologic, biochemical mosaics.
The probability of a neoplastic event following exposure of an organism to a carcinogenic stimulus is modified by a large series of factors. In regard to the stimulus, among the more evident influences are those of dose, route of exposure, physical state of the material, and length and schedule of exposure. In regard to the host, the probability of the neoplastic reaction is influenced by the genetic background, age, sex, nutritional status, and intercurrent infections. The longer the period between initial exposure and the end point of the reaction, the more opportunities occur for the introduction of additional modifying factors. Some of these factors may appear quite subtle. Sociologic and psychologic influences upon physiological processes are realities, and some influences of these factors upon the occurrence of cancers should occasion no surprise.
The long chain of events that occur in the cancer reaction and factors that may influence the reaction also suggest the many points at which the reaction can be entered or interrupted for possible preventive or therapeutic effects.