WILLIAM B. COLEY, M.D. NEW YORK

*Read before the Tri-State Medical Association, Memphis, Tenn., February, 1931

 

DURING the last quarter of a century, the subject of cancer control has been receiving an ever- increasing amount of attention, and in the last decade, it has taken first place in the discussions of medical societies throughout the world. Interest on the part of the laity has likewise been rapidly increasing, due in a measure, to the widespread propaganda of the numerous societies for the control of cancer, and also to the fact that the mortality of cancer has been steadily rising, especially during the last twenty-five years. A pioneer among these societies is the American Society for the Control of Cancer, and it is gratifying to note that this society in 1927 had little difficulty in raising the million-dollar endowment which it set out to do. Among the many proofs of this growing interest in the problem of cancer may be mentioned the International Lake Mohonk Conference held in 1926, and the International Cancer Conference held in London in 1928 under the auspices of the British Empire Cancer Committee. Both of these were attended by delegates from all over the world, and interesting and valuable papers were presented by the leading specialists in the field of clinical and laboratory cancer research. I had the privilege of being present at both of these conferences.

A careful review of the reports made on these two occasions leaves one in a very pessimistic frame of mind. No new light was thrown upon the fundamental problem, i.e., the etiology of cancer; and practically nothing new was offered in the way of more efficient methods of controlling cancer or of checking its steadily rising mortality.

A most important paper at the Lake Mohonk Conference was that of Ewing on "The Prevention of Cancer." In the opinion of Ewing, prevention plays a prominent role in the control of cancer. He stated there were many reasons for the current neglect of the prevention of cancer. Among the obstacles enumerated was, "the general assumption that cancer is the result of hereditary tendencies which express themselves in the so-called spontaneous outbreak of the disease." However, he did not believe that hereditary could be dismissed from a practical consideration of the origin of cancer; and he approved of the public being advised that, when there is a strong tendency to cancer in the family, the other members take unusual precautions against the disease.

A far more serious obstacle he believed was the widespread assumption of the parasitic theory of the origin of cancer. He stated:

If cancer is due to the action of an unknown microscopic, perhaps ultramicroscopic, universal parasite, then effectual prevention must await upon its discovery. At the present day, I have no hesitation in committing myself without reservation against this theory. With most general pathologists, I regard it as incompatible with the known facts about cancer. The assumption of a universal cancer parasite can be held only by those who assume in addition that cancer is a single disease, comparable to syphilis or tuberculosis. This assumption appears to be untenable. Cancer is not a single pathological entity, but a great group of diseases, of very varied origin and course.

And yet, Ewing admits that neoplastic reaction of tissue cells is comparable to inflammatory reaction, which latter, is due to a microbic cause.

According to Ewing, a rational basis for the prevention of cancer lies in the fact that the major forms of the disease are due to some form of chronic irritation. In closing, he pointed out that early diagnosis alone is not capable of accomplishing the desired reduction in the death rate; that the patient coming with an early diagnosis all too often fails completely of a cure. He states, "The experience of patients accomplishing a cure of early cancer is generally a severe one, while the fate of the failures is passed over in silence. The public knows these facts and therefore any plan of squarely meeting the problem of cancer control must eventually lean heavily upon cancer prevention.”

While it may be possible to avoid or prevent a certain amount of chronic irritation, in the great majority of cases the chronic irritation which is supposed to be the exciting cause of the malignant tumor is associated with the occupation or routine activity of the individual, and cannot well be avoided unless one is able or willing to give up work or all physical activities and lead a life of leisure.

RESULTS OF SURGICAL TREATMENT OF CANCER

According to some statistics, it has been found possible to obtain from 30 to 50 per cent five-year recoveries in cancer of the breast by early operation, and 33 per cent five-year recoveries in cancer of the cervix by surgery and also by radiation. However the occasional results reported by a few surgeons of large experience and unusual skill fail to give a true picture of the actual average results following the surgical treatment of cancer. This was recognized by Gibson.1 His unique report of 583 cases of malignant disease observed on the First Surgical (Cornell) Service of, the New York Hospital, from February, 1913 to January 1, 1926, showed that of 437 patients operated upon, 308 had died; only 64 were known to be alive and free from recurrence at the time of the report, and only 13 were alive for a period of five years after operation. According to Gibson: "No sadder report of the disheartening status of cancer surgery has come to our attention. It is however inevitable, dealing only with facts. . . Similar research emanating from other institutions would doubt- less furnish surprises. We have been living in a fool's paradise."

One month after the publication of Gibson's paper, the end-results of treatment by operation and radiation in 376 cases of cancer of the mouth, observed at the Massachusetts General and Collis P. Huntington Memorial Hospital during the years 1918, 1919 and 1920, were brought out by Simmons.2 In brief, only 20 patients were well for a period of three years; 16 of these were treated by surgery, and 4 by radiation. In the surgical cases only was the diagnosis confirmed by microscopical examination. Of 108 cases in which the regional glands were involved at the time the treatment was begun, there were no cures.

Shore,3 has given us one of the most complete and valuable set of statistics dealing with the operability of cancer. This paper is based upon a study of 1000 cases of cancer observed at St. Luke's Hospital, New York City, from July 1923 to January 1927.

1. 62 or 6.2 per cent refused admission as hopeless.

 2. 306 or 30.6 per cent admitted but not regarded as suitable for operative surgery.

3. 632 or 30.6 per cent admitted for treatment.

Of the 320 patients operated upon 26 or 8.11 per cent died. Only 29.4 per cent left the hospital with an assurance that the malignant process had been removed. Shore observes: “These are indeed discouraging figures in the surgical treatment of cancer.”

The conclusion is that 68 per cent of the 1000 patients were found inoperable when they reached the hospital. Many more of the remaining group died of recurrence. Wood, in a later paper, based on a 1 study of 748 cases of cancer, found only 135, or 21.8 per cent mechanically removable at time of observation.

These statistics represent the results obtained by a few men of unusual skill and experience. The average results throughout the country give an even more gloomy picture, and it is upon the average results that the opinion of the laity is based. Realizing this, the reason for their pessimism becomes apparent. In spite of all the recent propaganda urging the patient to consult a doctor on the slightest suspicion of a cancer, it has been found that in many cases, even of the breast or tongue, where early recognition would seem certain, the patient himself has not been aware of the existence of a cancer until it had involved the regional nodes.

The most discouraging fact in connection with cancer control is, that more than one- half of all cases of cancer occur within the abdominal cavity; and in this large group of cases, the disease has usually progressed too far before the diagnosis is made to justify much hope of curing it by any known method. While a considerable number of cases of cancer of the stomach have been cured by surgery performed by men who have devoted years to the perfection of their technique in this field, the fact remains that the vast majority of patients with cancer of the stomach die of the disease in spite of any treatment they are able to obtain. The best statistics do not show more than 2½ per cent of five-year cures in cancer of the stomach. The same is true of cancer of the liver and intestines. Therefore, we see that practically one-half of all patients with intra-abdominal cancer are destined to die of the disease in spite of any improvement in diagnosis and treatment.

The most discouraging feature about cancer is its steadily rising mortality, which, according to Hoffman, is apparently universal. In 1885 the recorded cancer death-rate of Scotland was only 32 per 100,000, increasing to 41 by 1865 and to 48 by 1875. In other words, during the twenty years intervening, the cancer deaths increased exactly 50 per cent. The rate increased to 57 by 1885 and during the next ten years to 72, by 1895. The relative rise in the rate during these twenty years was, therefore, again exactly 50 per cent. By 1905 the rate had increased to 91 and by 1915 it had reached 111 or an actual increase in twenty years of about 39 per 100,000, or 54 per cent of the rate at the beginning of the next period. During the five years following, the rate increased to 119 and during the last three years, ending with 1923, to 130. This, in other words, is the highest rate ever reported in the history of Scotland. Thus, in forty years, or between 1884 and 1923, the cancer rate has more than doubled, or, precisely, has increased from 56 to 130 per 100,000, equivalent to about 132 per cent. Hoffman has placed the increase in the cancer death-rate at about 100 per cent during the last forty years.

I would direct attention to some equally interesting statistics for the province of Ontario. According to the last registration report, the cancer rate of that province in 1914 was 69.6 per 100,000, which by 1918 had increased to 75.5 and by 1923 to 90. Cancer of the buccal cavity underwent slight changes, but there was a material increase in cancer of the stomach, from 22.8 in 1914 to 31.0 in 1923. Still more suggestive is the increase in cancer of the peritoneum and the intestines: from 9.8 in 1914 to 14.1 in 1923. Cancer of the female generative organs increased from 5.2 in 1914 to a rate of 9.4 in 1923. Cancer of the breast, which is certainly an easily diagnosed portion of the body, increased from 4.5 in 1914 to 9.2 in 1923. Cancer of the skin also increased, from 0.99 in 1914 to 2.7 in 1923. Other cancers, including unspecified ones, have remained practically stationary, the rate in 1914 having been 20.5, while in 1923 the rate was 19.2. It is pointed out in the report that the total number of deaths from cancer in the Province reached 2724 for 1923, or probably nearly 3000 for 1924. During the last 8 decade the rate has increased at the rate of 20.4 per 100,000 of population. Comparing the mean rate of the last decade with that of the preceding decade, there has been a relative increase in the cancer death-rate of 31 per cent.

A study of the mortality statistics of the United States shows practically the same steady and marked increase in the death-rate from cancer in every state. Cancer has now reached second place among the causes of death and is rapidly approaching first place.

Before the Senatorial Committee in 1930 Dr. Joseph C. Bloodgood said:

Twenty years ago cancer occupied eighth: place among the principal causes of death in this country. Its ascendancy to second place presents a vivid picture of the steadily increasing devastation caused by the disease. Cancer now is responsible for 10 per cent of the deaths from all causes in this country.

This increase has taken place in spite of the educational campaign of the American Society for the Control of Cancer during the seventeen years of its existence, in spite of cancer weeks and cancer days, in spite of much publicity through the daily press and the magazines, in I spite of an increasing amount of radium available for the treatment of cancer, in spite of the improvement in and the standardization of hospitals in this country, in spite of the immense improvement in medical teaching, in spite of the fact that a large number of people are enlightened and come earlier.

Therefore, if more than 100,000 die annually, there must be from 300,000 to 500,000 cases of cancer living in any one year, who have either received treatment or not. I am inclined to I think that more than 50 per cent of them have reached the hopeless stage, not withstanding treatment by surgery or radiation. I think that a consensus of most of the writers on this subject is that there are about 360,000 cases existent in the United States today. All cancer students agree that the cause of cancer is undiscovered, except that chronic irritation can produce cancer in experimental animals and in the human being, but we do not know why it is so produced. All cancer students agree that at present there is no specific cure for cancer outside of its removal by surgery or its treatment by radiation and we have demonstrated that the results of such treatment in late cases are less than 10 per cent and in the earliest cases have reached about 50 per cent; but in a large per cent of all cancers -- at least 30 -- the disease is hopeless from the onset, in spite of surgery or radiation.

We have sufficient evidence to know that cancer will never be eradicated completely as the cause of death until we have discovered the cause of the disease and the preventative and curative treatment. No matter how enlightened the individual is or how early he seeks treatment after the first warning, certain types of cancer are inaccessible, or become hopeless before the first symptom is noticed. Or, the cancer cell is of such a type that it disseminates at once before surgical removal is possible, and it is not radio- sensitive which means that radiation does not kill the cancer cells. Just as we need more education, continuous, systematic and organized, we need more and better organized research.

Before the same committee Dr. James Ewing stated: "My own experience and general judgment as a pathologist relates to the proportion of cases of cancer that are now cured in the average good hospital service in this country, and I find that it is not more than ten per cent."

Moynihan, in the Hastings Lecture (Lancet, January 2, 1927) discusses cancer and how to fight it. He believes that the oft-repeated attempt to explain away the steadily rising mortality of cancer on the ground that (1) people live to a greater age now than formerly, and (2) that in earlier times there were more errors in diagnosis, has no foundation in fact. While he emphasizes the importance of the early diagnosis and the early surgical removal of cancer, he does not rest content with this alone. He would concentrate our attention upon two points, laying greater stress upon the second: (1) to make the very utmost of our present methods which are applicable to all cases of accessible growths and (2) to undertake research so that we may discover the cause or causes of cancer and so be enabled to do something for the prevention of the disease or for its cure by methods other than surgery.

The present methods of cancer control are based entirely on the theory that cancer is of intrinsic origin, that is, due to something inherent in the cell itself, and not to the influence of some extrinsic agent such as a microorganism.

The great majority of pathologists today are firm believers in the intrinsic origin of cancer. They dominate not only the laboratory field of cancer research, but the clinical field as well; and they are on record as deprecating the time and money spent in further research along the lines of a microparasitic origin of malignant tumors. As a matter of fact, very little time and money have been spent in research along the latter lines in recent years, for the simple reason that the directors of cancer research laboratories are so strongly opposed to the extrinsic theory that young men entering the field of cancer research hesitate to engage in a study so strongly condemned. Furthermore there has been a steady attempt to regard the question as closed and definitely settled. No field of medical research has ever been so completely under the influence of authority as the field of cancer research.

Tradition and authority are of great value in medicine as in any other science, but there comes a time when the investigator must forget the existence of both and concentrate his whole thought and energy upon the discovery of new facts; he must apply the principles of logic, both inductive and deductive, and endeavor to arrive at some definite conclusions. When these have been reached and they seem to be founded on indisputable facts, he should be willing to stand by them, irrespective of whether or not they conflict with opinions handed down by tradition and supported by authority.

Ewing's assumption that "If cancer is ... really caused by an unknown parasite, then prevention is not to be considered" does not seem to rest upon any logical basis. If we have found out by experience that cancer is often the result of chronic irritation or of local trauma, it seems both logical and natural that we should take all possible steps to avoid these predisposing factors, quite irrespective of whether we I do or do not know the real or underlying cause of the disease. One might as well argue that we should not have made any attempt to improve the hygienic conditions  in the prevention of tuberculosis before  the germ was discovered, although it was well known that persons living under very I bad hygienic conditions were more likely to develop the disease. All the known methods of preventing cancer and all the improved methods of treating it should be carried out while we are making most strenuous efforts to discover the cause of the disease. I believe it is a mistake to discourage investigation along the lines that offer the most hope of ever gaining control of the disease, and decidedly discouraging is the statement from a recent leaflet of the American Society for the Control of Cancer:

If cancer is really due to an unknown parasite, then prevention is not to be considered; and we are not much better off than the ancient Egyptians, since we may at any time be stung to death by the hobgoblin parasite in its own mysterious way. The distressing difference in favor of the Egyptians is that they were beautifully embalmed like Tut-Ankah-men, and we are not.

 Those of us who still firmly believe that cancer is due to some extrinsic cause will not be seriously disturbed by the substitution of ridicule for argument; but such statements coming from one of the most brilliant workers in the field of cancer research, cannot fail to discourage investigations along these lines.

In the Annals of Internal Medicine, May, 1930 (reprinted in the Journal American Medicine Association, June 7, 1930) an editorial writer (Warthin) on "Cancer Cures" makes a most uncompromising attack upon the germ theory of cancer. While in his opinion there was some reason for entertaining the idea that cancer might be of an infectious nature in the latter part of the last century, he states that by the end of the first decade of the present century experimental work on animals together with the fuller bacteriologic and pathologic studies of neoplasms had convinced us that cancer is not an infectious disease; that no specific agent exists for the production of cancer." He claims:

Mistaken conceptions regarding certain infectious growths in the lower animals, particularly the so-called Rous chicken sarcoma, have been in part responsible for the persistence of such views. By many workers the chicken sarcoma is regarded as an infectious granuloma and not comparable to the true neoplasms in man. Furthermore, there exists today absolutely no proof that infection plays any specific part in the production of neoplasms; and when once the public mind has become cognizant of the fact that cancer has no infectious etiology, much will have been accomplished to prevent patients from falling victims to cancer cures.

While this writer brushes aside as of little importance the vast amount of research work done during the past twenty years on filterable fowl tumors on the assumption that these tumors are not infectious granulomata but are comparable to true neoplasms in man, it is curious that Murray, the Director of the Imperial Cancer Research Fund, regards these tumors as by far the most promising field in cancer research today. In his opinion these tumors now have been proved to be true neoplasms comparable to malignant tumors in mammals. A large part of the ninth scientific report of the Imperial Cancer Research Fund is devoted to further studies of the filterable fowl tumors.

Furthermore, the etiology of the filterable fowl tumors still occupies the first I place in the cancer research work of the Rockefeller Institute, carried on by Murphy and his co-workers.

If it is true that these tumors are probably due to a living agent, it might be wise to be a little less positive than was the writer of the editorial referred to, that cancer in man cannot possibly be due to a similar living agent.

That the filterable fowl tumors arc true malignant tumors and not infectious granulomata as claimed by the editorial writer in the Annals of Internal Medicine is proved by the fact (as shown by Murray; that metastases in other organs, e.g., the liver or lungs, are always made up of cells corresponding to the type of cell found in the primary tumor, exactly as occurs in human cancer.

These tumors are now recognized in all the important Cancer Research laboratories of the world as true tumors and not infectious granulomata.

Let us study for a moment the main points in favor of the microbic origin of cancer:

1. The close resemblance of certain cases of cancer to other diseases of known microbic origin, for example, tuberculosis and syphilis.
2. The frequency with which there is both a local and general rise in temperature in cases of malignant tumor. This is especially true of endothelioma of bone and other types of cancer with rapid generalization, in which the general temperature often rises to 1O3-1O4°F.
3. The difficulty in distinguishing tuberculosis from sarcoma and chronic osteomyelitis, or osteitis from endothelial myeloma, either by the clinical evidence or, in some cases, even by a study of the histological appearance. In a few cases it is impossible for the most experienced pathologist to differentiate these conditions.
4. Single antecedent trauma is now generally conceded to be an important causative factor in malignant tumors, especially bone sarcoma. No satisfactory explanation of the influence of trauma has ever been offered except on the assumption that the malignant tumor as due to some microorganism. Granting this, the explanation is easy. We have an exact parallel in cases of osteomyelitis as well as tuberculosis of bone, following local trauma.
5. The antagonistic action of the streptococcus of erysipelas, and the toxins of n erysipelas and Bacillus prodigiosus, upon various types of malignant tumors, is difficult to explain except on the theory that malignant tumors themselves are caused by some microorganism.
6. The remarkable variation in the geographical distribution of cancer. The most recent illustration of this may be found in the Westmoreland County Survey. 12

Ewing and many other pathologists I have stated that cancer is not a single disease, but a group comprising a large number of separate and distinct diseases, and in order to assume a microbic cause one must assume a large number of different microorganisms.

This assumption is by no means necessary. Sir Charles Ballance, at a meeting of the American Surgical Association twenty-five years ago, stated that he was firmly convinced not only that malignant tumors were due to some microorganism, but further that it would probably be shown that the same microorganism, under different conditions, gave rise to the different types of cancer.

Erwin Smith proved this to be true as regards malignant tumors in plants. By making superficial injections he produced epithelioma; by deep injections, a sarcoma; and by injecting a latent bud he produced a teratoma.

In a series of experiments recently carried out at the Laboratory of the Hospital for the Ruptured and Crippled by Dr. Richard F. Berg, using the dried virus of the filterable endothelioma fowl tumors, kindly furnished me by W. E. Gye, of London, found it possible to produce 5 different varieties of bone sarcoma, by injecting the virus into the marrow of the tibia of very young Rhode Island red chicks, one to two weeks old. These tumors correspond very closely to the several types of bone sarcoma in man.

The filterable fowl tumors, which are now almost universally accepted as true neoplasms, comparable to malignant tumors in man, are believed, by the majority of investigators, to be due to some microorganism or living agent, although up to the present time it has never been isolated.

Conner, of San Francisco, had previously been able to produce endothelial myeloma by the same dried virus. Our five varieties corresponded exactly with the different varieties of bone sarcoma in man.

During the past year further experiments with this virus and with the tumors produced by it have been carried on by Dr. J. E. Sullivan (Dr. Berg's successor to the Gibney Memorial Fellowship, the Hospital for Ruptured and Crippled) and myself. These experiments I believe offer convincing evidence that these tumors are caused by a living agent. The full results will be published in the near future.

It would be possible to explain the different types of malignant disease in two ways:

1. The microorganism attacks a particular cell which, by reason of some specific factor, e.g., local trauma, chronic irritation or some chemical changes in the body fluids, offers a lowered resistance to the microorganism. By reason of this lowered resistance, the organism gains a foothold or finds a favorable soil for development, and thus begins the early stages of a malignant tumor. The micro-organism enters the cell itself, becoming an intracellular parasite which by its constant irritation stimulates the cell to growth and division. This growth is naturally made up of the multiplication of the cell in which the organism first enters, and this very naturally accounts for the remarkable fact that the metastases of a malignant tumor always correspond morphologically with the type of cell of the primary tumor. This has been one of the main arguments of the pathologists against the parasitic theory, but if we assume that we are dealing with an intracellular parasite, then the difficulty is at once overcome.
2. We may suppose that instead of a single strain there may be a number of strains of this unknown microorganism, closely related and, perhaps, indistinguishable under the microscope or by variations in the manner of growth upon culture media, which strains may produce different types of tumors.

Personally I am inclined to believe in a single microorganism possessing different degrees of virulence, which, having gained access to the blood of a healthy individual, remains latent until the natural resistance of the cells has been lowered in some local area by some special factor, e.g., local trauma, chronic irritation or chemical changes in the body fluids, furnishing the organism with a favorable soil for its development.

In addition, there may be some inherent or inherited susceptibility of the cells of a certain individual, or an unusual lack of resistance that may account for the varying degrees of what we call malignancy, without assuming a great change in the virulence of the microorganism.

What determines the type of the tumor may not be the germ but the soil or some inherent factor in the cell itself which attracts the organism to one type of cell in one individual and to another type in another individual. If the Rous sarcoma virus is injected into the blood of a susceptible fowl, no tumor develops. Not until an injection is made into areas in which connective tissue is present and a connective tissue reaction is produced, is it possible to produce new tumors. The Rous sarcoma is a spindle cell sarcoma.

The theory that cancer is due to some infectious or microbic cause received great impetus and support by the discovery of the filterable fowl tumors by Fujinami of Japan and Peyton Rous of the Rockefeller Institute (1910). It was found that these tumors could not only be produced by living transplants, as had been done in rat sarcoma and mouse carcinoma, but, by a cell-free filtrate. Then for a time an, attempt was made to explain these tumors as not being true neoplasms but similar to the so-called infectious granulomata. The virus itself has been studied in many laboratories, and various opinions have been held as to its true nature, one group of investigators, the larger, believing, it to be a living agent, and another, a so-called enzyme or a chemical agent. The remarkable work of Gye of London in July, 1925, gave a new impetus to the opinion that cancer is due to a microbic cause, and the London Lancet gave his address a four-page editorial and intimated that at last the age-long problem of the cause of cancer had been solved.

Gye, basing his views upon a long and elaborate study of the Rous sarcoma, evolved the theory that cancer (malignant tumors in general) is not due to a single cause but to a dual cause. He believed: (1) that the Rous sarcoma tumor was due to a "living agent" which had not yet been isolated; (2) that this living agent alone could not produce malignant tumors, but required some other agent, which he called a specific factor, to produce the disease. His experiments seemed to prove the truth of this brilliant theory. He first rendered inert the active virus of the Rous sarcoma by chloroform or some antiseptic, until it could no longer produce any tumors on inoculation. Then by adding some one of several substances to the inert material, it once more became active and produced the Rous tumor on inoculation. This discovery was regarded at first as epoch-making, and seemed to go far toward solving the age-long problem of the cause of cancer; but, unfortunately, no one was able to repeat these experiments; and the most rational explanation would seem to be that the living virus was not entirely destroyed by the chloroform and that the addition of the specific factor simply helped to revive the inactive but not entirely dead virus, and rendered it capable of producing typical tumors.

Gye's4 theory differs but little from my own in a general way. According to my theory5 there is a dual cause: (1) the living agent which may be widely disseminated and is often present in the tissues or in the circulation of many individuals but doing no harm; (2) Some other factor, not so highly specific as Gye's, for example, local trauma, chronic irritation, or some chemical change in the body fluid, that damages some local area thereby causing a tissue reaction or a favorable soil for the organ- e ism to gain a foothold; and thus the beginning of a malignant tumor.

Hereditary susceptibility is doubtless and additional factor in many cases. It means that certain individuals have inherited tissues or cells that are less resistant to the cancer organism when exposed to it.  

My theory does not depend on the success or failure of a series of laboratory experiments such as Gye described. It has numerous clinical facts to support it. Furthermore, recent laboratory investigations furnish evidence in its favor.

Cramer, in the ninth Scientific Report on the Investigations of the Imperial Cancer Research Fund, reports a series of experiments in which he was able to transmit three strains of mouse sarcoma and Rous sarcoma without the presence of living cells. While numerous experiments along these lines have been made in the past, all seemed to prove that the trans- mission or mammalian neoplasm depended on living cells. Three different tumor strains were used by Cramer in his experiments: (I) rat sarcoma (2) different strains of mouse sarcoma, and (3) mouse carcinoma. The mouse carcinoma gave negative results.

These experiments would seem to place neoplasms in mammals measurably nearer, if not in the same-class as the filterable sarcomas of Rous and Murray.

Among the clinical facts that furnish evidence in favor of the theory that malignant tumors in man are due to a micro-organism or an infectious virus of some kind, the most important and convincing is the case reported by Lecene and Lacassagne.6 Here a medical student, while attempting to aspirate a collection of lymph from under the scar of an amputated breast (removed for cancer eleven days previously) pricked the palm of his left hand, at the same time discharging a small quantity of the liquid that was in the syringe. Three-quarters of an hour later the wound was cauterized with a galvanic current. The accident was forgotten until two years later when he began to have diffuse pain in the same hand. Shortly after he noticed a bosselated swelling at the exact site of the previous puncture. One month later an enlarged gland, the size of a nut, appeared in the axilla. This gland was removed and was thought to be tuberculous, although no evidence of the disease was found. Four months later the skin of the hand became invaded and ulcerated. One month later the tumor was excised under ether anesthesia. It recurred promptly and showed unmistakable evidence of being a malignant tumor. About five weeks later, four nodules appeared in the skin of the forearm; and a week later, disarticulation of the arm at the shoulder was performed. A careful histological examination of all the nodules and tissue removed at the various operations showed the same general structure. A diagnosis of spindle cell sarcoma was made.

Lecene and Lacassagne, after discussing various hypotheses expressed their conviction that:

The transmission of cancer is accomplished by the inoculation of a virus analogous to the filterable virus admitted by Gye, possibly of an infectious principle analogous to that admitted by Carrel. The change in the histology of the type of the tumor is entirely “reconcilable" with this hypothesis.*

*In our experimental production of bone sarcoma in young chickens at the laboratory of the Hospital for Ruptured and Crippled, Dr. Richard Berg, by injections of the virus of the endothelial fowl sarcoma was able to produce not only several varieties of sarcoma but, in two cases, typical carcinoma).

In my own early experience I recall a case that points to the possibility of a lymphosarcoma of the axilla having been caused by a microorganism introduced in the finger by the prick of a thorn. A young woman while picking roses, stuck a thorn into her index finger. Two or three weeks later, a swelling developed in the axilla of the same side, and grew rapidly. I removed it by operation and it proved to be a round cell sarcoma. It recurred promptly and finally caused death by general dissemination. I was greatly impressed at the time of a causal relationship between the wound of the finger and the rapid development of the malignant tumor in an axillary gland of the same side. I still believe that a microorganism was introduced into the circulation by the thorn-wound of the finger and that it found in the gland of the axilla a favorable soil for development, causing the malignant tumor that proved fatal in less than a year.

Relative to the Part Played by Trauma. If we inject the active virus of a cell-free filtrate of the Rous sarcoma intravenously into an animal, no tumor develops. On the other hand, Pentimalli7 has shown that in a fowl bearing filterable sarcoma No. I, a new tumor can be started by applying trauma to some other part of its body. The only explanation of this that can be offered is that the trauma produced the connective tissue reaction that is essential to the development of a Rous tumor.

This experiment has a very important bearing on the question of trauma as a causative factor in the development of malignant tumors in man. Also it seems to furnish a striking analogy to the experimental production of osteomyelitis in rabbits. It has been found that a sharp blow to the tibia of a rabbit produces nothing unusual; but if the animal is first inoculated intravenously with a small quantity of Staphylococcus aureus and then struck a blow on the tibia, an acute osteomyelitis will develop. In other words, the local trauma furnishes the tissue reaction or soil favorable for the localization of the germ which is present in the circulation but quite harmless until the local resistance of the tissues has been lowered. If this theory is correct, then we have for the first time a rational and scientific explanation of how trauma can be a causative factor in the development of malignant tumors.

One of the most recent and strongest arguments in favor of the microbic origin of malignant tumors is found in the monograph of Prof. Leon Bouveret8 of Lyon, France. In addition to the clinical evidence in favor of the theory, he believes the most important and most conclusive evidence is found in the inhibitory, and often curative, influence of the streptococcus of erysipelas upon malignant tumors. This influence, recognized as far back as the time of Hippocrates and reported in a gradually increasing number of cases, is indisputable. Bouveret reports some personal cases and refers to others in French literature. He has carefully examined the various explanations of the curative effect of erysipelas on cancer, and can find only one that is at all satisfactory, i.e., that malignant tumors, themselves, are caused by some strain of streptococcus, and that the curative action of erysipelas is brought about by antibodies in the blood. This is the explanation that I offered in my first paper 1893 in which I reported the results of treatment of 12 cases of inoperable cancer with the living cultures of the streptococcus of erysipelas; and I have continued to hold this opinion ever since. I would not go so far as Bouveret in assuming that the causal agent of cancer is a strain of streptococcus of erysipelas. I do not believe it essential to assume such a high degree of specificity that it could be acted on only by the same, or a strain of the same organism. We know that the tumors of leprosy have been caused to disappear by a combination of the toxins (vaccines) of the tubercle bacillus and the Bacillus pyocyaneus, quite as rapidly and completely as by a combination of the B. leprosy and B. pyocyaneus. The fact that we have been able to cause in a considerable number of cases the total disappearance of malignant tumors of practically all types by the mixed toxins of erysipelas and B. prodigiosus (which I began using in 1892) proves that heterogenous toxins may have the same curative action as the autogenous. I would leave open the question as to the type of organisms that is the causative agent of malignant tumors; but that such tumors are caused by some living extrinsic agent, ( I believe, with Bouveret, that this action of the erysipelas furnishes the strongest kind of proof. Bouveret states:

“It is a fact generally admitted without dispute, that an acute febrile malady may cause a malignant tumor to regress.” He believes it still remains to interpret this fact under the influence of diverse febrile conditions: a sarcoma, a lymphoma, and an epithelioma undergo a certain reduction in size. But, he adds, there is found a fairly I pronounced regression that may occasionally go on to complete disappearance of the tumor. One finds this among the published cases only in those associated with an attack of erysipelas. To this general rule, Bouveret has found in his researches only two exceptions. Riffel9 has seen cases of cancer of the face and of the stomach, whose evolution was suspended by an attack of intercurrent variola. In one case of Kutzner,10 an attack of pneumonia caused a regression of a cancerous neoplasm. I have had a case of sarcoma of the tonsil nearly disappear after an attack of pneumonia, but it promptly recurred and proved fatal within a few months. There can be no doubt that the influence of erysipelas upon malignant tumors is much I more powerful than any other febrile disease. Bouveret states that he has never noted regression following attacks of la grippe, typhoid fever, eruption fevers, or septicemia. On careful inquiry in two hospitals for incurable cancer, he found no history of a case of cancer showing notable regression following intercurrent febrile disease such as bronchopneumonia, although there were many examples of such association. Bouveret asks: "Why is the influence of the erysipelas so preponderating?" and he adds, “it not permitted to presume that there exists some relation between the erysipelas and the cancer?" In other words, may not malignant tumors themselves be caused by some microorganism, possibly by some strain of streptococcus closely allied to the streptococcus of erysipelas?

Bouveret has given the most careful study and discussion of the explanation of the regressive action of erysipelas upon malignant tumors that has ever been published, and a brief review of this argument should prove of interest. He points out that the explanation that has usually been accepted in the past, and the one that for the most part obtains today, is that of "hyperthermie" or high temperature. This is the explanation that is offered by Willy Meyer in his recently published book on "Cancer."13 It is one, however, which, at the time of my early experiments with inoculations of living cultures of erysipelas (1891-1892) I stated could not satisfactorily account for the inhibitory and often curative influence of an attack of erysipelas. The theory that I offered as the only one that could rationally explain this action, was: that malignant tumors themselves are due to some microorganism; and that the antagonistic and inhibitive action of the erysipelas streptococcus is due to certain changes in the blood serum brought about by antibodies which destroy the causative agent of the tumor and that later, the generating tumor is either absorbed or undergoes extensive necrosis requiring incision and evacuation.

Bouveret, in agreeing with my early explanation, offers the argument that, if fever were the true, efficient and only cause of the marked regression of a malignant neoplasm, such regression would always follow an attack of erysipelas, especially in cases associated with a severe attack of erysipelas with unusually high temperature. As a matter of fact, he states that the very severe attack with high fever does not generally produce the most marked regression.

Again, if high temperature alone were the sole cause, then all febrile diseases associated with high temperature and also with malignant tumors, should show marked regression. Such we know is not the case. Clinical observation shows, without any question, that the greatest number of cases of regression are associated with erysipelas alone. I do not agree with Bouveret that the most marked regressions are not associated with the more severe attacks of erysipelas. In my experience I have found that, as a general rule, the more severe the attack of erysipelas, the more marked the regression, but this I would by no means attribute to the higher temperature but to the production of antibodies in greater numbers and of higher antagonistic power. Bouveret states:

Erysipelas has the power of curing rebellious ulcers, lupus or syphilis. Is fever the exclusive agent of these cures? These old ulcers have not healed because they are the seat of a chronic, tenacious, microbic infection nor can it be the fever of the erysipelas that destroys the microbes of these infectious processes.

Borrel has made a study of the effect of I fever upon pathogenic microbes: "The fever has a salutary significance, not, because the microbes suffer from the elevation of temperature, but because it seems to destroy the processes of defense."

Duclaux11 gives, after the work of Sternberg, the temperature necessary to destroy pathogenic bacteria. The staphylococcus dies at a temperature of 62°C. and the streptococcus at 54°C. Hence it is readily seen that the temperature of erysipelas, 39°. to 40°C., falls far short of the degree necessary to destroy the micro-organisms present in these old ulcers. Bouveret concludes that it is not the fever then that cures these rebellious, old, tuberculous or syphilitic ulcers, but some other mode of action. Why cannot this other mode of action, or agent, accordingly, cause a regression of malignant tumors?

If, then, the fever is not the essential cause of the regression, what is the cause? It must be the infectious state or infectious nature of the erysipelas. How does it act? The organism tries to defend itself against the erysipelas infection. The general defense of the organism against an attack of erysipelas is assured, not only by the phagocytes, but also by the antibodies in the blood, antibodies which reduce or suppress the virulence of the streptococcus and neutralize the toxins it has produced.

Although Borrel in 1908 found that a transplanted rat sarcoma when subjected to a temperature of 37°C. for twenty-four hours lost its virulence, Haaland (Contamin Le Cancer Experimental, p. 130) studied the effect of heat upon mixed tumors, epithelioma and sarcoma, and found that a temperature of 44°C. for twenty-five minutes did not destroy the virulence of the sarcomatous cells. Fujinami and Inamaoto (Menetrier: Le Cancer, p. 164) held that although it destroyed the sarcoma cells of filterable fowl tumors, it did not destroy the virulence of a fragment of tumor subjected to a temperature 40° to 50°C. since it was found possible to inoculate successfully other fowls with the heated fragment. Doyen (Congress de Chirurgie, 1908) concluded from a series of researches, that the cancer cell was destroyed by a temperature of 6° to 8°C. Of course, such a temperature is vastly higher than that which occurs in an attack of erysipelas.

Once the truth of the parasitic theory of cancer is proven then a vast new field of scientific research is opened up, and this field holds out the promise of greater success in controlling the disease. We do not need to look for an immediate serum or vaccine that will effect a cure in a large proportion of malignant tumors that have already developed. Such a cure was looked for immediately on the discovery of the tubercle bacillus, and some of us remember the disappointment that followed Koch's failure to produce such a serum. We must expect the long period of observation and further investigation that followed the discovery of the tubercle bacillus. While no vaccine has been found that has proved of great value in the treatment of tuberculosis, knowledge of the habitat of the tubercle bacillus and of its probable mode of access in the human body, has enabled us to develop preventative measures of great value, and now the mortality of tuberculosis is scarcely more than one- third of what it was before the germ was discovered. Observation revealed that the organism was found frequently in other animals especially in cows, and that many cases of tuberculosis in children resulted from the use of unpasteurized milk from infected cows. This discovery enabled us to institute preventative measures of great value.

In my own opinion, it would seem most probable that the microorganism supposed to be the cause of malignant tumors is very widely distributed; that practically everyone is exposed to it; that it is much more prevalent in certain countries and localities than in others, and that that is the chief reason why we find the wide variations in the incidence of cancer in different localities some of them in close proximity to one another. This theory is strongly supported by the recent survey of Westmorland County Survey.

In this connection, the investigation into the incidence, epidemiology, and ecology of cancer in Westmorland County, England,12 is of interest. This survey, made under the auspices of the Westmoreland Field Commission for Cancer Research, was carried out under the direction of Dr. Louis Sambon, whose cancer investigations in the American tropics, in Iceland, Holland, and Italy, are well known. This report states:

In the first place, it was decided to discountenance all theories concerning causation, and to ignore, for the time being, all views based solely on clinical or pathological data. It seemed better to approach the disease as if totally unknown and collect only sheer facts...

Our investigations indicate very forcibly that we are confronted with a systemic infection characterized by long latency and exhibiting varied local manifestations, determined in type and site by the most diverse physical, chemical, mechanical, or animate irritants. Insidious invasion, long quiescence, complexity of secondary factors, muteness of symptoms, and protean nature of manifestations are the peculiar features which have baffled inquiry so persistently.

Dr. Sambon believes that cancer is brought about by the activities of an exceedingly minute endocellular parasite strictly adapted to its specific cell host. He points to Rickettsia, the germ of typhus, which inhabits almost exclusively the endothelial cells of the capillaries, giving rise to marked defensive proliferation, the formation of intravascular overgrowths (Fraenkel’s “nodules”) and metastases. He looks upon the gross tumour – the actual cancer – as a late, often fatal, manifestation of the cancerous infection not unlike the gumma intertiary syphilis. As in tuberculosis, so also in cancer, the specific germ appears to be universally prevalent and of wide zoological distribution.

Perhaps the most striking, the most interesting discovery our investigations have elicited is the great prevalence of all types of malignant and other neoplasms in animals. We have met with cancer in all kinds of wild and domestic animals, in pet animals, and in animals used as food. We might mention horse, cow, sheep, hog, cat, dog, hedgehog, rabbit, vole, rat, mouse, fowl, turkey, goose, owl, canary, and toad. Fowls of all varieties (Leghorns, Wyandottes, Plymouth Rocks, Bantams, Anconas, Rhode Island Reds in our series) are particularly liable and we have noticed small outbreaks of the disease in certain poultry flocks similar to the cage epidemics of rats and mice mentioned by Borrel in France and by Gaylord in America. This unexpected, remarkable prevalence of the disease in all kinds of animals opens up a wide field of possibilities which we are already exploiting.”

In his lectures at the London School of hygiene and Tropical Medicine, and in his epidemiological papers, Dr. Sambon has repeatedly stressed the importance of a full knowledge of the zoological distribution of disease. Without a knowledge of the animal sources of infection it is not possible to understand, far less to control, outbreaks of disease in man. A memorable example is that of Russian artillery sent to destroy all the inhabitants of a plague-stricken border village in order to stay the epidemic. The unfortunate people were sacrificed, but no one troubled about rats, and the disease soon spread like wildfire. Is it not known now that many cases of tuberculosis are not of human source, but, on examination prove to be of either bovine or avian type?

When once the organism which proves to be the cause of human cancer is discovered we should at once make a careful study of our domestic animals, especially dogs, cats, fowl, pigeons, mice and rabbits and especially raw vegetables and sources of water supply. If the organism is found frequently in these animals, then the proper steps for lessening the chances of infection from such sources should be taken. It is quite possible that some of these animals might be carriers of the disease without themselves being affected by it. It is possible that the organism may be soil organism and enter the system through uncooked vegetables or raw fruit.

Finally it might be possible to produce a vaccine or serum that would have an inhibitory or curative action upon malignant tumors, or would be of great prophylactic value when used after operation or in connection with irradiation.

These are some of the numerous and attractive fields of investigation that would be immediately opened up once the germ has been discovered.

To those who are thoroughly convinced that cancer cannot be due to a microorganism, this may all seem scarcely more than a beautiful dream. To those who are still open-minded, I have tried to show that the microbic theory of cancer rests upon a solid foundation of clinical facts and is further supported by a large and steadily increasing amount of experimental and laboratory evidence especially in the field of filterable fowl tumors. It will require much further work and investigation before this theory can be definitely established; but when we consider that this is a matter of such vast moment to the million or more sufferers from cancer and to the still greater number who are certain to suffer from cancer in the years to come, we who are engaged in cancer research, both: - clinical and laboratory, should feel stimulated to do our utmost to solve the problem of the cause of cancer. If the cause proves to be a specific living agent, as I believe it to be, then for the first time we shall have some satisfactory foundation for our hope of gaining control of the disease.

After you have listened to the arguments for and against the parasitic nature of cancer, you may feel yourself in the position of Rabbi Ben Ezra of Browning's poem:

Now, who shall arbitrate?

Ten men love what I hate,

Shun what I follow, slight what I receive;

Ten, who in ears and eyes

Match me; we all surmise,

They this thing, and I that: whom shall my soul believe?

Unlike Rabbi Ben Ezra, I freely admit that the question of the etiology of cancer is not so nicely balanced, and that I belong to a small minority. Still one may remember that in medical science as well as in politics, minorities have a surprising way of becoming majorities in a very short time. What we need most is more facts rather than opinions.

CONCLUSIONS

1. Every means, including those already adopted by the American Society for the Control of Cancer, should be employed to encourage early recognition of malignant tumors as well as prompt and proper treatment of them. This entails not only continuous propaganda to persuade the patient to consult a physician as soon as a tumor is discovered, but, and what is more important, entails greater effort in educating physicians in the difficult field of the diagnosis of malignant tumors.
2. A considerable portion of the funds raised for cancer control should be expended in the field of collecting more facts, especially more accurate firsthand information as to the geographical distribution of cancer, and especially, data that may account for the wide variation in incidence; in other words, surveys should be made similar to the one of the Westmorland County, England, conducted by Dr. Sambon in 1926-1928.
3. Little advance may be expected in the surgical treatment of cancer, but in the field of roentgen-ray and radium, I believe, that, with increasing knowledge of the action of these agents, and with improved technique, considerable advance may be anticipated. Radiation has already displaced surgery in certain fields, i.e., cancer of the skin, of the cervix, and oral cancer. It is rapidly becoming recognized that most cases of cancer can best be treated by a combination of surgery and radiation, rather than by the use of either alone.
4. A further study of the treatment of inoperable tumors, especially sarcoma, by the toxins of erysipelas and bacillus prodigiosus, should be carried out. It is found that a considerable number of these cases beyond hope from any other method of treatment can be cured by toxins, and that in many operable cases in which the toxins have been used as a prophylactic after operation a larger percentage of patients have remained well for five years; therefore this method should receive more general adoption.
5. A systematic attempt should be made to give the undergraduate medical student much better clinical instruction in the early diagnosis of cancer than he has had in the past; and even more important is it to provide numerous centers where intensive post-graduate instruction may be obtained. Cancer departments should be organized in the larger general hospitals, to be in charge of men who have had wide experience in the diagnosis and treatment e of cancer.
6. A number of so-called Cancer Institutes should be established in a few of the principal cities of the country, as recommended by Ewing at the meeting of the American College of Surgeons in October, 1929. Such institutes should have a large endowment to make them educational centers; they should be equipped to give the most approved methods of treatment; and they should contain research laboratories in which the whole field of cancer is studied.
7. The results of study by the various organizations and individuals working on the problem of cancer, including both the clinical and laboratory aspects, should be given wide publicity. At least every three to five years there should be an international cancer congress, bringing together the leading workers in cancer, and furnishing an opportunity for an interchange of ideas.

REFERENCES

1. GIBSON, C.L. Final results in the surgery of malignant disease Ann. Surg., 84: 108, 1926.
2. SIMMONS, C. C. Cancer of the mouth; the results of treatment by operation and radiation. Surg. Gynec. Obst. 43: 377, 1926.
3. SHORE. B. Operability in cancer; statistical study of 1000 cases. J. A. M. A., 90: 1690, 1928.
4. GYE, W. E. Aetiology of malignant new growths. Lancet. II: 109, 1921
5. COLEY. W. B. Clinical evidence in favor of extrinsic origin of cancer. Surg. Gynec. Obst.. 40: 353, 1925.
6. LECENE, P. and LACASSAGNE, A. Accidental inoculation of malignant tumor in man. Ann. D’anat. Path., 3: 97, 1926.
7. PENTIMALI, F. Ueber den Stoffwechsel des regenerierenden Gewebes. Zucbr. J. Krebsforscb., 25: 347, 1927.
8. BOUVERET, L. Essai sur Ie Pathogenie duCancer. Paris, Bailliere, 1930.
9. RIFFEL. In: Duroux: Les Cancers. 1923.
10. KUTZNER. In: Menetrier: Le Cancer. Paris, 1908, p.261.
11. DuCLAUX. Traite de MicrobioIogie. I: 603.
12. SAMBON, L. Brit. M. J., I: 1062, 1929.
13. MEYER, W. Cancer. N. Y. Hoeha, 1931.