Kt‘|>rinied from 
THE POST-GRADUATE. 
A U£. 189.*. 
A PRELIMINARY NOTE 
ON THE 
TREATMENT OF INOPERABLE SARCOMA 
BY THE 
TOXIC PRODUCTS OF ERYSIPELAS. 
By WILLIAM B. COLEY, M. D., 
OF NEW YORK. 
Instructor in Clinical Surgery, New York Post-Graduate Medical School; 
Surgeon to the New York Cancer Hospital; Assistant Surgeon 
to the Hospital for Ruptured and Crippled. 
The Meriden Gravure Co., Meriden, Conn. A Preliminary Note on the Treatment of Inoperable Sarcoma by 
the Toxic Products of Erysipelas. 
BY 
WILLIAM B COLEY^JL-®^ 
Instructor in Clinical Surgery, New York Post-Graduate Medical School; 
Surgeon to the Yew York Cancer Hospital; Assistant Surgeon 
to the Hospital for Ruptured and Crippled. 
In a recent paper published in the American Journal of Medical Sciences, 
May, 1893, I have given the results in a series of cases of malignant tumors 
treated by repeated inoculations of erysipelas. The paper also contains an 
abstract of all the cases that I had been able to find in literature, when an 
attack of erysipelas had occurred in patients suffering from sarcoma or car- 
cinoma. The clinical and experimental data above referred to conclusively 
prove that the germ of erysipelas contains a powerful principle, which has 
a marked antagonistic effect upon malignant tumors, more especially sar- 
coma, and which, in a certain proportion of cases, is curative of the same. 
The results of my experiments and investigations are in brief as follows: 
Of forty-three cases of malignant disease associated with erysipelas, 
twenty were carcinoma, nineteen sarcoma, and four either sarcoma or car- 
cinoma. In twenty-three cases the erysipelas was accidental, and in fifteen 
the result of inoculation. 
Of the twenty cases of carcinoma, three were permanently cured. In 
addition one case of probable carcinoma (Hutchinson’s) was well five 
years after the attack of erysipelas. One died from the inoculation. Most 
of the remaining cases showed more or less improvement, which, though 
temporary, undoubtedly added to the life of the patient. 
Sarcoma.—In the nineteen cases of sarcoma we find eight, or forty-two 
per cent, well from one to seven years after the attack of erysipelas. Most 
of these cases have a remarkable history, the details of which may be 
found in the paper alluded to. Two died as a result of the erysipelas. 
These cases have been in no way selected, and every effort has been made 
to include all cases resting upon competent authority. 
When we reflect that in nearly e\ery instance the tumor was not a pri- 
mary growth amenable to operative treatment, but either a recurrence after 
operation had been tried and failed, or from its nature and situation in- 
operable, then we are able to appreciate the importance of erysipelas as a 
curative agent. 
During the past two years I have treated twelve cases of inoperable, ma- 
lignant tumors with repeated inoculations of the living cultures of the strep- 
tococcus of erysipelas. Four of these cases were carcinoma and eight were 
sarcoma. 
In but four of the twelve cases did I succeed in producing a true attack of 
erysipelas, although in nearly all cases the frequently repeated injections of 2 
liquid cultures had more or less effect in diminishing the size of the 
tumors. The four cases in which erysipelas was produced were all sar- 
coma. In two, the tumors entirely disappeared. One was very large, 
7x5 inches, situated in back, with secondary tumor, size of goose egg, in 
groin, and the second was a recurrent sarcoma of the neck, with a very 
large secondary tumor in the tonsil. In the first case both primary and 
secondary tumors disappeared, while in the latter the primary tumor disap- 
peared and the tonsil tumor diminished in size, and although nearly two 
years have passed it has apparently not increased, a fact which seems to 
prove its malignant nature must have been changed. 
Recurrence followed in a few months in the first case. The tumors grew 
rapidly, but again disappeared under a second attack of erysipelas. Still 
another recurrence followed. The focus in the back was removed with the 
knife, but before the wound healed the disease had reappeared both in the 
back and the groin. Injections of the Toxic Products of erysipelas were 
then resorted to, and repeated at intervals of forty-eight hours for three 
weeks. The sarcomatious nodules disappeared, and the patient has en- 
tirely regained his normal weight and general health. No recurrence has 
yet taken place, and it is now fourteen months since the first inoculation, 
and four months since the injections were discontinued. 
Of the remaining two cases where erysipelas was produced, one showed 
marked decrease in the size of the tumor, but the effect in checking the 
growth was temporary. The fourth case died on the fifth day as a result 
of the erysipelas. In these five days the sarcoma had decreased markedly 
in size. 
The uncertainty of being able to produce erysipelas when desired (failure 
having occurred in two-thirds of my cases) coupled with the fact that fre- 
quently repeated injections of the liquid cultures of the erysipelas germs 
had produced a marked beneficial effect upon the tumors, led me to attempt 
some plan of isolating and using the active principle of the germ. My first 
experiments were made with liquid bouillon cultures, prepared by simply 
subjecting them to a temperature of ioo° C. 
Injection of cultures thus sterilized produced similar reactions to the 
living cultures, though the reactions were somewhat less severe. The in- 
jections were not continued sufficiently long to ascertain the relative effect 
upon the tumors, yet it was apparently less marked than the living cultures. 
Considering the complex nature of the Toxic products, there seems good 
reason to suppose that certain changes were produced by heat in sterilizing 
the cultures. I therefore determined to obtain, if possible, a preparation of 
the Toxic principles of the erysipelas germ without the use of heat. This I 
was finally able to do through the skill of Dr. Alexander Lambert of the La- 
boratory of the College of Physicians and Surgeons. The streptococcus 
was grown in bouillon for three weeks and then filtered through a modified 
Kitasato filter. The fluid thus obtained was found to be perfectly sterile. 
It was then putin glass-stoppered bottles, and kept in a dark, cool place. 
This preparation I have now used in six cases of Sarcoma. Some of the 
cases are still under treatment. 3 
Case I—A boy sixteen years of age, German, with good family and per- 
sonal history; was referred to me by Dr. L. B. Bangs of this city, in January, 
1893, for the treatment of a large, inoperable sarcoma of the abdomen and 
pelvis. Its transverse diameter was seven inches and its vertical five. It 
extended nearly to the umbilicus above and was deeply attached to the 
pelvis below. In consistence it was exceedingly hard. Examination of a 
section removed proved it to be sarcoma. He was sent to me with the view 
of producing artificial erysipelas, but I decided to try first the effect of the 
Toxic products. Injections of the preparation just described were begun 
early in February and continued in slightly increasing doses, at intervals of 
two or three days, for ten weeks. The doses varied from ,5 c.c. to 1.5 c,c., 
and ,2 to .3 c.c. of the Toxic products of the Bacillus Prodigiosus were 
given to increase the reaction. 
The treatment was discontinued May 5, 1893. Careful measurements, 
made from time to time, showed the tumor to be steadily decreasing in 
size and the general condition of the patient correspondingly improved. 
He had been confined to bed much of the time before the injections were 
begun. June 1, 1893, the tumor had so far disappeared as to be no longer 
visible. Examination showed a small, movable mass in the abdominal wall, 
at site of original tumor, extending but a short distance above the pubes, 
measuring less than two inches in diameter. When the treatment was 
begun the measurements recorded were seve?i inches by five. 
The boy gained ten pounds during the last month, and all bladder irrita- 
tion, which was marked at the beginning, had entirely disappeared. Dur- 
ing four weeks the injections were omitted, and as the tumor began to show 
signs of enlargement the injections were resumed. The case is still under 
treatment. The reaction following these injections has corresponded al- 
most exactly with the constitutional reaction occurring at the beginning of 
an attack of true erysipelas. Within one-half hour to twelve hours after 
the injection a more or less defined chill occurs, usually accompanied by 
nausea and vomiting. Severe headache and general muscular weakness 
likewise are present. Within twelve hours the temperature rises from ioip 
to 1040, depending upon the amount used. Locally the parts about the site 
of puncture become swollen and reddened. The local and constitutional 
reaction generally subsides in twenty-four to forty-eight hours. July 17th, 
the tumor had almost entirely disappeared, and there now remains scarcely 
more than some enlarged glands in the right groin. 
Case II—Was the sarcoma of the back and groin, given in detail in the 
paper referred to. The sarcoma had twice disappeared under attack of 
artificial erysipelas, and had twice recurred. The final disappearance, 
caused by injections of Toxic products, is all the more remarkable in the 
light of the previous history of the case. Four months have passed and 
there is absolutely no trace of recurrence. 
Case III—Recurrent Sarcoma of Thigh. The primary growth first ap- 
peared in the soft part just above the left patella, in July, 1889, It was 
removed in November, 1891. Local recurrence followed. A second opera- 
tion was done in July, 1892, and a third in January, 1893. The interval be- 4 
tween the recurrences became shorter and the rapidity of growth more 
rapid. 
On March 12, 1893, examination showed a tumor, two inches in diameter, 
attached to skin, but movable upon deeper parts, situated just above outer 
condyle of left femur, in region of old cicatrix. Just above the patella 
there was an induration two inches in area; a little farther up there was a 
distinct, separate nodule, % of an inch in diameter. 
The Toxic products of erysipelas were begun March 12, 1893, and 
given in the same way as in the two preceding cases. Much larger doses 
were given, and even then no appreciable rise of temperature followed. 
The only reaction that followed was severe headache, general muscular 
pain, and weakness. Locally, the tumors became swollen, even after the 
injections, and later decreased in size- 
While the filtered cultures (prepared without heat) were used the tumors 
not only did not increase in size but decreased considerably. The smallest 
one could scarcely be detected after six weeks. At the end of eight 
weeks, having no more of the filtered cultures on hand, cultures sterilized 
by heat were tried. A severe reaction, both general and local, followed. 
The temperature rose to 103.40 F., and the patient had asevere chill. The 
cultures were exactly the same as previously used, except in the method of 
sterilizing—no heat having been used in the filtered cultures. 
The injections were continued three weeks, but the tumors began to grow 
again, and on May 31, had attained the original size of March 12. The 
tumors were then removed by the knife, with the view of removing the 
larger part of the disease in the hope of destroying the remainder with ery- 
sipelas. The tumors removed were of the soft myxomatous variety, 
similar to case VIII1, loc. cit, when erysipelas produced only temporary 
improvement, therefore the prognosis is not so favorable. 
Case IV—Recurrent Sarcoma of Hand. This case, twenty-two years 
of age, a patient of Dr. Samuel Lloyd, had been operated upon several 
times for a sarcoma in the soft parts of the palm of the right hand. Local 
recurrence had followed every operation, and amputation of the arm had 
been advised by the surgeon and refused by the patient. 
The last operation was done in January, 1893. The tumor recurred 
within a few weeks and grew rapidly. Injections of the Toxic products of 
erysipelas were begun in March and continued up to the present time. The 
tumor not only ceased to grow but has diminished considerably in size. 
Case V—Recurrent Sarcoma of Axilla. The primary tumor was re- 
moved two years ago from the back. Recurrence followed in the axilla, 
and a second operation was performed in January, 1892. Local recurrence 
was observed in April, 1893, and grew rapidly, June 1, 1893, there was a 
flat, movable mass, 4j£ x 3 inches and 1 inch thick, occupying thoracic side 
of axilla at site of former cicatrix. Several nodules could also be detected 
in apex of axilla. 
The general condition of patient, who was a man forty-six years of age, 
was excellent. Injection of Toxic products of erysipelas, the same as in 
1 American Journal of Medical Science, May, 1893. 5 
previous cases was begun June i, 1893, and continued at intervals of 
twenty-four to forty-eight hours. Scarcely any reaction followed, although 
the doses were increased very much, and the Prodigiosus products were 
added to the erysipelas; 100.50 F, was the highest temperature reached. 
The injections were kept up only three weeks, and as no decided reaction 
was obtained, the tumors were removed by operation, as far as possible; a 
nodule was found firmly attached to the axillary vein and almost encircling 
it. This could not be removed without resecting the vein, which, con- 
sidering the improbability of being able to remove all the disease, was not 
thought advisable. 
Inoculation with a virulent culture of erysipelas will soon be made. 
While the Toxic products were being used the growth was apparently 
checked, although no diminution in the size of the tumor took place. The 
fact that so little reaction occurred in this case may have been due in part 
to the non-susceptibility of the patient; but the fact that the cultures from 
which the products are made had become attenuated by age, probably was 
the chief cause, 'l he cultures themselves as well as the Toxic products 
lose their strength rapidly with age. 
Case VI—Sarcoma of Foot atid Groin, The patient, a woman thirty - 
two years of age, had observed a small wart-like swelling upon the left in- 
step two years previously. It had twice been removed, but quickly re- 
turned. It remained very small, and never became larger than a cherry. 
In April, 1893, there was a tumor in. the left groin the size of two fists, 
protuberous and ulcerated over a large area. A hard swelling extended 
above Poupart’s ligament to within two inches of the umbilicus and to the 
median line in front. The ulcerating mass in the groin was removed by 
Dr. C. A. Powers at the New York Cancer Hospital, in April, 1893, and the 
following week injections of the Toxic products of erysipelas were begun. 
Good reaction followed doses of 2 c.c. The effect upon the tumors had not 
become noticeable when an attack of accidental erysipelas occurred. ’ The 
erysipelas ran the usual course. The attack was not severe. The effect upon 
the tumor was only a temporary check, and no marked decrease in size was 
observed. As soon as the attack had subsided the tumor grew very rapidly. 
The Toxic products were again given in large doses. The reaction 
gradually became less marked in spite of increasing doses. The largest 
dose was 3 c.c. of erysipelas products and .75 c.c. of Bacillus Prodigiosus 
products. The growth of the tumor was unmistakably retarded by the 
injections. They were discontinued for two weeks and then resumed. The 
short interval seemed to have restored the susceptibility, and good reaction 
followed. The case is still under treatment, but in view of the great size 
of the tumors when treatment was begun, nothing more than temporary 
improvement could be hoped for. 
The results from the use of the Toxic principles of erysipelas, in the 
limited number of cases described, are sufficiently encouraging to make 
further experiments exceedingly desirable, and there is good reason to be- 
lieve that most of the value of the erysipelas lies in these Toxic principles. 
Similar experiments have very recently been conducted by Dr. C. H. H. 6 
Spronck, Professor at the University of Utrecht, and published in the An- 
nales d LInstitut Pasteur. October, 1892. 
Spronck and his co-workers treated twenty-six cases of malignant tumors 
with a preparation of the Toxic products of erysipelas. 
Spronck, believing that the action of erysipelas upon tumors was for the 
most part systematic, made all of his injections remote from the tumors 
(usually in the gluteal regions). 
He prepared the Toxic products in the same way that Koch did his 
Tuberculin, viz., by heating liquid bouillon cultures to ioo° C, after the 
addition of 5 per cent glycerine, then evaporating to one-tenth the volume 
and finally filtering through porcelain. 
This preparation was first tried on animals In certain doses, varying 
with the virulence of the cultures from which it was derived, when injected 
into the circulation of a rabbit, caused death in from 24 to 48 hours. 
The more virulent the culture the more rapid the death and the smaller 
the fatal dose. 
A dose of 10 to 15 c. c. per kilogramme weight of animal usually caused 
death within 48 hours. The symptoms preceding death were elevation of 
temperature, loss of appetite, diarrhoea, albuminuria and paresis of poste- 
rior extremities. No changes worthy of note were seen at autopsy. 
Spronck then experimented upon tumors in dogs. Some of the tumors 
were benign, others malignant. Subcutaneous injections of the Toxic 
products were made in two cases at a distance from the tumors. The doses 
were repeated several times and increased in quantity yet no effect was 
observed upon the tumors. They were removed and examination showed 
the one to be a lipoma and the other a carcinoma (anus). 
With five other dogs, however, there was a more or less marked effect 
upon the tumors. After several injections (remote from tumors) the color 
of the tumors became of a bluish red and the tumors warmer and more sen- 
sitive to touch. Subsequently there occurred a softening and necrosis of 
the neoplasms. 
“In the two cases these latter changes manifested themselves with aston- 
ishing rapidity.” After the treatment had been continued a short time the 
remaining portion of the tumors were extirpated and examined. Two were 
found to be sarcoma and three carcinoma. In all there were found a con- 
dition of fatty degeneration and necrosis with an infiltration of polynuclear 
leucocytes from the blood. 
Spronck very properly admits that such changes frequently occur in 
malignant tumors, hence it would be difficult to prove the entire effect due 
to the injections, yet in view of the symptoms following the injections and 
the change in the tumors dating from the beginning of such injections, a 
causative relation seems indisputable. 
One of his experiments upon dogs is worthy of special note. A large dog 
had a very hard tumor the size of an infant’s head, situated in the region 
of the right shoulder. It was a local recurrence from a small tumor which 
had been extirpated the previous year. On the fourth day of the treatment 
the tumor developed a sliqht swelling, and at the same time it became 7 
of a bluish red color, warmer and more sensitive to touch. Shortly after- 
wards the consistence of the tumor diminished and on the eighth day fluc- 
tuation was present. Three days later the entire tumor was transformed 
into a soft fluctuating mass, covered only by the skin. Incision was fol- 
lowed by the escape of a voluminous sac filled with a puriform liquid in 
which were suspended necrosed fragments of tumors which proved on 
microscopic examination to be composed of detritus of the tumor and a 
large number of leucocytes. Somewhat later the dog died. 
The autopsy showed that of the entire tumor there remained only a 
small fragment the size of a pullet’s egg, where the tumoi had been adher- 
ent to the skin. Microscopical examination proved it to be adeno-carci- 
noma. There can be scarcely any doubt that these wonderful changes 
were produced by the Toxic products of erysipelas, and when we reflect 
that the tumor was carcinoma and not sarcoma, the significance is far 
greater. 
With the co-operation of Meulen, Renssen, Salzer and seven other physi- 
cians, Spronck extended his experiments to twenty-five inoperable malig- 
nant umois in man, eight sarcoma and the remainder carcinoma. The 
same preparation of Toxic products already described was used and the 
injections were made without exception remote from the tumor. The 
utmost care was used in these experiments and the doses were very small. 
The effect upon the tumors was by no means constant. 
In a certain number there was no appreciable change; in a second class 
they continued to increase in size but the growth appeared retarded; in a 
third class there was complete arrest of growth, and in a fourth diminu- 
tion in size. 
In the same patient, with several tumors, the tumors did not diminish 
with rapidity. While some disappeared entirely, a total disappear- 
ance ot all tumors present was not ooserved in a single case. In most of 
them the improvement was only temporary. The effect upon Carcinoma 
was much less marked than upon Sarcoma. 
Still in one case (recurrent cancer of the breast) there was produced an 
almost complete absorption of the lymphatic enlargements, though they 
afterwards returned. 
The most favorable effect was observed in a case of inoperable sarcoma 
treated by Renssen. After several weeks’ treatment the primary tumor, of 
very considerable size, completely disappeared, and the secondary tumors 
were much diminished in size. They were no larger than a pea. with one 
exception, and that the size of a pigeon’s egg. Failing to decrease further 
they were extirpated. Sufficient time had not elapsed to determine 
whether or not a radical cure had been obtained. The symptoms following 
the injections were very similar to those observed in my own cases and 
already noted. Elevation of temperature varying with the doses used, and 
with different subjects, was the most constant symptom. 
In most cases the rise in temperature began 4 or 5 hours after the injec- 
tions, never went beyond 103. i° and usually returned to normal at the end 
of 12 hours. In my cases the temperature has a few times risen beyond 
104°, but usually it has ranged between ioi° and 103.50. 
I have frequently observed a much more rapid onset, a chill occurring 
within a quarter of an hour after the injection, quickly followed by a sharp 
rise of temperature, nausea, vomiting, severe headache, general muscular 
pain, and malaise were more or less constantly seen. Spronck found that 
the relative susceptibility of the different patients varied greatly, some 
being entirely satisfactory, I have noted the same thing in my cases. Yet 
in one case that had failed to show any rise of temperature after frequent 
injections of my preparation of Toxic products (sterilized without heat) 
showed a marked rise, 103.4°, when cultures from the same source and of 
the same virulence, sterilized by heat, were used The important differ- 
ence, however, in the effect of the two preparations, was that the fimt 
although producing less constitutional reaction than the heated cul 8 
tures, caused a diminution in the size of the tumors (sarcoma) while the 
second failed to even check their growth. My experiments in this case 
together with others carried on in 1892, with cultures sterilized by heat, 
have convinced me that a portion at least of the value of the Toxic princi- 
ples is destroyed by a temperature of 100 C. 
Although I had practically arrived at this conclusion almost a year ago 
(before the publication of Spronck’s cases), I was unable to obtain a good 
prearation of Toxic products of erysipelas sterilized without heat until 
February, 1893. My results although they cover a smaller number of cases 
than Spronck’s are yet more remarkable. 
In one case the sarcomata have entirely disappeared under injections of 
Toxic products, and the patient is still free from recurrence. 
In a second case, a very large inoperable sarcoma of the abdominal wall, 
the tumor has so far diminished in size that only a small nodule remains. 
In two others there was decrease in the size, and they are still under treat- 
ment. The remaining cases are too recent to more than mention. In one, 
however, the improvement has been very doubtful. 
The elaborate investigations of Roger with the Toxic products of erysip- 
elas, upon animals, recently published in Rev. de Med., December, 1891, 
show that instead of a single Toxic principle the erysipelas germ grown in 
bouillon forms a considerable number of soluble products, some of which 
differ so widely from others that they are directly antagonistic. 
He further found that these Toxic products were capable of killing ani- 
mals in doses of 13 to 20 c. c. per kilogramme weight and that they were 
precipitated by alcohol and destroyed by heat. 
He also found that filtered cultures not heated in doses of 5 to 12 c. c., 
injected into the circulation of rabbits exercised a predisposing action, 
making the animal more susceptible to inoculation. On the other hand, 
heated cultures (104-110° C.) rendered the animals immune. 
In extremely large doses these products produced a more or less chronic 
form of Septicaemia, which caused death in a few hours to several weeks. 
Space will not permit a more detailed description of these experiments. 
They are cited to prove that in the soluble products of erysipelas we have a 
very powerful and complex combination of active principles; that some are 
directly antagonistic to others; and finally that heat has a very decided 
effect upon them. 
Aside from the difference in the method of preparation of the soluble 
products of erysipelas, my experiments differ from Spronck’s in the fact that 
he sought fora systemic action alone, whereas I have endeavored to obtain 
both systemic and local action by making the injections into the tumor or 
its immediate vicinity, and I believe my results show this to be the best 
plan. 
Spronck in conclusion says: “ If in man the results have been less satis- 
factory (than in dogs) one should not lose sight of the numerous difficulties 
that lie in the way of their first attempts towards a clinical application, or 
of the unfavorable conditions in which we have been placed, or finally of 
the fact that in nearly all of the patients the tumors were in a very 
advanced stage.” He further states: “ I believe that I am authorized to 
conclude that the products of erysipelas, in certain cases at least, can act 
in a curative manner upon malignant tumors.” 
While the treatment of malignant tumors by erysipelas inoculations or by 
injections of the Toxic principles of erysipelas is still in the experimental 
stage, we cannot but recognize that a long step in advance has been made 
and there is good ground for confidence that further investigation along 
the«e lines will be followed by even more brilliant results than those 
already obtained. 
18 East Thirty-second street.