Intracavitary Bacillus Calmette-guerin in the Treatment of Superficial Bladder Tumors

THE JOURNAL OF UROLOGY

Vol. 116, August

Printed in U.S.A.

Copyright© 1976 by The Williams & Wilkins Co.

INTRACAVITARY BACILLUS CALMETTE-GUERIN IN THE TREATMENT OF SUPERFICIAL BLADDER TUMORS A. MORALES,* D. EIDINGER

AND

A. W. BRUCE

From the Departments of Urology, and Microbiology and Immunology, Queen's University, Kingston, Ontario, Canada

ABSTRACT

Patients with recurrent.superficial bladder tumors have been treated by vesical and intradermal administration of Bacillus Calmette-Guerin. The pattern of recurrence in 9 patients has been altered favorably. Although the findings are still preliminary they appear to hold promise of a new therapeutic approach to the treatment of a group of neoplasms for which effective therapy is still lacking. Four to 6 weeks after the last immunization cystoscopy was performed. Any areas suggesting the presence of tumor were biopsied, otherwise random samples were taken. Recheck cystoscopies were performed periodically thereafter.

Instillation of oncolytic agents has been used for many years in the treatment and prophylaxis of superficial bladder tumor recurrences with variable success. 1 • 3 The location and natural history of these neoplasms and the easy accessibility of the bladder make this type of therapy particularly attractive. The antigenicity of bladder tumors has been demonstrated repeatedly.•. 5 This would suggest that immunotherapy may be useful in the eradication of non-invasive bladder neoplasms. Successful Bacillus Calmette-Guerin (BCG) immunotherapy must meet several criteria: 1) ability to develop an immune response to mycobacteria antigens, 2) adequate numbers of living bacilli, 3) close contact between BCG and tumor, 4) relatively small tumor load and 5) freedom from major systemic side effects.• Superficial bladder tumors appear to be ideally suited to this approach. The results presented summarize our initial experience with the use of BCG in the treatment and prophylaxis of these neoplasms.

RESULTS

Delayed cutaneous hypersensitivity. All patients exhibited a cutaneous reaction to mycobacteria antigens. In 4 cases the administration of tuberculin caused no reaction but strong reactivity was obtained 3 weeks after administration of BCG. Response to other recall antigens was found in 6 patients, the most common being to streptokinase-streptodornase. All patients tested with dinitrochlorobenzene showed reactivity to this substance. Quantitative lymphocyte studies. The mean numbers of absolute peripheral lymphocytes and the T and B subpopulations are illustrated in figure 1. The initial values are not different from the ones found in normal subjects. This finding may well reflect the limited tumor load. Although an increase in lymphocyte populations was noted after the onset of therapy the values lack statistical significance. Recurrence rate. The number of recurrences found in the 12-month period immediately before BCG therapy and during the post-vaccination period is illustrated in table 1. Before therapy 9 patients demonstrated a total of 22 recurrences during 77 patient months. After vaccination these 9 patients yielded 1 recurrence during a followup of 41 patient months. However, these 2 total distributions do not permit a valid statistical analysis. On the other hand, looking at 5 cases in which the pre-BCG and post-BCG periods were identical (25 patient months), it was found that 12 recurrences were detected during the pre-BCG period, while after immunization no recurrences were present. With the chi-square test in these 5 patients for the pre-BCG versus the post-BCG periods, a statistically significant difference was obtained (p less than 0.01). Of the 9 patients who have now received BCG immunotherapy 5 were treated for prevention of recurrence and 4 for residual tumor.

MATERIALS AND METHODS

There were 2 groups of patients considered candidates for BCG immunotherapy. In group 1 were patients with a history of persistent tumor recurrences but in whom all gross evidence of cancer was eliminated by endoscopic fulguration prior to the onset of immunotherapy. In group 2 were patients with tumor recurrence in whom complete endoscopic eradication of the neoplasm was not achieved. In every case the tumor was clinically and histologically staged and considered to be superficial (To to Ti). Immunological evaluation was performed in vivo and in vitro. In vivo studies consisted of the determination of delayed cutaneous hypersensitivity to a battery of recall antigens: tuberculin, histoplasmin, dermatophytin and streptokinasestreptodornase. In vitro studies included absolute peripheral lymphocyte counts and determination of T and B subpopulations as previously described. 7 ECG administration. The vaccine was given by intradermal and per urethram routes. Five mg. BCG (lnstitut Armand Frappier, Montreal) was administered to the upper thigh using a multiple puncture apparatus. For the intracavitary administration 120 mg. of the vaccine reconstituted in 50 cc normal saline was injected into the bladder through a No. 8 urethral catheter. The patient was advised to ratain the fluid for not less than 2 hours. Treatments were repeated at weekly intervals for 6 weeks, alternating thighs for the intradermal administration.

CASE REPORTS

Accepted for publication October 31. 1975. Supported by the Ontario Cancer Foundation Grant 292. Requests for reprints: Etherington Hall, Queen's University, Kingston, Ontario, Canada. * Current address: Laboratory of Immunodiagnosis, Building 8, Room 118, National Institutes of Health, Bethesda, Maryland 20014. 180

Case 1. An 80-year-old man was found to have a transitional cell carcinoma of the bladder in 1970. Elimination of tumor was achieved by endoscopic fulguration but rechecks every 3 to 4 months demonstrated persistent recurrences. In June 1974 a course of intradermal and intracavitary BCG was decided upon after conversion to purified protein derivative positivity by intradermal injection of BCG. It was well tolerated although the patient experienced fever, malaise and dysuria for 48 to 72 hours after the vaccination. After immunotherapy no recurrences have been detected endoscopically and random biopsies

181

INTRACAVITARY BCG IN TREATMENT OF SUPERFICIAL BLADDER TUMORS

:woo 1800 1600 1400 1200

1000 800

y----- ----

600

--i:>--l

--------t -------Toial No. Cell, T- Cell~
400 -

200

®--4'

B-Celle - ·

I

,--·-----r-----,---, D

2

4

8

i" WEEKS or minus 1 standard error of T and B subpopulations from onset

TiME

FIG. l, Absolute numbers of circulating lymphocytes, therapy.

further persistent recurrences were found. After a second course of thio-tepa prevention of tumor recurrence was achieved for another 9 months. However, at the beginning of 1969 recurrences were again noted. A third course of thio-tepa proved totally unsuccessful. Multiple recurrences were found persistently at 3-month intervals. In June 1972 the patient was started on a fourth course of thio-tepa followed by monthly instillations of the drug. This regimen kept her free of tumor until December 1973 when new tumors were detected. Twelve instillations of epody! failed to prevent additional recurrences. In February 1975 the immunotherapy trial was begun. During therapy symptoms of cystitis and low grade fever developed for 2 days after each instillation, Three months after completion of the bladder instillations cystoscopy was done. showed areas of mucosa! inflammation and necrosis associated multinucleated giant cells type and 2, 6), 6 months later did not reveal tumor recurrence. man had a transitional cell carcinoma located at the vault of the bladder for which he underwent cystectomy in June 1971. For the next 18 months recurrences were detected at the bladder vault. a cardiac condition precluded a operation the patient received 5,500 rads cobalt which failed to eliminate the persistent tumor. A course of thio-tepa was equally unsuccessfuL In November 1973 the patient entered the BCG trial. Since of therapy no additional tumor has been found in bladder. Random biopsies have disclosed chronic inflammation with areas of necrosis and calcification. In May 1975 the last cystoscopy showed an area of necrosis in which tumor had been present formerly. The bladder mucosa was otherwise normal and this has been confirmed histologically. The patient has had no recurrence for TABLE

1. Effect of BCG on the rate of tumor recurrence Calendar of Cystoscopic Examinations

Pt.* Post-BCG

Pre-BCG

FIG. 2. Granulomatous reaction seen in biopsy of bladder. Arrows indicate giant cells.

74 (2):j:

2

have failed to show neoplastic changes in the mucosa (table 2, patient 4). The patient has been free of tumor for 13 months. Case 2_ A 54-year-old man with a strong family history of bladder cancer was found to have a tumor immediately above the left ureteral orifice. Because of the location and size of the neoplasm he was treated by open excision and diathermy. The tumor was superficial and well differentiated but random samples taken at a distance from the lesion showed epithelial dysplasia. At 3 months no recurrence was seen but the mucosa had a patchy "angry" looking appearance. Three months later 5 small recurrences were detected, 4 of which were fulgurized and 1 of which was biopsied, The patient then received irnmunotherapy. Ten weeks after termination of the trial cystoscopy was done. No evidence of tumor was found except for the presence of a lesion with the gross appearance suggestive of severe inflammation. The lesion, which was located at the site of the previous recurrence that had been dealt with biopsy only, was treated by excisional biopsy, Histological examination disclosed a granuloma with multinucleated giant cells, lymphocytes and plasma cell infiltration (fig_ 2, table 2, patient 5), Repeat cystoscopy 22 weeks after of therapy revealed no recurrence, Case 3. A 55-year-old woman was found to have a well differentiated superficial transitional cell carcinoma of the which was succ,es1;tu rem.oved

3

4 5 6

7

Jan.73 Pos. (1) Sept. 73 Pos. (2) July 74 Pos. (1) July 74 Pos. (1) Sept. 73 Pos. (1) Feb. 73 Pos. (3)

Jan. 75 Pos. (1) Oct. 73 Pos. (1) Sept. 74 Pos. (2) Jan.75 Pos. (1) Jan. 75 Pos. (4) Aug. 74 Pos. (3) Sept. 73 Pos. (l.)

May75 Ned J'an. 74 Pos. (i) Jan. 75 Neg. Apr. 75 Neg. May75 Neg. June 75 Neg. Nov. 74

Neg.

Aug. 75 Neg. Mar. 74 Died 75 June 75 Neg. July 75 Neg. Aug. 75 Neg. May75 Neg.

* Two patients with insufficient followup are omitted.

t Cystoscopy positive for tumor recurrence. :j: Numbers in parentheses indicate number of recurrences in each examination. § Cystoscopy negative for tumor recurrence. TABLE

2. Summary of biopsy data

Pt.

3

4 5 6

Post-BCG Biopsy Not done Marked inflammatory cell infiltrate, squamous cell carcinoma present Granulomatous reaction and fibrosis with inflammatory infiltrate, no evidence of malignancy Inflammatory reaction, epithelial dysplasia, no tumor present Granulomatous reaction with multinucleated giant cells, infiltration with lymphocytes and plasma cells, no tumor Focal granulomatous inflammation, giant cells a.nd histiocytes about areas of no tumor Inflammatory focal areas of necrosis and calcification, no

B

rendered her fiee of tun:.or for 9

Aug. 74 Pos. (2) June 73 Pos. (1) Feb. 74 Pos. (1) Sept. 74 Pos. (1) Nov. 74 Neg. Feb. 74 Neg. June 73 Pos. (2)

9

Chronic

inflA.mrnation, no tu111or r::o evidence of mBlio2:na11cy

i
182

MORALES, EIDINGER AND BRUCE

20 months, the longest followup since the initial discovery of the vesical lesion. DISCUSSION

Intracavitary therapy of bladder cancer has attracted particular interest over the years. The easy accessibility of the bladder permits the delivery of a high local concentration of therapeutic agents with concomitant low systemic effects and allows an objective evaluation of their efficacy. Unfortunately, the sensitivity of established tumors to chemotherapeutic agents is poor and prophylaxis of recurrence has met with a similar lack of success. BCG is known to produce immune stimulation"· 9 and has been found capable of inducing regression of some solid tumors. 1 <>- 12 As previously mentioned the effectiveness of the vaccine depends on several factors, including ability of the host to react to mycobacteria antigens, small tumor load, adequate numbers of living bacilli and close contact between them and the tumor cells.• Evidently, no other tumor, with the exception of intralesional injection of cutaneous neoplasms, fulfills the requirements of adequate dose and close, intimate contact between BCG and tumor as does superficial bladder cancer. While our clinical study was in progress Bloomberg and associates reported on the cellular infiltrates in the bladders of normal dogs receiving intracavitary BCG. 13 Although the animals did not bear tumors local injury had been induced. Their findings lend additional support to the view that intracavitary administration of BCG induces a vigorous inflammatory reaction that would appear to auger well for immunotherapy via this route. The preliminary results presented herein indicate that this form of treatment is able to alter the pattern in patients with persistent recurrences without inducing adverse or long-lasting side effects. Furthermore, the presence of necrosis and formation of granulomas are highly suggestive of tumor destruction. 14 These 2 findings, while not conclusive and requiring additional followup to establish long-term efficacy, are of promise for a neoplasm for which adequate therapy is lacking. The limited tumor load appears to be a crucial factor in determining the effectiveness of BCG therapy. We have treated 5 additional patients with limited metastatic disease and have not detected objective remission. Although other factors may be involved the large mass with associated immune depression and the lack of direct contact of the vaccine with the neoplasm seem to be the major limiting factors. All patients included in this trial were initially responsive to tuberculin or became so after administration of BCG. In addition, most exhibited delayed cutaneous hypersensitivity to 1 or more antigens. This fact may well reflect the localized nature of the tumor. Absolute peripheral lymphocyte counts and T and B subpopulations have been considered reliable parameters of immune reactivity. The limited extent of the disease in our patients correlates well with the lack of depression in peripheral lymphoid cell populations. Patients with metastatic tumors show not only an impaired cutaneous reactivity but also a progressive diminution in lymphocyte populations that parallels the advance of their neoplasm. 15 • 11 Immunotherapy has raised many great expectations in the past but a more sober attitude on its effectiveness is now emerging. The role it has to play in the destruction and prevention of bladder tumors certainly deserves serious consideration. Although a much larger patient population and a prolonged followup are needed to provide definitive answers the early evidence points to a facorable outcome for the use of BCG immunotherapy in superficial bladder cancer. Dr. F. O'Cleireachain aided with the histological studies and Miss Pauline Coppleson provided technical assistance.

REFERENCES

1. Veenema, R. J., Dean, A. L., Jr., Uson, A. C., Roberts, M. and Longo, F.: Thiotepa bladder instillations: therapy and prophylaxis for superficial bladder tumors. J. Urol., 101: 711, 1969. 2. Riddle, P. R.: The management of superficial bladder tumours with intravesical epodyl. Brit. J. Urol., 45: 84, 1973. 3. Durrant, K. R. and Laing, A.H.: Treatment of multiple superficial papillary tumors of the bladder by intracavitary yttrium-90. J. UroL, 113: 480, 1975. 4. Bubenik, J., Perlmann, P., Helmstein, K. and Moberger, G.: Immune response to urinary bladder tumours in man. Int. J. Cancer, 5: 39, 1970. 5. Catalona, W. J. and Chretien, P. B.: Correlation among host immunocompetence and tumor stage, tumor grade and vascular permeation in transitional carcinoma. J. Urol., 110: 526, 1973. 6. Zbar, B. and Rapp, H. J.: Immunotherapy of guinea pig cancer with BCG. Cancer, 34: 1532, 1974. 7. Jondal, M., Holm, G. and Wigzell, H.: Surface markers on human T and B lymphocytes. I. A large population of lymphocytes forming nonimmune rosettes with sheep red blood cells. J. Exp. Med., 136: 207, 1973. 8. Ferluga, J .: Increased cytolytic activity of a subcellular fraction from mouse liver after B.C.G. injection. Lancet, 2: 1476, 1973. 9. Gutterman, J., Mavligit, G., McBride, C., Frei, E., III and Hersh, E. M.: BCG stimulation of immune responsiveness in patients with malignant melanoma. Cancer, 32: 321, 1973. 10. Grant, R. M., Mackie, R., Cochran, A. J ., Murray, E. L., Hoyle, D. and Ross, C.: Results of administering B.C.G. to patients with melanoma. Lancet, 2: 1096, 1974. 11. Donaldson, R. C.: Chemoimmunotherapy for cancer of the head and neck. Amer. J. Surg., 126: 507, 1973. 12. Morales, A. and Eidinger, D.: Bacillus Calmette-Guerin in the treatment of adenocarcinoma of the kidney. J. Urol., 115: 377, 1976. 13. Bloomberg, S. D., Brosman, S. A., Hausman, M. S., Cohen, A. and Battenberg, J. D.: The effects of BCG on the dog bladder. Invest. Urol., 12: 423, 1975. 14. Hanna, M. G., Jr., Snodgrass, M. J., Zbar, B. and Rapp, H.J.: Histopathology of Mycobacterium bovis (BCG)-mediated tumor regression. Nat. Cancer Inst. Monogr., 35: 345, 1972. 15. Falk, R. E., Mann, P. and Langer, B.: Cell-mediated immunity to human tumors. Abrogation by serum factors and nonspecific effects of oral BCG therapy. Arch. Surg., 107: 261, 1973. 16. Morales, A. and Eidinger, D.: Immune reactivity in renal cancer: a sequential study. J. Urol., 115: 510, 1976. 17. Papatestas, A. E. and Kark, A. E.: Peripheral lymphocyte counts in breast carcinoma. An index of immune competence. Cancer, 34: 2014, 1974.

COMMENT The authors are to be congratulated on their courage as well as their results. As I understand the thesis after BCG was given to immunoreactive patients and then instilled in their bladders the tumor was either destroyed or the incidence of new tumors was profoundly diminished. The basis for this lies presumably in the assumption that cell-mediated toxicity against BCG, heightened by injections in the thigh, will produce similar reactions in the bladder but only when tumor exists. One must assume that tumor-associated (or specific) antigens are identified with BCG because the tumor and only the tumor allows BCG contact with surveillance cells. It also suggests that tumor-associated or specific antigens are in some way released as BCG contacts the tumor. Thus, the surveying cells, identifying both antigens, cause heterogeneous response to be developed. Since I cannot understand the treatment protocol (the number and periodicity of BCG injection/instillations) and ancillary data to support the thesis further are lacking, I must conclude that the observations are largely empiric. This is not to say that I doubt either their importance or their validity. Experimental animal models and human therapy for cutaneous neoplasms have depended on intralesion injections and the local response can be most vigorous and debilitating. I would have probably taken sterile but dead autochthonous tumor and injected it at least once with the BCG in the patient's thigh. Let us hope the authors will provide us with more immunological