Tolerability of bacille Calmette-Guérin maintenance therapy for superficial bladder cancer

ADULT UROLOGY

TOLERABILITY OF BACILLE CALMETTE-GUE´RIN MAINTENANCE THERAPY FOR SUPERFICIAL BLADDER CANCER FABIEN SAINT, JACQUES IRANI, JEAN JACQUES PATARD, LAURENT SALOMON, ANDRE´ HOZNEK, SYLVIE ZAMMATTIO, HENRI DEBOIS, CLAUDE C. ABBOU, AND DOMINIQUE K. CHOPIN

ABSTRACT Objectives. To study the influence of adverse reactions on adherence to an immunotherapy maintenance schedule and the recurrence rate of bladder cancer. Bacille Calmette-Gue´rin immunotherapy has documented efficacy in the management of high-risk superficial bladder cancer. However, the optimal duration of intravesical bacille Calmette-Gue´rin therapy and the risk/benefit ratio of maintenance therapy are controversial. Methods. From April 1996 to April 2000, 72 patients with superficial bladder cancer were treated with Immucyst (six consecutive weekly instillations of 81 mg) and then received maintenance therapy consisting of three consecutive weekly instillations 3, 6, 12, 18, 24, 30, and 36 months later. Adverse reactions, studied during 518 instillations, were classified in four categories using a scale based on the World Health Organization recommendations, and their impact on the adherence to therapy was analyzed. Results. After an average follow-up of 24 months, a durable disease-free response was observed in 84.9% of the patients; 12.5% of patients had a relapse and 2.6% had disease progression. The response rate was similar in patients with and without adverse reactions. Only 14 patients (19%) received all the scheduled maintenance instillations. The dose was reduced in 41 patients (57%), and treatment was stopped in 28 patients (39%). In multivariate analysis, an adverse event score of 1.5 or greater during induction therapy was significantly associated with cessation or modification of maintenance therapy (P ⫽ 0.01). Conclusions. The scale developed in this study to monitor the adverse reactions to bacille Calmette-Gue´rin and their impact on the adherence to maintenance therapy may be helpful for tailoring maintenance regimens or implementing protective measures (dose reduction or treatment postponement). UROLOGY 57: 883–888, 2001. © 2001, Elsevier Science Inc.

T

he efficacy of bacille Calmette-Gue´rin (BCG) in the treatment of superficial bladder cancer was first reported by Morales et al.1 in 1976. Several investigators have since demonstrated the efficacy of BCG in the prophylaxis and treatment of highrisk superficial bladder tumors (pT1, carcinoma in situ).2 Although BCG is now recommended as an adjunctive treatment for superficial bladder tumors, the optimal treatment schedule remains to be defined. The main differences among the proposed protocols concern the concomitant use of From the Service d’Urologie, Hoˆpital Henri Mondor, Cre´teil; and Pharmacovigilance, Aventis-Pasteur, Lyon, France Reprint requests: Dominique K. Chopin, M.D., Service d’Urologie, Hoˆpital Henri Mondor, 51 Avenue du Mare´chal de Lattre de Tassigny, 94000 Cre´teil, France Submitted: October 12, 2000, accepted (with revisions): December 6, 2000 © 2001, ELSEVIER SCIENCE INC. ALL RIGHTS RESERVED

intradermal BCG, the BCG strain, and the use of maintenance therapy.3–5 The results reported by Lamm et al.5 suggest that an initial induction cycle of six weekly intravesical BCG instillations is suboptimal unless maintenance therapy (three consecutive weekly instillations) is given 3, 6, 12, 18, 24, 30, and 36 months later. However, the use of maintenance therapy is hindered by troublesome adverse reactions.5,6 During a course of six weekly BCG instillations, approximately 5% of patients experienced severe adverse reactions and 30% to 80% had minor local or systemic reactions.7 Tolerability falls substantially with treatment intensification, but severe adverse reactions have also been observed after only a few instillations.8 Isoniazid has been proposed to prevent adverse reactions, but no difference in local or systemic events after intravesical immunotherapy with BCG was ob0090-4295/01/$20.00 PII S0090-4295(00)01117-1 883

served in a randomized European Organization for Research and Treatment of Cancer trial of isoniazid protection.7 The frequency of complications differs from one BCG preparation to another (Armand Frappier, Tice, Connaught, and Pasteur strains). These differences are due not only to variations in the unit dose and treatment schedule but also to differences in the criteria used to assess the bladder symptoms and to quantify the adverse reactions.6 We evaluated the efficacy and tolerability of the maintenance schedule proposed by Lamm et al.5 by developing a scoring system of the adverse reactions. This system may prove useful for designing individually tailored maintenance schedules (dose reduction, treatment postponement) and for evaluating potentially protective drugs. MATERIAL AND METHODS PATIENTS From April 1996 to April 2000, 72 consecutive patients with a history of multifocal or recurrent papillary transitional cell carcinoma were enrolled in this study, with their informed consent. The average age of the 5 women and 67 men was 64 years (range 37 to 83). The patients were followed up for a median of 24 months (range 4 to 48). The inclusion criteria included recent complete resection of a histologically intermediate-risk (Stage Ta/T1, grade 1 and 2, n ⫽ 37) or high-risk (carcinoma in situ primary/secondary, n ⫽ 2; Stage Ta/T1 associated with grade 3 carcinoma in situ, n ⫽ 9; Stage Ta/T1, grade 3, n ⫽ 24) superficial bladder tumor.

MYCOBACTERIOLOGIC CONTACT STATUS The patients were tested with intermediate-strength purified protein derivative (PPD skin test) using 10 U tuberculin before entry into the protocol. A positive PPD skin test was defined as an area of induration at least 6 mm in diameter 72 hours after the test. Standard chest radiography, a tuberculosis vaccination questionnaire, and renal and hepatic biochemical tests were also performed before starting the BCG protocol.

TREATMENT PROTOCOL The instillation schedule was performed according to Lamm et al.5 The first induction cycle consisted of six consecutive weekly instillations of 81 mg BCG Connaught (Immucyst, Aventis Pasteur) dissolved in 50 mL physiologic saline solution and instilled using a 14F catheter, 3 to 6 weeks after the last transurethral resection of the bladder tumor. The maintenance therapy consisted of three consecutive weekly instillations given 3, 6, 12, 18, 24, 30, and 36 months after the beginning of the induction therapy, for a total of 21 maintenance instillations. Intradermal BCG inoculation was not given. Urine was cultured before each series of instillations. Patients were instructed to retain the solution for approximately 2 hours, to drink large amounts of fluid, and to avoid physical activity on the day of instillation. Patients were subsequently seen at 4-month intervals. The follow-up evaluations consisted of voided urinary cytologic analysis and cystoscopy. Transurethral resection and biopsy were performed for all overt or suspicious areas and when recurrence was suspected. The maintenance therapy was stopped after severe adverse reactions occurred and after the diagnosis of recurrence. 884

RESPONSE CRITERIA The responses were assessed on the basis of the histologic and cytologic results and were classified as complete when the cytologic and biopsy histologic findings were negative. Tumor recurrence was defined as the occurrence of transitional cell carcinoma without progression of stage. Tumor progression was defined as progression of stage, infiltration of muscle, or metastasis.

ADVERSE EVENT SCORE Using the World Health Organization’s adverse reaction terminology (Geneva 1979), the National Cancer Institute’s terminology, and a review of published reports on adverse events to BCG, we developed a four-class rating scale for local-regional and systemic adverse events (Fig. 1). Any symptoms were recorded by the patients on a self-administered questionnaire each day for 1 week after each instillation. Patients subjectively assessed the severity of the adverse reactions as mild, moderate, or severe. The questionnaires were returned at the next instillation appointment and the adverse effects documented on a standard form derived from events known to be associated with BCG instillation (Fig. 2).6,9,10 The questionnaires were used to define the classes of adverse reactions associated with each instillation, and the adverse reactions were summarized in the form of a scale index, classified on the basis of severity and duration. For each patient, an average adverse event score (sum of the adverse event classes observed after each instillation, divided by the number of instillations) was calculated for the six initial instillations (initial adverse event score, iAES) and for the entire protocol (total adverse event score, tAES).

STATISTICAL ANALYSIS Statistical analyses were performed with Stat-View software. The following tumor and patient variables were analyzed: age, sex, stage, grade, multifocality, prior intravesical BCG therapy, PPD test result, vaccination against tuberculosis, intercurrent urinary tract infections, BCG regimen (full dose, treatment stopped, or dose reduced because of adverse events), prior transurethral resection, iAES, and tAES. The time to recurrence was estimated using the Kaplan-Meier method. Cox’s proportional hazards regression model was used for univariate and multivariate analyses of the effects of covariables on the time to recurrence. Correlations between these variables were determined using the Spearman rank correlation. All P values were two-sided (P ⬍0.05 was considered significant).

RESULTS A complete response to BCG was observed in 61 patients (84.9%); 9 patients (12.5%) had a relapse and 2 (2.6%) progression. The recurrences were diagnosed an average of 18.4 months (range 4 to 48) after the initial transurethral resection, and the progressions occurred after 4 and 28 months. In multivariate analysis, disease-free status was not associated with age, sex, pathologic stage, tumor grade, vaccination against tuberculosis, PPD status before treatment, incidental urinary tract infections, iAES, tAES, or the duration of the maintenance protocol. Of the 72 patients enrolled and treated, 14 (19%) received the entire full-dose maintenance schedule. Treatment was stopped in 28 patients (39%), UROLOGY 57 (5), 2001

FIGURE 1. Adverse event scale index form. WHO ⫽ World Health Organization.

25 (35%) before the third maintenance course (instillation 13). Treatment was stopped most frequently at instillations 6 (3 patients), 9 (6 patients), and 12 (9 patients). The unit dose at each instillation was reduced in 41 patients (57%) because of adverse events (to 54 mg when severe adverse events occurred for the first time, 40 mg for the second time, and 27 mg for the third time). The dose decrements were most frequent at instillations 7 (8 patients), 8 (13 patients), and 9 (7 patients). Fourteen patients (19%) had a unit dose reduction before being withdrawn permanently from therapy, and 18 patients (25%) had at least UROLOGY 57 (5), 2001

two dose reductions. During maintenance therapy, all 72 patients had symptoms of toxicity (518 adverse-event questionnaires were evaluated). The most frequent local reaction was cystitis. Class I cystitis (lasting 2 to 48 hours) was noted in 46.7% of instillations, class II (48 hours to 7 days) in 38%, and class III (greater than 7 days) in 7%. Systemic adverse events were less frequent during maintenance therapy than during induction therapy and were principally fever and fatigability. During the BCG induction course (six weekly instillations), the severity of the adverse events associated with each instillation increased after each instillation, 885

FIGURE 2. Daily medical synoptic questionnaire.

reaching a peak at the sixth instillation. No class IV adverse events were noted. Class I reactions were most frequently reported after the fourth instillation (46% of patients had class IB and 11% class IA events). Class II reactions were most frequently reported after the fifth instillation (38% of patients had class IIB and 2% class IIA events). Class III adverse effects were most frequently reported after the sixth instillation (12% of patients had class IIIB and 2% class IIIA events). A similar pattern of increasingly severe adverse effects after successive 886

doses was observed during maintenance therapy. Three peaks of class II reactions occurred, at instillations 8 (47% of patients class IIB and 11% class IIA), 11 (44% class IIB and 10% class IIA), and 14 (36% class IIB and 13% class IIA); and three peaks of class III adverse events occurred, at instillations 9 (24% class IIIB and 2% class IIIA), 12 (12% class IIIB), and 14 (20% class IIIB). The iAES and tAES were independent of age, sex, pathologic stage, tumor grade, vaccination against tuberculosis, PPD status before treatment, and time between BCG inUROLOGY 57 (5), 2001

TABLE I. Factors predictive of tolerability of BCG maintenance therapy

Previous BCG treatment Yes No Intercurrent-related urinary infection Yes No Previous vaccination against BCG Yes No PPD skin test ⱖ6 mm ⬍6 mm iAES ⬍1.5 ⱖ1.5 Recurrence Yes No

P Value

Treatment Stopped (n)

Treatment Not Stopped (n)

1 (12) 36 (59)

0.01

5 (55) 23 (38)

4 (45) 38 (62)

NS

17 (55) 16 (41)

14 (45) 23 (59)

NS

10 (32) 18 (46)

21 (68) 21 (54)

NS

19 51

6 (31) 27 (53)

13 (69) 24 (47)

NS

6 (32) 22 (43)

13 (68) 29 (57)

NS

30 14

14 (47) 5 (36)

16 (53) 9 (64)

NS

10 (33) 4 (29)

20 (67) 10 (71)

NS

43 27

31 (74) 2 (7)

12 (26) 25 (93)

0.0001

11 (26) 16 (60)

31 (74) 11 (40)

0.01

11 59

6 (54) 27 (46)

5 (46) 32 (54)

NS

5 (46) 23 (39)

6 (54) 36 (61)

NS

Patients Evaluated (n)

tAES <1.5 (n)

tAES >1.5 (n)

9 61

8 (88) 25 (41)

31 39

P Value

KEY: tAES ⫽ total adverse event score; BCG ⫽ bacille Calmette-Gue´rin; NS ⫽ not significant; PPD ⫽ purified protein derivative; iAES ⫽ initial adverse event score. Numbers in parentheses are percentages.

stillations. Nosocomial lower urinary tract infections were noted in 6% of instillations (31 patients), and the tAES did not correlate with intercurrent urinary tract infection. BCG treatment before this protocol correlated with low iAES and tAES values (P ⫽ 0.01 and 0.007, respectively). In multivariate analysis, the iAES was the only factor predictive of the tAES during maintenance therapy (P ⫽ 0.0001) and was also the only factor predictive of treatment cessation (P ⫽ 0.01) (Table I). COMMENT Although the efficacy of BCG as an anti-bladdertumor agent is well documented, the durability of responses and the recurrence pattern have not been established with respect to the initial tumor stage and grade. Continued susceptibility to tumor recurrence in successfully treated patients was shown by Nadler et al.11 Although BCG efficacy should be enhanced by one or two additional courses,12,13 randomized trials have failed to demonstrate the efficacy of monthly maintenance instillations.14,15 Using a maintenance schedule of three consecutive weekly instillations every 3 to 6 months, Lamm et al.5 reported a long-term benefit, with an 83% 5-year survival rate. Although our study was not controlled and involved a relatively short follow-up, our results compare favorably with those of the Southwest Oncology Group (SWOG) 8507 trial, demonstrating that maintenance therapy yielded a significant reduction in the recurrence rate in a high-risk patient group, withUROLOGY 57 (5), 2001

out percutaneous BCG. Although several small trials have shown some benefit to adjunctive percutaneous BCG administration, an appropriately powered randomized trial has not yet been performed to settle this issue conclusively. Maintenance therapy was associated with a high frequency of adverse reactions, necessitating a dose reduction or treatment cessation in most patients, although this did not appear to affect efficacy.15–17 In our study, 19% of patients were given all scheduled instillations, similar to the SWOG 8507 trial (16%).5 Efforts to optimize the maintenance dose or administration schedule are hampered by a lack of consensus on the definition of adverse reactions. The variation in the reported frequency of adverse reactions to BCG may be partly caused by variations in the tumor burden, completeness of resection, and method of instillation and by the use of different definitions. Some adverse events related to BCG therapy may be indicative of efficacy. Some investigators have observed a positive prognostic impact of fever (greater than 37.5°C) on the time to first recurrence,18 and others have observed no such influence with bladder irritation, fever, history of tuberculosis, PPD skin test, or granulomatous findings in bladder biopsies.13,15,19 The use of statistical analyses based on adverse reaction scores may provide more relevant information on the relation between adverse reactions and BCG efficacy. Cystitis is the most common symptom associated with BCG instillation. In our study, all patients in the maintenance protocol complained 887

of cystitis. The symptoms generally resolved, without therapy, within 48 hours. Intolerance of maintenance therapy was mainly associated with localregional symptoms of irritation, which accounted for most treatment cessations. Fever is the most frequent systemic adverse reaction to BCG instillations, with low-grade fever affecting 28% to 60% of patients and moderate-grade fever affecting 9%.6 In our study, low and moderate-grade fever occurred in only 11% and 2% of instillations, respectively, during induction therapy (fourth and fifth instillations). The incidence of fever increased with successive maintenance instillations, but this systemic effect was never responsible for treatment cessation. These differences may be explained by differences in the monitoring of adverse reactions and by variations in unit doses and treatment schedules, and underline the need for a consensus classification of BCG-related adverse events. The iAES correlated with the tolerability of maintenance therapy, meaning that it may be suitable for tailoring maintenance therapy to individual cases. Adverse reaction scores can provide information different than that given by quality-of-life questionnaires, which are more closely related to patient satisfaction.20 Our failure to identify a relationship among adverse reactions, time between maintenance courses, and mycobacteriologic status (PPD skin test, tuberculosis vaccination, and previous BCG treatment) points to individual differences in the vulnerability to adverse reactions. The value of Nramp1 gene status as a marker of susceptibility to adverse effects of BCG should now be examined, in conjunction with efficacy, in clinical trials of BCG maintenance therapy for high-risk superficial bladder cancer.21 CONCLUSIONS A maintenance protocol of three consecutive weekly instillations every 3 to 6 months for 3 years prevented the recurrence of high-risk superficial bladder cancer, but adherence to the maintenance therapy was hindered by the increasing severity of adverse reactions with each instillation. We tested a new scoring system for adverse reactions during 518 instillations as a possible approach to individual tailoring of BCG maintenance therapy and to evaluating protective measures. REFERENCES 1. Morales A, Eidinger D, and Bruce AW: Intracavity bacillus Calmette-Gue´rin in the treatment of superficial bladder tumors. J Urol 116: 180 –183, 1976. 2. Herr HW, Pinsky CM, Whitmore WF, et al: Experience with intravesical bacillus Calmette-Gue´rin therapy of superficial bladder tumors. Urology 25: 119 –123, 1985. 3. Melekos MD: Intravesical bacillus Calmette-Gue´rin prophylactic treatment for superficial bladder tumors: results of a controlled prospective study. Urol Int 45: 137–141, 1990. 4. DeKernion JB, Huang MY, Lindner A, et al: The man888

agement of superficial bladder tumors and carcinoma in situ with intravesical bacillus Calmette-Gue´rin. J Urol 133: 598 – 601, 1985. 5. Lamm DL, Blumenstein BA, Crissman JD, et al: Maintenance bacillus Calmette-Gue´rin immunotherapy for recurrent Ta, T1 and carcinoma in situ transitional cell carcinoma of the bladder: a randomized Southwest Oncology Group study. J Urol 163: 1124 –1129, 2000. 6. Lamm DL, Van der Meijden APM, Morales A, et al: Incidence and treatment of complications of bacillus Calmette-Gue´rin intravesical therapy in superficial bladder cancer. J Urol 147: 596 – 600, 1992. 7. Vegt PDJ, Van der Meijden APM, Sylvester R, et al: Does isoniazid reduce side effects of intravesical bacillus CalmetteGue´rin therapy in superficial bladder cancer? Interim results of European Organization for Research and Treatment of Cancer protocol 30911. J Urol 157: 1246 –1249, 1997. 8. Brosman SA: Experience with bacillus CalmetteGue´rin in patients with superficial bladder carcinoma. J Urol 128: 27–30, 1982. 9. Van der Meijden APM: Practical approaches to the prevention and treatment of adverse reactions to BCG. Eur Urol 27(suppl 1): 23–28, 1995. 10. Steg A, Leleu C, Debre´ B, et al: Systemic bacillus Calmette-Gue´rin infection, “BCGitis”, in patients treated by intravesical bacillus Calmette-Gue´rin therapy for bladder cancer. Eur Urol 16: 161–164, 1989. 11. Nadler RB, Catalona WJ, Hudson MA, et al: Durability of tumor free response for intravesical bacillus CalmetteGue´rin therapy. J Urol 152: 367–373, 1994. 12. Sarosdy MF, and Lamm DL: Long term results of intravesical bacillus Calmette-Gue´rin therapy for superficial bladder cancer. J Urol 142: 719 –724, 1989. 13. Okamura T, Tozawa K, Yamada Y, et al: Clinicopathological evaluation of repeated courses of intravesical bacillus Calmette-Gue´rin instillation for preventing recurrence of initially resistant superficial bladder cancer. J Urol 156: 967–971, 1996. 14. Badalament RA, Herr HW, Wong GY, et al: A prospective randomized trial of maintenance versus nonmaintenance intravesical bacillus Calmette-Gue´rin therapy of superficial bladder cancer. J Clin Oncol 5: 441– 449, 1987. 15. Hudson MA, Ratliff TL, Gillen DP, et al: Single course versus maintenance bacillus Calmette-Gue´rin therapy for superficial bladder tumors: a prospective randomized trial. J Urol 138: 295–298, 1987. 16. Brosman SA: The use of bacillus Calmette-Gue´rin in the therapy of bladder carcinoma in situ. J Urol 134: 36 –39, 1985. 17. Gruenwald IE, Stein A, Rashcovitsky R, et al: A 12 versus 6-week course of bacillus Calmette-Gue´rin prophylaxis for the treatment of high risk superficial bladder cancer. J Urol 157: 487– 491, 1997. 18. Lu¨ftenegger W, Ackermann DK, Futerlieb A, et al: Intravesical versus intravesical plus intradermal bacillus Calmette-Gue´rin: a prospective randomized study in patients with recurrent superficial bladder tumours. J Urol 155: 483– 487, 1996. 19. Torrence RJ, Kavoussi LR, Catalona WJ, et al: Prognostic factors in patients treated with intravesical bacillus Calmette-Gue´rin for superficial bladder cancer. J Urol 139: 941–944, 1988. 20. Bo¨hle A, Balck F, Von Wietersheim J, et al: The quality of life during intravesical bacillus Calmette-Gue´rin therapy. J Urol 155: 1221–1226, 1996. 21. Kadhim SA, Chin JL, Batislam E, et al: Genetically regulated response to intravesical bacillus Calmette-Gue´rin immunotherapy of orthotopic murine bladder tumour. J Urol 158: 646 – 652, 1997. UROLOGY 57 (5), 2001