- Top of page
Objectives: We carried out a prospective, randomized, controlled trial to investigate the efficacy and safety of both induction and maintenance therapy with intravesical instillation of bacillus Calmette-Guérin (BCG) for high-risk non-muscle invasive bladder cancer (NMIBC).
Methods: Intravesical instillation of 80 mg Tokyo strain was given to patients with high-risk NMIBC, including carcinoma in situ (CIS), once weekly for eight consecutive weeks as induction therapy. Patients who achieved complete response (CR) were randomly assigned to either the maintenance group or the observation group.
Results: A total of 90 patients were enrolled. After induction therapy, 75% of the patients achieved CR and 53 of them were enrolled in the randomized comparative phase. A total of four maintenance instillations were given. Median follow-up was 26.5 and 28.7 months after randomization in the maintenance and the observation group, respectively. Although it was not statistically significant, the 2-year recurrence-free survival rate in the maintenance group (95.8%) was higher than that in the observation group (74.1%, P = 0.078). Univariate analysis identified maintenance therapy as a significant factor influencing recurrence. During induction therapy, 82.2% of patients experienced urination-related adverse drug reactions, but most events were not serious. There were fewer adverse drug reactions with maintenance therapy than with induction therapy. Neither induction therapy nor maintenance therapy reduced patients’ quality of life (QOL).
Conclusions: These findings show high levels of efficacy and safety of BCG induction treatment for high-risk NMIBC, and suggest that the number of maintenance instillations could probably be reduced without reducing treatment efficacy or influencing QOL.
- Top of page
Intravesical instillation of bacillus Calmette-Guérin (BCG) has been reported to have an advantage over intravesical chemotherapy in terms of therapeutic benefit and prophylaxis of non-muscle invasive bladder cancer (NMIBC).1,2 A meta-analysis study also showed that this treatment method reduced the risk of progression of NMIBC.3 Intravesical instillation of BCG is currently regarded as a standard modality for the treatment of high-risk Ta, T1 bladder cancer and carcinoma in situ (CIS) of the bladder, as well as for the prophylaxis of recurrence after the resection of high-risk Ta, T1 bladder cancer.4,5 The South West Oncology Group (SWOG) randomized clinical trial showed that 5-year recurrence-free survival and 5-year progression-free survival were higher in patients who received BCG induction therapy followed by maintenance therapy than those who received only BCG induction therapy.6 A meta-analysis showed that only BCG instillation as maintenance therapy prevented the progression of NMIBC.7 Thus, BCG maintenance therapy is recommended to prevent recurrence or progression and to prolong survival in patients with high-risk NMIBC.4,5 In the SWOG randomized study, the maintenance period was longer than two years and a total number of 21 maintenance instillations were given, although a 3-week maintenance schedule seems to be regarded as the standard maintenance instillation method. This might be one of the reasons why only a small number of patients completed the maintenance BCG instillation. It shows that efforts should be made to develop another BCG maintenance schedule.
In this time, we carried out a prospective, randomized, comparative study to investigate the efficacy and safety of both induction and maintenance BCG therapy, as well as another schedule of maintenance instillation for high-risk Ta, T1 cancer and CIS of the bladder.
- Top of page
This is a multicenter, prospective, open-label, randomized, comparative study. The present clinical trial protocol was approved by the institutional review board of each participating institution. Patients with histologically-confirmed Ta, T1 transitional cell carcinoma or CIS of the bladder were eligible to participate in the present study. In the case of Ta and T1, the tumor had to be diagnosed as not completely resectable by transurethral resection.
Other inclusion criteria were: (i) performance status (PS) of 0 to 2; (ii) age between 20 and 79 years; (iii) able to undergo assessment, such as cystoscopy and urinary cytology, on a regular basis; (iv) presence of at least one cystoscopically measurable or evaluable lesion in the case of unresectable NMIBC; and (v) intact function of the main organs (i.e. bone marrow, liver, kidney, heart and lung) Exclusion criteria were: (i) strongly positive tuberculin reaction or active tuberculous lesion; (ii) history of BCG intravesical instillation therapy; (iii) severe bladder irritation before the start of drug administration; (iv) intravesical instillation therapy with an anticancer drug within 3 weeks before the start of BCG administration; (v) history of bladder-sparing treatment for invasive bladder cancer; (vi) history of upper urinary tract carcinoma; and (vii) active double cancer or serious medical complications. Before treatment, patients provided a history and underwent physical examination, urinalysis, urine cytology examination, complete blood count, blood urea nitrogen, serum creatinine, liver function tests and electrocardiography. Chest X-ray and excretory urography were also carried out.
All eligible patients had to receive a full explanation of this study and sign an informed consent form before registration by facsimile to the central registration center. Induction therapy with 80 mg of BCG Tokyo strain started within 4 weeks after bladder biopsy or TUR for histological diagnosis. Patients were required to hold their urine for approximately 2 h after instillation. BCG treatment was repeated weekly for eight consecutive weeks. In the case of adverse drug reactions (ADR), administration could be delayed, but not for longer than 3 weeks from the scheduled administration day. Efficacy was evaluated at 4–12 weeks after the completion of this induction therapy. Patients who achieved CR on the induction therapy were required to receive a full explanation of the randomized study before re-registration at the same central registration center as for the induction therapy. The registered patients were randomly assigned to either the maintenance or the observation group. Maintenance therapy involved a single intravesical instillation of BCG Tokyo strain 80 mg within 3 months after the randomization, followed by instillations at 3, 6 and 9 months after the initial dose (a total of 4 doses). The dosing interval of maintenance therapy was set as 3 months based on the approved regimen of a similar drug that is already on the market in the USA. The total number of maintenance doses was determined as follows; the maximum number of doses was considered to be 12 which is 1.5 times the approved regimen of BCG Tokyo strain in Japan and because eight doses would be given as induction therapy, an additional four doses were to be given as maintenance therapy.
Patients in both groups underwent urinalysis, urinary cytology and cystoscopy 2 months after randomization. These examinations were repeated for observing the presence or absence of recurrence every 3 months for the first 3 years, and thereafter every 6 months. Clinical chemistry examination was carried out before the induction therapy, and before the fifth instillation and 2 weeks after the eighth instillation of induction therapy. In the maintenance group, clinical laboratory tests were carried out at 2, 5, 8 and 11 months after randomization. ADR were monitored at every visit and assessed according to the Common Terminology Criteria of Adverse Events version 3.0 (CTC AE v3.0). ADR that were difficult to grade by CTC AE, such as pain on urination and difficulty to urinate, were classified into grade 1 (mild), grade 2 (moderate) and grade 3 (severe). Whenever disease progression was suspected, imaging tests such as CT were carried out. Quality of life (QOL) was assessed according to the Japanese version of the European Organization for Research and Treatment of Cancer Quality of Life Questionnaire-Core 30 (EORTC QLQ-C30) v2.08 before the induction therapy, after the fifth instillation of induction therapy, 4 weeks after the completion of induction therapy and 14 months after randomization. Before using the QOL scale, authorization was acquired by the prescribed procedure.
The primary endpoint was the efficacy of induction therapy, and the secondary endpoints were recurrence-free survival, progression-free survival, overall survival, QOL and the frequency and severity of ADR in both groups. The efficacy was assessed according to the third edition of the General Rules for Clinical and Pathological Studies on Bladder Cancer.9 For Ta and T1 cancer, CR was defined as the complete clinical disappearance of target legions with negative biopsy and urine cytology. For CIS, CR was defined as negative biopsy and urine cytology. Recurrence was examined with cystoscopy and urine cytology, and then confirmed histologically. Disease progression was defined as the emergence of muscle invasive cancer or distant metastasis. The last observation day was 31 March 2007, when the present study was closed.
The sample size was determined as follows: CIS is the rarest type, representing just 4% of bladder cancer. Because there have been few reports of maintenance therapy given after BCG treatment for bladder cancer, as in the present study, it was considered difficult to determine the sample size from a statistical viewpoint. Therefore, the sample size was set as 30 patients in each group (a total of 60 patients) in the comparative phase of the study, taking feasibility into consideration. In a previous clinical study using BCG Tokyo strain, the CR rate in partly unresectable Ta, T1 bladder cancer and CIS of the bladder was 70.1%.10 If the number of patients required for random assignment is 60, then 82.3 patients must be enrolled in induction therapy, taking an estimation error of 10% into account. Taking patients who withdrew from the study and those who refused to participate in the comparative phase of study into consideration, the number of patients enrolled was set as 90.
The significance of differences in patient background factors between the two groups was examined by the χ2-square test. Recurrence-free, progression-free and overall survival curves were calculated by the Kaplan–Meier method and statistical significance was determined by the log–rank test. The Cox proportional hazard model was used to carry out univariate and multivariate analysis. QOL was evaluated using EORTC QOL-C30 scoring manual and examined by the paired t-test. Differences were considered to be significant at P < 0.05.
- Top of page
A total of 90 patients were enrolled between October 2002 and June 2005. The characteristics of enrolled patients are shown in Table 1. As shown in Figure 1, efficacy was evaluated for 84 patients. For the induction therapy, 6 of 10 patients with Ta and T1 cancer and 57 of 74 patients with CIS of the bladder achieved CR, resulting in CR rates of 60.0% and 77.0%, respectively. The overall CR rate was 75.0%. Among 63 patients who achieved CR, 53 patients were enrolled in the maintenance therapy phase and randomized: 26 to the maintenance group and 27 to the observation group. Two patients randomized to the maintenance group withdrew their consent before the start of maintenance instillation (Fig. 1). There were no significant differences in patient characteristics between the two groups (Table 2). The number of doses given to 24 patients randomized to the maintenance group was one in two patients (8.3%), two in three patients (12.5%), three in one patient (4.2%) and four in 18 patients (75.0%), showing a good completion rate of the maintenance regimen. There was not a significant difference in the follow-up period between the two groups. Recurrence was confirmed in one patient in the maintenance group and seven patients in the observation group, which showed 2-year recurrence-free survival rates of 95.8% and 74.1%, respectively. This indicates a lower recurrence rate in the maintenance therapy than in the observation group (Fig. 2), but the difference between the groups was not significant (P = 0.078). Disease progression was not seen in the maintenance group and was seen in one patient in the observation group. There was no significant difference in progression free survival between the groups (P = 0.383). Two patients each in the maintenance group and the observation group died during the study. One of the two in the observation group died of bladder cancer. Two-year overall survival rate was 91.7% in the maintenance group and 92.6% in the observation group, without a significant difference between the groups (P = 0.885). When the independent contribution of each background factor to recurrence was examined using univariate analysis, maintenance therapy was identified as a significant predictor (Table 3), but when multivariate analysis was used, it was not an individually significant factor influencing recurrence (P = 0.0676).
Table 1. Patient characteristics
| Younger than 70 years||39|
| 70 years or older||45|
|Performance status|| |
| 1, 2||4|
|Tumor status|| |
| Primary tumor||60|
| Recurrent tumor||24|
| Ta,T1 tumor||10|
|Smoking habit|| |
Table 2. Patient characteristics of the maintenance and observation groups
| ||Maintenance group||Observation group||P-value|
|Sex|| || || |
| Female||5||6|| |
|Age|| || || |
| Younger than 70 years||9||14||0.3039|
| 70 years or older||15||13|| |
|Performance status|| || || |
| 1, 2||1||2|| |
|Tumor status|| || || |
| Primary tumor||17||20||0.9558|
| Recurrent tumor||7||7|| |
|Stage|| || || |
| Ta,T1 tumor||3||2|| |
|Smoking habit|| || || |
| Yes||15||20|| |
|Median follow-up period||26.5||28.7||0.7199|
Figure 2. Recurrence-free survival curves after randomization into two groups. A lower recurrence rate was observed in the maintenance group than in the observation group, but the difference between the two groups was not significant (P = 0.078). , Maintenance group (n = 24); , observation group (n = 27).
Download figure to PowerPoint
Table 3. Univariate and multivariate analysis of factors influencing recurrence
|Variables||Hazard ratio||95% Confidential interval||P-value|
|Univariate analysis|| || || |
|Age (70 years or older/younger than 70 years)||0.4951||0.1183–2.0719||0.3264|
|Smoking habit (yes/no)||4.1639||0.5115–33.9003||0.1159|
|Tumor status (primary/recurrent)||0.3291||0.0405–2.6770||0.2354|
|Multivariate analysis|| || || |
|Group (observation/maintenance)|| || || |
| Adjusted for smoking habit||5.1078||0.6254–41.7154||0.0676|
Safety was evaluated in 90 patients. Table 4 shows the ADR with frequency ≥10% or grade ≥3 during the induction therapy. Although urination-related local symptoms occurred in 82.2% of patients, grade 3 urinary events had low frequencies. The frequencies of ADR not related to urination were low, except for pyrexia (38°C or higher), with an incidence rate of 30.0%. These ADR resolved with/without anti-inflammatory agents or corticosteroid preparations. Table 4 shows ADR with a frequency ≥10% or grade ≥3 in the maintenance group. These ADR also resolved with/without anti-inflammatory agents. The frequency of ADR during maintenance therapy was lower than that during induction therapy.
Table 4. Adverse drug reactions during the induction therapy and adverse drug reactions in the maintenance group
|Adverse event||Frequency (%)||Grade|
|Grade 1||Grade 2||Grade 3|
|Induction therapy|| || || || |
| Pain on urination||82.2||52.2||25.6||4.4|
| Urinary frequency||82.2||47.8||27.8||6.7|
| Gross hematuria||72.2||62.2||8.9||1.1|
| Difficulty with urination||52.2||46.7||4.4||1.1|
| Fever (38°C or higher)||30.0||27.8||2.2||0.0|
| Muscle pain||1.1||0.0||0.0||1.1|
| ALT elevation||7.8||4.4||2.2||1.1|
| AST elevation||7.8||6.7||0.0||1.1|
| gamma-GTP elevation||12.2||11.1||0.0||1.1|
|Maintenance group|| || || || |
| Urinary frequency||20.8||8.3||4.2||8.3|
| Pain on urination||16.7||8.3||4.2||4.2|
According to EORTC QLQ-C30, higher scores on the functioning scales and lower scores on the symptom scales indicate a better QOL. There were no significant differences in the QOL scores during induction therapy for all categories (data not shown). QOL on the emotional functioning scale was improved significantly (P = 0.004, data not shown). Most symptom scores tended to improve, although significant differences were not found. QOL after the induction therapy was also compared with that at 14 months after randomization in each group (Table 5). In the maintenance group, none of the functioning scales showed a reduction in QOL after the maintenance therapy compared with before. All the symptom scales, except the nausea and vomiting scale and the dyspnoea symptom scale, indicated an improvement in QOL after BCG treatment compared with before, although none of the improvements showed a significant difference. In the observation group, a significant difference in QOL after randomization compared with that before randomization was not observed.
Table 5. Quality of life scores of the maintenance and the observation groups
|Scale||Maintenance group||Observation group|
|After induction therapy||14 months after randomization||P-value||After induction therapy||14 months after randomization||P-value|
|Physical functioning scale||90.0 (14.7)||95.8 (6.4)||0.145||89.5 (16.6)||89.8 (18.5)||0.945|
|Role functioning scale||89.4 (22.7)||96.9 (9.1)||0.222||88.5 (14.0)||92.1 (22.7)||0.508|
|Emotional functioning scale||89.4 (11.8)||88.0 (12.9)||0.736||88.1 (11.1)||85.3 (15.8)||0.476|
|Cognitive functioning scale||84.9 (17.0)||84.4 (16.6)||0.932||80.1 (14.2)||76.2 (19.4)||0.426|
|Social functioning scale||88.6 (22.7)||92.7 (17.2)||0.55||89.7 (15.0)||91.3 (22.7)||0.784|
|Global health status||68.2 (22.1)||77.6 (17.4)||0.165||66.0 (17.6)||63.9 (21.5)||0.709|
|Fatigue symptom scale||17.7 (17.0)||16.7 (15.2)||0.851||19.2 (14.9)||20.6 (21.5)||0.793|
|Nausea and vomiting scale||0.0 (0.0)||1.0 (4.2)||0.246||0.0 (0.0)||0.0 (0.0)||—|
|Pain symptom scale||12.9 (23.0)||5.2 (10.0)||0.22||18.0 (25.4)||8.7 (22.7)||0.201|
|Dyspnoea symptom scale||9.1 (15.2)||10.4 (16.0)||0.796||9.0 (17.8)||7.9 (14.6)||0.83|
|Insomnia symptom scale||13.6 (24.5)||12.5 (16.7)||0.873||16.8 (19.4)||9.5 (15.4)||0.177|
|Appetite loss symptom scale||6.1 (13.2)||2.1 (8.3)||0.295||3.9 (10.9)||4.8 (12.0)||0.785|
|Constipation symptom scale||16.7 (30.4)||14.6 (27.1)||0.829||18.0 (27.1)||7.9 (14.6)||0.134|
|Diarrhoea symptom scale||4.6 (11.7)||4.2 (11.4)||0.921||2.6 (9.1)||1.6 (7.3)||0.691|
|Financial difficulties symptom scale||13.6 (16.8)||10.4 (16.0)||0.555||6.4 (13.4)||6.4 (17.1)||0.989|
- Top of page
The present study showed a CR rate of 77.0% for CIS of the bladder and 60.0% for Ta and T1 bladder cancer, resulting in an overall CR rate of 75.0%. This result was similar to that of a previous Japanese report.10 Although BCG intravesical instillation is often associated with undesirable adverse reactions, it was reported to be a highly safe treatment causing few serious ADR.11 A similar ADR profile was seen in the present study. The frequency of grade ≥3 ADR was low and there were no serious ADR resulting in death.
In the present study, only patients who had achieved CR with the induction therapy were randomly assigned to the maintenance or the observation group. This design allowed the advance exclusion of those who failed to achieve CR with induction therapy and the even assignment to the two groups, as shown in Table 2. In contrast, the design requiring reacquisition of consent before randomization might constitute a selection bias, because not all patients who have achieved CR will provide consent for participation again. Concerning the study design of clinical trials of BCG maintenance instillation, further investigation might be needed.
Although recurrence was confirmed in one patient in the maintenance group and seven patients in the observation group, there was no significant difference in recurrence-free survival between the groups. Univariate analysis showed that maintenance therapy reduced recurrence significantly, but multivariate analysis did not show a significant difference. This might be a result of the small sample size and the existence of various confounding factors.
According to the results of the SWOG randomized clinical trial,6 a 3-week maintenance schedule seems to be regarded as the standard maintenance instillation method. Decobert et al. retrospectively studied the relationship between the number of cycles of 3-week maintenance and recurrence12 and suggested that a minimum of three maintenance cycles of BCG (nine instillations) was required to significantly improve the no recurrence rate. In the present study, however, the total number of maintenance instillations was set as four instillations, because this study had to be carried out by an approved method for BCG Tokyo strain in Japan. Surprisingly, maintenance consisting of just four instillations decreased the recurrence rate as compared with that in the observation group, and was also identified as a significant factor in preventing recurrence using univariate analysis, although the present study failed to show significant superiority of the 4-instillation maintenance schedule. The SWOG 8507 trial is the only trial in which efficacy of maintenance therapy had been proved to be significant.6 This maintenance schedule consists of 81 mg Connaught strain by the 3-week maintenance method at 3- or 6-month intervals, and a total of 21 instillations. Akaza et al.10 carried out a prospective randomized trial using 40 mg Tokyo strain, and 107 patients were randomly assigned to the observation group and the maintenance group which consisted of 40 mg BCG monthly 12 times. There was no significant difference in the 3-year non-recurrence rate. In the present study, the maintenance schedule consisted of 80 mg Tokyo strain at 3-month intervals and a total of four instillations. These findings show that a 3-week maintenance method might be a critical factor for maintenance and that 3-monthly “1-week maintenance” might not be adequate to achieve an anticancer effect. This result, however, shows that it might be possible to decrease the number of maintenance instillations and further investigation is needed.
The completion rate of maintenance therapy in the SWOG trial was 16%,6 whereas another study reported that only one of 111 patients completed seven cycles of maintenance therapy.12 In contrast, a high completion rate of 75% was achieved in the present study. This is not only attributable to the small number of instillations for maintenance therapy, but also to a lower frequency and severity of ADR occurring during maintenance. In the EORTC prospective study, the majority of discontinuations as a result of adverse events were seen during induction therapy and the first 6 months of maintenance therapy, suggesting that BCG maintenance does not necessarily increase the occurrence of adverse events.13 Patients who show good compliance to BCG treatment during the induction therapy can receive maintenance therapy safely.
Concerning cancer treatment, it is essential to evaluate treatment from the patient's point of view, besides the clinical outcome, and therefore QOL assessment was incorporated in the present study. Induction therapy did not deteriorate QOL and the emotional functioning scale improved significantly. It seems that patients’ good acceptance of their disease state and treatment using BCG improved their own QOL. Furthermore, BCG maintenance did not deteriorate the QOL, which might be the reason for the good completion rate of maintenance treatment.
In conclusion, the present study, despite having a small scale, is a meaningful prospective investigation suggesting both the usefulness of BCG maintenance in patients with high-risk NMIBC and the probability that a schedule with fewer maintenance instillations would still be effective. The maintenance therapy prevented recurrence with fewer ADR and no deteriorations in QOL. Our future task is to continue exploring both the benefit and the optimal treatment schedule of maintenance therapy.