Abstract
Background/Aim: Intravesical bacillus Calmette–Guérin (BCG) therapy remains the standard adjuvant treatment for high-risk non–muscle-invasive bladder cancer (NMIBC), yet real-world maintenance delivery is frequently incomplete. Contemporary benchmarks that link BCG exposure patterns to clinically meaningful outcomes are needed. We evaluated real-world BCG treatment delivery and oncological outcomes in a tertiary cancer center cohort.
Patients and Methods: We retrospectively analyzed 120 consecutive patients with NMIBC who initiated intravesical BCG between June 2019 and May 2025. The institution-defined minimum BCG exposure was induction (≥5/6 instillations) plus initial maintenance (≥2/3 instillations) or re-induction (≥2/6 instillations). High-risk recurrence-free survival (HR-RFS) was defined as time from BCG initiation to high-risk NMIBC recurrence or pathological upstaging. Progression-free survival (PFS), metastasis-free survival (MFS), and overall survival (OS) were estimated using the Kaplan–Meier method.
Results: Median age was 73 years; pT1 disease occurred in 45.8%, concomitant carcinoma in situ in 29.2%, and high-grade tumors in 98.3%. Induction (≥5/6) was delivered in 94.2% of patients, and an initial maintenance cycle (≥2 instillations in cycle 1) in 68.3%; 73.3% achieved the minimum BCG exposure. Only 2.5% completed maintenance through cycle 7. At a median follow-up of 33.5 months, median HR-RFS, PFS, MFS, and OS were not reached. One-, three-, and five-year HR-RFS rates were 91.4%, 76.7%, and 73.0%; corresponding PFS rates were 98.3%, 92.5%, and 92.5%; MFS rates were 99.2%, 93.3%, and 93.3%; and OS rates were 98.3%, 97.3%, and 86.2%.
Conclusion: These data provide contemporary real-world benchmarks for outcomes achieved with conventional intravesical BCG in high-risk NMIBC when long-term maintenance completion is uncommon.
- Non–muscle-invasive bladder cancer
- bacillus Calmette–Guérin
- intravesical therapy
- maintenance therapy
- real-world outcomes
- recurrence-free survival
Introduction
Non–muscle-invasive bladder cancer (NMIBC) accounts for approximately 70%-75% of newly diagnosed bladder cancer cases worldwide, and it is characterized by marked heterogeneity in oncological outcomes, ranging from indolent disease to aggressive tumors with a high risk of recurrence and progression (1-3). Despite treatment by transurethral resection of bladder tumor, recurrence occurs in a substantial proportion of patients, and progression to muscle-invasive disease remains a critical concern, particularly in high-risk NMIBC (4, 5).
For several decades, intravesical bacillus Calmette–Guérin (BCG) therapy has been the cornerstone of adjuvant treatment for high-risk NMIBC and it remains the standard of care across major international guidelines, including those of the European Association of Urology (EAU), the American Urological Association (AUA), and the Japanese Urological Association (JUA) (1, 6, 7). BCG therapy aims to reduce tumor recurrence and progression while preserving bladder function, and no alternative intravesical therapy has consistently demonstrated superior long-term oncological outcomes in this setting. However, accumulating evidence indicates that BCG induction therapy alone is insufficient to achieve durable disease control in many patients. Randomized clinical trials, most notably the Southwest Oncology Group (SWOG) 8507 study, demonstrated that the addition of maintenance BCG therapy improved recurrence-free survival compared with induction therapy alone (8). Consequently, maintenance BCG therapy has been incorporated into guideline recommendations, particularly for patients with high-risk disease.
Despite these recommendations, the real-world implementation of BCG maintenance therapy remains highly variable. Differences in induction schedules, maintenance intensity, treatment duration, dose reductions, and discontinuation criteria exist across institutions, largely reflecting concerns regarding tolerability, adverse events, and patient preference (9-12). Accordingly, in routine clinical practice, many patients do not complete prolonged maintenance schedules, and the degree of BCG exposure achieved varies substantially across settings. Contemporary real-world data that jointly describe treatment delivery (including early maintenance exposure) and clinically meaningful oncological outcomes remain important as benchmarks for conventional intravesical BCG therapy. In parallel, the therapeutic landscape for NMIBC is evolving rapidly. Novel bladder-sparing strategies, including the administration of immune checkpoint inhibitors, gene therapy, and antibody–drug conjugates, have emerged as promising options for patients with BCG-unresponsive disease (13-15). More recently, combination strategies incorporating immune checkpoint inhibitors with intravesical BCG as part of a first-line treatment for BCG-naïve patients with high-risk NMIBC have also been investigated in randomized clinical trials (16-19). These developments underscore the importance of establishing contemporary real-world benchmarks for outcomes achieved with conventional BCG therapy, which serve as a critical reference for evaluating emerging salvage therapies and novel upfront combination approaches.
Therefore, to provide contemporary real-world benchmarks for conventional intravesical BCG therapy and to clarify its current limitations in routine practice, we conducted a retrospective single-center analysis of patients with NMIBC treated with BCG. We specifically described treatment delivery – including achievement of an institution-defined minimum BCG exposure (induction plus initial maintenance exposure or re-induction) – and evaluated oncological outcomes using clinically meaningful endpoints, including high-risk recurrence-free survival.
Patients and Methods
Patient population. This retrospective study included patients with NMIBC who were treated with intravesical BCG therapy at the Kyushu Cancer Center. Overall, 124 consecutive patients who initiated intravesical BCG therapy between June 2019 and May 2025 were initially identified. Before June 2019, intravesical BCG therapy at our institution was limited to induction therapy alone; therefore, only patients treated during the period in which maintenance BCG therapy was routinely implemented were considered eligible for inclusion in the present study. Of these 124 patients, four were excluded for the following reasons: concurrent upper tract urothelial carcinoma (UTUC) at the time of BCG initiation, insufficient clinical data for outcome evaluation, or receipt of intravesical BCG therapy during systemic treatment for metastatic urothelial carcinoma. Consequently, 120 patients were included in the final analysis. The observation period spanned from the initiation of intravesical BCG therapy to the date of the last follow-up or death, with a data cutoff of November 30, 2025.
Clinical, pathological, and treatment-related variables extracted from patient medical records included age at the initiation of intravesical BCG therapy, sex, tumor status (primary or recurrent), tumor size, tumor multifocality, pathological stage, presence of concomitant carcinoma in situ (CIS), tumor grade assessed according to the World Health Organization (WHO) 1973 classification (20) and the WHO 2004 classification (21), histological variant, lymphovascular invasion (LVI), history of previous intravesical BCG therapy, and history of UTUC.
BCG treatment protocol and adequate intravesical BCG therapy (definition). Intravesical BCG therapy was administered using the Immunobladder® formulation (Tokyo 172 strain) (22) according to an institutional protocol based on the Southwest Oncology Group (SWOG) 8507 trial. Induction therapy consisted of weekly intravesical BCG instillations for six consecutive weeks. Following the completion of induction therapy, patients were scheduled to receive maintenance BCG therapy consisting of three weekly instillations at 3, 6, 12, 18, 24, 30, and 36 months after the initiation of induction therapy, in line with the SWOG 8507 maintenance schedule (8).
The standard initial dose of BCG was 80 mg per instillation. Dose reduction or treatment modification was performed at the discretion of the treating physician based on patient tolerability. Maintenance therapy was withheld or discontinued in cases of disease recurrence, progression, unacceptable adverse events, or patient preference.
Adequate intravesical BCG therapy was defined according to institutional criteria and was considered achieved if patients met either of the following conditions: (i) completion of BCG induction therapy (≥5 of 6 scheduled instillations) and at least one cycle of maintenance therapy (≥2 of 3 scheduled instillations); or (ii) completion of BCG induction therapy (≥5 of 6 scheduled instillations) and BCG re-induction therapy (≥2 of 6 scheduled instillations) (13, 14). This operational definition was used to describe an institution-defined minimum exposure in routine practice and should not be equated with completion of the full SWOG maintenance schedule.
Ethics statement. This retrospective study was conducted in accordance with the ethical standards of the Declaration of Helsinki and was approved by the ethics review board of the National Hospital Organization Kyushu Cancer Center (approval no. 2014-99). In accordance with the guidelines of the ethics committee and national ethical standards in Japan, written informed consent was waived because this was a retrospective observational study using existing materials, including medical records and documents, with an opt-out approach.
Statistical analysis. All statistical analyses were performed using EZR (version 1.68; Saitama Medical Center, Jichi Medical University, Saitama, Japan) (23). Continuous variables are summarized as median (interquartile range), and categorical variables as number (percentage). High-risk recurrence-free survival (HR-RFS) was defined as the time from the initiation of intravesical BCG therapy to the first recurrence of high-risk NMIBC or progression to a higher pathological stage (upstaging); high-risk disease was defined according to the JUA guidelines (T1, high-grade, and/or CIS) (7). Because HR-RFS focuses on high-risk recurrences defined by the JUA risk stratification, comparability with conventional RFS definitions may be limited. Progression-free survival (PFS) was defined as the time from the initiation of BCG therapy to progression to muscle-invasive disease or the development of distant metastasis. Metastasis-free survival (MFS) was defined as the time from the initiation of BCG therapy to the first documentation of distant metastasis. Overall survival (OS) was defined as the time from the initiation of BCG therapy to death from any cause; cause-specific mortality data were not available in the institutional cancer registry. Survival outcomes were estimated using the Kaplan–Meier method. Patients without an event were censored at the date of the last follow-up. Given the limited number of outcome events, multivariable survival regression analyses were not performed to avoid model overfitting.
Results
Patient characteristics. Overall, 120 patients with NMIBC were included in this study (Table I). The median age was 73 years [interquartile range (IQR)=69-79], and most patients were men (85.0%). Primary tumors accounted for 80.0% of cases, and 20.0% of patients had recurrent disease. Most tumors were smaller than 3 cm in diameter (85.8%), and tumor multifocality was observed in 92.5% of patients. Regarding pathological stage, 13.3% of tumors were classified as pTis, 40.8% as pTa, and 45.8% as pT1. Concomitant CIS was present in 29.2% of patients.
Baseline clinicopathological characteristics of patients with non–muscle-invasive bladder cancer treated with intravesical Bacillus Calmette–Guérin (BCG) therapy.
According to the WHO 1973 classification, tumor grades were distributed as follows: grade 1 in 1.7%, grade 2 in 50.0%, and grade 3 in 48.3% of patients. Based on the WHO 2004 classification, 98.3% of tumors were classified as high grade. Histological variants were absent in 87.5% of the cases, and lymphovascular invasion was identified in one patient (0.8%).
A history of intravesical BCG therapy and upper tract urothelial carcinoma was noted in 5.0% and 11.7% of the patients, respectively.
BCG treatment delivery. Intravesical BCG therapy was administered to all patients in the study cohort (Table II). The initial BCG dose was 80 mg for 119 patients (99.2%), and one patient (0.8%) received a reduced dose of 40 mg. Regarding induction therapy, 101 patients (84.2%) completed the planned six instillations, 12 patients (10.0%) received five instillations, and seven patients (5.8%) received four or fewer instillations.
Bacillus Calmette–Guérin (BCG) treatment details.
With respect to maintenance therapy, 82 patients (68.3%) received at least two instillations during the first maintenance cycle (C1), whereas eight patients (6.7%) received only one instillation and 30 patients (25.0%) did not receive maintenance therapy in C1. Overall, the institution-defined minimum BCG exposure was achieved in 88 patients (73.3%). Beyond the first maintenance cycle, 21 patients (17.5%) received BCG maintenance therapy beyond C2, and only three patients (2.5%) completed maintenance therapy up to C7 (the full SWOG maintenance schedule).
Survival outcomes. The Kaplan–Meier survival curves for HR-RFS, PFS, MFS, and OS are shown in Figure 1. The median follow-up duration was 33.5 months (range, 4.4-70.3 months). The median HR-RFS, PFS, MFS, and OS were not reached during the study period.
Kaplan–Meier survival curves for patients with non–muscle-invasive bladder cancer treated with intravesical Bacillus Calmette–Guérin (BCG) therapy. (A) High-risk recurrence-free survival. (B) Progression-free survival. (C) Metastasis-free survival. (D) Overall survival. Tick marks indicate censored observations. Point estimates at 1, 3, and 5 years (y) with 95% confidence intervals are shown within each panel. The numbers of patients at risk are shown below each panel.
The 1-, 3-, and 5-year HR-RFS rates were 91.4% [95% confidence interval (CI)=84.6-95.3], 76.7% (95% CI=66.6-84.1), and 73.0% (95% CI=60.4-82.2), respectively (Figure 1A). For PFS, the 1-, 3-, and 5-year rates were 98.3% (95% CI= 93.4-99.6), 92.5% (95% CI=84.4-96.4), and 92.5% (95% CI=84.4-96.4), respectively (Figure 1B). The MFS rates at 1, 3, and 5 years were 99.2% (95% CI=94.2-99.9), 93.3% (95% CI=85.2-97.0), and 93.3% (95% CI=85.2-97.0), respectively (Figure 1C). Similarly, the OS rates at 1, 3, and 5 years were 98.3% (95% CI=93.3-99.6), 97.3% (95% CI=91.8-99.1), and 86.2% (95% CI=64.2-95.2), respectively (Figure 1D). During follow-up, HR-RFS events occurred in n=24; PFS events in n=7; MFS events (a subset of PFS events) in n=6; and deaths in n=6.
Discussion
In this retrospective single-center study, we evaluated real-world treatment delivery and the oncological outcomes of intravesical BCG therapy in patients with NMIBC. Despite the cohort was enriched with high-risk features – including a high prevalence of pT1 disease, concomitant CIS, and high-grade tumors – favorable survival outcomes were observed. The institution-defined minimum BCG exposure, incorporating induction therapy with at least one cycle of maintenance or re-induction therapy, was achieved in 73.3% of the patients. Although the completion of long-term maintenance therapy beyond the first cycle was limited, HR-RFS, PFS, MFS, and OS rates were encouraging. Notably, the median HR-RFS, PFS, MFS, and OS were not reached during a median follow-up of 33.5 months, and the 5-year progression-free and metastasis-free survival rates exceeded 90%. These findings provide contemporary real-world benchmarks for outcomes achieved with conventional intravesical BCG therapy and may help contextualize emerging bladder-sparing strategies in the evolving treatment landscape of NMIBC.
Compared with historical benchmarks, the present cohort demonstrated favorable oncological outcomes despite being enriched with well-established high-risk features, including pT1 disease in 45.8%, concomitant CIS in 29.2%, high-grade tumors in 98.3% of patients, and a high prevalence of multifocal disease. In the randomized Southwest Oncology Group (SWOG) 8507 trial, the addition of a 3-week maintenance schedule to induction BCG therapy significantly improved recurrence-related outcomes and worsening-free survival compared with induction therapy alone, establishing maintenance BCG as a critical component of optimal intravesical therapy whenever feasible (8).
From a guideline-based perspective, the EAU 2021 risk stratification model estimated a 5-year progression probability of approximately 9%-10% for patients classified as high risk (WHO 2004/2016 classification) and up to approximately 40% for those categorized as very high risk, underscoring the marked heterogeneity within high-risk NMIBC (1). In this context, the 5-year PFS rate of 92.5% observed in our cohort – corresponding to an event probability of 7.5% for muscle-invasive progression or distant metastasis – falls within the expected range for adequately treated high-risk disease. This finding suggests that the low incidence of progression observed in the present study is not unexpectedly favorable and may be influenced, at least in part, by the relatively high proportion of patients who received induction BCG therapy followed by at least an initial cycle of maintenance therapy in routine clinical practice; however, this hypothesis could not be formally tested in this descriptive analysis.
Recurrence outcomes are inherently more difficult to compare across studies, as recurrence risk is strongly influenced by tumor multiplicity, prior recurrence history, treatment delivery and, critically, by the definition of the recurrence endpoint itself. The European Organization for Research and Treatment of Cancer risk tables, derived from pooled analyses of multiple randomized trials, report widely varying 5-year recurrence probabilities ranging from approximately 31% to 78% across different risk-score strata, highlighting substantial heterogeneity even among patients with broadly similar baseline risk profiles (2, 24-26). Subsequent analyses incorporating maintenance BCG have confirmed persistent variability in recurrence risk despite prolonged intravesical therapy (3).
Importantly, insights from Japanese real-world data further contextualize the present findings. In a nationwide claims-based analysis of Japanese patients with NMIBC receiving intravesical BCG therapy, approximately three-quarters of patients were reported to have completed guideline-compliant induction therapy; however, the long-term oncological outcomes remained modest, with a 5-year recurrence-free rate of approximately 50% and a substantial proportion of patients requiring subsequent treatment (27). Notably, maintenance BCG therapy could not be reliably captured in that administrative database, and recurrence was broadly defined as the initiation of any subsequent bladder cancer treatment, including transurethral resection, regardless of tumor grade or risk category.
In contrast, the present single-center cohort demonstrated more favorable recurrence and progression outcomes, despite a high-risk disease profile. Several factors may explain this discrepancy. First, a higher proportion of patients in our study received induction therapy, as well as at least an initial cycle of maintenance or re-induction BCG, which has been shown to confer incremental benefit over induction therapy alone. Second, HR-RFS in our analysis was defined based on high-risk NMIBC recurrence or progression, thereby focusing on clinically meaningful events and excluding low-grade recurrences that may inflate recurrence rates in database-driven studies. Third, treatment at a tertiary cancer center may have facilitated closer surveillance, standardized pathological assessment, and optimized treatment delivery, all of which may have contributed to improved disease control. However, these potential contributors could not be formally evaluated in the present descriptive analysis and should be interpreted as hypotheses rather than causal explanations.
Although the completion of long-term maintenance therapy according to the full SWOG 8507 maintenance schedule was uncommon, approximately two-thirds of the patients in our cohort received two or more instillations during the first maintenance cycle. This real-world treatment pattern remains conceptually consistent with the core findings of SWOG 8507, which demonstrated an incremental benefit of maintenance BCG over induction therapy alone (8). Recent analyses of BCG-exposed NMIBC populations have similarly suggested that achieving induction therapy followed by at least limited maintenance exposure may be sufficient to maintain low risks of progression and metastasis in a meaningful proportion of patients (10, 28-30). Collectively, these observations indicate that the quality and completeness of BCG delivery – rather than induction therapy alone – may be an important contributor to long-term oncological outcomes in high-risk NMIBC and provide contemporary real-world benchmarks against which emerging bladder-sparing strategies can be evaluated.
Limitations. First, this analysis was retrospective in nature and was conducted at a single tertiary cancer center, which may limit the generalizability of the findings to broader clinical settings. Patient selection, treatment delivery, and follow-up intensity at a specialized cancer center may differ from those in community-based practice, potentially contributing to more favorable outcomes. Second, although the cohort size was moderate, subgroup analyses – particularly according to detailed treatment exposure patterns – were limited by the sample size and the relatively small number of progression events. As a result, the present study was not powered to identify independent predictors of recurrence or progression, and the findings should be interpreted primarily as descriptive real-world outcomes rather than causal associations. Third, the duration of follow-up, while sufficient to assess intermediate-term oncological outcomes, may not fully capture late recurrences or disease progression. Studies with a longer follow-up period will be required to confirm the long-term durability of the disease control observed in this cohort. Finally, HR-RFS in this study was defined based on high-risk NMIBC recurrence or progression, whereas low-grade or low-risk recurrences were not considered events. Although this approach was intentionally chosen to focus on clinically meaningful outcomes, it may limit direct comparability with prior studies that used broader recurrence definitions. This methodological difference should be considered when interpreting the relatively favorable recurrence outcomes reported here. In addition, reasons for maintenance discontinuation (e.g., adverse events or patient preference) were not systematically captured, precluding a detailed evaluation of protocol non-adherence.
Conclusion
In this real-world, single-center retrospective study, favorable oncological outcomes were observed in patients with NMIBC treated with intravesical BCG therapy, despite a high prevalence of adverse pathological features. Outcomes appeared favorable among patients who achieved the institution-defined minimum BCG exposure, with low risks of disease progression and distant metastasis, even though completion of the full SWOG maintenance schedule was uncommon.
Acknowledgements
The Authors thank J. Ludovic Croxford, PhD, from Edanz (https://jp.edanz.com/ac) for editing a draft of this manuscript.
Footnotes
Authors’ Contributions
Conceptualization, N.F., M.N., and T.N.; methodology, N.F. and T.N.; formal analysis, N.F. and T.N.; investigation, N.F., J.T., A.T., Y.S., M.N., and T.N.; resources, M.N. and T.N.; data curation, N.F.; writing – original draft preparation, N.F., M.N., and T.N.; writing – review and editing, all authors; visualization, N.F.; supervision, M.N. and T.N. All Authors have read and agreed to the published version of the manuscript.
Conflicts of Interest
The Authors declare no conflict of interest in relation to this study.
Artificial Intelligence (AI) Disclosure
No artificial intelligence (AI) tools, including large language models or machine learning software, were used in the preparation, analysis, or presentation of this manuscript.
- Received February 5, 2026.
- Revision received March 2, 2026.
- Accepted March 4, 2026.
- Copyright © 2026 The Author(s). Published by the International Institute of Anticancer Research.
This is an open access article under the terms of the Creative Commons Attribution License, which permits use, distribution and reproduction in any medium, provided the original work is properly cited.
References
In this issue
Jump to section
Related Articles
Cited By...
- No citing articles found.