Abstract
Purpose
To evaluate the safety, toxicity, efficacy, and health-economic context of intravesical gemcitabine (GEM) in patients with non–muscle-invasive bladder cancer (NMIBC), and compare outcomes with Bacillus Calmette–Guérin (Onco-BCG; Cipla Ltd.), particularly during shortages.
Materials and Methods
This prospective observational study included 128 patients with intermediate- or high-risk NMIBC. Sixty-six received intravesical GEM after transurethral resection of bladder tumor (TURBT), while 62 received Onco-BCG. GEM was administered according to the GUBGEM protocol. Safety was assessed through clinical, laboratory, I-PSS (International Prostate Symptom Score), and CTCAE (Common Terminology Criteria for Adverse Events) v5.0 grading. Efficacy was evaluated by recurrence-free survival (RFS), progression rates, and Kaplan–Meier analysis.
Results
Baseline demographic and pathological characteristics were similar. GEM demonstrated a favorable safety profile, with 90.9% experiencing none or grade 1 adverse events. Urinary tract infection (20.0%), dysuria (17.3%), and musculoskeletal pain (10.8%) were the most common symptoms. No severe systemic toxicities were observed. Recurrence occurred in 34.8% of GEM patients versus 16.1% with Onco-BCG (p=0.016). Median RFS was 32 months for GEM and 60 for Onco-BCG. Log-rank analysis showed significantly lower RFS with GEM (p=0.017), confirmed by Cox regression (hazard ratio 2.92, 95% confidence interval 1.30–6.54; p=0.009). Most recurrences occurred within the first 12 months.
Conclusions
Intravesical GEM was safe, well tolerated, and associated with high adherence. Although recurrence prevention was inferior to Onco-BCG—particularly within the first 12 months, with no difference in progression—GEM represents a pragmatic, budget-sensitive alternative during Onco-BCG shortages or intolerance.
Keywords: BCG vaccine, Gemcitabine, Urinary bladder neoplasms
INTRODUCTION
Bladder cancer ranks as the ninth most common malignancy worldwide, with an estimated 614,298 new cases and 220,596 deaths in 2022 [1]. Among histological subtypes, approximately 75% of diagnosed cases are non–muscle-invasive bladder cancer (NMIBC), usually of low histological grade [2]. Despite transurethral resection of bladder tumor (TURBT) being the standard initial treatment, this tumor group shows a high recurrence rate, which leads to increased morbidity, repeated surgical interventions, and high costs to healthcare systems [3].
For patients with intermediate- or high-risk NMIBC, adjuvant intravesical chemotherapy constitutes an important complementary strategy. Studies have shown that a single intravesical chemotherapy instillation within the first 24 hours after TURBT can reduce tumor recurrence risk by up to 40% compared with TURBT alone [2,3].
In this context, gemcitabine (GEM) (2′,2′-difluorodeoxycytidine) has emerged as a promising option. This drug acts as a DNA synthesis inhibitor, interfering with the replication of tumor cells; it was initially developed for systemic use in advanced bladder carcinoma and other malignancies [4]. GEM has also been investigated for intravesical application due to its favorable safety, good tolerability, and local efficacy [5,6].
Furthermore, GEM-based regimens have shown encouraging results as second-line therapy in high-grade NMIBC after failure of Bacillus Calmette–Guérin (Onco-BCG; Cipla Ltd.) immunotherapy [2,5,6,7,8]. The recent global shortage of Onco-BCG, coupled with its potential adverse effects and failure rates, underscores the need for safe and viable alternatives such as intravesical GEM. In recent years, a worldwide shortage of Onco-BCG has been reported, substantially affecting clinical practice. A 2025 global real-world survey found that 45% of physicians reported being impacted by BCG shortages, with marked regional variation (from 23% in Japan to 71% in Canada), leading many centers to reserve BCG primarily for high-risk patients [9]. This global supply crisis, combined with the well-recognized toxicity profile and risk of treatment failure, underscores the urgent need for safe and effective alternatives such as intravesical GEM.
This study aims to contribute to the literature by evaluating the safety, toxicity, efficacy and health-economic context of intravesical GEM in NMIBC patients treated at a public referral institution, and by comparing its outcomes with those achieved by using Onco-BCG in patients with similar characteristics.
MATERIALS AND METHODS
This prospective observational study, with a minimum follow-up of 24 months, was conducted at the Urology Outpatient Clinic of the Irmandade da Santa Casa de Misericórdia de São Paulo, Brazil (ISCM-SP) between January 2021 and July 2025. The protocol was reviewed and approved by the institutional Research Ethics Committee of ISCM-SP Hospital (approval number: CAAE 36369020.0.0000.5479), and all patients provided written informed consent.
A total of 66 adult patients diagnosed with intermediate- or high-risk NMIBC, according to the European Association of Urology (EAU) 2025 guidelines [2], were included. Eligible patients were those intolerant, unresponsive, or ineligible to intravesical Onco-BCG therapy, provided they had undergone an index TURBT with macroscopic complete resection and detrusor muscle present in the pathological specimen; both were mandatory inclusion criteria [10,11]. Re-TURBT is not routinely performed at our institution; however, it is selectively employed when there is concern regarding resection quality or tumor burden (e.g., tumor size >4 cm, multiple lesions, uncertain completeness or clearly incomplete resection, or absence of detrusor muscle in the initial specimen). Exclusion criteria were known GEM allergies, persistent gross hematuria, untreated active urinary tract infection, or clinical suspicion of bladder perforation.
The intervention followed the GUBGEM protocol from BC Cancer [12], consisting of intravesical instillation of 2,000 mg GEM diluted in 60 mL of saline, with a target intravesical bladder retention time of approximately 1–2 hours, when tolerated. Induction involved weekly instillations for six consecutive weeks, followed by monthly maintenance up to month 12 and quarterly instillations until month 24.
Each session was documented, including date, intended retention time, and any complications. Although exact dwell times for every instillation were not uniformly available for all treatment cycles, clinical records indicated that the majority of patients were able to retain the intravesical agent for at least 1 hour, and episodes of shorter retention were generally sporadic rather than persistent across treatment cycles.
Patients underwent regular clinical and laboratory monitoring, including complete blood count, renal function (creatinine), liver enzymes (alanine aminotransferase/aspartate aminotransferase [ALT/AST]), and the International Prostate Symptom Score (I-PSS) questionnaire. Toxicity was graded according to the Common Terminology Criteria for Adverse Events (CTCAE) v5.0 [13], documenting all local and systemic adverse events.
Oncological response was assessed with serial urine cytology and cystoscopy every three months. Recurrence was defined as the reappearance of a urothelial lesion confirmed by biopsy or repeat TURBT. For comparison, we used a historical Onco-BCG cohort with 62 patients, with comparable clinical and pathological characteristics. That cohort received a standard six-week induction course, followed by maintenance consisting of three weekly instillations for consecutive three weeks, scheduled at month 3, 6, 12, 18, 24, 30, and 36. Each instillation contained one 40 mg ampoule of Onco-BCG. Maintenance duration was risk-adapted: 1 year for intermediate-risk NMIBC and up to 3 years for high-risk disease, contingent on patient tolerance [14].
Clinical and demographic characteristics of the GEM and Onco-BCG groups are summarized in Table 1. Gender and ethnicity were recorded according to patient self-identification. No significant differences were observed between groups in mean age (68.0 years vs. 66.0 years, p=0.258) or sex distribution (male; 66.7% vs. 72.6%, p=0.468).
Table 1. Patient and tumor characteristics for GEM and Onco-BCG groups.
| Variable | Statistical measure | Treatment | Total (n=128) | p-value | |
|---|---|---|---|---|---|
| GEM (n=66, 51.6%) | Onco-BCG (n=62, 48.4%) | ||||
| Age of diagnosis (y) | Range | 28.0–92.0 | 40.0–90.0 | 28.0–92.0 | 0.258a |
| Median (IQR) | 67.0 (63.2–75.0) | 64.5 (60.0–71.0) | 67.0 (61.0–74.0) | ||
| Mean±SD | 68.0±10.8 | 66.0±9.8 | 67.0±10.4 | ||
| Number | 66 | 62 | 128 | ||
| Sex | Female | 22 (33.3) | 17 (27.4) | 39 (30.5) | 0.468b |
| Male | 44 (66.7) | 45 (72.6) | 89 (69.5) | ||
| Ethnicity | White | 46 (69.7) | - | 46 (35.9) | - |
| Mixed-race | 17 (25.8) | - | 17 (13.3) | ||
| Black | 3 (4.5) | - | 3 (2.3) | ||
| RLW | Yes | 37 (56.1) | 22 (35.5) | 59 (46.1) | 0.020b |
| LLW | Yes | 27 (40.9) | 17 (27.4) | 44 (34.4) | 0.108b |
| Base | Yes | 18 (27.3) | 6 (9.7) | 24 (18.8) | 0.011b |
| Floor | Yes | 27 (40.9) | 15 (24.2) | 42 (32.8) | 0.044b |
| Neck | Yes | 9 (13.6) | 5 (8.1) | 14 (10.9) | 0.313b |
| Trigone | Yes | 12 (18.2) | 4 (6.5) | 16 (12.5) | 0.045b |
| Dome | Yes | 17 (25.8) | 14 (22.6) | 31 (24.2) | 0.675b |
| Quantity | 1 | 33 (50.0) | 27/45 (60.0) | 60/111 (54.1) | 0.255b |
| 2–7 | 21 (31.8) | 8/45 (17.8) | 29/111 (26.1) | ||
| >8 | 12 (18.2) | 10/45 (22.2) | 22/111 (19.8) | ||
| Size | <3 cm | 37 (56.1) | 19/43 (44.2) | 56/109 (51.4) | 0.225b |
| >3 cm | 29 (43.9) | 24/43 (55.8) | 53/109 (48.6) | ||
| Invades lamina propria/superficial | Invades lamina propria | 28 (42.4) | 24 (38.7) | 52 (40.6) | 0.566b |
| Superficial | 38 (57.6) | 38 (61.3) | 76 (59.4) | ||
| Histological grade | High | 63 (95.5) | 60 (96.8) | 123 (96.1) | >0.999c |
| Low | 3 (4.5) | 2 (3.2) | 5 (3.9) | ||
| Staging | CIS | 3 (4.5) | 1 (1.6) | 4 (3.1) | 0.615c |
| T1 | 28 (42.4) | 24 (38.7) | 52 (40.6) | ||
| Ta | 35 (53.0) | 37 (59.7) | 72 (56.3) | ||
| Intermediate risk | 23 (34.8) | 23 (37.1) | 46 (35.9) | 0.626c | |
| High risk | 40 (60.6) | 38 (61.3) | 78 (60.9) | ||
| Very high risk | 3 (4.5) | 1 (1.6) | 4 (3.1) | ||
| Treatment lines | 1 | 43 (65.2) | 54 (87.1) | 97 (75.8) | 0.006c |
| 2 | 19 (28.8) | 8 (12.9) | 27 (21.1) | ||
| 3 | 4 (6.1) | 0 (0.0) | 4 (3.1) | ||
Values are presented as range, median (IQR), mean±SD, number only, or number (%).
This table summarizes baseline demographic, tumor, and pathological features of patients treated with GEM or Onco-BCG. Age distribution, sex, ethnicity, tumor site, size, number, histological grade, and stage are reported, along with calculated risk stratification. No major imbalances were observed between groups, although GEM patients presented more tumors in the RLW and trigone and some GEM patients had higher treatment lines compared to Onco-BCG patients.
Data were unavailable for some patients; therefore, denominators vary across variables, and percentages were calculated using only patients with available data.
GEM, gemcitabine; BCG, Bacillus Calmette–Guérin; IQR, interquartile range; SD, standard deviation; RLW, right lateral wall; LLW, left lateral wall; CIS, carcinoma in situ.
a:Two sample t-test. b:Pearson’s chi-squared test. c:Fisher’s exact test for count data.
Statistical analyses were performed using RStudio v4.4.2 and supplementary Python scripts. Descriptive statistics summarized clinical, laboratory, and demographic variables. Continuous variables were compared using Student’s t-test or Wilcoxon’s test, categorical variables with chi-squared or Fisher’s exact tests. Survival was estimated with Kaplan–Meier curves and compared via log-rank test. Cox regression provided hazard ratios (HR) with 95% confidence intervals (CIs). A p-value <0.05 was considered statistically significant.
RESULTS
A total of 128 patients diagnosed with NMIBC were included in the study, 66 treated with intravesical GEM and 62 with Onco-BCG. Clinical and demographic characteristics are shown in Table 1.
Tumors on the right lateral wall (RLW) were more frequent in the GEM group (56.1%) than in the Onco-BCG group (35.5%) (p=0.020), as were lesions located in the bladder base (27.3% vs. 9.7%, p=0.011), bladder floor (40.9% vs. 24.2%, p=0.044), and trigone (18.2% vs. 6.5%, p=0.045). The number of tumors, tumor size (<3 cm; 56.1% GEM vs. 44.2% Onco-BCG, p=0.225), and histologic grade (>95% high grade) were similar between groups.
In the GEM group, 65.2% of patients received only one treatment line, compared to 87.1% in the Onco-BCG group (p=0.006), suggesting that these patients were more likely to be managed with a single regimen.
As shown in Table 2, the recurrence rate was significantly higher in the GEM group (34.8%) compared to the Onco-BCG group (16.1%) (p=0.016), while progression rates were similar between groups (6.1% vs. 6.5%, p>0.999). No significant difference was observed in the meantime to recurrence (12.6 months vs. 19.3 months, p=0.950) or to progression (6.5 months vs. 5.2 months, p=0.765). Most recurrences occurred within the first 12 months in both groups, indicating that most events were early recurrences.
Table 2. Clinical outcomes in GEM and Onco-BCG groups.
| Variable | Statistical measure | Treatment | Total | p-value | |
|---|---|---|---|---|---|
| GEM | Onco-BCG | ||||
| Recurrence | No | 43 (65.2) | 52 (83.9) | 95 (74.2) | 0.016a |
| Yes | 23 (34.8) | 10 (16.1) | 33 (25.8) | ||
| Progression | No | 62 (93.9) | 58 (93.5) | 120 (93.8) | >0.999b |
| Yes | 4 (6.1) | 4 (6.5) | 8 (6.3) | ||
| Time to recurrence (mo) | Range | 2.0–49.0 | 2.0–67.0 | 2.0–67.0 | 0.950c |
| Median (IQR) | 7.0 (3.5–14.0) | 5.0 (4.0–21.0) | 6.5 (3.8–15.5) | ||
| Mean±SD | 12.6±13.1 | 19.3±25.7 | 14.5±17.4 | ||
| Number | 23 | 9 | 32 | ||
| Time to progression (mo) | Range | 0.0–14.0 | 1.0–13.0 | 0.0–14.0 | 0.765d |
| Median (IQR) | 6.0 (3.0–9.5) | 3.5 (2.5–6.2) | 4.0 (2.5–9.2) | ||
| Mean±SD | 6.5±6.0 | 5.2±5.3 | 5.9±5.3 | ||
| Number | 4 | 4 | 8 | ||
Values are presented as number (%), range, median (IQR), mean±SD, or number only.
This table presents recurrence, progression, and time-to-event outcomes stratified by treatment group. Recurrence occurred more frequently in GEM patients, whereas progression rates were similar across groups. Time-to-recurrence and time-to-progression are displayed using mean, median, and range measures, with statistical comparisons provided for each.
GEM, gemcitabine; BCG, Bacillus Calmette–Guérin; IQR, interquartile range; SD, standard deviation.
a:Pearson’s chi-squared test. b:Fisher’s exact test for count data. c:Wilcoxon rank sum test. d:Two sample t-test.
Regarding toxicity, GEM demonstrated a favorable safety profile, depicted in Table 3. Among GEM patients, 98.5% had mild I-PSS scores and only 1.5% moderate. CTCAE grading showed 59.1% Grade 1 toxicity, mostly due to dysuria, 9.1% Grade 2, with cases of isolated symptomatic hematuria, cystitis requiring bladder irrigation, or worsening of pre-existing urinary incontinence. The remainder of patients had no adverse effects.
Table 3. Distribution of toxicity and clinical changes in patients treated with gemcitabine.
| Variable | Classification | Value |
|---|---|---|
| I-PSS | Mild | 65 (98.5) |
| Moderate | 1 (1.5) | |
| CTCAE v5.0 | None | 21 (31.8) |
| Grade 1 | 39 (59.1) | |
| Grade 2 | 6 (9.1) | |
| Symptoms moment during treatment | Induction | 16/42 (38.1) |
| Induction and maintenance | 11/42 (26.2) | |
| Maintenance | 15/42 (35.7) | |
| Laboratory | Creatinine | 5/73 (6.8) |
| AST or ALT | 2/71 (2.8) | |
| Hyperkalemia | 2/71 (2.8) | |
| Blood count | 0 (0.0) | |
| Symptom | Urinary tract infection (culture) | 15/75 (20.0) |
| Dysuria | 13/75 (17.3) | |
| Musculoskeletal pain | 8/74 (10.8) | |
| Hematuria | 4/75 (5.3) | |
| Pollakiuria | 4/75 (5.3) | |
| Urgency incontinence | 3/68 (4.4) | |
| Weak stream | 2/71 (2.8) | |
| Nocturia | 2/71 (2.8) | |
| Mental confusion | 1/77 (1.3) |
Values are presented as number (%).
This table describes treatment-related toxicity, with grading scales such as CTCAE v5.0 and I-PSS, clinical symptoms, and laboratory abnormalities in patients receiving intravesical gemcitabine. Most patients experienced only mild or moderate adverse effects, with no determined predominance over induction or maintenance phases. Laboratory alterations and systemic toxicities were rare, and no severe or life-threatening events were observed.
Data were unavailable for some patients; therefore, denominators vary across variables, and percentages were calculated using only patients with available data.
I-PSS, International Prostate Symptom Score; CTCAE, Common Terminology Criteria for Adverse Events; AST, aspartate aminotransferase; ALT, alanine aminotransferase.
Symptom occurrence was distributed across treatment phases, with 38.1% during induction, 35.7% during maintenance, and 26.2% during both phases, without a clear predominance between the two stages. The most frequent adverse events were urinary tract infection (20.0%), dysuria (17.3%), musculoskeletal pain (10.8%), and hematuria (5.3%). Notably, all patients who reported hematuria subsequently experienced tumor recurrence after symptom onset, confirmed by cystoscopy.
Laboratory abnormalities were rare, limited to mild creatinine elevation (6.8%) representing a KDIGO (Kidney Disease Improving Global Outcomes) I classification in all cases, nonspecific increases in AST or ALT, and 2 cases of hyperkalemia, all without any clinical repercussions.
Table 4 summarizes recurrence-free survival (RFS) by treatment. The Onco-BCG group (n=62; 10 events) had a median RFS of 60 months, whereas the GEM group (n=66; 23 events) had a median RFS of 32 months. The 95% CIs for the medians were not estimable because the upper Kaplan–Meier confidence band never fell to ≤0.50 before the end of follow-up—an effect driven by few events and substantial censoring, including our conservative handling of missing follow-up (patients documented as recurrence-free for ≥24 months were censored at 24 months).
Table 4. Comparison of RFS between treatment groups in patients with bladder cancer.
| Group | No. of patients | Censored (n) | Event (n) | Median RFS (mo)a | 95% CIb,c | p log-rankd | HR (95% CI) |
|---|---|---|---|---|---|---|---|
| Onco-BCG | 62 | 52 | 10 | 60 | - | 0.017 | Reference |
| GEM | 66 | 43 | 23 | 32 | - | 2.92 (1.30–6.54)e |
This table compares RFS between GEM and Onco-BCG groups using Kaplan–Meier estimates, Cox regression, and log-rank testing. Median RFS was longer for Onco-BCG, with HR indicating a significantly higher risk of recurrence in the GEM group. Methodological notes explain censoring and CI calculations.
RFS, recurrence-free survival; BCG, Bacillus Calmette–Guérin; GEM, gemcitabine; CI, confidence interval; HR, hazard ratio; -, not estimable because the confidence band criterion was not met within the observed follow-up period.
a:Kaplan–Meier estimates were used for RFS.
b:The 24-month censoring rule avoids imputing events but can slightly bias long-term survival downward and widen CIs.
c:Providing actual last-contact times for censored patients would likely tighten CIs and may slightly shift medians and p-values.
d:Medians and their CIs follow Brookmeyer–Crowley (log–log Greenwood); two-sided p-values are reported.
e:Cox regression; p=0.009.
Between-group comparison by log-rank test showed a significant difference in RFS (χ2=5.66, p=0.017). Consistently, the Cox model estimated a higher recurrence hazard for GEM vs. Onco-BCG (HR=2.92, 95% CI 1.30–6.54, p=0.009), using Onco-BCG as reference.
Expanding the analysis to Table 5 and the Kaplan–Meier curve (Fig. 1) generated from it, which depicts RFS probability over time (6, 12, and 24 months), we observed sustained high efficacy of Onco-BCG treatment over time, with a low incidence of therapeutic failure. The cumulative failure rate was more than twice as high in the GEM group across all evaluated time points, being particularly pronounced in the first six months of follow-up.
Table 5. RFS probability in patients with bladder cancer by treatment group at specific time points.
| Treatment | Time (mo) | No. at riska | Event (n)b | %RFS | 95% CI |
|---|---|---|---|---|---|
| Onco-BCG | 6 | 57 | 5 | 93.5 | 83.7–97.5 |
| 12 | 51 | 1 | 90.3 | 79.7–95.5 | |
| 24 | 50 | 1 | 88.7 | 77.8–94.5 | |
| GEM | 6 | 55 | 11 | 83.3 | 71.9–90.4 |
| 12 | 50 | 5 | 74.2 | 61.9–83.1 | |
| 24 | 47 | 3 | 69.7 | 57.1–79.3 |
This table shows RFS probabilities at 6, 12, and 24 months for each treatment group. Onco-BCG patients maintained higher survival probabilities across all timepoints, while GEM patients demonstrated progressive reductions over time. CIs, number at risk, and events at each interval are reported to contextualize survival differences.
RFS, recurrence-free survival; BCG, Bacillus Calmette–Guérin; GEM, gemcitabine; CI, confidence interval.
a:Number of patients remaining in the cohort without recurrence or censoring at each interval.
b:Number of recurrence events observed at each time point.
Fig. 1. Kaplan–Meier curve showing recurrence-free survival probability for the treatments (gemcitabine and Onco-BCG). BCG, Bacillus Calmette–Guérin.
Fig. 2 includes only patients who experienced recurrence during the follow-up period, allowing observation of the temporal distribution of therapeutic failure events in both groups. Most recurrences occurred early, within the first 12 months after treatment initiation, regardless of the therapeutic group. After this period, both curves tended to stabilize/plateau, indicating a lower incidence of new events over time.
Fig. 2. Curve showing time (months) to recurrence among patients who experienced this outcome. BCG, Bacillus Calmette–Guérin.
DISCUSSION
In this prospective observational study, intravesical GEM exhibited a favorable safety profile but lower efficacy than Onco-BCG in preventing recurrence, with similar progression rates. These findings align with the literature demonstrating Onco-BCG’s superiority in recurrence control when maintenance is completed, albeit at the cost of higher local and systemic toxicity [2,5,15].
The distribution of adverse events—predominantly mild lower urinary tract symptoms, urinary tract infection, musculoskeletal pain, and hematuria—suggests an association with urethral catheterization trauma rather than GEM’s systemic pharmacologic effect. Catheterization can cause microtrauma, bladder spasms, and increase infection risk [16]. Current AUA/SUNA recommendations advise avoiding Onco-BCG instillation after traumatic catheterization or in the presence of gross hematuria [3]. The absence of relevant systemic toxicity and the predominance of Grade 1 events contributed to high treatment adherence, contrasting with real-world Onco-BCG maintenance completion rates (<50%).
Compared with classical and recent Onco-BCG toxicity data, GEM regimens show lower rates of dysuria (12.6%), hematuria (10.9%), urgency (11.5%), and urinary tract infection (6.9%), with fewer systemic adverse events [2]. Updated systematic reviews confirm that GEM offers equivalent or superior tolerability to Onco-BCG [4,5].
Another relevant finding was the concentration of recurrences within the first 12 months, followed by stabilization of Kaplan–Meier curves, consistent with NMIBC’s natural history and EAU 2025 guideline recommendations for quarterly cystoscopy and cytology during the first two years in high-risk cases [2,7]. This underscores the need for intensified surveillance during the first year of GEM therapy [17].
The NMIBC therapeutic landscape is evolving. For low-risk disease, a single immediate post-TURBT instillation reduces recurrence risk. In intermediate and high-risk NMIBC, Onco-BCG with induction and maintenance remains the gold standard, but global shortages have encouraged alternative regimens. GEM, alone or in combination with docetaxel, has shown promising RFS and lower discontinuation rates [5,6,18]. Other bladder-preserving options include pembrolizumab for Onco-BCG-unresponsive carcinoma in situ [19], nadofaragene firadenovec [20], and N-803 plus Onco-BCG, U.S. Food and Drug Administration-approved in 2023 [21]. Device-assisted intravesical therapies, such as EMDA (electromotive drug administration)-mitomycin or chemo-hyperthermia, are also gaining traction [2].
From a health-economic perspective, the trade-off between Onco-BCG and intravesical GEM hinges on three domains: drug and regimen costs with clinic utilization, toxicity-related costs, and downstream costs driven by recurrence and progression.
A formal cost-effectiveness analysis based on cohort-level data was not feasible in this study. As a public, SUS (Sistema Unico de Saude – Unified Health System)-integrated, residency-driven institution, comprehensive accounting of direct and indirect healthcare costs—including personnel, infrastructure, procedures, and longitudinal follow-up—would require access to centralized governmental reimbursement databases that are not available at the institutional level. Moreover, labor costs in a teaching-hospital environment are inherently underestimated when compared with non-teaching or private healthcare systems, limiting the external validity of cohort-based economic modeling. Nonetheless, by integrating Brazilian real-world cost data from comparable public services with contemporary global and U.S. health-economic analyses, it is possible to contextualize our findings within a broader health-economic framework and to interpret the economic implications of intravesical therapies beyond institution-specific cost structures.
Contemporary cost-utility modeling in high-grade T1 shows 5-year mean management costs around US$26,000 for initial intravesical Onco-BCG versus ~US$39,700 for immediate cystectomy, with sensitivity analyses indicating that the costs of Onco-BCG toxicity, TURBTs, and cystectomy exert the greatest leverage on incremental cost-effectiveness ratios [22].
Although GEM is a low-cost generic and is generally less toxic than Onco-BCG, most cost data are reported for the sequential GEM-docetaxel regimen rather than GEM monotherapy. In a recent analysis, 2-year mean costs were lower for GEM-docetaxel than for Onco-BCG (US$7,090 vs. US$12,363) with similar quality-adjusted life-years (1.76 in both arms), suggesting that GEM-based intravesical therapy can be economically favorable in the short to intermediate term, particularly when clinic throughput and adherence are optimized [23]. A 2024/2025 European Urology review likewise summarizes that GEM-docetaxel tends to be less expensive than Onco-BCG at 1–2 years of follow-up, while new bladder-sparing agents carry substantially higher acquisition costs [24,25].
Brazil-specific data reinforce the budgetary relevance of intravesical strategies in the SUS. In a public oncology hospital cohort, stage I bladder cancer generated median costs of Brazilian real (BRL) 48,627 per patient over short follow-up, with estimated annualized costs of ~BRL 83,360 for continued surveillance and treatment. Intravesical Onco-BCG was associated with a median treatment-cost component of ~BRL 25,040; intravesical chemotherapy added ~BRL 10,175, highlighting that instillation-based regimens are visible line items in SUS budgets [26]. National expenditure profiles further show that ~77% of cancer costs accrue in the ambulatory setting (e.g., chemo/instillations, radiotherapy), underscoring the importance of regimen selection and adherence for cost containment (contextual SUS data).
Toxicity and adherence patterns influence real-world cost–benefit. Onco-BCG’s superior efficacy in recurrence prevention must be balanced against higher rates of local/systemic adverse events and frequent early discontinuation in practice, both of which add clinic visits, medications, occasional hospitalizations, and provider time. Conversely, GEM’s favorable tolerability and patient adherence may reduce toxicity-related resource use. However, as seen in our cohort and others, can be offset by a higher early recurrence burden, triggering additional TURBTs and intensified surveillance in the first year—costs that can erase initial savings if recurrences cluster early. Preventing progression is paramount economically: in US data, patients who progressed incurred ~US$233k over 5 years versus ~US$95k without progression [27].
Taken together, and especially in the current context of Onco-BCG shortages and constrained operating room/endoscopy capacity, our results favor intravesical GEM as a budget-sensitive alternative: it offers low toxicity, high adherence, and an acceptable recurrence profile if surveillance is intensified during the first year to mitigate recurrence-driven downstream costs.
In high-risk disease where Onco-BCG supply is reliable and maintenance feasible, Onco-BCG remains cost-effective in many models; however, where availability or tolerability is limiting, GEM represents a health-system-aligned strategy that balances clinical outcomes with resource stewardship [22,23,24,25].
Despite its strengths, this study has limitations that should be acknowledged. The use of a historical Onco-BCG cohort may introduce selection or management biases, and the single-center design limits generalizability. The sample size may also be insufficient to detect small differences in progression or long-term survival outcomes. In addition, although intravesical retention time was recorded at instillation, granular cycle-by-cycle dwell-time data were not uniformly available, precluding robust subgroup analyses based on retention-time thresholds; notably, episodes of shorter dwell time were sporadic, non-persistent, and infrequent, involving few patients and thus limiting statistical power to assess their independent impact on recurrence. Nevertheless, the prospective design, systematic follow-up, and direct comparison with Onco-BCG treated patients provide valuable real-world evidence.
CONCLUSIONS
Intravesical GEM demonstrated low systemic toxicity, predominantly mild local adverse effects, and high adherence, making it an attractive therapeutic alternative in Onco-BCG shortage or intolerance [4]. Compared to Onco-BCG, efficacy for recurrence prevention was lower, particularly in the first 12 months, with no significant difference in progression rates [2,5].
The temporal pattern of recurrences reinforces the need for rigorous surveillance during the first year, with shorter follow-up intervals and timely rescue interventions. Combined strategies—such as GEM plus docetaxel—or incorporation of newly approved immuno-/gene therapies may enhance efficacy while maintaining a favorable safety profile.
From an economic standpoint, and particularly amid Onco-BCG shortages, GEM represents a pragmatic, budget-sensitive option: lower acquisition and toxicity-related resource use can partially offset the higher early recurrence burden when paired with intensified first-year surveillance; where Onco-BCG supply is reliable and maintenance feasible, Onco-BCG remains cost-effective, and preventing progression is the principal driver of long-term costs [14,22,23,24,26]. In SUS and other publicly funded systems, intravesical instillations are major ambulatory expenditures; within this context, GEM offers a safe, adherent, and cost-conscious pathway when Onco-BCG is unavailable or poorly tolerated [27]. Our results therefore support GEM as a bladder-preserving option for intermediate-/high-risk NMIBC, offering a safe and adherent alternative in the face of Onco-BCG’s logistic and tolerability limitations.
ACKNOWLEDGMENTS
The authors would like to thank the patients and staff of the Urology Outpatient Clinic of the ISCM-SP for their participation and support throughout this study. We also acknowledge the contributions of colleagues and collaborators who assisted with data collection, patient follow-up, and administrative support.
Footnotes
CONFLICTS OF INTEREST: The authors have nothing to disclose.
FUNDING: None.
- Research conception and design: Tiago Mestriner Costa, Giovanni Demartino, and Roni de Carvalho Fernandes.
- Data acquisition: Tiago Mestriner Costa, Gabriela Dias Vetorazzi, Felipe Falero Albano, Vinicius Alves de Andrade, and Rafael Bali Moreira.
- Statistical analysis: Tiago Mestriner Costa and Luis Gustavo Morato de Toledo.
- Data analysis and interpretation: Tiago Mestriner Costa, Giovanni Demartino, and Rafael Bali Moreira.
- Drafting of the manuscript: Tiago Mestriner Costa and Giovanni Demartino.
- Critical revision of the manuscript: Luis Gustavo Morato de Toledo, Gabriela Dias Vetorazzi, Felipe Falero Albano, Vinicius Alves de Andrade, Rafael Bali Moreira, and Roni de Carvalho Fernandes.
- Administrative, technical, or material support: Gabriela Dias Vetorazzi, Felipe Falero Albano, and Vinicius Alves de Andrade.
- Supervision: Roni de Carvalho Fernandes and Luis Gustavo Morato de Toledo.
- Approval of the final manuscript: all authors.
References
- 1.Bray F, Laversanne M, Sung H, Ferlay J, Siegel RL, Soerjomataram I, et al. Global cancer statistics 2022: GLOBOCAN estimates of incidence and mortality worldwide for 36 cancers in 185 countries. CA Cancer J Clin. 2024;74:229–263. doi: 10.3322/caac.21834. [DOI] [PubMed] [Google Scholar]
- 2.Babjuk M, Burger M, Capoun O, Cohen D, Compérat EM, Dominguez Escrig JL, et al. European Association of Urology guidelines on non-muscle-invasive bladder cancer (Ta, T1, and carcinoma in situ) Eur Urol. 2022;81:75–94. doi: 10.1016/j.eururo.2021.08.010. [DOI] [PubMed] [Google Scholar]
- 3.Chang SS, Boorjian SA, Chou R, Clark PE, Daneshmand S, Konety BR, et al. Diagnosis and treatment of non-muscle invasive bladder cancer: AUA/SUO guideline. J Urol. 2016;196:1021–1029. doi: 10.1016/j.juro.2016.06.049. [DOI] [PubMed] [Google Scholar]
- 4.Ciccolini J, Serdjebi C, Peters GJ, Giovannetti E. Pharmacokinetics and pharmacogenetics of Gemcitabine as a mainstay in adult and pediatric oncology: an EORTC-PAMM perspective. Cancer Chemother Pharmacol. 2016;78:1–12. doi: 10.1007/s00280-016-3003-0. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 5.Matloubieh JE, Hanelin D, Agalliu I. Comparisons of intravesical treatments with mitomycin C, gemcitabine, and docetaxel for recurrence and progression of non-muscle invasive bladder cancer: updated systematic review and meta-analysis. Cancers (Basel) 2024;16:4125. doi: 10.3390/cancers16244125. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 6.McElree IM, Steinberg RL, Martin AC, Richards J, Mott SL, Gellhaus PT, et al. Sequential intravesical Gemcitabine and Docetaxel for bacillus Calmette-Guerin-naive high-risk nonmuscle-invasive bladder cancer. J Urol. 2022;208:589–599. doi: 10.1097/JU.0000000000002740. [DOI] [PubMed] [Google Scholar]
- 7.Mariappan P, Smith G, Lamb AD, Grigor KM, Tolley DA. Pattern of recurrence changes in noninvasive bladder tumors ob served during 2 decades. J Urol. 2007;177:867–875. doi: 10.1016/j.juro.2006.10.048. discussion 875. [DOI] [PubMed] [Google Scholar]
- 8.Addeo R, Caraglia M, Bellini S, Abbruzzese A, Vincenzi B, Montella L, et al. Randomized phase III trial on gemcitabine versus mytomicin in recurrent superficial bladder cancer: evaluation of efficacy and tolerance. J Clin Oncol. 2010;28:543–548. doi: 10.1200/JCO.2008.20.8199. [DOI] [PubMed] [Google Scholar]
- 9.Bedke J, Eccleston A, Brinkmann J, Chang J, Koshy A, Milloy N, et al. Global real-world patients characteristics, treatment patterns, and impact of BCG shortage in patients with high-risk non-muscle invasive bladder cancer. J Clin Oncol. 2025;43(5 Suppl):717 [Google Scholar]
- 10.Lerner SP, Dinney C, Kamat A, Bivalacqua TJ, Nielsen M, O’Donnell M, et al. Clarification of bladder cancer disease states following treatment of patients with intravesical BCG. Bladder Cancer. 2015;1:29–30. doi: 10.3233/BLC-159002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 11.Paik I, Lee G, Lee J, Kwak TY, Ha HK. Artificial intelligence-driven digital pathology in urological cancers: current trends and future directions. Prostate Int. 2025;13:181–190. doi: 10.1016/j.prnil.2025.02.002. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 12.BC Cancer. BC Cancer Protocol Summary GUBGEM [Internet] BC Cancer; 2019. [cited 2024 Nov 2]. Available from: https://www.bccancer.bc.ca/chemotherapy-protocols-site/Documents/Genitourinary/GUBGEM_Protocol.pdf. [Google Scholar]
- 13.U.S. Department of Health and Human Services (HHS) Common Terminology Criteria for Adverse Events (CTCAE) version 5.0 [Internet] HHS; 2017. [cited 2024 Sep 15]. Available from: https://dctd.cancer.gov/research/ctep-trials/for-sites/adverse-events/ctcae-v5-5x7.pdf. [Google Scholar]
- 14.Lamm DL, Blumenstein BA, Crissman JD, Montie JE, Gottesman JE, Lowe BA, et al. Maintenance bacillus Calmette-Guerin immunotherapy for recurrent TA, T1 and carcinoma in situ transitional cell carcinoma of the bladder: a randomized Southwest Oncology Group Study. J Urol. 2000;163:1124–1129. [PubMed] [Google Scholar]
- 15.Kim GK, Jo YH, Lee J, Han HH, Ham WS, Jang WS, et al. Short-term outcomes of intravesical gemcitabine for non-muscle-invasive bladder cancer after recent approval for use in Korea. Investig Clin Urol. 2024;65:435–441. doi: 10.4111/icu.20240047. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 16.Centers for Disease Control and Prevention (CDC) Catheter-associated urinary tract infections (CAUTI): summary of recommendations [Internet] CDC; 2024. [cited 2025 Jan 5]. Available from: https://www.cdc.gov/infection-control/hcp/cauti/summary-of-recommendations.html. [Google Scholar]
- 17.Kim JS, Lee J, Lee CU, Nguyen TT, Choi SY. Impact of regular cystoscopy on prognosis in non-muscle-invasive bladder cancer: a nationwide study. Investig Clin Urol. 2025;66:302–310. doi: 10.4111/icu.20240362. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 18.Steinberg RL, Thomas LJ, O’Donnell MA, Nepple KG. Sequential intravesical gemcitabine and docetaxel for the salvage treatment of non-muscle invasive bladder cancer. Bladder Cancer. 2015;1:65–72. doi: 10.3233/BLC-150008. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 19.Necchi A, Roumiguié M, Kamat AM, Shore ND, Boormans JL, Esen AA, et al. Pembrolizumab monotherapy for high-risk non-muscle-invasive bladder cancer without carcinoma in situ and unresponsive to BCG (KEYNOTE-057): a single-arm, multicentre, phase 2 trial. Lancet Oncol. 2024;25:720–730. doi: 10.1016/S1470-2045(24)00178-5. [DOI] [PubMed] [Google Scholar]
- 20.Boorjian SA, Alemozaffar M, Konety BR, Shore ND, Gomella LG, Kamat AM, et al. Intravesical nadofaragene firadenovec gene therapy for BCG-unresponsive non-muscle-invasive bladder cancer: a single-arm, open-label, repeat-dose clinical trial. Lancet Oncol. 2021;22:107–117. doi: 10.1016/S1470-2045(20)30540-4. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 21.Claps F, Pavan N, Ongaro L, Tierno D, Grassi G, Trombetta C, et al. BCG-unresponsive non-muscle-invasive bladder cancer: current treatment landscape and novel emerging molecular targets. Int J Mol Sci. 2023;24:12596. doi: 10.3390/ijms241612596. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 22.Huelster HL, Mason NT, Davaro F, Naqvi SMH, Kim Y, Gilbert SM. Cost-utility of initial management of high-grade T1 bladder cancer with intravesical BCG vs immediate radical cystectomy. Urology. 2024;187:106–113. doi: 10.1016/j.urology.2024.02.033. [DOI] [PubMed] [Google Scholar]
- 23.Bukavina L, Bell S, Packiam VT, Smaldone M, Abbosh P, Uzzo R, et al. Sequential intravesical gemcitabine-docetaxel vs. bacillus Calmette-Guerin (BCG) in the treatment of non-muscle invasive bladder cancer: a preliminary cost-effectiveness analysis. Urol Oncol. 2023;41:391.e1–391.e4. doi: 10.1016/j.urolonc.2023.04.005. [DOI] [PubMed] [Google Scholar]
- 24.Scilipoti P, Moschini M, Li R, Lerner SP, Black PC, Necchi A, et al. The financial burden of localized and metastatic bladder cancer. Eur Urol. 2025;87:536–550. doi: 10.1016/j.eururo.2024.12.002. [DOI] [PubMed] [Google Scholar]
- 25.D’Andrea D, Mostafid H, Gontero P, Shariat S, Kamat A, Masson-Lecomte A, et al. Unmet need in non-muscle-invasive bladder cancer failing Bacillus Calmette-Guerin therapy: a systematic review and cost-effectiveness analyses from the International Bladder Cancer Group. Eur Urol Oncol. 2025;8:216–229. doi: 10.1016/j.euo.2024.10.012. [DOI] [PubMed] [Google Scholar]
- 26.Korkes F, Timóteo F, Soledade LCB, Bugalho LS, Peixoto GA, Teich VD, et al. Stage-related cost of treatment of bladder cancer in Brazil. Pharmacoecon Open. 2022;6:461–468. doi: 10.1007/s41669-022-00325-7. [DOI] [PMC free article] [PubMed] [Google Scholar]
- 27.Williams SB, Howard LE, Foster ML, Klaassen Z, Sieluk J, De Hoedt AM, et al. Estimated costs and long-term outcomes of patients with high-risk non-muscle-invasive bladder cancer treated with Bacillus Calmette-Guerin in the Veterans Affairs Health System. JAMA Netw Open. 2021;4:e213800. doi: 10.1001/jamanetworkopen.2021.3800. [DOI] [PMC free article] [PubMed] [Google Scholar]