Welcome, this website is intended for all international healthcare professionals in uro-oncology. By clicking the link below you are declaring and confirming that you are a healthcare professional.

You are here

Prognostic Factors and Risk Groups in T1G3 Non–Muscle-invasive Bladder Cancer Patients Initially Treated with Bacillus Calmette-Guérin: Results of a Retrospective Multicenter Study of 2451 Patients

European Urology



The impact of prognostic factors in T1G3 non–muscle-invasive bladder cancer (BCa) patients is critical for proper treatment decision making.


To assess prognostic factors in patients who received bacillus Calmette-Guérin (BCG) as initial intravesical treatment of T1G3 tumors and to identify a subgroup of high-risk patients who should be considered for more aggressive treatment.

Design, setting, and participants

Individual patient data were collected for 2451 T1G3 patients from 23 centers who received BCG between 1990 and 2011.

Outcome measurements and statistical analysis

Using Cox multivariable regression, the prognostic importance of several clinical variables was assessed for time to recurrence, progression, BCa-specific survival, and overall survival (OS).

Results and limitations

With a median follow-up of 5.2 yr, 465 patients (19%) progressed, 509 (21%) underwent cystectomy, and 221 (9%) died because of BCa. In multivariable analyses, the most important prognostic factors for progression were age, tumor size, and concomitant carcinoma in situ (CIS); the most important prognostic factors for BCa-specific survival and OS were age and tumor size. Patients were divided into four risk groups for progression according to the number of adverse factors among age ≥70 yr, size ≥3 cm, and presence of CIS. Progression rates at 10 yr ranged from 17% to 52%. BCa-specific death rates at 10 yr were 32% in patients ≥70 yr with tumor size ≥3 cm and 13% otherwise.


T1G3 patients ≥70 yr with tumors ≥3 cm and concomitant CIS should be treated more aggressively because of the high risk of progression.

Patient summary

Although the majority of T1G3 patients can be safely treated with intravesical bacillus Calmette-Guérin, there is a subgroup of T1G3 patients with age ≥70 yr, tumor size ≥3 cm, and concomitant CIS who have a high risk of progression and thus require aggressive treatment.

Take Home Message

In T1G3 patients initially treated with at least an induction course of bacillus Calmette-Guérin, the subgroup aged >70 yr with tumor size >3 cm and concomitant carcinoma in situ has an unfavorable outcome and should be considered for early, more aggressive treatment.

Keywords: Bacillus Calmette-Guérin, BCG, Non–muscle-invasive bladder cancer, Prognostic factors, T1G3.

1. Introduction

T1G3 is considered to be a high-risk subgroup of non–muscle-invasive bladder cancer (NMIBC). Its natural history suggests an unfavorable long-term outcome, as documented by early untreated series reporting 27–65% disease progression rates [1] and a 34% cancer-specific death rate [2] .

Bacillus Calmette-Guérin (BCG) is currently viewed as the gold standard conservative treatment option for T1G3 tumors [3] and [4]. This policy is based on several inconsistent retrospective series reporting a 5-yr disease-specific survival of up to 80% [5] . Meta-analyses including relatively small numbers of T1G3 patients have reached conflicting conclusions concerning the ability of BCG to reduce the risk of progression [6] and [7]. One-third of T1G3 patients will eventually progress under BCG [5] with a considerable risk of dying of their disease because of delaying radical surgery [8] . Early cystectomy, advocated by some authors as an alternative to BCG, has shown long-term survival rates not exceeding 80% [9] , meaning that it does not guarantee cure for T1G3 tumors. The current understanding of T1G3 disease suggests that BCG is a feasible first-line treatment option, while cystectomy should be recommended in the presence of unfavorable prognostic factors [3], [4], [10], and [11]. Integrating prognostic factors to develop risk groups would be particularly helpful in clinical decision making for T1G3 patients. Current scoring systems that predict clinical outcomes in NMIBC [12] and [13] are inadequate for this purpose because of the low rate of T1G3 patients in these series.

The aim of the current study is to assess outcome-related prognostic factors in a large cohort of patients who received BCG as initial treatment of T1G3 tumors and to identify subgroups of high-risk patients who should be considered for more aggressive treatment.

2. Patients and methods

2.1. Study design

This is a multicenter retrospective study including patients from 23 different centers.

2.2. Inclusion criteria

Patients who have histologically confirmed T1G3 tumors (World Health Organization [WHO] 1973) or T1 high-grade tumors (International Society of Urological Pathology 1998/WHO 2004) on bladder biopsy or transurethral resection (TUR) and who received at least an induction course of BCG as their initial intravesical treatment for a T1G3/high-grade tumor from 1990 to 2011 were eligible. Patients with a previous NMIBC that was not T1G3/high grade were eligible as long as they did not receive BCG for that tumor.

2.3. Exclusion criteria

Patients with a history of muscle-invasive disease (T2 or higher) or upper tract urothelial cancer, patients with nonurothelial carcinoma, patients who previously received BCG for a tumor that was not T1G3/high grade, or patients who did not receive BCG as initial intravesical treatment for a T1G3/high-grade tumor were ineligible.

2.4. Database construction, prognostic factors, and outcome variables

Based on an updated literature search on NMIBC and T1G3 conducted in 2010, potential prognostic factors were identified. Individual patient data were requested for the following patient and tumor characteristics and were included in the database: age, gender, smoking history and intensity, exposure to chemical compounds, tumor status (primary or recurrent), previous intravesical chemotherapy, tumor size (<3 cm vs ≥3 cm), tumor focality (solitary vs multiple), presence of carcinoma in situ (CIS), urethral involvement with or without stromal invasion, presence of muscle in the tissue specimen, restaging TUR, and results of pathology at restaging.

Information on BCG dose, total number of instillations, toxicity, dose reduction for toxicity, and reasons for stopping BCG was also recorded. Any instillations beyond six were defined as maintenance BCG.

The following end points were assessed: time to first recurrence, progression to muscle-invasive disease, and the duration of cancer-specific survival (CSS) and overall survival (OS).

Individual patient data were transferred to the secretariat in electronic format. Data quality control was carried out, and queries for inconsistent and missing data were sent back to the participating centers for resolution. All patients from centers with insufficient data quality were excluded.

2.5. Statistical analysis

Times to events were calculated taking the date of starting BCG as time zero. OS was estimated using the Kaplan-Meier technique. To take into account patients who died before observing the event of interest (competing risk), times to the other events were estimated using cumulative incidence functions. Patients without an event or death prior to the event were censored at the last date of follow-up.

Univariable and step-down multivariable Cox proportional hazards regression models using a significance level of 0.01 were used to identify prognostic factors related to the end points of interest. Risk groups were formed based on the factors found to be of prognostic significance for progression and bladder cancer (BCa)–specific survival in the multivariable models.

3. Results

In November 2010, 25 centers agreed to take part in the study. Individual patient data from each center were checked electronically, and queries were sent back to investigators to provide missing data and correct inconsistencies. All patients from two centers were excluded for quality control reasons. Patients from 23 centers were retained for the study, with between 9 and 396 patients per center, for a total of 2451 patients who met the eligibility criteria (Supplemental Table 1).

3.1. Baseline characteristics and treatment

Baseline patient information is reported in Table 1 . Median age of patients was 68 yr, 82% were male, 89% were primary T1G3, 56% had multifocal disease, 46% had tumors <3 cm, 24% had concomitant CIS, and 38% had a restaging TUR.

Table 1 Baseline characteristics of the 2451 T1G3 patients

Variable No. (%)
Age, yr
 <70 1390 (56.7)
 ≥70 1061 (43.3)
 68 Median
 60–74 Interquartile range
 Male 2012 (82.1)
 Female 439 (17.9)
Smoking history
 Never smoked 469 (19.1)
 Stopped smoking 618 (25.2)
 Current smoker 465 (19.0)
 Missing/unknown 899 (36.7)
Exposure to chemical compounds
 No 968 (39.5)
 Yes 86 (3.5)
 Missing/unknown 1397 (57.0)
Tumor status
 Primary T1G3 2179 (88.9)
 Recurrent after non-T1G3 272 (11.1)
Previous intravesical chemotherapy
 No 2320 (94.7)
 Yes 131 (5.3)
Muscle in TUR specimen
 No 416 (17.0)
 Yes 1768 (72.1)
 Missing/unknown 267 (10.9)
Tumor grade
 WHO 1973 grade 3 1703 (69.5)
 WHO 2004 high grade 1780 (72.6)
 Grade 3 and/or high grade 2451 (100)
Tumor focality
 Solitary 964 (39.3)
 Multiple 1365 (55.7)
 Missing/unknown 122 (5.0)
Largest tumor diameter, cm
 <3 1137 (46.4)
 ≥3 560 (22.9)
 Missing/unknown 754 (30.8)
Concomitant CIS
 No 1852 (75.6)
 Yes 599 (24.4)
Invasion of prostatic urethra
 No 1337 (54.6)
 Yes, without stromal invasion 44 (1.8)
 Yes, with stromal invasion 5 (0.2)
 Missing/unknown 1065 (43.4)
Restaging TUR before BCG
 No 1342 (54.8)
 Yes 935 (38.2)
 Missing/unknown 174 (7.1)
Pathology at restaging TUR *
 No residual tumor 267 (28.6)
 Ta 378 (40.4)
 T1 289 (30.9)
 Missing/unknown 1 (0.1)

* Information on patients with muscle invasion at restaging TUR was not available, as muscle invasion before starting BCG was an exclusion criterion.

BCG = bacillus Calmette-Guérin; CIS = carcinoma in situ; TUR = transurethral resection; WHO = World Health Organization.

Table 2 provides information on the number of BCG instillations, the dose, and reasons for stopping treatment. Thirty-eight percent of the cohort received some sort of maintenance.

Table 2 Bacillus Calmette-Guérin treatment schedule and tolerability

Variable No. (%)
Number of BCG instillations
 6 (induction only) 1515 (61.8)
 7–15 618 (25.2)
 16–21 226 (9.2)
 ≥22 92 (3.8)
BCG dose
 Full 2337 (97.0)
 One-third 49 (2.0)
 One-fourth 17 (0.7)
 Missing/unknown 8 (0.3)
Reasons for stopping BCG
 Planned treatment completed 1875 (76.5)
 Local toxicity 84 (3.4)
 Systemic toxicity 32 (1.3)
 Local and systemic toxicity 21 (0.9)
 Patient refusal 25 (1.0)
 Patient lost to follow-up 15 (0.6)
 Recurrence 141 (5.8)
 Death 8 (0.3)
 Other 48 (2.0)
 Missing/unknown 202 (8.2)

BCG = bacillus Calmette-Guérin.

3.2. Clinical outcome

Clinical outcome is reported in Table 3 . At a median follow-up of 5.2 yr and a maximum follow-up of 18.7 yr (interquartile range: 2.6–8.5), 1244 patients (51%) had a recurrence, and 465 patients (19%) progressed to muscle-invasive disease. Progression to extravesical, locoregional, and/or systemic disease was documented in 221 patients (9%). Five hundred and nine patients (21%) eventually underwent cystectomy, 205 (40%) for non–muscle-invasive recurrence and 288 (57%) for muscle-invasive disease. Pathology at cystectomy revealed CIS in 206 patients (40%) and nodal disease in 72 of cases (14%). A total of 596 patients (24%) died, 221 (9%) because of BCa. Of the 221 BCa-specific deaths, 113 (51%) occurred in patients who received cystectomy. As shown in Fig 1, Fig 2, Fig 3, and Fig 4, the 10-yr recurrence, progression, overall death, and BCa-specific death rates were 58.3% (95% confidence interval, 55.7–60.9), 23.3% (95% CI, 21.0–25.6), 41.5% (95% CI, 38.3–44.9), and 14.8% (95% CI, 12.2–17.4), respectively.

Table 3 Clinical outcome of T1G3 patients treated with bacillus Calmette-Guérin

Variable No. (%)
 No 1207 (49.2)
 Yes 1244 (50.8)
Stage at first recurrence
 Ta 450 (36.2)
 T1 455 (36.6)
 T2 229 (18.4)
 T3 30 (2.4)
 T4 18 (1.4)
 CIS alone 27 (2.2)
 Extravesical 15 (1.2)
 Missing/unknown 20 (1.6)
Grade at first recurrence
 G1–G2/low grade 379 (30.5)
 G3/high grade 804 (64.6)
 CIS alone 27 (2.2)
 Extravesical 15 (1.2)
 Missing/unknown 19 (1.5)
CIS at first recurrence
 No 798 (64.2)
 Yes 418 (33.6)
 Missing/unknown 28 (2.2)
CIS recurrence
 No 1956 (79.8)
 Yes 495 (20.2)
Progression to muscle invasion (≥pT2)
 No 1986 (81.0)
 Yes 465 (19.0)
Stage at first progression
 T2 347 (74.6)
 T3 54 (11.6)
 T4 23 (5.0)
 Missing/unknown 41 (8.8)
Extravesical recurrence
 No 1857 (75.8)
 Locoregional 73 (3.0)
 Systemic 87 (3.6)
 Locoregional and systemic 61 (2.5)
 Missing/unknown 373 (15.2)
Radical cystectomy
 No 1942 (79.2)
 Yes 509 (20.8)
Age at cystectomy, yr
 20–49 38 (7.47)
 50–59 79 (15.52)
 60–69 193 (37.92)
 70–79 164 (32.22)
 80–89 34 (6.68)
 ≥90 1 (0.20)
Pathologic stage at cystectomy
 T0 103 (20.2)
 T1 92 (18.1)
 T2 135 (26.5)
 T3 116 (22.8)
 T4 37 (7.3)
 CIS alone 10 (2.0)
 Missing/unknown 16 (3.1)
Timing of cystectomy (clinical tumor status at cystectomy)
 No recurrence 9 (1.8)
 TaT1 76 (14.9)
 <T1G3 30 (5.9)
 T1G3 33 (6.5)
 Unknown 13 (2.6)
 CIS 87 (17.1)
 Progression 337 (66.2)
 Before cystectomy 226 (44.4)
 At cystectomy 111 (21.8)
CIS at cystectomy
 No 284 (55.8)
 Yes 206 (40.5)
 Missing/unknown 19 (3.7)
Nodal status at cystectomy
 N0 254 (49.9)
 N+ 72 (14.1)
 Missing/unknown 183 (36.0)
Survival status
 Alive 1855 (75.7)
 Dead (all causes) 596 (24.3)
 Dead (bladder cancer) 221 (9.0)
Cause of death
 Bladder cancer 221 (37.1)
 Other neoplasms 81 (13.6)
 Treatment related 14 (2.4)
 Chronic disease 126 (21.1)
 Missing/unknown 154 (25.8)

CIS = carcinoma in situ.


Fig. 1 Cumulative incidence curve for time to recurrence.


Fig. 2 Cumulative incidence curve for time to progression.


Fig. 3 Kaplan-Meier curve for duration of overall survival.


Fig. 4 Cumulative incidence curve for time to bladder cancer death.

3.3. Prognostic factors

Table 4, Table 5, Table 6, and Table 7 report the results of the univariable and multivariable analyses of prognostic factors for recurrence, progression, OS, and BCa-specific survival, respectively. Gender was found to be significantly associated with progression in the univariable analysis but not in the multivariable one. Similarly, T1G3 progressing from an NMIBC of lower stage and/or lower grade significantly affected both CSS and OS solely in the univariable analysis. In the multivariable models, tumor size was the only variable that independently predicted all four outcome measures. Tumor multiplicity predicted only recurrence and concomitant CIS predicted only tumor progression, while age ≥70 yr had an independent negative impact on progression and both overall and BCa-specific survival. When BCG was administered with maintenance, an independent protective effect on all outcome measures was observed.

Table 4 Univariable and multivariable analyses of time to recurrence

  Univariable Multivariable
Factor HR (95% CI) p value HR (95% CI) p value
Age, yr       NS
 <70, ≥70 1.12 (1.00–1.25) 0.047    
Sex       NS
 Male, female 1.07 (0.93–1.24) 0.32    
Tumor status       NS
 Primary, recurrent 1.02 (0.85–1.22) 0.82    
No. of tumors
 Single, multiple 1.38 (1.23–1.56) <0.001 1.28 (1.12–1.47) <0.001
Tumor size, cm
 <3, ≥3 1.37 (1.19–1.58) <0.001 1.33 (1.15–1.53) <0.001
Concomitant CIS       NS
 No, yes 1.24 (1.09–1.40) 0.001    
Maintenance BCG
 No, yes 0.60 (0.53–0.67) <0.001 0.61 (0.53–0.70) <0.001

BCG = bacillus Calmette-Guérin; CI = confidence interval; CIS = carcinoma in situ; HR = hazard ratio; NS = excluded from final model because not statistically significant.

Table 5 Univariable and multivariable analyses of time to progression

  Univariate Multivariate
Factor HR (95% CI) p value HR (95% CI) p value
Age, yr
 <70, ≥70 1.44 (1.20–1.73) <0.001 1.36 (1.11–1.67) 0.003
Sex       NS
 Male, female 1.31 (1.05–1.64) 0.015    
Tumor status       NS
 Primary, recurrent 1.15 (0.87–1.53) 0.32    
No. of tumors       NS
 Single, multiple 1.04 (0.86–1.26) 0.66    
Tumor size, cm
 <3, ≥3 1.91 (1.55–2.34) <0.001 1.85 (1.51–2.28) <0.001
Concomitant CIS
 No, yes 1.41 (1.16–1.71) 0.001 1.46 (1.17–1.82) 0.001
Maintenance BCG
 No, yes 0.78 (0.64–0.94) 0.01 0.73 (0.59–0.90) 0.004

BCG = bacillus Calmette-Guérin; CI = confidence interval; CIS = carcinoma in situ; HR = hazard ratio; NS = excluded from final model because not statistically significant.

Table 6 Univariable and multivariable analyses of duration of survival

  Univariate Multivariate
Factor HR (95% CI) p value HR (95% CI) p value
Age, yr
 <70, ≥70 2.75 (2.33–3.24) <0.001 2.45 (2.03–2.97) <0.001
Sex       NS
 Male, female 0.91 (0.74–1.13) 0.40    
Tumor status       NS
 Primary, recurrent 1.46 (1.16–1.85) 0.002    
No. of tumors       NS
 Single, multiple 1.04 (0.88–1.23) 0.65    
Tumor size, cm
 <3, ≥3 1.61 (1.34–1.94) <0.001 1.52 (1.26–1.83) <0.001
Concomitant CIS       NS
 No, yes 1.08 (0.91–1.29) 0.38    
Maintenance BCG        
 No, yes 0.76 (0.64–0.90) 0.001 0.74 (0.61–0.90) 0.002

BCG = bacillus Calmette-Guérin; CI = confidence interval; CIS = carcinoma in situ; HR = hazard ratio; NS = excluded from final model because not statistically significant.

Table 7 Univariable and multivariable analyses of duration of bladder cancer–specific survival

  Univariate Multivariate
Factor HR (95% CI) p value HR (95% CI) p value
Age, yr
 <70, ≥70 2.03 (1.56–2.66) <0.001 1.84 (1.36–2.49) <0.001
Sex       NS
 Male, female 1.10 (0.79–1.53) 0.56    
Tumor status       NS
 Primary, recurrent 1.52 (1.04–2.22) 0.03    
No. of tumors       NS
 Single, multiple 1.30 (0.98–1.73) 0.07    
Tumor size, cm        
 <3, ≥3 2.34 (1.74–3.15) <0.001 2.22 (1.65–2.99) <0.001
Concomitant CIS       NS
 No, yes 1.16 (0.87–1.55) 0.32    
Maintenance BCG
 No, yes 0.72 (0.54–0.96) 0.023 0.68 (0.50–0.93) 0.015

BCG = bacillus Calmette-Guérin; CI = confidence interval; CIS = carcinoma in situ; HR = hazard ratio; NS = excluded from final model because not statistically significant.

3.4. Risk groups

Patients were divided into four risk groups for progression according to the number of adverse prognostic factors among age ≥70 yr, tumor size ≥3 cm, and presence of CIS. Progression rates at 10 yr were 17.3% (95% CI, 13.0–21.6), 25.3% (95% CI, 20.9–29.7), 32.2% (95% CI, 26.5–37.9), and 52.0% (95% CI, 37.7–66.3) for patients with zero, one, two, and three adverse factors, respectively ( Fig. 5 ). Dividing patients into two risk groups for BCa-specific survival according to the presence of both age ≥70 yr and tumor ≥3 cm, the BCa-specific death rates at 10 yr were 31.7% (95% CI, 21.0–42.4) for patients with both factors present, as opposed to 12.9% (95% CI, 10.0–15.8) for patients without both of these factors. Patients with none of these factors and those with one factor had similar CSS ( Fig. 6 ).


Fig. 5 Cumulative incidence curves for time to progression according to the number of adverse prognostic factors for progression among patients ≥70 yr, tumor size ≥3 cm, and presence of carcinoma in situ.


Fig. 6 Cumulative incidence curves for bladder cancer death according to the presence of both age ≥70 yr and tumor size ≥3 cm.

4. Discussion

In the largest-ever reported series of T1G3 patients receiving BCG as primary intravesical treatment, with a median follow-up of 5.2 yr and a maximum follow-up of 18.7 yr, 19% of patients progressed and 9% died because of BCa, with 37% of progressing patients dying because of malignant disease or its treatment. Seventy-nine percent of patients retained their bladders. These favorable outcomes are comparable to recent smaller series of conservatively managed T1G3 patients [5], [11], and [14] and confirm that most T1G3 tumors can be safely treated with first-line BCG. Maintenance BCG, documented in 38% of our patients, appeared to have a significant positive impact on all outcome measures. Recently, induction-only BCG was found to have high treatment success rates in a series of patients with high-risk NMIBC that was restaged with a second TUR [15] . Our results are the first in a large series of T1G3 patients to suggest that when BCG is chosen, it should be given with maintenance. This is not a randomized comparison, however, so definitive conclusions cannot be drawn.

Since NMIBC progressing to muscle-invasive disease is known to have an unfavorable prognosis even when treated with cystectomy [8] , early identification of T1G3 patients who are destined to progress becomes of utmost importance. Apart from the benefit of BCG, a number of clinical and pathologic variables beyond grade and stage have long been recognized to affect the prognosis of NMIBC [16] . Concerning the T1G3 subcategory, treatment decision making has usually been based on a number of prognostic factors that have been assessed in small series or subgroups of this disease category [10], [11], [12], and [17]. Defining the impact of the most common clinicopathologic variables associated with T1G3 constituted the primary objective of this study.

The most striking finding was the independent prognostic value of tumor size on all outcome measures. The role of tumor size has been a matter of controversy in the recent literature. Some reports have recognized a tumor size >3 cm to be predictive of both recurrence and progression [12] or of recurrence only [18] , while other reports have failed to show any clinical value [11] and [17]. Notably, in all the reported series, T1G3 patients were largely underrepresented and considerably fewer in number compared with the present series. In our analysis, age affected not only OS but also progression and BCa-specific death. This finding is in line with that of a large series of NMIBC patients treated with BCG that also included T1G3 [14] , suggesting that BCG may be less effective in elderly patients [18] .

Concomitant CIS, another important prognostic factor in this study, is a common finding in T1G3 tumors; however, its prevalence has been shown to vary considerably, from 10% in early series [12] up to 50% in more recent reports in which multiple biopsies [11] or restaging TUR [15] were routinely adopted. This situation may account for the changing view on the role of CIS, traditionally considered a major determinant of unfavorable outcomes [12] and [19], as a factor with little [15] or no [11] prognostic importance in T1G3 patients. In our series, the rate of concomitant CIS (24%) was in line with that of a subgroup of T1G3 patients from a recent prospective series [14] . In contrast to this latter series that included a large number of intermediate- to high-risk NMIBC patients in whom CIS was linked to only disease recurrence, we confirmed the role of CIS as an independent predictor of tumor progression in this cohort of high-grade T1 tumors. Similar to previous findings [15] , tumor multiplicity predicted only disease recurrence. We failed to confirm that female gender has a negative impact on outcomes [11] . Recently, high-risk NMIBC progressing from lower-risk categories was found to be associated with significantly worse outcomes than primary high-risk NMIBC [20] . We were unable to show that T1G3 recurring from an NMIBC of lower stage and/or grade had an independent prognostic impact on outcomes as compared with primary T1G3.

Predicting the risk of disease progression or the probability of dying of disease in T1G3 patients based on their individual profile of prognostic factors remains a clinical challenge. Two previous studies have attempted to do so in large series of NMIBC patients mostly naive to intravesical therapies [12] or treated with BCG [13] in which clinical and pathologic variables were attributed a score that reflected their weight in predicting recurrence and progression. As for T1G3, the small number of T1G3 patients represents a significant limitation to the clinical value of both scoring systems. By dividing our cohort into risk groups based on the number of prognostic variables that were found to independently predict clinical outcomes (“adverse prognostic factors”), a progressive worsening of the most critical outcome measure (disease progression) was observed as the number of adverse factors that were simultaneously present in a patient increased. More specifically, the probability of progression at 10 yr increased from <17% in T1G3 patients with no adverse prognostic factors for progression to 25%, 32%, and 52% in T1G3 patients with one, two, or all three adverse prognostic factors, respectively. Age and size, but not CIS, were retained as independent predictors of CSS. The corresponding survival curves showed that patients ≥70 yr having a primary T1G3 tumor ≥3 cm are at significant risk of dying of BCa at 10 yr (31.7%), as opposed to the more promising 12.9% BCa-specific death rate in patients with either smaller tumors (<3 cm) or aged <70 yr.

This novel approach to predicting the prognosis of T1G3 tumors has significant clinical implications to guide clinicians in choosing the most appropriate treatment. The strength of our findings is supported by the fact that they are derived from a large, homogeneous series of T1G3 patients who were exposed to at least an induction course of BCG as their initial intravesical treatment. One criticism of the current study is related to the fact that our series is a retrospectively selected sample of favorable-prognosis T1G3 patients in whom BCG was more likely to be successful. Because of the retrospective design, the accuracy in reporting prognostic factors such as tumor size and CIS suffers from missing data and a lack of standardized assessment, there was no central pathology review, a second TUR was carried out in only 38% of the patients, and the assessment of BCG maintenance was not based on randomization. The validation of these findings in a prospective setting is therefore advisable.

5. Conclusions

T1G3 patients treated with at least an induction course of BCG show excellent long-term CSS, with 79% of patients retaining their bladders. The simultaneous presence of three adverse prognostic factors—age ≥70 yr, tumor size ≥3 cm, and concomitant CIS—identifies a subgroup of patients with an unfavorable outcome who should be considered for more aggressive treatment.

Author contributions: Paolo Gontero had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis.

Study concept and design: Gontero, Sylvester.

Acquisition of data: Gontero, Sylvester, Pisano, Joniau, Vander Eeckt, Serretta, Larré, Di Stasi, Van Rhijn, Witjes, Grotenhuis, Kiemeney, Colombo, Briganti, Babjuk, Malmström, Oderda, Irani, Malats, Baniel, Mano, Cai, Cha, Ardelt, Varkarakis, Bartoletti, Spahn, Johansson, Frea, Soukup, Xylinas, Dalbagni, Karnes, Shariat, Palou.

Analysis and interpretation of data: Gontero, Sylvester.

Drafting of the manuscript: Gontero.

Critical revision of the manuscript for important intellectual content: Sylvester, Joniau, Serretta, Di Stasi, Van Rhijn, Witjes, Grotenhuis, Kiemeney, Colombo, Briganti, Babjuk, Malmström, Irani, Malats, Baniel, Mano, Cai, Cha, Ardelt, Varkarakis, Bartoletti, Spahn, Johansson, Frea, Soukup, Xylinas, Dalbagni, Karnes, Shariat, Palou.

Statistical analysis: Sylvester.

Obtaining funding: Gontero.

Administrative, technical, or material support: Pisano.

Supervision: Palou.

Other (specify): None.

Financial disclosures: Paolo Gontero certifies that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: None.

Funding/Support and role of the sponsor: Banca del Piemonte, Torino, Italy, supported the data management of the study with an unrestricted grant.

Appendix A. Supplementary data



  • [1] A.M. Nieder, M. Brausi, D. Lamm, et al. Management of stage T1 tumors of the bladder: International Consensus Panel. Urology. 2005;66(Suppl 1):108-125 Crossref
  • [2] M.S. Cookson, H.W. Herr, Z.F. Zhang, S. Soloway, P.C. Sogani, W.R. Fair. The treated natural history of high risk superficial bladder cancer: 15-year outcome. J Urol. 1997;158:62-67 Crossref
  • [3] M. Babjuk, M. Burger, R. Zigeuner, et al. EAU guidelines on non-muscle-invasive urothelial carcinoma of the bladder: update 2013. Eur Urol. 2013;64:639-653 Crossref
  • [4] M.C. Hall, S.S. Chang, G. Dalbagni, et al. Guideline for the management of nonmuscle invasive bladder cancer (stages Ta, T1, and Tis): 2007 update. J Urol. 2007;178:2314-2330 Crossref
  • [5] G.S. Kulkarni, O.W. Hakenberg, J.E. Gschwend, et al. An updated critical analysis of the treatment strategy for newly diagnosed high-grade T1 (previously T1G3) bladder cancer. Eur Urol. 2010;57:60-70 Crossref
  • [6] R.J. Sylvester, A.P. van der Meijden, D.L. Lamm. Intravesical bacillus Calmette-Guerin reduces the risk of progression in patients with superficial bladder cancer: a meta-analysis of the published results of randomized clinical trials. J Urol. 2002;168:1964-1970
  • [7] P.-U. Malmström, R.J. Sylvester, D.E. Crawford, et al. An individual patient data meta-analysis of the long-term outcome of randomised studies comparing intravesical mitomycin C versus bacillus Calmette-Guérin for non-muscle-invasive bladder cancer. Eur Urol. 2009;56:247-256
  • [8] S. van den Bosch, J.A. Witjes. Long-term cancer-specific survival in patients with high-risk, non-muscle-invasive bladder cancer and tumour progression: a systematic review. Eur Urol. 2011;60:493-500 Crossref
  • [9] R.E. Hautmann, B.G. Volkmer, K. Gust. Quantification of the survival benefit of early versus deferred cystectomy in high-risk non-muscle invasive bladder cancer (T1 G3). World J Urol. 2009;27:347-351 Crossref
  • [10] S. Denzinger, H.M. Fritsche, W. Otto, A. Blana, W.-F. Wieland, M. Burger. Early versus deferred cystectomy for initial high-risk pT1G3 urothelial carcinoma of the bladder: do risk factors define feasibility of bladder-sparing approach?. Eur Urol. 2008;53:146-152 Crossref
  • [11] J. Palou, R.J. Sylvester, O. Rodríguez Faba, et al. Female gender and carcinoma in situ in the prostatic urethra are prognostic factors for recurrence, progression, and disease-specific mortality in T1G3 bladder cancer patients treated with bacillus Calmette-Guérin. Eur Urol. 2012;62:118-125 Crossref
  • [12] R.J. Sylvester, A.P. van der Meijden, W. Oosterlinck, et al. Predicting recurrence and progression in individual patients with stage Ta T1 bladder cancer using EORTC risk tables: a combined analysis of 2596 patients from seven EORTC trials. Eur Urol. 2006;49:466-475
  • [13] J. Fernandez-Gomez, R. Madero, E. Solsona, et al. Predicting non muscle invasive bladder recurrence and progression in patients treated with bacillus Calmette-Guerin: the CUETO scoring model. J Urol. 2009;182:2196-2203
  • [14] J. Fernandez-Gomez, E. Solsona, M. Unda, et al. Prognostic factors in patients with non-muscle-invasive bladder cancer treated with bacillus Calmette-Guérin: multivariate analysis of data from four randomized CUETO trials. Eur Urol. 2008;53:992-1001
  • [15] H.W. Herr, G. Dalbagni, S.M. Donat. Bacillus Calmette-Guérin without maintenance therapy for high-risk non-muscle-invasive bladder cancer. Eur Urol. 2011;60:32-36 Crossref
  • [16] R.J. Sylvester. Natural history, recurrence, and progression in superficial bladder cancer. Sci World J. 2006;6:2617-2625 Crossref
  • [17] G. Jancke, J. Rosell, S. Jahnson. Impact of tumour size on recurrence and progression in Ta/T1 carcinoma of the urinary bladder. Scand J Urol Nephrol. 2011;45:388-392 Crossref
  • [18] F.N. Joudi, B.J. Smith, M.A. O’Donnell, B.R. Konety. The impact of age on the response of patients with superficial bladder cancer to intravesical immunotherapy. J Urol. 2006;175:1634-1639
  • [19] R. Hurle, A. Losa, A. Ranieri, P. Graziotti, A. Lembo. Low dose Pasteur bacillus Calmette-Guerin regimen in stage T1, grade 3 bladder cancer therapy. J Urol. 1996;156:1602-1605
  • [20] F. Thomas, A.P. Noon, N. Rubin, J.R. Goepel, J.W.F. Catto. Comparative outcomes of primary, recurrent, and progressive high-risk non-muscle-invasive bladder cancer. Eur Urol. 2013;63:145-154 Crossref


a Department of Surgical Sciences, Molinette Hospital, University of Studies of Turin, Turin, Italy

b EORTC Headquarters, Brussels, Belgium

c Oncologic and Reconstructive Urology, Department of Urology, University Hospitals Leuven, Leuven, Belgium

d Department of Urology, Paolo Giaccone General Hospital, Palermo, Italy

e Department of Surgical Science, John Radcliffe Hospital, University of Oxford, Oxford, UK

f Policlinico Tor Vergata-University of Rome, Rome, Italy

g Department of Urology, Netherlands Cancer Institute, Antoni van Leeuwenhoek Hospital, Amsterdam, The Netherlands

h Department of Urology and Department for Health Evidence, Radboud University Medical Centre, Nijmegen, The Netherlands

i Dipartimento di Urologia, Università Vita-Salute, Ospedale S. Raffaele, Milan, Italy

j Department of Urology, Motol Hospital, 2nd Faculty of Medicine, Charles, University of Prague, Prague, Czech Republic

k Department of Surgical Sciences, Uppsala University, Uppsala, Sweden

l Department of Urology, Centre Hospitalier Universitaire La Milétrie, University of Poitiers, Poitiers, France

m Department of Urology, Spanish National Cancer Research Centre-Madrid, Madrid, Spain

n Institute of Urology, Rabin Medical Center, Petach Tikva and Sackler Faculty of Medicine, Tel Aviv University, Tel Aviv, Israel

o Department of Urology, Santa Chiara Hospital, Trento, Italy

p Department of Urology, Weill Medical College of Cornell University, New York, NY, USA, and Department of Surgery, Memorial Sloan-Kettering Cancer Center, New York, NY, USA

q Facharzt fur Urologie, Abteilung fur Urologie, Chirurgische Universitatsklinik, Freiburg, Germany

r Department of Urology, Sismanoglio Hospital, University of Athens, Athens, Greece

s Urology Unit, S. Maria Annunziata Hospital, University of Florence, Florence, Italy

t Department of Urology, University Hospital of Wuerzburg, Wuertzburg, Germany

u Department of Radiation Sciences, Oncology, Umeå University, Umeå, Sweden

v Department of Urology, General Teaching Hospital and 1st Faculty of Medicine, Charles University in Praha, Praha, Czech Republic

w Cochin Hospital, Paris Descartes University, Paris, France

x Urology Service, Department of Surgery, Memorial Sloan-Kettering Cancer Center, New York, NY, USA

y Department of Urology, Mayo Clinic, Rochester, MN, USA

z Department of Urology, Comprehensive Cancer Center Medical University Vienna, Vienna, Austria

aa Department of Urology, Fundacio Puigvert, University of Barcelona, Barcelona, Spain

lowast Corresponding author. Urology Clinic, Department of Surgical Sciences, San Giovanni Battista Hospital, C.so Bramante 88/90, 10126 Turin, Italy. Tel.: +390116335581; Fax: +390116706670.