Richard P. Meijer, Department of Urology, University Medical Centre Utrecht, Utrecht, the Netherlands. e-mail: firstname.lastname@example.org
Study Type – Therapy (individual cohort)
Level of Evidence 2b
What’s known on the subject? and What does the study add?
Although it has been reported that patients with primary carcinoma in situ of the bladder (CIS) have a better prognosis than patients with concomitant or secondary CIS, the risk profiles of these three clinical types of CIS have not yet been fully clarified.
The current study was performed to give further insight into the risk profiles of these three clinical types of CIS. In clinical practice it would be helpful to be able to discriminate ‘high-risk’ from ‘low-risk’ CIS before the start of intravesical therapy and as such to tailor the treatment and follow-up to this risk-profile.
• To further clarify the risk profiles of three clinical types of carcinoma in situ (CIS) of the bladder.
MATERIALS AND METHODS
• Population-based data from the Comprehensive Cancer Centre Middle Netherlands, as part of the nationwide Netherlands Cancer Registry, were used for patients presenting with CIS in the period from 1987 to 2009.
• Patients with muscle-invasive bladder cancer on primary diagnosis were excluded. The patients were divided into three groups according to their ‘clinical type’, being primary, concomitant or secondary CIS.
• Overall, 90 patients with CIS were identified with a mean age of 63.4 years, predominantly men (91.1%). Primary CIS (P-CIS) was found in 43 patients (47.8%), concomitant CIS (C-CIS) in 21 patients (23.3%) and secondary CIS (S-CIS) in 26 patients (28.9%). Mean follow up was 81.3 months (range 8–222 months). Recurrence of disease was observed in 68.9% of patients, with significantly more recurrences in the S-CIS group (88.5%).
• Progression to muscle-invasive disease was seen in 17 patients (18.9%): eight patients (18.7%) with P-CIS, four (19.0%) with C-CIS and five (19.2%) with S-CIS. Overall, 29 patients underwent a cystectomy, equally distributed over the three groups. The duration of bladder preservation was worse in the C-CIS group but did not differ significantly between the groups.
• Overall survival at 5 years was 79.6% for the total group, with poorer results for the C-CIS group, although the difference was not significant.
• Carcinoma in situ is clearly an entity that requires meticulous treatment and thorough follow up because of its high recurrence rate (68.9%) and high rate of progression to muscle-invasive bladder cancer (18.9%).
• The C-CIS group appears to have a poorer prognosis with a shorter duration of bladder preservation and a worse overall survival.
Non-muscle-invasive bladder cancer can be subdivided into several sub-types with varying risk factors for recurrence and progression. In this broad range of malignant bladder tumours, urothelial carcinoma in situ (CIS) has been recognized as a distinctive lesion. When left untreated, CIS of the bladder may result in progression to muscle-invasive disease in approximately 50% of patients and in disease recurrence in up to 90% of patients [1,2]. CIS can be found as a primary urothelial lesion (i.e. solely CIS in primary pathology specimen) or in combination with another form of transitional cell carcinoma either concomitantly (i.e. CIS concurrent with another form of transitional cell carcinoma) or secondarily (i.e. another form of transitional cell carcinoma and discovery of CIS during follow up) [3,4]. Although, it has been reported that patients with primary CIS have a better prognosis than patients with concomitant or secondary CIS, the risk profiles of these three clinical types of CIS have not yet been fully clarified [5–7]. The standard treatment for CIS of the bladder primarily consists of intravesical BCG instillations [7–12]. In case of BCG failure, a selection of patients with CIS will be offered early cystectomy. In general, CIS is considered to be a disease that requires life-long follow up because of a considerable risk of recurrence anywhere in the urinary tract [13,14]. The current study was performed to give further insight into the risk profiles of the three clinical types of CIS of the bladder. In clinical practice it would be helpful to be able to discriminate ‘high-risk’ from ‘low-risk’ CIS before the start of intravesical therapy and as such to tailor the treatment and follow up to this risk profile.
MATERIALS AND METHODS
Population-based data from the Comprehensive Cancer Centre Middle Netherlands (CCCMN), as part of the nationwide Netherlands Cancer Registry, were used for patients presenting with CIS in the period from 1987 to 2009. The Netherlands Cancer Registry is based on notification of all newly diagnosed malignancies in the Netherlands by the automated pathological archive. Additional sources are the national registry of hospital discharge diagnoses, haematology departments and radiotherapy institutions . Information on patient characteristics such as gender, date of birth and tumour characteristics such as date of diagnosis, tumour type, International Classification of Diseases for Oncology , histopathology, stage (TNM) classification , tumour grade and initial treatment, are extracted routinely from the medical records by data managers, 6–18 months after diagnosis . The quality of the data is high because of the thorough training of the data managers and computerized consistency checks at regional and national levels. Completeness is estimated to be at least 95%. The internal review board of the CCCMN approved the disclosure of registered data from the individual urology departments of all hospitals of the CCCMN region. For completion of the study database, the individual urology departments of the CCCMN hospitals agreed to the retrospective extraction of follow up data from the patient files. These additional data consisted of follow up investigations (i.e. imaging, urethrocystoscopy, urinary cytology, histopathology), regimens of intravesical therapy, re-interventions and adjuvant therapy. Patients with muscle-invasive bladder cancer on primary diagnosis were excluded from this analysis.
The patients were divided into three groups according to their ‘clinical type’, being primary (P-), concomitant (C-) or secondary (S-) CIS. Overall survival, progression rate and recurrence rate were analysed using the Kaplan–Meier method. Cumulative probabilities and median survival are given wherever appropriate. Overall survival is defined as the time to any cause of death: patients who were alive at the end of follow up were censored at that date. Recurrence was defined as first relapse of disease in the bladder. Progression of disease was defined as the development of muscle-invasive bladder cancer. The Kaplan–Meier curves were compared using the log-rank test. Statistical significance was defined as a P value <0.05. Data analysis was performed with the Statistical Package for Social Sciences, version 15.0 for Windows (SPSS Inc., Chicago, IL, USA).
Of all patients diagnosed with bladder cancer in the CCCMN region between 1987 and 2009, 104 had CIS. Fourteen of these patients were also diagnosed with muscle-invasive bladder cancer and were excluded from this study. The remaining 90 patients corresponded to the inclusion criteria of this study. These patients had a mean age of 63.4 years (range 24–85 years), and 91.1% were men. P-CIS was found in 47.8% of patients, C-CIS in 23.3% and S-CIS in 28.9%. Urine cytology was performed in 74.4% of patients of which 64.2% was high grade and therefore suspicious for transitional cell carcinoma. Mean follow up was 81.3 months (range 8–222). The patient characteristics are shown in Table 1. The results from the initial pathological evaluation are summarized in Table 2.
Table 1. Patient characteristics
LUTS, lower urinary tract symptoms; UTI, urinary tract infection; P-CIS, primary carcinoma in situ; C-CIS, concomitant carcinoma in situ; S-CIS, secondary carcinoma in situ.
Overall, 78 of 90 patients (86.7%) were treated with intravesical BCG immunotherapy and 21 of 90 (23.3%) patients were treated with intravesical mitomycin C (MMC). As 76.2% of the patients who were treated with MMC also received BCG, either before or subsequent to MMC therapy, there is some overlap between the MMC and BCG treatment groups, leaving seven patients (7.8%) who did not receive any intravesical therapy. In the group with P-CIS 52 courses of intravesical therapy were started (four MMC; 48 BCG, median 12 instillations per course) compared with 26 courses in the C-CIS group (four MMC; 22 BCG, median of 9 instillations per course) and 44 courses in the group with S-CIS (16 MMC; 28 BCG, median of 9 instillations per course).
Table 3 shows the recurrence and progression rates. In the group with S-CIS an 88.5% recurrence rate was found, which is significantly higher than the rate found in the other two groups (vs P-CIS: P= 0.02; vs C-CIS: P= 0.04). For the P-CIS group the median time to recurrence was 34 months (range 2–136 months), for the C-CIS group it was 29 months (range 2–97 months) and for the S-CIS group it was 12 months (range 2–66 months), respectively. The median time to recurrence overall was 29 months (range 2–136 months). Despite the apparently shorter time to recurrence in the S-CIS group, the differences among the three groups, did not reach statistical significance (P= 0.076). As was stated earlier, progression was defined as the development of muscle-invasive disease. Progression to muscle invasive disease occurred in 18.9% of patients, with no significant differences among the three groups (P= 0.96). Recurrence in the upper urinary tract occurred in six patients (6.7%; two in the ureter and four in the renal pelvis).
Overall, 29 patients (32.2%) underwent a radical cystectomy, equally distributed over the three groups. In 21 of these patients (72.4%) CIS was found in the cystectomy specimen, either solitary or concomitantly. Lymph node metastasis was found in three patients: two with P-CIS and one with S-CIS. The pathology results after radical cystectomy are shown in Table 4.
Table 4. Pathology results after radical cystectomy
The duration of bladder preservation did not differ significantly among the groups (P= 0.37), although it was markedly shorter in the C-CIS group (Fig. 1). At 5 years follow up the bladder preservation rate was 67.1% in the P-CIS group, 56.9% in the C-CIS group and 80.4% in the S-CIS group.
Survival was analysed with the Kaplan–Meier method. The 5-year overall survival is 79.6% for the total group. For the sub-groups P-CIS, C-CIS and S-CIS, the 5-year overall survival is 81.2%, 65.4% and 87.0% respectively (P= 0.11; Fig. 2). Hence the overall survival is notably worse for the C-CIS group compared with the other two groups, although this difference did not reach significance.
This study substantiates the aggressive nature of CIS of the bladder, with a high risk of recurrence (68.9%) and progression to muscle-invasive disease (18.9%). Intravesical BCG immunotherapy is considered to be the first-line approach for CIS of the bladder. Several studies report on the success of intravesical BCG maintenance with disease-free rates of 45–71% and with progression rates decreasing to 14–22%[9,12,20,21]. In the present study most patients (86.7%) were treated with intravesical BCG immunotherapy, with a median of nine instillations per course. The overall recurrence rate was 68.9% with significantly more recurrences in the S-CIS group (88.5%). However, this may be because in the S-CIS group intravesical treatment with MMC was administered more frequently than in the other two groups or because the S-CIS group has already been shown to be prone to recurrence. The time to recurrence was notably shorter for the S-CIS group (median time to recurrence 12 months) than for the other two groups (overall median time to recurrence 29 months). Nevertheless no statistically significant differences were found (P= 0.076), this may be because of the relatively small sample size.
When comparing progression rates between studies one should be aware of the different definitions of progression that are being used: either progression to a higher stage of disease (≥T1) or progression to muscle-invasive disease (≥T2). The overall progression rate of 18.9% in our study is markedly higher than the rate of 13.9% that was reported in the meta-analysis of Sylvester et al., who also defined progression as the development of muscle-invasive disease . This difference could be because the meta-analysis included mostly patients who underwent BCG maintenance therapy, whereas the patients in the present study only received a median of nine BCG instillations. However, a more likely explanation may be found in the different follow up periods. A relatively short follow up of 2.5 years was reported in the meta-analysis, whereas we report a median follow up of 6.0 years. Kaasinen et al. also reported a progression rate of 17.8% but they defined progression as the development of T1 disease or higher .
The sub-classification of CIS into three clinical types (P-CIS, C-CIS and S-CIS) has been used by Jakse et al. in their study evaluating the efficacy of BCG in CIS of the bladder. In this study the three groups of CIS responded equally well to BCG. In contrast to this, several studies have reported a significantly higher risk of progression and a worse outcome for patients with CIS associated with a T1 papillary tumour compared with patients with P-CIS [5,7,9,23]. Ovesen et al. reported that progression of disease was found in 8% of patients with primary CIS compared with 57% of patients with concomitant or secondary CIS. However, in interpreting such data one should consider the possibility of a lead-time bias. In our study the progression rates did not differ significantly among the three groups.
Radical cystectomy was performed in 29 patients (32.2%), equally distributed over the three groups. In 21 of these patients (72.4%) CIS was found in the cystectomy specimen, either solitarily or concomitantly. In the literature a wide range of cystectomy rates (up to 50%) has been reported [12,24]. In the present study, the duration of bladder preservation did not differ significantly among the groups. Although the C-CIS group appears to have a worse prognosis, with a bladder preservation rate that was 12–24% lower than that in the other two groups, significance was not achieved. However, the relatively small number of patients in the three groups, which is a problem with most sub-group analyses in CIS patient groups, should be taken into account when interpreting these data.
The C-CIS group also showed a worse overall survival compared with the other two groups, although statistically significant differences were not found. At 5 years follow up the overall survival for the total group was 79.6%. This is in concordance with previous reports, quoting 5-year survival rates that vary between 67 and 87%[11,25,26].
To provide a risk profile for CIS of the bladder, Hudson and Herr  suggested a different definition of patients with high-risk CIS (i.e. diffuse CIS, prostatic urethra involvement, overexpression of p53). These patients should undergo radical cystectomy without delay in case of BCG failure, whereas those with low-risk CIS (i.e. focal CIS, lack of overexpression of p53) can be offered bladder-preserving therapy. Furthermore Shariat et al. suggested a risk profile for CIS, without concomitant muscle-invasive disease, based on p21 and p53 expression. In their study progression was defined as development of T1 disease or higher. They found that p21 expression is independently associated with recurrence and progression, whereas combined p21 and p53 overexpression resulted in the poorest recurrence, progression and survival rates. These results suggest that early definitive therapy could be advisable in this poor-risk group.
Hudson and Herr  found upper urinary tract tumours in 20% of patients with CIS with a median interval to recurrence of 38 months (range 6–81 months). Likewise, Solsona et al. reported recurrence in the upper urinary tract in 21% of patients with CIS, significantly higher than in patients without CIS (2.3%). In comparison, in the present study markedly fewer recurrences in the upper urinary tract were seen: in 6.7% of patients. Because our analysis was performed in a population-based registry, the higher recurrence rates in the reports of Solsona et al. and Hudson and Herr  may reflect the more aggressive tumour biology in patients who are treated in tertiary referral centres.
It is clear that CIS is an entity that requires meticulous treatment and thorough follow up because a recurrence rate of 68.9% was found and progression to muscle-invasive bladder cancer occurred in 18.9% of patients. There were no significant differences among the three groups with respect to duration of bladder preservation and progression to muscle-invasive disease. Nevertheless, the C-CIS group appears to have a poorer prognosis with a shorter duration of bladder preservation and a worse overall survival, although not statistically significant.
We thank the urology departments of all seven hospitals of the CCCMN region for kindly granting access to the patient files for retrospective extraction of additional data for completion of the study database.