British Journalof Urology(1992),70, 1 5 6 1 6 0
01992 British Journal of Urology
Urological Audit: the Role for an Aggressive Approach to High Grade Superficial Bladder Tumours S. R. MOHAMED, S. F. MISHRIKI, R. A. PERSAD, P. ABRAMS, J. C.GINGELL and R. C.L. FENELEY Department of Urology, Southmead Hospital, Bristol
Summary-A retrospective study was undertaken of the different treatment modalities for bladder tumours under the care of 3 consultants in the urology department of a district general hospital. The aim was to review the results of the various forms of treatment. In all, 261 patients’ case records were reviewed and 19 variables extracted. There was an average delay of 4.2 months from the onset of symptoms to the initial cystoscopy. Over 50% of high grade tumours were invasive on initial presentation (G3T2/G3T3). A range of treatments for the more aggressive tumours was adopted by the urologists, ranging from a conservative resection (TURBT) to an aggressive approach (cystourethrectomy) at the earliest sign of progression. A strong association between aggressive treatment and higher survival was noted. This study has proved valuable in demonstrating to the urology team the value of routine audit and questioning “established” surgical practice. As a result, a more standard regime for the treatment of bladder tumours has been advocated and a prospective randomised controlled trial will be introduced.
The treatment of carcinoma of the bladder is controversial. In particular, the management of superficially invasive bladder cancer remains a challenge for the urologist. The results of a cautious approach using transurethral resection and radiotherapy for invasive carcinoma and high grade superficial carcinoma have been unsatisfactory (Wallace and Bloom, 1976; Herr, 1991). Birch and Harland (1989) reported some success with immunotherapy for recurrent multiple superficial carcinoma owing to the field change concept of multiple tumour origin. They advocated 2 courses of Bacillus Calmette-Guerin (BCG), reserving radical cystectomy if there was no response. There is a natural reluctance to adopt a more aggressive approach; prolonged surgery, the fear of high morbidity, abdominal stoma and impotence, result in a tendency to avoid performing cystectomies. The good response following continence and potencyAccepted for publication 9 March 1992
preserving cystoprostatectomy and substitution cystoplasty reported by Mundy et al. (1986) has further served to question the role of radiotherapy and the value of chemotherapy. Radical surgery in the management of high grade superficial cancer (G3T1) requires an early decision in order to prevent progression of tumour, especially when associated with carcinoma in situ. It was with a view to analyse the results of our management, find the outcome of different treatments for high risk and invasive bladder carcinomas, select the best agents, regimes and alternative treatments that this retrospective study was undertaken. Patients and Methods A retrospective study of patients with bladder cancer in the Bristol area from January 1985 to October 1990 was conducted. The study included 261 new cases of carcinoma of the bladder managed by different modes of treatment. The following
156
157
UROLOGICAL AUDIT: TREATMENT MODALITIES FOR BLADDER TUMOURS
items were recorded after studying the case notes and histology reports : patient's age, sex, district of referral, symptoms and signs (microscopic haematuria, macroscopic haematuria, pain, dysuria and obstruction), interval between onset of symptoms and presentation to the general practitioner, interval between presentation to the hospital and time of first cystoscopy after investigations, relevant drug history, social history, findings on first cystoscopy and examination under anaesthesia (number, site, size of tumours and degree of fixation), treatment method, follow-up and progression. All histology results were retrieved from the histopathology reports. Information was databased (dbase I11 Plus) in order to allow sequential review and computer analysis of the data. There were 76 cases of high grade tumours (G3) and these were analysed for progression of disease.
0Not
80
known
60
40
20
0
GI
Glll
Gll
Grade
Fig. 2 Cytological findings and grades. 100%Disease-free
Results It was found that 75% of the patients were older than 65 years on presentation (Fig. 1). There was a delay in presentation of 2.7 months from the onset of symptoms and a further doctors' delay (period between the time first seen to the time management was instituted) of 1.5 months. More than 20% of the tumours were above 5 cm in size. Positive urine cytology was noted in 66% of high, 26% of moderate and 8% of low grade tumours (Fig. 2). The diseasefree interval after first transurethral resection of grade 3 tumours at 6 months was 60% and at 2 years was 10% (Fig. 3). The grade 3 tumours were analysed with regard to their staging. The majority (5 1%)invaded muscle. The non-invasive superficial group included 22 patients with carcinoma in situ (Cis). There were 39 patients with invasive disease lo. of patients 100
2o "
\
t
0:oo
0:06
0:12
0:24
Follow-up (months)
Fig. 3 Disease-free interval following cystoscopy and TURBT.
and 37 with superficial disease. Of the 39 patients in the invasive group, 18 had radiotherapy and 7 of them died within 2 years. There were 17 patients on follow-up with transurethral resection alone and 7 patients died over a follow-up period of 1 to 5 years. Four patients had a cystectomy; 1 died from post-operative peritonitis (Fig. 4). The progression of tumour in the superficial group was then analysed. There were 25 patients whose tumour progressed; 22 patients had Cis associated with the tumour. Seven patients had G3 Cis alone which progressed and the other 15 patients had G3T1 with Cis. There were 6 patients with G3T1 without Cis whose tumour progressed, thus giving a total of 25 patients whose tumour progressed to muscle-invasive disease over a 2-year period. Five of these patients had radiotherapy and all died within 2 years. Seven patients who underwent cystectomy are alive and well. Of the 3 patients who had chemotherapy with Epodyl, 1 died after completion of a single course.
2
80 -
60 37 40 -
I
20 -
30-40
41-50
51-60
61-70
71-80
Age (years)
Fig, 1 Age of patients on initial treatment.
158
BRITISH JOURNAL OF UROLOGY
G3 n.76
Superficial 3 7
T 1 & Cis
Ta & T1 15
A
Progr. 19
No progr.
A A13
D 6
lnvasive 3 9
A A 2
D 1
Progr.
Alive 24 DXT 11 TUR 10 cyst. 3
\
Dead 15 DXT 7 TUR 7 cyst. 1
No progr.
A A 5
D 1
A 8
D 1
DXT 5 (all dead) Chemotherapy 3 (1 dead) Cystectomy 7 (all alive and well) A=Alive
D=Dead
Fig. 4 Disease progression.
Discussion More than 50% of the high grade tumours were invasive at first cystoscopy. There was a total delay of 3 to 6 months before they were first cystoscoped. Wallace and Harris (1965) pointed out that most bladder tumours were probably curable when symptoms first appeared and that the survival rate of patients with infiltrating tumours fell from 65 to 25% when treatment was delayed for more than 1 month after the onset of bleeding. Eighteen patients with invasive disease had radiotherapy, mostly palliative. Their results were poor. Five patients died within 2 years. Only 3 patients had cystectomy and the rest had transurethral resection. Skinner and Leiskovsky (1984) advocated radical cystectomy with en bloc pelvic iliac node dissection as being the best surgical therapy for high grade invasive carcinoma. Droller (1985) listed the problems of radiotherapy for invasive disease. Few patients respond without recurrence for a substantial length of time. For those who respond temporarily, post-radiation cystectomy may carry a higher morbidity and mortality. Progression has been indicated by the development of muscle invasion or metastasis. Twenty-five patients had progression of the tumour. Althausen et al. (1976) recognised the difficulty in deciding the
correct managment for G3T1 tumours. There was a 40% chance of developing invasive disease. With concurrent Cis there was an 80% chance of developing invasive disease within 18 months. The present study included 19 patients with Cis and this is consistent with other reports in the literature. In this study the average time to progression was 1.9 years. Different modalities of treatment were adopted by different clinicians in the management of high grade superficial tumours. Cystectomy was performed only when other modalities failed and/or there was clear evidence of muscle invasion. The experience of Riddle et al. (1976) indicates that radiotherapy has little to offer in the management of widespread superficial Cis. It is ineffective in controlling superficial tumours and may make matters worse by inducing telangiectasia and encouraging bleeding (Bracken et al., 1981). Epodyl, the only drug used, produced dismal results. Fitzpatrick et al. (1979) also showed disappointing results with Epodyl. The anti-tumour activity of vaccine has been established as most dramatic on Cis, with a response rate of 75%. Sarosdy and Lamm (1989) found long-term benefits in 120 patients treated with Bacillus CalmetteGukrin who were followed up for 13 to 120 months ; 78% had a good initial response. Ten patients required an intensive course of therapy with a final
159
UROLOGICAL AUDIT: TREATMENT MODALITIES FOR BLADDER TUMOURS G3 T I (Histology)
I
At six weeks repeat cystoscopy random mucosal biopsiesand cytology
+ ve Cis
- ve Cis
+ ve cytology
+ ve cytology/dysplasia - ve secondaries (MRI
I
or CT)
Cystectomy
B.C.G. x 1 course
I
Recurrence/progression at 6 weeks
Fit
(MRI or CT)
- ve
I
secondaries
Cystectomy
Unfit
1
Repeat B.C.G.
Surveillance
Fig. 5 Management flow chart.
success rate of 89%. A reduction in the recurrence rate is useful in disease of low grade and stage. In high grade disease the aim is long-term prevention of progression to muscle-invasive disease. No controlled study has yet demonstrated that intravesical chemotherapy can achieve this (Heney et al., 1982). Soloway (1980) and Prout (1984) showed that one-third of patients with superficial tumours will not respond to chemotherapy. These were the poorly differentiated tumours with basement membrane breakthrough (T1 tumours). Herr (1991) emphasised that the T1 tumour category (lamina propria invasion) emerged as the only significant adverse prognostic variable before and after treatment. All 25 patients in this series were potentially curable at the time of presentation.
It is our opinion that patients should be considered for radical cystectomy when there is a threat of aggression of the primary tumour and the likelihood of spread into the deeper layers of the bladder. Once the high grade tumour has penetrated the lamina propria, radical cystoprostatectomy designed to preserve the innervation of the distal sphincter mechanism and the corpora cavernosa and substitution cystoplasty as described by Mundy et al. (1986) should be the preferred treatment. It is the usual practice to remove the urethra in patients with widespread field change. This operation (cystoprostatectomy) satisfies the field changes (Cis) which are common in this series. Malkowicz et al. (1990) showed that cystectomy was highly effective in curing patients with high grade superficial disease, including those with
160 superficially invasive disease associated with nodal metastasis. If there is clear evidence of disease progression, especially if symptoms are severe, the bladder should be removed, since the disease is more extensive than apparent at cystoscopy. Magnetic resonance imaging (MRI) or CT scanning may help to define the depth of invasion, as there is a tendency to understage rather than overstage these tumours clinically (Nijhout et al., 1985). All 7 patients who underwent cystectomy for progression are alive and well. The results of early cystectomy are impressive. Stockle et al. (1987), in a series of 55 patients with G3 tumours alone, reported a 90% 5-year survival rate for patients with cystectomy for initial invasive T1 tumours, but for T1 patients with late cystectomies the 5year survival rate was 61.5%. Thus there is a need to identify patients at risk of progression and also to select the best agents, regimes and alternative strategies for the high grade superficial tumour with lamina propria invasion and associated Cis. G3TI tumours have been described as the “wolf in sheep’s clothing”. We would call it as the “silver lining amongst the clouds” and take the opportunity to offer the patient the option of cure with cystoprostatectomy, especially when associated with Cis. Our flow chart (Fig. 5) may be helpful, although a controlled clinical trial is necessary. References Athausen, A. F., Prout, G. R., Jr and Daly, J. J. (1976). Noninvasive papillary carcinoma of the bladder associated with carcinoma in situ.J . Urol., 116, 575-580. Birch, B. R. P. and Harland, S. J. (1989). The pTlG3 bladder tumour. Br. J . Urol.,64, 109-116. Bracken, R. B., McDonald, W. M. and Johnson, D. E. (1981). Cystectomy for superficial bladder cancer. Urology, 18, 459463. Droller, M. J. (1985). The natural history of invasive bladder cancer and the case against definitive radiation therapy. World J . Urol., 3, 86-93. Fitzpatrick,J. M., Khan, O., Oliver, R. T. D. etal. (1979). Longterm follow-up in patients with superficial bladder tumours treated with intravesical Epodyl. Br. J . Urol.,51, 545-548.
BRITISH JOURNAL OF UROLOGY
Heney, N. M., Nocks, B. N., Daly, J. J. e t d (1982). Ta and TI bladder cancer: location, recurrence and progression. Br. J . Urol.,54, 152-157. Herr, H. W. (1991). Progression of stage T1 bladder tumours after intravesical Bacillus Calmette-Gutrin. J . Urol., 145,4044. Malkowicz, S. B., Nichols, P., Lieskovsky, G. et al. (1990). The role of radical cystectomy in the management of high grade superficial bladder cancer (PA, P1, PIS and P2). J . Urol., 144, 641-645. Mundy, A. R., Nurse, D. E., Dick, J. A. et al. (1986). Continence and potency preserving cystoprostatectomy and substitution cystoplasty for patients with bladder cancer. Br. J . Urol.,58, 664-668. Nijhout, M. A. E. M., Falke, T. H. M., Jones, B. et al. (1985). Magnetic resonance imaging of the bladder and prostate. World J . Urol.,3, 6 6 7 2 . Prout, G . R. (1984). Superficial bladder cancer. In Bladder Cancer, ed. Smith, P. H. and Prout, G. R. Pp.151-171. London : Butterworth. Riddle, P. R., Chisholm, G . D., Trott, P. A. et d (1976). Flat carcinoma in situ of bladder. Br. J . Urol.,47,829-833. Sarosdy, M. F. and Lamm, D. L. (1989). Long-term results of intravesical Bacillus Calmette-Gutrin therapy for superficial bladder cancer. J . Urol., 142, 719-722. Skinner, D. G . and Leiskovsky, G . (1984). Contemporary cystectomy with pelvic node dissection compared to preoperative radiation therapy plus cystectomy in management of invasive bladder cancer. J . Urol., 131, 1069-1072. Soloway, M. S. (1980). The management of superficial bladder cancer. Cancer, 45, 18561865. Stockle, M., Alken, P., Engeimann, U. et d (1987). Radical cystectomy--often too late? Eur. Urol., 13,361-367. Wallace, D. M. and Harris, D. L. (1965). Delay in treating bladder tumours. Lancet, 2, 332-334. Wallace, D. M. and Bloom, H. J. G. (1976). The management of deeply infiltrating (T3) bladder carcinoma : controlled trial of radical radiotherapy versus pre-operative radiotherapy and radical cystectomy. Br. J . Urol.,48, 587-594.
The Authors S. R. Mohamed, MS, Commonwealth Medical Fellow S. F. Mishriki, FRCS, Research Fellow. R. A. Persad, FRCS, Registrar. P. Abrams, FRCS, Consultant Urologist. J. C. Gingell, FRCS, Consultant Urologist. R. C. L. Feneley, ChM, FRCS, Consultant Urologist. Requests for reprints to: S. F. Mishriki, Lithotriptor Unit, Southmead Hospital, Westbury-on-Trym, Bristol BSlO 5NB.
01992 British Journal of Urology
Urological Audit: the Role for an Aggressive Approach to High Grade Superficial Bladder Tumours S. R. MOHAMED, S. F. MISHRIKI, R. A. PERSAD, P. ABRAMS, J. C.GINGELL and R. C.L. FENELEY Department of Urology, Southmead Hospital, Bristol
Summary-A retrospective study was undertaken of the different treatment modalities for bladder tumours under the care of 3 consultants in the urology department of a district general hospital. The aim was to review the results of the various forms of treatment. In all, 261 patients’ case records were reviewed and 19 variables extracted. There was an average delay of 4.2 months from the onset of symptoms to the initial cystoscopy. Over 50% of high grade tumours were invasive on initial presentation (G3T2/G3T3). A range of treatments for the more aggressive tumours was adopted by the urologists, ranging from a conservative resection (TURBT) to an aggressive approach (cystourethrectomy) at the earliest sign of progression. A strong association between aggressive treatment and higher survival was noted. This study has proved valuable in demonstrating to the urology team the value of routine audit and questioning “established” surgical practice. As a result, a more standard regime for the treatment of bladder tumours has been advocated and a prospective randomised controlled trial will be introduced.
The treatment of carcinoma of the bladder is controversial. In particular, the management of superficially invasive bladder cancer remains a challenge for the urologist. The results of a cautious approach using transurethral resection and radiotherapy for invasive carcinoma and high grade superficial carcinoma have been unsatisfactory (Wallace and Bloom, 1976; Herr, 1991). Birch and Harland (1989) reported some success with immunotherapy for recurrent multiple superficial carcinoma owing to the field change concept of multiple tumour origin. They advocated 2 courses of Bacillus Calmette-Guerin (BCG), reserving radical cystectomy if there was no response. There is a natural reluctance to adopt a more aggressive approach; prolonged surgery, the fear of high morbidity, abdominal stoma and impotence, result in a tendency to avoid performing cystectomies. The good response following continence and potencyAccepted for publication 9 March 1992
preserving cystoprostatectomy and substitution cystoplasty reported by Mundy et al. (1986) has further served to question the role of radiotherapy and the value of chemotherapy. Radical surgery in the management of high grade superficial cancer (G3T1) requires an early decision in order to prevent progression of tumour, especially when associated with carcinoma in situ. It was with a view to analyse the results of our management, find the outcome of different treatments for high risk and invasive bladder carcinomas, select the best agents, regimes and alternative treatments that this retrospective study was undertaken. Patients and Methods A retrospective study of patients with bladder cancer in the Bristol area from January 1985 to October 1990 was conducted. The study included 261 new cases of carcinoma of the bladder managed by different modes of treatment. The following
156
157
UROLOGICAL AUDIT: TREATMENT MODALITIES FOR BLADDER TUMOURS
items were recorded after studying the case notes and histology reports : patient's age, sex, district of referral, symptoms and signs (microscopic haematuria, macroscopic haematuria, pain, dysuria and obstruction), interval between onset of symptoms and presentation to the general practitioner, interval between presentation to the hospital and time of first cystoscopy after investigations, relevant drug history, social history, findings on first cystoscopy and examination under anaesthesia (number, site, size of tumours and degree of fixation), treatment method, follow-up and progression. All histology results were retrieved from the histopathology reports. Information was databased (dbase I11 Plus) in order to allow sequential review and computer analysis of the data. There were 76 cases of high grade tumours (G3) and these were analysed for progression of disease.
0Not
80
known
60
40
20
0
GI
Glll
Gll
Grade
Fig. 2 Cytological findings and grades. 100%Disease-free
Results It was found that 75% of the patients were older than 65 years on presentation (Fig. 1). There was a delay in presentation of 2.7 months from the onset of symptoms and a further doctors' delay (period between the time first seen to the time management was instituted) of 1.5 months. More than 20% of the tumours were above 5 cm in size. Positive urine cytology was noted in 66% of high, 26% of moderate and 8% of low grade tumours (Fig. 2). The diseasefree interval after first transurethral resection of grade 3 tumours at 6 months was 60% and at 2 years was 10% (Fig. 3). The grade 3 tumours were analysed with regard to their staging. The majority (5 1%)invaded muscle. The non-invasive superficial group included 22 patients with carcinoma in situ (Cis). There were 39 patients with invasive disease lo. of patients 100
2o "
\
t
0:oo
0:06
0:12
0:24
Follow-up (months)
Fig. 3 Disease-free interval following cystoscopy and TURBT.
and 37 with superficial disease. Of the 39 patients in the invasive group, 18 had radiotherapy and 7 of them died within 2 years. There were 17 patients on follow-up with transurethral resection alone and 7 patients died over a follow-up period of 1 to 5 years. Four patients had a cystectomy; 1 died from post-operative peritonitis (Fig. 4). The progression of tumour in the superficial group was then analysed. There were 25 patients whose tumour progressed; 22 patients had Cis associated with the tumour. Seven patients had G3 Cis alone which progressed and the other 15 patients had G3T1 with Cis. There were 6 patients with G3T1 without Cis whose tumour progressed, thus giving a total of 25 patients whose tumour progressed to muscle-invasive disease over a 2-year period. Five of these patients had radiotherapy and all died within 2 years. Seven patients who underwent cystectomy are alive and well. Of the 3 patients who had chemotherapy with Epodyl, 1 died after completion of a single course.
2
80 -
60 37 40 -
I
20 -
30-40
41-50
51-60
61-70
71-80
Age (years)
Fig, 1 Age of patients on initial treatment.
158
BRITISH JOURNAL OF UROLOGY
G3 n.76
Superficial 3 7
T 1 & Cis
Ta & T1 15
A
Progr. 19
No progr.
A A13
D 6
lnvasive 3 9
A A 2
D 1
Progr.
Alive 24 DXT 11 TUR 10 cyst. 3
\
Dead 15 DXT 7 TUR 7 cyst. 1
No progr.
A A 5
D 1
A 8
D 1
DXT 5 (all dead) Chemotherapy 3 (1 dead) Cystectomy 7 (all alive and well) A=Alive
D=Dead
Fig. 4 Disease progression.
Discussion More than 50% of the high grade tumours were invasive at first cystoscopy. There was a total delay of 3 to 6 months before they were first cystoscoped. Wallace and Harris (1965) pointed out that most bladder tumours were probably curable when symptoms first appeared and that the survival rate of patients with infiltrating tumours fell from 65 to 25% when treatment was delayed for more than 1 month after the onset of bleeding. Eighteen patients with invasive disease had radiotherapy, mostly palliative. Their results were poor. Five patients died within 2 years. Only 3 patients had cystectomy and the rest had transurethral resection. Skinner and Leiskovsky (1984) advocated radical cystectomy with en bloc pelvic iliac node dissection as being the best surgical therapy for high grade invasive carcinoma. Droller (1985) listed the problems of radiotherapy for invasive disease. Few patients respond without recurrence for a substantial length of time. For those who respond temporarily, post-radiation cystectomy may carry a higher morbidity and mortality. Progression has been indicated by the development of muscle invasion or metastasis. Twenty-five patients had progression of the tumour. Althausen et al. (1976) recognised the difficulty in deciding the
correct managment for G3T1 tumours. There was a 40% chance of developing invasive disease. With concurrent Cis there was an 80% chance of developing invasive disease within 18 months. The present study included 19 patients with Cis and this is consistent with other reports in the literature. In this study the average time to progression was 1.9 years. Different modalities of treatment were adopted by different clinicians in the management of high grade superficial tumours. Cystectomy was performed only when other modalities failed and/or there was clear evidence of muscle invasion. The experience of Riddle et al. (1976) indicates that radiotherapy has little to offer in the management of widespread superficial Cis. It is ineffective in controlling superficial tumours and may make matters worse by inducing telangiectasia and encouraging bleeding (Bracken et al., 1981). Epodyl, the only drug used, produced dismal results. Fitzpatrick et al. (1979) also showed disappointing results with Epodyl. The anti-tumour activity of vaccine has been established as most dramatic on Cis, with a response rate of 75%. Sarosdy and Lamm (1989) found long-term benefits in 120 patients treated with Bacillus CalmetteGukrin who were followed up for 13 to 120 months ; 78% had a good initial response. Ten patients required an intensive course of therapy with a final
159
UROLOGICAL AUDIT: TREATMENT MODALITIES FOR BLADDER TUMOURS G3 T I (Histology)
I
At six weeks repeat cystoscopy random mucosal biopsiesand cytology
+ ve Cis
- ve Cis
+ ve cytology
+ ve cytology/dysplasia - ve secondaries (MRI
I
or CT)
Cystectomy
B.C.G. x 1 course
I
Recurrence/progression at 6 weeks
Fit
(MRI or CT)
- ve
I
secondaries
Cystectomy
Unfit
1
Repeat B.C.G.
Surveillance
Fig. 5 Management flow chart.
success rate of 89%. A reduction in the recurrence rate is useful in disease of low grade and stage. In high grade disease the aim is long-term prevention of progression to muscle-invasive disease. No controlled study has yet demonstrated that intravesical chemotherapy can achieve this (Heney et al., 1982). Soloway (1980) and Prout (1984) showed that one-third of patients with superficial tumours will not respond to chemotherapy. These were the poorly differentiated tumours with basement membrane breakthrough (T1 tumours). Herr (1991) emphasised that the T1 tumour category (lamina propria invasion) emerged as the only significant adverse prognostic variable before and after treatment. All 25 patients in this series were potentially curable at the time of presentation.
It is our opinion that patients should be considered for radical cystectomy when there is a threat of aggression of the primary tumour and the likelihood of spread into the deeper layers of the bladder. Once the high grade tumour has penetrated the lamina propria, radical cystoprostatectomy designed to preserve the innervation of the distal sphincter mechanism and the corpora cavernosa and substitution cystoplasty as described by Mundy et al. (1986) should be the preferred treatment. It is the usual practice to remove the urethra in patients with widespread field change. This operation (cystoprostatectomy) satisfies the field changes (Cis) which are common in this series. Malkowicz et al. (1990) showed that cystectomy was highly effective in curing patients with high grade superficial disease, including those with
160 superficially invasive disease associated with nodal metastasis. If there is clear evidence of disease progression, especially if symptoms are severe, the bladder should be removed, since the disease is more extensive than apparent at cystoscopy. Magnetic resonance imaging (MRI) or CT scanning may help to define the depth of invasion, as there is a tendency to understage rather than overstage these tumours clinically (Nijhout et al., 1985). All 7 patients who underwent cystectomy for progression are alive and well. The results of early cystectomy are impressive. Stockle et al. (1987), in a series of 55 patients with G3 tumours alone, reported a 90% 5-year survival rate for patients with cystectomy for initial invasive T1 tumours, but for T1 patients with late cystectomies the 5year survival rate was 61.5%. Thus there is a need to identify patients at risk of progression and also to select the best agents, regimes and alternative strategies for the high grade superficial tumour with lamina propria invasion and associated Cis. G3TI tumours have been described as the “wolf in sheep’s clothing”. We would call it as the “silver lining amongst the clouds” and take the opportunity to offer the patient the option of cure with cystoprostatectomy, especially when associated with Cis. Our flow chart (Fig. 5) may be helpful, although a controlled clinical trial is necessary. References Athausen, A. F., Prout, G. R., Jr and Daly, J. J. (1976). Noninvasive papillary carcinoma of the bladder associated with carcinoma in situ.J . Urol., 116, 575-580. Birch, B. R. P. and Harland, S. J. (1989). The pTlG3 bladder tumour. Br. J . Urol.,64, 109-116. Bracken, R. B., McDonald, W. M. and Johnson, D. E. (1981). Cystectomy for superficial bladder cancer. Urology, 18, 459463. Droller, M. J. (1985). The natural history of invasive bladder cancer and the case against definitive radiation therapy. World J . Urol., 3, 86-93. Fitzpatrick,J. M., Khan, O., Oliver, R. T. D. etal. (1979). Longterm follow-up in patients with superficial bladder tumours treated with intravesical Epodyl. Br. J . Urol.,51, 545-548.
BRITISH JOURNAL OF UROLOGY
Heney, N. M., Nocks, B. N., Daly, J. J. e t d (1982). Ta and TI bladder cancer: location, recurrence and progression. Br. J . Urol.,54, 152-157. Herr, H. W. (1991). Progression of stage T1 bladder tumours after intravesical Bacillus Calmette-Gutrin. J . Urol., 145,4044. Malkowicz, S. B., Nichols, P., Lieskovsky, G. et al. (1990). The role of radical cystectomy in the management of high grade superficial bladder cancer (PA, P1, PIS and P2). J . Urol., 144, 641-645. Mundy, A. R., Nurse, D. E., Dick, J. A. et al. (1986). Continence and potency preserving cystoprostatectomy and substitution cystoplasty for patients with bladder cancer. Br. J . Urol.,58, 664-668. Nijhout, M. A. E. M., Falke, T. H. M., Jones, B. et al. (1985). Magnetic resonance imaging of the bladder and prostate. World J . Urol.,3, 6 6 7 2 . Prout, G . R. (1984). Superficial bladder cancer. In Bladder Cancer, ed. Smith, P. H. and Prout, G. R. Pp.151-171. London : Butterworth. Riddle, P. R., Chisholm, G . D., Trott, P. A. et d (1976). Flat carcinoma in situ of bladder. Br. J . Urol.,47,829-833. Sarosdy, M. F. and Lamm, D. L. (1989). Long-term results of intravesical Bacillus Calmette-Gutrin therapy for superficial bladder cancer. J . Urol., 142, 719-722. Skinner, D. G . and Leiskovsky, G . (1984). Contemporary cystectomy with pelvic node dissection compared to preoperative radiation therapy plus cystectomy in management of invasive bladder cancer. J . Urol., 131, 1069-1072. Soloway, M. S. (1980). The management of superficial bladder cancer. Cancer, 45, 18561865. Stockle, M., Alken, P., Engeimann, U. et d (1987). Radical cystectomy--often too late? Eur. Urol., 13,361-367. Wallace, D. M. and Harris, D. L. (1965). Delay in treating bladder tumours. Lancet, 2, 332-334. Wallace, D. M. and Bloom, H. J. G. (1976). The management of deeply infiltrating (T3) bladder carcinoma : controlled trial of radical radiotherapy versus pre-operative radiotherapy and radical cystectomy. Br. J . Urol.,48, 587-594.
The Authors S. R. Mohamed, MS, Commonwealth Medical Fellow S. F. Mishriki, FRCS, Research Fellow. R. A. Persad, FRCS, Registrar. P. Abrams, FRCS, Consultant Urologist. J. C. Gingell, FRCS, Consultant Urologist. R. C. L. Feneley, ChM, FRCS, Consultant Urologist. Requests for reprints to: S. F. Mishriki, Lithotriptor Unit, Southmead Hospital, Westbury-on-Trym, Bristol BSlO 5NB.
