Clinical Oncology 26 (2014) 506e514 Contents lists available at ScienceDirect
Clinical Oncology journal homepage: www.clinicaloncologyonline.net
Original Article
Curative Therapy for Bladder Cancer in Routine Clinical Practice: A Population-based Outcomes Study C.M. Booth *y, D.R. Siemens yz, G. Li *, Y. Peng *x, W. Kong *, D.M. Berman x, W.J. Mackillop *y{ * Division
of Cancer Care and Epidemiology, Queen’s University Cancer Research Institute, Queen’s University, Kingston, Canada y Department of Oncology, Queen’s University, Kingston, Canada z Department of Urology, Queen’s University, Kingston, Canada x Department of Pathology and Molecular Medicine, Queen’s University, Kingston, Canada { Department of Public Health Sciences, Queen’s University, Kingston, Canada Received 30 January 2014; received in revised form 8 April 2014; accepted 13 May 2014
Abstract Aims: Definitive therapy of bladder cancer involves cystectomy or radiotherapy; controversy exists regarding optimal management. Here we describe the management and outcomes of patients treated in routine practice. Materials and methods: Treatment records were linked to the Ontario Cancer Registry to identify all cases of bladder cancer in Ontario treated with cystectomy or radiotherapy in 1994e2008. Practice patterns are described in three study periods: 1994e1998, 1999e2003, 2004e2008. Logistic regression, Cox model and propensity score analyses were used to evaluate factors associated with treatment choice and survival. Results: In total, 3879 cases (74%) underwent cystectomy and 1380 (26%) were treated with primary radiotherapy. Cystectomy use increased over time (66, 75, 78%), whereas radiotherapy decreased (34, 25, 22%), P < 0.001. There was substantial regional variation in the proportion of cases undergoing radiotherapy (range 16e51%). Five year cancer-specific survival (CSS) and overall survival were 40 and 36% for surgical cases and 35 and 26% for radiotherapy cases (P < 0.001). In multivariate Cox model and propensity score analyses, there was no significant difference in CSS between surgery and radiotherapy (hazard ratio 0.99, 95% confidence interval 0.91e1.08); radiotherapy was associated with slightly inferior overall survival (hazard ratio 1.08, 95% confidence interval 1.00e1.16). Conclusion: Utilisation of cystectomy for bladder cancer in routine practice has increased over time with no evidence of a significant difference in CSS between radiotherapy and cystectomy. Ó 2014 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved. Key words: Bladder cancer; cystectomy; health services; outcomes; quality of care; radiation
Introduction Initial local therapies for muscle-invasive bladder cancer include cystectomy or radiotherapy. Outcomes achieved with either modality are good in highly selected series from centres of excellence [1,2] and there is no strong evidence showing the superiority of one modality over the other. Accordingly, controversy exists as to what treatment is Author for correspondence: C.M. Booth, Division of Cancer Care and Epidemiology, Queen’s University Cancer Research Institute, 10 Stuart Street, Kingston ON, K7L 3N6, Canada. Tel: þ1-613-533-6895; Fax: þ1-613-5336794. E-mail address: [email protected] (C.M. Booth).
preferred and whether in fact there is an optimal treatment strategy. This controversy is evident in the discordance of international treatment guidelines, with some strongly recommending cystectomy over radiotherapy [3e5] and others concluding that both forms of treatment are acceptable [6,7]. These recommendations are not based on highlevel evidence, as there are no contemporary randomised controlled trials (RCTs) that have systematically compared these two treatment modalities. Population-based outcome studies provide insight into patterns of care and outcomes among patients treated in routine clinical practice [8]. They can minimise sources of selection and referral bias that often limit single institutionbased studies. These studies may also contribute to an
0936-6555/$36.00 Ó 2014 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.clon.2014.05.007
C.M. Booth et al. / Clinical Oncology 26 (2014) 506e514
understanding about effectiveness of medical therapies when there are no randomised data to guide patient care. Only a handful of population-based studies have reported practice and outcomes of cystectomy and radiotherapy for bladder cancer and these studies primarily described patients treated in the 1980s and 1990s [9e12]. It remains unclear if clinical features of bladder cancer have changed over time and how surgical and radiotherapy treatments may have evolved. We undertook the current populationbased cohort study to evaluate the use of surgery and radiotherapy for patients with bladder cancer in the contemporary era and provide insights into the outcomes achieved in routine clinical practice.
Materials and Methods Study Design and Population This was a population-based, retrospective cohort study to describe the management and outcome of muscleinvasive bladder cancer in the Canadian province of Ontario. Ontario has a population of about 13.5 million people and a single-payer universal health insurance programme. All incident cases of bladder cancer in Ontario with transitional cell, adenocarcinoma and squamous cell histology who underwent cystectomy or radical radiotherapy in 1994e2008 were included. The study population was classified into three temporal periods: 1994e1998, 1999e2003, 2004e2008. Data Sources The Ontario Cancer Registry (OCR) is a passive, population-based cancer registry that captures diagnostic and demographic information on at least 98% of all incident cases of cancer in the province of Ontario [13]. The OCR does not compile information about the extent of disease or treatment. Accordingly, we obtained surgical pathology reports for patients with cystectomy. Indicators of the socioeconomic status of the community in which patients resided at diagnosis were linked, as described previously [14]. A variety of electronic administrative health databases were linked to the OCR. Records of hospitalisation from the Canadian Institute for Health Information (CIHI) provided information about surgical interventions; these records are known to be consistent and complete [15]. The clinical databases of Ontario’s comprehensive cancer centres provided records of radiotherapy and chemotherapy. These centres are the only providers of radiotherapy in the province and the electronic radiotherapy records are known to be 95% complete and 99% accurate with respect to total dose, number of fractions, date of therapy, body region irradiated and treatment intent [16]. Provincial physician billing records and treatment records from regional cancer centres were used to identify chemotherapy utilisation e these records are created in the act of prescribing treatment. Surgical pathology reports were
507
obtained from the OCR and reviewed by a team of data abstractors. Definitions of Comorbidity, Management and Outcomes Comorbidity was classified using the Charlson Index modified for administrative data based on all non-cancer diagnoses recorded during any hospital admission within 5 years before surgery [17]. Cases treated with radical radiotherapy were identified from the radiotherapy treatment records of the regional cancer centres. Cases treated to the bladder or pelvis with curative intent were included, as were those with missing intent who were treated with 59.5 Gy over 30þ fractions. Twelve per cent (167/1380) of cases received concurrent chemotherapy with radiotherapy. Neoadjuvant and adjuvant chemotherapy were delivered to 17% (230/1380) and 5% (70/ 1380) of radiotherapy cases, respectively. A small number of surgical cases received preoperative (11/3879, 0.2%) and postoperative (37/3879, 1%) radiotherapy. Factors Associated with Treatment Modality Factors associated with treatment choice are shown in Table 2. The results from the multivariate analysis show that patients with advanced age and greater comorbidity were less likely to have cystectomy. Utilisation rates of radiotherapy ranged three-fold across geographical regions (range 16e51%). These differences remained significant in multivariate analysis (P < 0.001).
C.M. Booth et al. / Clinical Oncology 26 (2014) 506e514
509
Table 2 Variables associated with surgical resection among 5259 cases with bladder cancer treated with curative intent in Ontario 1994e2008 Characteristic
Patient-related Gender Male (n ¼ 3963) Female (n ¼ 1296) Age, years 20e49 (n ¼ 181) 50e59 (n ¼ 560) 60e69 (n ¼ 1185) 70e79 (n ¼ 1996) 80þ (n ¼ 1337) Socioeconomic status, quintile* 1 (n ¼ 1053) 2 (n ¼ 1218) 3 (n ¼ 1122) 4 (n ¼ 974) 5 (n ¼ 884) Charlson comorbidity score 0 (n ¼ 3449) 1e2 (n ¼ 1435) 3þ (n ¼ 375) Disease-related Histology TCC papillary (n ¼ 2448) TCC non-papillary (n ¼ 2462) Adenocarcinoma (n ¼ 189) Squamous carcinoma (n ¼ 160) Interval diagnosis to treatment (unit ¼ 1month) Treatment-related Study period 1994e1998 (n ¼ 1415) 1999e2003 (n ¼ 1640) 2004e2008 (n ¼ 1721) Geographical region of Ontario A (n ¼ 2115) B (n ¼ 672) C (n ¼ 639) D (n ¼ 363) E (n ¼ 92) F (n ¼ 176) G (n ¼ 381) H (n ¼ 817)
Proportion with surgery
Univariate analysis
73% 75%
Reference 1.10 (0.95e1.27)
87% 84% 83% 76% 56%
Reference 0.81 (0.50e1.32) 0.73 (0.46e1.15) 0.49 (0.31e0.76) 0.20 (0.13e0.31)
71% 72% 77% 76% 73%
0.90 (0.74e1.10) 0.95 (0.79e1.16) 1.22 (1.00e1.50) 1.15 (0.93e1.41) Reference
79% 67% 54%
Reference 0.53 (0.47e0.61) 0.31 (0.25e0.39)
73% 74% 78% 83% N/A
Reference 1.05 (0.93e1.20) 1.31 (0.92e1.86) 1.84 (1.21e2.81) 0.998 (0.996e1.000)
66% 75% 78%
Reference 1.59 (1.36e1.85) 1.88 (1.62e2.18)
81% 84% 49% 62% 62% 74% 70% 76%
Reference 1.24 (0.99e1.57) 0.23 (0.19e0.28) 0.39 (0.31e0.50) 0.39 (0.25e0.61) 0.68 (0.48e0.97) 0.56 (0.44e0.71) 0.75 (0.62e0.91)
Odds ratio (95% confidence interval)
Multivariate analysis P value
Odds ratio (95% confidence interval)
0.189
P value
0.282 Reference 1.09 (0.93e1.28)
Clinical Oncology journal homepage: www.clinicaloncologyonline.net
Original Article
Curative Therapy for Bladder Cancer in Routine Clinical Practice: A Population-based Outcomes Study C.M. Booth *y, D.R. Siemens yz, G. Li *, Y. Peng *x, W. Kong *, D.M. Berman x, W.J. Mackillop *y{ * Division
of Cancer Care and Epidemiology, Queen’s University Cancer Research Institute, Queen’s University, Kingston, Canada y Department of Oncology, Queen’s University, Kingston, Canada z Department of Urology, Queen’s University, Kingston, Canada x Department of Pathology and Molecular Medicine, Queen’s University, Kingston, Canada { Department of Public Health Sciences, Queen’s University, Kingston, Canada Received 30 January 2014; received in revised form 8 April 2014; accepted 13 May 2014
Abstract Aims: Definitive therapy of bladder cancer involves cystectomy or radiotherapy; controversy exists regarding optimal management. Here we describe the management and outcomes of patients treated in routine practice. Materials and methods: Treatment records were linked to the Ontario Cancer Registry to identify all cases of bladder cancer in Ontario treated with cystectomy or radiotherapy in 1994e2008. Practice patterns are described in three study periods: 1994e1998, 1999e2003, 2004e2008. Logistic regression, Cox model and propensity score analyses were used to evaluate factors associated with treatment choice and survival. Results: In total, 3879 cases (74%) underwent cystectomy and 1380 (26%) were treated with primary radiotherapy. Cystectomy use increased over time (66, 75, 78%), whereas radiotherapy decreased (34, 25, 22%), P < 0.001. There was substantial regional variation in the proportion of cases undergoing radiotherapy (range 16e51%). Five year cancer-specific survival (CSS) and overall survival were 40 and 36% for surgical cases and 35 and 26% for radiotherapy cases (P < 0.001). In multivariate Cox model and propensity score analyses, there was no significant difference in CSS between surgery and radiotherapy (hazard ratio 0.99, 95% confidence interval 0.91e1.08); radiotherapy was associated with slightly inferior overall survival (hazard ratio 1.08, 95% confidence interval 1.00e1.16). Conclusion: Utilisation of cystectomy for bladder cancer in routine practice has increased over time with no evidence of a significant difference in CSS between radiotherapy and cystectomy. Ó 2014 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved. Key words: Bladder cancer; cystectomy; health services; outcomes; quality of care; radiation
Introduction Initial local therapies for muscle-invasive bladder cancer include cystectomy or radiotherapy. Outcomes achieved with either modality are good in highly selected series from centres of excellence [1,2] and there is no strong evidence showing the superiority of one modality over the other. Accordingly, controversy exists as to what treatment is Author for correspondence: C.M. Booth, Division of Cancer Care and Epidemiology, Queen’s University Cancer Research Institute, 10 Stuart Street, Kingston ON, K7L 3N6, Canada. Tel: þ1-613-533-6895; Fax: þ1-613-5336794. E-mail address: [email protected] (C.M. Booth).
preferred and whether in fact there is an optimal treatment strategy. This controversy is evident in the discordance of international treatment guidelines, with some strongly recommending cystectomy over radiotherapy [3e5] and others concluding that both forms of treatment are acceptable [6,7]. These recommendations are not based on highlevel evidence, as there are no contemporary randomised controlled trials (RCTs) that have systematically compared these two treatment modalities. Population-based outcome studies provide insight into patterns of care and outcomes among patients treated in routine clinical practice [8]. They can minimise sources of selection and referral bias that often limit single institutionbased studies. These studies may also contribute to an
0936-6555/$36.00 Ó 2014 The Royal College of Radiologists. Published by Elsevier Ltd. All rights reserved. http://dx.doi.org/10.1016/j.clon.2014.05.007
C.M. Booth et al. / Clinical Oncology 26 (2014) 506e514
understanding about effectiveness of medical therapies when there are no randomised data to guide patient care. Only a handful of population-based studies have reported practice and outcomes of cystectomy and radiotherapy for bladder cancer and these studies primarily described patients treated in the 1980s and 1990s [9e12]. It remains unclear if clinical features of bladder cancer have changed over time and how surgical and radiotherapy treatments may have evolved. We undertook the current populationbased cohort study to evaluate the use of surgery and radiotherapy for patients with bladder cancer in the contemporary era and provide insights into the outcomes achieved in routine clinical practice.
Materials and Methods Study Design and Population This was a population-based, retrospective cohort study to describe the management and outcome of muscleinvasive bladder cancer in the Canadian province of Ontario. Ontario has a population of about 13.5 million people and a single-payer universal health insurance programme. All incident cases of bladder cancer in Ontario with transitional cell, adenocarcinoma and squamous cell histology who underwent cystectomy or radical radiotherapy in 1994e2008 were included. The study population was classified into three temporal periods: 1994e1998, 1999e2003, 2004e2008. Data Sources The Ontario Cancer Registry (OCR) is a passive, population-based cancer registry that captures diagnostic and demographic information on at least 98% of all incident cases of cancer in the province of Ontario [13]. The OCR does not compile information about the extent of disease or treatment. Accordingly, we obtained surgical pathology reports for patients with cystectomy. Indicators of the socioeconomic status of the community in which patients resided at diagnosis were linked, as described previously [14]. A variety of electronic administrative health databases were linked to the OCR. Records of hospitalisation from the Canadian Institute for Health Information (CIHI) provided information about surgical interventions; these records are known to be consistent and complete [15]. The clinical databases of Ontario’s comprehensive cancer centres provided records of radiotherapy and chemotherapy. These centres are the only providers of radiotherapy in the province and the electronic radiotherapy records are known to be 95% complete and 99% accurate with respect to total dose, number of fractions, date of therapy, body region irradiated and treatment intent [16]. Provincial physician billing records and treatment records from regional cancer centres were used to identify chemotherapy utilisation e these records are created in the act of prescribing treatment. Surgical pathology reports were
507
obtained from the OCR and reviewed by a team of data abstractors. Definitions of Comorbidity, Management and Outcomes Comorbidity was classified using the Charlson Index modified for administrative data based on all non-cancer diagnoses recorded during any hospital admission within 5 years before surgery [17]. Cases treated with radical radiotherapy were identified from the radiotherapy treatment records of the regional cancer centres. Cases treated to the bladder or pelvis with curative intent were included, as were those with missing intent who were treated with 59.5 Gy over 30þ fractions. Twelve per cent (167/1380) of cases received concurrent chemotherapy with radiotherapy. Neoadjuvant and adjuvant chemotherapy were delivered to 17% (230/1380) and 5% (70/ 1380) of radiotherapy cases, respectively. A small number of surgical cases received preoperative (11/3879, 0.2%) and postoperative (37/3879, 1%) radiotherapy. Factors Associated with Treatment Modality Factors associated with treatment choice are shown in Table 2. The results from the multivariate analysis show that patients with advanced age and greater comorbidity were less likely to have cystectomy. Utilisation rates of radiotherapy ranged three-fold across geographical regions (range 16e51%). These differences remained significant in multivariate analysis (P < 0.001).
C.M. Booth et al. / Clinical Oncology 26 (2014) 506e514
509
Table 2 Variables associated with surgical resection among 5259 cases with bladder cancer treated with curative intent in Ontario 1994e2008 Characteristic
Patient-related Gender Male (n ¼ 3963) Female (n ¼ 1296) Age, years 20e49 (n ¼ 181) 50e59 (n ¼ 560) 60e69 (n ¼ 1185) 70e79 (n ¼ 1996) 80þ (n ¼ 1337) Socioeconomic status, quintile* 1 (n ¼ 1053) 2 (n ¼ 1218) 3 (n ¼ 1122) 4 (n ¼ 974) 5 (n ¼ 884) Charlson comorbidity score 0 (n ¼ 3449) 1e2 (n ¼ 1435) 3þ (n ¼ 375) Disease-related Histology TCC papillary (n ¼ 2448) TCC non-papillary (n ¼ 2462) Adenocarcinoma (n ¼ 189) Squamous carcinoma (n ¼ 160) Interval diagnosis to treatment (unit ¼ 1month) Treatment-related Study period 1994e1998 (n ¼ 1415) 1999e2003 (n ¼ 1640) 2004e2008 (n ¼ 1721) Geographical region of Ontario A (n ¼ 2115) B (n ¼ 672) C (n ¼ 639) D (n ¼ 363) E (n ¼ 92) F (n ¼ 176) G (n ¼ 381) H (n ¼ 817)
Proportion with surgery
Univariate analysis
73% 75%
Reference 1.10 (0.95e1.27)
87% 84% 83% 76% 56%
Reference 0.81 (0.50e1.32) 0.73 (0.46e1.15) 0.49 (0.31e0.76) 0.20 (0.13e0.31)
71% 72% 77% 76% 73%
0.90 (0.74e1.10) 0.95 (0.79e1.16) 1.22 (1.00e1.50) 1.15 (0.93e1.41) Reference
79% 67% 54%
Reference 0.53 (0.47e0.61) 0.31 (0.25e0.39)
73% 74% 78% 83% N/A
Reference 1.05 (0.93e1.20) 1.31 (0.92e1.86) 1.84 (1.21e2.81) 0.998 (0.996e1.000)
66% 75% 78%
Reference 1.59 (1.36e1.85) 1.88 (1.62e2.18)
81% 84% 49% 62% 62% 74% 70% 76%
Reference 1.24 (0.99e1.57) 0.23 (0.19e0.28) 0.39 (0.31e0.50) 0.39 (0.25e0.61) 0.68 (0.48e0.97) 0.56 (0.44e0.71) 0.75 (0.62e0.91)
Odds ratio (95% confidence interval)
Multivariate analysis P value
Odds ratio (95% confidence interval)
0.189
P value
0.282 Reference 1.09 (0.93e1.28)
