You are on page 1of 10

European Urology

European Urology 43 (2003) 53±62

Gender-Related Differences in Clinical and Pathological Characteristics and Therapy of Bladder Cancer
 nc, R. Garcõ Âa-Closasd, C. Serrae, D. Puentea, N. Malatsa,*, L. Cecchinib, A. Tardo A. Carratof, M. Salaa, R. Boixedaa, M. Dosemecig, F.X. Reala, M. Kogevinasa For the EPICURO Study Group Investigators1

 Me Á dica, Universitat Pompeu Fabra, Carrer del Dr. Aiguader 80, E-08003, Barcelona, Spain Institut Municipal d'Investigacio Hospital Germans Tries i Pujol, Badalona, Spain c Universidad de Oviedo, Asturias, Spain d Hospital Universitario, Tenerife, Spain e Â, Sabadell, Spain Consorci Hospitalari Parc Taulõ f Hospital General de Elche, Alicante, Spain g National Cancer Institute, Bethesda, USA

Accepted 26 August 2002

Abstract Objective: To con®rm the very high male:female ratios previously observed among Spanish bladder cancer patients and to assess gender differences in tumoral characteristics, diagnostic procedures, and treatment in a large series of consecutive bladder cancer patients. Patients and Methods: All newly diagnosed bladder cancer patients (n ˆ 615) in 17 Spanish hospitals, between 1997±2000, were included. Information was collected both through personal interviews to patients and from medical records using a structured form. Results: Seventy-six percent of tumours were super®cial. The male:female ratio was 6.7 and it was similar for super®cial and in®ltrating tumours. Women were older than men at the diagnosis of bladder cancer (68:2 Æ 9:4 years versus 65:7 Æ 9:7 years, p ˆ 0:01). Ten percent of super®cial tumours in women, versus 3% in men, were classi®ed as ``other histological types'' ( p ˆ 0:008). T1GIII tumours were more frequent among men (17% versus 7%, p ˆ 0:047). On the other hand, women were more likely to present with 0a-stage tumours (48.6% versus 35.5%, p ˆ 0:04), multiple tumours (50% versus 29%, trend test: 0.005), multi-centric tumours (54% versus 38%, p ˆ 0:019), and larger in®ltrating masses (5.2 cm versus 3.8 cm, p ˆ 0:03) than men. Among 0a-stage tumours, only 23% of women compared to 54% of men received transurethral resection (TUR) alone ( p ˆ 0:002). Women were almost ®ve-fold more likely to receive additional therapies to TUR ( p ˆ 0:004) after adjusting for age, geographical area, stage, tumoral size, nuclear grade, and multiplicity. Conclusion: The study con®rms the very high male:female ratio of bladder cancer in Spain. We found substantial differences in the pathological characteristics of tumours from men and women. There was a tendency for women to receive more frequently non-standard, more aggressive, therapy than men. # 2002 Elsevier Science B.V. All rights reserved. Keywords: Bladder cancer; Sex ratio; Women; Treatment; Stage; Pathological characteristics 1. Introduction Bladder cancer is the fourth most common malignancy in men from developed countries. It is less frequent in women, especially in Southern European countries, where it ranks at the 10th position [1]. The excess risk of bladder cancer in men compared to

Abbreviations: TUR, transurethral resection; BCG, Bacille CalmetteGuerin; TNM, tumour, lymph node, and metastases system; OR, odds ratio; 95% CI, 95% con®dence interval * Corresponding author. Tel. ‡34-92-221-1009; Fax: ‡34-93-221-3237. E-mail address: (N. Malats). 1 Members of the Spanish Multicenter Study on Genetic and Clinical Factors Associated with the Prognosis of Bladder Cancer (EPICURO) Study Group are listed in Appendix A.

0302-2838/02/$ ± see front matter # 2002 Elsevier Science B.V. All rights reserved. PII: S 0 3 0 2 - 2 8 3 8 ( 0 2 ) 0 0 4 9 6 - 7

®rst treatment. standard . This report concentrates on the evaluation of gender-differences in bladder cancer risk in a large patient series and a detailed clinical description of gender-related differences in patient management at ®rst diagnosis and clinical±pathological characteristics of patients. we have taken advantage of a large prospective series of consecutive bladder cancer patients to begin to study in greater detail the gender-associated differences previously observed among Spanish bladder cancer patients and to assess whether there are differences between genders regarding tumour characteristics. This work is the basis to study in greater detail the causes of such differences. including bladder. Only 7. and tumoral characteristics were collected from medical records through reviews conducted by physicians using a structured questionnaire.9]. On the basis of the pT-stage. most of them because of a transurethral resection (TUR) procedure in Urology Departments.19]. was computed as the time between the patient's ®rst awareness of symptoms to the TUR. when more than one area was affected. grade. Macroscopic haematuria. All patients were approached by a study investigator during the ®rst hospital admission. 9th revision. Tumour. incontinence.8. Data collection Information on sociodemographics and symptoms at presentation was collected through personal computer-assisted questionnaires conducted by trained interviewers during the ®rst hospital admission. The study is part of a larger project (EPICURO Study) aiming at: (1) analysing bladder cancer risk in relation with genetic susceptibility factors (i. / European Urology 43 (2003) 53±62 women occurs in all areas of the world regardless of age [2]. intravenous pyelogram. stage. tumour location was grouped as one versus !2 areas (multicentricity) and as trigone versus non-trigone areas. vesical irritation. chest X-ray. and metastases (TNM) system was used for pathologic staging [21]. were selected. environmental exposures. number and location of masses. this issue has not yet been solved.14]. Patients with a previous diagnosis of lower urinary tract cancer. and grade [13. diagnostic procedures. Clinical information related to diagnostic procedures. all bladder cancer patients newly diagnosed in 17 hospitals from ®ve regions of Spain (two in Catalunya and one each in Alicante. and past medical history.2% of women refused to participate in the study. For this analysis.and microscopical tumoral features were recorded including. mucosal appearance. diet.10±12]. Tenerife. Subsequently. Detailed macro. drugs. Use of diagnostic procedures such as abdominal ultrasonography. the sex ratio appears to vary widely. normal versus abnormal. The largest tumour was classi®ed as sessile or pediculated. and hormonal factors between sexes [1. and neck. mixed. and high (III) [22]. survival) [18. size. gross tumour appearance. occupation. and non-speci®c tumoral histology. However. cystoscopy. the ®rst category (grade I) was compared to all others regardless of the classi®cation scheme used. urination habits. In summary. and dysuria were recorded as symptoms at presentation. it was appropriately registered. Time lag to diagnosis. code 188 [20]). 2. Grade of nuclear atypia was recorded as low (I). and mucosal appearance was considered as a two-category factor. tumoral stage [7. and radioisotope bone scan were recorded. 2. it has been suggested that tumour characteristics could partly explain gender-related differences in bladder cancer prognosis (i. Number of tumours and size of the largest mass were determined by three methods: abdominal ultrasonography. Treatment management was categorised according to conventional criteria [24. These data were sent to the coordinating centre where they were edited and coded. whose residence was in the catchment's area of each hospital. and the main ®ndings registered. Age was also computed to the TUR date. posterior wall. In this work. patients were subsequently classi®ed as suffering from a super®cial (Ta. Tis) or an in®ltrating tumour (T2±4). ranging from 3:1 in the US to 7:1 in Spain and other Southern European countries [3. Histology was categorised as ``pure'' transitional or ``other''.e. urinary tract infection. gross tumour appearance was classi®ed as papillary.e. between 20 and 80 years of age. To what extent tumour characteristics re¯ect diverse environmental exposure patterns has not been extensively studied [15±17]. and endoscopy.5±7]. the latter including squamous. medium (II). Less work has been performed comparing gender differences related to clinical and pathological characteristics of bladder tumours. trigone.54 D. Tumour location was recorded as lateral right wall. Cases with histologically con®rmed bladder cancer (International Classi®cation of Diseases. dome. tumour growth pattern. and clinical management.1. and (2) identifying molecular prognostic markers for bladder cancer. CT scan. lymph node. polymorphisms of genes coding for metabolic and DNA repair enzymes) of tobacco. or mixed. ureter and urethra were not eligible for inclusion in the study. solid.2. Since TUR also provided diagnostic data. Study population Between May 1997 and December 2000. renal pelvis. Several authors have studied the causes of this difference pointing at factors such as gender differences in smoking habits or occupational exposures as well as anatomic features. The study made no attempt to impose a common diagnostic or therapeutic strategy in the participating hospitals and re¯ects clinical practice at these centers. and Asturias) were included in the study. lateral left wall. and histology of the largest mass. Written informed consent for study participation was requested from all patients.8% of men and 8. However. adenocarcinoma. anterior wall. computed tomography scan of the abdomen and pelvis. Puente et al. stage. However.2. Two cases with HIV infection were also excluded to avoid risk of contamination during biological sample handling. information obtained from cystoscopy and TUR procedures was pooled under ``endoscopy''. expressed in months.25] and strati®ed by stage. Patients and methods 2. A small fraction of tumours was graded in four categories according to Ash's classi®cation [23]. T1.4]. Several studies have reported differences in histological type [8.

9) (1.7 19 213 171 73 25 6 28 173 233 35 25 22 45 (3. Results A total of 615 patients were considered for this analysis.1) (46.5) 0. Vesical irritation was the presenting symptom in 34 and 61% of patients with super®cial and in®ltrating tumours. Chicago. Two superficial cases with stage IV: TaN1 and T1N2 (ring cells). the chi-square test and the chi-square test for trend were applied. Women with super®cial tumours had a higher frequency of 0a-stage tumours than men (p ˆ 0:04).7 (Table 1). grade. IL. Number of masses measured by ultrasonography and CT scan was compared to those obtained by endoscopy and agreement between procedures was assessed by estimating kappa indexes. for patients with T1 or Ta depending on tumour grade. radiotherapy.5) (13) (5..1) (33.4) (7) (5. Treatment strategies different from those speci®ed above were grouped under the ``other'' category. / European Urology 43 (2003) 53±62 55 treatments were considered to be the following: TUR for all super®cial tumours. Results were considered signi®cant at the two-sided p of 0. yielding a sex ratio of 6. respectively (Table 2). either chemotherapy or Bacille Calmette-Guerin (BCG).4) (4.001 (37. 87% were men and 13% were women. number of masses. Puente et al. occurring in 80 and 92% of patients.7) (14. p ˆ 0:05).D.1) (35.2) (5. Four hundred and seventy patients (76%) had a super®cial tumour and 145 (24%) had an in®ltrating tumour. size and multiplicity.7) (7. Data analysis The distribution of patient and tumoral characteristics among men and women was assessed. Outliers were removed from these analyses. 41% of women with in®ltrating tumours.1 Æ 9. The models were adjusted for potential confounding factors such as age. 1997) and statistical analysis was performed using version 9.1) (5. By contrast.1.05 level. To assess the independence of two categorical variables.1) (4. 3.4 12 33 25 3 5 0 2 34 26 4 5 1 10 (15. Level of education Illiterate Primary school incomplete Primary school ®nished High school University Others Missing information Stageb 0a I II III IV Missing information a b a 3.D.2) Women n (%) 80 (13) 68.6) (37. and systemic chemotherapy. Correlation coef®cients were computed to compare ultrasonography and CT scan with endoscopy measurements of tumoral size and number of masses.1) (1) 0. mainly when the tumour was super®cial (40% of women versus 27% of men. Tumour symptoms and diagnostic procedures Macroscopic haematuria was the commonest ®rst symptom in patients with super®cial or in®ltrating tumours.8) (6. additional intravesical therapy. stage.186 One man with missing data. Analyses were conducted separately for super®cial and in®ltrating tumours.4) (4.4) p 615 66. Both groups of patients were highly similar according to sex and age. Fisher's exact test was used. Time from occurrence of ®rst symptom to diagnosis was similar in men and women with super®cial tumours. Data were managed using ACCESS databases (Microsoft Corporation. To determine the relationship between a categorical variable with two levels and normally or non-normally distributed quantitative variables.3) (42.01 <0.8) (42.4) (42. Table 1 Description of the study population by gender Total n (%) Total Age Mean Æ S.7 Æ 9.3.7 31 246 196 76 30 6 30 207 259 39 30 23 55 (5. for patients with in®ltrating tumours.5) (47. 1999).7) (7. When 20% of cells had expected counts of less than ®ve. were diagnosed in Men n (%) 535 (87) 65. Similarly.3) (32. Women were signi®cantly older and more illiterate than men. compared to 13% of men. respectively. dysuria was more commonly referred by patients with in®ltrating tumours.1) (1. 2. and size of the largest mass measured by endoscopy. Incontinence was the only ®rst symptom that women described more frequently.1) (3.2 Æ 9.5) (48. Student's t-test and Mann± Whitney U tests were applied. radical cystectomy. respectively.0) (33. Logistic regression models were ®t to achieve adjusted odds ratios (OR) of ``having received only TUR'' versus ``having received TUR plus other treatments''. These symptoms did not show any association with gender.0 SPSS statistical package (SPSS Inc. . geographical area.

1) 8 10 (20.4) 12 (70. T4 (TNM).7) 32 19 (17. . CT scan in 8 and 86%.5) 104 (28. Diagnostic tests applied according to gender and invasiveness.8) 96 229 (73.2) 44 (45.6) 8 45 (84.56 D.292 0.3) 13 p Total n (%) 9 (7.4) 292 (79. infiltrating tumours.2) 59 (51.8) 31 58 (51.8) 30 12 (13.1) 32 55 (48.1) 31 44 (38.2) 89 (91.4) 21 (39. radioisotope bone scan in 4 and 57%.1) 2 6 (35.3) 11 (64.4) 8 32 (60.5) 104 75 (23.6) 97 246 (78.8) 42 (79.9) 105 (92.5) 75 (20.6) 103 243 (66. and chest X-ray in 97 and 99%.1) 21 (22.1) 34 Women n (%) 1 (5.6) 58 (60.151 0. endoscopy in 99 and 98%. 1.4) 96 65 (24) 60 (22.5) 29 52 (54.7) 2 5 (29.6) 95 204 (65. T1 (TNM).5) 47 (48. Tis.6) 14 (26.2) 2 (11.6) 104 261 (71.5) 71 (22. Puente et al. T2.950 1. Ta. Diagnostic procedures were similarly used in women and men (Fig.6) 2 6 (35. respectively.8) 11 (22.7) 29 38 (39.230 0. Fig.092 0.1) 2 p Men n (%) 64 (20.4) 83 (26. / European Urology 43 (2003) 53±62 Table 2 Symptoms at presentation and time-interval to diagnosis according to sex and tumour invasiveness Super®cial tumoursa Total n (%) Macroscopic haematuria No Yes Missing information Vesical irritation No Yes Missing information Incontinence No Yes Missing information Dysuria No Yes Missing information Symptom-TUR interval 1 month 1À3 months >3 months Missing information a In®ltrating tumoursa Women n (%) 11 (20.5) 36 Men n (%) 8 (8. the course of the ®rst month since the ®rst symptom occurred (p ˆ 0:02).6) 67 (21.5) 105 291 (79.4) 250 (79.9) 8 (15.2) 109 (34.000 0.2) 8 39 (73. although the number of cases was small.4) 23 (21.8) 59 (64.052 0. T3.3) 173 (54.3) 55 (48. 1).893 0. intravenous pyelogram in 44 and 41%.738 0.1) 146 (53.018 Superficial tumours.9) 27 (56.2) 151 0.1) 67 (61.9) 69 (61.9) 16 (94.4) 123 (33.7) 2 7 (41. Ultrasonography was conducted in 81% of super®cial and 83% of in®ltrating tumours.9) 138 75 (20.4) 30 50 (51.3) 11 (64.7) 8 (47.

8) 1 11 (91.5) 5 84 (20. women with super®cial tumours were more commonly affected of multicentricity (54% versus 38% of men.1) 3 49 (92.2) 12 (10) 43 (35.6) 14 (10. CT scan (5. / European Urology 43 (2003) 53±62 Table 3 Pathological characteristics of tumours according to sex and tumour invasiveness Super®cial tumour Total n (%) Histology Transitional Others Missing information Grade GI GII ‡ GIII Missing information Tumour location >1 localization Trigone Others Missing information Tumour growth Papillary Solid Mixed Missing information Tumour gross appearance Sessile Pediculated Missing information Mucosal appearance Normal Others Missing information a b 57 In®ltrating tumour Women n (%) 52 (89. Table 3). Women were also more likely to present multiple super®cial tumours (50% versus 29% of men. 10% of super®cial tumours in women.019a 0.2.9) 32 (38.005) and larger in®ltrating masses than men regardless of the method used to estimate tumoral size: ultrasonography (5. A comparison of CT scan and endoscopy yielded correlation coef®cients lower than 0.847 0.5) 11 (64.8) 140 (46. Nevertheless. Treatment Treatment administered to patients with super®cial tumours is displayed as shown in Table 4.2) 25 (56.5) 24 (42.6) 78 (27. Two subjects were diagnosed of Tis.3) 3 151 (38. compared to 23% of .542 for super®cial and 0.1 in both super®cial and in®ltrating tumours.4) 204 (51. p ˆ 0:052).1) 2 4 (23.097.5) 5 (45.5) 165 (47. 95% CI 1. 2). compared to 3% in men. trigone. Tumoral characteristics Most tumours were ``pure'' transitional carcinomas. trend test: 0.2) 43 (9.5) 30 (8.3) 2 10 (16.188 0. p ˆ 0:047).2) 101 207 (72.9) 55 163 (53.7) 6 (10.6) 7 19 (43.3) 6 6 (54.2) 6 (6. In addition.5) 2 (10.8) 16 33 (71. and endoscopy (5. p ˆ 0:019.9) 14 (13) 16 82 (93.2) 371 (79. This ®nding was mainly observed among 0a-stage tumours: 54% of men.5) 117 240 (72.6) 2 0 133 (100) 12 72 (52. However.991 0. p ˆ 0:003). p ˆ 0:012). Table 3). T1GIII tumours were more frequent among men (17% versus 7% in women. and 0.1) 41 (10.7) 20 (5) 62 182 (52.391 and 0.7) 322 (79.2) 80 (64) 16 (12. Although rare. the latter ®nding lost signi®cance when the number of tumours was considered in the model. The agreement on the number of masses estimated by ultrasonography and CT scan.5) 8 302 (86. There were 32 additional super®cial carcinomas in which in situ tumour was present together with Ta or T1 masses.367 0.899 0.9) 49 (83.4) 28 (19. compared with those obtained by endoscopy. 3.9) 3 (5.1) 1 31 (54. others.7) 8 (47. p ˆ 0:03.3) 31 (7.6) 17 (4.7) 13 (28.8) 5 25 (23.5 cm versus 3.5) 139 0. Only 21% of super®cial tumours in men and 17% in women were grade I. 3. showed Kappa indexes of 0. showing coef®cients of 0. TaGIII tumours were found to be more frequent among women (7% versus 2% in men.533 for in®ltrating tumours.5) 91 (27.745 >1 Localization vs.4) 124 446 (96. Tumoral size measurements by ultrasonography and endoscopy were highly correlated. were classi®ed as ``other'' histologies (p ˆ 0:008.5) 228 (50.2) 2 (11. Puente et al.3%.5) 16 (3. p ˆ 0:025).D.22±4.8) 17 93 (93) 7 (7) 43 58 (61) 37 (39) 48 Men n (%) 98 (79) 26 (21) 2 0 115 (100) 11 65 (54. In contrast.7) 1 (8.008 0.3.4) 2 (3.5) 7 p Men n (%) 394 (97.5) 7 94 (20.185 for super®cial and in®ltrating tumours.231 0.315a 1.328 and 0.5) 10 (2.394b 0.8) 37 52 (61.7% versus 19.5 cm in men.1) 69 (63. Men were more likely to undergo a TUR alone than women (39.5) 1 (1. respectively.000b 0. Fig.3) 15 p Total n (%) 115 (80.503 0.2 cm versus 3.5) 0 0 18 (100) 1 7 (41.2) 7 29 (23.6 cm in men.2) 51 (37. Trigone vs.8 cm in men. crude OR ˆ 2:45.3) 11 351 (87.1) 41 Women n (%) 17 (89.93.7) 2 (11.3 cm versus 3.8) 4 182 (40.

endoscopy-measured tumoral size. geographical area.3) (23. While we did not ®nd major gender-associated . gender-associated differences in treatment were also observed though they did not reach statistical signi®cance. Among patients with stage-I tumours. Discussion The main goal of this study was to assess differences in clinical±pathological characteristics and management among men and women recruited at a large prospective series of bladder cancer patients from Spain.6) (23.3) (26.002 T1 TUR `alone' TUR ‡ BCG TUR ‡ CT endovesical TUR ‡ BCG ‡ CT endovesical Others Missing a b (26.6) (23.2) (40.8) (9.9) (39. women.6) (1. 95% CI 1.4) (7. The p-value without stratifying by stage was 0.4) (27. stage. Women received BCG as an additional treatment more often than men: 47% versus 22% (p ˆ 0:004).7) (24.6) Women n (%) 7 7 8 7 1 4 4 9 7 2 0 4 (23. Differences between genders on the number of tumours and size of the largest mass according to diagnostic test used.8) (1.067. Puente et al.6) (0.5) (9.7) (23. men being ®ve times more likely than women to undergo only TUR (OR ˆ 5:19. nuclear grade.1) 0.770 Two cases with superficial tumours with stage IV: TaN1 and T1N2 (ring cells) have been excluded from the analysis. 4.0) (23.3) (3. underwent TUR alone (p ˆ 0:002).2) (7.9) (39.58 D.5) (18.9) (16) (1) Men n (%) 86 11 37 23 1 15 55 78 46 15 3 36 (54. and multiplicity. The OR became even more signi®cant after adjustment. / European Urology 43 (2003) 53±62 Fig.69±15. Table 4 Treatment of patients with super®cial tumours according to gender Total n (%)a Ta TUR `alone' TUR ‡ BCG TUR ‡ CT endovesical TUR ‡ BCG ‡ CT endovesical Others Missing 93 18 45 30 2 19 59 87 53 17 3 40 (49.4) (14.4) pb 0.4) (7. The risk of being treated with ``TUR alone'' was adjusted for co-factors that could explain a different treatment strategy among women: age.9) (31.91). 2.

Many studies have demonstrated the bene®t of adjuvant treatment. Nevertheless.12. While this ®nding may be due to differences in exposure patterns or to the presence of genetic and endogenous protective factors in women. medical doctors may be less prone to consider this diagnosis at the woman's ®rst visit [35]. Puente et al. Such similarity between the two groups of tumours has also been reported by other Spanish authors [28]. this difference disappeared when multiplicity was taken into account. were diagnosed in men. Thus. We have searched for sex-related differences in super®cial and in®ltrating tumours.26] who have reported that super®cial tumours represent 60±80% of all bladder cancers [27].8. This represents a lower percentage than that described in the literature whose etiological signi®cance is presently unknown [34]. In the near future. in decreasing the recurrence rate or in prolonging the disease-free interval of patients with super®cial tumours [34]. When the analysis was strati®ed by sex. thus in¯uence prognosis [37].31]. geographical area. grade. the study should be able to provide some clues to better understand the high men:women ratio observed in Spain. or when several tumours are located in different areas. The association persisted even after adjusting for age. Whether urologists are more aggressive when treating women with bladder cancer because the disease is less common and has worse prognosis in women deserves further investigation. there were differences in the pathological characteristics of tumours and in the treatment given to patients. Our results are in agreement with the above statement. The men:women ratio observed in our study was almost 7 and it was similar for both super®cial and in®ltrating cancers. These data are in agreement with other authors [11. Mulders et al.D. Several authors have described that women are diagnosed at advanced stages more frequently than men [7. In the present study. This is probably due to the fact that urologists report the presence of tumour in more than one location either in the case of large masses that occupy more than one bladder area. among others. the higher frequency of urinary tract infections among women may lead to less extensive investigation of haematuria [8]. where it ranges from 3 to 4 [10. The proportion of ``pure'' transitional cell carcinomas was signi®cantly higher among super®cial tumours. Further analyses of the reasons and consequences of differential therapeutic strategies . one super®cial and two in®ltrating tumours.10±12] and have suggested that.28±30]. Nevertheless. In addition. several observations from our study do not provide support to this hypothesis and suggest that the differences in stage at presentation are not due to diagnostic delay: (1) the proportion of Ta tumours was signi®cantly higher in women than in men. We are only aware of one study evaluating differences in treatment according to gender and bladder cancer prognosis: women with in®ltrating bladder tumours were treated more aggressively but there was no effect on survival [11]. we observed that the ``mixed histology'' diagnosis was more common in women than in men and that only three cases with squamous cell carcinoma. It has been described that treatment of super®cial bladder cancer is similar for men and women [7]. other authors described higher men:women ratios among super®cial tumours [8. treating women with Ta tumours more aggressively might result in an increased recurrence-free interval. we found that women with Ta tumours were more likely to receive TUR plus adjuvant immunotherapy than men with the same T-stage tumours. because bladder cancer is less common in women. size.12. more work is necessary to rule out the effect of a higher life expectancy among women. especially for in®ltrating tumours. Depth of in®ltration and grade of differentiation are important prognostic parameters in bladder cancer. since it has been proposed these two neoplasms may be different molecular entities [32]. Our study also con®rms that women were consistently older than men at the time of diagnosis [11. and tumour size. independently. (2) the proportion of Ta tumours in men is in fact lower than that reported in other series (42% versus approximately 70%) [28. Unlike other studies. Overall. such as BCG. tumour location. This case series comprised 76% of super®cial and 24% of in®ltrating tumours. Our ®ndings con®rm in a very large series that the highest men:women bladder cancer ratio are found in Spain. and multiplicity in a multivariate model.33]. the ratio being almost twice of that reported in Northern European countries and North America.18]. we have observed differences regarding these variables related to gender. (3) the ®rst symptom to diagnosis interval was shorter for women than for men. have also been proposed as independent prognostic factors of recurrence and progression of bladder tumours [37. Multiplicity.38]. observed that treatment may alter the biological behaviour of the tumour and. / European Urology 43 (2003) 53±62 59 differences concerning diagnostic procedures. Although.36]. a ®nding that is consistent with other reports [8] and could be explained by the fact that invasive tumours tend to be composed of mixed histologies. There are no known reasons to account for this ®nding. transitional cell carcinoma is the predominant bladder cancer histology in industrialised countries [9. women had tumours in more than one location more commonly than men. stage.

Reala. Hernando. is one of the largest series in which gender-associated differences in bladder cancer characteristics have been analysed. M. Lloretae. Martõ  ceres. Peiro S. J. N.  nb. del Mayod. Y. C. J. Nogue Á nech. A. C. Tenerife) Hospital General de Elche (Elche. Perezc. Kogevinasa. M. E. Alfaroc. Sa  d. Puente et al. Herrero. C. Institut Municipal d'Investigacio Universitat Pompeu Fabra (coordinating center) Member M. Ä ob. F. Ull. Miralles. P Barbas. Tardo N. A. L. and G Perello  n Sanitaria (Grants funded by the Fondo de Investigacio 96/1998-01 and 00/0745). M. R. Alicante) Universidad de Oviedo (Oviedo.M. Guate. women being treated more aggressively.8% of men and 8. J. O. DeLamod. J. selection bias is very unlikely and the study population should be considered as fully representative of bladder cancer patients in the ®ve areas examined. A. clinical data were collected retrospectively. A. Additional funding was provided by BIOMED (Grant PL 963243ÐConcerted Action) and National Cancer Institute. Deu. J. D. T. C. Monte  sc. Bautistad. Castillejoc. Tenerife) Hospital La Candelaria (Santa Cruz. Serrab. This work was partially L Ruiz. Rodrõ Âguez. C. the study shows that bladder tumours in women presents different pathological characteristics than those in men. Mene  ndez.M. Teruel. Ce Âa. Hospital del Mar (Barcelona) Hospital Germans Tries i Pujol (Badalona. L. C. Barcelona) Hospital de Sant Boi (Sant Boi. In this study.  . Blancoc. A. R. M. T. A. A. Tan Âa-Lo  pez. R. we observed that standard treatment was less commonly applied to women than to men. Acknowledgements Technical assistance was provided by DJ McFarlane. E. Pujol. Carretero.  . F. Ariasc. M. Juand.60 D. In addition to con®rming the very high men:women ratio among patients with bladder cancer in Spain compared to Northern Europe and the US. since all newly diagnosed bladder cancer patients in the participating hospitals were prospectively identi®ed and consecutively included in the study. M. A. Gelabert. Nadal. Galbis. P. Á . C. Bustos. / European Urology 43 (2003) 53±62 between genders as related to bladder cancer prognosis are planned. Joverc.J. including 615 patients. Benito .  ndez. J. Tora G. G. J. F. J. while lack of detailed information is a major problem when using medical records as a source. This study. L. D. Boixeda. Centers and members of the EPICURO Study Group Center  Me Á dica. Ferrere. C. A. Serranoe. Lao. Barcelona)  (Sabadell. Ä o. Garcõ Ânez. Puente. Mene  ndezc. M. Villanuevac. A. Ca F. Asturias) Hospital Central Covadonga (Oviedo. Ibarz.X. Martõ Â. F. Malatsa. Manzano. J. Abascal. Barcelona) Centre Hospitalari Parc Taulõ Á gic (Manresa. Ferna R. M. Andrada. Pujadas. Asturias) Hospital Central General (Oviedo. J. Velasco. Villadiego. A. F. J. Herna  ndez. Lo  pez. Pipo d  nchez-Aguilera . Rodriguez de Vera. the fact that all physicians used the same structured questionnaire minimises the possibility of misclassi®cation and the large size of the study allows for strati®cation of results according to tumour subgroups. Asturias) Hospital San Agustin (Aviles. Furthermore. Cecchini. Garcõ E. Salab. K. Barcelona) Centre Hospitalari i Cardiolo Hospital Universitario (La Laguna. J. Velasco.L. Bielsa. Dome Âa-Closasb. Although we cannot yet assess the association of such differences with tumour recurrence. Saladie  spedes. Ve  lezc. A. H. Cabezuelo. Castan  ndez. A. M.I. Ferna  c.M. Garcõ c Ân. Carles. Badal. A. Carratob. M. J. This ®nding should be taken into account when analysing gender-related differences in bladder cancer prognosis. D. N.2% of women refused to participate excludes the possibility that factors such as motivation and attitudes towards health could impair the study's external validity. Lanzas. Asturias) . The fact that only 7. Amoros. Appendix A.

Reuter VR. Eur J Cancer 1998. Epidemiology of bladder cancer. Maintenance Bacillus Calmette-Guerin immunotherapy for recurrent Ta. Bladder cancer incidence and survival in the south-eastern part of The Netherlands. Garcia Escudero C. Acta Urol Esp 1995. O. Gatta G. Mosto® FK. Giorgi Rossi A. Estimates of the worldwide incidence of 25 major cancers in 1990.13:291±8.78:1505±13.41:30±6. Gender differences in stage-adjusted bladder cancer survival.. Eur J Cancer 1994. Knudsen JB. Ferlay J. et al. [21] AJCC. Associations between bladder cancer risk factors and tumor stage and grade at diagnosis. 1156±79. [2] Hartge P. Diaz. Mettlin C. Pascual Bueno Ân J. Mosquera. Mariotto A. [10] Kiemeney LA.44:135±45. V. Epithelial tumors of the bladder. Schiffman Z. Aben KK. [12] Mungan N. [9] Johansson SL. Fischer N.32:183±5. Hoover RN. Asturias) de Cruz Roja (Gijo Alvarez-Buylla (Mieres. [27] Rubben H. Muti P.34:2271±8. J. Soloway MS. 9th revision. et al. [19] Mungan NA. Asturias) Ä es (Gijo de Cabuen  n. Mateos. Bladder cancer. Intravesical instillation of epirubicin. T1 and carcinoma in situ transitional cell carcinoma of the bladder: a randomized Southwest Oncology Group Study. Stewart AK. Coebergh JW. Cohen SM. Kiemeney L. Bladder cancer in women. Arch Esp Urol 1992. Van der Poel H. Ä a del ca Espan [5] Miller AB. Natural history and treatment of low and high risk super®cial bladder tumors. [26] Baniel J.36:660±72. Principal investigator. Lynch C.82:1636±40. Sex-dependent variations in natural history of transitional cell carcinoma of bladder. Gender differences in stage distribution of bladder cancer. Pauwels RP. 1993.166: 476±81. Van der Heijden LH. 2nd ed. Lagrange W.55:368±71. Crissman JD. [7] Van der Poel HG. / European Urology 43 (2003) 53±62 61 Appendix A. Alfonso Gil R. Does invasive bladder cancer differ between women and men? Urology 1988. Van Dick J. Vendrell R. In¯uencia de la  gica del tumor urotelial diferencia del sexo en la agresividad biolo  stico. Carretero P. Lutzeyer W. Fraumeni J. Linehan WM.22:34±6. Devesa S. Fernandez Ruiz M. Witjes JA. Lowe BA. [15] Jensen OM. F. 1997. Visser O. The Etiology of Bladder Cancer from the Epidemiological Viewpoint. The National Cancer Data Base report on bladder carcinoma: the American College of Surgeons Commission on Cancer and the American Cancer Society. M. e Study reference pathologist. [14] Laor E.23:237±9. [3] Parkin DM. Asturias) de Jove (Gijo  n. Gottesman JE. bacillus Calmette-Guerin and bacillus Calmette-Guerin plus isoniazid for intermediate and high risk Ta. editors. Epidemiology 1994. [28] Fernandez Fernandez A. P. [11] Fleshner NE. Unexplained Excess Risk of Bladder Cancer in Men. Tolia BM. Silverman DT. Invasive bladder cancer in women. [17] Aben KK. Kiemeney LA. Asturias) M. Sa  nchez. Giani G. Deutz F. In: Schottenfeld D. Cancer 1996. [24] Van der Meijden AP. C. Arguelles. Area coordinator. Amin MB. The Copenhagen case-control study of bladder cancer: role of smoking in invasive and non-invasive bladder tumours. Hartge P. Urology 1984. [4] Ruiz Cerda JL. Fraumeni Jr JF. Asturias) Ä a. Int J Cancer 1999. code 188. (Continued ) Center Hospital Hospital Hospital Hospital Hospital Hospital a b Member  n. T1 papillary carcinoma of the bladder: a European Organization for Research and Treatment of Cancer genito-urinary group randomized phase III trial. [18] Micheli A. J Urol 1988. Reid RE.163:1124±9. Wahrendorf J.19:196±202. Pravia. Muntan A. Witjes JA. J Epidemiol Commun Health 1987. Bladder Consensus Conference Committee.10:207±12.30A:1134±7. Am J Surg Pathol 1998. Schoenberg MP. Semin Surg Oncol 1997. Cancer Epidemiology and Prevention. 1975±1989.10:399±404. et al. Cancer Res 1977. Epidemiol Rev 1981. Huescar. Int Urogynecol J Pelvic Floor Dysfunct 1999.45:837±9. Blettner M. J Urol 2000. Acta Urol Esp 1998. Sylvester R. Montie JE. Witjes J. Menck HR. Pisani P. Silverman DT. [13] Romero JA. Harvey EB.37:2939±42. J.55:876±80. Martinez Jabaloyas J. Sorensen BL. [25] Lamm DL. [23] Ash J. Urinary bladder: AJCC Cancer Staging Manual. [22] Epstein JI. The Prognostic Role of Gender in Survival of Adult Cancer Patients. Kirkels W. [8] Hickey D. p. Urology 2000. [20] AJCC. Puente et al. J Urol 1940. Linehan WM. Gil-Vernet Jr JM. [16] Sturgeon SR. Huergo. Ä ola. New York: Oxford University Press Inc. Brausi M. 1996. Kantor AF. Schapers RF.M. Herr HW. Epidemiology and etiology of bladder cancer. Gutierrez R. Dominguez Benito P. [6] Matanoski GM. J Urol 2001. Koper NP. 3:203±29.22:1435±48. d Laboratory technician. Otero Mauricio G. vesical en su primer diagno . M. Morrison A. Mungan NA. Freed SZ. Eur Urol 1999. Coebergh JW. Urology 2000. J Natl Cancer Inst 1990. Dõ A. Frade. Domenech Ferrando E. c Interviewer.D. Blumenstein BA. References [1] Silverman D. de Balincourt C. 5th ed. Zambon V.139:283±5. Rivas. et al. Alvarez-Vijande R. Sullivan JW. Int Urogynecol J Pelvic Floor Dysfunct 1999. Âaz. The World Health Organization/International Society of Urological Pathology consensus classi®cation of urothelial (transitional cell) neoplasms of the urinary bladder. Bladder cancer in women. Gil Fabra J. Evolucio  n de masculinidad en temporal (1960±1990) de la mortalidad y la razo  ncer vesical. Vera Donoso CD. Bladder Cancer Epidemiology. Jimenez Cruz JF. Philadelphia: Lippincott-Raven.5: 218±25.80:827±41. International Classi®cation of Diseases. Asturias) Jarrio (Coan Carmen y Severo Ochoa (Cangas. Murphy GP.

Puente et al. Cancer incidence in Five Continents. [35] Anton-Culver H. Prognostic factors in pTa±pT1 super®cial bladder tumours treated with intravesical instillations. Oosterhof GO. Debruyne FM. and smoking.91:144±8. 1998. Whelan S. Vicente-Rodriguez J. J Natl Med Assoc 1999. J Natl Med Assoc 2000.3:429±33. Meyden AP. Epidemiological characteristics of squamous cell carcinoma and adenocarcinoma of the bladder. New York: Cold Spring Harbor Laboratory Press.62: 235±9. Verbeek AL. [33] Kantor AF. / European Urology 43 (2003) 53±62 [34] Metts MC. Palou J. Coptcoat MJ. 92:285±94. Taylor TH. Chechile-Toniolo G. Br J Cancer 1993. editors. [38] Millan-Rodriguez F. Klein L. Ann Epidemiol 1993. Milito SJ. [31] Hoke GP. Cancer Res 1988. Molecular genetics of bladder cancer: pathways of development and progression. Hartge P. [37] Mulders PF. .62 D. Young J. Williams KN. Metts JC. Stone BA. vol 31.48:3853±5. Thomas Jr CR. 49±76. Heijbroek RP. Br J Urol 1994. Salvador-Bayarri J. The in¯uence of gender on incidence and outcome of patients with bladder cancer in Harlem. Hall RR. p. Should pT1 transitional cell cancers of the bladder still be classi®ed as super®cial? Br J Urol 1988. no. editors. [36] Abel PD.163:73±8. Williams G. Hoover RN. The association of bladder cancer risk with ethnicity. Multivariate analysis of the prognostic factors of primary super®cial bladder cancer.67:806±12. Fraumeni Jr JF. Lee-Feldstein A. The clinical epidemiology of super®cial bladder cancer: Dutch SouthEast Cooperative Urological Group.73:403±8. Bladder cancer: a review of diagnosis and management. Witjes JA. [30] Parkin D. Lyon: IARC Scienti®c Publications. 1997. Ferlay J. [29] Kiemeney LA. In: Oliver RTD. gender. Debruyne FM. The Dutch South-Eastern Urological Collaborative Group. [32] Knowles M. Raymond L. J Urol 2000. 143. Bladder cancer. Doesburg WH.