Impact of Smoking on Outcomes of Patients with a History of Recurrent Nonmuscle Invasive Bladder Cancer

Impact of Smoking on Outcomes of Patients with a History of Recurrent Nonmuscle Invasive Bladder Cancer Michael Rink,* Evanguelos Xylinas, Marko Babjuk, Jens Hansen, Armin Pycha, Evi Comploj, Yair Lotan, Maxine Sun, Pierre I. Karakiewicz, Joual Abdennabi, Harun Fajkovic, Wolfgang Loidl, Felix K. Chun, Margit Fisch, Douglas S. Scherr and Shahrokh F. Shariat†,‡ From the Department of Urology (MR, EX, DS, SFS) and Division of Medical Oncology, Weill Cornell Medical College-New York Presbyterian Hospital (SFS), New York, New York, and Departments of Urology, University Medical Center Hamburg-Eppendorf (MR, JH, FKC, MF), Hamburg, Germany, Cochin Hospital, Assistance Publique-Hôpitaux de Paris, Paris Descartes University (EX), Paris, France, Hospital Motol, Second Faculty of Medicine, Charles University (MB), Prague, Czech Republic, General Hospital of Bolzano (AP, EC), Bolzano, Italy, University of Texas Southwestern Medical Center (YL), Dallas, Texas, University of Montreal (MS, PIK), Montreal, Quebec, Canada, University Hospital Ibn Rochd (JA), Casablanca, Morocco, and General Hospital St. Poelten (HF), St. Poelten and St. Vincent’s Hospital (WL), Linz, Austria

Abbreviations and Acronyms BCG ⫽ bacillus Calmette-Guérin CIS ⫽ carcinoma in situ CPD ⫽ cigarettes per day NMIBC ⫽ nonmuscle invasive bladder cancer TURB ⫽ transurethral resection of bladder UCB ⫽ urothelial carcinoma of bladder Submitted for publication April 17, 2012. Study received institutional review board approval at each participating institution. Supported by The Frederick J. and Theresa Dow Wallace Fund of the New York Community Trust (MR). * Financial interest and/or other relationship with Pfizer Pharma. † Correspondence: Division of Medical Oncology, Department of Urology, Weill Medical College of Cornell University, New York Presbyterian Hospital, 525 East 68th St., Box 94, Starr 900, New York, New York 10065 (telephone: 212-746-5562; e-mail: [email protected]). ‡ Financial interest and/or other relationship with Ferring Pharma.

For another article on a related topic see page 2398.

Purpose: We investigated the effects of cigarette smoking status, cumulative exposure and time from cessation on disease recurrence and progression in patients with a history of recurrent nonmuscle invasive bladder cancer. Materials and Methods: A total of 390 patients with recurrent nonmuscle invasive bladder cancer were treated with transurethral resection of the bladder, of whom 159 (41%) received instillation therapy immediately postoperatively and 73 (19%) received adjuvant intravesical immunotherapy or chemotherapy. Smoking history included smoking status, number of cigarettes per day, smoking duration in years and years since smoking cessation. Cumulative smoking exposure was categorized as light short-term—19 or fewer cigarettes per day and 19.9 years or less, moderate—all combinations except light short-term and heavy long-term, and heavy long-term—20 or greater cigarettes per day and 20 years or greater. Results: A total of 91 (23%), 192 (49%) and 107 patients (28%) were never, former and current smokers, respectively. Of ever smokers 56 (19%), 156 (52%) and 87 (29%) were light short-term, moderate and heavy long-term smokers, respectively. There was no difference in the risk of disease recurrence and progression among current, former and never smokers. On univariable analyses in ever smokers the risk of disease recurrence and progression increased with augmented smoking intensity (p ⱕ0.015), duration (p ⬍0.001) and cumulative exposure (p ⬍0.001). On multivariable analyses cumulative smoking exposure was an independent risk factor for disease recurrence and progression (p ⱕ0.003). Smoking cessation greater than 10 years before treatment was independently associated with decreased disease recurrence compared to current smoking (HR 0.4, p ⬍0.001). In addition, current smokers had worse survival than former smokers, who in turn had worse survival than never smokers (p ⬎0.05). Conclusions: There is a dose-response relationship of smoking exposure and smoking cessation with disease recurrence and progression in ever smokers with a history of recurrent nonmuscle invasive bladder cancer. These findings support counseling on smoking cessation benefits. Key Words: urinary bladder, urothelium, carcinoma, smoking, survival

2120

www.jurology.com

0022-5347/12/1886-2120/0 THE JOURNAL OF UROLOGY® © 2012 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION

AND

http://dx.doi.org/10.1016/j.juro.2012.08.029 Vol. 188, 2120-2128, December 2012 RESEARCH, INC. Printed in U.S.A.

IMPACT OF SMOKING AND HISTORY OF RECURRENT NONMUSCLE INVASIVE BLADDER CANCER

UROTHELIAL carcinoma of the bladder is the sixth most common malignancy in each gender. In 2012 in the United States alone an estimated 73,510 new cases and 14,880 deaths have occurred.1 Cigarette smoking is the most established risk factor for UCB, increasing the risk by twofold to fourfold.2 Although smoking has decreased slowly but steadily in the last 40 years, 40% of American adults are still current or former smokers.3,4 In patients with primary NMIBC several studies have shown that smokers have worse outcomes than nonsmokers.5– 8 The risk of disease recurrence among ever smokers is increased up to twofold and it seems to be higher in current than in former smokers. However, to our knowledge no group has analyzed the effect of smoking and smoking cessation in patients with a history of recurrent NMIBC. We hypothesized that in patients with a history of recurrent NMIBC smoking would be associated with disease recurrence and progression to muscle invasive UCB. Moreover, we hypothesized that there would be a dose-response relationship between smoking intensity and adverse outcomes. Therefore, to determine the relationship of smoking status, intensity and time from smoking cessation to cancer outcomes, we investigated smoking habits and history in a multicenter cohort of patients treated with TURB for recurrent NMIBC.

METHODS Patient Population All studies were performed with the approval and oversight of the institutional review board at each institution. All participating sites provided the necessary data sharing agreements before initiation. A total of 446 patients with a history of recurrent NMIBC underwent TURB at 4 international centers between 1987 and 2007. In all patients previous TURB was performed elsewhere before the index TURB, ie the TURB used for this study. The 56 patients with missing variables or followup were excluded from analysis, leaving 390 available for study.

TURB and Instillation Therapy All patients had histopathologically proven, recurrent NMIBC and underwent complete TURB according to criteria consistent with guideline recommendations.9 –11 Resection was repeated according to guideline recommendations and at surgeon discretion within 2 to 6 weeks after initial treatment based on pathology and intraoperative findings.12 Instillation of chemotherapy (40 mg mitomycin, 80 mg epirubicin or 50 mg doxorubicin) was administered immediately postoperatively at clinician discretion based on guideline criteria. Similarly, based on guideline recommendations, and physician and/or patient preference adjuvant intravesical immunotherapy or chemotherapy was administered. Briefly, the first instillation was

2121

given 7 to 21 days after final TURB and repeated once weekly for 6 consecutive weeks. Maintenance therapy was not given.13

Pathological Evaluation All surgical specimens were processed according to standard pathological procedures. Genitourinary pathologists assigned tumor grade according to the 1973 WHO grading system. Pathological stage was reassigned according to the 2002 American Joint Committee on Cancer TNM staging system. Concomitant CIS was defined as CIS in conjunction with a tumor other than CIS alone.

Smoking Assessment Smoking history was routinely assessed at presentation to the study center before the index TURB. Patients were only considered ever smokers if they had smoked 100 cigarettes during their lifetime. Data on self-reported cigarette smoking included smoking status (current, former or never smoker), average CPD, ie quantity (1 to 9 cigarettes, 10 to 19, 20 to 29, or 30 or greater), smoking duration in years (9.9 or less, 10 to 19.9, 20 to 29.9, 30 to 39.9, or 40 or greater) and years from smoking cessation to TURB in former smokers (4.9 or less, 5 to 9.9, or 10 or greater). Patients who reported smoking cessation within 1 year before TURB were considered current smokers. Patients who reported tobacco use other than cigarette smoking, eg tobacco chewing, cigars and pipes, were excluded from analysis.

Followup Regimen Patients were followed according to guidelines, generally every 3 to 4 months during the first 2 years after TURB, biannually up to 5 years and annually thereafter.9,10 Followup consisted of a history, physical examination, routine blood study, urinary cytology, cystoscopy and biopsy of suspicious lesions. Radiographic evaluation of the upper urinary tract to rule out synchronous upper tract urothelial carcinoma was done at NMIBC diagnosis and repeated in cases of disease recurrence, suspicion of upper tract urothelial carcinoma, eg positive cytology, during followup or yearly. When disease recurrence was detected, the tumor was resected. When disease recurrence was not detected but urinary cytology was positive, bladder and prostatic urethral biopsies were performed, in addition to upper tract evaluation. Disease recurrence was defined as the first tumor relapse in the bladder regardless of stage. Progression was defined as tumor relapse with an increase to disease stage T2 or higher in the bladder. Patients alive and without disease recurrence or progression were censored at the date of the last followup. In case of death the cause of death was determined by treating physicians, chart review corroborated by death certificates or death certificates alone.14 Tumor recurrence in the upper urinary tract was not considered tumor recurrence but as a second primary tumor.

Statistical Analysis For statistical analyses, tumor size (less than 3 cm vs 3 or greater) and the number of tumors (solitary vs multifocal) were categorized. The prior recurrence rate represents the

2122

IMPACT OF SMOKING AND HISTORY OF RECURRENT NONMUSCLE INVASIVE BLADDER CANCER

number of disease recurrences in the last year before the index TURB. Smoking quantity (never vs 19 or fewer vs 20 or greater CPD), duration (never vs 19 years or less vs 20 or greater) and years since cessation (never vs 9.9 years or less vs 10 or greater vs current smoking) were grouped based on preliminary analyses (data not shown) and previous studies.15,16 Based on cumulative smoking exposure, we categorized ever smokers into 3 groups, including light short-term—19 or fewer CPD and 19.9 years or less, moderate—19 or fewer CPD and 20 years or greater, or 20 or greater CPD and 19.9 years or less) and heavy longterm (20 or greater CPD and 20 years or greater). Preliminary analyses using different cutoffs revealed that this stratification provided the best discrimination of cumulative smoking exposure. The Kolmogorov-Smirnov test was used to assess variable distribution variables. Differences in variables with a continuous distribution across categories were assessed using the Mann-Whitney U test (2 categories) and the Kruskal-Wallis test (3 or more categories). The Fisher exact and chi-square tests were used to evaluate associations between categorical variables. Differences between actuarial recurrence-free and cancer specific survival probabilities were assessed with the log rank statistic. The Kaplan-Meier method was used to graphically show survivor functions. Univariable and multivariable Cox regression models addressed time to disease recurrence, time to progression and time to all cause mortality after TURB. All analyses were adjusted for the effect of institutions. In all models proportional hazards assumptions were systematically verified using the Grambsch-Therneau residual based test. All reported p values are 2 sided with statistical significance considered at p ⬍0.05. Statistical tests were performed with SPSS® Statistics 20.

RESULTS Smoking Association Clinicopathological findings. Table 1 shows descriptive characteristics of the study cohort. Of the 390 patients 91 (23%) had never smoked, while 192 (49%) and 107 (28%) were former and current smokers, respectively. Median age was 67 (IQR 59, 74), 68 (IQR 60, 72), 66 (IQR 58, 74) and 67 years (IQR 60, 73), respectively (p ⫽ 0.9). Of ever smokers 56 (19%), 156 (52%) and 87 (29%) had light short-term, moderate and heavy long-term cumulative smoking exposure, respectively. On univariable analyses current smoking status was associated with more concomitant CIS (p ⫽ 0.045). There was no association between cumulative smoking exposure and clinicopathological findings. Disease recurrence and progression. Median followup at censorship was 66 months (IQR 34, 96). Of 390 patients 192 (49%) and 64 (16%) experienced disease recurrence and progression, respectively. Median time from TURB to disease recurrence was 7 months (IQR 4, 23). Median time from TURB to disease progression was 36 months (IQR 18, 78). A

total of 159 patients (41%) received instillation chemotherapy immediately postoperatively and 73 (19%) received adjuvant intravesical immunotherapy or chemotherapy. Intravesical chemotherapy immediately postoperatively was not associated with disease recurrence (p ⫽ 0.7). Adjuvant intravesical immunotherapy or chemotherapy was associated with decreased disease recurrence (p ⫽ 0.004) but not with disease progression (p ⫽ 0.3). There were no differences in disease recurrence and progression among never, former and current smokers (fig. 1). Of ever smokers those who smoked 19 or fewer CPD and those who smoked for 19.9 years or less were less likely to experience disease recurrence and disease progression than their counterparts (p ⱕ0.015 and ⬍0.001, respectively, fig. 2). Patients with heavy long-term cumulative smoking exposure were at significantly increased risk for disease recurrence and progression compared to patients with moderate (p ⫽ 0.008 and 0.001, respectively) or light short-term (each p ⬍0.001) cumulative exposure (fig. 2). In turn, patients with moderate cumulative smoking exposure were at significantly increased risk for disease recurrence and progression than patients with short-term cumulative exposure (p ⫽ 0.026 and 0.001, respectively, fig. 2). On multivariable Cox regression analysis adjusted for the effects of standard clinicopathological features and smoking status, increasing cumulative smoking exposure was significantly associated with disease recurrence and progression in ever smokers (p ⱕ0.003, table 2). Smoking Cessation Association with Outcomes Of 192 former smokers 119 (62%) quit smoking less than 10 years (recent former smokers) and 73 (38%) quit smoking more than 10 years (distant former smokers) before the current TURB. Compared to distant former smokers, recent former smokers experienced more disease recurrence (p ⬍0.001, fig. 3). Recent former smokers had an incidence of disease recurrence similar to that of current smokers, who in turn also experienced more disease recurrence than distant former smokers (p ⫽ 0.004). There were no differences between distant and recent former smokers or current smokers in disease progression on univariable analysis (fig. 3). On multivariable Cox regression analysis of ever smokers adjusted for standard clinicopathological features, distant former smokers were at significantly decreased risk for disease recurrence compared to current smokers (HR 0.4, 95% CI 0.2– 0.7, p ⬍0.001, table 2). In contrast, recent former smokers were not significantly different from current smokers in disease recurrence. There was no differ-

IMPACT OF SMOKING AND HISTORY OF RECURRENT NONMUSCLE INVASIVE BLADDER CANCER

2123

Table 1. Descriptive characteristics of 390 patients with history of recurrent urothelial NMIBC treated with TURB with or without intravesical therapy No. Pts (%) Overall Smoking quantity (CPD): 1–9 10–19 20–29 30 or Greater Smoking duration (yrs): Less than 10 10–19.9 20–29.9 30–30.9 40 or Greater Smoking cessation (yrs): 4.9 or Less 5–9.9 10 or Greater Gender: M F Prior recurrence rate/yr: 1 or Less 2 or Greater Pathological stage: pTa pTis pT1 Pathological grade: G1 G2 G3 Concomitant CIS: Absent Present Tumor size (cm): 1 or Less 1.1–2.9 3 or Greater No. tumors: 1 2–7 8 or Greater Immediate postop intravesical chemotherapy: No Yes Adjuvant intravesical therapy: No Yes BCG Chemotherapy

390

No. Never Smoker (%) 91 —

93 (31) 92 (31) 63 (21) 51 (17)

No. Former Smoker (%) 192

No. Current Smoker (%) 107

— 63 (33) 62 (32) 39 (20) 28 (15)

30 (28) 30 (28) 24 (22) 23 (22)

25 (13) 44 (23) 52 (27) 70 (36) 1 (1)

2 (2) 12 (11) 42 (39) 26 (24) 25 (24) —

— 27 (9) 56 (19) 94 (31) 96 (32) 26 (9)

—

— 44 (23) 75 (39) 73 (38)

p Value

—

44 (23) 75 (39) 73 (38) 0.8

284 (73) 106 (27)

64 (70) 27 (30)

140 (73) 52 (27)

80 (75) 27 (25)

195 (50) 195 (50)

47 (52) 44 (48)

95 (49) 97 (51)

53 (49) 54 (51)

265 (68) 6 (2) 119 (30)

59 (65) 2 (2) 30 (33)

134 (70) 1 (1) 57 (29)

72 (67) 3 (3) 32 (30)

144 (37) 112 (29) 134 (34)

26 (28) 29 (32) 36 (40)

71 (37) 58 (30) 63 (33)

47 (44) 25 (23) 35 (33)

360 (92) 30 (8)

85 (93) 6 (7)

182 (95) 10 (5)

93 (87) 14 (13)

228 (58) 129 (33) 33 (9)

47 (51) 37 (41) 7 (8)

107 (56) 68 (35) 17 (9)

74 (69) 24 (23) 9 (8)

168 (43) 150 (39) 72 (18)

40 (44) 35 (38) 16 (18)

72 (37) 82 (43) 38 (20)

56 (52) 33 (31) 18 (17)

231 (59) 159 (41)

51 (56) 40 (44)

111 (58) 81 (42)

69 (65) 38 (35)

317 (81) 73 (19) 60 (15) 13 (4)

68 (75) 23 (25) 16 (18) 7 (7)

164 (85) 28 (15) 23 (12) 5 (3)

85 (79) 22 (21) 21 (20) 1 (1)

0.9

0.5

0.2

0.045

0.07

0.2

0.4

0.08

ence in time from smoking cessation and disease progression. Smoking Association with Survival Overall 113 (29%) patients died of any cause and 40 (10%) died of UCB. Mean ⫾ SE actuarial estimated overall survival at 2, 5 and 10 years was 89% ⫾ 2%, 74% ⫾ 3% and 59% ⫾ 4%, respectively. Current smokers had worse survival than former smokers, who in turn had worse survival than never smokers.

0.1

However, these differences were not statistically significant (p ⬎0.05).

DISCUSSION We found smoking intensity to be associated with disease recurrence and progression in ever smokers. After adjusting for the effects of standard clinicopathological factors and smoking status, cumulative smoking exposure remained a strong, independent

Smoking status

1,0 0,9 0,8 0,7 0,6 0,5

Former Never Current

0,4 0,3

P-value (Logrank-Test):

0,2

Never vs. Former 0.5

0,1

Never vs. Current 0.4

0,0

Former vs. Current 0.7 0

12 24 36 48 60 72 84 96 108 120 Time from Transurethral Resection in Months

Patient numbers at risk for disease recurrence 0 12 24 36 48 60 72 84 96 Never 91 54 39 30 20 16 11 8 5 Former 192 112 93 78 65 51 38 27 20 Current 107 69 52 39 29 23 15 9 8

108 4 14 5

120 2 7 2

Probability of Progression-free Survival

IMPACT OF SMOKING AND HISTORY OF RECURRENT NONMUSCLE INVASIVE BLADDER CANCER

Probability of Recurrence-free Survival

2124

Smoking status

1,0 0,9 0,8

Never

0,7

Former

0,6

Current

0,5 0,4 0,3

P-value (Logrank-Test):

0,2

Never vs. Former 0.7

0,1

Never vs. Current 0.2

0,0

Former vs. Current 0.2 0

12 24 36 48 60 72 84 96 108 120 Time from Transurethral Resection in Months

Patient numbers at risk for disease recurrence 0 12 24 36 48 60 72 84 Never 91 73 62 52 41 34 26 20 Former 192 160 141 126 110 90 71 52 Current 107 90 81 69 58 48 36 27

96 13 37 22

108 9 26 14

120 4 11 6

Figure 1. Kaplan-Meier plots of estimated recurrence-free and progression-free survival by cigarette smoking status in 390 patients with history of recurrent urothelial NMIBC treated with TURB. Green curves indicate former. Blue curves indicate never. Yellow curves indicate current.

risk factor for each end point. The detrimental dose relationship of increasing smoking intensity and unfavorable outcomes was underlined by the increasing odds of moderate to heavy long-term smokers. In addition, we found more unfavorable outcomes in smokers with high smoking quantity and duration. This finding intuitively seems reasonable since dose escalation and longer duration might increase not only the risk of urothelial carcinoma development but also its aggressiveness.2,16,17 Interestingly, light short-term smokers showed better outcomes than never smokers on univariable Kaplan-Meier analysis. This might be explained by different toxins inducing carcinogenesis and/or genetic profiles of the tumors between ever and never smokers. However, although to date we consider that smoking might adversely affect tumor biology and prognosis, the exact biological and molecular mechanisms of smoking induced urothelial carcinogenesis are poorly understood.6 Distinct smoke toxin induced genetic defects may lead to more aggressive UCB clones and might accumulate with increasing smoking intensity.16,17 Moreover, the field effect is probably higher in smokers with heavy exposure than in patients with no or light exposure. These associations are of particular importance since they not only lead to better understanding of tumor biology but also have direct implications for patient counseling regarding smoking cessation. Former smokers who quit more than 10 years before treatment had significantly decreased disease recurrence compared to current and more recent former smokers. These findings are in accordance with those of Chen et al, who found that former smoking was associated with a lower recurrence rate in patients with primary NMIBC.18 Although

we could not adjust for whether this effect was specific at the diagnosis of recurrent NMIBC or whether it would have been the same at the patient primary diagnosis, we confirmed the need for long-term smoking cessation to decrease the effect of smoking in patients with UCB.6,18 Therefore, our results underscore the need for more smoking cessation programs as well as the vital role of urologists and general health care practitioners in influencing patient knowledge regarding smoking risk and the benefits of smoking cessation.19 However, a combination of smoking with other inherent/genetic, environmental, behavioral or lifestyle factors that our study unfortunately could not adjust for might have influenced our results.20,21 The overwhelmingly reported association of smoking with different unhealthy diet and lifestyle factors is a serious concern that must be considered because these factors also seem to be associated with an increased cancer incidence and particularly with decreased life expectancy.20 –22 In addition, the multivariable analysis of disease progression was overfitted, which may have masked the benefits of smoking cessation and decreased disease progression. Smoking status was not associated with disease recurrence and progression in patients with recurrent NMIBC. In addition, we found no difference in overall survival according to smoking status. While previous studies in patients with primary NMIBC demonstrated that former and current smokers have an increased likelihood of disease recurrence and worse outcomes,5– 8 to our knowledge this is the first study to analyze the impact of smoking on outcomes in patients with recurrent NMIBC.

IMPACT OF SMOKING AND HISTORY OF RECURRENT NONMUSCLE INVASIVE BLADDER CANCER

2125

1,0

0,9

0,9

0,8 0,7 0,6 0,5

≤19 CPD

0,4 0,3

≥20 CPD P-value (Logrank-Test): 0.015

0,2 0,1 0,0

Patient numbers at risk for disease recurrence 185 117 95 77 60 51 42 30 24 114 64 51 40 34 22 12 6 4 0

17 2

8 1

Probability of Progression-free Survival

Probability of Recurrence-free Survival

Smoking quantity 1,0

≤19 CPD

0,8 0,7 0,6

≥20 CPD

0,5 0,4 0,3

P-value (Logrank-Test): <0.001

0,2 0,1 0,0

12 24 36 48 60 72 84 96 108 120 Time from Transurethral Resection in Months

Patient numbers at risk for disease progression 185 154 137 122 106 89 73 56 43 30 114 96 85 73 63 48 34 22 16 11 0

13 4

12 24 36 48 60 72 84 96 108 120 Time from Transurethral Resection in Months

1,0

0,9

0,9

0,8 0,7

≤19.9 years

0,6 0,5 0,4 0,3 0,2 0,1 0,0

≥20 years

P-value (Logrank-Test): <0.001 Patient numbers at risk for disease recurrence 83 60 55 49 40 33 28 20 15 216 121 91 68 55 40 26 16 13 0

10 9

5 4

Probability of Progression-free Survival

Probability of Recurrence-free Survival

Smoking duration 1,0

≤19.9 years

0,8

≥20 years

0,7 0,6 0,5 0,4 0,3 0,2 0,1 0,0

12 24 36 48 60 72 84 96 108 120 Time from Transurethral Resection in Months

P-value (Logrank-Test): <0.001 Patient numbers at risk for disease progression 83 77 72 66 60 53 44 34 25 17 216 173 149 129 109 85 63 44 34 23 0

8 9

12 24 36 48 60 72 84 96 108 120 Time from Transurethral Resection in Months

1,0

0,9

0,9

0,8

Light-short-term

0,7 0,6 0,5

Moderate

0,4 0,3 0,2 0,1 0,0

P-value (Logrank-Test): <0.001 Patient numbers 56 42 37 156 94 75 87 46 33 0

Heavy-long-term at risk for disease progression 33 28 25 22 17 13 9 5 59 45 35 25 16 13 9 4 25 22 14 6 3 2 1 1

12 24 36 48 60 72 84 96 108 120 Time from Transurethral Resection in Months

Probability of Progression-free Survival

Probability of Recurrence-free Survival

Cumulative smoking exposure 1,0

Light-short-term

0,8 0,7

Moderate

0,6

Heavy-long-term

0,5 0,4 0,3 0,2 0,1 0,0

P-value (Logrank-Test): <0.001 Patient numbers 56 52 48 156 128 113 87 71 61 0

at risk for disease recurrence 45 41 37 33 27 20 99 84 68 52 37 29 51 44 33 23 15 11

15 17 9

7 7 3

12 24 36 48 60 72 84 96 108 120 Time from Transurethral Resection in Months

Figure 2. Kaplan-Meier plots of recurrence-free and progression-free survival estimates by cigarette smoking quantity and duration, and cumulative smoking exposure in 299 ever smokers with history of recurrent urothelial NMIBC treated with TURB.

Failure to show a significant association between smoking status and outcomes in this study may be explained by the selection process of patients treated with TURB for recurrent NMIBC, which introduced a selection bias. Patients with NMIBC and features of aggressive biology, such as early or frequent tumor recurrence, failure to respond to intravesical therapy or aggressive pathological features, might undergo early radical cystectomy.23,24 Indeed, cumulative evidence suggests an association of smoking with more advanced disease stage and

grade as well as with disease progression.6,7,17,19 Additionally, intravesical therapy might have influenced our results.10 While a previous study showed that smoking did not affect the response to BCG therapy or long-term oncological outcomes,25 the effect of smoking intensity was not assessed in that series. Our study is not devoid of limitations. First and foremost are limitations inherent to the multicenter and retrospective design, including a lack of data on prior treatments (particularly the date of first TURB

2126

IMPACT OF SMOKING AND HISTORY OF RECURRENT NONMUSCLE INVASIVE BLADDER CANCER

Table 2. Multivariable Cox regression analyses predicting disease recurrence and progression in 299 ever smokers with history of recurrent urothelial NMIBC treated with TURB with or without intravesical therapy, adjusted for effects of center Recurrence

Female gender Age Prior recurrence rate Pathological stage (pT1 vs pTa/Tis) Pathological grade (trend): G2 vs G1 G3 vs G1 Concomitant CIS Tumor size No. tumors Intravesical therapy (BCG vs chemotherapy) Smoking cessation vs current (yrs):* 10 or Greater 9.9 or Less Cumulative smoking exposure (trend): Moderate vs light short-term Heavy long-term vs light short-term

Progression

HR (95% CI)

p Value

1.314 (0.908–1.902) 0.992 (0.976–1.008) 1.363 (0.964–1.928) 1.033 (0.341–2.035)

0.148 0.306 0.079 0.688 0.018 0.004 0.403 0.655 0.009 0.002 0.101 ⬍0.001 ⬍0.001 0.053 ⬍0.001 0.006 ⬍0.001

1.827 (1.206–2.768) 1.472 (0.594–3.647) 0.863 (0.454–1.643) 2.004 (1.191–3.372) 1.761 (1.232–2.516) 0.401 (0.235–0.683) 0.403 (0.241–0.671) 1.438 (0.995–2.079) 2.075 (1.231–3.496) 4.307 (2.434–7.622)

HR (95% CI) 1.947 1.036 1.652 1.860

p Value

(1.073–3.530) (1.003–1.071) (0.875–3.119) (0.245–3.017)

0.028 0.032 0.121 0.814 0.072 0.081 0.038 0.640 0.862 0.086 0.973 0.085 0.108 0.475 0.003† — —

2.208 (0.907–5.374) 4.352 (1.082–17.501) 1.253 (0.487–3.225) 1.083 (0.441–2.663) 1.746 (0.923–3.303) 0.986 (0.446–2.179) 0.509 (0.223–1.161) 1.263 (0.666–2.393) — —

* Including time from smoking cessation in former smokers. † Only trend was calculated because no patient progressed in comparative group of light short-term smokers.

garding surgical therapy, introducing a selection bias. Another bias may be the lack of accounting for the consumption of other tobacco products (eg cigars, pipes and tobacco chewing) and different forms of tobacco exposure (eg secondhand smoke and occupational exposure). Finally, smoking history was self-reported and, therefore, subject to recall bias.

and the exact number of previous recurrences) and treatment delay, a lack of routine repeat transurethral resection,12 failure to control for the quality of transurethral resection26 and failure to control for other preoperative prognostic factors, such as lymphovascular invasion.27 We also could not adjust for the number and experience of surgeons and pathologists at each institution, and a central pathology review was not performed. However, all surgeons and pathologists operated at tertiary care centers where there is experience with UCB. We did not adjust our analysis for patient comorbidities, although smokers have a higher risk of noncancer comorbidities.28 In addition, these comorbidities might have influenced the decision making re-

CONCLUSIONS While we found no difference among never, former and current smokers, heavy long-term smoking was a strong risk factor for disease recurrence and progression in patients with a history of recurrent

Years from smoking cessation

0,9 0,8

1,0

P-value (Logrank-Test): ≥10 vs. ≤9.9 years ≥10 vs. Current

<0.001 0.004

≤9.9 years vs. Current 0.2

≤9.9 years

0,7

Current

0,6 0,5 0,4

≥10 years

0,3 0,2 0,1

Incidence of Disease Recurrence

Incidence of Disease Recurrence

1,0

0,9 0,8

P-value (Logrank-Test): ≥10 vs. ≤9.9 years

0.3

≥10 vs. Current

0.1

≤9.9 years vs. Current 0.6

0,7 0,6 0,5 0,4

Current

0,3 0,2

≤9.9 years ≥10 years

0,1 0,0

0,0 0

12 24 36 48 60 72 84 96 Time from Transurethral Resection in Months

108

120

0

12 24 36 48 60 72 84 96 Time from Transurethral Resection in Months

108

120

Figure 3. Cumulative incidence of disease recurrence and progression in 299 ever smokers with history of recurrent urothelial NMIBC treated with TURB by smoking status, considering time from cessation in former smokers.

IMPACT OF SMOKING AND HISTORY OF RECURRENT NONMUSCLE INVASIVE BLADDER CANCER

NMIBC. In addition, a smoking cessation history of more than 10 years seemed to abrogate the detrimental effect of cigarette smoking. These findings should encourage all physicians, particularly urologists, to emphatically counsel their patients to quit

2127

smoking. We believe that cigarette smoking cessation counseling using integrated multidisciplinary programs should be an integral part of the management of NMIBC. Nevertheless, our findings must be confirmed in large, prospective studies.

REFERENCES 1. Siegel R, Naishadham D and Jemal A: Cancer statistics, 2012. CA Cancer J Clin 2012; 62: 10.

der cancer (stages Ta, T1, and Tis): 2007 update. J Urol 2007; 178: 2314.

2. Zeegers MP, Tan FE, Dorant E et al: The impact of characteristics of cigarette smoking on urinary tract cancer risk: a meta-analysis of epidemiologic studies. Cancer 2000; 89: 630.

12. Divrik RT, Sahin AF, Yildirim U et al: Impact of routine second transurethral resection on the long-term outcome of patients with newly diagnosed pT1 urothelial carcinoma with respect to recurrence, progression rate, and disease-specific survival: a prospective randomised clinical trial. Eur Urol 2010; 58: 185.

3. Global Health Risks. Geneva: World Health Organization 2009. 4. Vital signs: current cigarette smoking among adults aged ⱖ18 years—United States, 2005– 2010. MMWR Morb Mortal Wkly Rep 2011; 60: 1207.

13. Herr HW, Dalbagni G and Donat SM: Bacillus Calmette-Guérin without maintenance therapy for high-risk non-muscle-invasive bladder cancer. Eur Urol 2011; 60: 32.

5. Allard P, Fradet Y, Tetu B et al: Tumor-associated antigens as prognostic factors for recurrence in 382 patients with primary transitional cell carcinoma of the bladder. Clin Cancer Res 1995; 1: 1195.

14. Rink M, Fajkovic H, Cha EK et al: Death certificates are valid for the determination of cause of death in patients with upper and lower tract urothelial carcinoma. Eur Urol 2012; 61: 854.

6. Aveyard P, Adab P, Cheng KK et al: Does smoking status influence the prognosis of bladder cancer? A systematic review. BJU Int 2002; 90: 228.

15. Freedman ND, Silverman DT, Hollenbeck AR et al: Association between smoking and risk of bladder cancer among men and women. JAMA 2011; 306: 737.

7. Fleshner N, Garland J, Moadel A et al: Influence of smoking status on the disease-related outcomes of patients with tobacco-associated superficial transitional cell carcinoma of the bladder. Cancer 1999; 86: 2337.

16. Jiang X, Castelao JE, Yuan JM et al: Cigarette smoking and subtypes of bladder cancer. Int J Cancer 2012; 130: 896.

8. Lammers RJ, Witjes WP, Hendricksen K et al: Smoking status is a risk factor for recurrence after transurethral resection of non-muscle-invasive bladder cancer. Eur Urol 2011; 60: 713. 9. National Comprehensive Cancer Network Clinical Practice Guidelines in Oncology: Bladder Cancer 2011. Available at http://www.nccn.org/index.asp. Accessed April 17, 2012. 10. Babjuk M, Oosterlinck W, Sylvester R et al: EAU guidelines on non-muscle-invasive urothelial carcinoma of the bladder, the 2011 update. Eur Urol 2011; 59: 997. 11. Hall MC, Chang SS, Dalbagni G et al: Guideline for the management of nonmuscle invasive blad-

17. Sturgeon SR, Hartge P, Silverman DT et al: Associations between bladder cancer risk factors and tumor stage and grade at diagnosis. Epidemiology 1994; 5: 218. 18. Chen CH, Shun CT, Huang KH et al: Stopping smoking might reduce tumour recurrence in nonmuscle-invasive bladder cancer. BJU Int 2007; 100: 281. 19. Strope SA and Montie JE: The causal role of cigarette smoking in bladder cancer initiation and progression, and the role of urologists in smoking cessation. J Urol 2008; 180: 31. 20. Ferrucci L, Izmirlian G, Leveille S et al: Smoking, physical activity, and active life expectancy. Am J Epidemiol 1999; 149: 645.

21. Zeegers MP, Kellen E, Buntinx F et al: The association between smoking, beverage consumption, diet and bladder cancer: a systematic literature review. World J Urol 2004; 21: 392. 22. Ma J, Betts NM and Hampl JS: Clustering of lifestyle behaviors: the relationship between cigarette smoking, alcohol consumption, and dietary intake. Am J Health Promot 2000; 15: 107. 23. Karakiewicz PI, Shariat SF, Palapattu GS et al: Precystectomy nomogram for prediction of advanced bladder cancer stage. Eur Urol 2006; 50: 1254. 24. Shariat SF, Weizer AZ, Green A et al: Prognostic value of P53 nuclear accumulation and histopathologic features in T1 transitional cell carcinoma of the urinary bladder. Urology 2000; 56: 735. 25. Sfakianos JP, Shariat SF, Favaretto RL et al: Impact of smoking on outcomes after intravesical bacillus Calmette-Guérin therapy for urothelial carcinoma not invading muscle of the bladder. BJU Int 2011; 108: 526. 26. Mariappan P, Zachou A and Grigor KM: Detrusor muscle in the first, apparently complete transurethral resection of bladder tumour specimen is a surrogate marker of resection quality, predicts risk of early recurrence, and is dependent on operator experience. Eur Urol 2010; 57: 843. 27. Cho KS, Seo HK, Joung JY et al: Lymphovascular invasion in transurethral resection specimens as predictor of progression and metastasis in patients with newly diagnosed T1 bladder urothelial cancer. J Urol 2009; 182: 2625. 28. Cokkinides V, Bandi P, McMahon C et al: Tobacco control in the United States–recent progress and opportunities. CA Cancer J Clin 2009; 59: 352.

EDITORIAL COMMENT The relationship of smoking with NMIBC recurrence and progression is clear. This study provides an important contribution by focusing only on patients with recurrent disease. While a dose-response relationship may be no surprise, the magnitude is worth noting with an HR of 2.1 and 4.3 for recurrence in moderate and heavy smokers, respectively,

compared to light smokers. This is particularly salient, considering that these patients had previously experienced disease recurrence. Limitations due to the few events and absent important information, such as the number of prior recurrences, time since diagnosis and associated unhealthy behaviors, necessitate caution when interpreting the data, eg the

2128

IMPACT OF SMOKING AND HISTORY OF RECURRENT NONMUSCLE INVASIVE BLADDER CANCER

increased rate of progression in never vs light or short-term smokers. However, the key question at this point is no longer whether cumulative tobacco exposure impacts bladder cancer outcomes but how we can use these data to help improve our care of patients with NMIBC. To that end, increasing evidence suggests that we have a unique opportunity to curb tobacco use when we treat patients with

NMIBC (ask, advise, assist and arrange).1–3 These efforts may be as impactful as many of our current evidence-based management strategies for improving outcomes in patients with NMIBC. Todd M. Morgan Department of Urology University of Michigan Ann Arbor, Michigan

REFERENCES 1. Bassett JC, Gore JL, Chi AC et al: Impact of a bladder cancer diagnosis on smoking behavior. J Clin Oncol 2012; 30: 1871.

2. Guzzo TJ, Hockenberry MS, Mucksavage P et al: Smoking knowledge assessment and cessation trends in patients with bladder cancer presenting to a tertiary referral center. Urology 2012; 79: 166.

3. Gritz ER, Fingeret MC, Vidrine DJ et al: Successes and failures of the teachable moment: smoking cessation in cancer patients. Cancer 2006; 106: 17.