Background Smoking is a major risk factor for bladder cancer, but the relationship between smoking cessation after initial treatment and bladder cancer recurrence has been investigated less frequently and not prospectively yet. Methods 722 non-muscle-invasive bladder cancer (NMIBC) patients (pTa, pT1, and CIS) from the prospective Bladder Cancer Prognosis Programme (BCPP) cohort, selected in the UK between 2005 and 2011, provided complete data on smok- ing behavior before and up to 5 years after diagnosis. The impact of smoking behavior on NMIBC recurrence was explored by multivariable Cox regression models investigating time-to-first NMIBC recurrence. Results Over a median follow-up period of 4.21 years, 403 pathologically confirmed NMIBC recurrences occurred in 210 patients. Only 25 current smokers at diagnosis quit smoking (14%) during follow-up and smoking cessation after diagnosis did not decrease risk of recurrence compared to continuing smokers (p = 0.352). Conclusions Although quitting smoking after diagnosis might reduce the risk of recurrence based on retrospective evidence, this is not confirmed in this prospective study because the number of NMIBC patients quitting smoking before their first recurrence was too low. Nevertheless, this indicates an important role for urologists and other health care professionals in promoting smoking cessation in NMIBC. Keywords Smoking · Smoking cessation · Bladder cancer · Prognosis · Recurrence · Epidemiology Introduction majority of BCs diagnosed are non-muscle-invasive blad- der cancers (NMIBC) . However, NMIBC often recurs Bladder cancer (BC) is estimated to be the ninth most fre-  and has a risk of progressing to muscle-invasive bladder quent cancer worldwide with approximately 400,000 newly cancer (MIBC) , events which impact on the quality of diagnosed cases per year . Compared to other cancers, life of the patient  and generate high disease management mortality rates are generally lower for BC  since the costs . * Frits H. M. van Osch University Hospital Birmingham, NHS Foundation Trust, email@example.com Birmingham, UK Department of Pharmacology and Toxicology, School Unit of Nutritional and Cancer Epidemiology, Chairgroup for Nutrition and Translational Research in Metabolism of Complex Genetics and Epidemiology, School (NUTRIM), Maastricht University, Maastricht, for Nutrition and Translational Research in Metabolism The Netherlands (NUTRIM), Maastricht University, Maastricht, The Netherlands Chairgroup of Complex Genetics and Epidemiology, Care and Public Health Research Institute (CAPRHI), Maastricht Institute of Cancer and Genomic Sciences, University University, Maastricht, The Netherlands of Birmingham, Birmingham, UK Department of Public Health and Epidemiology, University of Birmingham, Birmingham, UK Vol.:(0123456789) 1 3 676 Cancer Causes & Control (2018) 29:675–683 Although smoking is an established risk factor for BC, its per day), smoking cessation (in years), and tobacco type effects has been less frequently investigated in relation to BC (filter, non-filter, or rolled cigarettes, cigar, or pipe). Monthly prognosis [7–10]. Although many studies investigated effec- smoking status was also assessed retrospectively by postal tiveness of treatment for NMIBC and MIBC with regard to questionnaires that were sent out to participants yearly until recurrence, progression, and mortality, most studies did not the end of follow-up. investigate the effect of smoking or other factors modifi- able by patients on BC prognosis . Nevertheless, the Smoking status at diagnosis and during follow‑up number of studies also reporting hazard ratios (HRs) for BC recurrence by smoking status at diagnosis has increased A combined smoking status variable was created indicat- recently and the current body of evidence consistently shows ing continuing smokers, former smokers who consistently that there is a small association between smoking and BC abstained, never smokers, former smokers who started recurrence when comparing current smokers to never smok- smoking again, and current smokers who quit smoking ers at diagnosis [10, 12]. However, the impact of smoking post-diagnosis. Patients were considered quitters when they cessation after BC diagnosis on recurrence and mortality abstained consistently, so smokers who quit for 3 months has not yet been quantified prospectively . Studies have and then started again were considered as continuing smok- investigated the impact of smoking cessation within 1 year ers. Furthermore, for each participant that reported smoking after diagnosis on BC recurrence, showing a slight decrease cessation during follow-up, it was confirmed whether this in risk of recurrence [14, 15], and one study indicates no occurred before or after their first recurrence. If patients quit effect of quitting after diagnosis on overall or bladder can- smoking after their first recurrence, they were considered as cer-specific mortality . continuing smokers in the time-to-first recurrence analysis. The Bladder Cancer Prognosis Programme (BCPP) fol- lowed up BC patients for 5 years post-diagnosis and investi- Population at risk gated changes in smoking behavior in relation to the course of the disease . The principal aim of this study was to Of the 1,550 cases who agreed to participate, 231 were investigate whether smoking cessation post-diagnosis and subsequently identified as not having BC. Patients who smoking behavior pre-diagnosis influences BC recurrence. presented with MIBC (n = 275) disease at diagnosis were excluded from analysis because they are fundamentally dif- ferent from NMIBC with regard to recurrence. Patients with Methods squamous or adenocarcinomas of non-urothelial origin or with bladder cancer as secondary carcinoma were excluded The Bladder Cancer Prognosis programme (n = 41). In addition to patients presenting with Ta and T1 tumors, carcinoma in situ (CIS) tumors were included This study was conducted within the framework of the West (n = 16) since they have an increased risk of recurrence . Midlands Bladder Cancer Prognosis Programme (BCPP), a In total, 846 (84%) of these patients had provided data on cohort study in the United Kingdom. Details of the study are smoking behavior at diagnosis and during follow-up and described elsewhere . In brief, individuals were included remained under follow-up within the cohort study. Of the between December 2005 and October 2011 after referral to included 846 NMIBC patients, there were 116 patients with participating urology centers due to symptoms suspicious unknown recurrent tumor stage. These 116 unconfirmed of BC and followed for a maximum of 5 years from diag- events were excluded for other analyses as well as 8 cases nosis. Patients with previous cancer of the urethra, bladder, who had radiotherapy (on suspicion of being MIBC cases) ureter, or renal pelvis within the last decade were excluded. resulting in a NMIBC patient population at risk of recur- The study was ethically approved (06/MRE04/65) and all rence of 722. participants gave written informed consent. No systematic guidance or tools were provided to enable patients to quit smoking after diagnosis, so care as usual was Data collection applied by all participating urologists. At or around time of diagnosis, trained research nurses used Statistical analysis semi-structured face-to-face interviews and questionnaires to collect data on social support, health-related quality of BC recurrence was defined as a new tumor that was at the life, sociodemographics, medical history, and health-related same stage as the primary tumor (Ta or T1) but also when behaviors including smoking behavior. Variables on smok- a primary Ta patient had a T1 recurrence. Patients that pro- ing behavior included current smoking status (never, former, gressed from T1 to T2 disease were not counted as a recur- current), duration (years of smoking), intensity (cigarettes rence but as a progression event. Unfortunately, there were 1 3 Cancer Causes & Control (2018) 29:675–683 677 not enough events to also consider biological progression 3 years. Over this period of follow-up, 210 NMIBC patients within this sample of NMIBC patients, as defined in the experienced at least one confirmed recurrence event. These BCPP cohort . Therefore, this study only focussed on 210 NMIBC patients accumulated a total of 403 confirmed confirmed recurrence events and patients who experienced recurrence events in the cohort. a progression event were censored in the survival analysis Most cases were male (79%) and around the age of 70 when the progression event was diagnosed. (Table 1). Furthermore, continuing smokers seemed to be The impact of smoking behavior on BC recurrence was underrepresented in the low EAU risk group (12%), those explored by Cox regression models—with time since initial who quit smoking seemed more likely to be younger and transurethral resection of the bladder tumor (TURBT) as female, and continuing smokers seemed more likely to the time-metric—investigating possible differences in like- present with multiple tumors at diagnosis (Table 1). In lihood of a first recurrence. We explored two different Cox the multivariate models, 26 patients were not included in regression models: one adjusted for age at diagnosis and sex the analysis due to missing data on age (n = 7), number of (model 1) and one additionally adjusted for BC stage, grade, tumors at diagnosis (n = 15), and tumor size (n = 4). Because tumor size, and number of tumors at diagnosis (model 2). participants were recruited from multiple centers, patients This set of confounders was chosen since they are mark- were treated by multiple urologists with different individual ers of NMIBC prognosis and are factors that contribute to thresholds to perform certain therapies. Therefore, not all European Association of Urology (EAU) risk stratification patients were treated exactly according to the EAU guide- for clinical decisions . Moreover, they are potentially lines , which is often the case in actual clinical practice associated with smoking behavior at diagnosis . Conse- . quently, conditional risk set modeling was applied to inves- tigate time between multiple recurrent events and analysis time was reset at each event . For this analysis, re-resec- Associations between smoking behavior pre‑ tion of tumors was added to model 2 as a confounder. The and post‑diagnosis and BC recurrence proportional hazards assumption was checked in all models using Schoenfeld residuals. Cumulative incidence functions Although HR estimates for smoking cessation pre-diag- (CIF) corrected for competing risks (death) were made . nosis indicated a protective association with BC recur- Furthermore, the differences in mean number of recur - rence, the p for linear trend was not statistically significant rences over 5 years between never smokers, former smokers, (p = 0.126) and therefore the association cannot be con- trend and continuing smokers were compared using a multivari- sidered as strong (Table 2). No association between smok- able ANOVA model correcting for pairwise comparisons ing status and risk of recurrence was observed in the mul- using Tukey’s HSD. There were not enough BC-related tivariable model (Table 2). Interestingly, when compared death events (45) or confirmed progression events (19) to continuing smokers (HR 1.04, 95% CI 0.65–1.66), HRs to allow for separate analyses. A similarly low number of were similar for those who quit smoking (p = 0.352) and progression events has been observed in a large (n = 718) former smokers who started again post-diagnosis (p = 0.431) NMIBC patient sample before . (Table 2). Additionally, the cumulative incidence function NMIBC patients who died before the end of follow-up shows that cumulative incidence of BC recurrence was low- (n = 157) were censored at time of death, and patients who est for former smokers and never smokers (Fig. 1). underwent cystectomy (n = 15) were censored at the date Only 25 smokers (14%) of the 174 current smokers origi- of cystectomy (13). Other patients were considered lost to nally recorded at diagnosis quit smoking at any point during follow-up when the date on which patients were last seen in follow-up. Three quitters were excluded for full analysis for the hospital for bladder cancer-related therapy or the date on not having information on their date last seen and another which they filled in their last follow-up questionnaire was five had missing data regarding the invasiveness of their before the end of follow-up (5 years). recurrent events. Of the 480 former smokers at diagnosis, 172 (36%) started smoking (any form of tobacco) again post- diagnosis in all included 846 NMIBC patients. Results Exposure to environmental tobacco smoke during child- hood (HR 1.17, 95% CI 0.81–1.68) or adulthood (HR 1.02, Number of recurrences and characteristics 95% CI 0.76–1.36) did not seem to have any impact on time- of population at risk to-first recurrence (Table 2). Table 3 shows HRs for time-to-first recurrence by smok - All 722 patients at risk of recurrence were followed over ing intensity, duration, and pack-years. No linear trends were a median period of 4.21 years (IQR 2.64–5.00 years). The observed although the highest categories showed the highest majority of patients (506, 70%) were followed for at least point estimates for both smoking intensity and pack-years. 1 3 678 Cancer Causes & Control (2018) 29:675–683 Table 1 Patient characteristics at diagnosis and number of recurrences over 5 years for 722 NMIBC patients treated with transurethral resection by smoking category Overall (n = 722) Combined smoking status Never Former Continu- Former smoker who Quitters after p value* smoker smoker ing smoker started again (n = 150) diagnosis (n = 103) (n = 266) (n = 186) (n = 17) Age in years < 0.001 Median (25th– 71 (63–77) 72 (61–79) 72 (67–79) 67 (57–74) 72 (64–77) 62 (56–67) 75th percentile) Sex < 0.001 Male 573 (79%) 63 (61%) 231 (87%) 139 (75%) 129 (86%) 11 (65%) Female 149 (21%) 40 (39%) 35 (13%) 47 (25%) 21 (14%) 6 (35%) EAU risk group < 0.001 Low 128 (18%) 28 (27%) 71 (27%) 23 (12%) 4 (3%) 2 (12%) Intermediate 383(53%) 50 (49%) 131 (49%) 97 (52%) 91 (61%) 14 (82%) High 211 (29%) 25 (24%) 64 (24%) 66 (36%) 55 (37%) 1 (6%) Number of tumors < 0.001 1 429 (61%) 70 (70%) 179 (69%) 100 (55%) 69 (46%) 11 (65%) 2–7 258 (36%) 27 (27%) 74 (28%) 76 (42%) 75 (50%) 6 (35%) ≥ 8 22 (3%) 3 (3%) 8 (3%) 6 (3%) 5 (3%) 0 (-) Tumor size 0.068 < 3 cm 445 (63%) 68 (68%) 174 (67%) 105 (58%) 85 (57%) 13 (76%) ≥ 3 cm 260 (37%) 32 (32%) 84 (33%) 77 (42%) 63 (43%) 4 (24%) Grade 0.001 1 212 (30%) 34 (34%) 99 (38%) 51 (28%) 26 (17%) 2 (13%) 2 257 (36%) 34 (34%) 75 (28%) 73 (40%) 66 (44%) 9 (56%) 3 245 (34%) 33 (33%) 90 (34%) 60 (32%) 57 (38%) 5 (31%) Stage 0.590 pTa 476 (66%) 68 (66%) 184 (69%) 115 (62%) 95 (63%) 14 (82%) pT1 239 (33%) 35 (34%) 79 (30%) 69 (37%) 53 (35%) 3 (18%) pCis 7 (1%) 0 (-) 3 (1%) 2 (1%) 2 (1%) 0 (-) No of recurrences 0.337 1 108 (51%) 18 (62%) 28 (46%) 33 (53%) 27 (52%) 2 (33%) 2 46 (22%) 6 (21%) 16 (26%) 16 (26%) 6 (11%) 2 (33%) > 3 56 (27%) 5 (17%) 17 (28%) 13 (21%) 19 (37%) 2 (33%) Smoking intensity 0.076 1–9 cigarettes 128 (29%) NA 55 (30%) 23 (21%) 42 (34%) 8 (50%) 10–19 cigarettes 140 (32%) NA 53 (28%) 42 (38%) 42 (34%) 3 (19%) >20 cigarettes 167 (38%) NA 78 (42%) 45 (41%) 39 (32%) 5 (31%) Smoking duration < 0.001 1–9 years 45 (10%) NA 26 (14%) 2 (2%) 16(14%) 1 (6%) 10–19 years 83 (19%) NA 43 (23%) 10 (9%) 29 (25%) 1 (6%) 20–29 years 87 (20%) NA 46 (25%) 12 (11%) 27 (23%) 2 (13%) 30–39 years 88 (21%) NA 37 (20%) 28 (25%) 19 (16%) 4 (25%) > 40 years 127 (30%) NA 32 (17%) 60 (54%) 27 (23%) 8 (50%) Smoking cessation 0.051 < 20 years 48 (12%) NA 23 (9%) NA 25 (17%) NA 21–40 years 208 (51%) NA 134 (51%) NA 74 (49%) NA > 40 years 155 (38%) NA 104 (40%) NA 51 (34%) NA *Kruskal–Wallis test for continuous and Chi-square test for categorical variables 1 3 Cancer Causes & Control (2018) 29:675–683 679 Table 2 Cox regression analysis investigating the association between combined smoking status, smoking cessation before diagnosis and passive smoking, and time-to-first recurrence in NMIBC patients treated with TURBT Age and sex adjusted Multivariable model* HR 95% CI Number of events/ HR 95% CI Number of patients at risk events/patients at risk Combined smoking status Never smoker 1.00 Ref. 29/103 1.00 Ref. 28/99 Former smoker 0.79 0.51–1.24 61/266 0.78 0.48–1.24 59/254 Continuing smoker 1.17 0.75–1.83 62/186 1.04 0.65–1.66 61/180 Former smoker who started again** 1.04 0.65–1.64 51/150 0.87 0.53–1.41 49/146 Current smoker who quit smoking*** 1.25 0.52-3.00 6/17 1.47 0.63–3.41 6/17 Smoking cessation (in years)**** < 20 years 0.81 0.46–1.43 15/48 0.82 0.46–1.46 15/47 21–40 years 0.76 0.53–1.08 57/208 0.74 0.51–1.08 54/200 > 40 years 0.67 0.44–1.02 39/155 0.71 0.46–1.09 38/148 p for trend 0.070 0.126 Exposed to passive smoking during childhood? No 1.00 Ref. 36/142 1.00 Ref. 35/138 Yes 1.23 0.86–1.75 173/576 1.17 0.81–1.68 168/554 Exposed to passive smoking during adulthood? No 1.00 Ref. 74/261 1.00 Ref. 74/261 Yes 1.03 0.77–1.38 135/454 1.02 0.76–1.36 135/454 *All estimates adjusted for age, sex, stage, grade, tumor size, and number of tumors **Former smoker who started again and current smoker who quit smoking not included in former smokers at diagnosis ***Smokers who quit after their first event are considered as current smokers ****Reference category = current smokers at diagnosis, estimates also include former smokers who started again after diagnosis recurrence analysis (HR for continuing vs. never smokers is 1.10, 95% CI = 0.72–1.69). However, continuing smokers seemed to have experienced more recurrences than never smokers on average over 5 years on average, however not significantly (0.64 vs. 0.45, p = 0.308). Discussion Smoking cessation post‑diagnosis and BC recurrence and clinical implications The reported HRs give reason to believe that quitting smok- ing does not influence the likelihood of NMIBC recurrence over 5 years when compared to continuing smokers in our Fig. 1 Cumulative incidence functions with correction for competing sample. However, the number of quitters in our prospec- risk (death) indicating cumulative incidence of first recurrence per category of smoking status in NMIBC patients treated with TURBT tive sample was small which complicates drawing conclu- sions for this group. Another (retrospective) patient cohort study which assessed smoking cessation post-diagnosis concluded that quitting smoking significantly reduced risk For smoking duration, the HRs were divergent and did not indicate any trend (p = 0.729) at all. of recurrence (HR 0.45, 95% CI 0.25–0.83, comparing trend quitters to continuing smokers); however, the proportion of When considering multiple events that have occurred in patients (Table 4), the HRs are similar to the time-to-first quitters (~ 43% of current smokers at diagnosis) was also 1 3 680 Cancer Causes & Control (2018) 29:675–683 Table 3 Multivariable Cox Age and sex adjusted Multivariable model* regression analysis concerning the association between HR 95% CI Number of events/ HR 95% CI Number of smoking pack-years, intensity, patients at risk events/patients and duration (recorded at at risk diagnosis) with time-to-first Never smoker 1.00 Ref. 29/103 1.00 Ref. 28/99 recurrence in NMIBC patients treated with TURBT Pack-years 1–9 pack-years 0.86 0.53–1.42 36/141 0.81 0.48–1.37 34/134 10–19 pack-years 0.95 0.54–1.67 22/81 0.92 0.51–1.65 22/80 20–29 pack-years 0.93 0.49–1.77 15/58 0.81 0.42–1.60 15/57 30–39 pack-years 0.70 0.35–1.43 11/55 0.60 0.30–1.22 11/53 > 40 pack-years 1.28 0.76–2.14 30/86 1.14 0.66–1.97 29/83 p for trend 0.365 0.688 Smoking intensity (cigarettes/day) 1–9 cigarettes 0.83 0.50–1.38 32/128 0.81 0.47–1.38 30/122 10–19 cigarettes 0.75 0.45–1.28 31/140 0.61 0.35–1.07 31/138 20+ cigarettes 1.24 0.79–1.96 55/167 1.16 0.72–1.85 54/160 p for trend 0.112 0.198 Smoking duration (in years) 1–9 years 1.03 0.52–2.05 12/45 0.97 0.48–1.95 12/43 10–19 years 0.94 0.54–1.62 22/83 0.85 0.48–1.50 21/78 20–29 years 0.79 0.45–1.39 21/87 0.79 0.44–1.44 20/85 30–39 years 1.08 0.61–1.89 26/88 0.93 0.52–1.66 25/85 40+ years 1.00 0.60–1.64 36/127 0.88 0.52–1.49 36/124 p for trend 0.917 0.729 *All estimates adjusted for age, sex, stage, grade, tumor size, and number of tumors at diagnosis Table 4 Conditional risk HR* 95% CI Number of Mean number of recur- set model investigating time events/patients rences over 5 years (95% between multiple recurrence at risk CI) events in NMIBC patients treated with TURBT by Smoking status smoking status at diagnosis and Never smoker 1.00 Ref. 43/99 0.45 (0.28–0.63) after diagnosis Former smoker 0.71 0.47–1.08 108/254 0.45 (0.33–0.57) Continuing smoker 1.10 0.72–1.69 116/180 0.64 (0.47–0.81) Former smoker who started again 0.89 0.56–1.43 108/146 0.82 (0.57–1.06) Current smoker who quit smoking** 0.85 0.35–2.04 18/19 0.84 (0.10–1.58) *All estimates adjusted for age, sex, stage, grade, tumor size, number of tumors, and re-resection of recur- rent tumor **Smokers who have quit after their first event (n = 2) are also included considerably larger . In another retrospective cohort as lung cancer where this association is stronger, evidence to study, Fleshner et al. concluded that it remained unclear imply a strong, causal relationship between smoking behav- whether smoking cessation at time of diagnosis is benefi- ior after diagnosis and recurrence is still limited ; so cial with regard to BC recurrence  although Aveyard more prospective research is needed. et al. estimated that the Fleshner study shows a HR of 0.71 Considering the prolonged latency period for the develop- (95% CI 0.48–1.05) when comparing quitters to continuing ment of BC after exposures , it is credible that the asso- smokers , which is similar to the estimate observed in ciation between altering smoking behavior post-diagnosis the study by Chen et al. Taken together, the limited evidence and likelihood of a first recurrence or multiple recurrences at this point seems to indicate that quitting smoking at or over 5 years is not as strong as the association between closely after diagnosis could reduce risk of recurrence. How- smoking and carcinogenesis. Similarly, epidemiological ever, even across several smoking-related cancer sites such evidence suggests that pre-diagnostic smoking cessation 1 3 Cancer Causes & Control (2018) 29:675–683 681 does not immediately lower the risk of BC , also indi- tumors. This theory of “field cancerization” proposes that cating a longer latency period than 5 years. Furthermore, it (pre-)malignant transformation of cells has already occurred is considered that a first BC recurrence is often the result of at different sites across the urothelium, explaining why incomplete resection and/or tumor cell re-implantation, and (changing) smoking exposure will not have a large impact that genuine new tumor formation only plays a more impor- on disease prognosis . tant role in later recurrences . It is therefore reasonable Additionally, given that recent reviews indicate no con- to suggest that, because of the DNA-damaging effects of siderable heterogeneity between studies that do not show an cigarette smoke , modifying smoking behavior may only association between environmental tobacco smoke and risk influence later recurrences and possibly those that may occur of BC, it is unlikely that we would have shown any substan- beyond the follow-up period of 5 years reported here. tial association with BC recurrence either [37, 38]. Notwithstanding the results from our study, when consid- Because no substantial association between smoking ering the impact of comorbidities on overall survival in BC status pre-diagnosis and BC recurrence was observed in patients  which include several smoking-related diseases adjusted models it is possible that the tumor characteristics  and other evidence indicating beneficial and significant associated with BC recurrence (stage, grade, tumor size, results of post-diagnostic smoking cessation in retrospective number of tumors) included as confounders in these models studies [14, 15], it is evident that smoking cessation should overshadow the effects of smoking behavior in determining be encouraged for NMIBC patients at diagnosis. risk of BC recurrence  and possible also mortality since It is striking that only 14% of current smokers at diag- no association between quitting smoking after diagnosis and nosis in our sample quit smoking post-diagnosis. There are all-cause or bladder cancer-specific mortality was observed examples of successful smoking cessation interventions in in a large retrospective cohort study . Moreover, since urology , and several studies found that when patients current smokers at diagnosis in our cohort have been asso- were diagnosed with BC they were more likely to quit ciated with having a higher stage, higher grade, and larger smoking [34, 35]. Therefore, urologists should continue to tumor size compared to never smokers , smoking behav- improve smoking cessation counseling in newly diagnosed ior might play a more crucial role in determining risk of NMIBC patients and be updated on the available tools to recurrence already before diagnosis through promoting unfa- improve smoking cessation figures. Moreover, more inter - vorable tumor characteristics associated with BC recurrence vention clinical research investigating smoking cessation at diagnosis, although in a Dutch cohort of 323 UBC patients programmes in NMIBC patients is warranted. there was only a weak association between smoking inten- sity and increased risk of a more aggressive tumor type . Smoking behavior pre‑diagnosis and exposure to environmental tobacco smoke Strengths and weaknesses Smoking cessation was most beneficial, with regard to Despite the prospective nature of our study there were some reducing the risk of recurrence, the longer before diagnosis limitations restricting the analyses. Due to the relatively it happened compared to continuing smokers. This was the short follow-up of this study, long-term effects of smoking strongest association observed in our study and has been cessation post-diagnosis could not be assessed and the num- observed in other studies as well, although not consistently ber of deaths due to BC in the NMIBC patients within our . Other results were in line with earlier studies inves- cohort was too low for Cox regression analysis. Also, it was tigating smoking status at diagnosis and BC recurrence as not possible to obtain detailed information on adjuvant ther- well, by indicating a slightly increased risk of recurrence apy for all patients, so differences in adjuvant therapy could in NMIBC patients for current smokers compared to never not be considered in the statistical analysis. Additionally, smokers in a meta-analysis . we did not correct for biomarkers of BC recurrence such as Another recent study not included in the aforemen- mutations in the FGFR3 or TP53 genes , although they tioned meta-analysis shows similar HRs (HR 1.49, 95% might work together with smoking intensity in predicting CI 0.95–2.33) for current smokers at diagnosis . How- BC outcome . ever, when including this study and our study (data from Furthermore, one of the caveats of using only self- continuing smokers) in the meta-analysis, the pooled HR reported questionnaire data to assess smoking exposure barely changes from 1.27 (95% CI 1.09–1.46) to 1.26 (95% was likely demonstrated in our sample of NMIBC patients. CI 1.12–1.40) , indicating a significantly increased risk The large proportion (about 1 in 3) of former smokers pre- of recurrence for current smokers at diagnosis compared to diagnosis who reported to have started smoking again post- never smokers. Possibly, the lack of association for continu- diagnosis is implausible and is probably observed due to ing smokers in this study can be explained through multiple misclassification of either the questionnaire at baseline or synchronous tumors being present at diagnosis in epithelial during follow-up. A high misclassification rate (47%) when 1 3 682 Cancer Causes & Control (2018) 29:675–683 5. Roychowdhury DF, Hayden A, Liepa AM (2003) Health-related comparing self-reported data on smoking behavior to coti- quality-of-life parameters as independent prognostic factors in nine values in blood was also shown in another sample of advanced or metastatic bladder cancer. J Clin Oncol 21:673–678 bladder cancer patients undergoing surveillance . Prefer- 6. Svatek RS, Hollenbeck BK, Holmäng S, Lee R, Kim SP, Stenzl ably, future studies should consider more reliable ways of A et al (2014) The economics of bladder cancer: costs and con- siderations of caring for this disease. Eur Urol 66:253–262. https verifying smoking exposure through biochemical analysis. ://doi.org/10.1016/j.eurur o.2014.01.006 Unfortunately, at the start of the study we did not antici- 7. Hou L, Hong X, Dai M, Chen P, Zhao H, Wei Q et al (2016) pate this small proportion of quitters after diagnosis which Association of smoking status with prognosis in bladder cancer: is why the analysis concerning quitters is underpowered. a meta-analysis. Oncotarget Impact J. https ://doi.org/10.18632 / oncot arget .13606 8. Li HM, Azhati B, Rexiati M, Wang WG, Li XD, Liu Q et al (2016) Impact of smoking status and cumulative smoking exposure on tumor recurrence of non-muscle-invasive bladder cancer. Int Urol Conclusion Nephrol. https ://doi.org/10.1007/s1125 5-016-1441-6 9. Wilcox AN, Silverman DT, Friesen MC, Locke SJ, Russ DE, Although quitting smoking after diagnosis might reduce Hyun N et al (2016) Smoking status, usual adult occupation, and risk of recurrent urothelial bladder carcinoma: data from The probability of recurrence based on retrospective evidence, Cancer Genome Atlas (TCGA) Project. Cancer Causes Control the number of NMIBC patients quitting smoking in our pro- 27:1429–1435. https ://doi.org/10.1007/s1055 2-016-0821-7 spective study was low. This indicates an important role for 10. van Osch FHM, Jochems SHJ, van Schooten FJ, Bryan RT, urologists and other health care professionals in promoting Zeegers MP (2016) Significant role of lifetime cigarette smok - ing in worsening bladder cancer and upper tract urothelial carci- smoking cessation in NMIBC. Based on the current evi- noma prognosis: a meta-analysis. J Urol 195:872–879. https://doi. dence, smoking cessation pre-diagnosis seems to have the org/10.1016/j.juro.2015.10.139 largest impact on reducing risk of recurrence after NMIBC 11. Gritz ER, Dresler C, Sarna L. Smoking (2005) The missing drug diagnosis. interaction in clinical trials: ignoring the obvious. Cancer Epide- miol Biomark Prev 14:2287–2293. https://doi.or g/10.1158/1055- 9965.EPI-05-0224 Acknowledgments We thank all the West Midlands Consultant Urolo- 12. Simonis K, Shariat SF, Rink M (2014) Smoking and smoking gists and their units involved with the Bladder Cancer Prognosis Pro- cessation effects on oncological outcomes in nonmuscle invasive gramme (BCPP), as well as the BCPP research nurses. The BCPP bladder cancer. Curr Opin Urol 24:492–499 programme was funded by Cancer Research UK and supported by 13. Crivelli JJ, Xylinas E, Kluth LA, Rieken M, Rink M, Shariat SF the Institute of Applied Health Research (IAHR) and the Institute of (2014) Effect of smoking on outcomes of urothelial carcinoma: a Cancer of Genomic Sciences, University of Birmingham. We would systematic review of the literature. Eur Urol 65:742–754. https :// also like to thank Professor Lucinda Billingham’s group and Dr. Ton doi.org/10.1016/j.eurur o.2013.06.010 Ambergen from the University of Birmingham and Maastricht Uni- 14. Chen C-H, Shun C-T, Huang K-H, Huang C-Y, Tsai Y-C, Yu versity, respectively, for their advice regarding the statistical analyses. H-J et al (2007) Stopping smoking might reduce tumour recur- rence in nonmuscle-invasive bladder cancer. BJU Int. https ://doi. Open Access This article is distributed under the terms of the Crea- org/10.1111/j.1464-410X.2007.06873 .x tive Commons Attribution 4.0 International License (http://creat iveco 15. Fleshner N, Garland J, Moadel A, Herr H, Ostroff J, Trambert R mmons.or g/licenses/b y/4.0/), which permits unrestricted use, distribu- et al (1998) Influence of smoking status on the disease-related out- tion, and reproduction in any medium, provided you give appropriate comes of patients with tobacco-associated superficial transitional credit to the original author(s) and the source, provide a link to the cell carcinoma of the bladder. Cancer 86:2337–2345 Creative Commons license, and indicate if changes were made. 16. Koshiaris C, Aveyard P, Oke J, Ryan R, Szatkowski L, Stevens R et al (2017) Smoking cessation and survival in lung, upper aero- digestive tract and bladder cancer: cohort study. Br J Cancer Nat Publ Group 117:1224–1232. https://doi.or g/10.1038/bjc.2017.179 17. Zeegers MP, Bryan RT, Langford C, Billingham L, Murray P, References Deshmukh NS et al (2010) The west midlands bladder cancer prognosis programme: rationale and design. BJU Int 105:784– 1. Antoni S, Ferlay J, Soerjomataram I, Znaor A, Jemal A, Bray F 788. https ://doi.org/10.1111/j.1464-410X.2009.08849 .x (2017) Bladder cancer incidence and mortality: a global overview 18. Sylvester RJ, van der Meijden APM, Oosterlinck W, Witjes JA, and recent trends. Eur Urol 71:96–108. https ://doi.org/10.1016/j. Bouffioux C, Denis L et al (2006) Predicting recurrence and pro- eurur o.2016.06.010 gression in individual patients with stage Ta T1 bladder cancer 2. Bryan RT, Zeegers MP, van Roekel EH, Bird D, Grant MR, Dunn using EORTC risk tables: a combined analysis of 2596 patients JA et al (2013) A comparison of patient and tumour characteristics from seven EORTC trials. Eur Urol 49:466–477. https ://doi. in two UK bladder cancer cohorts separated by 20 years. BJU Int org/10.1016/j.eurur o.2005.12.031 112:169–175. https ://doi.org/10.1111/bju.12032 19. Bryan RT, Cheng KK, James ND, Zeegers MP, Wallace DMA 3. Yan Y, Andriole GL, Humphrey PA, Kibel AS (2002) Patterns of (2014) Re: defining progression in nonmuscle invasive bladder multiple recurrences of superficial (Ta/T1) transitional cell carci- cancer: it is time for a new, standard definition: D. Lamm, R. noma of bladder and effects of clinicopathologic and biochemical Persad, M. Brausi, R. Buckley, J. A. Witjes, J. Palou, A. Böhle, factors. Cancer 95:1239–1246. https://doi.or g/10.1002/cncr.10822 A. M. Kamat, M. Colombel and M. Soloway J Urol 2014; 191: 4. Kaufman DS, Shipley WU, Feldman AS. Bladder cancer. Lancet 20–27. J Urol. https ://doi.org/10.1016/j.juro.2014.01.011 374:239–249. https ://doi.org/10.1016/S0140 -6736(09)60491 -8 20. Babjuk M, Böhle A, Burger M, Capoun O, Cohen D, Compé- rat EM et al (2017) EAU guidelines on non–muscle-invasive 1 3 Cancer Causes & Control (2018) 29:675–683 683 urothelial carcinoma of the bladder: Update 2016 [Internet]. Eur comorbidity in a hospital-based cancer registry. JAMA 291:2441. Urol. https ://doi.org/10.1016/j.eurur o.2016.05.041https ://doi.org/10.1001/jama.291.20.2441 21. Babjuk M, Oosterlinck W, Sylvester R, Kaasinen E, Böhle A, 33. Bjurlin MA, Cohn MR, Kim DY, Freeman VL, Lombardo L, Palou-Redorta J et al (2012) EAU guidelines on non-muscle- Hurley SD et al (2013) Brief smoking cessation intervention: a invasive urothelial carcinoma of the bladder, the 2011 update. prospective trial in the urology setting. J Urol 189:1843–1849. Actas Urológicas Españolas 36:389–402https ://doi.org/10.1016/j.juro.2012.11.075 22. Amorim LDAF, Cai J (2015) Modelling recurrent events: a tuto- 34. Bassett JC, Gore JL, Chi AC, Kwan L, McCarthy W, Chamie K rial for analysis in epidemiology. Int J Epidemiol 44:324–333. et al (2012) Impact of a bladder cancer diagnosis on smoking https ://doi.org/10.1093/ije/dyu22 2 behavior. J Clin Oncol 30:1871–1878. https ://doi.org/10.1200/ 23. Latouche A, Allignol A, Beyersmann J, Labopin M, Fine JP JCO.2011.36.6518 (2013) A competing risks analysis should report results on all 35. Ostroff J, Garland J, Moadel A, Fleshner N, Hay J, Cramer L cause-specific hazards and cumulative incidence functions. et al (2000) Cigarette smoking patterns in patients after treat- J Clin Epidemiol 66:648–653. https ://doi.or g/10.1016/j.jclin ment of bladder cancer. J Cancer Educ 15:86–90. https ://doi. epi.2012.09.017org/10.1080/08858 19000 95286 63 24. Lammers RJM, Witjes WPJ, Hendricksen K, Caris CTM, Janzing- 36. Braakhuis BJM, Tabor MP, Kummer JA, Leemans CR, Braken- Pastors MHC, Witjes JA (2011) Smoking status is a risk factor for hoff RH (2003) A genetic explanation of Slaughter’s concept of recurrence after transurethral resection of non–muscle-invasive field cancerization: evidence and clinical implications. Cancer Res bladder cancer. Eur Urol 60:713–720. https ://doi.org/10.1016/j. 63:1727–1730 eurur o.2011.07.010 37. Osch F van, Jochems S, Wesselius A, Schooten F van, Bryan 25. Hendricksen K, Aziz A, Bes P, Chun FK-H, Dobruch J, Kluth R, Zeegers M (2018) A stratified meta-analysis of the associa- LA et al (2017) Discrepancy between European Association of tion between exposure to environmental tobacco smoke during urology guidelines and daily practice in the management of non– childhood and adulthood and urothelial bladder cancer risk. Int muscle-invasive bladder cancer: results of a European survey. Eur J Environ Res Public Heal 2018 15:569. https ://doi.org/10.3390/ Urol Focus. https ://doi.org/10.1016/j.euf.2017.09.002IJERP H1504 0569 26. Aveyard P, Adab P, Cheng KK, Wallace DMA, Hey K, Murphy 38. Lee PN, Thornton AJ, Hamling JS, Lee PN (2016) Epidemiologi- MFG (2002) Does smoking status influence the prognosis of blad- cal evidence on environmental tobacco smoke and cancers other der cancer? A systematic review. BJU Int 90:228–239. https://doi. than lung or breast. Regul Toxicol Pharmacol 80:134–163. https org/10.1046/j.1464-410X.2002.02880 .x://doi.org/10.1016/j.yrtph .2016.06.012 27. Warren GW, Sobus S, Gritz ER (2014) The biological and clinical 39. van Roekel EH, Cheng KK, James ND, Wallace DMA, Billingham effects of smoking by patients with cancer and strategies to imple- LJ, Murray PG et al (2013) Smoking is associated with lower ment evidence-based tobacco cessation support. Lancet Oncol age, higher grade, higher stage, and larger size of malignant blad- 15:e568–e580. https ://doi.org/10.1016/S1470 -2045(14)70266 -9 der tumors at diagnosis. Int J Cancer 133:446–454. https ://doi. 28. van Osch FH, Jochems SH, van Schooten F-J, Bryan RT, org/10.1002/ijc.28017 Zeegers MP (2016) Quantified relations between exposure to 40. Barbosa ALA, Vermeulen SHHM., Aben KK, Grotenhuis AJ, tobacco smoking and bladder cancer risk: a meta-analysis of 89 Vrieling A, Kiemeney LA (2018) Smoking intensity and blad- observational studies. Int J Epidemiol 45:857–870. https ://doi. der cancer aggressiveness at diagnosis. PLoS ONE 13:e0194039. org/10.1093/ije/dyw04 4https ://doi.org/10.1371/journ al.pone.01940 39 29. Bryan RT, Collins SI, Daykin MC, Zeegers MP, Cheng K, Wal- 41. Bryan RT, Zeegers MP, James ND, Wallace DMA, Cheng KK lace DMA et al (2010) Mechanisms of recurrence of Ta/T1 (2010) Biomarkers in bladder cancer. BJU Int 105:608–613. https bladder cancer. Ann R Coll Surg Engl 92:519–524. https ://doi.://doi.org/10.1111/j.1464-410X.2009.08880 .x org/10.1308/00358 8410X 12664 19207 6935 42. Mitra AP, Castelao JE, Hawes D, Tsao-Wei DD, Jiang X, Shi S-R 30. Corral R, Lewinger JP, Van Den Berg D, Joshi AD, Yuan J-M, et al (2013) Combination of molecular alterations and smoking Gago-Dominguez M et al (2014) Comprehensive analyses of intensity predicts bladder cancer outcome. Cancer 119:756–765. DNA repair pathways, smoking and bladder cancer risk in Los https ://doi.org/10.1002/cncr.27763 Angeles and Shanghai. Int J Cancer 135:335–347. https ://doi. 43. Thong AE, Petruzella S, Orlow I, Zabor EC, Ehdaie B, Ostroff JS org/10.1002/ijc.28693 et al (2016) Accuracy of self-reported smoking exposure among 31. Megwalu II, Vlahiotis A, Radwan M, Piccirillo JF, Kibel AS bladder cancer patients undergoing surveillance at a tertiary refer- (2008) Prognostic impact of comorbidity in patients with blad- ral center. Eur Urol Focus 2:441–444. https ://doi.org/10.1016/j. der cancer. Eur Urol 53:581–589. https ://doi.org/10.1016/j.eurur euf.2015.12.002 o.2007.10.069 32. Piccirillo JF, Tierney RM, Costas I, Grove L, Edward L. Spitznagel J, AR F et al (2004) Prognostic importance of 1 3
Cancer Causes & Control – Springer Journals
Published: May 30, 2018
It’s your single place to instantly
discover and read the research
that matters to you.
Enjoy affordable access to
over 18 million articles from more than
15,000 peer-reviewed journals.
All for just $49/month
Query the DeepDyve database, plus search all of PubMed and Google Scholar seamlessly
Save any article or search result from DeepDyve, PubMed, and Google Scholar... all in one place.
Get unlimited, online access to over 18 million full-text articles from more than 15,000 scientific journals.
Read from thousands of the leading scholarly journals from SpringerNature, Elsevier, Wiley-Blackwell, Oxford University Press and more.
All the latest content is available, no embargo periods.
“Hi guys, I cannot tell you how much I love this resource. Incredible. I really believe you've hit the nail on the head with this site in regards to solving the research-purchase issue.”Daniel C.
“Whoa! It’s like Spotify but for academic articles.”@Phil_Robichaud
“I must say, @deepdyve is a fabulous solution to the independent researcher's problem of #access to #information.”@deepthiw
“My last article couldn't be possible without the platform @deepdyve that makes journal papers cheaper.”@JoseServera