Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 14-Day Trial for You or Your Team.

Learn More →

Accuracy of the EORTC risk tables and of the CUETO scoring model to predict outcomes in non-muscle-invasive urothelial carcinoma of the bladder

Accuracy of the EORTC risk tables and of the CUETO scoring model to predict outcomes in... FULL PAPER British Journal of Cancer (2013) 109, 1460–1466 | doi: 10.1038/bjc.2013.372 Keywords: bladder cancer; recurrence; progression; risk assessment; bcg; validation Accuracy of the EORTC risk tables and of the CUETO scoring model to predict outcomes in non-muscle-invasive urothelial carcinoma of the bladder 1,2 3 1,4 5 5 6 6 7 7 E Xylinas , M Kent , L Kluth , A Pycha , E Comploj , R S Svatek , Y Lotan , Q-D Trinh , P I Karakiewicz , 8 1 2 3 ,1,9,10 S Holmang , D S Scherr , M Zerbib , A J Vickers and S F Shariat 1 2 Department of Urology, Weill Cornell Medical College, New York, NY, USA; Department of Urology, Cochin Hospital, APHP, Paris Descartes University, Paris, France; Department of Statistics, Memorial Sloan Kettering Cancer Center, 4 5 New York, NY, USA; Department of Urology, University Medical Centre Hamburg-Eppendorf, Hamburg, Germany; Department of Urology, General Hospital of Bolzano, Bolzano, Italy; Department of Urology, University of Texas Southwestern 7 8 Medical Center, Dallas, TX, USA; Department of Urology, University of Montreal, Montreal, Quebec, Canada; Department of Urology, Sahlgrenska University Hospital, Goteborg, Sweden; Division of Medical Oncology, Weill Cornell Medical College, New York, NY, USA and Department of Urology, Medical University of Vienna, Vienna, Austria Background: The European Organization for Research and Treatment of Cancer (EORTC) risk tables and the Spanish Urological Club for Oncological Treatment (CUETO) scoring model are the two best-established predictive tools to help decision making for patients with non-muscle-invasive bladder cancer (NMIBC). The aim of the current study was to assess the performance of these predictive tools in a large multicentre cohort of NMIBC patients. Methods: We performed a retrospective analysis of 4689 patients with NMIBC. To evaluate the discrimination of the models, we created Cox proportional hazard regression models for time to disease recurrence and progression. We incorporated the patients calculated risk score as a predictor into both of these models and then calculated their discrimination (concordance indexes). We compared the concordance index of our models with the concordance index reported for the models. Results: With a median follow-up of 57 months, 2110 patients experienced disease recurrence and 591 patients experienced disease progression. Both tools exhibited a poor discrimination for disease recurrence and progression (0.597 and 0.662, and 0.523 and 0.616, respectively, for the EORTC and CUETO models). The EORTC tables overestimated the risk of disease recurrence and progression in high-risk patients. The discrimination of the EORTC tables was even lower in the subgroup of patients treated with BCG (0.554 and 0.576 for disease recurrence and progression, respectively). Conversely, the discrimination of the CUETO model increased in BCG-treated patients (0.597 and 0.645 for disease recurrence and progression, respectively). However, both models overestimated the risk of disease progression in high-risk patients. Conclusion: The EORTC risk tables and the CUETO scoring system exhibit a poor discrimination for both disease recurrence and progression in NMIBC patients. These models overestimated the risk of disease recurrence and progression in high-risk patients. These overestimations remained in BCG-treated patients, especially for the EORTC tables. These results underline the need for improving our current predictive tools. However, our study is limited by its retrospective and multi-institutional design. *Correspondence: Professor SF Shariat; E-mail: sfshariat@gmail.com or sfs2002@med.cornell.edu Received 2 April 2013; revised 13 June 2013; accepted 22 June 2013; published online 27 August 2013 & 2013 Cancer Research UK. All rights reserved 0007 – 0920/13 1460 www.bjcancer.com | DOI:10.1038/bjc.2013.372 Accuracy of the EORTC risk tables and CUETO scoring model BRITISH JOURNAL OF CANCER Bladder cancer (BCa) is the most common malignancy of the reassigned according to the 2002 American Joint Committee on urinary tract with an estimated 73 510 new cases and 14 880 deaths Cancer TNM staging system. The presence of concomitant cis was in 2012 in the United States (Siegel et al, 2013). At initial diagnosis, defined as the presence of cis in conjunction with another tumour. 470% of patients have non-muscle-invasive bladder cancer (NMIBC), which is managed by transurethral resection of the Follow-up regimen. Patients were followed every 3–6 months bladder (TURB) with or without intravesical therapy (Babjuk et al, during the first 2 years after TURB, biannually up to 5 years, and 2011). Recurrence rates for NMIBC range from 50% to 70%, and annually thereafter (Babjuk et al, 2011; Burger et al, 2012). Follow- B10–15% of patients experience disease progression to muscle- up consisted of a history, physical examination, urinary cytology, invasive disease over a 5-year period (Sylvester et al, 2006; cystoscopy, and biopsy of suspicious lesions. Radiographic Fernandez-Gomez et al, 2009; Babjuk et al, 2011). These statistics evaluation of the upper urinary tract to rule out UTUC was done underscore the need for a continuous, costly follow-up, making at NMIBC diagnosis in every patient and yearly or in case of BCa the most expensive malignancy per patient (Stenzl et al, 2008). disease recurrence or suspicion, such as positive cytology during To predict the short-term and long-term probabilities of disease follow-up. When disease recurrence was detected, the tumour was recurrence and progression, the European Organization for resected. When no evidence of cancer was seen but urinary Research and Treatment of Cancer (EORTC) Genitourinary cytology was positive, bladder and prostatic urethra biopsies in (GU) group has developed a scoring system and risk addition to upper urinary tract work-up were performed. Disease tables (Sylvester et al, 2006), built on data from 2596 patients recurrence was defined as first tumour relapse in the bladder diagnosed with Ta/T1 tumours, who were randomised in seven regardless of tumour stage. Disease progression was defined as previous EORTC-GU group trials. The proposed scoring system tumour relapse at tumour stage T2 or higher in the bladder or in was based on the six most relevant clinical and pathological the prostatic urethra. In case of death, cause of death was predictors of outcomes, such as tumour stage and grade, number of determined by treating physicians, by chart review corroborated by tumours, size, carcinoma in situ (cis), and prior recurrence rate. death certificates, or by death certificates alone (Rink et al, 2012). The main limitation of this scoring system was the low number of Tumour recurrence in the upper urinary tract was not considered patients treated with bacillus Calmette Guerin (BCG). To over- as tumour recurrence but rather as a second primary tumour. come this limitation, the Club Urolo´gico Espan˜ol de Tratamiento Oncologico (CUETO) developed a scoring model, which predicts Statistical analyses. The previously published EORTC model the short- and long-term probability of disease recurrence and incorporated the number of tumours (single, 2–7 or X8), tumour progression in BCG-treated patients (n¼ 1062) (Fernandez- size (o3cm or X3), prior recurrence rate (primary, p1 Gomez et al, 2009). Despite their potential usefulness in daily recurrence per year, 41 recurrence per year), T stage (Ta or T1), practice, to date, few studies have externally validated these models concomitant carcinoma in situ, and grade (G1, G2, or G3). Each of (Fernandez-Gomez et al, 2011; Hernandez et al, 2011). these variables received a scoring value as previously described The aim of the present multicentre study was to externally (Sylvester et al, 2006). Patient scores were then categorized into validate the EORTC risk tables and the CUETO scoring model for risk groups and a probability of disease recurrence and progression predicting disease recurrence and progression in NMIBC patients were given for both end points of 1 and 5 years. We applied this with or without BCG therapy. scoring methodology to our cohort to calculate each patient’s EORTC risk score. Using Kaplan–Meier methods, we then calculated each patient’s risk for disease recurrence and progres- PATIENTS AND METHODS sion at the same time points of 1 and 5 years. To evaluate the discrimination of the EORTC risk tables, we created Cox proportional hazard regression models for time to recurrence Patients. The study was performed with the approval and oversight of the institutional review board at each institution, with and time to progression. We incorporated the patients calculated risk score as a predictor into both of these models and then all participating sites providing the necessary data-sharing agreements before initiation. Templates for NMIBC data collection calculated the concordance indexes. We compared the concor- dance index of our models with the concordance index reported were sent out to eight international centres. In total, 4784 patients underwent TURB for NMIBC between 2000 and 2007. Patients for the 5-year EORTC model. Unfortunately, the methodology for calculating separate 1-year and 5-year concordance indexes was who had a pure pTis disease (n¼ 75) were also excluded, as this group was too small for separate analyses, leaving a final cohort of not reported in the EORTC paper, so we could not compare our models with both time points. Calibration plots were used to 4689 patients. The data of these patients were frozen for analyses in January 2011. compare the EORTC predicted risks with the actual risks seen in our cohort. Calibration plots usually show the predicted versus TURB and instillation therapy. All patients had cystoscopically actual risk among different risk groups of patients. Our calibration proven UCB and underwent complete TURB according to plots were slightly modified to show the EORTC predicted risk vs guideline recommendations (Babjuk et al, 2011). A re-resection the actual risk for each individual risk score and not just by the risk was performed according to guideline recommendations and at score categories. surgeons’ discretion within 2–6 weeks after initial treatment based We then repeated our analyses for the CUETO risk tables. The on pathologic and intraoperative findings. The first adjuvant previously published CUETO model incorporated gender, age instillation was given within 7–21 days of the diagnostic TURB and (o60, 60–70, 470), recurrent tumour, number of tumours (p3 repeated once weekly for 6 weeks. In all, 51% of patients (n¼ 2405) or 43), T stage (Ta or T1), concomitant carcinoma in situ, and received a single immediate postoperative instillation of che- grade (G1, G2, and G3). Similarly to the EORTC method, each of motherapy (essentially mitomycin C). All BCG patients were these variables received a risk scoring value, patients were proposed some form of maintenance therapy (at least 1 year). categorised according to this value, and probabilities of recurrence None of the patients had upper tract urothelial carcinoma (UTUC) and progression were given as previously described (Fernandez- at diagnosis. Gomez et al, 2009). Concordance indexes and calibrations plots Pathologic evaluation. All surgical specimens were processed were again utilised to evaluate the accuracy of the CUETO risk according to standard pathologic procedures. Genitourinary tables. Unfortunately, it was not reported how separate c-indexes pathologists assigned tumour grade according to the 1973 World were calculated at 1 and 5 years, so the 1-year c-index was used for Health Organisation grading system. Pathological stage was our comparisons. www.bjcancer.com | DOI:10.1038/bjc.2013.372 1461 BRITISH JOURNAL OF CANCER Accuracy of the EORTC risk tables and CUETO scoring model We repeated the above analyses in the following subgroups of EORTC c-indexes for disease recurrence and progression at 1 year patients to explore the robustness of both risk tables: treated with were 0.660 and 0.740, respectively (Table 2A). When the CUETO BCG, not treated with BCG, primary and recurrent tumours. All risk groupings were applied to our cohort, the c-indexes for disease statistical analyses were performed using Stata 12.0 (StataCorp, recurrence and progression were 0.523 and 0.616, respectively. For College Station, TX, USA). comparison, the published CUETO c-indexes were 0.636 and 0.687 for disease recurrence and progression at 1 year, respectively (Table 2B). RESULTS Calibration plots were designed to compare the predicted (red) and the actual risk (black) among each individual risk score and for each end point (at 1 and 5 years, respectively). Modified calibration Patient’s clinico-pathologic characteristics and oncological out- plots were used due to the defined risk groupings used in the both comes. Table 1 shows the clinico-pathologic characteristics of the models. Calibrations plots for disease recurrence are displayed in 4689 patients included in the study. The BCG immunotherapy was Figure 1A and B (EORTC) and Figure 2A and B (CUETO), administered to 538 (11%) patients. In all, 2110 patients respectively. With regard to disease recurrence, the EORTC model experienced disease recurrence and 591 patients experienced had fairly good calibration in low-risk patients (score¼ 0). In disease progression. Median follow-up for patients who did not intermediate-risk patients (score¼ 1–9), the EORTC model was experience disease recurrence was 46 months and was 57 months hindered by the heterogeneity of this risk group (underestimation for those who did not experience disease progression. and overestimation). For example, intermediate-risk patients with External validation of the EORTC and the CUETO models to an EORTC risk score from 1 to 9 were included in the same group but had a 1-year risk of disease recurrence ranging from 20% to predict disease recurrence and progression in the entire cohort. The application of the EORTC risk groupings to our 50%. In high-risk patients (score410), the EORTC model cohort resulted in c-indexes for disease recurrence and progression overestimated the risk of disease recurrence, both at 1 and at 5 of 0.597 and 0.662, respectively. For comparison, the published years. The CUETO model had poor calibration for disease recurrence, with underestimation of the risk for low-risk patients (score 0–6) and overestimation for high-risk patients (score410). The calibration for intermediate-risk patients (score¼ 7–9) was Table 1. Clinical characteristics and pathologic features of the 4689 satisfactory. patients with non-muscle-invasive urothelial carcinoma of the bladder Calibration plots for the EORTC and CUETO models for disease progression are displayed in Figures 1C, D, 2C, and D, Patients (n¼ 4689) respectively. Both models resulted in overestimation of the risk of disease progression in high-risk patients (EORTC score47 and Age CUETO score 410) at 1 and 5 years, respectively. Moreover, both Median, IQR 67 (59–74) models resulted in overestimation of the risk of disease progression in intermediate-risk patients (EORTC score¼ 2–6 and CUETO Gender (n,%) score¼ 7–9) at 5 years. Male 3721 (79%) Female 968 (21%) Application to BCG patients. When including only patients who were treated with BCG (n¼ 538), similar results were found for Prior recurrence rate (n,%) Primary 3284 (70%) Table 2A. C-indexes for application of EORTC model in the entire cohort Recurrent 1405 (30%) of 4689 patients with non-muscle-invasive urothelial carcinoma of the p1 recurrence per year 727 (16%) bladder 41 recurrence per year 678 (14%) Number of tumours (n,%) Published EORTC estimates 1 2865 (61%) 2–7 1816 (39%) Current study 1 Year 5 Years X88(o1%) Disease recurrence 0.597 0.660 0.660 Disease progression 0.662 0.740 0.750 Tumour size (n,%) o3 cm 3698 (79%) Abbreviation: EORTC¼ European Organization for Research and Treatment of Cancer. X3 cm 991 (21%) Tumour stage (n,%) Ta 3030 (65%) Table 2B. C-indexes for application of CUETO model in the entire cohort T1 1659 (35%) of 4689 patients with non-muscle-invasive urothelial carcinoma of the bladder Tumour grade (n,%) G1 1419 (30%) Published CUETO estimates G2 1428 (30%) G3 1842 (39%) Current study 1 Year 5 Years Concomitant CIS (n, %) 223 (5%) Disease recurrence 0.523 0.636 0.644 Adjuvant treatment with BCG (n, %) 538 (11%) Disease progression 0.616 0.687 0.700 Abbreviations: BCG¼ Bacille Calmette Gue´ rin; Cis¼ Carcinoma in situ;IQR¼ interquartile Abbreviation: CUETO¼ Urological Club for Oncological Treatment. range. 1462 www.bjcancer.com | DOI:10.1038/bjc.2013.372 Accuracy of the EORTC risk tables and CUETO scoring model BRITISH JOURNAL OF CANCER AB 100% 100% 80% 80% 60% 60% 40% 40% 20% 0% 20% 01 2345 67 89 10 0 1 2345 67 89 10 EORTC score EORTC score CD 50% 50% 40% 40% 30% 30% 20% 20% 10% 10% 0% 0% 0 12345 6 7 8 9 10 11 12 13 14 1516 012345 6 7 8 9 10111213141516 EORTC score EORTC score Figure 1. (A–D) Calibration plots for the EORTC model in the entire cohort of 4689 patients with non-muscle-invasive urothelial carcinoma of the bladder (Grey – EORTC risk, Black – actual risk). (A) One-year risk of recurrence. (B) Five-year risk of recurrence. (C) One-year risk of progression. (D) Five-year risk of progression. AB 80% 100% 80% 60% 60% 40% 40% 20% 20% 0% 0% 01 2345 67 8 9 012345 678 9 10 CUETO score CUETO score CD 50% 80% 70% 40% 60% 30% 50% 40% 20% 30% 20% 10% 10% 0% 0% 0123456789 10 01 2345 67 89 10 CUETO score CUETO score Figure 2. (A–D) Calibration plots for the CUETO model in the entire cohort of 4689 patients with non-muscle-invasive urothelial carcinoma of the bladder (Grey – CUETO risk; Black – actual risk). (A) One-year risk of recurrence. (B) Five-year risk of recurrence. (C) One-year risk of progression. (D) Five-year risk of progression. both models. The c-indexes for the EORTC model when restricted The calibration plots for these patients indicate even poorer to only BCG-treated patients were even lower than those from the calibration than the full cohort. The EORTC model overestimated development cohorts. The EORTC model had c-indexes of 0.554 both risks of disease recurrence and progression in high-risk and 0.576 for disease recurrence and progression, respectively patients at 1 year and in intermediate- and high-risk patients at (Table 3). Conversely, the discrimination of the CUETO model 5 years (Figure 3A–D). The CUETO model was more accurate increased with c-indexes of 0.597 and 0.645 for disease recurrence concerning prediction of disease recurrence at 1 year, while and progression, respectively (Table 3). overestimating this risk at 5 years in high-risk patients (score410) www.bjcancer.com | DOI:10.1038/bjc.2013.372 1463 Probability of progression Probability of recurrence Probability of progression Probability of recurrence Probability of progression Probability of recurrence Probability of progression Probability of recurrence BRITISH JOURNAL OF CANCER Accuracy of the EORTC risk tables and CUETO scoring model (Figure 4A and B). With regard to disease progression, it disease to allow evidence-based risk stratification with regard to overestimated this risk in intermediate (score¼ 7–9) and high- administration of therapy and follow-up scheduling. The EORTC risk patients (score410), both at 1 and at 5 years (Figure 4C and D). and CUETO models were born out of this clinical need to When including only patients who were not treated with BCG individualise and optimise therapy for NMIBC patients. These (n¼ 4151), the discrimination of the EORTC tables improved for tables are indeed the best currently existing prognostic models, and both disease recurrence and progression (c-indexes of 0.603 and their use, as recommended by various guidelines (Babjuk et al, 0.672, respectively; Table 3). The discrimination of the CUETO 2011; Burger et al, 2012) has become the standard of care. External model decreased (c-indexes of 0.522 and 0.626 for disease validation of a prognostic model on a new data set of more recurrence and progression, respectively; Table 3). contemporaneous patients is crucial to assess its generalisability. Frequently, the performance of a predictive model is typically Application to primary and recurrent tumour patients. Sub- overestimated in the original data used to develop the model. group analyses in patients with primary tumours or recurrent Therefore, we set out to test the discrimination and calibration of tumours only revealed no large differences compared with the data these models in an external multicentre cohort. Despite the ability in the combined group (Table 3). of a prognostic model to improve the decision-making process, the validity of predictive models is based both on the agreement between observed and predicted probabilities of an event (i.e., DISCUSSION calibration) and its ability to distinguish subjects with different prognoses (i.e., discrimination). Non-muscle-invasive bladder cancer is a common disease with We found that the EORTC tables exhibited a poor discrimina- highly variable behaviour and outcomes. Many factors predicting tion in predicting both disease recurrence and progression (Seo disease recurrence and progression have been reported in this et al, 2010; Fernandez-Gomez et al, 2011). Similarly to two previous external validations (Fernandez-Gomez et al, 2011; Hernandez et al, 2011), we found that the EORTC tables overestimated these risks in high-risk patients. This finding could Table 3. C-indexes of the EORTC and the CUETO model when applied be explained by the major limitation of the EORTC series: the low to BCG only, no BCG, primary and recurrent tumours, respectively number of patients who received adjuvant BCG instillations indeed the model was constructed based on the data from EORTC trials Recurrence Progression testing the efficacy of various intravesical chemotherapy regimens in the post-TUR setting. Intravesical chemotherapies are con- EORTC CUETO EORTC CUETO sidered to be inferior to BCG therapy for high-risk tumours (Malmstrom et al, 2009). Only a single EORTC trial, used for the BCG patients 0.554 0.597 0.576 0.645 EORTC tables, included patients managed with induction No BCG patients 0.603 0.522 0.672 0.626 intravesical BCG, albeit without any form of maintenance therapy. Primary tumours 0.598 0.505 0.637 0.621 As BCG therapy has been shown to be more effective in preventing Recurrent tumours 0.585 0.547 0.660 0.638 disease recurrence (and may delay disease progression) than is intravesical chemotherapy, the overestimation by the EORTC Abbreviations: BCG¼ Bacille Calmette Gue´ rin; CUETO¼ Urological Club for Oncological tables of the risk of disease recurrence and progression was Treatment; EORTC¼ European Organization for Research and Treatment of Cancer. expected (Sylvester et al, 2002; Bohle and Bock, 2004; Malmstrom AB 100% 100% 80% 80% 60% 60% 40% 40% 20% 20% 0% 0% 012345 6789 10 012345 6789 10 EORTC score EORTC score CD 100% 100% 80% 80% 60% 60% 40% 40% 20% 20% 0% 0% 012345 6 7 8 9 10 11 12 13 14 1516 012345 6 7 8 9 10 11 12 13 14 1516 EORTC score EORTC score Figure 3. (A–D) Calibration plots for the EORTC model in 538 patients with non-muscle-invasive urothelial carcinoma of the bladder treated with BCG (Grey – EORTC risk; Black – actual risk). (A) One-year risk of recurrence. (B) Five-year risk of recurrence. (C) One-year risk of progression. (D) Five-year risk of progression. 1464 www.bjcancer.com | DOI:10.1038/bjc.2013.372 Probability of progression Probability of recurrence Probability of progression Probability of recurrence Accuracy of the EORTC risk tables and CUETO scoring model BRITISH JOURNAL OF CANCER AB 100% 100% 80% 80% 60% 60% 40% 40% 20% 20% 0% 0% 012345 6789 10 0123456789 10 CUETO score CUETO score CD 100% 100% 80% 80% 60% 60% 40% 40% 20% 20% 0% 0% 012345 6789 10 012345 6789 10 CUETO score CUETO score Figure 4. (A–D) Calibration plots for the CUETO model in 538 patients with non-muscle-invasive urothelial carcinoma of the bladder treated with BCG (Grey – CUETO risk; Black – actual risk). (A) One-year risk of recurrence. (B) Five-year risk of recurrence. (C) One-year risk of progression. (D) Five-year risk of progression. et al, 2009). Moreover, when we restricted the analyses to the factors such as lymphovascular invasion. In particular, given the patients who did not receive BCG immunotherapy, the discrimina- heterogeneous treatment pattern across centres, it is difficult to tion of the EORTC tables increased for both disease recurrence and ascertain whether patients and which patients received re-TURB, a progression. treatment that is instrumental in the proper management of some Due to the limitations in the designing of the EORTC tables and high-risk NMIBC. Similarly, we could not ascertain the number of to improve prediction of outcomes in BCG-treated patients, the patients who completed adjuvant intravesical instillation therapy. CUETO group developed a separate scoring model. Based on These limitations could have impacted the occurrence of events calculations of concordances indices, the authors found that such as disease recurrence and progression. In addition, we could disease recurrence and progression predictions were lower than not adjust for the number and experience of surgeons and those reported by Sylvester et al (2006). We performed the first, to pathologists at each institution; no central pathology review was date, external validation of this scoring model. While, it was not performed, possibly leading to differences in patients’ treatment designed for patients without BCG, we evaluated its performance strategies. However, all surgeons and pathologists operated at in predicting disease recurrence and progression in our overall tertiary care centres with experience in UCB. Comorbidities might cohort of patients, which included only 11% of BCG-treated also have influenced the decision making regarding surgical patients. Not surprisingly, the discriminative ability of this model therapy, introducing a selection bias. However, this series comprise was significantly lower than that of the original study (Sylvester a contemporary and large sample size cohort, which are definitely et al, 2006). Moreover, this scoring model exhibited poor strengths of the study. calibration for both disease recurrence and progression. To better evaluate its discrimination, we restricted our analyses to BCG- treated patients. While the discrimination of this score improved, CONCLUSION we found that it still overestimated the risk of disease progression. Regarding the prediction of disease progression, there was no The EORTC risk tables and the CUETO scoring system exhibit a definite difference between the values presented by the EORTC risk poor discrimination for both disease recurrence and progression in tables and the CUETO scoring model. Both overestimated the NMIBC patients. These models overestimated the risk of disease progression rates in our cohort. Several studies that have compared recurrence and progression in high-risk patients. These over- BCG therapy and intravesical chemotherapy failed to find a estimations remained in BCG-treated patients, especially for the difference between BCG therapy and intravesical chemotherapy in EORTC tables. These results underline the need for improving our preventing disease progression (Lundholm et al, 1996; Shelley et al, current predictive tools. Our study is limited by its retrospective 2004). Additional factors not included in the EORTC or the and multi-institutional design. CUETO models could be added to new prognostic models to enhance their usefulness. Lymphovascular invasion has been proven to be associated with poorer outcomes (Kunju et al, 2008; Streeper et al, 2009). Moreover, molecular factors or grading REFERENCES could also be of interest (van Rhijn et al, 2010). Our study is not devoid of limitations such as its multicentre Babjuk M, Oosterlinck W, Sylvester R, Kaasinen E, Bohle A, Palou-Redorta J, and retrospective study design. We did not control for treatment Roupret M (2011) EAU guidelines on non-muscle-invasive urothelial delay, effect of repeat TURB, quality of the TURB, and prognostic carcinoma of the bladder, the 2011 update. Eur Urol 59(6): 997–1008. www.bjcancer.com | DOI:10.1038/bjc.2013.372 1465 Probability of progression Probability of recurrence Probability of progression Probability of recurrence BRITISH JOURNAL OF CANCER Accuracy of the EORTC risk tables and CUETO scoring model Bohle A, Bock PR (2004) Intravesical Bacille Calmette-Guerin versus Rink M, Fajkovic H, Cha EK, Gupta A, Karakiewicz PI, Chun FK, Lotan Y, mitomycin C in superficial bladder cancer: formal meta-analysis of Shariat SF (2012) Death certificates are valid for the determination of comparative studies on tumor progression. Urology 63(4): 682–686; cause of death in patients with upper and lower tract urothelial carcinoma. discussion 686-7. Eur Urol 61(4): 854–855. Burger M, Oosterlinck W, Konety B, Chang S, Gudjonsson S, Pruthi R, Seo KW, Kim BH, Park CH, Kim CI, Chang HS (2010) The efficacy of the Soloway M, Solsona E, Sved P, Babjuk M, Brausi MA, Cheng C, Comperat E, EORTC scoring system and risk tables for the prediction of recurrence and Dinney C, Otto W, Shah J, Thu¨rof J, Witjes JA (2012) ICUD-EAU progression of non-muscle-invasive bladder cancer after intravesical international consultation on bladder cancer 2012: non-muscle-invasive Bacillus Calmette-Guerin instillation. Korean J Urol 51(3): 165–170. urothelial carcinoma of the bladder. Eur Urol 63(1): 36–44. Shelley MD, Wilt TJ, Court J, Coles B, Kynaston H, Mason MD (2004) Fernandez-Gomez J, Madero R, Solsona E, Unda M, Martinez-Pineiro L, Intravesical Bacillus Calmette-Guerin is superior to mitomycin C in Gonzalez M, Portillo J, Ojea A, Pertusa C, Rodriguez-Molina J, Camacho JE, reducing tumour recurrence in high-risk superficial bladder cancer: Rabadan M, Astobieta A, Montesinos M, Isorna S, Muntanola P, Gimeno A, a meta-analysis of randomized trials. BJU Int 93(4): 485–490. Blas M, Martinez-Pin˜eiro JA (2009) Predicting nonmuscle invasive Siegel R, Naishadham D, Jemal A (2013) Cancer Statistics, 2013. CA Cancer bladder cancer recurrence and progression in patients treated with J Clin 63(1): 11–30. Bacillus Calmette-Guerin: the Cueto Scoring Model. J Urol 182(5): Stenzl A, Hennenlotter J, Schilling D (2008) Can we still afford bladder 2195–2203. cancer? Curr Opin Urol 18(5): 488–492. Fernandez-Gomez J, Madero R, Solsona E, Unda M, Martinez-Pin˜eiro L, Ojea A, Streeper NM, Simons CM, Konety BR, Muirhead DM, Williams RD, Portillo J, Montesinos M, Gonzalez M, Pertusa C, Rodriguez-Molina J, O’Donnell MA, Joudi FN (2009) The significance of lymphovascular Camacho JE, Rabadan M, Astobieta A, Isorna S, Muntan˜ola P, Gimeno A, invasion in transurethral resection of bladder tumour and cystectomy Blas M, Martinez-Pin˜eiro JA. Club Urolo´gico Espan˜ol de Tratamiento specimens on the survival of patients with urothelial bladder cancer. Oncologico (2011) The EORTC tables overestimate the risk of recurrence BJU Int 103(4): 475–479. and progression in patients with non-muscle-invasive bladder cancer Sylvester RJ, van der Meijden AP, Oosterlinck W, Witjes JA, Bouffioux C, treated with Bacillus Calmette-Guerin: external validation of the EORTC Denis L, Newling DW, Kurth K (2006) Predicting recurrence and risk tables. Eur Urol 60(3): 423–430. progression in individual patients with stage Ta T1 bladder cancer using Hernandez V, De La Pena E, Martin MD, Blazquez C, Diaz FJ, Llorente C eortc risk tables: a combined analysis of 2596 patients from seven EORTC (2011) External validation and applicability of the EORTC risk tables for trials. Eur Urol 49(3): 466–5discussion 475-477. non-muscle-invasive bladder cancer. World J Urol 29(4): 409–414. Sylvester RJ, van der Meijden AP, Lamm DL. Intravesical Bacillus Calmette- Kunju LP, You L, Zhang Y, Daignault S, Montie JE, Lee CT (2008) Guerin reduces the risk of progression in patients with superficial bladder Lymphovascular invasion of urothelial cancer in matched transurethral cancer: a meta-analysis of the published results of randomized clinical bladder tumor resection and radical cystectomy specimens. J Urol 180(5): trials (2002) J Urol 168(5): 1964–1970. 1928–1932; discussion 1932. van Rhijn BW, Zuiverloon TC, Vis AN, Radvanyi F, van Leenders GJ, ´ ´ Lundholm C, Norlen BJ, Ekman P, Jahnson S, Lagerkvist M, Lindeborg T, Ooms BC, Kirkels WJ, Lockwood GA, Boeve ER, Jo¨bsis AC, Zwarthoff EC, Olsson JL, Tveter K, Wijkstrom H, Westberg R, Malmstro¨m PU (1996) van der Kwast TH (2010) Molecular grade (Fgfr3/Mib-1) and EORTC risk A randomized prospective study comparing long-term intravesical scores are predictive in primary non-muscle-invasive bladder cancer. Eur instillations of mitomycin C and bacillus calmette-guerin in patients with Urol 58(3): 433–441. superficial bladder carcinoma. J Urol 156(2 Pt 1): 372–376. Malmstrom PU, Sylvester RJ, Crawford DE, Friedrich M, Krege S, Rintala E, This work is published under the standard license to publish agree- Solsona E, Di Stasi SM, Witjes JA (2009) An individual patient data ment. After 12 months the work will become freely available and meta-analysis of the long-term outcome of randomised studies comparing the license terms will switch to a Creative Commons Attribution- intravesical mitomycin C versus Bacillus Calmette-Guerin for non-muscle-invasive bladder cancer. Eur Urol 56(2): 247–256. NonCommercial-Share Alike 3.0 Unported License. 1466 www.bjcancer.com | DOI:10.1038/bjc.2013.372 http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png British Journal of Cancer Springer Journals

Accuracy of the EORTC risk tables and of the CUETO scoring model to predict outcomes in non-muscle-invasive urothelial carcinoma of the bladder

Loading next page...
 
/lp/springer-journals/accuracy-of-the-eortc-risk-tables-and-of-the-cueto-scoring-model-to-ksvtRBzrbj

References (38)

Publisher
Springer Journals
Copyright
Copyright © 2013 by The Author(s)
Subject
Biomedicine; Biomedicine, general; Cancer Research; Epidemiology; Molecular Medicine; Oncology; Drug Resistance
ISSN
0007-0920
eISSN
1532-1827
DOI
10.1038/bjc.2013.372
Publisher site
See Article on Publisher Site

Abstract

FULL PAPER British Journal of Cancer (2013) 109, 1460–1466 | doi: 10.1038/bjc.2013.372 Keywords: bladder cancer; recurrence; progression; risk assessment; bcg; validation Accuracy of the EORTC risk tables and of the CUETO scoring model to predict outcomes in non-muscle-invasive urothelial carcinoma of the bladder 1,2 3 1,4 5 5 6 6 7 7 E Xylinas , M Kent , L Kluth , A Pycha , E Comploj , R S Svatek , Y Lotan , Q-D Trinh , P I Karakiewicz , 8 1 2 3 ,1,9,10 S Holmang , D S Scherr , M Zerbib , A J Vickers and S F Shariat 1 2 Department of Urology, Weill Cornell Medical College, New York, NY, USA; Department of Urology, Cochin Hospital, APHP, Paris Descartes University, Paris, France; Department of Statistics, Memorial Sloan Kettering Cancer Center, 4 5 New York, NY, USA; Department of Urology, University Medical Centre Hamburg-Eppendorf, Hamburg, Germany; Department of Urology, General Hospital of Bolzano, Bolzano, Italy; Department of Urology, University of Texas Southwestern 7 8 Medical Center, Dallas, TX, USA; Department of Urology, University of Montreal, Montreal, Quebec, Canada; Department of Urology, Sahlgrenska University Hospital, Goteborg, Sweden; Division of Medical Oncology, Weill Cornell Medical College, New York, NY, USA and Department of Urology, Medical University of Vienna, Vienna, Austria Background: The European Organization for Research and Treatment of Cancer (EORTC) risk tables and the Spanish Urological Club for Oncological Treatment (CUETO) scoring model are the two best-established predictive tools to help decision making for patients with non-muscle-invasive bladder cancer (NMIBC). The aim of the current study was to assess the performance of these predictive tools in a large multicentre cohort of NMIBC patients. Methods: We performed a retrospective analysis of 4689 patients with NMIBC. To evaluate the discrimination of the models, we created Cox proportional hazard regression models for time to disease recurrence and progression. We incorporated the patients calculated risk score as a predictor into both of these models and then calculated their discrimination (concordance indexes). We compared the concordance index of our models with the concordance index reported for the models. Results: With a median follow-up of 57 months, 2110 patients experienced disease recurrence and 591 patients experienced disease progression. Both tools exhibited a poor discrimination for disease recurrence and progression (0.597 and 0.662, and 0.523 and 0.616, respectively, for the EORTC and CUETO models). The EORTC tables overestimated the risk of disease recurrence and progression in high-risk patients. The discrimination of the EORTC tables was even lower in the subgroup of patients treated with BCG (0.554 and 0.576 for disease recurrence and progression, respectively). Conversely, the discrimination of the CUETO model increased in BCG-treated patients (0.597 and 0.645 for disease recurrence and progression, respectively). However, both models overestimated the risk of disease progression in high-risk patients. Conclusion: The EORTC risk tables and the CUETO scoring system exhibit a poor discrimination for both disease recurrence and progression in NMIBC patients. These models overestimated the risk of disease recurrence and progression in high-risk patients. These overestimations remained in BCG-treated patients, especially for the EORTC tables. These results underline the need for improving our current predictive tools. However, our study is limited by its retrospective and multi-institutional design. *Correspondence: Professor SF Shariat; E-mail: sfshariat@gmail.com or sfs2002@med.cornell.edu Received 2 April 2013; revised 13 June 2013; accepted 22 June 2013; published online 27 August 2013 & 2013 Cancer Research UK. All rights reserved 0007 – 0920/13 1460 www.bjcancer.com | DOI:10.1038/bjc.2013.372 Accuracy of the EORTC risk tables and CUETO scoring model BRITISH JOURNAL OF CANCER Bladder cancer (BCa) is the most common malignancy of the reassigned according to the 2002 American Joint Committee on urinary tract with an estimated 73 510 new cases and 14 880 deaths Cancer TNM staging system. The presence of concomitant cis was in 2012 in the United States (Siegel et al, 2013). At initial diagnosis, defined as the presence of cis in conjunction with another tumour. 470% of patients have non-muscle-invasive bladder cancer (NMIBC), which is managed by transurethral resection of the Follow-up regimen. Patients were followed every 3–6 months bladder (TURB) with or without intravesical therapy (Babjuk et al, during the first 2 years after TURB, biannually up to 5 years, and 2011). Recurrence rates for NMIBC range from 50% to 70%, and annually thereafter (Babjuk et al, 2011; Burger et al, 2012). Follow- B10–15% of patients experience disease progression to muscle- up consisted of a history, physical examination, urinary cytology, invasive disease over a 5-year period (Sylvester et al, 2006; cystoscopy, and biopsy of suspicious lesions. Radiographic Fernandez-Gomez et al, 2009; Babjuk et al, 2011). These statistics evaluation of the upper urinary tract to rule out UTUC was done underscore the need for a continuous, costly follow-up, making at NMIBC diagnosis in every patient and yearly or in case of BCa the most expensive malignancy per patient (Stenzl et al, 2008). disease recurrence or suspicion, such as positive cytology during To predict the short-term and long-term probabilities of disease follow-up. When disease recurrence was detected, the tumour was recurrence and progression, the European Organization for resected. When no evidence of cancer was seen but urinary Research and Treatment of Cancer (EORTC) Genitourinary cytology was positive, bladder and prostatic urethra biopsies in (GU) group has developed a scoring system and risk addition to upper urinary tract work-up were performed. Disease tables (Sylvester et al, 2006), built on data from 2596 patients recurrence was defined as first tumour relapse in the bladder diagnosed with Ta/T1 tumours, who were randomised in seven regardless of tumour stage. Disease progression was defined as previous EORTC-GU group trials. The proposed scoring system tumour relapse at tumour stage T2 or higher in the bladder or in was based on the six most relevant clinical and pathological the prostatic urethra. In case of death, cause of death was predictors of outcomes, such as tumour stage and grade, number of determined by treating physicians, by chart review corroborated by tumours, size, carcinoma in situ (cis), and prior recurrence rate. death certificates, or by death certificates alone (Rink et al, 2012). The main limitation of this scoring system was the low number of Tumour recurrence in the upper urinary tract was not considered patients treated with bacillus Calmette Guerin (BCG). To over- as tumour recurrence but rather as a second primary tumour. come this limitation, the Club Urolo´gico Espan˜ol de Tratamiento Oncologico (CUETO) developed a scoring model, which predicts Statistical analyses. The previously published EORTC model the short- and long-term probability of disease recurrence and incorporated the number of tumours (single, 2–7 or X8), tumour progression in BCG-treated patients (n¼ 1062) (Fernandez- size (o3cm or X3), prior recurrence rate (primary, p1 Gomez et al, 2009). Despite their potential usefulness in daily recurrence per year, 41 recurrence per year), T stage (Ta or T1), practice, to date, few studies have externally validated these models concomitant carcinoma in situ, and grade (G1, G2, or G3). Each of (Fernandez-Gomez et al, 2011; Hernandez et al, 2011). these variables received a scoring value as previously described The aim of the present multicentre study was to externally (Sylvester et al, 2006). Patient scores were then categorized into validate the EORTC risk tables and the CUETO scoring model for risk groups and a probability of disease recurrence and progression predicting disease recurrence and progression in NMIBC patients were given for both end points of 1 and 5 years. We applied this with or without BCG therapy. scoring methodology to our cohort to calculate each patient’s EORTC risk score. Using Kaplan–Meier methods, we then calculated each patient’s risk for disease recurrence and progres- PATIENTS AND METHODS sion at the same time points of 1 and 5 years. To evaluate the discrimination of the EORTC risk tables, we created Cox proportional hazard regression models for time to recurrence Patients. The study was performed with the approval and oversight of the institutional review board at each institution, with and time to progression. We incorporated the patients calculated risk score as a predictor into both of these models and then all participating sites providing the necessary data-sharing agreements before initiation. Templates for NMIBC data collection calculated the concordance indexes. We compared the concor- dance index of our models with the concordance index reported were sent out to eight international centres. In total, 4784 patients underwent TURB for NMIBC between 2000 and 2007. Patients for the 5-year EORTC model. Unfortunately, the methodology for calculating separate 1-year and 5-year concordance indexes was who had a pure pTis disease (n¼ 75) were also excluded, as this group was too small for separate analyses, leaving a final cohort of not reported in the EORTC paper, so we could not compare our models with both time points. Calibration plots were used to 4689 patients. The data of these patients were frozen for analyses in January 2011. compare the EORTC predicted risks with the actual risks seen in our cohort. Calibration plots usually show the predicted versus TURB and instillation therapy. All patients had cystoscopically actual risk among different risk groups of patients. Our calibration proven UCB and underwent complete TURB according to plots were slightly modified to show the EORTC predicted risk vs guideline recommendations (Babjuk et al, 2011). A re-resection the actual risk for each individual risk score and not just by the risk was performed according to guideline recommendations and at score categories. surgeons’ discretion within 2–6 weeks after initial treatment based We then repeated our analyses for the CUETO risk tables. The on pathologic and intraoperative findings. The first adjuvant previously published CUETO model incorporated gender, age instillation was given within 7–21 days of the diagnostic TURB and (o60, 60–70, 470), recurrent tumour, number of tumours (p3 repeated once weekly for 6 weeks. In all, 51% of patients (n¼ 2405) or 43), T stage (Ta or T1), concomitant carcinoma in situ, and received a single immediate postoperative instillation of che- grade (G1, G2, and G3). Similarly to the EORTC method, each of motherapy (essentially mitomycin C). All BCG patients were these variables received a risk scoring value, patients were proposed some form of maintenance therapy (at least 1 year). categorised according to this value, and probabilities of recurrence None of the patients had upper tract urothelial carcinoma (UTUC) and progression were given as previously described (Fernandez- at diagnosis. Gomez et al, 2009). Concordance indexes and calibrations plots Pathologic evaluation. All surgical specimens were processed were again utilised to evaluate the accuracy of the CUETO risk according to standard pathologic procedures. Genitourinary tables. Unfortunately, it was not reported how separate c-indexes pathologists assigned tumour grade according to the 1973 World were calculated at 1 and 5 years, so the 1-year c-index was used for Health Organisation grading system. Pathological stage was our comparisons. www.bjcancer.com | DOI:10.1038/bjc.2013.372 1461 BRITISH JOURNAL OF CANCER Accuracy of the EORTC risk tables and CUETO scoring model We repeated the above analyses in the following subgroups of EORTC c-indexes for disease recurrence and progression at 1 year patients to explore the robustness of both risk tables: treated with were 0.660 and 0.740, respectively (Table 2A). When the CUETO BCG, not treated with BCG, primary and recurrent tumours. All risk groupings were applied to our cohort, the c-indexes for disease statistical analyses were performed using Stata 12.0 (StataCorp, recurrence and progression were 0.523 and 0.616, respectively. For College Station, TX, USA). comparison, the published CUETO c-indexes were 0.636 and 0.687 for disease recurrence and progression at 1 year, respectively (Table 2B). RESULTS Calibration plots were designed to compare the predicted (red) and the actual risk (black) among each individual risk score and for each end point (at 1 and 5 years, respectively). Modified calibration Patient’s clinico-pathologic characteristics and oncological out- plots were used due to the defined risk groupings used in the both comes. Table 1 shows the clinico-pathologic characteristics of the models. Calibrations plots for disease recurrence are displayed in 4689 patients included in the study. The BCG immunotherapy was Figure 1A and B (EORTC) and Figure 2A and B (CUETO), administered to 538 (11%) patients. In all, 2110 patients respectively. With regard to disease recurrence, the EORTC model experienced disease recurrence and 591 patients experienced had fairly good calibration in low-risk patients (score¼ 0). In disease progression. Median follow-up for patients who did not intermediate-risk patients (score¼ 1–9), the EORTC model was experience disease recurrence was 46 months and was 57 months hindered by the heterogeneity of this risk group (underestimation for those who did not experience disease progression. and overestimation). For example, intermediate-risk patients with External validation of the EORTC and the CUETO models to an EORTC risk score from 1 to 9 were included in the same group but had a 1-year risk of disease recurrence ranging from 20% to predict disease recurrence and progression in the entire cohort. The application of the EORTC risk groupings to our 50%. In high-risk patients (score410), the EORTC model cohort resulted in c-indexes for disease recurrence and progression overestimated the risk of disease recurrence, both at 1 and at 5 of 0.597 and 0.662, respectively. For comparison, the published years. The CUETO model had poor calibration for disease recurrence, with underestimation of the risk for low-risk patients (score 0–6) and overestimation for high-risk patients (score410). The calibration for intermediate-risk patients (score¼ 7–9) was Table 1. Clinical characteristics and pathologic features of the 4689 satisfactory. patients with non-muscle-invasive urothelial carcinoma of the bladder Calibration plots for the EORTC and CUETO models for disease progression are displayed in Figures 1C, D, 2C, and D, Patients (n¼ 4689) respectively. Both models resulted in overestimation of the risk of disease progression in high-risk patients (EORTC score47 and Age CUETO score 410) at 1 and 5 years, respectively. Moreover, both Median, IQR 67 (59–74) models resulted in overestimation of the risk of disease progression in intermediate-risk patients (EORTC score¼ 2–6 and CUETO Gender (n,%) score¼ 7–9) at 5 years. Male 3721 (79%) Female 968 (21%) Application to BCG patients. When including only patients who were treated with BCG (n¼ 538), similar results were found for Prior recurrence rate (n,%) Primary 3284 (70%) Table 2A. C-indexes for application of EORTC model in the entire cohort Recurrent 1405 (30%) of 4689 patients with non-muscle-invasive urothelial carcinoma of the p1 recurrence per year 727 (16%) bladder 41 recurrence per year 678 (14%) Number of tumours (n,%) Published EORTC estimates 1 2865 (61%) 2–7 1816 (39%) Current study 1 Year 5 Years X88(o1%) Disease recurrence 0.597 0.660 0.660 Disease progression 0.662 0.740 0.750 Tumour size (n,%) o3 cm 3698 (79%) Abbreviation: EORTC¼ European Organization for Research and Treatment of Cancer. X3 cm 991 (21%) Tumour stage (n,%) Ta 3030 (65%) Table 2B. C-indexes for application of CUETO model in the entire cohort T1 1659 (35%) of 4689 patients with non-muscle-invasive urothelial carcinoma of the bladder Tumour grade (n,%) G1 1419 (30%) Published CUETO estimates G2 1428 (30%) G3 1842 (39%) Current study 1 Year 5 Years Concomitant CIS (n, %) 223 (5%) Disease recurrence 0.523 0.636 0.644 Adjuvant treatment with BCG (n, %) 538 (11%) Disease progression 0.616 0.687 0.700 Abbreviations: BCG¼ Bacille Calmette Gue´ rin; Cis¼ Carcinoma in situ;IQR¼ interquartile Abbreviation: CUETO¼ Urological Club for Oncological Treatment. range. 1462 www.bjcancer.com | DOI:10.1038/bjc.2013.372 Accuracy of the EORTC risk tables and CUETO scoring model BRITISH JOURNAL OF CANCER AB 100% 100% 80% 80% 60% 60% 40% 40% 20% 0% 20% 01 2345 67 89 10 0 1 2345 67 89 10 EORTC score EORTC score CD 50% 50% 40% 40% 30% 30% 20% 20% 10% 10% 0% 0% 0 12345 6 7 8 9 10 11 12 13 14 1516 012345 6 7 8 9 10111213141516 EORTC score EORTC score Figure 1. (A–D) Calibration plots for the EORTC model in the entire cohort of 4689 patients with non-muscle-invasive urothelial carcinoma of the bladder (Grey – EORTC risk, Black – actual risk). (A) One-year risk of recurrence. (B) Five-year risk of recurrence. (C) One-year risk of progression. (D) Five-year risk of progression. AB 80% 100% 80% 60% 60% 40% 40% 20% 20% 0% 0% 01 2345 67 8 9 012345 678 9 10 CUETO score CUETO score CD 50% 80% 70% 40% 60% 30% 50% 40% 20% 30% 20% 10% 10% 0% 0% 0123456789 10 01 2345 67 89 10 CUETO score CUETO score Figure 2. (A–D) Calibration plots for the CUETO model in the entire cohort of 4689 patients with non-muscle-invasive urothelial carcinoma of the bladder (Grey – CUETO risk; Black – actual risk). (A) One-year risk of recurrence. (B) Five-year risk of recurrence. (C) One-year risk of progression. (D) Five-year risk of progression. both models. The c-indexes for the EORTC model when restricted The calibration plots for these patients indicate even poorer to only BCG-treated patients were even lower than those from the calibration than the full cohort. The EORTC model overestimated development cohorts. The EORTC model had c-indexes of 0.554 both risks of disease recurrence and progression in high-risk and 0.576 for disease recurrence and progression, respectively patients at 1 year and in intermediate- and high-risk patients at (Table 3). Conversely, the discrimination of the CUETO model 5 years (Figure 3A–D). The CUETO model was more accurate increased with c-indexes of 0.597 and 0.645 for disease recurrence concerning prediction of disease recurrence at 1 year, while and progression, respectively (Table 3). overestimating this risk at 5 years in high-risk patients (score410) www.bjcancer.com | DOI:10.1038/bjc.2013.372 1463 Probability of progression Probability of recurrence Probability of progression Probability of recurrence Probability of progression Probability of recurrence Probability of progression Probability of recurrence BRITISH JOURNAL OF CANCER Accuracy of the EORTC risk tables and CUETO scoring model (Figure 4A and B). With regard to disease progression, it disease to allow evidence-based risk stratification with regard to overestimated this risk in intermediate (score¼ 7–9) and high- administration of therapy and follow-up scheduling. The EORTC risk patients (score410), both at 1 and at 5 years (Figure 4C and D). and CUETO models were born out of this clinical need to When including only patients who were not treated with BCG individualise and optimise therapy for NMIBC patients. These (n¼ 4151), the discrimination of the EORTC tables improved for tables are indeed the best currently existing prognostic models, and both disease recurrence and progression (c-indexes of 0.603 and their use, as recommended by various guidelines (Babjuk et al, 0.672, respectively; Table 3). The discrimination of the CUETO 2011; Burger et al, 2012) has become the standard of care. External model decreased (c-indexes of 0.522 and 0.626 for disease validation of a prognostic model on a new data set of more recurrence and progression, respectively; Table 3). contemporaneous patients is crucial to assess its generalisability. Frequently, the performance of a predictive model is typically Application to primary and recurrent tumour patients. Sub- overestimated in the original data used to develop the model. group analyses in patients with primary tumours or recurrent Therefore, we set out to test the discrimination and calibration of tumours only revealed no large differences compared with the data these models in an external multicentre cohort. Despite the ability in the combined group (Table 3). of a prognostic model to improve the decision-making process, the validity of predictive models is based both on the agreement between observed and predicted probabilities of an event (i.e., DISCUSSION calibration) and its ability to distinguish subjects with different prognoses (i.e., discrimination). Non-muscle-invasive bladder cancer is a common disease with We found that the EORTC tables exhibited a poor discrimina- highly variable behaviour and outcomes. Many factors predicting tion in predicting both disease recurrence and progression (Seo disease recurrence and progression have been reported in this et al, 2010; Fernandez-Gomez et al, 2011). Similarly to two previous external validations (Fernandez-Gomez et al, 2011; Hernandez et al, 2011), we found that the EORTC tables overestimated these risks in high-risk patients. This finding could Table 3. C-indexes of the EORTC and the CUETO model when applied be explained by the major limitation of the EORTC series: the low to BCG only, no BCG, primary and recurrent tumours, respectively number of patients who received adjuvant BCG instillations indeed the model was constructed based on the data from EORTC trials Recurrence Progression testing the efficacy of various intravesical chemotherapy regimens in the post-TUR setting. Intravesical chemotherapies are con- EORTC CUETO EORTC CUETO sidered to be inferior to BCG therapy for high-risk tumours (Malmstrom et al, 2009). Only a single EORTC trial, used for the BCG patients 0.554 0.597 0.576 0.645 EORTC tables, included patients managed with induction No BCG patients 0.603 0.522 0.672 0.626 intravesical BCG, albeit without any form of maintenance therapy. Primary tumours 0.598 0.505 0.637 0.621 As BCG therapy has been shown to be more effective in preventing Recurrent tumours 0.585 0.547 0.660 0.638 disease recurrence (and may delay disease progression) than is intravesical chemotherapy, the overestimation by the EORTC Abbreviations: BCG¼ Bacille Calmette Gue´ rin; CUETO¼ Urological Club for Oncological tables of the risk of disease recurrence and progression was Treatment; EORTC¼ European Organization for Research and Treatment of Cancer. expected (Sylvester et al, 2002; Bohle and Bock, 2004; Malmstrom AB 100% 100% 80% 80% 60% 60% 40% 40% 20% 20% 0% 0% 012345 6789 10 012345 6789 10 EORTC score EORTC score CD 100% 100% 80% 80% 60% 60% 40% 40% 20% 20% 0% 0% 012345 6 7 8 9 10 11 12 13 14 1516 012345 6 7 8 9 10 11 12 13 14 1516 EORTC score EORTC score Figure 3. (A–D) Calibration plots for the EORTC model in 538 patients with non-muscle-invasive urothelial carcinoma of the bladder treated with BCG (Grey – EORTC risk; Black – actual risk). (A) One-year risk of recurrence. (B) Five-year risk of recurrence. (C) One-year risk of progression. (D) Five-year risk of progression. 1464 www.bjcancer.com | DOI:10.1038/bjc.2013.372 Probability of progression Probability of recurrence Probability of progression Probability of recurrence Accuracy of the EORTC risk tables and CUETO scoring model BRITISH JOURNAL OF CANCER AB 100% 100% 80% 80% 60% 60% 40% 40% 20% 20% 0% 0% 012345 6789 10 0123456789 10 CUETO score CUETO score CD 100% 100% 80% 80% 60% 60% 40% 40% 20% 20% 0% 0% 012345 6789 10 012345 6789 10 CUETO score CUETO score Figure 4. (A–D) Calibration plots for the CUETO model in 538 patients with non-muscle-invasive urothelial carcinoma of the bladder treated with BCG (Grey – CUETO risk; Black – actual risk). (A) One-year risk of recurrence. (B) Five-year risk of recurrence. (C) One-year risk of progression. (D) Five-year risk of progression. et al, 2009). Moreover, when we restricted the analyses to the factors such as lymphovascular invasion. In particular, given the patients who did not receive BCG immunotherapy, the discrimina- heterogeneous treatment pattern across centres, it is difficult to tion of the EORTC tables increased for both disease recurrence and ascertain whether patients and which patients received re-TURB, a progression. treatment that is instrumental in the proper management of some Due to the limitations in the designing of the EORTC tables and high-risk NMIBC. Similarly, we could not ascertain the number of to improve prediction of outcomes in BCG-treated patients, the patients who completed adjuvant intravesical instillation therapy. CUETO group developed a separate scoring model. Based on These limitations could have impacted the occurrence of events calculations of concordances indices, the authors found that such as disease recurrence and progression. In addition, we could disease recurrence and progression predictions were lower than not adjust for the number and experience of surgeons and those reported by Sylvester et al (2006). We performed the first, to pathologists at each institution; no central pathology review was date, external validation of this scoring model. While, it was not performed, possibly leading to differences in patients’ treatment designed for patients without BCG, we evaluated its performance strategies. However, all surgeons and pathologists operated at in predicting disease recurrence and progression in our overall tertiary care centres with experience in UCB. Comorbidities might cohort of patients, which included only 11% of BCG-treated also have influenced the decision making regarding surgical patients. Not surprisingly, the discriminative ability of this model therapy, introducing a selection bias. However, this series comprise was significantly lower than that of the original study (Sylvester a contemporary and large sample size cohort, which are definitely et al, 2006). Moreover, this scoring model exhibited poor strengths of the study. calibration for both disease recurrence and progression. To better evaluate its discrimination, we restricted our analyses to BCG- treated patients. While the discrimination of this score improved, CONCLUSION we found that it still overestimated the risk of disease progression. Regarding the prediction of disease progression, there was no The EORTC risk tables and the CUETO scoring system exhibit a definite difference between the values presented by the EORTC risk poor discrimination for both disease recurrence and progression in tables and the CUETO scoring model. Both overestimated the NMIBC patients. These models overestimated the risk of disease progression rates in our cohort. Several studies that have compared recurrence and progression in high-risk patients. These over- BCG therapy and intravesical chemotherapy failed to find a estimations remained in BCG-treated patients, especially for the difference between BCG therapy and intravesical chemotherapy in EORTC tables. These results underline the need for improving our preventing disease progression (Lundholm et al, 1996; Shelley et al, current predictive tools. Our study is limited by its retrospective 2004). Additional factors not included in the EORTC or the and multi-institutional design. CUETO models could be added to new prognostic models to enhance their usefulness. Lymphovascular invasion has been proven to be associated with poorer outcomes (Kunju et al, 2008; Streeper et al, 2009). Moreover, molecular factors or grading REFERENCES could also be of interest (van Rhijn et al, 2010). Our study is not devoid of limitations such as its multicentre Babjuk M, Oosterlinck W, Sylvester R, Kaasinen E, Bohle A, Palou-Redorta J, and retrospective study design. We did not control for treatment Roupret M (2011) EAU guidelines on non-muscle-invasive urothelial delay, effect of repeat TURB, quality of the TURB, and prognostic carcinoma of the bladder, the 2011 update. Eur Urol 59(6): 997–1008. www.bjcancer.com | DOI:10.1038/bjc.2013.372 1465 Probability of progression Probability of recurrence Probability of progression Probability of recurrence BRITISH JOURNAL OF CANCER Accuracy of the EORTC risk tables and CUETO scoring model Bohle A, Bock PR (2004) Intravesical Bacille Calmette-Guerin versus Rink M, Fajkovic H, Cha EK, Gupta A, Karakiewicz PI, Chun FK, Lotan Y, mitomycin C in superficial bladder cancer: formal meta-analysis of Shariat SF (2012) Death certificates are valid for the determination of comparative studies on tumor progression. Urology 63(4): 682–686; cause of death in patients with upper and lower tract urothelial carcinoma. discussion 686-7. Eur Urol 61(4): 854–855. Burger M, Oosterlinck W, Konety B, Chang S, Gudjonsson S, Pruthi R, Seo KW, Kim BH, Park CH, Kim CI, Chang HS (2010) The efficacy of the Soloway M, Solsona E, Sved P, Babjuk M, Brausi MA, Cheng C, Comperat E, EORTC scoring system and risk tables for the prediction of recurrence and Dinney C, Otto W, Shah J, Thu¨rof J, Witjes JA (2012) ICUD-EAU progression of non-muscle-invasive bladder cancer after intravesical international consultation on bladder cancer 2012: non-muscle-invasive Bacillus Calmette-Guerin instillation. Korean J Urol 51(3): 165–170. urothelial carcinoma of the bladder. Eur Urol 63(1): 36–44. Shelley MD, Wilt TJ, Court J, Coles B, Kynaston H, Mason MD (2004) Fernandez-Gomez J, Madero R, Solsona E, Unda M, Martinez-Pineiro L, Intravesical Bacillus Calmette-Guerin is superior to mitomycin C in Gonzalez M, Portillo J, Ojea A, Pertusa C, Rodriguez-Molina J, Camacho JE, reducing tumour recurrence in high-risk superficial bladder cancer: Rabadan M, Astobieta A, Montesinos M, Isorna S, Muntanola P, Gimeno A, a meta-analysis of randomized trials. BJU Int 93(4): 485–490. Blas M, Martinez-Pin˜eiro JA (2009) Predicting nonmuscle invasive Siegel R, Naishadham D, Jemal A (2013) Cancer Statistics, 2013. CA Cancer bladder cancer recurrence and progression in patients treated with J Clin 63(1): 11–30. Bacillus Calmette-Guerin: the Cueto Scoring Model. J Urol 182(5): Stenzl A, Hennenlotter J, Schilling D (2008) Can we still afford bladder 2195–2203. cancer? Curr Opin Urol 18(5): 488–492. Fernandez-Gomez J, Madero R, Solsona E, Unda M, Martinez-Pin˜eiro L, Ojea A, Streeper NM, Simons CM, Konety BR, Muirhead DM, Williams RD, Portillo J, Montesinos M, Gonzalez M, Pertusa C, Rodriguez-Molina J, O’Donnell MA, Joudi FN (2009) The significance of lymphovascular Camacho JE, Rabadan M, Astobieta A, Isorna S, Muntan˜ola P, Gimeno A, invasion in transurethral resection of bladder tumour and cystectomy Blas M, Martinez-Pin˜eiro JA. Club Urolo´gico Espan˜ol de Tratamiento specimens on the survival of patients with urothelial bladder cancer. Oncologico (2011) The EORTC tables overestimate the risk of recurrence BJU Int 103(4): 475–479. and progression in patients with non-muscle-invasive bladder cancer Sylvester RJ, van der Meijden AP, Oosterlinck W, Witjes JA, Bouffioux C, treated with Bacillus Calmette-Guerin: external validation of the EORTC Denis L, Newling DW, Kurth K (2006) Predicting recurrence and risk tables. Eur Urol 60(3): 423–430. progression in individual patients with stage Ta T1 bladder cancer using Hernandez V, De La Pena E, Martin MD, Blazquez C, Diaz FJ, Llorente C eortc risk tables: a combined analysis of 2596 patients from seven EORTC (2011) External validation and applicability of the EORTC risk tables for trials. Eur Urol 49(3): 466–5discussion 475-477. non-muscle-invasive bladder cancer. World J Urol 29(4): 409–414. Sylvester RJ, van der Meijden AP, Lamm DL. Intravesical Bacillus Calmette- Kunju LP, You L, Zhang Y, Daignault S, Montie JE, Lee CT (2008) Guerin reduces the risk of progression in patients with superficial bladder Lymphovascular invasion of urothelial cancer in matched transurethral cancer: a meta-analysis of the published results of randomized clinical bladder tumor resection and radical cystectomy specimens. J Urol 180(5): trials (2002) J Urol 168(5): 1964–1970. 1928–1932; discussion 1932. van Rhijn BW, Zuiverloon TC, Vis AN, Radvanyi F, van Leenders GJ, ´ ´ Lundholm C, Norlen BJ, Ekman P, Jahnson S, Lagerkvist M, Lindeborg T, Ooms BC, Kirkels WJ, Lockwood GA, Boeve ER, Jo¨bsis AC, Zwarthoff EC, Olsson JL, Tveter K, Wijkstrom H, Westberg R, Malmstro¨m PU (1996) van der Kwast TH (2010) Molecular grade (Fgfr3/Mib-1) and EORTC risk A randomized prospective study comparing long-term intravesical scores are predictive in primary non-muscle-invasive bladder cancer. Eur instillations of mitomycin C and bacillus calmette-guerin in patients with Urol 58(3): 433–441. superficial bladder carcinoma. J Urol 156(2 Pt 1): 372–376. Malmstrom PU, Sylvester RJ, Crawford DE, Friedrich M, Krege S, Rintala E, This work is published under the standard license to publish agree- Solsona E, Di Stasi SM, Witjes JA (2009) An individual patient data ment. After 12 months the work will become freely available and meta-analysis of the long-term outcome of randomised studies comparing the license terms will switch to a Creative Commons Attribution- intravesical mitomycin C versus Bacillus Calmette-Guerin for non-muscle-invasive bladder cancer. Eur Urol 56(2): 247–256. NonCommercial-Share Alike 3.0 Unported License. 1466 www.bjcancer.com | DOI:10.1038/bjc.2013.372

Journal

British Journal of CancerSpringer Journals

Published: Aug 27, 2013

There are no references for this article.