15Archivio Italiano di Urologia e Andrologia 2021; 93, 1 ORIGINAL PAPER No conflict of interest declared. DOI: 10.4081/aiua.2021.1.15 results published in the first years of this century were obtained in patients treated in a time period spanning from 1995 to 2003, with a median age of 65 to 67 years, and no mention with regards to the presence of co-mor- bidities is available. In recent years, however, global life expectancy has increased in both sexes causing an expansion of the elderly segment of the population. Ageing is one of the reason for increase in cancer incidence worldwide and, with regards specifically to BC, a 1.5-fold rise has been observed in subjects of 70 years of age and beyond (7). In the ageing population systemic diseases are concurrently diagnosed, namely, cardio-vascular, respira- tory, metabolic, etc., therefore the need for chronic med- ications often represents the norm rather than the excep- tion (8-9). The present study was aimed at documenting the association of advanced age together with the presence of co-morbidities on clinical outcomes of RC. MATERIALS AND METHODS Study design and patient population The study population consisted of 334 patients submit- ted consecutively to RC from 01/01/2005 to 31/12/2015 at our tertiary care center and teaching institution, the University of Genova, Italy. All the charts were examined and data were extracted with regards to pre-, intra-, and post-operative parameters. Follow-up records and life status were retrieved from the internal follow-up data- base of our Institute, the Liguria Hospitalization Records, the Regional Mortality Registry, and the Genova Cancer Registry. Patients who received neo-adjuvant or adjuvant chemotherapy were excluded from analysis, whereas patients who received salvage (post-RC) chemotherapy and/or radiotherapy were included. Patients’ characteristics The recorded baseline patients’ characteristics were: age at RC (ARC), gender, full medical history, 12 channel Serum Multiple Analysis Compound (SMAC), physical examination, self-reported comorbidities and chronic Objective: To assess the joint effect of age and comorbidities on clinical outcomes of radical cystectomy (RC). Methods: 334 consecutive patients undergoing open RC for bladder cancer (BC) during the years 2005-2015 were ana- lyzed. Pre-, peri- and post-operative parameters, including age at RC (ARC) and Charlson Comorbidity Index (CCI), were evaluated. Overall and cancer-specific survivals (OS, CSS) were assessed by univariate and multivariate modelling. Furthermore, a three-knot restricted cubic spline (RCS) was fitted to survival data to detect dependency between death-rate ratio (HR) and ARC. Results: Median follow-up time was 3.8 years (IQR = 1.3-7.5) while median OS was 5.9 years (95%CL = 3.8-9.1). Globally, 180 patients died in our cohort (53.8%), 112 of which (62.2%) from BC and 68 patients (37.8%) for unrelated causes. After adjusting for preoperative, pathological and perioperative parameters, patients with CCI > 3 showed significantly higher death rates (HR = 1.61; p = 0.022). The highest death rate was recorded in ARC = 71-76 years (HR = 2.25; p = 0.034). After fitting an RCS to both OS and CSS rates, two overlapping non- linear trends, with common highest risk values included in ARC = 70-75 years, were observed. Conclusions: Age over 70 years and CCI > 3 were significant factors limiting the survival of RC and should both be consid- ered when comparing current RC outcomes. KEY WORDS: Radical cystectomy; Comorbidity; Age; Frail patient; Elderly. Submitted 12 November 2020; Accepted 7 December 2020 INTRODUCTION Bladder cancer is one of the most common urological neoplasms and often requires multiple surgical treat- ment, as well as radical and invasive therapies (1-3). Surgery in the form of Radical Cystectomy (RC) repre- sents the mainstay treatment for organ-confined and locally advanced muscle-invasive BC (MIBC). Reference studies of RC showed long-term survival rates around 60% and 40% at 5 and 10 years, respectively (4-6). Such Age above 70 years and Charlson Comorbidity Index higher than 3 are associated with reduced survival probabilities after radical cystectomy for bladder cancer. Data from a contemporary series of 334 consecutive patients Summary Massimo Maffezzini 1, Vincenzo Fontana 2, Andrea Pacchetti 3, Federico Dotta 3, Mattia Cerasuolo 3, Davide Chiappori 3, Giovanni Guano 3, Guglielmo Mantica 3, Carlo Terrone 3 1 Department of Urology, Ospedale Nuovo Legnano, Ospedale Fornaroli Magenta, Milano, Italy; 2 Clinical Epidemiology Unit, IRCCS Policlinico San Martino, University of Genova, Genova, Italy; 3 Department of Urology, IRCCS Policlinico San Martino, University of Genova, Genova, Italy. Archivio Italiano di Urologia e Andrologia 2021; 93, 1 M. Maffezzini, V. Fontana, A. Pacchetti, F. Dotta, M. Cerasuolo, D. Chiappori, G. Guano, G. Mantica, C. Terrone 16 need for medications, ECG and cardiologic assessment, anesthesiology assessment and American Society of Anesthesiology (ASA) score, and clinical stage of disease. In addition, the Charlson comorbidity index (CCI) score was assigned based on the recorded data. Clinical staging was based on the pathology report of staging trans-urethral resection and by chest and abdomen CT scan. Additional tests (i.e. bone scans) were requested at the discretion of the treating physician. TREATMENT All patients underwent RC with bilateral pelvic lym- phadenectomy with the standard technique (4; 5; 6). Briefly, in the male patient the bladder, prostate, and sem- inal vesicles, and in the female patient the bladder, uterus, ovaries and anterior wall of the vagina, were removed en bloc. Pelvic lymphadenectomy included all the tissue overlaying the major pelvic vessels from the crossing of the ureter over the iliac vessels proximally to the internal inguinal ring distally, and from the genito-femoral nerve laterally to the obturator fossa medially. A form of urinary diversion (i.e. orthotopic reservoirs, ileal conduits, and uretero-cutaneostomy) was selected based on disease stage, ARC, co-morbidities, and the surgeon’s preferences. Additional data searched included pathological stage and grade, post-operative mortality (POM), 90-day post-oper- ative incidence of complications using the Clavien- Dindo scale (10), and length of hospitalization. Follow-up data comprised evidence of local and/or dis- tant recurrence, need for further treatment, vital status, and cause of death. Outcomes Two survival outcomes were considered: OS when all deceased patients were assumed to die from BC regard- less of the certified cause of death, and CSS when patients died from causes other than BC were considered as censored at the date of death. Patients’ characteristics taken into consideration for analysis were: ARC, gender, CCI score, baseline hemoglobin (Hb) and creatinine (Cr) levels, tumor size, lymph-node involvement, histotype, type of urinary diversion and complications. Statistical analysis Patients and disease-related prognostic factors were explored using descriptive statistics. Continuous vari- ables were described using median values and ranges of variation (min-max) and categorized according to statis- tically or clinically meaningful thresholds, namely quin- tiles for ARC and a level of 10.0 g/dL, and 1.2 mg/dL for Hb and serum Cr, respectively. A CCI score of 3 was used as a cut-off in order to separate lower from higher comorbid patients. All categorical factors were finally expressed in terms of absolute and relative frequencies. Univariate survival comparisons were carried out using the Kaplan-Meier method and the statistical significance of each comparison was assessed by the log-rank test. The association of ARC and CCI score with survival probabili- ty was estimated using the Cox regression modelling and expressed as death rate ratio (HR) and corresponding 95% confidence limits (95% CL). For each prognostic factor the Cox regression allows to obtain an HR value adjusted for the potential confounding effect of other variables entered the same equation. The likelihood ratio test was applied to evaluate the statistical significance of each prog- nostic variable included in the Cox model (11). In order to point out a non-linear dose-response relation- ship between ARC and death rates, potentially blurred by the categorization process, a three-knot restricted cubic spline (RCS) was fitted to survival data with the Cox regression equation. RCS is a flexible fitting procedure that allows the observed data to determine a smoothed func- tional form of dependency between the death rate (i.e., response) and a continuous prognostic variable (i.e., dose) (12). A two-tailed p-value < 0.05 was considered as sig- nificant. All analyses were performed using Stata (Stata Corp. Statistical Software, release 14. Statistical Software. College Station, TX: StataCorp LP, 2015). Table 1. Frequency distribution of patients’ baseline characteristics. Patients’ characteristics No. % Age (yrs) at radical cystectomy (median, range) 71, 46-87 - 46-65 66 19.8 66-70 68 20.4 71-76 66 19.8 77-80 67 20.0 81-87 67 20.0 Gender Male 288 86.2 Female 46 13.8 Raw Charlson comorbidity index ≤ 3 269 80.5 > 3 59 17.7 Missing 6 1.8 Tumor stage T0-T1 203 60.8 T2-T3 93 27.8 T4 38 11.4 Histotype Urothelial 263 78.7 Non urothelial/rare variant 71 21.3 Lymph node involvement N0 227 68.0 N1-N3 73 21.9 NX 34 10.2 Preoperative hemoglobin ≤ 10.0 g/dL 10 3.0 > 10.0 g/dL 302 90.4 Missing 22 6.6 Preoperative creatinine ≤ 1.2 mg/dL 247 74.0 > 1.2 mg/dL 65 19.5 Missing 22 6.6 Early complications No 289 86.5 Yes 45 13.5 Late complications No 289 86.5 Yes 45 13.5 Urinary diversion Ureterocutaneostomy/ileal conduit 203 60.8 Orthotopic neobladder 131 39.2 Days of hospitalization (median, range) 18, 2-80 - 2-13 88 26.4 14-17 73 21.9 18-23 94 28.1 24-80 79 23.7 Whole sample 334 100.0 RESULTS Baseline patients’ characteristics are listed in Table 1. The median follow-up time was 3.8 years (IQR = 1.3-7.5 years). During the study period, a total of 180 patients died (53.8%), 68 (62.2%) from all causes whereas the remaining 112 patients (37.8%) from BC. The median OS was 5.9 years (95%CL = 3.9- 8.1 years). 288 (86.2%) patients were male whereas 46 (13.8%) patients were female. Table 2 shows the results of uni- variate OS analysis. ARC, CCI, tumor stage, histotype, preopera- tive hemoglobin and creatinine levels, and urinary diversion showed a strong association with OS. A significant decreasing ten- dency in OS probabilities was found to be associated to an increasing ARC (p-value = 0.006). In addition, comorbid patients with CCI score > 3 showed a twofold increased death rate (p-value < 0.001) when com- pared to patients with lower CCI score (≤ 3). The cumulative effect on OS of ARC and CCI score was assessed in a multivariate context by mod- elling data using the Cox regres- sion. After adjusting for gender, tumor stage, lymph node involve- ment, preoperative Hb and Cr levels, complications, and urinary diversion the significant associa- tion of ARC and CCI score with OS was confirmed (Table 3, Model 1). Specifically, patients with CCI score > 3 showed a death rate excess of about 60% when compared to patients with CCI score ≤ 3 (HR = 1.61; 95%CL 17Archivio Italiano di Urologia e Andrologia 2021; 93, 1 Age above 70 years and Charlson Comorbidity Index higher than 3 are associated with reduced survival probabilities after radical cystectomy... Table 2. One-, three- and five-year overall survival probabilities (Pr) estimated through the Kaplan-Meier method according to levels of study prognostic factors. Patients’ Overall survival characteristics Follow-up One-year Three-year Five-year T D D% Med IQR Pr 95%CL Pr 95%CL Pr 95%CL P-value Age at radical cystectomy 46-65 66 26 39.4 6.8 2.4-10.0 0.89 0.79-0.94 0.71 0.58-0.80 0.66 0.53-0.76 0.006 66-70 68 34 50.0 4.1 1.6-8.3 0.84 0.72-0.90 0.62 0.48-0.71 0.55 0.41-0.65 71-76 66 38 57.6 3.5 0.9-6.0 0.70 0.57-0.79 0.54 0.41-0.65 0.47 0.34-0.58 77-80 67 40 59.7 3.3 1.0-6.5 0.75 0.62-0.83 0.58 0.45-0.68 0.47 0.34-0.59 81-87 67 42 62.7 2.3 0.9-6.2 0.72 0.59-0.80 0.47 0.34-0.58 0.43 0.30-0.54 Gender Male 288 155 53.8 3.7 1.3-7.5 0.78 0.72-0.82 0.58 0.51-0.63 0.52 0.46-0.57 0.882 Female 46 25 54.3 3.9 1.1-7.2 0.76 0.60-0.85 0.63 0.47-0.75 0.50 0.33-0.63 Raw Charlson comorbidity index ≤ 3 269 131 48.7 4.4 1.7-8.1 0.82 0.76-0.85 0.63 0.56-0.68 0.57 0.50-0.62 < 0.001 > 3 59 43 72.9 1.6 0.9-5.2 0.66 0.52-0.76 0.44 0.31-0.56 0.33 0.21-0.45 Missing 6 6 100.0 0.1 0.0-0.2 0.17 0.00-0.51 0.17 0.00-0.51 0.17 0.00-0.51 Tumor stage T0-T1 203 90 44.3 5.2 2.3-8.4 0.85 0.79-0.89 0.72 0.65-0.77 0.65 0.57-0.71 < 0.001 T2-T3 93 63 67.7 1.9 0.9-4.8 0.70 0.59-0.78 0.38 0.28-0.48 0.33 0.23-0.42 T4 38 27 71.1 1.3 0.4-4.2 0.58 0.40-0.71 0.34 0.19-0.48 0.28 0.14-0.42 Histotype Urothelial 263 137 52.1 4.0 1.5-8.3 0.81 0.75-0.84 0.62 0.55-0.67 0.53 0.46-0.59 0.032 Non urothelial/rare variant 71 43 60.6 2.2 0.5-6.3 0.68 0.55-0.77 0.46 0.33-0.57 0.46 0.33-0.57 Lymph node involvement N0 227 101 44.5 8.4 2.1-8.4 0.85 0.79-0.89 0.70 0.63-0.75 0.63 0.55-0.68 < 0.001 N1+N2 73 53 72.6 4.7 0.8-4.6 0.64 0.52-0.74 0.35 0.24-0.46 0.29 0.19-0.40 NX 34 26 76.5 5.2 0.4-5.2 0.59 0.40-0.73 0.32 0.17-0.48 0.29 0.15-0.44 Preoperative hemoglobin ≤ 10 10 8 80.0 0.4 0.2-1.0 0.40 0.12-0.67 0.20 0.03-0.47 0.20 0.03-0.47 > 10 302 156 51.7 3.9 1.4-7.4 0.80 0.75-0.84 0.61 0.55-0.66 0.54 0.48-0.59 0.002 Missing 22 16 72.7 1.5 0.2-8.2 0.59 0.36-0.76 0.41 0.20-0.60 0.32 0.14-0.51 Preoperative creatinine ≤ 1.2 247 118 47.8 4.4 1.7-7.9 0.83 0.77-0.87 0.65 0.58-0.70 0.58 0.51-0.63 < 0.001 > 1.2 65 46 70.8 1.9 0.6-5.6 0.65 0.51-0.74 0.40 0.27-0.51 0.36 0.24-0.48 Missing 22 16 72.7 1.5 0.2-8.2 0.59 0.36-0.76 0.41 0.20-0.60 0.32 0.14-0.51 Early complications No 289 151 52.2 4.0 0.7-0.8 0.81 0.75-0.85 0.61 0.55-0.66 0.54 0.47-0.59 0.009 Yes 45 29 64.4 1.7 0.4-0.6 0.58 0.42-0.70 0.42 0.27-0.55 0.40 0.25-0.53 Late complications No 312 167 53.5 3.9 1.3-7.5 0.79 0.73-0.82 0.60 0.54-0.64 0.53 0.46-0.58 0.241 Yes 22 13 59.1 2.1 0.9-4.5 0.68 0.44-0.83 0.41 0.20-0.60 0.41 0.20-0.60 Urinary diversion Ureterocutanostomy/ileal conduit 203 124 60.9 2.5 0.9-6.3 0.70 0.64-0.76 0.50 0.43-0.56 0.44 0.37-0.51 < 0.001 Orthotopic neobladder 131 56 42.7 5.2 2.2-8.9 0.89 0.83-0.94 0.72 0.64-0.79 0.65 0.56-0.72 Whole sample 334 180 53.9 3.8 1.3-7.5 0.78 0.73-0.82 0.58 0.53-0.64 0.52 0.46-0.57 . T: sample size; D/D%: number/percent of deceased patients; Med: median of follow-up time (years); IQR: inter-quartile range (min-max); 95%CL: 95% confidence limits for estimated Pr; P-value: probability level associated with the log-rank test. Table 3. Joint effect of age at radical cystectomy (ARC) and Charlson comorbidity index (CCI) on overall and bladder cancer- specific survival estimated through the Cox regression model. Regression model Patients’ characteristics Overall survival Cancer-specific survival Deaths = 164 (52.7%) Deaths = 101 (32.5%) HR 95%CL P-value HR 95%CL P-value 1 Age at radical cystectomy 0.034 0.750 46-65 1.00 (Ref.) 1.00 (Ref.) 66-70 1.92 1.06-3.47 1.57 0.77-3.16 71-76 2.25 1.28-3.94 1.41 0.69-2.84 77-80 2.08 1.19-3.64 1.44 0.73-2.84 81-87 2.03 1.16-3.57 1.41 0.70-2.81 Charlson comorbidity index 0.022 0.253 ≤ 3 1.00 (Ref.) 1.00 (Ref.) > 3 1.61 1.07-2.41 1.38 0.80-2.38 2 Age at radical cystectomy 0.155 0.431 46-69 1.00 (Ref.) 1.00 (Ref.) 70-80 1.39 0.97-1.99 1.09 0.70-1.71 81-87 1.04 0.59-1.83 0.68 0.31-1.48 Charlson comorbidity index 0.054 0.437 ≤ 3 1.00 (Ref.) 1.00 (Ref.) > 3 1.50 0.98-2.26 1.26 0.72-2.21 HR: death rate ratio adjusted for gender, tumor size, lymph nodes involvement, pre-operative hemoglobin and creatinine, early and late complications, urinary diversion; 95%CL: 95% confidence limits for HR; Ref.: reference category; P-value: probability level associated with the likelihood ratio test. Archivio Italiano di Urologia e Andrologia 2021; 93, 1 M. Maffezzini, V. Fontana, A. Pacchetti, F. Dotta, M. Cerasuolo, D. Chiappori, G. Guano, G. Mantica, C. Terrone 18 = 1.07-2.41; p-value = 0.022). An increase in death rates by ARC was observed although the estimated dose- response relationship appeared to be non-linear. In other words, assuming the death rate of the lowest ARC cate- gory (46-65 years) as a reference (HR = 1.00) the high- est rate was estimated in the intermediate category 71-76 years (HR = 2.25, 95%CL = 1.28-3.94). In order to point out better such a non-monotonic ten- dency, a Cox model was fitted to OS data after rearrang- ing ARC in three categories using 70 and 80 years as threshold values (Table 3, Model 2; Figure 1). In this case, the intermediate category (70-80 years) showed a death rate which was about 40% higher than those in the other two categories (46-69 and 81-87 years), even though a Figure 1. Joint effect of age at radical cystectomy (ARC) and raw Charlson comorbidity index (CCI) on life expectancy estimated through the Cox regression model. Overall survival probabilities are adjusted for gender, tumor size, lymph node involvement, pre-operative hemoglobin and creatinine, early and late complications, urinary diversion. Figure 2. Relationship between overall (OS) and cancer specific (CSS) death rate ratio (HR) and age at radical cystectomy (ARC) estimated using the Cox regression model adjusted for gender, tumor size, lymph node involvement, pre-operative hemoglobin and creatinine, early and late complications, urinary diversion. Smoothed HR point estimates were obtained by fitting three-knot cubic spline functions to OS and CSS data. 19Archivio Italiano di Urologia e Andrologia 2021; 93, 1 Age above 70 years and Charlson Comorbidity Index higher than 3 are associated with reduced survival probabilities after radical cystectomy... loss in statistical power was pointed out (p-value = 0.155). In order to assess the influence of death on the relation- ship between ARC and CCI on survival, all previous analyses were repeated using CS mortality as an out- come. After fitting a three-knot RCS (Figure 2) to both survival data series, two overlapping non-linear trends with common upmost risk values included in the ARC group 70-75 years were pointed out. DISCUSSION Multiple systemic diseases known as co-morbidities are diagnosed in the ageing men and women whom, in addi- tion, are also exposed to the risk of developing cancer (8- 9). Notably, the presence of multiple systemic diseases necessitating chronic treatment can also undermine the results of cancer treatment. The effect of ARC on the prognosis has been investigat- ed previously. The conclusion reported so far are hetero- geneous probably because, among other reasons, an objective threshold for dichotomizing this risk factor is not clearly identifiable and, accordingly, the issue remains controversial. Nevertheless, in several studies increased ARC showed an association with poor progno- sis and survival (13-16) and although RC is currently performed in advanced age subjects a systematic review of the literature has outlined a decline in both OS and CSS beginning at the age of 70 years (17). The association of comorbidities and survival after RC has also been investigated. The Adult Co-morbidity Evaluation 27 instrument (ACE-27) was used retrospec- tively in some studies (18-19) showing that co-morbidi- ty score and pathologic stage of disease significantly cor- related with reduced OS. Moreover, in the severe co- morbidity group the number needed to harm was 6, that is, for every 6 patients dying after RC 1 death occurred due to co-morbidity per se. The CCI score was used also and correlated with the out- comes of RC. Koppie et al. classified RC patients with the age-adjusted CCI (AA-CCI) into three groups, namely i) low AA-CCI score ≤ 2 , ii) moderate AA-CCI score of 3 to 5 , and iii) a high AA-CCI score > 5 (20). They found a median OS time of 6.3 years, 3.9 years, and 1.7 years in the low, moderate, and high score groups, respective- ly. Mayr et al. compared the ASA score, the ACE 27 instrument, the ECOG scale, and the AA-CCI (21). The Authors found that none of the comorbidity indices were significant predictors for CSS, whereas each index was a significant predictor for cancer-independent mortality. Importantly, based on ARC and comorbidity a weighed prognostic risk model was developed where after 3 years 47% of the patients within the high-risk group died of causes other than BC, compared with 8% of patients within the low-risk group. In a recently published study, D'Andrea et al. analysed a cohort of 46 patients with localized MIBC who were con- sidered unfit for RC or TT, and therefore sent to RT alone (22). Their survival outcomes were compared to an equal number of patients treated with RC. After per- forming a propensity score analysis CSS and OS of the patients undergoing RT were substantially the same as those who were treated with RC. The main findings of our study can be summarised as follows. First, after adjusting for known prognostic fac- tors (Table 1), patients in the age group 70-80 years showed the greater mortality risk. Second, after elimi- nating the competitive risk of death due to causes differ- ent from BC, the curves of OS and CSS show a parallel profile, as outlined by the RCS analysis (Figure 2). Finally, in patients with 70-80 years and CCI score > 3 the mortality risk is remarkably higher (HR = 2.09, 95%CL = 1.23-3.53) than all other groups (Figure 1). These findings reproduce closely what reported previ- ously by other authors (21). Certain limitations of our study are to be acknowledged. Firstly, inherent to the retrospective nature of the inves- tigation (study period: 2005-2015), although the comor- bidities were listed at the moment of patient enrolment, the CCI score was retrospectively assigned. Secondly, the CCI score has been commonly used to assess survival for various cancers since it offers a general patient evalua- tion. However, some particular conditions, i.e., hyper- tension, lung diseases in the absence of chronic obstruc- tive disease, coronary artery disease in the absence of myocardial infarction, etc., are overlooked. Thirdly, based on our dataset we were unable to address several concurrent factors that may have had an influence on survival such as a delay from symptoms-onset and diag- nosis/treatment, nutritional status, and post-operative hospitalization in intensive care units, among others. On the other hand, the strengths of our study are to be acknowledged as well. Two factors known to influence prognosis are addressed jointly, namely age and co-mor- bidities, and we were able to confirm their cumulative effect on survival after RC, as previously observed by other researchers. Moreover, the study cohort consists in series of consecutive subjects treated at a single referral, high volume center, offering a homogeneous treatment to all patients, that is, RC alone (with pelvic lym- phadenectomy), and no selection was made to exclude from treatment older and sicker patients. CONCLUSIONS In conclusion, in the present cohort analysis of BC patients consecutively submitted to RC we were able to identify age 70-80 years and higher CCI score (> 3) as significant factors limiting a better prognosis. Although in principle we believe that surgery should not be precluded in the presence of advanced age and co- morbidities both factors should be attentively considered when comparing outcomes after RC in contemporary series. Implementation of ad hoc trials, focusing on these patients, should be encouraged. ACKNOWLEDGMENTS The authors thank Peter De Ville, MA, for linguistic revision. REFERENCES 1. Jue JS, Koru-Sengul T, Miao F, et al. Neoadjuvant vs adjuvant chemotherapy for muscle-invasive bladder cancer: a propensity matched analysis. Minerva Urol Nefrol. 2020 Jan 30. Archivio Italiano di Urologia e Andrologia 2021; 93, 1 M. Maffezzini, V. Fontana, A. Pacchetti, F. Dotta, M. Cerasuolo, D. Chiappori, G. Guano, G. Mantica, C. Terrone 20 2. Capece M, Spirito L, La Rocca R, et al. Hexaminolevulinate blue light cystoscopy (Hal) assisted transurethral resection of the bladder tumour vs white light transurethral resection of the bladder tumour in non-muscle invasive bladder cancer (NMIBC): a retrospective analysis. Arch Ital Urol Androl. 2020; 92:17-20. 3. Mantica G, Simonato A, Du Plessis DE, et al. The pathologist's role in the detection of rare variants of bladder cancer and analysis of the impact on incidence and type detection. Minerva Urol Nefrol. 2018; 70:594-597. 4. Stein JP, Lieskovsky G, Cote R, et al. Radical cystectomy in the treatment of invasive bladder cancer: long-term results in 1,054 patients. J Clin Oncol. 2001; 19:666-75. 5. Madersbacher S, Hochreiter W, Burkhard F, et al. Radical cys- tectomy for bladder cancer today - a homogeneous series without neoadjuvant therapy. J Clin Oncol. 2003; 21:690-6. 6. Hautmann RE, Gschwend JE, de Petriconi RC, et al. Cystectomy for transitional cell carcinoma of the bladder: results of a surgery only series in the neobladder era. J Urol. 2006; 176:486-92 7. Dy GW, Gore JL, Forouzanfar MH, et al. Global burden of uro- logic cancers, 1990-2013. Eur Urol. 2017; 71:437-446. 8. Valderas JM, Starfield B, Sibbald B, et al. Defining comorbidity: implications for understanding health and health services. Ann Fam Med. 2009; 7:357-63. 9. Sgrò P, Sansone M, Sansone A, et al. Physical exercise, nutrition and hormones: three pillars to fight sarcopenia. Aging Male. 2019; 22:75-88. 10. Dindo D, Demartines N, Clavien PA. Classification of surgical complications: a new proposal with evaluation in a cohort of 6336 patients and results of a survey. Ann Surg. 2004; 240:205-13. 11. Marubini E, Valsecchi MG. Analyzing survival data from clin- ical trials and observational studies. John Wiley and Sons, New York, 1995. 12. Harrell FE Jr. Regression Modelling Strategies. 2nd Ed. New York, Springer Verlag, 2015. 13. Clark PE, Stein JP, Groshen SG, et al. Radical cystectomy in the elderly: comparison of clinical outcomes between younger and older patients. Cancer. 2005; 104:36-43. 14. Nielsen ME, Shariat SF, Karakiewicz PI, et al. Advanced age is associated with poorer bladder cancer-specific survival in patients treated with radical cystectomy. Eur Urol. 2007; 51:699-706. 15. Donat SM, Siegrist T, Cronin A, et al. Radical cystectomy in octogenarians--does morbidity outweigh the potential survival bene- fits? J Urol. 2010; 183:2171-7. 16. Chromecki TF, Mauermann J, Cha EK, et al. Multicenter vali- dation of the prognostic value of patient age in patients treated with radical cystectomy. World J Urol. 2012; 30:753-759. 17. Fonteyne V, Ost P, Bellmunt J, et al. Curative treatment for muscle invasive bladder cancer in elderly patients: a systematic review. Eur Urol. 2018; 73:40-50. 18. Megwalu II, Vlahiotis A, Radwan M, et al. Prognostic impact of comorbidity in patients with bladder cancer. Eur Urol. 2008; 53:581-9. 19. Fairey AS, Jacobsen NE, Chetner MP, et al. Associations between comorbidity, and overall survival and bladder cancer spe- cific survival after radical cystectomy: results from the Alberta Urology Institute Radical Cystectomy database. J Urol. 2009; 182:85-92. 20. Koppie TM, Serio AM, Vickers AJ, et al. Age-adjusted Charlson comorbidity score is associated with treatment decisions and clinical outcomes for patients undergoing radical cystectomy for bladder cancer. Cancer. 2008; 112:2384-92. 21. Mayr R, May M, Martini T, et al. Comorbidity and performance indices as predictors of cancer-independent mortality but not of can- cer-specific mortality after radical cystectomy for urothelial carci- noma of the bladder. Eur Urol. 2012; 62:662-70. 22. D'Andrea D, Soria F, Zehetmayer S, et al. Comparative effec- tiveness of radical cystectomy and radiotherapy without chemother- apy in frail patients with bladder cancer (published online ahead of print, Scand J Urol. 2020; 54:52-57. Correspondence Massimo Maffezzini, MD massimo.maffezzini@gmail.com Department of Urology, Ospedale Nuovo Legnano, Ospedale Fornaroli Magenta Via Papa Giovanni Paolo II, 20025 Legnano (Italy) Vincenzo Fontana, MD vincenzo.fontana@hsanmartino.it Clinical Epidemiology Unit, IRCCS Policlinico San Martino, University of Genova Largo Benzi 10, 16132 Genova (Italy) Andrea Pacchetti, MD a.pacchetti.90@gmail.com Federico Dotta, MD fededotta@hotmail.it Mattia Cerasuolo, MD mattiacerasuolo@hotmail.it Davide Chiappori, MD chiappori.davide@gmail.com Giovanni Guano, MD giovanni.guano@gmail.com Guglielmo Mantica, MD (Corresponding author) guglielmo.mantica@gmail.com Carlo Terrone, MD carlo.terrone@med.uniupo.it Department of Urology, IRCCS Policlinico San Martino, University of Genova Largo Benzi 10, 16132 Genova (Italy)