The Predictors of Local Recurrence after Radical Cystectomy in Patients with Invasive Bladder Cancer Hiroki Ide, Eiji Kikuchi, Akira Miyajima, Ken Nakagawa, Takashi Ohigashi, Jun Nakashima and Mototsugu Oya Department of Urology, Keio University School of Medicine, Tokyo, Japan Received March 2, 2008; accepted April 10, 2008 Jpn J Clin Oncol 2008;38(5)360 364 doi:10.1093/jjco/hyn036 Objectives: To examine the association between local recurrence and distant metastasis or disease-specific survival and identify independent factor predictors for local recurrence. Methods: We identified a study population of 146 consecutive patients treated surgically for invasive bladder cancer at our institution between 1987 and 2003. We clarified the relationship among local recurrence, distant metastasis and disease-specific survival and identified significant predictors for local recurrence. Results: Local recurrence developed in 26 (17.8%) of the 146 patients at a median of 10 months (range, 1 73 months) after cystectomy. It was independently associated with distant metastasis in addition to the number of retrieved lymph nodes. The 2- and 5-year metastasisfree rates were 86.7 and 76.5% in patients without local recurrence and 26.5 and 0% in those with local recurrence (P, 0.001), respectively. The presence or absence of local recurrence and tumor grade were independent predictors of disease-specific survival. The 2- and 5-year disease-specific survival rates were 93.5 and 88.3% in patients without local recurrence and 55.1 and 35.4% in those with local recurrence (P, 0.001). The presence of concomitant adenocarcinoma component, pathological nodal involvement and the number of retrieved lymph nodes were independent predictors of local recurrence. Conclusions: Local recurrence was independently associated with distant metastasis and disease-specific survival. Patients who have the predictive factors described above may benefit from integrated surgical therapies with post-operative adjuvant chemotherapy. Key words: bladder cancer local recurrence distant metastasis INTRODUCTION Although radical cystectomy has been considered the gold standard curative treatment for invasive bladder cancer, unfortunately around one-third of the patients relapse and die of the disease after surgery. According to previous studies, the 5-year survival rates were 76 85% for pt1pn0, 64 84% for pt2pn0, 25 56% for pt3pn0 and 19 44% for pt4pn0 (1,2). The main reason for cancer death in patients treated with radical cystectomy has been associated with the development of distant metastasis. Meanwhile, long-term survival after local recurrence is also extremely rare. Thus, local recurrence is thought to be as important as distant metastasis for disease-specific survival. However, few studies have discussed the clinical course and natural history of local recurrence in patients with bladder cancer. The association between local recurrence and distant metastasis For reprints and all correspondence: Eiji Kikuchi, Department of Urology, Keio University School of Medicine, 35 Shinanomachi, Shinjuku-ku, Tokyo 160-8582, Japan. E-mail: eiji-k@kb3.so-net.ne.jp has also not been fully evaluated. To determine the significance of local recurrence on distant metastasis or diseasespecific survival, understand the clinical characteristics of local recurrence and identify risk factors for predicting, it might provide clues for improving the treatment of invasive bladder cancer. In this study, we retrospectively reviewed patients who underwent radical cystectomy for invasive bladder cancer in order to analyze the clinical course according to patterns of local recurrence and distant metastasis. Furthermore, we examined the independent association between local recurrence and distant metastasis or disease-specific survival and identified significant factors for predicting local recurrence. PATIENTS AND METHODS A total of 150 consecutive patients underwent radical cystectomy with pelvic lymph node dissection for recurrent or invasive bladder cancer at our institution between 1987 # The Author (2008). Published by Oxford University Press. All rights reserved.
Jpn J Clin Oncol 2008;38(5) 361 and 2003. Exclusion criteria were apparent regional lymph node involvement and distant metastasis at the time of diagnosis, the presence of concomitant upper urinary tract tumor or incomplete data. We identified a study population of 146 consecutive patients who had complete data based on pathological features. Mean patient age was 67 years (range, 36 87 years). Simultaneous urethrectomy was performed in 72 of these 146 patients. Eighty-eight patients received ileal conduit, 37 patients continent ileal reservoir and 21 patients orthotopic neobladder. Cisplatin-based neoadjuvant and adjuvant chemotherapy regimens were administered to 8 (5.5%) and 17 (11.6%) patients, respectively. Patients were followed post-operatively with urinary cytology, and laboratory studies every 3 months for 2 years and then every 6 months. Chest X-rays and CT scans and/or MRI and/or excretory urograms were performed every 6 months until 5 years, and then annually. The median follow-up period was 42 months (range, 1 220). The tumor was staged and graded according to the criteria of the third edition of General Rules for Clinical and Pathological Studies on Bladder Cancer of the Japanese Urological Association and Japanese Society of Pathology (3). Local recurrence was defined as recurrence in the pelvic soft tissue or pelvic lymph nodes detected with imaging studies. Involvement of lymph nodes above the level of the iliac bifurcation and inguinal lymph nodes was classified as distant metastasis. The associations between the occurrence of local recurrence or distant metastasis and clinical or pathologic features were assessed with the chi-square test for trends. The actuarial probabilities obtained using Kaplan Meier analysis are reported as the median + 2 standard errors (95% CI, confidence intervals) and were compared using the log-rank test. Prognostic factors assessed were age (greater versus less than the patients mean age of 65 years), gender, tumor grade (G1/2 versus G3), pathological T (pt) stage (pta-t2 versus pt3/4), nodal involvement, lymphovascular invasion (LVI), number of retrieved nodes, concomitant adenocarcinoma (AC) component and concomitant squamous cell carcinoma (SCC) component. Cox proportional hazard regression analysis was used to assess the prognostic indicators. The endpoints of the univariate and multivariate analyses were local recurrence, distant metastasis and death from bladder cancer. P value of,0.05 was considered to indicate statistical significance. These analyses were performed with the SPSS w, version 11.0, statistical software package. RESULTS The patient characteristics are summarized in Table 1. The percentage of tumor Grade 3 increased as the pathological stage increased (48.1, 88.6, 94.1 and 95.7% for T1, T2, T3 and T4, respectively). LVI was present in 42.5% (62 of 146) overall, involving 38.1% (51 of 134) and 91.7% (11of 12) of node-negative and node-positive patients, respectively. Neoadjuvant chemotherapy was administered to eight patients (5.5%), most of whom had clinical stage T3 or greater and/or nodal involvement. Seventeen (11.6%) patients with pt3 or more and/or nodal involvement received adjuvant chemotherapy according to the judgment of the urologist and the performance status of the patient. The Table 1. Patient characteristics Characteristic Number of patients Gender Male 109 Female 37 Age Median 68 Pathological T classification pt1 50 pt2 36 pt3 35 pt4 25 Grade G1 1 G2 34 G3 111 Nodal involvement Yes 12 No 134 Number of retrieved lymph nodes Median 8,8 92 8 46 Unknown 8 LVI Yes 62 No 40 Unknown 44 Perioperative chemotherapy Neoadjuvant 8 Adjuvant 17 Histology Pure UC 111 Pure SCC 1 UC þ SCC 19 UC þ AC 10 UC þ SCC þ AC 5 LVI, lymphovascular invasion; UC, urothelial carcinoma; SCC, squamous cell carcinoma; AC, adenocarcinoma; pt, pathological T.
362 The predictor of recurrence after cystectomy histological type was pure urothelial carcinoma in 111 patients (76.0%), pure SCC in one patient (0.7%) and urothelial carcinoma associated with other histological components in 34 patients (23.3%). Among the other concomitant histological components, SCC component alone, AC component alone and both SCC and AC components were found in 19, 10 and 5 patients, respectively (Table 1). Local recurrence developed in 26 (17.8%) of the 146 patients at a median of 10 months (range, 1 73) after cystectomy. Of these, seven had local recurrence without distant metastasis, and 10 had local recurrence preceding distant metastasis for follow-up time. Local recurrence with synchronous distant metastasis was found in seven patients. Distant metastasis preceding local recurrence was observed in two patients. Distant metastasis without local recurrence developed in 24 patients (16.4%) at a median of 10 months (range, 1 67) after total cystectomy. The disease recurred in the upper urinary tract and urethra in eight and two patients, respectively. pt stage, grade, LVI, nodal involvement, occurrence of local recurrence, presence of concomitant SCC and AC component were significantly associated with distant metastasis in univariate analysis. Multivariate analysis revealed that occurrence of local recurrence and the number of retrieved nodes were independent predictors of distant metastasis (Table 2). Analyses were rerun after exclusion of patients who were treated with neoadjuvant chemotherapy or had AC component and SCC component, and this resulted in consistent patterns and P values. The 2- and 5-year metastasis-free rates were 86.7 and 76.5% in patients without local recurrence and 26.5 and 0% in those with local recurrence (P, 0.001). Age, pt stage, LVI, nodal involvement and occurrence of local recurrence were significantly associated with disease-specific survival in univariate analysis. In multivariate analysis, tumor grade and occurrence of local recurrence were independently associated with disease-specific survival (Table 2). Analyses were rerun after exclusion of patients who were treated with neoadjuvant chemotherapy or had AC component and SCC component, and this resulted in consistent patterns and P values. Kaplan Meier analysis demonstrated that the local recurrence preceding distant metastasis (Pre Rec) group, local recurrence with synchronous distant metastasis (Syn Meta) group and distant metastasis preceding local recurrence (Pre Meta) group were each at significantly increased risk for disease-specific survival compared with neither local recurrence nor distant metastasis (Rec-Meta-) group (each; P, 0.001). Meanwhile, there were no significant differences in survival among these groups with recurrence and/or distant metastasis (Fig. 1). The 2- and 5-year disease-specific survival rates were 100 and 100% in Rec-Meta- group, 54.3 and 17.1% in Pre Rec group, 75.3 and 46.8% in Pre Meta group, and 68.6 and 45.7% in Syn Meta group. The association between patient characteristics and occurrence of local recurrence and/or distant metastasis was assessed with the chi-square test for trends. There was no significant difference for any characteristic, with the exception of gender. The percentage of females to males was greater only in Syn Meta group. Predictors for local recurrence are shown in Table 3. pt stage, LVI, nodal involvement and presence of concomitant AC component were significantly associated with local recurrence in univariate analysis. Furthermore, the presence of concomitant AC component, nodal involvement and the number of retrieved nodes were independent predictors of local recurrence in multivariate analysis. Table 2. Univariate and multivariate analysis of parameters predicting distant metastasis and disease-specific survival Parameter Distant metastasis Disease-specific survival Univariate Multivariate Univariate Multivariate P value HR 95% CI P value P value HR 95% CI P value Age (,70 versus 70) 0.115 0.037 Gender (male versus female) 0.261 0.493 Pathological T stage (T2 versus.t2) 0.001,0.001 Grade (G1 G2 versus G3) 0.043 0.224 5.495 1.019 29.412 0.047 LVI (negative versus positive) 0.008 0.019 Nodal involvement (presence or absence) 0.005 0.017 Number of retrieved lymph nodes (,8 versus 8) 0.335 4.545 1.093 26.316 0.005 0.438 Concomitant SCC (without versus with) 0.003 0.152 Concomitant AC (without versus with),0.001 0.189 Local recurrence (presence or absence),0.001 4.773 1.635 13.933 0.004,0.001 10.059 2.642 41.093 0.001 HR, hazard ratio; CI, confidence interval.
Jpn J Clin Oncol 2008;38(5) 363 Figure 1. Disease-specific survival rates according to local recurrence and distant metastasis. Rec-Meta-, neither local recurrence nor distant metastasis; Pre Meta, distant metastasis preceding local recurrence; Pre Rec, local recurrence preceding distant metastasis; Syn Meta, local recurrence with synchronous distant metastasis. The results of log rank test. Rec-Meta- versus Pre Meta P, 0.001, Pre Meta versus Pre Rec P ¼ 0.370; Rec-Meta- versus Pre Rec P, 0.001, Pre Meta versus Syn Meta P ¼ 0.393; Rec-Meta- versus Syn Meta P, 0.001, Pre Rec versus Syn Meta P ¼ 0.669. Table 3. Univariate and multivariate analysis of parameters predicting local recurrence Parameter Univariate Multivariate P value HR 95% CI P value Age (,70 versus 70) 0.218 Gender (male versus female) 0.750 Pathological stage (T2,0.001 versus.t2) Grade (G1 G2 versus G3) 0.228 LVI (negative versus positive),0.001 Nodal involvement (negative 0.005 7.879 1.636 37.938 0.010 versus positive) Number of retrieved lymph 0.388 5.376 1.093 26.316 0.039 nodes (,8 versus 8) Concomitant SCC (without 0.155 versus with) Concomitant AC (without versus with),0.001 12.422 2.238 68.958 0.004 DISCUSSION Owing to frequent post-treatment imaging study after radical cystectomy, recent studies have demonstrated that local recurrence is much more pronounced than previously believed and have reported that the rate of local recurrence after cystectomy ranged between 5.0 and 18.6% (2,4 7). In the present study, 26 (17.8%) of the 146 patients who underwent radical cystectomy had local recurrence. Honma et al. (4) reported that local recurrence developed in 27 (18.6%) of the 145 patients and 8 (5.5%) had local recurrence alone and 19 (13.1%) had concurrent distant metastasis. These results are almost similar to ours. Our study revealed that local recurrence is an independent predictor of distant metastasis. Pollack et al. (5) have also reported that local recurrence, in addition to pt stage and nodal involvement, is an independent predictor of distant metastasis. These findings indicate a close association between local recurrence and distant metastasis. We have demonstrated that local recurrence is also independently associated with disease-specific survival. In the present study, the presence of concomitant AC component, nodal involvement and the number of retrieved nodes were independent predictors of local recurrence in multivariate analysis. It is apparent from some reports that primary AC of the bladder has an aggressive behavior, poor prognosis and spreads locoregionally. According to one report, local recurrence rate was 31% in patients with primary bladder AC who were followed-up for.6 months after surgery (8). The other studies also reported that the prognosis of patients with bladder AC was poor and 5-year overall survival rate is 11 55% for patients with bladder AC (8,9). These studies may in part suggest that the existence of an AC component in a bladder specimen would have a great impact on the development of local recurrence. According to our study, pathological nodal involvement was a predictor of local recurrence. Some studies have also reported that pelvic node involvement is an important predictor of disease-free survival (10 12). Herr et al. (7) reported that 5-year local recurrence rates were 12 and 29% in patients without and with nodal involvement, respectively. In this study, the number of retrieved nodes (,8) was also an independent predictor. Recently, the importance of pelvic lymph node dissection during radical cystectomy and the prognostic significance of the number of retrieved nodes have been emphasized. Herr et al. (7) demonstrated that the number of retrieved nodes (,10) was a poor surgical predictor of survival and 5-year disease-specific survival rates were 61 and 44% in patients having 10 or more nodes removed and fewer than 10, respectively (P ¼ 0.007). Stein et al. also reported 10-year recurrence-free survival rates of 25 and 36% in patients with 15 or less nodes removed and.15 nodes removed, respectively. Furthermore, they suggested that lymph node density was a significant prognostic factor and reported 10-year recurrence-free survival rates of 43 and 17% in patients with a lymph node density of 20% or less and greater than 20%, respectively (P, 0.001) (12). In the present study, local recurrence occurred in 3 of 10 (30%) and 2 of 2 (100%) patients with nodal involvement having eight or more and fewer than eight nodes removed, respectively. The prognostic information for predicting local recurrence after cystectomy might help us select a subgroup of patients who would benefit from optimal local control, subsequently reducing distant metastasis and improving disease-specific survival. There are some strategies for local control such as aggressive surgical extirpation of tissue adjacent to bladder, extended lymph node dissection and adjuvant therapy that includes a new chemotherapeutic regimen or novel
364 The predictor of recurrence after cystectomy molecular targeting therapy in the future. On the basis of our results, extensive lymph node dissection would be of benefit to reduce local recurrence. CONCLUSION In our population of 146 patients who underwent radical cystectomy, local recurrence developed in 26 (17.8%). Local recurrence was independently associated with distant metastasis and disease-specific survival. The presence of concomitant AC component, pathological nodal involvement and the number of retrieved nodes were independent predictors of local recurrence. These results would be helpful for selecting a subgroup of patients who may benefit from appropriate local control such as adjuvant therapy that includes a new chemotherapeutic regimen or novel molecular targeting therapy in the future. On the basis of our results, extensive lymph node dissection would be of benefit to reduce local recurrence. Conflict of interest statement None declared. References 1. Freiha F, Reese J, Torti FM. A randomized trial of radical cystectomy versus radical cystectomy plus cisplatin, vinblastine and methotrexate chemotherapy for muscle invasive bladder cancer. J Urol 1996;155:495 9; discussion 499 500. 2. Stein JP, Lieskovsky G, Cote R, Groshen S, Feng AC, Boyd S, 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. 3. Japanese Urological Association, Japanese Society of Pathology. General Rules for Clinical and Pathological Studies on Bladder Cancer. 3rd edn. Tokyo: Kanehara 2001 (in Japanese). 4. Honma I, Masumori N, Sato E, Takayanagi A, Takahashi A, Itoh N, et al. Local recurrence after radical cystectomy for invasive bladder cancer: an analysis of predictive factors. Urology 2004;64: 744 8. 5. Pollack A, Zagars GK, Cole CJ, Dinney CP, Swanson DA, Grossman HB. The relationship of local control to distant metastasis in muscle invasive bladder cancer. JUrol1995;154:2059 63; discussion 2063 4. 6. Greven KM, Spera JA, Solin LJ, Morgan T, Hanks GE. Local recurrence after cystectomy alone for bladder carcinoma. Cancer 1992;69:2767 70. 7. Herr HW, Bochner BH, Dalbagni G, Donat SM, Reuter VE, Bajorin DF. Impact of the number of lymph nodes retrieved on outcome in patients with muscle invasive bladder cancer. JUrol2002;167:1295 8. 8. el-mekresh MM, el-baz MA, Abol-Enein H, Ghoneim MA. Primary adenocarcinoma of the urinary bladder: a report of 185 cases. Br J Urol 1998;82:206 12. 9. Wright JL, Porter MP, Li CI, Lange PH, Lin DW. Differences in survival among patients with urachal and nonurachal adenocarcinomas of the bladder. Cancer 2006;107:721 8. 10. Bassi P, Ferrante GD, Piazza N, Spinadin R, Carando R, Pappagallo G, et al. Prognostic factors of outcome after radical cystectomy for bladder cancer: a retrospective study of a homogeneous patient cohort. J Urol 1999;161:1494 7. 11. Dalbagni G, Genega E, Hashibe M, Zhang ZF, Russo P, Herr H, et al. Cystectomy for bladder cancer: a contemporary series. J Urol 2001;165:1111 6. 12. Stein JP, Cai J, Groshen S, Skinner DG. Risk factors for patients with pelvic lymph node metastases following radical cystectomy with en bloc pelvic lymphadenectomy: concept of lymph node density. J Urol 2003;170:35 41.