Intravesical Gemcitabine for High Risk, Nonmuscle Invasive Bladder Cancer after Bacillus Calmette-Guerin Treatment Failure Itay A. Sternberg, Guido Dalbagni,* Ling Y. Chen, Sherri M. Donat, Bernard H. Bochner and Harry W. Herr From the Urology Service, Department of Surgery (IAS, GD, SMD, BHB, HWH) and Department of Epidemiology and Biostatistics (LYC), Memorial Sloan-Kettering Cancer Center, New York, New York Abbreviations and Acronyms BCG ¼ bacillus Calmette-Guerin CIS ¼ carcinoma in situ CR ¼ complete response CSS ¼ cancer specific survival MIBC ¼ muscle invasive bladder cancer NMIBC ¼ nonmibc NR ¼ no response OS ¼ overall survival PFS ¼ progression-free survival PR ¼ partial response RFS ¼ recurrence-free survival Accepted for publication April 29, 2013. Study received institutional review board approval. Supported by the Sidney Kimmel Center for Prostate and Urologic Cancers. * Correspondence: Memorial Sloan-Kettering Cancer Center, 1275 York Ave., New York, New York 10065 (telephone: 646-422-4394; FAX: 212-988-0760; e-mail: dalbagng@mskcc.org). For another article on a related topic see page 1912. Purpose: We report our experience with intravesical gemcitabine for bladder cancer after failed bacillus Calmette-Guerin treatment. Materials and Methods: We retrospectively reviewed the records of patients at our cancer center treated with intravesical gemcitabine after bacillus Calmette- Guerin failure. We estimated progression-free, recurrence-free and cancer specific survival using the cumulative incidence function, considering death from another cause as a competing risk. Comparisons were made using the Gray test. Overall survival was estimated using the Kaplan-Meier method and differences were compared with the log rank test. Results: Of 69 patients treated with intravesical gemcitabine 37 had bacillus Calmette-Guerin refractory disease. Median followup in progression-free patients was 3.3 years. Progression-free and cancer specific survival were similar in patients with refractory disease and those with other types of bacillus Calmette- Guerin failure. Overall survival was lower in patients with refractory disease (58% vs 71%) but this was not statistically significant (p ¼ 0.096). Of the patients 27 patients experienced a complete response. Progression-free, cancer specific and overall survival did not differ significantly between patients with and without a complete response. Cystectomy was subsequently performed in 20 patients. Those with a complete response had a delayed time to cystectomy and no muscle invasive bladder cancer at cystectomy. There were no serious adverse events and only a minority of patients discontinued treatment due to adverse events. Conclusions: In our experience intravesical gemcitabine should be considered after bacillus Calmette-Guerin failure in patients with bladder cancer who refuse radical cystectomy or who are not candidates for major surgery. Key Words: urinary bladder, urinary bladder neoplasms, BCG vaccine, treatment failure, gemcitabine BCG is considered the most effective intravesical agent for NMIBC and it is recommended for patients at high risk for progression. 1 BCG decreases the risk of recurrence, the development of distant metastasis, and the risk of death and bladder cancer related death compared to intravesical chemotherapy. 2 Failure to achieve a CR after an induction course of BCG or recurrence after BCG treatment is associated with an increased risk of disease progression and poor prognosis. 3,4 Standard treatment after BCG failure is radical cystectomy. Patients 1686 j www.jurology.com 0022-5347/13/1905-1686/0 THE JOURNAL OF UROLOGY 2013 by AMERICAN UROLOGICAL ASSOCIATION EDUCATION AND RESEARCH,INC. http://dx.doi.org/10.1016/j.juro.2013.04.120 Vol. 190, 1686-1691, November 2013 Printed in U.S.A.
GEMCITABINE FOR BLADDER CANCER AFTER BACILLUS CALMETTE-GU ERIN FAILURE 1687 who are not candidates for or who are unwilling to undergo cystectomy have limited options. 1 Intravesical valrubicin was approved by the United States Food and Drug Administration for BCG refractory bladder CIS. Several other intravesical therapies were also studied in patients with NMIBC who experience BCG failure, including immunotherapy, chemotherapy and thermochemotherapy. 5,6 Gemcitabine is a deoxycytidine with a broadspectrum antitumor effect. It is used in neoadjuvant and adjuvant settings in combination with other chemotherapeutic drugs for MIBC, and for advanced and metastatic tumors. 7 12 We previously reported the results of phase I and II studies of intravesical gemcitabine in patients after BCG failure. 13,14 In the current study we report our experience with intravesical gemcitabine after BCG failure. We explored the effect of the type of BCG failure on the patient response to gemcitabine and compared patient outcomes by the response to gemcitabine treatment. PATIENTS AND METHODS After obtaining institutional review board approval, we retrospectively reviewed the charts of patients treated with intravesical gemcitabine for NMIBC at our institution between January 1999 and October 2011. We identified 69 patients with NMIBC who were treated with intravesical gemcitabine after BCG treatment failed. BCG Failure A patient was considered to have BCG refractory disease when there was failure to achieve a disease-free state 6 months after initial BCG therapy with maintenance or retreatment at 3 months due to a persistent or rapidly growing recurrent tumor. BCG resistant disease was defined as recurrence 3 months after an induction cycle. BCG relapsing disease was defined as disease recurrence after the patient was disease-free for 6 months. BCG intolerant disease was defined as recurrence after administering a less than adequate course of therapy due to a serious adverse event or symptomatic intolerance requiring the discontinuation of BCG therapy. 15 Intravesical Gemcitabine Treatment Patients received 2 courses of intravesical gemcitabine twice weekly for 3 weeks with courses separated by a week of rest for a total of 12 instillations. Intravesical gemcitabine was given at a dose of 2,000 mg instilled in 100 ml saline. Patients were instructed to hold the instillation for 60 minutes. Responses to intravesical gemcitabine treatment were divided into 3 categories, including 1) CR when no tumor was seen 3 months after treatment and the patient had negative cytology results, 2) PR when no tumor was seen at 3 months but the patient had positive cytology results, and 3) NR when there was a viable tumor 3 months after treatment. Toxicity was assessed after each treatment and reported using National Cancer Institute (NCI) Common Toxicity Criteria (CTC), version 2.0. Statistical Analysis Separate analyses were done for OS, PFS, RFS and CSS. Progression was defined as progression to MIBC or the appearance of metastasis. Recurrence was defined as any recurrent bladder cancer. Kaplan-Meier curves were calculated to determine the difference in OS between patients with BCG refractory disease vs other types of BCG failure and compared with the log rank test. PFS, RFS and CSS were estimated using the cumulative incidence function, considering death from another cause as a competing risk, with comparisons made using the Gray test. The HR and 95% CI were estimated using a Cox proportional hazard model for OS and competing risk analysis for PFS, RFS and CSS. Survival was calculated from the date of BCG failure. We also analyzed PFS, CSS and OS by the response to gemcitabine therapy (CR vs PR and NR). Separate analyses were done for PFS, CSS and OS using the Kaplan-Meier method and the cumulative incidence function, as described. Landmark analysis was performed using a landmark time of 5 months after the date of gemcitabine initiation to ascertain the response to gemcitabine. Patients were evaluated 1 to 3 weeks after completing treatment. Given the absence of a clinically significant landmark, we chose 5 months as the landmark since this was the time point that excluded the fewest patients. Two patients who responded after the landmark as well as 3 lost to followup, 1 with progression and 1 who died before the landmark were excluded from PFS analysis. For CSS and OS analysis 2 patients who responded after the landmark and 3 lost to followup before the landmark were excluded. Statistical analysis was done using StataÒ 12 and R (R Foundation for Statistical Computing, Vienna, Austria) with the cmprsk package. RESULTS Between 1999 and 2011 at our institution 69 patients were treated with intravesical gemcitabine for NMIBC after BCG treatment failed. All patients were offered cystectomy, of whom 5 were not candidates for major surgery, 63 were interested in bladder sparing alternatives and 1 was treated after cystectomy was abandoned due to concern over injuring a single pelvic kidney. Table 1 lists study cohort descriptive characteristics. Therapy was done in 37 (54%) of the 69 patients for BCG refractory disease, in 5 (7%) for BCG resistant disease, in 20 (29%) for BCG recurrent disease and in 5 (7%) for BCG intolerant disease. The reason for BCG failure could not be determined in 2 patients due to insufficient data. There was no great difference between patients with BCG refractory disease and those with another type of BCG failure. However, patients with refractory disease were more likely to have had CIS before BCG (41% vs 29%). At last followup 11 patients had progressed, 10 had died of bladder cancer and 13 had died of another cause. Median followup was 3.3 years in
1688 GEMCITABINE FOR BLADDER CANCER AFTER BACILLUS CALMETTE-GU ERIN FAILURE Table 1. Patient characteristics BCG Refractory Other BCG Failure* No. pts 37 32 Median age at BCG failure (IQR) 71 (63e75) 73 (63e77) No. male (%) 27 (73) 26 (81) No. stage before BCG treatment (%): Ta 19 (51) 17 (55) CIS 15 (41) 9 (29) T1 3 (8) 5 (16) No. high grade before BCG treatment (%) 34 (92) 24 (86) No. stage at BCG failure (%): Ta 7 (19) 10 (33) Ta þ CIS 7 (19) 1 (3) T1 1 (3) 2 (7) T1 þ CIS 2 (5) 3 (10) CIS 20 (54) 14 (47) No. high grade at BCG failure (%) 37 (100) 28 (88) No. reason for deferring cystectomy (%): Patient unwilling 34 (92) 29 (91) Medically unfit 3 (8) 2 (6) Other 0 1 (3) Median No. recurrences after 1 (0e2) 1 (0e3) gemcitabine/pt (IQR) No. further intravesical treatment after 13 (35) 12 (38) gemcitabine (%) No. radical cystectomy (%) 12 (32) 8 (25) * BCG resistant, recurrent, intolerant or unknown due to insufficient data. progression-free patients and 3.6 years in diseasefree patients. Five-year OS was lower in patients with BCG refractory disease than in patients with another type of BCG failure but this was not statistically significant (58% vs 71%, log rank test p ¼ 0.096). Patients with BCG refractory disease were at twice the risk of death from any cause but this was not statistically significant (HR 2.05, 95% CI 0.87e4.84, p ¼ 0.1). Figure 1 shows the OS Kaplan- Meier curve. PFS, CSS and RFS did not differ significantly between patients with BCG refractory disease and those with another type of BCG failure (p ¼ 0.9, 0.7 and 0.2, respectively). The cumulative 5-year incidence of progression was 19% for BCG refractory disease and 22% for other types of BCG failure (HR 1.09, 95% CI 0.34e3.50). A similar cumulative incidence estimate with a wide CI was observed for CSS (HR 1.41, 95% CI 0.41e4.81). Survival analysis in patients with CIS at BCG failure showed results similar to those of the entire cohort. There was no significant difference in PFS, CSS or OS between patients with BCG refractory CIS and patients with other types of BCG failure (log rank test p ¼ 0.2, 0.3 and 0.2, respectively). Figure 2 shows the CSS Kaplan-Meier curve. Median time from BCG failure to the initiation of intravesical gemcitabine therapy was 8.1 weeks. The full treatment course was completed as planned in 61 patients (88%), while treatment was stopped prematurely in only 8 due to adverse events. After gemcitabine treatment, 27 patients had CR, 19 had PR and 20 had NR. Three patients could not be evaluated for a response to treatment due to insufficient data. The 5-year cumulative incidence of death from bladder cancer was 12% (95% CI 2e31) and 18% (95% CI 7e34) in patients with and without a CR, respectively. Figure 3 shows cumulative incidence curves. There was no statistically significant difference in PFS, CSS or OS between patients with vs without a CR (p ¼ 0.3, 0.3 and 0.8, respectively). Figure 4 shows the cumulative incidence of recurrent bladder cancer. Of the 46 patients who experienced recurrence after intravesical gemcitabine treatment 18 underwent cystectomy. Another 2 patients underwent cystectomy after intravesical gemcitabine was prematurely discontinued due to adverse events. Of these 20 patients 18 initially refused cystectomy to Survival probability 0.00 0.25 0.50 0.75 1.00 Cancer specific survival probability 0.00 0.25 0.50 0.75 1.00 0 2 4 6 8 10 0 2 4 6 8 Time from BCG failure (yrs) Time from BCG failure (yrs) Number at risk Other types of BCG failure 30 27 18 12 8 4 BCG refractory 37 27 14 8 8 3 Number at risk Other types of BCG failure 18 15 11 7 4 BCG refractory 29 21 10 5 5 Figure 1. Kaplan-Meier curve of OS by type of BCG failure. Solid curve indicates BCG refractory disease. Dashed curve indicates other type. Figure 2. Kaplan-Meier curve of CSS in 47 patients with CIS by type of BCG failure. Solid curve indicates BCG refractory disease. Dashed curve indicates other type.
GEMCITABINE FOR BLADDER CANCER AFTER BACILLUS CALMETTE-GU ERIN FAILURE 1689 Cumulative Incidence 0.2.4.6 0 1 2 3 4 5 6 7 Time from 5 months post gemcitabine initiation (years) Figure 3. Cumulative incidence curves of death from bladder cancer by response to intravesical gemcitabine. Solid curve indicates PR plus NR. Dashed curve indicates CR. pursue bladder preservation but they elected cystectomy after the condition progressed or recurred to high grade disease. Two patients were initially evaluated to be unfit for major surgery. One underwent aortic valve replacement and coronary artery bypass grafting in the interim, and his risk decreased. The other patient was evaluated again after gemcitabine failed and was found to be at moderate risk for surgery. The remaining 28 patients continued to refuse cystectomy or were not candidates for major surgery. Of these patients 19 were treated with additional intravesical agents (further intravesical gemcitabine in 5, BCG with or without interferon in 13 and intravesical mitomycin C in 3 with more than 1 agent in some patients), while for 9 patients no further treatment was documented in the medical records. Of the 20 patients who underwent cystectomy the surgery was done in 17 for NMIBC and in 3 for progression to MIBC. Ten of these patients had a CR to intravesical gemcitabine and 10 did not. Median time from intravesical gemcitabine therapy to cystectomy was 8.5 months in the entire cohort, including 20.9 and 5.3 months in patients with and without a CR, respectively. No patient with a CR had MIBC at cystectomy. In 1 patient treated with cystectomy after a CR metastasis developed 28 months after cystectomy. The patient died of bladder cancer. Patients generally tolerated intravesical gemcitabine well. According to Common Terminology Criteria for Adverse Events (CTCAE), there were no life threatening events or treatment related deaths (grade 4 or 5) (table 2). Adverse events were reported in 49 patients (71%) and most were mild or moderate (CTCAE grade 1 or 2). The most common mild and moderate adverse events reported were urinary symptoms in 25 cases and fatigue in 24. Of the patients 12% experienced moderately severe adverse events (CTCAE grade 3) that required discontinuation of therapy, including urinary symptoms in 3, infection in 2 and skin rash, deep vein thrombosis and thrombocytopenia in 1 each. DISCUSSION The safety and efficacy of intravesical gemcitabine therapy were previously demonstrated in phase I and II studies with a CR rate of 23% to 56%. 13,14,16e20 Most patients enrolled in these studies were previously treated with BCG. Intravesical gemcitabine instillation immediately postoperatively was also reported to be safe but with no proven RFS advantage. 21 Table 2. Adverse events reported after intravesical gemcitabine therapy Adverse Event* No. Grade 1 or 2 (%) No. Grade 3 (%) Figure 4. Cumulative incidence curves of recurrent bladder cancer by response type, excluding 25 patients with recurrence before 5-month landmark. Solid curve indicates CR. Dashed curve indicates NR plus PR. Urinary 25 (36) 3 (4) Fatigue 24 (35) 0 Diarrhea 7 (10) 0 Hematuria 5 (7) 0 Skin rash 6 (9) 1 (1) Nausea 4 (6) 0 Fever 2 (3) 0 Constipation 2 (3) 0 Pulmonary 2 (3) 0 Urinary tract infection 0 1 (1) Deep vein thrombosis 0 1 (1) Infection 0 2 (2) Thrombocytopenia 0 1 (1) * Including only adverse events in 2 or more patients and grade 3 events according to NCI CTC. Including irritative symptoms, urgency, frequency, dysuria and bladder outlet obstruction.
1690 GEMCITABINE FOR BLADDER CANCER AFTER BACILLUS CALMETTE-GU ERIN FAILURE A recent report of the ongoing SWOG S0353 trial, in which intravesical gemcitabine was administered after BCG failure, described outcomes in 47 treated patients with a followup of at least 12 months. 22 The CR rate was similar to that in our series (45% and 39%, respectively) and 62% of the patients with CR were disease free at 12 months. We report one of the largest series of patients with NMIBC treated with intravesical gemcitabine after failed BCG therapy. We saw indications that intravesical gemcitabine resulted in a lower survival probability in patients with BCG refractory disease than in those with another type of BCG failure, although this was not statistically significant. However, about 80% of patients treated with intravesical gemcitabine remained free of progression for 5 years regardless of the type of BCG failure, which is a higher rate than previously reported. 4 Thus, we saw no reason to suggest that the indication for intravesical gemcitabine depends on the type of BCG failure. Nevertheless, the heterogeneity of patients with another type of BCG failure could have impacted the lack of difference between the 2 groups. Huguet et al reported that patients treated with cystectomy for NMIBC after BCG failure fared worse if cancer was up-staged to MIBC at cystectomy (5-year CSS 38% vs 90%). 23 In our cohort no patient with CR was up-staged at cystectomy and overall only 15% treated with radical cystectomy after intravesical gemcitabine had MIBC. Moreover, in our study patients who received intravesical gemcitabine had a 5-year CSS rate of about 85%, similar to the rate reported by Huguet et al in patients who were not up-staged at cystectomy, regardless of the response to intravesical gemcitabine. This observation may merely reflect the selection of patients for intravesical gemcitabine but it may also reflect the efficacy of this treatment. In patients who underwent cystectomy after intravesical gemcitabine failed an advantage of achieving a CR to intravesical gemcitabine was longer time to cystectomy (20.9 vs 5.3 months). However, this should be viewed with caution since all patients in our cohort initially refused or were unfit for cystectomy. Thus, those who underwent cystectomy after intravesical gemcitabine represent a select patient group. Since none of these patients had MIBC at surgery, the longer time to cystectomy can be attributable to the effect of intravesical gemcitabine and should be considered an advantage rather than a delay in recognizing the need for cystectomy. It is also plausible that a CR to gemcitabine reflects less aggressive disease. Whether it was a treatment effect or less aggressive disease, patients with a CR to gemcitabine can probably be followed, while those who do not achieve a CR should probably be counseled to consider cystectomy. Our study also shows that intravesical gemcitabine is safe and well tolerated. Of the patients 88% completed the treatment protocol and 29 of the 69 did not experience an adverse event. Most adverse events experienced by the others were grade 1 or 2. Only 12% of patients had grade 3 adverse events and none had a grade 4 or 5 adverse event. While this is a retrospective study with a relatively small cohort and no comparison arm with standard treatment, its main strength is its homogenous cohort composed entirely of patients previously treated with BCG, of whom 54% had BCG refractory disease. CONCLUSIONS In our experience intravesical gemcitabine should be considered after BCG failure in patients with NMIBC who refuse radical cystectomy or are not candidates for major surgery. We found no indication to suggest that the efficacy of intravesical gemcitabine depends on the type of BCG failure. In patients who eventually elected cystectomy after intravesical gemcitabine treatment an advantage of achieving a CR was longer time to cystectomy and lack of progression to MIBC. REFERENCES 1. Babjuk M, Oosterlinck W, Sylvester R et al: EAU guidelines on non-muscle-invasive urothelial carcinoma of the bladder, the 2011 update. Eur Urol 2011; 59: 997. 2. Sylvester RJ, Brausi MA, Kirkels WJ et al: Long-term efficacy results of EORTC Genito- Urinary Group randomized phase 3 study 30911 comparing intravesical instillations of epirubicin, bacillus Calmette-Guerin, and bacillus Calmette-Guerin plus isoniazid in patients with intermediate- and high-risk stage Ta T1 urothelial carcinoma of the bladder. Eur Urol 2010; 57: 766. 3. Zlotta AR, Fleshner NE and Jewett MA: The management of BCG failure in non-muscleinvasive bladder cancer: an update. Can Urol Assoc J, suppl., 2009; 3: S199. 4. Lerner SP, Tangen CM, Sucharew H et al: Failure to achieve a complete response to induction BCG therapy is associated with increased risk of disease worsening and death in patients with high risk non-muscle invasive bladder cancer. Urol Oncol 2009; 27: 155. 5. Barlow LJ, Seager CM, Benson MC et al: Novel intravesical therapies for non-muscle-invasive bladder cancer refractory to BCG. Urol Oncol 2010; 28: 108. 6. Yates DR and Roupret M: Contemporary management of patients with high-risk nonmuscle-invasive bladder cancer who fail intravesical BCG therapy. World J Urol 2011; 29: 415. 7. Dash A, Pettus JA 4th, Herr HW et al: A role for neoadjuvant gemcitabine plus cisplatin in muscle-invasive urothelial carcinoma of the bladder: a retrospective experience. Cancer 2008; 113: 2471.
GEMCITABINE FOR BLADDER CANCER AFTER BACILLUS CALMETTE-GU ERIN FAILURE 1691 8. Jin JO, Lehmann J, Taxy J et al: Phase II trial of adjuvant gemcitabine plus cisplatin-based chemotherapy in patients with locally advanced bladder cancer. Clin Genitourin Cancer 2006; 5: 150. 9. Lorusso V, Pollera CF, Antimi M et al: A phase II study of gemcitabine in patients with transitional cell carcinoma of the urinary tract previously treated with platinum. Italian Co-operative Group on Bladder Cancer. Eur J Cancer 1998; 34: 1208. 10. Meliani E, Lapini A, Serni S et al: Gemcitabine plus cisplatin in adjuvant regimen for bladder cancer. Toxicity evaluation. Urol Int 2003; 71: 37. 11. Scosyrev E, Messing EM, van Wijngaarden E et al: Neoadjuvant gemcitabine and cisplatin chemotherapy for locally advanced urothelial cancer of the bladder. Cancer 2012; 118: 72. 12. Shelley M, Cleves A, Wilt TJ et al: Gemcitabine for unresectable, locally advanced or metastatic bladder cancer. Cochrane Database Syst Rev 2011; 4: CD008976. 13. Dalbagni G, Russo P, Sheinfeld J et al: Phase I trial of intravesical gemcitabine in bacillus Calmette-Guerin-refractory transitional-cell carcinoma of the bladder. J Clin Oncol 2002; 20: 3193. 14. Dalbagni G, Russo P, Bochner B et al: Phase II trial of intravesical gemcitabine in bacille Calmette-Guerin-refractory transitional cell carcinoma of the bladder. J Clin Oncol 2006; 24: 2729. 15. Lerner SP, Schoenberg MP and Sternberg CN: Textbook of Bladder Cancer. Abingdon, United Kingdom: Taylor & Francis 2006. 16. Witjes JA, van der Heijden AG, Vriesema JL et al: Intravesical gemcitabine: a phase 1 and pharmacokinetic study. Eur Urol 2004; 45: 182. 17. Gontero P, Casetta G, Maso G et al: Phase II study to investigate the ablative efficacy of intravesical administration of gemcitabine in intermediate-risk superficial bladder cancer (SBC). Eur Urol 2004; 46: 339. 18. Bartoletti R, Cai T, Gacci M et al: Intravesical gemcitabine therapy for superficial transitional cell carcinoma: results of a Phase II prospective multicenter study. Urology 2005; 66: 726. 19. Serretta V, Galuffo A, Pavone C et al: Gemcitabine in intravesical treatment of Ta-T1 transitional cell carcinoma of bladder: Phase I-II study on marker lesions. Urology 2005; 65: 65. 20. Laufer M, Ramalingam S, Schoenberg MP et al: Intravesical gemcitabine therapy for superficial transitional cell carcinoma of the bladder: a phase I and pharmacokinetic study. J Clin Oncol 2003; 21: 697. 21. Bohle A, Leyh H, Frei C et al: Single postoperative instillation of gemcitabine in patients with non-muscle-invasive transitional cell carcinoma of the bladder: a randomised, double-blind, placebo-controlled phase III multicentre study. Eur Urol 2009; 56: 495. 22. Skinner EC, Goldman B, Sakr WA et al: SWOG S0353: Phase II trial of intravesical gemcitabine in patients with non-muscle invasive bladder cancer who recurred following at least two prior courses of BCG. Presented at annual meeting of American Urological Association, Atlanta, Georgia, May 19-23, 2012, abstract 1666. 23. Huguet J, Crego M, Sabate S et al: Cystectomy in patients with high risk superficial bladder tumors who fail intravesical BCG therapy: pre-cystectomy prostate involvement as a prognostic factor. Eur Urol 2005; 48: 53.