The Impact of Adding Taxanes to Gemcitabine and Platinum Chemotherapy for the First-Line Therapy of Advanced or Metastatic Urothelial Cancer: A Systematic Review and Meta-analysis

EURURO-6474; No. of Pages 10 EUROPEAN UROLOGY XXX (2015) XXX–XXX

available at www.sciencedirect.com journal homepage: www.europeanurology.com

Platinum Priority – Review – Urothelial Cancer Editorial by XXX on pp. x–y of this issue

The Impact of Adding Taxanes to Gemcitabine and Platinum Chemotherapy for the First-Line Therapy of Advanced or Metastatic Urothelial Cancer: A Systematic Review and Meta-analysis Patrizia Giannatempo a,y, Gregory R. Pond b,y, Guru Sonpavde c, Daniele Raggi a, Gurudatta Naik c, Matthew D. Galsky d, Joaquim Bellmunt e, Andrea Necchi a,* a

Fondazione IRCCS Istituto Nazionale dei Tumori, Milan, Italy;

b

McMaster University, Hamilton, Ontario, Canada; c UAB Comprehensive Cancer Center,

Birmingham, AL, USA; d Mount Sinai School of Medicine, Tisch Cancer Institute, New York, USA; e Dana-Farber Cancer Institute and Harvard Medical School, Boston, MA, USA

Article info

Abstract

Article history: Accepted September 29, 2015

Context: Gemcitabine/platinum chemotherapy is the most widely used first-line regimen for metastatic urothelial carcinoma, and the potential improvement of adding taxanes needs to be clarified. Objective: To study the survival impact of taxane plus gemcitabine/platinum compared with gemcitabine/platinum alone as upfront therapy. Evidence acquisition: Literature was searched for studies including gemcitabine/platinum  taxanes (paclitaxel or docetaxel only). We pooled trial level data including the median, proportions, and confidence intervals on response-rate, progression-free survival, overall survival (OS), and side effects. Univariable and multivariable regression models evaluated the prognostic role of addition of taxanes after adjusting for platinum type, performance status 2, and the presence of visceral metastases. Data were weighted by the logarithm of the trial sample size. Evidence synthesis: Thirty-five arms of trials including 2,365 patients were selected (seven with taxanes [n = 617], and 28 arms without taxanes [n = 1,748]). Median OS was univariably significantly different (p = 0.019) between trials with and without taxanes. Across trials, the median ‘median OS’ amongst trials containing taxanes was 15.5 mo, compared with 12.5 mo in trials which did not. Multivariably, visceral disease and performance status were significantly associated with OS, and the addition of taxanes trended toward significantly better OS (p = 0.056) and increase in grade  3 neurotoxicity (p = 0.051), regardless of specific platinum agent used. Conclusions: In this meta-analysis, adding taxanes to gemcitabine and platinum showed a trend for improved OS and higher grade  3 neurotoxicity. Improvements in patient selection and the evaluation of a more potent and tolerable tubulin inhibitor in combination with gemcitabine/platinum in a well-powered trial are the critical next steps. Patient summary: In this report, a trend for improved overall survival and worse neurotoxicity was observed for adding a taxane to first-line gemcitabine/platinum chemotherapy for metastatic urothelial carcinoma. More effective taxanes should be investigated further in urothelial carcinoma in combination with gemcitabine/platinum. # 2015 European Association of Urology. Published by Elsevier B.V. All rights reserved.

Associate Editor: James Catto Keywords: First-line therapy Meta-analysis Systemic therapy Taxanes Urothelial carcinoma

y These authors contributed equally. * Corresponding author. Department of Medical Oncology, Fondazione IRCCS Istituto Nazionale dei Tumori, Via Giacomo Venezian 1, 20133 Milano, Italy. Tel. +39 02 2390 2402; Fax: +39 02 2390 3150. E-mail address: [email protected] (A. Necchi).

http://dx.doi.org/10.1016/j.eururo.2015.09.051 0302-2838/# 2015 European Association of Urology. Published by Elsevier B.V. All rights reserved.

Please cite this article in press as: Giannatempo P, et al. The Impact of Adding Taxanes to Gemcitabine and Platinum Chemotherapy for the First-Line Therapy of Advanced or Metastatic Urothelial Cancer: A Systematic Review and Meta-analysis. Eur Urol (2015), http://dx.doi.org/10.1016/j.eururo.2015.09.051

EURURO-6474; No. of Pages 10 2

EUROPEAN UROLOGY XXX (2015) XXX–XXX

1.

Introduction

Despite the remarkable progress in the therapeutic paradigm of urothelial cancer (UC), particularly in advanced stages, UC has suffered a 2-decade delay characterized by plenty of unsuccessful systemic treatments that have resulted from negative trials, often plagued by early closure due to poor accrual. With the use of methotrexate, cisplatin, doxorubicin, and vinblastine combination and gemcitabine/ cisplatin (GC) doublet, despite an overall objective response-rate of about 50%, the median progression-free survival (PFS) and overall survival (OS) attained are approximately 8 mo and 14 mo, respectively [1]. Furthermore, cisplatin is being administered in less than half of patients once they are diagnosed with advanced UC, and GC doublet is the alternative option; although, its efficacy seems to be inferior to that of cisplatin-based therapy [2–5]. In the attempt to improve the available results, the addition of a third drug to platinum doublets has been investigated. In particular, given the single agent activity of taxanes [6,7], the combination of GC with paclitaxel has been investigated in a phase 3 trial of 626 patients led by the European Organization for the Research and Treatment of Cancer (EORTC) [8]. A nonsignificant trend (p = 0.075) towards a better OS with the triple combination regimen was observed, although a larger trial may have potentially identified a statistically significant difference. Additional smaller phase 2 studies have reported results with either gemcitabine or platinum alone or in combination with paclitaxel or docetaxel [9–35]. Results with gemcitabineplatinum doublet as the standard arm of randomized phase 3 studies were also available [36–39]. Although promising new agents are emerging, there remains a role for combinations of chemotherapeutic drugs with proven single agent activity. Given that the aforementioned single phase 3 EORTC 30987 trial showed a trend for improved survival by the addition of paclitaxel to GC, we hypothesized that a pooled analysis of multiple trials will increase the power to identify a significant increment in OS. Hence, we conducted a systematic review and meta-analysis of the results from prospective studies which either did or did not contain paclitaxel or docetaxel in combination with gemcitabine plus platinum.

2.

Evidence acquisition

2.1.

Search strategy and data abstraction

From December 2014 to April 2015, we performed a systematic review and meta-analysis in accordance to the preferred reporting items for systematic reviews and metaanalyses guidelines [40]. Eligible randomized or nonrandomized phase 2 and 3 studies were searched in PubMed, EMBASE, and meeting abstracts presented at congresses of the American Society of Clinical Oncology, European Society for Medical Oncology, American Association of Cancer Research, and Genitourinary Cancers Symposiums.

The following inclusion criteria have been adopted: period of publication between 1990 and 2014, English language, retrospective and prospective trials/studies reporting data on gemcitabine and platinum doublet, alone or with a taxane (including paclitaxel or docetaxel only). The administration of prior perioperative chemotherapy was allowed. Principal exclusion criteria were overlapping publications, lack of relevant outcome data, less than 15 patients, studies entirely reporting on patients treated prior to 1990, and studies reporting on either gemcitabine/ platinum + nontaxane combination regimens. Also, trials that allowed the crossover between gemcitabine/platinum  new drugs were excluded (since new drugs may have unknown activity and benefit), as well as those where any agent had been administered sequentially or as a maintenance (since this approach is not standard). The population, intervention, comparison, and outcome (question: which are the outcomes of gemcitabine/platinum chemotherapy or gemcitabine/platinum plus a taxane as first-line therapy for advanced or metastatic UC?) strategy was conducted and the following search string was utilized: ‘transitional cell carcinoma’/exp OR transitional AND cell AND carcinoma:ab,ti AND ‘chemotherapy’/ exp OR ‘salvage therapy’/exp OR ‘single drug dose’/exp OR ‘cancer combination chemotherapy’/exp OR ‘salvage therapy’:ab,ti OR ‘combination chemotherapy’:ab,ti OR ‘single drug dose’:ab,ti AND ‘clinical effectiveness’/exp OR ‘overall survival’/exp OR ‘progression free survival’/exp OR ‘cancer staging’/exp OR ‘toxicity’/exp. Additional queries with relevant variants and filters have been added up, integrated by the search through the American Society of Clinical Oncology portal. Search results were independently reviewed by two authors (DR, AN). Full articles were retrieved for further qualitative review. 2.2.

Statistical analyses

The primary objective was to compare taxane- versus nontaxane studies, and the secondary objective was to compare cisplatin versus carboplatin studies. The primary endpoint was median OS, while secondary endpoints included 1-yr OS, response-rate, median PFS, and the rate of adverse events. Outcomes were defined as per each study definitions, however, PFS was commonly defined as the time from the date of starting treatment to the date of documented relapse, recurrence, and censoring patients who have died without progression, while OS was defined as the time from treatment start to death for any reason, with censoring alive patients at the date of last contact. Descriptive statistics were used to summarize information across all trials, and grouped by whether the treatment contained a taxane (docetaxel vs paclitaxel) and by platinum status (carboplatin vs cisplatin). Trial characteristics and outcomes were compared between trials using the Fisher’s exact test (dichotomous characteristics) or Wilcoxon rank sum test (continuous characteristics and outcomes). These analyses were univariable only. Bootstrapping was performed to evaluate the sensitivity of the

Please cite this article in press as: Giannatempo P, et al. The Impact of Adding Taxanes to Gemcitabine and Platinum Chemotherapy for the First-Line Therapy of Advanced or Metastatic Urothelial Cancer: A Systematic Review and Meta-analysis. Eur Urol (2015), http://dx.doi.org/10.1016/j.eururo.2015.09.051

EURURO-6474; No. of Pages 10 EUROPEAN UROLOGY XXX (2015) XXX–XXX

significance of the primary outcome result (ie, median OS). Only those characteristics and outcomes which were recorded in at least 10 trials were included for analysis. Multivariable regression models were added up whereby a single regression model was calculated for each outcome which was reported in at least 20 trials. Statistical power was limited for multivariable analyses, so 20 was arbitrarily chosen for pragmatic reasons. Regression models evaluated whether a taxane or platinum status was significantly prognostic for outcome, after adjusting for each other, for the proportion of patients in the trial who had an Eastern Cooperative Oncology Group-Performance Status (ECOGPS) 2, and the rate of patients with visceral metastases. Data were weighted using the logarithm of the trial sample size. Publication bias was evaluated by visually inspecting funnel plots and using Egger’s test for bias. Heterogeneity was assessed through visual inspection of forest plots, through the I2 statistic and through Cochran’s Q test. Evaluations were assessed within each subgroup of interest (taxane, no taxane, cisplatin, no cisplatin). Standard error of the median OS and median PFS was estimated as 1/H(1/[# of patients/3]), as the true standard error was not available for most trials. Statistical significance was defined as a p-value of 0.05 or less, and all tests and confidence intervals were two-sided. No adjustment for multiple testing was performed. Descriptive analyses were performed using SAS v9.0 (SAS Institute, Cary, NC, USA), while bootstrap analysis and meta-analytic diagnostic statistics were calculated using R v3.1.2. (www.r-project.org) using the packages boot and metafor. 3.

Evidence synthesis

3.1.

Search results

We identified a total of 8,676 studies using the search criteria (EMBASE: 3,340; MEDLINE: 310; EMBASE and MEDLINE: 5,026), published between 1999 and 2014. Figure 1 outlines the selection process and reasons for study exclusion, and Supplementary Table 1 provides the quality assessment of the included studies. A total of 32 studies accounting for 35 arms of trials were selected for the meta-analysis, including 34 prospective studies and one retrospective study resulting in a total of 2,365 evaluable patients (seven studies with a taxane [n = 617], 16 with carboplatin [n = 657], and 19 cisplatin [n = 1,708], Table 1). Three trials were 2-arm randomized controlled trials in which both arms were included in this analysis; no adjustment was performed to account for potential correlation between arms. The remaining 29 trials were either single-arm trials, or two-arm trials in which only one arm contained usable information. Summary statistics across all trials, and by taxane status, are presented in Table 2, while statistics by platinum status (ie, cisplatin vs carboplatin studies) is in Supplementary Table 2. Information on the site of the primary tumor was infrequent across the studies and only 16 arms provided it. The median percentage of patients with an upper tract UC was 14.3% overall (range, 2.4–28.6%), and there was a significant difference between the nontaxane

3

Fig. 1 – Preferred Reporting Items for Systematic Reviews and metaanalysis flowchart. GU-ASCO = American Society of Clinical Oncology (Genitourinary Cancers Symposium); PICO = population, intervention, comparison, and outcome.

(median 18.8%) and the taxane group (median 4.1%, p = 0.037). The median ‘median follow up’ was 24.3 mo (range, 12.9–54 mo) in the taxane group and 13.8 mo (range, 1–60 mo) in the nontaxane group. 3.2.

Uni- and multivariable meta-analyses for response and

survival outcome

In univariable analyses, the median response-rate was not statistically different between the taxane and nontaxane groups (p = 0.27), being 55.5% (range, 40–81%) for the taxane group and 45.7% (range, 35–67%) for the nontaxane group, respectively. Median PFS trended to significance in favor of the taxane group (p = 0.069). Among the 29 arms with evaluable information, the median ‘median PFS’ was 8.0 mo (range, 7.4–10.0 mo, I-square = 0%, p for heterogeneity = 0.99, Egger’s test p = 0.99) for the taxane group and 7.4 mo (range, 3.5–9.4 mo, I-square = 0%, p for heterogeneity = 1.00, Egger’s test p = 0.82) for the nontaxane group. Across the 34 evaluable arms, median OS was statistically significantly (p = 0.019) different between the arms which contained a taxane and those which did not. The 95% bias-corrected and accelerated (BCa) confidence interval for the p-value was <0.001–0.72 and 60.8% of bootstrapped

Please cite this article in press as: Giannatempo P, et al. The Impact of Adding Taxanes to Gemcitabine and Platinum Chemotherapy for the First-Line Therapy of Advanced or Metastatic Urothelial Cancer: A Systematic Review and Meta-analysis. Eur Urol (2015), http://dx.doi.org/10.1016/j.eururo.2015.09.051

Author

Yr

No.

Age (median, yr)

Timeframe

Study type

Randomized

ECOG-PS 2 no. (%)

Visceral metastases: no. (%)

Upper Prior peritract operative UC: N (%) chemotherapy: no. (%)

Treatment

No. of cycles (Median)

FUP Response(mo, rate median) (CR+PR): no. (%)

PFS (median, mo)

OS (median, mo)

Gemcitabine + platinum alone 1999 2000 1999 2000 2014 2013 2006 2012 2001

42 51 31 203 28 64 60 38 16

64 61 69 63 65.8 65 69 74 68

1995–1996 NR 1996–1997 1996–1998 2008–2011 2003–2008 2002–2003 2004–2009 1997–1998

Prospective Prospective Prospective Prospective Prospective Prospective Prospective Prospective Prospective

No No No Yes Yes Yes No No No

20 (47.6) 11 (21.6) 4 (12.9) 38 (18.7) 2 (7.1) 0 8 (13.3) 11 (28.9) 3 (18.7)

25 (59.5) 15 (29.4) 24 (77.4) 99 (48.8) 7 (25) 19 (29.7) 40 (66.7) 14 (36.8) 8 (50)

Carles et al [16] Hoschke et al [17] Khaled et al [18] Baitaret al [19] Krege et al [20] Sternberg et al [38] Lorusso et al [21] Bamias et al [22] Linardou et al [23] Nogue-Aliguer et al [24] Hudson et al [25] Shannon et al [26] Xu et al [27] De Santis et al [39] Dogliotti et al [28]

2000 2004 2008 2011 2014 2013 2000 2007 2004 2003 2010 2001 2007 2012 2007

Prospective Prospective Prospective Prospective Prospective Prospective Prospective Prospective Prospective Prospective Retrospective Prospective Prospective Prospective Prospective

No No No No Yes Yes No No No No No No No Yes Yes

2013

69 68 55 67 67.3 64 67 75.5 75 66 67 69 64.5 70 67 67 74

1997–1998 1999–2002 NR 2003–2007 2006–2010 2008–2010 1997–1998 2003–2006 2000–2002 1999 2003–2008 1998–1999 2003–2006 2001–2008 2000–2002

Park et al [29]

17 23 57 23 49 171 54 34 56 41 15 17 41 119 55 55 31

2005–2012

Prospective

No

7 (41.2) 5 (21.7) NR 6 (26.1) 1 (2.0) 7 (4.1) 18 (33.3) 23 (67.6) 26 (46.4) 15 (36.6) 3 (20.0) 5 (29.4) 5 (12.2) 53 (44.5) 3 (5.4) 8 (14.5) 2 (6.4)

6 (35.3) 13 (56.5) NR 19 (82.6) 23 (46.9) 130 (76.0) 32 (59.2) 15 (44.1) 33 (58.9) 13 (31.7) 10 (66.7) 15 (88.2) 16 (39.0) 55 (46.2) 44 (80) 31 (56.4) 9 (29)

2000–2005 2000–2003 1997–2000 1999–2002

Prospective Prospective Prospective Prospective

No No No Yes

2001–2004

Prospective

Yes

1997–1999 2004–2007

Prospective Prospective

No No

NR NR NR NR 7 (25) 12 (18.7) 9 (15) NR NR

0 9 4 0 6 0 5 0 3

GEM-CDDP GEM-CDDP GEM-CDDP GEM-CDDP (32.1) GEM-CDDP (DD)GEM-CDDP (8.3) GEM-CBDCA (AUC5) GEM-CDDP (18.7) GEM-CBDCA (AUC 4.5 or 5) NR 3 (17.6) GEM-CBDCA (AUC 5) 3 (13) 0 GEM-CBDCA (AUC4.5) NR NR GEM (c.i.)-CDDP 3 (13) 0 GEM-CBDCA (AUC5) 14 (28.6) 0 GEM-CDDP 40 (23.4) 36 (21.1) GEM-CDDP NR 7 (12.9) GEM-CDDP 4 (11.8) 0 GEM-CBDCA NR 0 GEM-CBDCA (AUC4) NR 6 (14.6) GEM-CBDCA (AUC 5) NR 0 CBDCA (AUC 5.6)-GEM 4 (23.5) 3 (17.6) GEM-CBDCA (AUC5) 11 (26.8) 0 GEM-CBDCA (AUC5) 24 (20.2) 0 GEM-CBDCA (AUC 4.5) NR 0 GEM-CDDP NR 0 GEM-CBDCA (AUC 5) NR 0 GEM-CBDCA (AUC 5)

4 5.5 5 6 6 6 6 3 4

60 NR 12 19 NR 52.1 18.4 14.14 NR

16 19 16 81 16 32 23 15 7

7.2 5.5 NR 7.4 8.5 7.8 7.6 3.5 NR

12.5 14.3 13.2 13.8 17.4 18 16.3 8.5 NR

4 4 4 3 4 4 4 6 6 5.5 4 4 4 4 4 4 6

NR 16.7 12 NR NR NR NR 8 13.5 11.2 NR 8.5 19.2 54 7.2 6.9 15

9 14 32 8 23 43 48 24 36 23 10 10 18 49 27 22 14

NR 7.8 7.2 NR 6 7.6 NR NR 4.8 7.2 NR 4.6 7.5 5.8 8.3 7.7 9.4

10 15.4 11.5 3.3 10.4 14.3 13.5 9.8 7.2 10.1 9 10.5 13.6 9.3 12.8 9.8 20

NR NR 2 (4.1) 1 (2.4) 3 (6.9) 40 (12.8) 43 (13.7) NR NR

5 4 6 NR NR 6 5 4 4

17.5 31 NR NR NR 55.2 55.2 12.9 NR

48 26 32 18 19 173 137 23 24

10 7.4 NR 8 6.5 8.3 7.6 8.9 7.7

22 11 14.7 15.2 12.2 15.8 12.7 15.5 21.4

(17.6) (12.9)

Gemcitabine + platinum and paclitaxel or docetaxel Ecke et al [30] Hainsworth et al [31] Hussain et al [32] Lorusso et al [33]

2006 2005 2001 2005

Bellmunt et al [8]

2012

Pectasides et al [34] Boukovinas et al [35]

2002 2012

59 60 49 42 43 312 314 35 60

70 63 63 69 68 61 61 65 65.5

9 0 5 8 6 0 0 8 5

(15.2) (10.2) (19.0) (13.9)

(22.8) (8.3)

9 35 24 23 19 145 155 27 26

(15.3) (58.3) (48.9) (54.8) (44.2) (46.5) (49.4) (77.1) (43.3)

0 4 4 4 5 0 0 8 4

(6.7) (8.2) (9.5) (11.6)

(22.8) (6.7)

TXL-CDDP-GEM TXL-CBDCA (AUC5)-GEM TXL-CBDCA (AUC5)-GEM TXL-CDDP-GEM GEM-CDDP TXL-CDDP-GEM GEM-CDDP TXT-CDDP-GEM TXT-CDDP-GEM

AUC = area under the curve; CBDCA = carboplatin; CDDP = cisplatin; c.i. = continuous infusion; CR = complete response; DD = dose-dense; ECOG = Eastern Cooperative Oncology Group; FUP = follow up; GEM = gemcitabine; NR = not reported; OS = overall survival; PFS = progression-free survival; PR = partial response; PS = performance status; TXL = paclitaxel; TXT = docetaxel; UC = urothelial cancer.

EUROPEAN UROLOGY XXX (2015) XXX–XXX

von der Maase et al [9] Kaufman et al [10] Moore et al [11] von der Maase et al [36] Hussain et al [12] Bamias et al [37] Bamias et al [13] Morales-Barrera et al [14] Bellmunt et al [15]

EURURO-6474; No. of Pages 10

4

Please cite this article in press as: Giannatempo P, et al. The Impact of Adding Taxanes to Gemcitabine and Platinum Chemotherapy for the First-Line Therapy of Advanced or Metastatic Urothelial Cancer: A Systematic Review and Meta-analysis. Eur Urol (2015), http://dx.doi.org/10.1016/j.eururo.2015.09.051

Table 1 – Patient, disease, and treatment characteristics of included studies

EURURO-6474; No. of Pages 10 5

EUROPEAN UROLOGY XXX (2015) XXX–XXX

Table 2 – Summary of characteristics and outcomes between the taxane and nontaxane groups N

Results

No taxane

With taxane

p value

(28.6) (22.9) (20.0) (28.6) (97.1) (34.3) (54.3) (45.7) (14.3) (5.7) (15, 314) (34.3)

28 8 (28.6) 5 (17.9) 7 (25.0) 8 (28.6) 27 (96.4) 10 (35.7) 14 (50.0) 14 (50.0) 0 (0.0) 0 (0.0) 42.5 (15, 314) 10 (35.7)

7 2 3 0 2 7 2 5 2 5 2 59 2

67 (55, 75) – 18.8 (0, 67.6) 49.4 (25.0, 88.2) 1 (6.7) 1 (6.7) 2 (15.4) 9 (69.2) 0 (0, 21.4) 18.8 (7.0, 28.6)

65 (61, 70) – 10.2 (0, 22.9) 49.0 (15.3, 77.1) 0 (0.0) 0 (0.0) 0 (0.0) 2 (100.0) 6.7 (0, 22.9) 4.1 (2.4, 12.8)

0.23 – 0.13 0.77 0.84

4 (3, 6) 13.8 (1, 60) 45.7 (24.0, 67.0) 7.4 (3.5, 9.4) 12.5 (3.3, 20.0) 52.8 (26, 82)

4.5 24.3 55.5 8.0 15.5 61.4

0.63 0.22 0.27 0.069 0.019 0.15

Baseline characteristics N N (%), yr

Prospective Randomized Cisplatin Carboplatin Paclitaxel Docetaxel Sample Size 2-arm trials

35 10 8 7 10 34 12 19 16 5 2 49 12

1999–2001 2002–2006 2007–2011 2012–2014 N (%) N (%) N (%) N (%) N (%) N (%) Med (range) N (%)

35

Med Med Med Med 2.5 4 4.5 5 Med Med

(range) (range) (range) (range)

35 7 34 34 15

(range) (range)

34 16

67 (55, 75.5) 62.5 (54, 82) 17.0 (0, 67.6) 49.2 (15.3, 88.2) 1 (6.7) 1 (6.7) 2 (13.3) 11 (73.3) 0 (0, 22.9) 14.3 (2.4, 28.6)

Med Med Med Med Med Med

(range) (range) (range) (range) (range) (range)

33 22 33 27 34 16

4 (3, 6) 14.6 (1, 60) 46.2 (24.0, 81.0) 7.5 (3.5, 10.0) 13.0 (3.3, 22.0) 55.0 (26.0, 82.0)

35 35 35 35 35 35 35 35

(28.6) (42.9) (0.0) (28.6) (100.0) (28.6) (71.4) (28.6) (71.4) (28.6) (35, 312) (28.6)

0.84

1.00 1.00 0.42 0.42 – – 0.16 1.00

Patient level characteristics Mean age, yr % MSKCC score 1–2 % ECOG PS=2 % with visceral metastases Carboplatin dose AUC

% prior peri-operative chemo % Upper tract UC

0.55 0.037

Outcomes Median no. of cycles Median mo of follow-up Response rate Median PFS Median OS 1-yr OS

(4, 6) (12.9, 54.0) (40, 81) (7.4, 10.0) (11, 22) (46, 73.3)

AUC = area under the curve; ECOG PS = Eastern Cooperative Oncology Group-Performance Status; OS = overall survival; PFS = progression-free survival; UC = urothelial cancer.

tests resulted in statistically significant p-values. The median ‘median OS’ among trials containing a taxane was 15.5 mo (range, 11–22 mo, I-square = 0%, p for heterogeneity = 0.55, Egger’s test p = 0.78), compared with 12.5 mo (range, 3.3–20 mo, I-square = 27%, p for heterogeneity = 0.25, Egger’s test p = 0.40) among trials which did not contain a taxane. When considering cisplatin studies only, the significant difference was confirmed (p = 0.025), while there were no differences between the taxane and nontaxane groups when selecting carboplatin studies only (p = 0.35). Forest plots for the OS outcome are provided in Figure 2A and 2B. Funnel plots for the taxane and nontaxane groups are provided as Supplementary Figure 1A and 1B. Median OS was also statistically significant (p = 0.031) between trials which used carboplatin versus cisplatin. The 95% bias-corrected and accelerated confidence interval for the p-value was <0.001–0.50 and 73.4% of bootstrapped tests resulted in statistically significant p-values. Among trials containing carboplatin, the median ‘median OS’ was 10.1 mo (range, 3.3–20 mo, I-square = 42%, p for heterogeneity = 0.061, Egger’s test p = 0.63), compared with 13.8 mo (range, 8.5–22 mo, I-square = 0%, p for heterogeneity = 0.82, Egger’s test p = 0.80) among trials containing cisplatin. Forest plots

for the OS outcome are provided in Figure 2C and 2D. Funnel plots for the carboplatin and cisplatin groups are provided as Supplementary Figure 1C and 1D. A detailed outcome assessment according to the combination chemotherapy is presented in Table 3. Multivariably, taxane status was not statistically significant for response-rate (p = 0.15) or PFS (p = 0.094), while it trended to significance for OS (p = 0.056) after adjusting for platinum status, percentage of patients with ECOG=2, and with visceral metastases (Table 4). The platinum type was not significant for any survival outcome. 3.3.

Incidence of acute toxicities

A summary of all grade and grade 3–4 side effects in the taxane and nontaxane groups is presented in Supplementary Table 3, while a detailed presentation of toxicities by study arm is provided in Supplementary Table 4. The only statistically significant difference between the taxane and nontaxane groups was in the rate of grade 3–4 neuropathy (p = 0.024) and grade 3–4 neutropenic fever (p = 0.035). After adjusting for the platinum chemotherapy, the presence of visceral disease and ECOG-PS 2 in the

Please cite this article in press as: Giannatempo P, et al. The Impact of Adding Taxanes to Gemcitabine and Platinum Chemotherapy for the First-Line Therapy of Advanced or Metastatic Urothelial Cancer: A Systematic Review and Meta-analysis. Eur Urol (2015), http://dx.doi.org/10.1016/j.eururo.2015.09.051

EURURO-6474; No. of Pages 10 6

EUROPEAN UROLOGY XXX (2015) XXX–XXX

Fig. 2 – Forest plots of (A) overall survival of taxane studies, (B) overall survival of nontaxane studies, (C) overall survival of cisplatin studies, and (D) overall survival of carboplatin studies. RE = random effects.

multivariable model, taxane status was trending toward significance for grade 3–4 neurotoxicity (p = 0.051, Table 4). The median percentage of drug-related mortality was 0.8% (range, 0–4.8%) in the taxane group and 0.5% (range, 0–11.8%) in the nontaxane group.

3.4.

Discussion

In the present systematic review and meta-analysis, unique to the authors’ knowledge, we observed a trend toward a better OS with the addition of taxanes to gemcitabine and

Table 3 – Efficacy outcomes according to the combination chemotherapy Outcome RR (%) Median PFS (mo) Median OS (mo) 1-yr OS (%)

GEM-CDDP 13, 12, 14, 6,

47 (38–65.3) 7.3 (3.5–8.5) 13 (8.5–18) 53.4 (28–82)

GEM-CBDCA 13, 9, 13, 5,

45.1 (24–67) 7.5 (4.6–9.4) 10 (3.3–20) 42 (26–58.5)

GEM-CDDP-taxane 5, 5, 5, 3,

55.5 (40–81) 8.3 (7.4–10) 15.8 (14–22) 68 (61.4–73.3)

GEM-CBDCA-taxane 2, 1, 2, 2,

55.5 (43–68) 7.4 (7.4–7.4) 12.9 (11–14.7) 52.3 (46–58.5)

CBDCA = carboplatin; CDDP = cisplatin; GEM = gemcitabine; OS = overall survival; PFS = progression-free survival; RR = response-rate.

Please cite this article in press as: Giannatempo P, et al. The Impact of Adding Taxanes to Gemcitabine and Platinum Chemotherapy for the First-Line Therapy of Advanced or Metastatic Urothelial Cancer: A Systematic Review and Meta-analysis. Eur Urol (2015), http://dx.doi.org/10.1016/j.eururo.2015.09.051

EURURO-6474; No. of Pages 10 7

EUROPEAN UROLOGY XXX (2015) XXX–XXX

Table 4 – Multivariate model p-values (reported in I20 trials) Outcome

Response rate Median PFS Median OS % Hemoglobin, grade 3–4 % Neutropenia, grade 3–4 % Platelets, grade 3–4 % Neuropathy, grade 3–4 % Renal, grade 3–4

ECOG-PS 2

Visceral metastases

Taxane (No vs yes)

Platinum (CBDCA vs CDDP)

0.19 0.030 0.048 0.91 0.28 0.88 0.50 0.55

0.35 0.57 0.018 0.50 0.31 0.089 0.39 0.076

0.15 0.094 0.056 0.33 0.27 0.11 0.051 0.73

0.86 0.64 0.16 0.28 0.62 0.36 0.096 0.53

CBDCA = carboplatin; CDDP = cisplatin; ECOG PS = Eastern Cooperative Oncology Group performance status; OS = overall survival; PFS = progression-free survival.

platinum, even when adjusting for platinum status and known applicable clinical confounders. These results did corroborate those obtained in the large EORTC Intergroup study 30987 and included a wider patient sample, since in order to be comprehensive we allowed the inclusion of cisplatin-unfit patients who had received carboplatin-based combinations. Unfortunately, we were unable to run subgroup analyses that would have allowed evaluating which patient category might benefit the most from the addition of a taxane, and this information will be critical for the next studies. The EORTC 30987 study was the only one reporting on the efficacy outcomes according to the primary tumor site [8]. In this trial, a longer OS was observed with the GC-paclitaxel combination in patients with bladder primary tumor compared with those with an upper tract primary [41]. A few other confounders will deserve evaluation for subgroup analyses: the comparison between the lymphnodal disease only and visceral disease, and between the locally advanced (ie, T4bN0M0 or T3N2-3M0) and distantly metastatic cases. A number of biases should be acknowledged and caution is needed when interpreting the present findings. Firstly, the characteristics of the analyzed studies was a major concern: the overall quality was rather low as it was the result of single-arm studies and with a small sample size for the majority of cases, outcome data were frequently lacking, and patient selection criteria was not the same across the studies. Only two randomized studies were included with a control arm, hence the classical meta-analytic techniques could not be applied (eg, the random effects model could not work). The disproportionate representation of the taxane and nontaxane groups is a further issue, with only 617 of 2,365 patients having received a taxane. Moreover, only two arms totaling 109 patients received a taxane in combination with carboplatin and gemcitabine. Thus, the subanalysis examining the differential impact of taxanes added to specific platinum agent was underpowered, and may have resulted in the lack of statistical significance for OS and some other outcomes, in particular myelosuppression and neuropathy based on platinum agent. Therefore, we have applied a conservative statistical approach and included endpoints that have been reported in at least 20 of the 35 arms of trials, and bootstrapping was performed to increase the sensitivity of the analyses. Secondly, the aim to

be comprehensive with the inclusion of cisplatin fit and unfit patients, as well as both paclitaxel and docetaxeltreated patients in the taxane group, actually accounted for a huge heterogeneity of the available data. This amount of heterogeneity is likely to have impacted the results of the regression model beyond what could be adjusted through the multivariate analyses. Thirdly, although the characteristics of both the studies and patients were well-balanced between the taxane and nontaxane groups, there was a significant difference regarding the percentage of patients with an upper tract carcinoma in favor of the doublet chemotherapy group. This factor could not be included in the multivariable model due to the relatively small numbers. Nevertheless, our pooled analysis included prospective trials and the multivariable analysis validated the prognostic impact of ECOG-PS (p = 0.048) and visceral metastasis (p = 0.018), which lends confidence to the reliability of other results observed. Accurate reporting of adverse events was generally infrequent as it emerges from Supplementary Table 4, and the grading system was not the same across the studies (data not shown). When comparing the incidence of adverse events across the studies, the strongest difference between the taxane and nontaxane group was the trend toward an increased incidence of severe neurotoxicity with taxanes, while the hematological toxicity seemed not to be statistically influenced by the treatment. Unfortunately, we were not able to find any data regarding long-term side effects in the studies analyzed. The secondary objective of the meta-analysis (ie, comparison of cislatin vs carboplatin chemotherapy) did corroborate univariably the superiority of cisplatin combinations for the OS outcome. Nevertheless, multivariably this difference was lost, probably meaning that our understanding on the efficacy of platinum type, when adjusting for clinical prognostic factors (especially ECOG-PS 2), needs to be improved. Novel tubulin-inhibitors have shown promising activity in the salvage setting: the nanoparticle albumin-boundpaclitaxel was investigated as second-line therapy in a Canadian multicenter, single-arm, phase 2 study [42]. The overall response rate was 27.7% and, most importantly, median OS was 10.8 mo despite a high proportion of patients with baseline poor prognostic features. Yet combining nanoparticle albumin-bound-paclitaxel with

Please cite this article in press as: Giannatempo P, et al. The Impact of Adding Taxanes to Gemcitabine and Platinum Chemotherapy for the First-Line Therapy of Advanced or Metastatic Urothelial Cancer: A Systematic Review and Meta-analysis. Eur Urol (2015), http://dx.doi.org/10.1016/j.eururo.2015.09.051

EURURO-6474; No. of Pages 10 8

EUROPEAN UROLOGY XXX (2015) XXX–XXX

GC chemotherapy exhibited significant myelosuppression; although, cisplatin may have been a potentially more optimal partner than carboplatin [43]. The same issue characterized the efforts of combining vinflunine or pemetrexed with doublet chemotherapy [44], and consequently further investigations of these compounds occurred through cisplatin-free combinations or a sequential use after response to standard first-line chemotherapy. Moving forward, eribulin is a novel microtubule inhibitor with a unique mechanism of action leading to G(2)/M phase cell-cycle arrest and apoptosis. Eribulin mesylate has been approved by the Food and Drug Administration for the treatment of metastatic breast cancer in patients who have been previously treated with an anthracycline and a taxane, and promising mature results in the salvage setting of UC have been presented [45]. The combination of eribulin with GC in the first-line setting is under investigation in a randomized phase 2 study, and early safety and activity results from the phase 1 trial have been already presented [46]. Multiple trials have investigated the combination of gemcitabine plus platinum with biologic agents. The combination of GC with bevacizumab is being evaluated in a phase 3 trial based on preliminary promise in a nonrandomized phase 2 trial [47]. However, the combination with apatorsen (antisense oligonucleotide designed to inhibit heat shock protein 27) in unselected patients and trastuzumab with gemcitabine-platinum in human epidermal growth factor receptor-2 overexpressing or amplified patients did not provide increments in randomized phase 2 trials [48,49]. The development of immune-checkpoint inhibitors that are likely to be approved in the salvage setting will impact the overall therapeutic strategy of UC in the next few years [50]. The possibility to obtain incrementally better outcomes with the triple chemotherapy combinations with taxanes, platinum, and gemcitabine may be considered in those with a performance status of 0–1, with the substitution of docetaxel or paclitaxel by a tubulin inhibitor with a batter therapeutic index. Moreover, UC is an aggressive malignancy that does not offer the opportunities to deliver multiple lines of therapy due to declines in performance status or comorbidities. Hence, rationale combination therapy development should occur in tandem with the development of new agents. For these patients, the possibility to administer the best therapeutic first-line and salvage options (ie, triple regimen, and immunotherapy at relapse) may now portend even better survival improvements than those observed in the present analysis.

unfortunately not available. The development of potent tubulin inhibitors should continue to be pursued, to complement efforts in the arena of immunotherapy and biologic agents. Administering the triple chemotherapy combination is likely the best option for patients who are fit for cisplatin and probably for those with bladder tumor primary, in the absence of more precise criteria for treatment selection. Author contributions: Andrea Necchi had full access to all the data in the study and takes responsibility for the integrity of the data and the accuracy of the data analysis. Study concept and design: Necchi, Sonpavde. Acquisition of data: Giannatempo, Raggi, Naik. Analysis and interpretation of data: Necchi, Pond, Sonpavde, Galsky, Bellmunt. Drafting of the manuscript: Giannatempo, Pond, Sonpavde, Raggi, Naik, Galsky, Bellmunt, Necchi. Critical revision of the manuscript for important intellectual content: Giannatempo, Pond, Sonpavde, Raggi, Naik, Galsky, Bellmunt, Necchi. Statistical analysis: Pond. Obtaining funding: None. Administrative, technical, or material support: None. Supervision: None. Other: None. Financial disclosures: Andrea Necchi certifies that all conflicts of interest, including specific financial interests and relationships and affiliations relevant to the subject matter or materials discussed in the manuscript (eg, employment/affiliation, grants or funding, consultancies, honoraria, stock ownership or options, expert testimony, royalties, or patents filed, received, or pending), are the following: None. Funding/Support and role of the sponsor: None.

Appendix A. Supplementary data Supplementary data associated with this article can be found, in the online version, at http://dx.doi.org/10.1016/j. eururo.2015.09.051.

References [1] Abida W, Bajorin DF, Rosenberg JE. First-line treatment and prognostic factors of metastatic bladder cancer for platinum-eligible patients. Hematol Oncol Clin North Am 2015;29:319–28. [2] Sonpavde G, Galsky MD, Vogelzang NJ. First-line systemic therapy trials for advanced transitional-cell carcinoma of the urothelium: should we stop separating cisplatin-eligible and -ineligible patients? J Clin Oncol 2010;28:e441–2. [3] Sonpavde G, Watson D, Tourtellot M, et al. Administration of cisplatin-based chemotherapy for advanced urothelial carcinoma

4.

Conclusions

in the community. Clin Genitourin Cancer 2012;10:1–5. [4] Galsky MD, Chen GJ, Oh WK, et al. Comparative effectiveness of

In conclusion, the results of the present meta-analysis remarkably parallels the results of the phase 3 trial comparing GC with or without paclitaxel. Indeed, the data suggests that a large meta-analysis of trials may provide data similar to a phase 3 trial, and in a highly cost-effective manner. The availability of individual patient-level data would have further enhanced our analysis, but were

cisplatin-based and carboplatin-based chemotherapy for treatment of advancd urothelial carcinoma. Ann Oncol 2012;23:406–10. [5] Galsky MD, Hahn NM, Rosenberg JE, et al. Treatment of patients with metastatic urothelial cancer ‘‘unfit’’ for cisplatin-based chemotherapy. J Clin Oncol 2011;29:2432–8. [6] McCaffrey JA, Hilton S, Mazumdar M, et al. Phase II trial of docetaxel in patients with advanced or metastatic transitional-cell carcinoma. J Clin Oncol 1997;16:1853–7.

Please cite this article in press as: Giannatempo P, et al. The Impact of Adding Taxanes to Gemcitabine and Platinum Chemotherapy for the First-Line Therapy of Advanced or Metastatic Urothelial Cancer: A Systematic Review and Meta-analysis. Eur Urol (2015), http://dx.doi.org/10.1016/j.eururo.2015.09.051

EURURO-6474; No. of Pages 10 EUROPEAN UROLOGY XXX (2015) XXX–XXX

[7] Roth BJ, Dreicer R, Einhorn LH, et al. Significant activity of paclitaxel

9

[24] Nogue´-Aliguer M, Carles J, Arrivi A, et al. Gemcitabine and carbo-

in advanced transitional-cell carcinoma of the urothelium: a phase

platin in advanced transitional cell carcinoma of the urinary tract:

II trial of the Eastern Cooperative Oncology Group. J Clin Oncol

an alternative therapy. Cancer 2003;97:180–6.

1994;12:2264–70.

[25] Hudson E, Lester JF. Gemcitabine and carboplatin in the treatment of

[8] Bellmunt J, von der Maase H, Mead GM, et al. Randomized phase III

transitional cell carcinoma of the urothelium: a single-centre expe-

study comparing paclitaxel/cisplatin/gemcitabine and gemcita-

rience and review of the literature. Eur J Cancer Care 2010;19:324–8.

bine/cisplatin in patients with locally advanced or metastatic

[26] Shannon C, Crombie C, Brooks A, et al. Carboplatin and gemcitabine

urothelial cancer without prior systemic therapy; EORTC Inter-

in metastatic transitional cell carcinoma of the urothelium: effec-

group Study 30987. J Clin Oncol 2012;30:1107–13. [9] von der Maase H, Andersen L, Crino` L, et al. Weekly gemcitabine and cisplatin combination therapy in patients with transitional cell carcinoma of the urothelium: a phase II clinical trial. Ann Oncol

tive treatment of patients with poor prognostic features. Ann Oncol 2001;12:947–52. [27] Xu N, Zhang XC, Xiong JP, et al. A phase II trial of gemcitabine plus cisplatin in advanced transitional cell carcinoma of the urothelium.

[10] Kaufman D, Raghavan D, Carducci M, et al. Phase II trial of gemci-

BMC Cancer 2007;7:98. [28] Dogliotti L, Cartenı` G, Siena S, et al. Gemcitabine plus cisplatin

tabine plus cisplatin in patients with metastatic urothelial cancer.

versus gemcitabine plus carboplatin as first-line chemotherapy in

1999;10:1–5.

J Clin Oncol 2000;18:1921–7. [11] Moore MJ, Winquist EW, Murray N, et al. Gemcitabine plus cisplatin, an active regimen in advanced urothelial cancer: a phase II trial

advanced transitional cell carcinoma of the urothelium: results of a randomized phase 2 trial. Eur Urol 2007;52:134–41. [29] Park JH, Lee SW, Kim HS, et al. Combination of gemcitabine and

of the National Cancer Institute of Canada Clinical Trials Group.

carboplatin as first line treatment in elderly patients or those unfit

J Clin Oncol 1999;17:2876–81.

for cisplatin-based chemotherapy with advanced transitional cell

[12] Hussain M, Daignault S, Agarwal N, et al. A randomized phase 2 trial of gemcitabine/cisplatin with or without ceuximab in patients with advanced urothelial carcinoma. Cancer 2014;120:2684–93.

carcinoma of the urinary tract. Cancer Chemother Pharmacol 2013; 71:1033–9. [30] Ecke TH, Bartel P, Koch S, Ruttloff J, Theissig F. Chemotherapy with

[13] Bamias A, Moulopoulos LA, Koutras A, et al. The combination of

gemcitabine, paclitaxel, and cisplatin in the treatment of patients

gemcitabine and carboplatin as first-line treatment in patients with

with advanced transitional cell carcinoma of the urothelium. Oncol

advanced urothelial carcinoma. A phase II study of the Hellenic Cooperative Oncology Group. Cancer 2006;106:297–303. [14] Morales-Barrera R, Bellmunt J, Suarez C, et al. Cisplatin and gemci-

Rep 2006;16:1381–8. [31] Hainsworth JD, Meluch AA, Litchy S, et al. Paclitaxel, carboplatin, and gemcitabine in the treatment of patients with advanced transitional

tabine every two weeks in patients with locally advanced or

cell carcinoma of the urothelium. Cancer 2005;103:2298–303.

metastatic urothelial carcinoma and impaired renal function. Eur

[32] Hussain M, Vaishampayan U, Du W, Redman B, Smith DC. Combi-

J Cancer 2012;48:1816–21.

nation paclitaxel, carboplatin, and gemcitabine is an active treat-

[15] Bellmunt J, de Wit R, Albanell J, Baselga J. A feasibility study of

ment in advanced urothelial cancer. J Clin Oncol 2001;19:2527–33.

carboplatin with fixed dose of gemcitabine in ‘‘unfit’’ patients with

[33] Lorusso V, Crucitta E, Silvestris N, et al. Randomized, open-label,

advanced bladder cancer. Eur J Cancer 2001;37:2212–5. [16] Carles J, Nogue´ M, Dome`nech M, et al. Carboplatin-gemcitabine

citabine and cisplatin as first-line chemotherapy in advanced tran-

phase II trial of paclitaxel, gemcitabine, and cisplatin versus gem-

treatment of patients with transitional cell carcinoma of the blad-

sitional cell carcinoma of the urothelium. Oncol Rep 2005;13:

der and impaired renal function. Oncology 2000;59:24–7.

283–7.

[17] Hoschke B, May M, Seehafer M, Helke C. Our experience with

[34] Pectasides D, Glotsos J, Bountouroglou N, et al. Weekly docetaxel,

23 consecutive patients on gemcitabine/carboplatin chemotherapy

gemcitabine and cisplatin in advanced transitional cell urothelial

for treatment of metastasized transitional cell carcinoma of the urothelium. Int J Urol 2004;11:461–6.

cancer: a phase II trial. Ann Oncol 2002;13:243–50. [35] Boukovinas I, Androulakis N, Kentepozidis N, et al. Chemotherapy

[18] Khaled H, Emara ME, Gaafar RM. Primary chemotherapy with low-

with gemcitabine, cisplatin, and docetaxel in the treatment for

dose prolonged infusion gemcitabine and cisplatin in patients with

patients with muscle-invasive bladder cancer: a multicenter phase

bladder cancer: a Phase II trial. Urol Oncol 2008;26:133–6. [19] Baitar A, De Vos M, Vandebroeck A, Schrijvers D. Carboplatin and gemcitabine in patients with advanced and/or metastatic urothelial cancers: a phase II study. J Ger Oncol 2011;2:31–5.

II study of the Hellenic Oncology Research Group (HORG). Cancer Chemother Pharmacol 2012;69:351–6. [36] von der Maase H, Hansen SW, Roberts JT, et al. Gemcitabine and cisplatin versus methotrexate, vinblastine, doxorubicin, and cisplatin

[20] Krege S, Rexer H, vom Dorp F, et al. Prospective randomizd double-

in advanced or metastatic bladder cancer: results of a large, random-

blind multicentre phase II study comparing gemcitabine and cis-

ized, multinational, multicenter, phase III study. J Clin Oncol 2000;

platin plus sorafenib chemotherapy with gemcitabine and cisplatin plus placebo in locally advanced and/or metastasized urohelial cancer: SUSE (AUO-AB 31/05). BJU Int 2014;113:429–36.

18:3068–77. [37] Bamias A, Dafni U, Karadimou A, et al. Prospective, open-label, rancomized, phase III study of two dose-dense regimens MVAC

[21] Lorusso V, Manzione L, De Vita F, et al. Gemcitabine plus cisplatin

versus gemcitabine/cisplatin in patients with inoperable, metastat-

for advanced transitional cell carcinoma of teh urinary tract: a

ic or relapsed urothelial cancer: a Hellenic Cooperative Oncology

phase II multicenter trial. J Urol 2000;164:53–6. [22] Bamias A, Lainakis G, Kastritis E, et al. Biweekly carbopaltin/gem-

Group study (HE 16/03). Ann Oncol 2013;24:1011–7. [38] Sternberg CN, Skoneczna IA, Castellano D, et al. Larotaxel with

citabine in patients with advanced urothelial cancer who are unfit

cisplatin in the first-line treatment of locally advanced/metastatic

for cisplatin-based chemotherapy: report of efficacy, quality of life

urothelial tract or bladder cancer: a randomized, active-controlled,

and geriatric assessment. Oncology 2007;73:290–7. [23] Linardou H, Aravantinos G, Efstathiou E, et al. Gemcitabine and

phase III trial (CILAB). Oncology 2013;85:208–15. [39] De Santis M, Bellmunt J, Mead G, et al. Randomized phase II/III trial

carboplatin combination as first-line treatment in elderly patients

assessing gemcitabine/carboplatin and methotrexate/carboplatin/

and those unfit for cisplatin-based chemotherapy with advanced

vinblastine in patients with advanced urothelial cancer who are

bladder carcinoma: Phase II study of the Hellenic Co-operative

unfit for cisplatin-based chemotherapy: EORTC study 30986. J Clin

Oncology Group. Urology 2004;64:479–84.

Oncol 2012;30:191–9.

Please cite this article in press as: Giannatempo P, et al. The Impact of Adding Taxanes to Gemcitabine and Platinum Chemotherapy for the First-Line Therapy of Advanced or Metastatic Urothelial Cancer: A Systematic Review and Meta-analysis. Eur Urol (2015), http://dx.doi.org/10.1016/j.eururo.2015.09.051

EURURO-6474; No. of Pages 10 10

EUROPEAN UROLOGY XXX (2015) XXX–XXX

[40] Liberati A, Altman DG, Tetzlaff J, et al. The PRISMA statement for

[46] Vogelzang NJ, Conkling P, Duran I, et al. Phase Ib/II study of

reporting systematic reviews and meta-analyses of studies that

eribulin mesylate administered in combination with gemcitabine/

evaluate health care interventions: explanation and elaboration.

cisplatin as first-line therapy for locally advanced or metastatic

PLoS Med 2009;339:b2700. [41] Bellmunt J, Petrylak DP. New therapeutic challenges in advanced bladder cancer. Semin Oncol 2012;39:598–607.

bladder cancer: phase Ib results. J Clin Oncol 2012;30(suppl 5; abstr 273). [47] Hahn NM, Stadler WM, ZonRT, et al. Phase II trial of cisplatin,

[42] Ko YJ, Canil CM, Mukherjee SD, et al. Nanoparticle albumin-bound

gemcitabine, and bevacizumab as first-line therapy for metastatic

paclitaxel for second-line treatment of metastatic urothelial carci-

urothelial carcinoma: Hoosier Oncology Group GU 04-75. J Clin

noma: a single group, multicentre, phase 2 study. Lancet Oncol 2013; 14:769–76.

Oncol 2011;29:1525–30. [48] Bellmunt J, Eigl BJ, Senkus-Konefka E, et al. First-line randomized

[43] Alva, Daignault S, Smith DC, Hussain M. Phase II trial of combination

phase II study of gemcitabine/cisplatin plus apatorsen or placebo in

nab-paclitaxel, carboplatin and gemcitabine in first-line therapy of

patients with advanced bladder cancer: the international Borealis-

advanced urothelial carcinoma. Invest New Drugs 2014;32:188–94.

1 trial. J Clin Oncol 2015;33(suppl; abstr 4503).

[44] Hutson TE, Vukelja S, Atienza D, et al. Phase I study of a 3-drug

[49] Oudard S, Culine S, Vano Y, et al. Multicentre randomised phase II

regimen of gemcitabine/cisplatin/pemetrexed in patients with

trial of gemcitabine+platinum, with or without trastuzumab, in

metastatic transitional cell carcinoma of the urothelium. Invest

advanced or metastatic urothelial carcinoma overexpressing Her2.

New Drugs 2008;26:151–8.

Eur J Cancer 2015;51:45–54.

[45] Quinn DI, Ruel N, Twardowski P, et al. Eribulin in advanced urothe-

[50] Powles T, Eder JP, Fine GD, et al. MPDL3280A (anti-PD-L1) treat-

lial cancer (AUC) patients (pts): a California Cancer Consortium trial

ment leads to clinical activity in metastatic bladder cancer. Nature

– NCI/CTEP 7435. J Clin Oncol 2015;33(suppl; abstr 4504).

2014;515:558–62.

Please cite this article in press as: Giannatempo P, et al. The Impact of Adding Taxanes to Gemcitabine and Platinum Chemotherapy for the First-Line Therapy of Advanced or Metastatic Urothelial Cancer: A Systematic Review and Meta-analysis. Eur Urol (2015), http://dx.doi.org/10.1016/j.eururo.2015.09.051